Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Chemical looping combustion : cold model hydrodynamics and modeling of methane combustion Xu, Min 2010

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

Item Metadata

Download

Media
24-ubc_2010_fall_xu_min.pdf [ 1.89MB ]
Metadata
JSON: 24-1.0058888.json
JSON-LD: 24-1.0058888-ld.json
RDF/XML (Pretty): 24-1.0058888-rdf.xml
RDF/JSON: 24-1.0058888-rdf.json
Turtle: 24-1.0058888-turtle.txt
N-Triples: 24-1.0058888-rdf-ntriples.txt
Original Record: 24-1.0058888-source.json
Full Text
24-1.0058888-fulltext.txt
Citation
24-1.0058888.ris

Full Text

Chemical Looping Combustion: Cold Model Hydrodynamics and Modeling of Methane Combustion by Min Xu B.Sc., Anhui University of Technology, 1994 M.Sc., Dalian University of Technology, 1997 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in The Faculty of Graduate Studies (Chemical and Biological Engineering) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) August 2010 © Min Xu, 2010 ii ABSTRACT A novel interconnected fluidized bed (IFB) reactor with a bypass line for chemical looping combustion (CLC) has been developed to overcome the problem of short residence time of oxygen carrier in the air reactor. A comprehensive hydrodynamic study was carried out on the cold-flow model of the proposed reactor. Detailed mapping of the operating conditions for the reactor system was studied. Pressure transducers were applied to investigate the pressure loops and the cross-sectional average solids hold-up along the air reactor. Solids circulation flux between the two reactors was measured using butterfly valves by estimating the time interval for collecting a given volume of solids. Helium was used as gas tracer for gas leakage measurement. The experiments examined the gas leakage from air reactor to fuel reactor, from fuel reactor to air reactor, from loop-seals to fuel reactor and from fuel reactor to the cyclone. For scaling consideration, the cold-flow reactor was operated with fluidizing gas mixture of helium and air to simulate the hydrodynamics of the hot unit. The effect of density ratio of solids to gas on the solids circulation flux, pressure loops and voidage distribution along the air reactor was investigated. The connection between cold unit and hot unit is achieved by applying a scaling law. It can be stated that the cold-flow model operated with fluidizing gas mixture of 96 vol% helium and 4 vol% air can be used to simulate the hydrodynamics of an atmospheric CLC hot unit. A comprehensive model for the investigation of the reactor is introduced by combining fluidization properties and a particle population balance for calculation of the bed particle conversion, considering the chemical reaction of a single particle. The dimensionless parameters, Mrfuel and Mrair, which represent the mass ratio of input oxidized-particles to the input fuel in unit time for the fuel reactor and the mass ratio of reduced-particles to the input oxygen in unit time for the air reactor, respectively, are introduced. The model shows that Mrfuel should be more than 50 for achieving fuel conversion of 90% in the fuel reactor and Mrair should be more than 60 for achieving oxygen conversion of 85% in the air reactor. A procedure for optimizing the performance of the atmospheric CLC reactor is developed. The modeling analysis indicated that the optimum operating condition of an atmospheric CLC reactor hot unit should be chosen as follows: fuel capacity is 80 kW, Ua0=6.6 m/s, Uf0= 0.076 m/s, UA1=4Umf, UA2=1Umf, and the temperature in air reactor is 1223 K and in fuel reactor is 1173 K. iii TABLE OF CONTENTS ABSTRACT ................................................................................................ .........................................ii TABLE OF CONTENTS ................................................................................................................... iii LIST OF TABLES................................................................................................ ............................ viii LIST OF FIGURES ..............................................................................................................................x NOMENCLATURE ....................................................................................................................... xviii ACKNOWLEDGEMENTS................................ ............................................................................xxxii CHAPTER 1 INTRODUCTION................................ ..........................................................................1 1.1 Introduction ................................................................................................ ................................1 1.2 Oxygen carrier development ......................................................................................................2 1.3 Reaction kinetic model ................................................................................................ ...............2 1.3.1 Unreacted shrinking core model..........................................................................................3 1.3.2 Power-law relation model ................................................................................................... 5 1.3.3 Nucleation and nuclei growth model...................................................................................5 1.3.4 Modified volumetric model................................ .................................................................6 1.4 Reactor design ............................................................................................................................6 1.5 Hydrodynamic study...................................................................................................................7 1.5.1 Operation condition and pressure profile ............................................................................ 7 1.5.2 Gas leakage................................ ..........................................................................................7 1.5.3 Solids hold-up in the reactors .............................................................................................. 8 1.6 Scale up ......................................................................................................................................8 1.7 Mathematical model for CLC reactor................................ .........................................................9 1.8 Summary and research objectives ..............................................................................................9 CHAPTER 2 DESIGN OF A COLD-FLOW CLC REACTOR.........................................................11 2.1 Requirements for the CLC reactor............................................................................................ 11 2.2 Literature review for CLC reactor design for gaseous fuel................................ ......................11 2.2.1 Four-compartment IFB................................................................ ......................................11 2.2.2 IFB with alternative valve .................................................................................................12 2.2.3 Packed bed with alternative valve..................................................................................... 13 2.2.4 IFB with combining fluidized bed and moving bed ..........................................................14 2.2.5 Two-compartment IFB ................................................................ ......................................14 2.2.6 IFB with combining two bubbling beds............................................................................ 15 iv 2.2.7 IFB with combining riser and turbulent bed................................ ......................................18 2.2.8 IFB with combining riser and bubbling bed................................ ......................................19 2.3 Flow regime of reactor ................................................................................................ .............21 2.4 Design of cold-flow CLC reactor at UBC ................................................................................22 CHAPTER 3 EXPERIMENTAL SET-UP.........................................................................................25 3.1 CLC reactor ................................................................................................ ..............................25 3.2 Particulate material ................................................................................................................... 27 3.3 Experimental unit at open system............................................................................................. 28 3.4 Experimental unit at close status ................................................................ ..............................30 3.5 Pressure measurement ................................................................................................ ..............30 3.6 Cross-sectional average solids hold-up along the air reactor ................................................... 33 3.7 Solids circulation flux measurement ........................................................................................33 3.8 Gas leakage measurement ................................ ........................................................................34 CHAPTER4 HYDRODYNAMIC STUDY WITH FLUIDIZING GAS OF AIR ............................. 35 4.1 Solids circulation flux................................................................ ...............................................35 4.1.1 Aeration velocity in loop-seal 2 at 0..................................................................................35 4.1.2 Aeration velocity in loop-seal 2 at 1Umf and 1.5Umf..........................................................38 4.1.2.1 Minimum superficial gas velocity for stable particle transportation..........................38 4.1.2.2 Solids circulation flux in loop 1 .................................................................................38 4.1.2.3 Solids circulation flux in loop 2 .................................................................................40 4.1.2.4 Total solids circulation flux........................................................................................42 4.1.3 Effect of aeration velocity in loop-seal 2 on solids circulation flux..................................42 4.1.3.1 Effect on solids circulation flux in loop 1 ..................................................................42 4.1.3.2 Effect on solids circulation flux in loop 2 ..................................................................42 4.1.3.3 Effect on total solids circulation flux .........................................................................45 4.1.4 Mapping of operation conditions for solids circulation flux ................................ .............46 4.2 Pressure loop ............................................................................................................................ 48 4.2.1 Effect of aeration velocity in loop-seal 1 ..........................................................................49 4.2.2 Effect of superficial gas velocity................................ .......................................................49 4.2.3 Effect of blockage in fitler.................................................................................................49 4.2.4 Effect of aeration velocity in loop-seal 2 ..........................................................................53 4.3 Cross-sectional average solids hold-up in the air reactor .........................................................54 4.3.1 Effect of aeration velocity in loop-seal 1 ..........................................................................54 4.3.2 Effect of superficial gas velocity................................ .......................................................56 v4.3.3 Effect of solids from loop 2................................ ...............................................................57 4.4 Summary................................................................................................................................... 57 CHAPTER5 GAS LEAKAGE ........................................................................................................... 59 5.1 Gas leakage from fuel reactor to air reactor ................................................................ .............59 5.1.1 Approaches to analyze gas leakage from fuel reactor to air reactor..................................59 5.1.2 Effect of pressure drop across loop-seal 1.........................................................................60 5.1.3 Effect of aeration velocity in loop-seal 1 ..........................................................................62 5.1.4 Effect of aeration velocity in loop-seal 2 ..........................................................................62 5.1.5 Summary for the gas leakage from fuel reactor to air reactor........................................... 64 5.2 Gas leakage from fuel reactor to cyclone .................................................................................64 5.2.1 Approaches to analyze gas leakage from fuel reactor to the cyclone................................64 5.2.2 Helium concentration fluctuation in air reactor and after-cyclones ..................................65 5.3 Gas leakage from air reactor to fuel reactor ................................................................ .............67 5.3.1 Approaches to analyze gas leakage ................................................................................... 67 5.3.2 Gas leakage through loop-seal 1........................................................................................68 5.3.3 Effect of solids circulation rate between air reactor and fuel reactor ................................68 5.3.4 Effect of aeration velocity in loop-seal 2 ..........................................................................69 5.3.5 Effect of aeration velocity in loop-seal 1 ..........................................................................70 5.3.6 Effect of filter blockage................................................................................................ .....71 5.4 Gas leakage from aeration gas of loop-seal 1 to fuel reactor ................................................... 72 5.4.1 Approaches in analyzing gas leakage................................................................................72 5.4.2 From F1 and F2 ports to fuel reactor.................................................................................73 5.4.3 From F3 and F4 ports to fuel reactor.................................................................................75 5.5 Gas leakage from aeration gas of loop-seal 2 to fuel reactor ................................................... 75 5.6 Summary................................................................................................................................... 76 CHAPTER 6 SCALING CONSIDERATIONS .................................................................................78 6.1 Equations for scaling ................................ ................................................................................78 6.2 Scaling of cold unit for a given particle size ............................................................................ 81 6.3 Scaling of cold unit for a given bed diameter........................................................................... 83 6.4 Scaling of cold unit for identical ratio of particles size to reactor size................................ .....85 6.5 Summary................................................................................................................................... 88 CHAPTER 7 HYDRODYNAMIC STUDY WITH FLUIDIZING GAS MIXTURE OF HELIUM AND AIR ..................................................................................................................................89 7.1 Experimental procedure............................................................................................................ 89 vi 7.2 Effect of helium concentration on the pressure ........................................................................90 7.3 Minimum superficial gas velocity for stable particle transportation ........................................91 7.4 Pressure profile................................ .........................................................................................92 7.5 Axial solids hold-up in air reactor ............................................................................................ 93 7.6 Solids circulation flux................................................................ ...............................................94 7.6.1 Effect of helium concentration on solids circulation flux in loop 1..................................94 7.6.2 Effect of helium concentration on solids circulation flux in loop 2..................................95 7.6.3 Effect of helium concentration on total solids circulation flux .........................................97 7.7 Summary................................................................................................................................... 97 CHAPTER 8 MODELING ANALYSIS OF METHANE COMBUSTION ....................................100 8.1 Modeling development................................................................................................ ...........100 8.1.1 Particle properties................................................................................................ ............101 8.1.2 General particle mass balance equation ..........................................................................101 8.1.3 Fuel reactor model................................................................................................ ...........103 8.1.3.1 Hydrodynamics of fuel reactor.................................................................................103 8.1.3.2 Mole balance of gas phase .......................................................................................105 8.1.3.3 Particle population balance ......................................................................................106 8.1.4 Air reactor model................................................................ .............................................106 8.1.4.1 Dense bottom zone................................................................................................ ...107 8.1.4.1.1 Mole balance of gas phase ................................................................................108 8.1.4.1.2 Particle population balance ...............................................................................109 8.1.4.2 Dilute transport zone ................................................................................................109 8.1.4.2.1 Hydrodynamics in dilute transport zone ........................................................... 110 8.1.4.2.2 Particle population balance ...............................................................................112 8.1.4.2.3 Mole balance of gas phase ................................................................................114 8.1.5 Reaction rate of single particle ........................................................................................115 8.1.6 Energy balance ................................................................................................................116 8.2 Computational procedure .......................................................................................................116 8.3 Results and discussion................................................................................................ ............122 8.3.1 Analysis on the fuel reactor................................ .............................................................122 8.3.2 Analysis on the air reactor...............................................................................................127 8.3.3 Linkage of air reactor and fuel reactor ................................................................ ............130 8.4 Procedure of model optimization of atmospheric CLC reactor..............................................137 8.5 Comparison of simulation results and experimental data .......................................................139 vii 8.6 Optimum operating condition for an atmospheric CLC reactor .............................................142 CHAPTER 9 CONCLUSIONS AND RECOMMENDATION .......................................................143 9.1 Conclusions from this study ................................................................................................ ...143 9.2 Recommendations for future work.........................................................................................145 REFERENCES .................................................................................................................................147 APPENDIX A SUMMARY OF PREVIOUS WORKS ON OXYGEN CARRIERS AND COLD- MODEL HYDRODYNAMICS ................................ ..............................................................160 APPENDIX B MEASUREMENT SYSTEM................................................................................... 182 B.1 Pressure measurement................................................................................................ ............182 B.2 Gas velocity measurement................................................................ .....................................186 B.3 Helium concentration measurement ......................................................................................187 APPENDIX C DRAWINGS OF THE MAIN COMPONENTS OF THE COLD-FLOW MODEL .................................................................................................................................................190 APPENDIX D ADDITIONAL TABLES AND FIGURES .............................................................202 APPENDIX E MODIFICATION OF THE ORIGINAL CLC COLD-FLOW MODEL .................227 APPENDIX F MEAUREMENT ERROR AND STANDARD DEVIATION ................................229 viii LIST OF TABLES Table 2.1 Features of fluidized beds in different flow regimes ..........................................................22 Table 3.1 Spent FCC particle properties............................................................................................. 27 Table 6.1 Hot unit modeled by a cold unit fluidized with 96 vol% helium and 4 vol% air at temperature of 40 oC and atmosphere pressure (based on a pre-determined particle size in hot unit) ........................................................................................................................................... 82 Table 6.2 Dimensionless group comparison for operation conditions in Table 6.1 ........................... 83 Table 6.3 Hot unit modeled by a cold unit fluidized with 96 vol% helium and 4 vol% air at temperature of 40 oC and atmosphere pressure (based on a given bed diameter in hot unit) .84 Table 6.4 Dimensionless group comparison for operation conditions in Table 6.3 ........................... 85 Table 6.5 Hot unit modeled by a cold unit fluidized with 96 vol% helium and 4 vol% air at temperature of 40 oC and atmosphere pressure (based on identical ratio of particle size to reactor size)................................................................ ...............................................................86 Table 6.6 Dimensionless group comparison for operation conditions in Table 6.5 ........................... 87 Table 7.1 Summary of Umstable and operating conditions of the experiments......................................92 Table 8.1 Expressions used in bubbling bed model for fuel reactor.................................................105 Table 8.2 Expressions used for the core-annulus flow structure in air reactor.................................111 Table 8.3 Kinetic parameters of single oxygen carrier (refer to Zafar et al. 2007a) ........................115 Table 8.4 Ralationship of the paramters between cold unit and hot unit..........................................131 Table 8.5 Main operating parameters of the hot unit........................................................................131 Table 8.6 Specific operation conditions for reference cases ................................ ............................132 Table 8.7 Performance parameters for the reference cases (Fuel reactor)........................................132 Table 8.8 Performance parameters for the reference cases (Air reactor) ........................................132 Table 8.9 Reactor geometry and typical operation conditions for CLC hot unit..............................139 Table 8.10 Comparison of dimensionless groups between the simulation hot model and the hot unit in Pröll et al. 2009b .................................................................................................................140 Table A.1. Summary of investigation on oxygen carrier (metal oxide) for chemical-looping combustion................................................................ ..............................................................161 Table A.2 Experiments about the hydrodynamics research on cold-flow model .............................181 Table B.1 Pressure transducers used in this study................................................................ ............182 Table D.1 Positions of ports for pressure measurement ................................................................ ...202 Table D.2 Combination of operating conditions ..............................................................................203 ix Table D.3 Hot unit modeled by a cold unit fluidized with 80 vol% helium and 20 vol% air at temperature of 40 oC and atmosphere pressure (based on identical ratio of particle size to reactor size)................................................................ .............................................................203 Table D.4 Dimensionless group comparison for operation conditions in Table D.3 .......................204 Table D.5 Hot unit modeled by a cold unit fluidized with 60 vol% helium and 40 vol% air at temperature of 40 oC and atmosphere pressure (based on identical ratio of particle size to reactor size)................................................................ .............................................................205 Table D.6 Dimensionless group comparison for operation conditions in Table D.5 .......................206 Table D.7 Hot unit modeled by a cold unit fluidized with 40 vol% helium and 60 vol% air at temperature of 40 oC and atmosphere pressure (based on identical ratio of particle size to reactor size)................................................................ .............................................................207 Table D.8 Dimensionless group comparison for operation conditions in Table D.7 .......................208 Table D.9 Hot unit modeled by a cold unit fluidized with 20 vol% helium and 80 vol% air at temperature of 40 oC and atmosphere pressure (based on identical ratio of particle size to reactor size)................................................................ .............................................................209 Table D.10 Dimensionless group comparison for operation conditions in Table D.9 .....................210 Table D.11 Hot unit modeled by a cold unit fluidized with 100 vol% air at temperature of 40 oC and atmosphere pressure (based on identical ratio of particle size to reactor size).......................211 Table D.12 Dimensionless group comparison for operation conditions in Table D.11 ................... 212 Table D.13 Operating conditions of the simulation runs for fuel reactor.........................................213 Table D.14 Operating conditions of the simulation runs for air reactor................................ ...........214 Table F.1 Standard deviation of fluctuaton of gas leakage from fuel reactor to air reactor for Figure 5.2 (Ua0=2.5 m/s)................................................................ .....................................................234 Table F.1 Standard deviation of fluctuaton of gas leakage from fuel reactor to air reactor for Figure 5.4 (Ua0=5.0 m/s)................................................................ .....................................................234 Table F.1 Standard deviation of fluctuaton of gas leakage from fuel reactor to air reactor for Figure 5.5 (Ua0=2.5m/s)................................ ......................................................................................234 xLIST OF FIGURES Figure 1.1 Principle of chemical looping combustion (M, MxOy and MxOy-z represent metal, metal oxide and reduced metal oxide)................................................................ ................................2 Figure 2.1 Schematic diagram of four-compartment IFB reactor (adapted from snieders et al.1999) ................................................................................................................................................... 12 Figure 2.2 Schematic diagram of IFB reactor with alternating valve.................................................12 Figure 2.3 Schematic diagram of periodically operated packed bed reactor (adapted from Noorman et al. 2007)................................ .................................................................................................13 Figure 2.4 Schematic diagram of IFB combining fluidized bed and moving bed..............................14 Figure 2.5 Schematic diagram of two-compartment IFB (adapted from Rydén et al.2008a) ............ 15 Figure 2.6 Schematic diagram of IFB combining two bubbling beds (adapted from Adánez et al. 2006a)................................................................ ........................................................................16 Figure 2.7 Schematic diagram of IFB combining two bubbling beds (adapted from Adánez et al. 2009)..........................................................................................................................................16 Figure 2.8 Schematic diagram of IFB combining two bubbling beds (adapted from Son et al. 2006) ................................................................................................................................................... 17 Figure 2.9 Schematic diagram of IFB combining two bubbling beds (adapted from Ryu 2008a) .....18 Figure 2.10 Schematic diagram of IFB combining riser and turbulent bed (adapted from Kolbitsch et al. 2009a)................................................................................................................................... 19 Figure 2.11 Schematic diagram of IFB combining riser and bubbling bed (adapted from Lyngfelt et al. 2004)................................................................................................................................ .....20 Figure 2.12 Schematic diagram of IFB combining riser and bubbling bed (adapted from Kronberger et al. 2005)................................ .................................................................................................21 Figure 2.13 Schematic diagram of the proposed cold-flow CLC reactor........................................... 24 Figure 3.1 Schematic diagram of the cold-flow CLC reactor at UBC ...............................................26 Figure 3.2 Schematic diagram of the fuel reactor................................ ...............................................27 Figure 3.3 Particle size distribution of spent FCC particles ...............................................................28 Figure 3.4 Experimental system at open status with fluidizing gas of air ..........................................29 Figure 3.5 Experimental system at close status with fluidizing gas of mixture of air and helium .....31 Figure 3.6 Port locations at the cold model CLC for pressure measurement ................................ .....32 Figure 4.1 Effect of superficial gas velocity in the air reactor and aeration velocity in loop-seal 1 on the solids circulation flux in loop 1, UA2=0Umf................................ .......................................36 xi Figure 4.2 Measurement error bar for the solids circulation flux in loop 1 (Ua0=3.0 and 4.0 m/s).... 37 Figure 4.3 Solids flow pattern from fuel reactor to air reactor through loop-seal 1 ........................... 37 Figure 4.4 Effect of superficial gas velocity in air reactor and aeration velocity in loop-seal 1 on the solids circulation flux in loop 1, UA2=1Umf ................................ ...............................................39 Figure 4.5 Effect of superficial gas velocity in air reactor and aeration velocity in loop-seal 1 on the solids circulation flux in loop 1, UA2=1.5Umf..........................................................................39 Figure 4.6 Solids flow between air reactor and primary downcomer in loop 2 .................................40 Figure 4.7 Effect of superficial gas velocity in air reactor on the solids circulation flux in loop 2, UA2=1Umf................................ .................................................................................................41 Figure 4.8 Effect of superficial gas velocity in air reactor on the solids circulation flux in loop 2, UA2=1.5Umf................................................................................................ ..............................41 Figure 4.9 Effect of superficial gas velocity in air reactor and aeration velocity in loop-seal 1 on the total solids circulation flux, UA2=1Umf.................................................................................... 43 Figure 4.10 Effect of superficial gas velocity in air reactor and aeration velocity in loop-seal 1 on the total solids circulation flux, UA2=1.5Umf .................................................................................43 Figure 4.11 Effect of aeration velocity in loop-seal 2 on solids circulation flux in loop 1, Ua0=4.0 m/s ................................................................................................................................ .............44 Figure 4.12 Effect of aeration velocity in loop-seal 2 on solids circulation flux in loop 2, Ua0=4.0 m/s ................................................................................................................................ .............44 Figure 4.13 Effect of aeration velocity in loop-seal 2 on total solids circulation flux, Ua0=4.0 m/s ................................................................................................................................................... 45 Figure 4.14 Mapping of operation conditions (Ua0=2.5 m/s, U f0=0.029 m/s) .................................... 46 Figure 4.15 Mapping of operation conditions (Ua0=3.0 m/s, U f0=0.035 m/s) .................................... 47 Figure 4.16 Mapping of operation conditions (Ua0=4.0 m/s, U f0=0.046 m/s) .................................... 47 Figure 4.17 Mapping of operation conditions (Ua0=5.0 m/s, U f0=0.058 m/s) .................................... 48 Figure 4.18 Pressure loop 1, U0=3.0 m/s, UA2=1.5Umf........................................................................50 Figure 4.19 Pressure loop 2, U0=3.0 m/s, UA2=1.5Umf........................................................................50 Figure 4.20 Pressure loop 1, UA1=4Umf, UA2=1.5Umf ..........................................................................51 Figure 4.21 Pressure loop 2, UA1=4Umf, UA2=1.5Umf ..........................................................................51 Figure 4.22 Pressure loop 1, U0=4.0 m/s, UA1=4Umf, UA2=1.5Umf................................ ......................52 Figure 4.23 Pressure loop 2, U0=4.0 m/s, UA1=4Umf, UA2=1.5Umf................................ ......................52 Figure 4.24 Pressure loop 1, U0=4.0 m/s, UA1=6Umf........................................................................... 53 Figure 4.25 Pressure loop 2, U0=4.0 m/s, UA1=6Umf........................................................................... 54 xii Figure 4.26 Cross sectional average solids hold-up distribution in air reactor U0=3.0 m/s, UA2=1.5Umf................................................................................................ ..............................55 Figure 4.27 Cross sectional average solids hold-up distribution in air reactor (Partial enlargement figure of Figure 4.24) U0=3.0 m/s, UA2=1.5Umf................................ ......................................55 Figure 4.28 Cross sectional average solids hold-up distribution in air reactor UA1=4Umf, UA2=1.5Umf ................................................................ ................................................................................56 Figure 4.29 Cross sectional average solids hold-up distribution in air reactor UA1=6Umf, Ua0=4.0 m/s ................................................................................................................................................... 57 Figure 5.1 Possible routes for gas leakage from fuel reactor to air reactor ........................................60 Figure5.2 Leakage from fuel reactor to air reactor versus the pressure drop across loop-seal 1, PD3- PB2, Ua0=2.5 m/s, Uf0=0.029 m/s ............................................................................................. 61 Figure 5.3 Leakage from fuel reactor to air reactor versus the pressure drop across loop-seal 1, PD3- PB2 (Ua0=4.0 m/s, Uf0=0.046 m/s, UA1=1Umf ~6Umf, UA2=0Umf ~1.5Umf).................................62 Figure 5.4 Leakage from fuel reactor to air reactor versus aeration velocity in loop-seal 1, Ua0=5.0m/s, Uf0=0.058 m/s ................................................................ .......................................63 Figure 5.5 Leakage from fuel reactor to air reactor versus aeration velocity for loop-seal 2, Ua0=2.5 m/s, Uf0=0.029 m/s .................................................................................................................... 63 Figure 5.6 Possible route for gas leakage from fuel reactor to cyclone................................ ..............65 Figure 5.7 Helium concentration fluctuation in air reactor and at the gas outlet of cyclone (Ua0=2.5 m/s, Uf0=0.029 m/s, UA1=2Umf, UA2=1Umf) ................................................................................66 Figure 5.8 Helium concentration fluctuation in air reactor and at the gas outlet of cyclone when the bottom end of the lower downcomer is opened completely (Ua0=2.5 m/s, Uf0=0.029 m/s, UA1=2Umf, UA2=1Umf)................................................................................................ ..............66 Figure 5.9 Possible routes for gas leakage from air reactor to fuel reactor ........................................67 Figure 5.10 Comparison of helium concentrations in fuel reactor when inject helium at A and B (Ua0=3.0 m/s, Uf0=0.035 m/s, UA1=4Umf, UA2=1Umf) ................................ ..............................68 Figure 5.11 Gas leakage from air reactor to fuel reactor versus solids circulation flux between air reactor and fuel reactor (Ua0=2.5 m/s, Uf0=0.029 m/s) .........................................................69 Figure 5.12 Leakage from fuel reactor to air reactor versus aeration velocity in loop-seal 2, UA2 (Ua0=3.0 m/s, Uf0=0.035 m/s) .................................................................................................70 Figure 5.13 Leakage from fuel reactor to air reactor versus aeration velocity in loop-seal 1, UA1 (Ua0=3.0 m/s, Uf0=0.035 m/s) .................................................................................................70 Figure 5.14 Effect of blockage in filter on the gas leakage from air reactor to fuel reactor...............71 xiii Figure 5.15 Comparison of gas leakage from air reactor to fuel reactor between no blockage in filter and blockage in filter (Ua0=4.0 m/s, Uf0=0.046 m/s, UA2=1.5Umf) .........................................72 Figure 5.16 Helium injection ports for measurement of gas leakage from loop-seal 1 to fuel reactor ................................................................................................................................................... 73 Figure 5.17 Aeration velocity in loop-seal 2, UA2, versus gas leakage from F1 port to fuel reactor (Ua0=2.5 m/s, Uf0=0.029 m/s) .................................................................................................74 Figure 5.18 Aeration velocity in loop-seal 2, UA2, versus gas leakage from F2 port to fuel reactor (Ua0=2.5 m/s, Uf0=0.029 m/s) .................................................................................................75 Figure 5.19 Helium injection ports for measurement of gas leakage from loop-seal 2 to fuel reactor .................................................................................................................................................. 76 Figure 7.1 Pressures at the bottom of the air reactor, before the orifice and across the orifice with increasing the helium concentration in the gas mixture, gas volumetric flow rate=136.4 Nm3/hr................................................................................................................................ .....91 Figure 7.2 Pressure loop 1 versus helium concentration, Ua0=5.0 m/s, UA1=4Umf, UA2=0Umf............ 92 Figure 7.3 Solids hold-up distribution in air reactor at different helium concentration (Ua0=5.0 m/s, UA1=4Umf, UA2=1.5Umf) ........................................................................................................... 93 Figure 7.4 Solids circulation flux in loop 1 versus helium concentration in gas mixture for different superficial gas velocities in air reactor, Ua0 (UA1=4.0 Umf, UA2=1.5Umf) ................................95 Figure 7.5 Solids circulation flux in loop 2 versus helium concentration in gas mixture for different superficial gas velocities in air reactor, Ua0 (UA1=4Umf, UA2=1.5Umf) .................................... 96 Figure 7.6 Total solids circulation flux versus helium concentration in gas mixture for different superficial gas velocities in air reactor, Ua0 (UA1=4Umf, UA2=1.5Umf) .................................... 97 Figure 8.1 Schematic diagram of CLC fluidized bed reactor development .....................................100 Figure 8.2 Solids population balance for a controlled volume.........................................................102 Figure 8.3 Flow structure and solid flows in fuel reactor.................................................................104 Figure 8.4 Flow structure in air reactor ................................................................ ............................107 Figure 8.5 Core-annulus structure in the dilute transport zone of air reactor ................................ ...110 Figure 8.6 Mole balance of gas and mass balance solids in one discrete element ........................... 113 Figure 8.7 Gas-solids flow in reactor system ................................................................................... 117 Figure 8.8 Energy balance in reactor system................................................................ ....................118 Figure 8.9 Flow chart for determining temperatures of the air and fuel reactors .............................119 Figure 8.10 Flow chart of the mass balance for the air reactor ........................................................120 Figure 8.11 Flow chart of the mass balance for the fuel reactor ......................................................121 xiv Figure 8.12 Effect of temperature of fuel reactor, T2 , on the fuel reactor performance (Fs0,fuel=0.12 kg/s, Ws,fuel=19 kg, F0,CH4=1.42 g/s, fuelinX , =40% .................................................................122 Figure 8.13 Minimum solids circulation flux and solids circulaton rate to achieve temperature in fuel reactor at 1173 K (T1=1223 K)................................................................................................123 Figure 8.14 Effect of input solids mass flow rate of fuel reactor, Fs0,fuel, on fuel reactor performance (Ws,fuel=18 kg, F0,CH4=1.42 g/s, fuelinX , =10%, T2=1173 K.....................................................124 Figure 8.15 Effect of average oxidation degree of input particles, fuelinX , , on fuel reactor performance (Fs0,fuel=0.48 kg/s, Ws,fuel=18 kg, F0,CH4=1.42 g/s, T2=1173 K) ........................124 Figure 8.16 Effect of superficial gas velocity of fuel reactor, Uf0, on fuel reactor performance (Fs0,fuel=0.48 kg/s, Ws,fuel=18 kg, fuelinX , =40%, T2=1173 K)..................................................125 Figure 8.17 Effect of bed materials in fuel reactor, Wfuel, on fuel reactor performance (Fs0,fuel=0.48 kg/s, F0,CH4=1.42 g/s, fuelinX , =10%, T2=1173 K) ..................................................................126 Figure 8.18 Effect of dimensionless parameter Mrfuel (T2=1173 K) ................................................126 Figure 8.19 Effect of dimensionless parameter Mrfuel (partial enlargement of Figure 8.17) ...........127 Figure 8.20 Effect of solids mas flow rate of input particles of air reactor, Fs0,air, on the performance of air reactor (Ua0=6.6 m/s, F0,O2=7.02 g/s, fuelinX , =60%, T1=1223 K)..............................128 Figure 8.21 Effect of average oxidation degree of input particles for air reactor, airinX , , on air reactor performance (Ua0=6.6 m/s, Fs0,air=0.49 kg/s, F0,O2=7.02 g/s, T1=1223 K) ........................... 128 Figure 8.22 Effect of superficial gas velocity in air reactor, Ua0, on air reactor performance (Fs0,air=1.15 kg/s, fuelinX , =0%, T1=1223 K)................................................................ ...........129 Figure 8.23 Effect of dimensionless parameter Mrair on the oxygen conversion .............................130 Figure 8.24 Oxygen conversion along air reactor................................ .............................................133 Figure 8.25 Fuel conversion along fuel reactor ................................................................................134 Figure 8.26 Fuel conversion along fuel reactor (effect of gas volume expansion)........................... 135 Figure 8.27 Particle distribution of the input particles of fuel reactor (Case1) ................................136 Figure 8.28 Particle distribution of the output particles of fuel reactor (Case1) ..............................136 Figure 8.29 Flow chart for optimization using model ......................................................................138 Figure 8.30 Comparison between simulation result and experimental data .....................................141 Figure B.1 Pressure transducer calibration system................................................................ ...........183 Figure B.2 Calibration results for pressure transducers (PX164-010D5V) ......................................183 Figure B.3 Calibration results for pressure transducers (PX142-001D5V) ......................................184 Figure B.4 Calibration results for pressure transducers (PX142-005D5V) ......................................184 xv Figure B.5 Calibration results for pressure transducers (PX142-015D5V) ......................................185 Figure B.6 Calibration results for pressure transducers (PX142-030A5V) ......................................185 Figure B.7 Location of the orifice meter ..........................................................................................186 Figure B.8 Typical curve produced by Micro GC for helium concentration....................................188 Figure B.9 Experimental system for Micro GC calibration on helium concentration......................188 Figure B.10 Calibration result of Micro GC for helium concentration measurement ......................189 Figure C.1 Distributor of air reactor .................................................................................................190 Figure C.2 Lower section of air reactor................................................................ ............................191 Figure C.3 Middle section of air reactor........................................................................................... 192 Figure C.4 Upper section of air reactor ................................................................ ............................193 Figure C.5 Lower section of fuel reactor..........................................................................................194 Figure C.6 Middle section of fuel reactor.........................................................................................195 Figure C.7 Upper section of fuel reactor ..........................................................................................196 Figure C.8 Loop-seal 1 ................................................................ .....................................................197 Figure C.9 Loop-seal 2 ................................................................ .....................................................198 Figure C.10 Primary cyclone................................................................................................ ............199 Figure C.11 Secondary cyclone........................................................................................................200 Figure C.12 Internal cyclone in fuel reactor................................ .....................................................201 Figure D.1 Leakage from fuel reactor to air reactor versus the pressure drop across loop-seal 1 (Ua0=3.0 m/s) ........................................................................................................................214 Figure D.2 Leakage from fuel reactor to air reactor versus the pressure drop across loop-seal 1 (Ua0=4.0 m/s) ........................................................................................................................215 Figure D.3 Leakage from fuel reactor to air reactor versus the pressure drop across loop-seal 1 (Ua0=5.0 m/s) ........................................................................................................................215 Figure D.4 Leakage from fuel reactor to air reactor versus aeration velocity for loop-seal 2, UA2 (Ua0=3.0 m/s) ........................................................................................................................216 Figure D.5 Leakage from fuel reactor to air reactor versus aeration velocity for loop-seal 2, UA2 (Ua0=4.0 m/s) ........................................................................................................................216 Figure D.6 Leakage from fuel reactor to air reactor versus aeration velocity for loop-seal 2, UA2 (Ua0=5.0 m/s) ........................................................................................................................217 Figure D.7 Gas leakage from air reactor to fuel reactor versus solids circulation flux between air reactor and fuel reactor (Ua0=3.0 m/s) ................................................................................217 Figure D.8 Gas leakage from air reactor to fuel reactor versus solids circulation flux between air reactor and fuel reactor (Ua0=4.0 m/s) ................................................................................218 xvi Figure D.9 Gas leakage from air reactor to fuel reactor versus solids circulation flux between air reactor and fuel reactor (Ua0=5.0 m/s) ................................................................................218 Figure D.10 Aeration velocity in loop-seal 2 versus gas leakage from F1 port to fuel reactor (Ua0=3.0 m/s) ........................................................................................................................219 Figure D.11 Aeration velocity in loop-seal 2 versus gas leakage from F2 port to fuel reactor (Ua0=3.0 m/s) ........................................................................................................................219 Figure D.12 Aeration velocity in loop-seal 2 versus gas leakage from F1 port to fuel reactor (Ua0=4.0 m/s) ........................................................................................................................220 Figure D.13 Aeration velocity in loop-seal 2 versus gas leakage from F2 port to fuel reactor (Ua0=4.0 m/s) ........................................................................................................................220 Figure D.14 Aeration velocity in loop-seal 2 versus gas leakage from F1 port to fuel reactor (Ua0=5.0 m/s) ........................................................................................................................221 Figure D.15 Aeration velocity in loop-seal 2 versus gas leakage from F2 port to fuel reactor (Ua0=5.0 m/s) ........................................................................................................................221 Figure D.16 Pressure loop 1 versus helium concentration (Ua0=5.0 m/s, UA1=4Umf, UA2=1Umf) .....222 Figure D.17 Pressure loop 2 versus helium concentration (Ua0=5.0 m/s, UA1=4Umf, UA2=1Umf) .....222 Figure D.18 Pressure loop 1 versus helium concentration (Ua0=5.0 m/s, UA1=4Umf, UA2=1.5Umf) ..223 Figure D.19 Pressure loop 2 versus helium concentration (Ua0=5.0 m/s, UA1=4Umf, UA2=1.5Umf) ..223 Figure D.20 Solids fraction distribution in air reactor at different helium concentration(Ua0=5.0 m/s, UA1=4Umf, UA2=0Umf) ................................................................................................ ............224 Figure D.21 Solids fraction distribution in air reactor at different helium concentration (Ua0=5.0 m/s, UA1=4Umf, UA2=1Umf) ................................................................................................ ............224 Figure D.22 Solids circulation flux in loop 1 versus helium concentration in gas mixture for different actual superficial gas velocities (UA1=4Umf, UA2=0Umf)........................................................225 Figure D.23 Solids circulation flux in loop 1 versus helium concentration in gas mixture for different actual superficial gas velocities (UA1=4Umf, UA2=1Umf)........................................................225 Figure D.24 Solids circulation flux in loop 2 versus helium concentration in gas mixture for different superficial gas velocities in air reactor, Ua0 (UA1=4Umf, UA2=1Umf) .....................................226 Figure D.25 Total solids circulation flux versus helium concentration in gas mixture for different actual superficial gas velocities. (UA1=4Umf, UA2=1Umf).......................................................226 Figure E.1 Schematic of the CLC cold-flow model before-modification .......................................228 Figure E.2 Schematic of the CLC cold-flow model after-modification ..........................................228 Figure F.1 Measurement error bar for Figure 4.1 ................................................................ ............229 Figure F.2 Measurement error bar for Figure 4.4 ................................................................ ............229 xvii Figure F.3 Measurement error bar for Figure 4.5 ................................................................ ............230 Figure F.4 Measurement error bar for Figure 4.7 ................................................................ ............230 Figure F.5 Measurement error bar for Figure 4.8 ................................................................ ............231 Figure F.6 Measurement error bar for Figure 4.9 ................................................................ ............231 Figure F.7 Measurement error bar for Figure 4.10 ..........................................................................232 Figure F.8 Measurement error bar for Figure 4.11 ..........................................................................232 Figure F.9 Measurement error bar for Figure 4.12 ..........................................................................233 Figure F.10 Measurement error bar for Figure 4.13 ........................................................................233 xviii NOMENCLATURE a )/( 0100100   , dimensionless denseba , Bubble surface area per unit volume in dense zone, 1/m fuelba , Bubble surface area per unit volume in fuel reactor, 1/m oxa Pre-exponential factor for oxidation, 1/s pta Pressure transducer constant, V reda Pre-exponential factor for reduction, 1/s 0A Intermediate parameter in calculating orifice discharge coefficient aA Cross-sectional area of annulus, m 2 airA Cross-sectional area of air reactor, m 2 cA Cross-sectional area of core, m 2 fuelA Cross-sectional area of fuel reactor, m 2 Ar Archimedes number, dimensionless b Stoichometric factor for chemical reaction, dimensionless 0b Intermediate parameter in calculating orifice discharge coefficient oxb Stoichometric factor for oxidation, dimensionless reb Stoichometric factor for reduction, dimensionless 0B Intermediate parameter in calculating orifice discharge coefficient C Gas reactant concentration, mol/m3 AC Concentration of gas reactant A, mol/m 3 xix 0AC Bulk concentration of gaseous reactant, mol/ m 3 0BC Initial concentration of solid reactant, mol/ m 3 2,0 OC Oxygen concentration in the feeding gas flow of air reactor, mol/m 3 4,0 CHC Methane concentration in the feeding gas flow of fuel reactor, mol/m 3 bCHC ,4 Methane concentration in bubble phase, mol/m 3 eCHC ,4 Methane concentration in emulsion phase, mol/m 3 inbCHC ,,4 Methane concentration in bubble phase when Z=0 in fuel reactor, mol/ m3 ineCHC ,,4 Methane concentration in emulsion phase when Z=0 in fuel reactor, mol/ m3 iinOcC ,,2, Oxygen concentration in the input gas flow of core, mol/m 3 ioutOcC ,,2, Oxygen concentration in the output gas flow of core, mol/m 3 eqC Gas concentration at equilibrium conditions, mol/m 3 bOC ,2 Oxygen concentration in bubble phase of dense zone, mol/ m 3 eOC ,2 Oxygen concentration in emulsion phase, mol/ m 3 inbOC ,,2 Oxygen concentration in bubble phase when Z=0 in air reactor, mol/ m3 ineOC ,,2 Oxygen concentration in emulsion phase when Z=0 in air reactor, mol/ m3 2,OoutC Oxygen concentration in the gas outflow from air reactor, mol/ m 3 pd Mean particle diameter, m pcd Mean particle diameter in cold unit, m xx phd Mean particle diameter in hot unit, m D Diameter of column, m 0D Diameter of orifice, m 1D Diameter of pipe, m acD Solid mass transfer coefficient from annulus to core, m/s densebD , Bubble size in dense zone, m fuelbD , Bubble size in fuel reactor, m cD Diameter of column in cold unit, m caD Solid mass transfer coefficient from core to annulus, m/s eD Effective diffusion coefficient, m 2/s hD Diameter of column in hot unit, m E Activation energy for chemical reaction, kJ/mol oxE Activation energy for oxidation, kJ/mol oxaE , Activation energy for oxidation, kJ/mol reE Activation energy for reduction, kJ/mol reaE , Activation energy for reduction, kJ/mol 0F Mass flow rate of the particles inputted into a controlled volume, kg/s 4,0 CHF Molar flow rate of input methane for fuel reactor, mol/s 2,0 NF Molar flow rate of input nitrogen for air reactor, mol/s 2,0 OF Molar flow rate of input oxygen for air reactor, mol/s xxi 1F Mass flow rate of the particles moving out from a controlled volume, kg/s iinasF ,, Input solid mass flow rates of the annulus, kg/s ioutasF ,, Output solid mass flow rate of the annulus, kg/s iincsF ,, Input solid mass flow rates of the core, kg/s ioutcsF ,, Output solid mass flow rate of the core, kg/s iinOcF ,,2, Input oxygen molar flow rate of core, mol/s ioutOcF ,,2, Output oxygen molar flow rate of core, mol/s 4,CHoutF Molar flow rate of output methane from fuel reactor, mol/s 2,COoutF Molar flow rate of output carbon dioxide from fuel reactor, mol/s OHoutF 2, Molar flow rate of output steam from fuel reactor, mol/s 2,NoutF Molar flow rate of output nitrogen from air reactor, mol/s 2,OoutF Molar flow rate of output oxygen from air reactor, mol/s airsF ,0 Mass flow rate of feeding solids flow into air reactor, kg/s fuelsF ,0 Mass flow rate of feeding solids flow for fuel rector, kg/s 1sF Solids mass flow rate in the loop 1, kg/s 2sF Solids mass flow rate in the loop 2, kg/s saF Internal downward solids mass flow rate in the annulus, kg/s scF Internal upward solids mass flow rate in the core, kg/s sdF Mass flow rate of solids flow returned into dense zone from dilute transport zone in air reactor , kg/s airoutsF ,, Mass flow rate of solids outflow from air reactor, kg/s xxii fueloutsF ,, Mass flow rate of solids outflow because of overflow from fuel reactor, kg/s stF Mass flow rate of solids flow entrained into the dilute transport zone from dense zone, kg/s g Acceleration due to gravity, m/s2 sG Circulation flux of solid, kg/ m 2.s 1sG Solids circulation flux in loop 1, kg/ m 2.s 2sG Solids circulation flux in loop 2, kg/ m 2.s scG Solids circulation flux in cold unit, kg/ m 2.s shG Solids circulation flux in hot unit, kg/ m 2.s stG Total solids circulation flux, kg/ m 2.s h Enthalpy of gas species or solids, kJ/mol or kJ/kg, respectively 4,0 CHh Enthalpy of output methane from fuel reactor, kJ/mol 2,0 Nh Enthalpy of output nitrogen from air reactor, kJ/mol 2,0 Oh Enthalpy of output oxygen from air reactor, kJ/mol 4,CHouth Enthalpy of output methane from fuel reactor, kJ/mol 2,COouth Enthalpy of output carbon dioxide from fuel reactor, kJ/mol OHouth 2, Enthalpy of output steam from fuel reactor, kJ/mol 2, Nouth Enthalpy of output nitrogen from air reactor, kJ/mol 2,Oouth Enthalpy of output oxygen from air reactor, kJ/mol airoutsh ,, Enthalpy of solids outflow from air reactor, kJ/kg fueloutsh ,, Enthalpy of solids outflow because of overflow from fuel reactor, kJ/kg xxiii H Height of column, m 1H Height of the accumulated particles in loop-seal 1, m 2H Height of the accumulated particles in primary downcomer, m denseH Height of dense zone of air reactor, m diluteH Height of dilute zone of air reactor, m 0k Pre-exponential factor of the chemical reaction rate constant, model specific oxk0 Pre-exponential factor of the chemical reaction rate constant for oxidation, m/s rek0 Pre-exponential factor of the chemical reaction rate constant for reduction, mol0.6m-0.8/s 0K Orifice discharge coefficient, dimensionless fuelbeK , Gas mass transfer coefficient between bubble and emulsion phase in fuel reactor, m/s densebeK , Gas mass transfer coefficient between bubble and emulsion phase in dense zone, m/s caK Gas mass transfer coefficient between core and annulus, m/s eK Orifice discharge coefficient when Re0=Ree, dimensionless gk Mass transfer coefficient of gaseous reactant, m/s sk Reaction rate constant a the surface of the unreacted core of single particle, m/s ptk Pressure transducer coefficient, kPa/V L Typical bed dimensions xxiv actualM Actual mass of oxygen carrier, kg oxfM , Mass of oxygen carrier in the fully-oxidized form, kg redfM , Mass of oxygen carrier in the fully-reduced form, kg fuelMr Mass ratio of oxidized input solids to the input fuel in unit time for fuel reactor, dimensionless airMr Mass ratio of reduced input particles to the input oxygen in unit time for air reactor, dimensionless n Reaction order, dimensionless 0n Intermediate parameter in calculating orifice discharge coefficient avn Avrami constant, dimensionless oxn Reaction order for oxidation, dimensionless ren Reaction order for reduction, dimensionless P Pressure, kPa 0P Particle distribution function of particles inputted into a controlled volume, 1/μm airP ,0 Particle distribution function of feeding solids flow into air reactor, 1/μm fuelP ,0 Particle distribution function of feeding solids flow for fuel reactor, 1/μm 1P Particle distribution function of particles moving out from a controlled volume, 1/μm iabP , Particle distribution functions of particles in the annulus, 1/μm iinasP ,, Particle distribution functions of input solids flow of the annulus, 1/μm ioutasP ,, Particle distribution functions of output solids flow of the annulus, 1/μm xxv bP Particle distribution function of particles in a controlled volume, 1/μm densebP , Particle distribution function of the particles in dense zone of air reactor, 1/μm fuelbP , Particle distribution function of the particles in fuel reactor, 1/μm cP Pressure in cold unit, bar icbP , Particle distribution functions of particles in the core, 1/μm iincsP ,, Particle distribution functions of input solids flow of the core, 1/μm ioutcsP ,, Particle distribution functions of output solids flow of the core, 1/μm dP Particle distribution function of solids flow returned into dense zone from dilute transport zone, 1/μm hP Pressure in hot unit, bar fueloutP , Particle distribution function of solids outflow from fuel reactor, 1/μm airoutP , Particle distribution function of the solids outflow from air reactor, 1/μm tP Particle distribution function of solids flow entrained into the dilute transport zone from dense zone, 1/μm 3DP Pressure at the measurement point D3 2BP Pressure at the measurement point B2 Q Volumetric gas flow rate, m3/s oxQ Reaction heat release rate because of exothermic reaction in air reactor, kJ/s reQ Reaction heat absorption rate because of endothermic reaction in fuel reactor, kJ/s airremoveQ , Heat removed from the air reactor, kJ/s xxvi fuelremoveQ , Heat removed from the fuel reactor, kJ/s r Grain radius in a single oxygen carrier, m cr Radius of core region in air reactor, m pr Core radius of a single oxygen carrier, m R Radius of oxygen carrier, μm 0R Oxygen transport capacity of oxygen carrier, kg/kg 0Re Reynolds number based on orifice diameter (ρgU0D0/μg), dimensionless dRe Reynolds number based on particle diameter (ρfU0dp/μf), dimensionless eRe Special Reynolds number, (10 6 D0)/15, dimensionless universalR Universal gas constant, J/mol.K S Cross-sectional area of column, m2 igS , Molar flow rates of oxygen transferred from core to annulus, mol/s isS , Mass flow rates of solid transferred from core to annulus, kg/s t Time, s T Temperature for reaction, K 1T Temperature in air reactor, K 2T Temperature in fuel reactor, K hT Temperature in hot unit, K mT Centring temperature in single oxygen carrier, K sinT Sinter temperature of particle, oC xxvii 1AU Aeration velocity in loop-seal 1, m/s 2AU Aeration velocity in loop-seal 2, m/s 0U Superficial gas velocity, m/s cU 0 Superficial gas velocity in cold unit, m/s hU 0 Superficial gas velocity in hot unit, m/s 0aU Superficial gas velocity in air reactor, m/s denseabsU , Absolute rise velocity of bubble in dense zone, m/s fuelabsU , Absolute rise velocity of bubble in fuel reactor, m/s 0fU Superficial gas velocity in fuel reactor, m/s mfU Minimum fluidization velocity, m/s airmfU , Minimum fluidization velocity in air reactor, m/s mfcU Minimum fluidization velocity in cold unit, m/s fuelmfU , Minimum fluidization velocity in fuel reactor, m/s mfhU Minimum fluidization velocity in hot unit, m/s mstableU Minimum superficial gas velocity in air reactor for stable particle transportation, m/s orU Gas velocity through the throat of the orifice plate, m/s saU Solid velocity in annulus region, m/s gaU Gas velocity in annulus region, m/s gcU Gas velocity in core region, m/s scU Solid velocity in core region, m/s xxviii tU Terminal velocity, m/s V Total volume of a group of particles, m3 gV Volumetric gas flow rate in air actor, m 3/s ptV Electrical Voltage from pressure transducer coefficient, V W Mass of particles in a controlled volume, kg iabW , Mass of the particles in the annulus, kg icbW , Mass of the particles in the core, kg densesW , Inventory of the bed materials in dense zone of air reactor, kg fuelsW , Total mass of the particles in fuel reactor, kg X Oxidation degree of oxygen carrier, dimensionless pX Solid conversion of oxygen carrier, dimensionless fuelinX , Average oxidation degree of input particles for fuel reactor, % fueloutX , Average oxidation degree of output particles for fuel reactor, % airinX , Average oxidation degree of input particles for air reactor, % fueloutX , Average oxidation degree of output particles for air reactor, % Y Helium concentration in the gas mixture (molar fraction), % Z Height, m P Pressure drop across the orifice, Pa P Pressure drop between two ports along air reactor, Pa 1T Absolute temperature different between the new T1 calculated from last iteration and the old T1 calculated from the former iteration, K xxix 2T Absolute temperature different between the new T2 calculated from last iteration and the old T2 calculated from the former iteration, K Z Vertical separation between two ports, m Greek Letters  Coefficient for power-law relation, dimensionless  Coefficient for power-law relation, dimensionless 0 Diameter ratio of throat-to-pipe, dimensionless T Rate of temperature increase, K/min  Reaction rate of a single oxygen carrier shown by dtdX / , 1/s 4CH Reduction rate of a single oxygen carrier, 1/s 2O Oxidation rate of a single oxygen carrier, 1/s dense Fraction of bed occupied by bubbles in dense zone, 1/s fuel Fraction of bed occupied by bubbles in fuel reactor, 1/s a Voidage in the annulus region, dimensionless c Voidage in the core region, dimensionless dense Radial averaged voidage in dense zone, dimensionless dilute Radial average voidage in dilute transport zone airmf , Voidage at minimum fluidization in air reactor, dimensionless fuelmf , Voidage at minimum fluidization in fuel reactor, dimensionless  Cross-sectional average voidage, dimensionless s Cross-sectional average solids hold-up, dimensionless xxx f Viscosity of gas, Pa.s fc Viscosity of gas in cold unit, Pa.s fh Viscosity of gas in hot unit, Pa.s λ Excess air coefficient, dimensionless 0 Apparent density of oxygen carrier with oxidation degree of 0 %, kg/ m 3 100 Apparent density of oxygen carrier with oxidation degree of 100 %, kg/ m3 air Density of air, kg/ m 3 B Molar density of the solid reactant B, mol/m3 f Gas density, kg/ m 3 fc Gas density in cold unit, kg/ m 3 fh Gas density in hot unit, kg/ m3 0fh Density of gas in hot unit at standard state, kg/ m 3 helium Density of helium, kg/ m3 m Molar density of the reacting material, mol/m 3 p Density of the particle, kg/ m 3 s Density of the solid, kg/ m3 sc Solids density in cold unit, kg/ m 3 sh Solids density in hot unit, kg/ m3 X Apparent density of oxygen carrier with oxidation degree X, kg/ m3 denseb, Average density of the particles in dense zone of air reactor, kg/ m 3 xxxi fuelb, Average density of the particles in fuel rector, kg/ m 3 iab, Average density of the particles in the annulus, kg/ m 3 iinas ,, Average density of particles in input solids flow of the annulus, kg/ m 3 ioutas ,, Average density of particles in output solids flow of the annulus, kg/ m 3 icb, Average density of the particles in the core, kg/ m 3 iincs ,, Average density of particles in the input solids flow of the core, kg/ m 3 ioutcs ,, Average density of particles in the output solids flow of the core, kg/ m 3  Sphericity of particles, dimensionless Subscripts c)( Cold unit h)( Hot unit i The ith discrete element in dilute transport zone 1i The (i+1)th discrete element in dilute transport zone top The top discrete element in dilute transport zone bottom The bottom discrete element in dilute transport zone xxxii ACKNOWLEDGEMENTS I would like to express my sincere gratitude to Dr. Naoko Ellis and Dr. C. Jim Lim for their distinguished supervision, patience, encouragement and financial support for the completion of this work. Thanks to Dr. Norman Epstein, Dr. Chad Bennington and Dr. Ho-jung Ryu for their invaluable advice and assistance and for being my committee members. I especially feel regret over Dr. Chad Bennington’s untimely passing and would like to condole with his family on the devastating loss. I am also thankful to Dr. Zhiwei Cheng, Dr. Jianjun Dai, Dr. Zhiguo Wang, and Dr. Tumuluru Jaya Shankar for their help in the experiments. Thanks are also due to the staff of the Department of Chemical & Biological Engineering Workshop and Store for all kinds of support. I also wish to acknowledge the financial support from the Ministry of Commerce, Industry and Energy (MOCIE) through the Electric Power Industry Technology Evaluation & Planning Center (ETEP), Korea, and through the collaboration with the Korea Institute of Energy Research, and from the Natural Science and Engineering Research Council of Canada. Finally, I am grateful to my wife and parents, for their understanding, patience and encouragement. 1CHAPTER 1 INTRODUCTION 1.1 Introduction Concerns over the global climate change have led to a need for new systems that produce electricity from fossil fuels with less CO2 emitted to the atmosphere. The fundamental problem with the current CO2 separation systems is the need to separate dilute CO2 from combustion flue gas stream, which is an energy intensive process that can reduce plant efficiency by 9-37%, and CO2 capture costs for projects reported to date can exceed $70 per tonne of CO2 [Herzog et al. 1997].As a promising way of integrating combustion, CO2 separation and pollution control with high efficiency and low cost, chemical looping combustion (CLC) has gained attention in recent years. A conventional combustion system normally takes place in a single reactor. The fuel is oxidized by air or enriched-O2 producing CO2 and H2O, while the released heat is used to generate power. In contrast with the conventional combustion, the CLC is comprised of two reactors: a fuel reactor and an air reactor, as shown schematically in Figure 1.1. In the fuel reactor, fuel is oxidized by the solid oxygen carrier, i.e., metal oxide, while the metal oxide is reduced to a metal or a metal oxide with a lower oxidation number. The reaction between fuel and oxygen carrier is endothermic. In the air reactor, the oxygen carrier is re-oxidized by air while emitting heat and producing a high temperature stream of gases. The heat from these two reactors can be used to drive a gas turbine to generate electricity. The circulation of solids between the two reactors in supplying of oxygen to the fuel could be effectively maintained by two interconnecting fluidized beds. The net heat of reaction over the two reactors is equivalent to that of the conventional combustion reaction of fuel [Mattisson et al. 2001a]. The process of CLC will produce an exhaust stream that contains mostly H2O and CO2 from the fuel reactor. After the water is removed by condensation, it is left with a highly concentrated, high pressure CO2 stream that can be inexpensively sequestered. Furthermore, CLC can be applied to such processes as hydrogen production and fuel cells [Ishida et al. 2001, Mattisson et al. 2001a, Brandvoll et al. 2004, Rydén et al. 2004]. In order to implement the CLC technique in practice, it is necessary to know the design and operation of the reactor, and the characteristics of the oxygen carrier. Although xonsiderable research on the CLC technology has been reported, most studies focus on the oxygen carrier development, and those related to reactor system operation and hydrodynamic properties are scarce. 2Knowledge of hydrodynamics and operating conditions is of fundamental importance for scale-up, reactor design, and process control of this kind of new reactor system. Air Air Reactor N2, O2 Fuel Steam ,CO2 Mor MxOy-Z MxOy Fuel Reactor Figure 1.1 Principle of chemical looping combustion (M, MxOy and MxOy-z represent metal, metal oxide and reduced metal oxide) In this study, a CLC reactor was designed and built to conduct the hydrodynamic studies. Furthermore, based on the hydrodynamic studies coupled with reaction kinetics of oxygen carrier from the literature, a comprehensive model for the reactor was developed to identify and investigate the critical components and process of the CLC. 1.2 Oxygen carrier development Most reported work on CLC has been related to oxygen carrier development. The metal oxide particle, used as an oxygen carrier in CLC, must meet the following requirements: sufficient rates of reduction and oxidation; high oxygen transport capacity; stability during repeated redox cycles at high temperature; and enough strength to limit particle breakage, attrition and agglomeration. It is also important that the oxygen carriers are inexpensive and environmentally benign [Lyngfelt et al. 2001, Villa et al. 2003, Adánez et al. 2004a]. As listed in Table A.1 in Appendix A, a number of metal oxides have been investigated for this purpose. These studies focus on the Ni-based, Cu-based, Fe-based, Mn-based and Co-based oxygen carriers with experiments being conducted in a thermogravimetric analyzer (TGA), fixed bed and fluidized bed in the research lab. 1.3 Reaction kinetic model The reaction rate of an oxygen carrier is one of the most important properties in the CLC system as it dictates the solids circulation flux between fuel reactor and air reactor, and the amount of bed materials required in the system [Mattisson et al. 2001a]. The reaction rate varies widely 3depending on the particle size, temperature, the conversion range, the composition and type of the metal oxide and binder, and gas composition [Ishida et al. 1996a, Jin et al. 1998, Jin et al.1999, de Diego et al. 2004]. In order to describe the reaction rate of single particle, the degree of oxidation, X , is introduced. redfoxf redfactual MM MM X ,, ,    (1.1) where actualM is the actual mass of the carrier in its partially oxidized state, redfM , is the mass of fully-reduced oxygen carrier, and oxfM , is the mass of fully-oxidized oxygen carrier, then the solid conversion X actually represents the oxidation degree of one solid oxygen carrier and the reaction rate could be presented by dt dX . Thus, reactivity data could be obtained in TGA tests from the weight variations during the reduction and oxidation as a function of time. On the other hand, researchers usually use solid conversion, pX , to describe the status of particles. According to their definitions of pX , the relationship between X and pX can be shown to be: For oxidation reaction pXX  (1.2) For reduction reaction pXX 1 (1.3) Generally four types of kinetic models have been applied to describe the reaction of single oxygen carrier in CLC based on experimental data from different experimental conditions, i.e., unreacted shrinking core model, power-law relation, nucleation and nuclei growth model, and modified volumetric model, which are described further. 1.3.1 Unreacted shrinking core model For a spherical particle, the unreacted shrinking core model takes into account the reaction resistances by the gas-film diffusion, the ash-layer diffusion and the chemical reaction [Levenspiel 1972] while the surface of the unreacted core is shrinking from outer layer to the inner layer of the particles. Thus, with assuming that a solid-gas reaction shown by Eq. 1.4 is first order with respect to the concentration of reactant gas, )()()()( sdDgcCsbBgA  (1.4) the unreacted-core shrinking model gives the following change of the core radius pr : 4sep ppp gp p BAp kDR rrR kR r bC dt dr 1)( / 2 2    (1.5) where the solid conversion pX is given by 3 1      p p p R rX (1.6) Ishida et al. (1996a) presented the mass transfer coefficient of gaseous reactant gk , effective diffusion coefficient eD and chemical reaction rate constant sk for oxidation and reduction through fitting the experimental curves with oxygen carrier of NiO/YSZ. The oxidation and reduction were conducted only at 1273 K and 873 K, respectively, and the reduction gas was Hydrogen. Ryu et al. (2001) reported that the oxidation reaction was controlled by product layer diffusion and the reduction reaction was chemical reaction controlled with Air and CH4 as reacting gases and the oxygen carrier was NiO/bentonite. Thus, the Eqs.1.7 and 1.8 can be used to represent the oxidation and reduction, respectively, with the assumptions of the spherical particle and the first order reaction.  )1(2)1(31 6 3/2 0 3 0 pp Ae pB XX CbD RC t  (1.7)  3/1 0 0 )1(1 p As pB X Cbk RC t  (1.8) where the effective diffusion coefficient eD and chemical reaction rate constant sk could be fitted to the Arrhenius equations as function of temperature. However, the oxidation curve showed that the reaction rate of oxidation was very small for their specific oxygen carrier of NiO/bentonite. A set of new equations, i.e., Eqs. 1.9 and 1.10, based on the unreacted shirking core model was developed by Garca-Labiano et al. (2006) and Abad et al. (2007b) considering H2 and CO as the gas reactants. The reaction rates for both of oxidation and reduction reaction were assumed under chemical reaction rate control. 3/1)1(1 pX t  where )( neqns m CCbk r    (1.9) The n is the reaction order, and it is assumed at 0.4 and 1.0 for the reduction and oxidation, respectively. The kinetic constant could be represented as Arrhenius-type function with temperature: TRE s universalekk /0  (1.10) 5Furthermore, the kinetics of reduction and oxidation of the NiO/MgAl2O4 and Mn-based oxygen carrier particles with spherical grain geometry were investigated by Zafar et al. (2007a,b); while CH4 and O2 were chosen to be the gas reactants for the reduction and oxidation reactions, respectively. Similarly, the following equations were used to describe the reaction kinetics: 3/1)1(1 pX t   where n s m Cbk r (1.11) In the work of García-Labiano et al. (2006) and Zafar et al. (2007a,b), it was assumed that the particles were composed of spherical grains. Thus, for the spherical shape of particles, the grain radius,r , was selected to show the effect of geometry size of particle on the reaction rate. Sedor et al. (2008a) provided a simplified equation based on the shrinking-core model for the only reduction reaction of Nickel-Alumina oxygen carrier with CH4. The rate of chemical reaction was considered as the limiting resistance, while the effect of particle size on the reduction reaction rate was neglected. Thus, the following Eq. 1.12 did not have any term related to the effect of particle size. 4 3/2 0 )1))( 11 (exp( CHp muniversal p CX TTR E k dt dX  (1.12) 1.3.2 Power-law relation model Sedor et al. (2008a) also tried to use power-law relation to describe the reduction rate of oxygen carrier particle as shown in Eq. 1.13:  )()1))(11(exp( 40 CHp muniversial p CX TTR Ek dt dX  (1.13) According to the equation, the effect of particle size was not considered in a simplified model. 1.3.3 Nucleation and nuclei growth model Nucleation and nuclei growth model [Hossain et al. 2007b, Sedor et al. 2008a] was applied to describe the kinetics of oxygen carrier based on the reaction-rate controlling Avrami-Erofeev model [Malecka et al. 2002, Malecka et al. 2004]. It was believed that the reduction and oxidation of metal oxide proceed through nucleation and crystal growth, thus, a reaction rate equation can be expressed as following:   avav nnpp mTunivesal av p XX TtTR Ekn dt dX /)1( 0 0 )1ln()1))( 11(exp(   (1.14) 6where nav is the Avrami exponent indicative of the reaction mechanism and crystal growth dimension and different values of the nav were reported by Hossain et al. (2007b). It was noted that there is no term of gas reactants concentration, because the gas reactants concentration was assumed constant during the process of reaction experiments. Further application of nucleation and nuclei growth model for reduction of Nickel-Alumina oxygen carrier with CH4 were carried out by Sedor et al. (2008a). Eq. 1.15 could be used to express the reduction rate considering the effect of gas reactant concentration.   )()1ln()1))(11(exp( 4 /)1( 0 CH nn pp muniversal av p CXX TTR E kn dt dX avav (1.15) Similar with the equation used with Power-law relation model, the geometry size of oxygen carrier particle show no influence on the reaction rate since the particle size in their experimental studies was constant. 1.3.4 Modified volumetric model Son et al. (2006) used modified volumetric model to describe the reaction rate of NiO- Fe2O3/bentomite particles with fuel gas of CH4. Thus, the reaction rate equations could be shown to be: )1)(exp( , X TR E a dt dX universal reda red    (1.16) 3/2, )1)(exp( X TR E a dt dX universal oxa ox   (1.17) The pre-exponential factor and activation energy could be correlated with only one parameter, i.e., percent value of the Fe2O3 ratio when the sum of NiO and Fe2O3 was 100%. Detailed description was present in Son et al. (2006). It was found that there was no term to represent the effect of particle size and gas reactants concentration. 1.4 Reactor design The concept of interconnected fluidized bed (IFB) is believed to be the most promising set-up for successful long-term operation of CLC system [Lyngfelt et al. 2001, Brandvoll et al. 2004]. Several attempts [Harvey et al.1994, Grönkvist 1995, Lyngfelt et al. (2001, 2004), Johansson et al. (2002, 2006a,b), Ryu et al. (2003a, 2008a,b), Kronberger et al. (2004, 2005), Adánez et al. (2004b, 2006a, 2009a,b), Abad et al. (2006a, 2007), Son et al. 2006, Noorman et al. 2007, de Diego et al. 2007, Kolbitsch et al. (2008,2009a,b), Linderholm et al. 2008, Pröll et al. 2009a, García-Labiano et 7al. 2009] have been made to design the interconnected fluidized bed (IFB) for the CLC reactor. Detailed discussion on the CLC reactor design is presented in Chapter 2. 1.5 Hydrodynamic study Hydrodynamic study is one of the key information sources for designing a fluidized bed reactor. It could provide important information for the selection of the flow regime, system configuration and operation of reactors. Some limited works on hydrodynamic studies have been done for the operation condition and gas leakage for the CLC reactor. 1.5.1 Operation condition and pressure profile For a packed bed of particles, if the fluidization velocity is below a critical gas velocity known as the minimum fluidization velocity, Umf, the particles remain fixed. At the minimum fluidization velocity, the particles are supported by the gas. Thus, the aeration velocity for loop-seals to control the solids flux can be set through the ratio between this velocity and Umf [Kronberger et al. 2005]. As summarized in Table A.2 in Appendix A, the pressure profiles and the relationships between superficial gas velocities, total solids inventory and net solids circulation rate have been reported [Johansson et al. 2002, Kronberger et al. 2004, Kronberger et al. 2005, Ryu et al. 2008a,b, Pröll et al. 2009a]. However, most reported hydrodynamic studies were conducted in a narrow range of operating conditions, and their cold-flow models were relatively small-scale fluidized beds. Thus, the detailed operating mapping of the CLC reactor is required to be further investigated. 1.5.2 Gas leakage The pressure in the two reactors should be approximately equal in order to minimize gas leakage between them. Gas leakage could lead to a reduced capture of CO2 in the process and N2 entering the gas recirculation loop, thus defeating the purpose of CLC [Lyngfelt et al. 2001]. Johansson et al. (2003) conducted some experiments to measure the gas leakage in their cold- flow models and investigated the effect of the operating conditions. The gas leakage up to 20% between air reactor and fuel reactor can be observed through the experiments for a specific two- compartment CLC reactor [Kronberger et al. 2004, Abad et al. 2007a, and Johansson et al. 2006b]. Ryu et al. (2008a,b) investigated the leakage from a hole on the solid injection nozzle to bubbling fluidized bed for their two-bubbling-bed CLC reactor, and believed that the leakage could be negligible for their reactor. However, as shown in Table A.2, a relatively narrow operating condition 8range was used for the experiments of gas leakage, and their experiments were conducted at relatively lower superficial gas velocities and solids circulation rate. 1.5.3 Solids hold-up in the reactors It is known that the gas-solid contact in the reactor is very important for the conversion of solid and gases in CLC reactor. However, the relationship between the solids hold-up distributions and the operating conditions of CLC has not been previously reported. 1.6 Scale up In order to achieve hydrodynamic similarity between a hot unit and the cold unit of CLC reactors, certain scaling rules are applied. Glicksman (1984) presented a full set of scaling relationships for fluidized beds. Based on the governing equations of bubble and interstitial gas dynamics, Horio et al. (1986) developed a simpler similarity rule. Glicksman (1988) showed that the parameters of Horio et al. (1986) are equivalent to his reduced governing parameters under the condition of viscous limit. Horio et al. (1989) also developed another scaling law for circulating fluidized beds. Glicksman et al. (1993) further explored a new set of simplified scaling laws. The detailed discussion about the scale up will be presented in Chapter 6. According to the scaling laws introduced by Glicksman et al. (1993), keeping the density ratio between solids and gas, ρs/ρf, identical for both the hot unit and cold unit is very important to achieve the hydrodynamic similarity between them. However, the ρs/ρf in an atmospheric CLC hot unit is relatively big, because the density of the popular oxygen carrier, i.e., Ni-based oxygen carrier, is around 2800~4500 kg/m3 and the density of gas at high temperature and atmospheric pressure is reduced to around 0.29~0.36 kg/m3. Thus, it is very difficult to use air and particles with density of less than 10,000 kg/m3 in a cold-flow model to simulate the hydrodynamics of an atmospheric hot unit. In the current hydrodynamic investigations, some of the CLC cold-flow models are operated with air at ambient temperature and pressure. Thus the CLC cold-flow model can only be used for simulating the pressurized hot unit [Johansson et al. 2002, 2003]. Owing to the limitations of the gas (helium) supply, the bench-scale CLC cold-flow model investigation using a mixture of helium and nitrogen as fluidizing gas was restricted to low gas velocity range (0.26 m/s~1.5 m/s) [Kronberger et al. 2004, Kronberger et al. 2005]. Furthermore, Pröll et al. (2009a) used heavy particles and air at ambient temperature and pressure to simulate the hydrodynamics of an atmospheric reactor, but ρs/ρf requirements were still not satisfied. 91.7 Mathematical model for CLC reactor Very few published papers have dealt with modeling of a CLC reactor as a whole. A one- dimensional adiabatic packed bed reactor model was developed by Noorman et al. (2007) based on an idea of packed bed CLC reactor where the solids are stationary and are alternately exposed to reducing and oxidizing conditions via periodic switching of the gas feed streams. Berguerand et al. (2008) suggested two simplified reactor model concepts for the solids flow in the fuel reactor, i.e., the continuous stirred-tank reactor and the plug flow reactor where the fuel and the oxygen carrier particles are in co-current flow with no backmixing. Jung et al. (2008) adapted a multiphase CFD-based model only to the fuel reactor. Kolbitschl et al. (2009c) introduced a model to simulate their two-riser reactor, where each riser was modeled with the assumption that the reacting gas was only in contact with a defined fraction of well mixed solids. Hydrodynamic profile of the reactor was described only by prescribed solids concentration profile along the reactor height. Energy and solids were balanced globally across the whole reactor and plug flow was assumed for the gases. 1.8 Summary and research objectives Based on the above review, the research status on CLC reactor with gas fuel can be stated as following:  Most studies of CLC focus on the oxygen carrier development. Different oxygen carriers (metal oxide) were tested in the TGA, fixed bed, fluidized bed and the particle properties including reactivity were measured.  Based on the reactivity data, some reaction kinetic models have been developed for the oxygen carrier particles.  Some approaches have been presented for the reactor design, and the interconnected fluidized beds are believed to have the advantages over alternative designs. However, the solid conversion in the high-velocity air reactor remains to be the problem.  Very limited work concerning the hydrodynamic studies is available for CLC. Most reported studies investigate the hydrodynamics in narrow range of operating conditions, and their cold-flow models are relatively small-scale fluidized beds. The cold-flow model should be carried out over a wider range of operating conditions.  Because of the difficulties in satisfying the scaling law, there are few attempts with very small bench-scale reactor and lower gas velocities to scale the cold-flow model to the atmospheric reactor. 10  Very few published papers have dealt with modeling of CLC reactor, the effect of particles properties and hydrodynamics on the reactor performance, and the linkage between air reactor and fuel reactor need to be investigated. Thus, the main objectives of the present study are:  to design and build a cold-flow CLC reactor;  to provide detailed mapping of the operating conditions of the CLC reactor in terms of control parameters such as superficial gas velocity, aeration velocities in loop-seals and investigate the effect of these parameters on the solids circulation flux, pressure profile, solids hold-up in air reactor;  to investigate gas leakage at a wider range of operating condition in detail and provide the suggestion and improvement for the reactor in order to minimize the gas leakage;  to use the gas with lower density, e.g. helium, instead of air as fluidization gas in the pilot-scale cold-flow model at UBC to simulate the hydrodynamics in the atmospheric pilot-scale hot unit at higher gas velocities;  to use gas mixture of helium and air with different helium concentration to simulate the different density ratio of particle to gas, ρs/ρf and analyze the effect of the density ratio on the hydrodynamics in CLC cold-flow model;  to scale up the cold-flow model running at different density ratios between particles and gases to atmospheric and pressurized hot units;  to develop a mathematical model for the CLC reactor considering the hydrodynamics and particles properties, and  to systematically analyze the influence of critical variables on the performance of the CLC reactor while gaining further understanding of the reactor design and operation. 11 CHAPTER 2 DESIGN OF A COLD-FLOW CLC REACTOR 2.1 Requirements for the CLC reactor The configuration of a CLC reactor requires particles to circulate between the air and fuel reactors while transferring oxygen for fuel combustion. Therefore these two reactors must be integrated in order for the oxygen carrier to be circulated continuously with successful separation between gases and solids. Based on the above requirements for the configuration of a CLC reactor and the advantages for the gas-solid reaction of fluidized beds, the concept of interconnected fluidized bed (IFB) is believed to be the most promising set-up for successful long-term operation of the CLC reactor [Lyngfelt et al. 2001, Brandvoll et al. 2004]. Following criteria are chosen for the reactor system design:  high conversion of fuel and oxygen;  sufficient conversion of solids in the reactors;  proper solids circulation flux for providing the sufficient oxygen for the complete conversion of fuel in the fuel reactor and supplying the needed energy for temperature balance between air reactor and fuel reactor;  limited gas leakage between air reactor and fuel reactor and  stable long-term operation and good contact between solid and gas. 2.2 Literature review for CLC reactor design for gaseous fuel A few attempts have been made to design the CLC reactor. 2.2.1 Four-compartment IFB In earlier years, a four-compartment interconnected fluidized bed (IFB) was proposed and investigated by various researchers. As shown in Figure 2.1, there are four sections: two of them were used as reactors; while the other two sections were used as downcomers. The orifice pressure drop forces the particles to transfer from one section to another [Kuramoto et al.1985, Fox et al.1989, Snip et al.1996, Abellon et al.1997, Snieders et al.1999]. However, in this system the gas leakage is proportional to the area of the orifice, thus limiting the solids transfer rate between the two reactors. 12 Figure 2.1 Schematic diagram of four-compartment IFB reactor (adapted from Snieders et al. 1999) 2.2.2 IFB with alternative valve As shown in the Figure 2.2, two isothermal fluidized bed reactors with alternating valves had been suggested for CLC [Harvey and Richter, 1994]. In this design, the valve allows the operation of reactors to be switched from oxidation to reduction and vice versa without transporting the solids. Figure 2.2 Schematic diagram of IFB reactor with alternating valve 13 2.2.3 Packed bed with alternative valve As shown in Figure 2.3, Noorman et al. (2007) provided a concept of CLC reactor of periodically operated packed bed reactor. In this reactor, the solids are stationary and are alternately exposed to reducing gases and oxidizing gas via periodic switching of the gas feed streams. In fact, if two of the single packed beds were operated in parallel, the reactor shown in Figure 2.3 could be considered as the same as the reactor shown in Figure 2.2. The main advantage of both of them is that particle transfer between reactors can be avoided intrinsically. Figure 2.3 Schematic diagram of periodically operated packed bed reactor (adapted from Noorman et al. 2007) However, there is a need for additional complicated systems to switch between reducing gases and oxidation gases. At the same time, in order to achieve the continuous high temperature gas stream supply to the downstream gas turbine, another complicated controlling system has to be installed to switch the reactors between reducing and oxidizing conditions. Furthermore, it is necessary to provide the inert gases for separation between switching the gases, and a large temperature drop will result in the reactors when the oxidation condition is switched to the reducing condition. 14 2.2.4 IFB combining fluidized bed and moving bed Grönkvist (1995) proposed a reactor system as shown in Figure 2.4. In this design, oxidation reaction takes place in a fluidized bed reactor while a counter-current moving bed is considered for the reduction reaction. However, in order to let the fuel reactor work as one counter-current moving bed, the superficial gas velocity must be limited. Thus, this system is not suitable for high capacity operation. Figure 2.4 Schematic diagram of IFB combining fluidized bed and moving bed 2.2.5 Two-compartment IFB Kronberger et al. (2004), Johansson et al. (2006a, b), Abad et al. (2006a, 2007a) and Rydén et al. (2008a) designed a two-compartment interconnected fluidized bed reactor for CLC, as shown in Figure 2.5. In this design, two adjacent fluidized beds are divided by a vertical wall with two orifices. On one side, the air reactor has a higher velocity causing particles to be thrown upwards, while some of them fall into the downcomer that has a return orifice leading to the fuel reactor. At the bottom of fuel reactor, the particles can flow through the slot and go back into the air reactor. The biggest problem of this reactor system is the difficulty in minimizing the gas leakage while increasing the solids circulation flux. In order to achieve a higher solids circulation flux, the pressure differences through the two orifices and the cross-sectional areas of the orifices should be 15 higher, but such a condition results in higher gas leakage which makes the main advantage of CLC difficult to achieve. Figure 2.5 Schematic diagram of two-compartment IFB (adapted from Rydén et al. 2008a) 2.2.6 IFB combining two bubbling beds Adánez et al. (2006a) and de Diego et al. (2007) proposed an IFB CLC reactor by combining two bubbling beds as shown in Figure 2.6. Because the oxidation reaction is not sufficiently fast to be completed in a fast fluidized bed riser, the air reactor is designed as a bubbling fluidized bed to increase the contact time between gas and solids. This results in requiring an additional riser for particle conveying. Figure 2.7 shows a CLC reactor combining two bubbling beds which is an improvement of the reactor shown in Figure 2.6. The difference is the position of the riser for solid conveying. As shown in Figure 2.7, the riser is installed on top of the air reactor and secondary air is used as pneumatic transport gas. 16 Figure 2.6 Schematic diagram of IFB combining two bubbling beds (adapted from Adánez et al. 2006a) Figure 2.7 Schematic diagram of IFB combining two bubbling beds (adapted from Adánez et al. 2009a) 17 Figure 2.8 shows another type of CLC reactor combining two bubbling beds. An annular shape reactor was designed to optimize heat transfer, where the inner column is an air reactor and the annular column is an fuel reactor. Two risers are required to transport the particles between the air and fuel reactors. Figure 2.8 Schematic diagram of IFB combining two bubbling beds (adapted from Son et al. 2006) As shown in Figure 2.9, a compacted IFB CLC reactor combining two bubbling beds was developed by Ryu et al. (2008a,b). Since the superficial gas velocity in a bubbling bed is not sufficient to achieve solid conveying, additional solid injection nozzles and short risers on top of the reactors are required for solid circulation. 18 Figure 2.9 Schematic diagram of IFB combining two bubbling beds (adapted from Ryu et al. 2008a) However, it is noted that the volumetric gas flow of air should be kept 10 times larger than that of the fuel in order to provide sufficient oxygen for complete conversion of fuel, while the diameter of the air reactor should be almost 3 times as large as that of the fuel reactor. Thus the IFB with combining two bubbling beds is not suitable for CLC reactor with high power capacity. 2.2.7 IFB combining riser and turbulent bed Kolbitsch et al. (2009a,b) proposed a CLC reactor shown in Figure 2.10, where the air reactor and fuel reactor were designed as fast-fluidization bed and turbulent bed, respectively. In order to avoid the bypassing of unconverted fuel through the bubbling fluidized bed fuel reactor, a turbulent fluidized bed is applied as a fuel reactor. However, one of the disadvantages is the shorter residence time of gas in a fuel reactor. Furthermore, the oxidization conversion of solids achieved in one single riser running at fast fluidization is may be insufficient for the complete conversion of fuel. 19 Figure 2.10 Schematic diagram of IFB combining riser and turbulent bed (adapted from Kolbitsch et al. 2009a) 2.2.8 IFB combining riser and bubbling bed As shown in Figure 2.11, Lyngfelt et al. (2001) proposed the first design of IFB combining riser and bubbling bed. In their studies, an atmospheric circulating system comprised of two connected fluidized beds, a high-velocity riser (air reactor) and a low-velocity bubbling fluidized bed (fuel reactor), was developed. Gas and solids are separated in a cyclone and particles fall down to the fuel reactor. After finishing the reduction reaction, the particles flow into the riser again. Two loop-seals are added to separate the gases in the two reactors and control the solids circulation rate. Similar reactor systems were investigated by other researchers [Johansson et al. 2002, Lyngfelt et al. 2004, Linderholm 2008]. 20 Figure 2.11 Schematic diagram of IFB combining riser and bubbling bed (adapted from Lyngfelt et al. 2004) The reactor shown in Figure 2.12 is identical to that shown in Figure 2.11 without the alternative particle separator. The “hat”separator is intended to decrease the exit effects on the solids flow and lower particle back-flow in the air reactor. Furthermore, the configuration also reduces the pressure drop across the particle separator. The disadvantage of the separator design is the reduction in particle separation efficiency. As shown in Figure 2.12, the gas exit of the fuel reactor is completely open to the atmosphere, resulting in particle loss due to entrainment. Thus, in the subsequent design of the IFB combining a riser and a bubbling bed, an additional cyclone has been added for the gas exit of fuel reactor [Ryu et al. 2002]. Notably, this kind of reactor system is similar to the current commercial circulating fluidized bed. Thus many existing industrial techniques could be retrofitted in this new reactor, with potential to be integrated with the current power generation system. However, the main disadvantage of the reactor is that the reduced residence time of particles in the air reactor will result in less oxygen conversion if the oxidation rate of the oxygen carrier is limited. 21 Figure 2.12 Schematic diagram of IFB combining riser and bubbling bed (adapted from Kronberger et al. 2005) 2.3 Flow regime of reactor As discussed in Section 2.2, different designs of the CLC reactor have advantages and disadvantages. It is found that the most influential factor for the CLC reactor design is the reaction rate of the oxygen carrier. When the reaction is slow, a longer residence time and better contact between gas and solids are favourable. This requires taller reactor, higher solids inventory, and lower superficial gas velocity. A reactor running at lower superficial gas velocities is not suitable for a CLC reactor with high fuel capacity. When the reaction rate is fast enough, a transport bed can be used for the reaction and solids conveying simultaneously [Ryu et al. 2008a], since the superficial gas velocity can remain high with less residence time of particles and gas. 22 Thus, the flow regimes for a CLC reactor are dependent on the reaction rate of the oxygen carrier, which influences the solids inventory, supplied power capacity, and reactor geometry. Table 2.1 shows the features of fluidization beds in different flow regimes. Bubbling and turbulent fluidized beds can be used for the reactor with lower volumetric flow rate of input gas while the reaction rate is slow. However, this is compensated by the high solids inventory. On the other hand, the gas bypass through the bubble phase leading to significant slip of unreacted gas in a bubbling bed must be dealt with. To solve the problem, a distributor with many small-diameter-orifices is preferred to produce a number of bubbles with small size, or some internal baffles could be installed to improve the contact between gas and solids. However, bubbling and turbulent fluidized beds need additional riser to transport particles. Fast fluidized beds can be used for solids conveying in the reactor with higher gas volumetric flow rate. However, the reaction rate in fast fluidized beds is required to be fast because of the short residence time of solids and gas. Table 2.1 Features of fluidized beds in different flow regimes Flow regime Superficial gas velocity Residence time of gas and solids Contact between solids and gas Height of reactor Reactor diameter Solids inventory Solid conveying Bubbling bed Low Long - Low Bigger More - Turbulent bed Medium Short + Medium Big More - Fast fluidization High Shorter + High Small Less + 2.4 Design of cold-flow CLC reactor at UBC The high volumetric gas flow in the air reactor (approximately 10 times larger than that of natural gas in the fuel reactor) provides the driving force for the circulation of particles between the two reactors [Lyngfelt et al. 2001]. Thus, in order to keep a reasonable size of the reactor system with higher gas volumetric flow rate and fuel capacity, the fluidization regimes in the air reactor and fuel reactor are fast fluidization and bubbling fluidization, respectively. As discussed above, the interconnected fluidized bed (IFB) is believed to be the most promising set-up for successful long-term operation of the CLC reactor. Some interconnected fluidized beds (IFB) have been discussed and compared with each other, and the disadvantages and advantages of fluidized bed running in different flow regimes have also been discussed. The IFB combining riser and bubbling bed (Figure 2.11) seems to be the best candidate for the CLC reactor. 23 However, in this kind of reactor, the lower residence time of particles in the riser limits the oxidation conversion of solids in one cycle. Thus, the riser has to be operated under higher temperature in order to achieve higher solids conversion. In order to overcome the problem, for the current cold-flow model CLC reactor used for hydrodynamic research in this study (presented in Figure 2.13), a bypass line (primary downcomer) is added to transport some particles out of the air reactor to return to the riser directly without reduction. In this manner, a portion of the particles can be continuously oxidized in the air reactor increasing the average residence time of particles. Thus, higher solid conversion together with higher gas velocity in the riser and global solids circulation rate, are achieved. Compared with the reactor without a bypass line, the solids flux between the air and fuel reactors will decrease. However, the residence time of the oxygen carrier in the fuel reactor will increase and compensate for the reduced solids flux between the air and fuel reactors. On the other hand, according to the experiments (shown in Chapter 7), the total solids flux flowing through the air reactor will be far more than in the conventional reactor without a bypass line. This means the performance of the air reactor will be significantly improved. Figure 2.13 shows the schematic diagram of cold-flow CLC reactor at the University of British Columbia, where the air reactor is a riser of ID 0.1 m and height 5.60 m; and the fuel reactor is a bubbling fluidized bed of ID 0.29 m and bed height 1.0 m, with an extended freeboard of 0.5 m high. This cold-flow model can simulate the hydrodynamics of an atmospheric chemical looping combustion reactor with fuel capacity of 80 ~ 105 kW. An internal cyclone is installed at the gas exit of the fuel reactor to collect the particles because of entrainment. Furthermore, in order to decrease the gas leakage from the fuel reactor to the cyclones, the bottom end of the lower downcomer is sealed, while a hole (referred to 10 in Figure 2.13) of 0.1 m in diameter is opened on the side. Thus, the lower downcomer in the fuel reactor is working as an internal loop-seal. As shown in Figure 2.13, the operating process of this reactor is described as follows:  in the air reactor, particles are oxidized by air and entrained out of the riser;  cyclones collect the particles and these particles are divided into two streams;  the flow rates of the two streams of particles are controlled by the aeration rates in the loop- seals;  one stream of particles will be reduced by natural gas in the fuel reactor and  two loop-seals are used to limit the leakage between air reactor and fuel reactor, and control the solids circulation flux. 24 Figure 2.13 Schematic diagram of the proposed cold-flow CLC reactor 1: air reactor, 2: primary cyclone, 3: secondary cyclone, 4: dipleg, 5: upper downcomer, 6: primary downcomer, 7: internal cyclone, 8: lower downcomer, 9: fuel reactor, 10: hole opened on the lower downcomer, 11: Seal at the bottom of the lower downcomer, 12: loop-seal 1, 13: loop-seal 2. 25 CHAPTER 3 EXPERIMENTAL SET-UP 3.1 CLC reactor Based on the reactor design discussed in Section 2.4, a schematic CLC reactor with the main geometric sizes is given in Figure 3.1. The air reactor with height of 5.6 m and ID of 0.1 m is connected to the fuel reactor with height of 1.5 m and ID of 0.3 m through the loop-seal 1. The primary downcomer with 4.29 m in height and 0.1 m in ID is connected with the air reactor through loop-seal 2. Thus, as shown in Figure3.1, there are two loops in the CLC reactor, i.e., loop 1 (as shown by the right view in Figure 3.1): air reactor - cyclone - upper section of primary downcomer - upper downcomer- dowcomer 3 - fuel reactor - loop-seal 1 - air rector; and loop 2 (as shown by the front view in Figure 3.1): air reactor –cyclone - primary downcomer - loop-seal 2 - air reactor. Loop 1 is used to achieve the solids circulation between the air and fuel reactors, while loop 2 transports a portion of particles out of air reactor back to the air reactor, directly bypassing the fuel reactor. The particles flow upward in the air reactor with the fluidizing gas and most of the particles can be collected by the primary cyclone at the top of the system after they are entrained out of the air reactor, while the fine particles are collected by the secondary cyclone and returned to the primary downcomer through a small dipleg. The particles collected by the primary cyclone are divided into two streams: one stream returning back into the air reactor for continuous oxidation; and the other stream entering the fuel reactor. The solids circulation fluxes in the two loops are controlled by varying the aeration velocities in the loop-seals. Two butterfly valves are used for the measurement of solids circulation flux. The fuel reactor of ID 0.29 m and height 1.0 m is shown in Figure 3.2 in detail. The extended freeboard is 0.5 m high and the diameters of its two bases are 0.53 m and 0.29 m, respectively. A small internal cyclone is designed to collect the fine particles due to entrainment. The bottom end of the lower downcomer is sealed completely and one hole (referred to 10 in Figure 2.13) with diameter of 0.1 m is open on the side of the downcomer. Thus, as shown by the shadow section in Figure 3.2, the accumulated particles will work as an internal loop-seal to prevent gas leakage between the fuel reactor and the cyclone through the lower downcomer. The particles leaving the fuel reactor are returned into the air reactor via an overflow pipe connected to the fuel reactor at a vertical angle of 20 o. 26 Figure 3.1 Schematic diagram of the cold-flow CLC reactor at UBC 1: gas inlet of air reactor, 2: air reactor, 3: primary cyclone, 4: secondary cyclone, 5: dipleg, 6: primary downcomer, 7: loop-seal 2, 8: aeration ports of loop-seal 2, 9: Butterfly valve, 10: upper downcomer, 11: internal cyclone, 12: fuel reactor, 13: lower downcomer, 14: loop-seal 1, 15: distributor of fuel reactor, 16: aeration ports of loop-seal 1, 17: gas inlet of fuel reactor 27 Figure 3.2 Schematic diagram of the fuel reactor 1: internal cyclone, 2: lower downcomer, 3: opened hole in the lower downcomer 3.2 Particulate material Table 3.1 shows the basic properties of the spent FCC particles used in the experiments, while the particle size distribution shown in Figure 3.3 were measured through the sieving mesh. Table 3.1 Spent FCC particle properties Particles Particle density (kg/m3) Bulk density (kg/m3) Sauter mean diameter (µm) Spent FCC 1560 * 870 78 * Referred to Ellis 2003 28 0 50 100 150 200 250 300 350 400 450 0 10 20 30 40 50 60 70 80 90 100 Particle size (m) C um ul at iv e pe rc en tu nd er si ze Figure 3.3 Particle size distribution of spent FCC particles 3.3 Experimental unit at open status To investigate the performance of the CLC cold-flow model under different operating conditions with the fluidization gas of air, the experimental unit at open status as shown in Figure 3.4 was used in this project. The fluidizing gas of air from atmosphere was provided by a blower (53 URAI-Gas) and released into the atmosphere after the gas ran through the experimental system. A filter was used to collect the very fine particles at the gas outlet before the air was released into the building ventilation system. The total volumetric flow rate of air was controlled by adjusting the bypass valve. The flow rate of inlet air for the air reactor was measured by an orifice meter of inner diameter 41.4 mm and outer diameter 102.3 mm. The flow rate of inlet air for the fuel reactor was measured by an orifice meter of inner diameter 17.8 mm and outer diameter 77.9 mm. The flow rate through each reactor was controlled by a flow control valve upstream of each orifice plate. Compressed building air was used as the aeration gas for the loop-seals. The flow rates of the aeration gas were controlled by six flowmeters. Since the system was operated at room temperature with the air being supplied at atmospheric conditions, the temperature of the system was measured only at two positions as shown in Figure 3.4. The results confirmed that the temperature in the system was constant at room temperature. 29 Figure 3.4 Experimental system at open status with fluidizing gas of air 30 3.4 Experimental unit at closed status The experimental unit conducted as a closed system, shown in Figure 3.5, was used to study the CLC reactor hydrodynamics with a fluidizing gas mixture of helium and air through recycling the gas by a blower. The recycling system was established to accommodate the maximum flow rate of helium at 136.4 Nm3/hr provided by one cylinder of helium, which can last for only a few minutes. The additional helium was continuously added to maintain a constant helium concentration through the experiment as the blower system inevitably leaked the gas mixture while in operation. A Micro-GC (Varian Inc.) was used to monitor the helium concentration at the gas inlets of the reactors. As indicated by the arrows in Figure 3.5, the blower forces gas mixture to circulate in two cycles, i.e., the internal cycle of blower to bypass valve to blower, and the outer cycle of blower to safety valve to filter to orifice meter to reactor to blower. By adjusting the bypass valve in internal cycle, the ratio of gas flow rate in the internal cycle to the outer cycle is altered. Thus, the superficial gas velocity in the reactor was changed while recirculating the same fluidizing gas in a closed system. An additional small gas compressor was added to the unit to create enough pressure drop for the gas mixture to aerate the loop-seals. Both temperature and differential pressure gauges were mounted across the blower to ensure that the blower operated within the safe temperature and pressure limits. In order to prevent a potential damage to the bearings or gears, a small secondary filter was installed before the inlet of gas blower to collect the very fine particles that escaped from the primary filter. 3.5 Pressure measurement The pressure measurements were obtained by PX142 and PX162 series pressure transducers supplied by OMEGA. These pressure transducers were positioned along the columns, as shown in Figure 3.6. A 38µm wire mesh was fixed at the entrance of every measurement port to prevent solids from entering the transducers. The pressure transducers were calibrated with a water manometer. A linear correlation between the voltage signal and pressure was observed, resulting in a linear calibration equation for each pressure transducer. The electrical voltage signals received by a computer via a DAS08 A/D converter board were converted into pressure values through the calibration equations. The pressure transducers A1~A15 were used to measure the pressure difference between two points along the height of the air reactor, while others were used to measure the gauge pressures at each specific position. Table D.1 in Appendix D shows the detailed information on the positions of the pressure transducers. 31 Figure 3.5 Experimental system at closed status with fluidizing gas of mixture of air and helium 32 Figure 3.6 Port locations at the cold model CLC for pressure measurement (A1~A15 are for differential pressure measurements. The numbers indicate the vertical distance relative to the distributor of air reactor of the ports, their units are mm. The vertical distances of the ports for gauge pressure measurement are referred to Table D.1 in the Appendix D) 33 3.6 Cross-sectional average solids hold-up along the air reactor For a fully developed flow, with assumption that the wall shear and solid stress are ignored, the vertical variation of the cross-sectional average solids hold-up can be inferred from the differential pressure along the wall of the air reactor [Ellis 2003], as shown by Eq. 3.1. ))1(( sfssgZ P     (3.1) where the P and Z are the differential pressure and the vertical distance between two measurement ports, respectively; s is the solids density; s is the cross-sectional average solids hold-up; and f is the gas density. 3.7 Solids circulation flux measurement Solids circulation flux between the two reactors was measured using two butterfly valves indicated in Figure 3.6 by estimating the time interval for collecting a given volume of solids. When the valve was closed, the particles were collected in a short time interval until a given volume was collected, and the solids flow rate would be calculated based on the pre-determined bulk density of the particle. Dividing the solids flow rate by the cross-sectional area of the air reactor, the solids circulation flux was calculated. The bulk density was measured through a short plexiglass column with ID of 0.1 m. The spent FCC particles were poured down into this column until the pre-determined volume was filled up without shaking. The total amount of the spent FCC in the short column was measured. The bulk density was estimated by dividing the measured mass by the volume. This procedure was repeated for 5 times, and the average bulk density was 870 kg/m3 with a deviation of +/-5.7%. The bulk density of the spent FCC particles colleted by the butterfly in the experiments could be higher than 870 kg/m3 if the pressure in the reactor was significantly higher than the atmospheric pressure. However, the pressure at the location of the butterfly was measured only 1.5 ~ 2 kPa above the atmospheric pressure (presented in Chapter 4). Thus, the effect of reactor pressure on the bulk density was considered negligible. As shown in Figure 3.6, closing the butterfly valves 1 or 2 will collect the solids for calculating the total solids circulation flux or the solids circulation flux in loop 1, respectively. Since the total solids circulation flux is the sum of the solids circulation flux in loop 1 and that in loop 2, the solids circulation flux in loop 2 can then be determined. 34 3.8 Gas leakage measurement Helium was used as gas tracer gas for measuring the gas leakage between the air reactor and fuel reactor. Helium was added into the inlet gases of the air reactor or fuel reactor or loop-seals. The corresponding stable helium concentration was measured by a Micro-GC (Varian Inc.) in the fuel reactor or air reactor or the outlet gas from the cyclones, respectively. The gas leakage between two reactors can be represented by the fraction of helium added into one reactor, which escapes to the other reactor. The experiments were carried out to examine the gas leakage from the air reactor to the fuel reactor, from the fuel reactor to the air reactor, from the loop-seal 1 to the fuel reactor and from the fuel reactor to the cyclone. The detailed experimental procedure is discussed in Chapter 5. 35 CHAPTER 4 HYDRODYNAMIC STUDY WITH FLUIDIZING GAS OF AIR The cold-flow model experimental system, as shown in Figure 3.4, was operated with spent FCC particles of density of 1560 kg/m3 with total solid inventory of 92 kg, fluidized with air. All experiments were conducted at room temperature and atmosphere pressure. The superficial gas velocity in the air reactor, Ua0, was varied between 2.5 and 5.0 m/s with the corresponding superficial gas velocity in the fuel reactor, Uf0, changed from 0.028 to 0.056 m/s in order to achieve the constant input gas volumetric flow rate ratio of air reactor to fuel reactor. The volumetric ratio is fixed at 11 simulating the excess air coefficient of 1.1. As shown in Figure 3.1, loop-seal 1 connecting the air and fuel reactors, and loop-seal 2 connecting the air reactor and primary downcomer are used to control the solids circulation, while preventing gas leakage between the two reactors. Thus, the solids circulation flux in the loops can be controlled by adjusting the aeration velocities in loop-seal 1, UA1, and loop-seal 2, UA2. In this research, UA1 was change from 1Umf to 6Umf, while UA2, was set at 0Umf, 1Umf and 1.5Umf. The solids circulation flux, pressure loop and cross-sectional average solids hold-up in the air reactor were studied under different operating conditions by changing the superficial gas velocities in the reactors and the aeration velocities in both loop-seals. 4.1 Solids circulation flux In a CLC system, the solids circulation dictates the oxygen transportation and energy transfer between the air and the fuel reactors. Thus it is very important to investigate the solids circulation flux in the two loops under different operating conditions. Generally, the initial total solids inventory in the reactor also has significant effect on the solids circulation flux. For a specific operating condition, the solids circulation flux increases with increasing initial solids inventory as confirmed by the experiments of Pröll et al. (2009a). Thus, the following discussion on solids circulation flux focuses on the influences of operating condition for a fixed initial solids inventory. 4.1.1 Aeration velocity in loop-seal 2 at 0 When UA2 = 0, particles are circulated only in loop 1 which connects the air and the fuel reactors, and no solid is transferred from the primary downcomer to the air reactor. Figure 4.1 shows 36 the net solids circulation flux of particles between the air and fuel reactors as functions of UA1 and Ua0. 2 2.5 3 3.5 4 4.5 5 5.5 0 5 10 15 20 25 Superficial gas velocity for air reactor (m/s) S ol id s ci rc ul at io n flu x in lo op 1 (k g/ m2 .s ) UA1=1Umf UA1=2Umf UA1=4Umf UA1=6Umf Figure 4.1 Effect of superficial gas velocity in the air reactor and aeration velocity in loop-seal 1 on the solids circulation flux in loop 1, UA2=0Umf (Measurement error bar is shown in Figure F.1 in Appendix F) It is found that the solids circulation flux in loop 1, Gs1, increases with increasing UA1 for a given Ua0, the maximum solids circulation flux occurs when Ua0 is around 4.0 m/s. However, when Ua0 is increased from 3.0 m/s to 4.0 m/s, the increasing degree of solids circulation flux is small. Figure 4.2 shows the error associated with the solids circulation flux measured for Ua0 of 3.0 and 4.0 m/s. It indicates that there is no significant difference on the solids circulation flux between 3.0 m/s and 4.0 m/s considering the measurement error. The reason of the above phenomena is that the hole (referred to 10 in Figure 2.13) at the bottom end of the lower downcomer has a crucial effect on the Gs1. Figure 4.3 shows the solids flow path between the fuel and air reactors; the shadow section represents the accumulated particles. The solids flow from the lower downcomer to fuel reactor through the hole opened at the bottom end of the lower downcomer, and then continuously overflow from the top of the fuel reactor into the standpipe of loop-seal 1. As discussed in Section 3.1, the particles in lower downcomer and fuel reactor act as an internal loop seal to prevent the gas leakage between the lower downcomer and fuel reactor. This internal loop-seal also is the bottleneck whose limited capacity of transferring particles reduced the Gs1. Thus, when Ua0 is increased from 3.0 to 4.0 m/s, the Gs1 actually is determined by the solids circulation rate through the hole in the lower downcomer. 37 0 5 10 15 20 25 30 1 2 4 6 Aeration velocity in loopseal1 UA1/Umf (-) S o lid s c irc u la tio n flu x in lo op 1 (k g/ m 2 .s ) Ua0=3.0m/s Ua0=4.0m/s Figure 4.2 Measurement error bar for the solids circulation flux in loop 1 (Ua0=3.0 and 4.0 m/s) Figure 4.3 Solids flow pattern from fuel reactor to air reactor through loop-seal 1 Moreover, Figure 4.1 shows that when the Ua0 is increased to 5.0 m/s, the solids circulation flux decreases. This is due to the reduced pressure-head of loop 1 under higher Ua0. Higher maximum carrying capacity of gas-solid suspension in the air reactor is achieved with higher Ua0, but at the same time, increasing Ua0 also results in the increase of pressure in the air reactor and decrease of the pressure-head in loop 1, i.e., the pressure differences between D6 and D3 shown in Figure 4.3. Thus, the achieved Gs1 is determined by the combination of the maximum carrying capacity of gas-solid suspension in the air reactor of a given Ua0, the pressure-head of the loop 1 and 38 the solids circulation rate flowing through the hole (referred to 10 in Figure 2.13) in the lower downcomer. The experiments showed that for UA1 > 6Umf, the Gs1 will not increase further because of the above mentioned bottleneck from internal loop-seal, so UA1 was limited to be below 6Umf for all experiments. 4.1.2 Aeration velocity in loop-seal 2 at 1Umf and 1.5Umf When UA2 ≥ 1Umf, transferring of particles occurs in both loop 1 and loop 2. However, for UA2 ≥ 2Umf, most of the particles are recirculated through loop 2, while the solids circulation flux in loop 1 remains very small. Thus, in this study UA2 was set as 1Umf and 1.5Umf to control the solids circulation flux in loop 2. The effects of operating conditions on the solids circulation flux in loop 1, Gs1, the solids circulation flux in loop 2, Gs2, and total solids circulation flux, Gst (sum of the solids circulation fluxes in loop 1 and loop 2), were further tested. 4.1.2.1 Minimum superficial gas velocity for stable particle transportation For a specific UA2, there is a minimum superficial gas velocity in the air reactor to achieve a stable particle transport in the system. If the Ua0 is less than the minimum value, the solids circulation flux from the primary downcomer to the air reactor will be more than the maximum carrying capacity of gas-solid suspension. This results in the accumulation of solids in the air reactor until the pressure balance in the whole system breaks down. The experiments showed that the minimum superficial gas velocities for stable particle transportation were 1.81 and 2.27 m/s when UA2 was 1Umf and 1.5Umf, respectively. 4.1.2.2 Solids circulation flux in loop 1 Figures 4.4 and 4.5 show Gs1 as a function of Ua0 and UA1 when UA2 is set at 1Umf and 1.5Umf, respectively. As discussed in Section 4.1.1, higher Ua0 results in higher carrying capacity of gas- solid suspension with less pressure-head for loop 1. Thus, these figures show Gs1 follows the similar pattern as shown in Figure 4.1, i.e., higher UA1 resulted in higher Gs1, and with the net solids circulation flux in loop 1 going through a maximum value when Ua0=4.0 m/s. This implies that the air reactor should be operated at Ua0 ≈4.0 m/s in order to achieve the maximum transfer of oxygen from the air reactor to the fuel reactor, because the amount of oxygen per unit time which can be transferred from the air reactor to the fuel reactor is directly proportional to the solids circulation flux between air reactor and fuel reactor. 39 2 2.5 3 3.5 4 4.5 5 5.5 0 5 10 15 20 Superficial gas velocity for air reactor (m/s) S o lid s ci rc u la tio n flu x in lo op 1 (k g/ m2 .s ) UA1 =Um f UA1 =2Umf U A1 =4U mf UA1 =6Umf Figure 4.4 Effect of superficial gas velocity in air reactor and aeration velocity in loop-seal 1 on the solids circulation flux in loop 1, UA2=1Umf (Measurement error bar is shown in Figure F.2 in Appendix F) 2 2.5 3 3.5 4 4.5 5 5.5 0 5 10 15 Superficial gas velocity for air reactor (m/s) S ol id s ci rc u la tio n flu x in lo o p1 (k g/ m2 .s ) UA1=1Um f UA1=2Um f U A1 =4U m f UA1=6Um f Figure 4.5 Effect of superficial gas velocity in air reactor and aeration velocity in loop-seal 1 on the solids circulation flux in loop 1, UA2=1.5Umf (Measurement error bar is shown in Figure F.3 in Appendix F) 40 4.1.2.3 Solids circulation flux in loop 2 Figure 4.6 shows the solids flow path in the loop 2 which is comprised of the primary downcomer and the air reactor. The solids circulation flux between the air reactor and primary downcomer is controlled by the loop-seal 2. The effect of Ua0 on the Gs2 is shown in Figures 4.7 and 4.8. Figure 4.6 Solids flow between air reactor and primary downcomer in loop 2 Similarly, Gs2 is determined by the balance between Ua0, pressure head (pressure drop between D5 and D4 in Figure 4.6) of loop 2, and UA2. As a result, as shown in Figures 4.7 and 4.8, for a specific UA1, Gs2 generally increases with increasing Ua0 between 2.5 and 4.0 m/s. Note that the maximum Gs1 also occurs at Ua0 of 4.0 m/s, thus, the phenomenon imply that Ua0 of 4.0 m/s is the best operation condition for achieving the maximum solids circulation flux in both of the loops. However, as shown in Figure 4.6, height of the accumulated particles in the primary downcomer, i.e., H2, was very high, this resulted in the large pressure at point D5 during the 41 operation. Pressure head of loop 2 is the main parameter to control Gs2. Thus, if UA2 > 1.5Umf, most of the particles are only recirculated in the loop 2. 2 2.5 3 3.5 4 4.5 5 5.5 0 1 2 3 4 5 6 7 8 9 10 Superficial gas velocity for air reactor (m/s) S ol id s ci rc ul at io n flu x in lo op 2 (k g/ m2 .s ) UA1=1Umf UA1=2Umf UA1=4Umf UA1=6Umf Figure 4.7 Effect of superficial gas velocity in air reactor on the solids circulation flux in loop 2, UA2=1Umf (Measurement error bar is shown in Figure F.4 in Appendix F) 2 2.5 3 3.5 4 4.5 5 5.5 0 2 4 6 8 10 12 14 16 18 20 Superficial gas velocity for air reactor (m/s) S ol id s ci rc ul at io n flu x in lo op 2 (k g/ m2 .s ) U A1 =1U mf U A1 =2U mf UA1=4Umf U A1 =6U mf Figure 4.8 Effect of superficial gas velocity in air reactor on the solids circulation flux in loop 2, UA2=1.5Umf (Measurement error bar is shown in Figure F.5 in Appendix F) 42 As shown in Figures 4.4, 4.5, 4.7 and 4.8, the solids flux in both of the loops increase with increasing Ua0 reaching maximum values at Ua0 = 4.0 m/s. If Ua0 is further increased, the solids flux will decrease. However, this trend does not apply for UA1=1Umf as shown on Figure 4.7.The most likely explanation is that at very small UA1 and UA2, the increasing carrying capacity of gas-solids does not compensate for the reduced pressure head of loop 2 due to higher Ua0. 4.1.2.4 Total solids circulation flux Figures 4.9 and 4.10 show Gst as a function of Ua0 and UA1 which varied in the range of 1Umf ~ 6Umf, while keeping UA2 at 1Umf and 1.5Umf, respectively. Because Gst is the sum of the net solids circulation fluxes in both loop 1 and loop 2, it follows the same fluctuation pattern as Gs1 with that a maximum value at Ua0 of 4.0 m/s and then decreases with increasing Ua0. At the same time, it also increases with increasing of UA1. 4.1.3 Effect of aeration velocity in loop-seal 2 on solids circulation flux When UA2 is increased from 0 to 1.5Umf, more and more particles will be transported in loop 2 which connects the air reactor and primary downcomer bypassing the fuel reactor. In order to investigate the effect of UA2, the typical profiles of Gs1, Gs2 and Gst when Ua0 = 4.0 m/s are further discussed. 4.1.3.1 Effect on solids circulation flux in loop 1 Figure 4.11 shows the effect of UA2 on Gs1. The solids circulation flux in loop 1 decreases when UA2 is increased from 0 to 1.5Umf. The phenomenon indicates that the solids circulation flux shifts from loop 1 to loop 2 with increasing UA2. Thus if UA2 is too high, there would be less amount of particles circulating in loop 1. 4.1.3.2 Effect on solids circulation flux in loop 2 Figure 4.12 presents the effect of UA2 on Gs2. A large increase of Gs2 is observed with UA2 changing from 1Umf to 1.5Umf as more particles circulate in the loop 2. Larger solids circulation flux was observed in loop 2 for UA2>2Umf which resulted in much higher solids circulation in loop 2 than the maximum collection capacity of cyclones, ending up in gas pipe blockage. 43 2 2.5 3 3.5 4 4.5 5 5.5 0 5 10 15 20 25 30 Superficial gas velocity for air reactor (m/s) T ot al so lid s ci rc ul at io n flu x (k g/ m2 .s ) UA1=1Um f U A1 =2U m f UA1=4Um f UA1=6Um f Figure 4.9 Effect of superficial gas velocity in air reactor and aeration velocity in loop-seal 1 on the total solids circulation flux, UA2=1Umf (Measurement error bar is shown in Figure F.6 in Appendix F) 2 2.5 3 3.5 4 4.5 5 5.5 0 5 10 15 20 25 30 35 Superficial gas velocity for air reactor (m/s) T ot al so lid s ci rc u la tio n flu x (k g/ m2 .s ) UA1=1Um f U A1 =2U m f UA1=4Um f UA1=6Um f Figure 4.10 Effect of superficial gas velocity in air reactor and aeration velocity in loop-seal 1 on the total solids circulation flux, UA2=1.5Umf (Measurement error bar is shown in Figure F.7 in Appendix F) 44 0 1 2 3 4 5 6 7 0 5 10 15 20 25 Aeration velocity in loopseal1 U A1 /U mf (-) S ol id s ci rc ul at io n flu x in lo op 1 (k g/ m2 .s ) U A2 =0 m/s U A2 =1U m f U A2 =1.5U mf Figure 4.11 Effect of aeration velocity in loop-seal 2 on solids circulation flux in loop 1, Ua0=4.0 m/s (Measurement error bar is shown in Figure F.8 in Appendix F) 0 1 2 3 4 5 6 7 0 2 4 6 8 10 12 14 16 18 20 Aeration velocity in loopseal1 UA1/Um f (-) S ol id s ci rc ul at io n flu x in lo op 2 (k g/ m2 .s ) UA2=1Umf UA2=1.5Umf Figure 4.12 Effect of aeration velocity in loop-seal 2 on solids circulation flux in loop 2, Ua0=4.0 m/s (Measurement error bar is shown in Figure F.9 in Appendix F) 45 4.1.3.3 Effect on total solids circulation flux The effect of UA2 on Gst is presented in Figure 4.13. It is found that Gst is increased with increasing UA2, but Figure 4.11 indicate that solids circulation flux in loop 1 is a decreasing function of UA2. This means when UA2 is increased, the increasing of solids circulation flux occurring in loop 2 is far higher than the reduced amount of the solids circulation flux in loop 1. The effect of UA1 on Gst is not significant when UA2 = 1.5Umf, because compared with the pressure-head of loop 1, the pressure-head of loop 2 is large, thus particles are mainly transferred between the air reactor and primary downcomer. The total solids circulation flux is mainly controlled by the pressure-head of loop 2. 0 1 2 3 4 5 6 7 0 5 10 15 20 25 30 35 Aeration velocity in loopseal1 U A1 /U mf (-) To ta ls ol id s ci rc ul at io n flu x (k g/ m2 .s ) U A2 =0 m/s U A2 =1U mf U A2 =1.5U mf Figure 4.13 Effect of aeration velocity in loop-seal 2 on total solids circulation flux, Ua0=4.0 m/s (Measurement error bar is shown in Figure F.10 in Appendix F) As discussed in Chapter 2, loop 2 utilizes the primary downcomer to transport a portion of the particles out of the air reactor to return to the air reactor, directly bypassing the fuel reactor in order to increase the average oxidation degree of the oxygen carrier. If UA2 is too high, most particles will circulate only between the air reactor and the primary downcomer, while a small amount of particles circulate between the two reactors, which will limit the oxygen transfer. Thus UA2 should be chosen at 1.5Umf for a specific Ua0 to maintain the ratio of solids circulation fluxes between loop 1 and loop 46 2 at 1:1. Based on the experiments, the desirable operating conditions which satisfy the above requirements can be mapped. 4.1.4 Mapping of operation conditions for solids circulation flux Figures 4.14 to 4.17 show the mapping of operation conditions for solids circulation flux in detail. For a specific Ua0 and U f0, there are 8 combinations of UA1 and UA2. The height of the bars represents the solids circulation flux in the two loops. These figures indicate that when UA2=1Umf, only small amount of particles circulate in loop 2 while most of particles circulating in loop 1. With UA2 changing from 1Umf to 1.5Umf, the significant shift of circulating particles from loop 1 to loop 2 occurs because the pressure head of loop 2 is increased significantly. Through these figures, it is helpful to choose a proper operating condition. As shown in Figures 4.15 and 4.16, when the superficial Ua0 = 3.0 or 4.0 m/s, UA1 = 2Umf ~ 6Umf and UA2 =1.5Umf, the solids circulation flux of 10 ~ 15 kg/m2.s for both of the loops can be achieved with the ratio of the Gs1 to Gs2 around 1:1. Thus, the best operating conditions for the solids circulation flux can be chosen as Ua0 = 3.0 or 4.0 m/s, UA1 = 2Umf ~ 6Umf and UA2 =1.5Umf. 0 2 4 6 8 10 12 14 16 18 20 S o lid s ci rc ul at io n fl ux (k g/ m 2 . s) UA1=1Umf UA2=1Umf Gs1 Gs2 UA1=2Umf UA2=1Umf UA1=4Umf UA2=1Umf UA1=6Umf UA2=1Umf UA1=1Umf UA2=1.5Umf UA1=2Umf UA2=1.5Umf UA1=4Umf UA2=1.5Umf UA1=6U mf UA2=1.5Umf Figure 4.14 Mapping of operating conditions (Ua0=2.5 m/s, Uf0=0.029 m/s) 47 0 2 4 6 8 10 12 14 16 18 20 S o li d s ci rc u la ti o n fl u x (k g /m 2 . s) UA1=1Umf UA2=1Umf Gs1 Gs2 UA1=2Umf UA2=1Umf UA1=4Umf UA2=1Umf UA1=6U mf UA2=1U mf UA1=1U mf UA2=1.5Umf UA1=2Umf UA2=1.5U mf UA1=4Umf UA2=1.5Umf UA1=6Umf UA2=1.5Umf Figure 4.15 Mapping of operating conditions (Ua0=3.0 m/s, Uf0=0.035 m/s) 0 2 4 6 8 10 12 14 16 18 20 S o li d s ci rc ul at io n flu x (k g /m 2 . s) UA1=1Umf UA2=1Umf Gs1 Gs2 UA1=2Umf UA2=1Umf UA1=4Umf UA2=1Umf UA1=6Umf UA2=1Umf UA1=1U mf UA2=1.5Umf UA1=2Umf UA2=1.5Umf UA1=4Umf UA2=1.5Umf UA1=6Umf UA2=1.5Umf Figure 4.16 Mapping of operating conditions (Ua0=4.0 m/s, Uf0=0.046 m/s) 48 0 2 4 6 8 10 12 14 16 18 20 S o li d s ci rc u la ti o n fl u x (k g /m 2 .s ) UA1=1Umf UA2=1Umf Gs1 Gs2 UA1=2Umf UA2=1Umf UA1=4Umf UA2=1Umf UA1=6U mf UA2=1U mf UA1=1U mf UA2=1.5Umf UA1=2Umf UA2=1.5Umf UA1=4Umf UA2=1.5Umf UA1=6Umf UA2=1.5Umf Figure 4.17 Mapping of operating conditions (Ua0=5.0 m/s, Uf0=0.058 m/s) 4.2 Pressure loop The pressure balance is very important for maintaining stable solids circulation in the CLC reactor. As shown in Figure 3.6, there are two solids loops in the cold-flow model, i.e., loop 1 of D2- A2-B9-B8-B7-B6-B2-D6-D3-D2, and loop 2 of A2-B9-B8-B7-B6-B5-D5-D4-A2. Loop 1 is used to achieve the solid circulation between the air reactor and the fuel reactor, while loop 2 utilizes primary downcomer to transport some particles out of the air reactor to return to the riser directly without catalyst reduction (without going into the fuel reactor). Points D2, A2, B9 show the pressures in the air reactor. Point B2 is shows the pressure at the solids exit of the fuel reactor. Points D6 and D3 represent the pressures in loop-seal 1. The pressures in loop-seal 2 are represented by points of D4 and D5. Thus, the pressure heads of the loop-seal 1 between D6 and D3 in loop 1 and of loop-seal 2 between D5 and D4 in loop 2 provide the forces to achieve a stable solid flow from fuel to air reactors and from primary downcomer to air reactor. 49 4.2.1 Effect of aeration velocity in loop-seal 1 Figures 4.18 and 4.19 present the typical profiles of pressure loops, when Ua0 = 3.0 m/s and UA2 = 1.5Umf, showing the effect of UA1 on the pressure loop 1 and pressure loop 2, respectively. These figures indicate the pressure in the cold-flow model is increased as the UA1 change from 1Umf to 6Umf, while the amount of bed material in the air reactor, which is given by the pressure drop between D2 and B9 (shown in Figure 4.18), is increased. 4.2.2 Effect of superficial gas velocity Figures 4.20 and 4.21 show the effect of Ua0 on the pressure loop 1 and loop 2 for Ua0 = 2.5 and 5.0 m/s when UA1 and UA2 were fixed at 4Umf and 1.5Umf, respectively. Less bed mass in the air reactor resulting from higher Ua0 can be indicated by the smaller pressure drop between D2 and B9 as shown in Figure 4.20. Therefore, there will be less bed material in the air reactor for reaction with oxygen at higher Ua0. 4.2.3 Effect of blockage in filter Figures 4.22 and 4.23 show the effect of the blockage in filter on the pressure loop 1 and loop 2. The pressure profiles which are represented by the symbol of ‘○’ were measured right after the accumulated particles in the filter were cleaned up, and the symbol of ‘×’represents the pressure profiles which were measured before the blockage was removed. It is shown that the blockage in the filter caused by the entrainment of very fine particles from the reactor has significant effect on the pressure in the reactor. The blockage was caused by the very fine particles accumulated on the filter paper element collected about 5 g. Since the total inventory in the reactor is 92 kg and the very fine particles accumulated on the paper filter element has been returned to the reactor immediately after the paper element was clean, it can be assumed that particle size distribution in the experiments has not varied significantly. 50 0 1 2 3 4 5 6 7 8 9 0 1000 2000 3000 4000 5000 6000 Pressure P-P atmo sphere (kPa) H ei gh ta bo ve th e ga s di st rib ut or of ai rr ea ct or (m m ) U A1 =1U m f U A1 =2U m f U A1 =4U m f U A1 =6U m f B9 B8 B7 B6 B2 A2 D2 D3 D6 Figure 4.18 Pressure loop 1, U0=3.0 m/s, UA2=1.5Umf (D2-A2-B9: air reactor; B9-B8: elbow connector and cyclone; B7-B6-B2: upper downcomer, lower downcomer and fuel reactor; D6-D3: loop-seal 1. The positions of the measurement points are shown in Figure 3.6) 1 2 3 4 5 6 7 8 9 10 0 1000 2000 3000 4000 5000 6000 Pressure P-P atmos phere (kPa) H ei gh ta bo ve th e ga s di st rib ut or of ai rr ea ct or (m m ) UA1=1Um f U A1 =2U m f U A1 =4U m f U A1 =6U m f B9 B8 B7 B6 B5 D4 A2 D5 Figure 4.19 Pressure loop 1, U0=3.0 m/s, UA2=1.5Umf (A2-B9: air reactor; B9-B8: elbow connector and cyclone; B7-B6-B5: primary downcomer; D5-D4: loop-seal 2. The positions of the measurement points are shown in Figure 3.6) 51 0 1 2 3 4 5 6 7 8 9 0 1000 2000 3000 4000 5000 6000 Pressure P-P atmos phere (kPa) H ei gh ta bo ve th e ga s di st rib ut or of ai rr ea ct or (m m ) U a0 =2.5m/s U a0 =5.0m/s B9 B8 B7 B6 B2 D3 D6 A2 D2 Figure 4.20 Pressure loop 1, UA1=4Umf, UA2=1.5Umf (D2-A2-B9: air reactor; B9-B8: elbow connector and cyclone; B7-B6-B2: upper downcomer, lower downcomer and fuel reactor; D6-D3: loop-seal 1.The positions of the measurement points are shown in Figure 3.6) 1 2 3 4 5 6 7 8 9 10 0 1000 2000 3000 4000 5000 6000 Pressure P-P atmosphere (kPa) H ei gh ta bo ve th e ga s di st rib ut or of ai rr ea ct or (m m ) U a0 =2.5m/s U a0 =5.0m/s B9 B8 B7 B6 B5 D4 A2 D5 Figure 4.21 Pressure loop 2, UA1=4Umf, UA2=1.5Umf (A2-B9: air reactor; B9-B8: elbow connector and cyclone; B8-B7-B6-B5:primary downcomer; D5-D4: loop-seal 2. The positions of the measurement points are shown in Figure 3.6) 52 0 2 4 6 8 10 0 1000 2000 3000 4000 5000 6000 Pressure P-P atmosphere (kPa) H ei gh ta bo ve th e ga s di st rib ut or of ai rr ea ct or (m m ) No blockage Blockage B9 B8 B7 B6 B2 D3 A2 D2 Figure 4.22 Pressure loop 1, Ua0=4.0 m/s, UA1=4Umf, UA2=1.5Umf (D2-A2-B9: air reactor; B9-B8: elbow connector and cyclone; B7-B6-B2: upper downcomer, lower downcomer and fuel reactor; D6-D3: loop-seal 1. The positions of the measurement points are shown in Figure 3.6) 1 2 3 4 5 6 7 8 9 10 0 1000 2000 3000 4000 5000 6000 Pressure P-P atmos phere (kPa) H ei gh ta bo ve th e ga s di st rib ut or of ai rr ea ct or (m m ) No blockage Blockage B9 B8 B7 B6 B5 D4 A2 D5 Figure 4.23 Pressure loop 2, Ua0=4.0 m/s, UA1=4Umf, UA2=1.5Umf (A2-B9: air reactor; B9-B8: elbow connector and cyclone; B8-B7-B6-B5: primary downcomer; D5-D4: loop-seal 2. The positions of the measurement points are shown in Figure 3.6) 53 4.2.4 Effect of aeration velocity in loop-seal 2 When UA2 is increased from 0 to 1.5Umf for a given Ua0, the amount of particles circulating in the system increase, as a result, the pressure in the air reactor will be increased to support the movement of the particles. This effect of UA2 on the pressure loop 1 and loop 1 is shown in Figure 4.24 and 4.25. 0 1 2 3 4 5 6 7 8 0 1000 2000 3000 4000 5000 6000 Pressure P-Pa tmosphere (kPa) H ei g ht ab ov e th e g as d is tr ib u to ro fa ir re ac to r( m m ) U A2 =0U mf UA2=1Umf UA2=1.5Umf B9 B8 B7 B6 B2 D3 A2 D2 D6 Figure 4.24 Pressure loop 1, Ua0=4.0 m/s, UA1=6Umf (D2-A2-B9: air reactor; B9-B8: elbow connector and cyclone; B7-B6-B2: upper downcomer, lower downcomer and fuel reactor; D6-D3: loop-seal 1. The positions of the measurement points are shown in Figure 3.6) 54 1 2 3 4 5 6 7 8 9 0 1000 2000 3000 4000 5000 6000 Pressure P-Patm osphere (kPa) H ei gh ta bo ve th e ga s di st rib ut or of ai rr ea ct or (m m ) UA2=1Um f U A2 =1.5U mf B9 B8 B7 B6 B5 D5 D4 A2 Figure 4.25 Pressure loop 2, Ua0=4.0 m/s, UA1=6Umf (A2-B9: air reactor; B9-B8: elbow connector and cyclone; B8-B7-B6-B5:primary downcomer; D5-D4: loop-seal 2. The positions of the measurement points are shown in Figure 3.6) 4.3 Cross-sectional average solids hold-up in the air reactor 4.3.1 Effect of aeration velocity in loop-seal 1 Typical profiles showing the solids hold-up varied with UA1 for constant Ua0 and UA2 are presented in Figures 4.26 and 4.27. Figure 4.27 is a partial enlargement of Figure 4.26 to show the solids hold-up distribution at upper section of air reactor more clearly because the solids hold-up in the bottom section is much more than in the upper section. Because the lower part of the air reactor (height of 0~500 mm) is connected with loop-seal 1 and loop-seal 2, there is a dense zone in the lower part of the air reactor with strong backmixing and splashing of solids caused by the input particles from loop-seal 1 and loop-seal 2. Above the dense zone, the solids hold-up decreases along the height of air reactor. On the other hand, the figures indicate that with increasing of UA1, the solid hold-up in the air reactor is also increased at all heights. This means there is more bed material in the air reactor to react with oxygen with higher oxygen conversion efficiency. As discussed in Section 4.1, Gs1 is also increased with increasing of UA1. Thus, 55 UA1 should be as high as possible to provide higher solids hold-up and Gs1 in order to transfer more oxygen from the air reactor to the fuel reactor for a given Ua0. 0 5 10 15 20 25 0 1000 2000 3000 4000 5000 6000 Cross sectional average solids hold-up (%) H e ig h ta b ov e th e ga s di st rib ut o ro fa ir re a ct o r( m m ) U A1 =1U mf U A1 =2U mf UA1=4Umf UA1=6Umf Figure 4.26 Cross sectional average solids hold-up distribution in air reactor U0=3.0 m/s, UA2=1.5Umf 0 1 2 3 4 5 6 7 0 1000 2000 3000 4000 5000 6000 Cross sectional average solids hold-up (%) H e ig ht ab ov e th e ga s di st rib ut or o fa ir re ac to r( m m ) UA1=1Umf U A1 =2U mf UA1=4Umf U A1 =6U mf Figure 4.27 Cross sectional average solids hold-up distribution in air reactor (Partial enlargement figure of Figure 4.26) U0=3 m/s, UA2=1.5Umf 56 Figure 4.27 indicates an evident increase of solids hold-up when UA1 is increased from 1Umf to 2Umf and from 4Umf to 6Umf. However, such increase of solids hold-up is not observed when UA1 is changed from 2Umf to 4Umf. This is because solids hold-up in the air reactor is proportional to the total solids flux flowing through the air reactor for a specific Ua0. However, as shown in Figure 4.10, when Ua0=3.0 m/s, the total solids flux at UA1=2Umf is close to that at UA1=4Umf. Thus, there is no significant increase of solids hold-up in the air reactor between UA1=2Umf and UA1=4Umf. 4.3.2 Effect of superficial gas velocity Figure 4.28 shows the effect of Ua0 on the solids hold-up in the air reactor when UA1 andUA2 were set as 4Umf and 1.5Umf, respectively. The solids hold-up could be significantly reduced when Ua0 was changed from 2.5 to 5.0 m/s. This means the solids fraction in the air reactor which could react with oxygen would be reduced at higher Ua0. However, as shown in Figure 4.1, 4.4 and 4.5, for Ua0 in the range of 2.5 m/s~4.0 m/s, Gs1 increased with increasing Ua0. For ensuring enough oxygen transferred from the air reactor to fuel reactor, Ua0 has to be chosen to provide the maximum oxygen transfer considering the combined influence of solids circulation flux and solids hold-up. 0 5 10 15 20 25 30 35 0 1000 2000 3000 4000 5000 6000 Cross sectional average solids hold-up (%) H ei gh ta bo ve th e ga s di st rib ut or of ai rr ea ct or (m m ) U a0 =2.5m/s U a0 =3.0m/s U a0 =4.0m/s U a0 =5.0m/s Figure 4.28 Cross sectional average solids hold-up distribution in air reactor UA1=4Umf, UA2=1.5Umf 57 4.3.3 Effect of solids from loop 2 As shown in Figure 4.29 the solids from loop 2 result in an increase of the solids hold-up in the air reactor. Especially for the section at height of 0~500 mm where the particles are transferred through loop-seal 2, evident accumulation of particles can be observed. It is clear that when UA2 is increase from 0Umf to 1.5Umf, the overall average solids hold-up in the air reactor can be increased by about 50%. Therefore, adding solids from loop 2 into the air reactor is a good solution to achieve higher oxygen conversion in the air reactor considering the total solids circulation flux also increased with increasing UA2. 0 2 4 6 8 10 12 14 16 0 1000 2000 3000 4000 5000 6000 Cross sectional average solids hold-up (%) H ei gh ta bo ve th e g as di st rib ut o ro fa ir re ac to r( m m ) U A2 =0U mf U A2 =1U mf UA2=1.5Umf Figure 4.29 Cross sectional average solids hold-up distribution in air reactor UA1=6Umf, Ua0=4.0 m/s 4.4 Summary The solids circulation flux, pressure loop and cross-sectional average solids hold-up in the air reactor were studied under different operating conditions.  Solids circulation flux The solids circulation flux increased with increasing superficial gas velocity for Ua0 ≤ 4.0 m/s, and then decreased when Ua0 was further increased from 4.0 to 5.0 m/s. The maximum solids circulation flux occurred at around Ua0 = 4.0 m/s. The achieved solids circulation flux was found to 58 be determined by the combination of the maximum carrying capacity of gas-solid suspension at the superficial gas velocity in the air reactor, the pressure-head of the loop, and the solids circulation rate flowing through the hole (referred to 10 in Figure 2.13) in the lower downcomer. For a specific superficial gas velocity, higher UA1 and UA2 would result in higher solids circulation flux in both loops.  Pressure profile The pressure in the cold-flow model increased with increasing superficial gas velocity.  Vertical solids hold-up distribution in the air reactor The solids hold-up in the air reactor decreased with increasing superficial gas velocity, while higher UA1 and UA2 could provide higher solids hold-up in the air reactor.  Good operating condition for the cold–flow model with fluidizing gas of air Since higher UA1 and UA2 would give higher solids hold-up and higher solids circulation flux, the recommended operation conditions for the cold-flow model are UA1 at 4Umf or 6Umf, and UA2 at 1.5 Umf. The superficial gas velocity in the air reactor, Uao, should be as small as possible to maintain a high solids hold-up. However, higher Ua0 is desirable for more oxygen input to the air reactor, and higher solids circulation flux (for Ua0 ≤ 4.0 m/s). Thus, Ua0 = 3.0 or 4.0 m/s is suggested for the reactor operation, while Uf0 should be set as 0.034 or 0.045 m/s, correspondingly. 59 CHAPTER 5 GAS LEAKAGE The investigation of gas leakage is very important for an interconnected fluidized bed CLC reactor, because the gas leakage between the reactors will reduce the CO2 separation efficiency of the CLC process. In this study, the dependency of gas leakage in the CLC reactor system on the operating parameters and pressure balance were examined. Helium was used as a gas tracer for the CLC system fluidized with air at steady-state operating conditions. The stable helium concentration in the air reactor, fuel reactor and outlet gas flow from the cyclone were measured with a Micro-GC (Varian CP4900). The gas leakage between the two reactors is defined as the fraction of helium added to the reactor, which escapes to the other reactor. In order to understand the influence of the operating conditions on the gas leakage, the superficial gas velocity in the air reactor, Ua0, was varied in the range between 2.5 and 5.0 m/s with the corresponding superficial gas velocity in the fuel reactor, Uf0, varying between 0.028 to 0.056 m/s to achieve a constant input gases volumetric flow rate ratio of the air reactor to the fuel reactor. The volumetric flow rate ratio was fixed at 11 simulating the excess air coefficient of 1.1. The aeration velocity in loop-seal 1 and loop-seal 2, UA1 and UA2, were changed from 1Umf to 6Umf, and between 0Umf, 1Umf and 1.5Umf, respectively. Thus, there were 48 combinations of operating conditions in which the gas leakages from fuel reactor to air reactor, from fuel reactor to the cyclone, from air reactor to fuel reactor and from loop-seals to fuel reactor were studied. Table D.2 in Appendix D shows these operating condition combinations. 5.1 Gas leakage from fuel reactor to air reactor 5.1.1 Approaches to analyze gas leakage from fuel reactor to air reactor As shown in Figure 5.1, there are two possible pathways for gas leaking from fuel reactor to air reactor, i.e., fuel reactor - loop-seal 1 - air reactor, and fuel reactor - lower downcomer - upper downcomer - primary downcomer - loop-seal 2 - air reactor. In order to investigate the gas leakage from the fuel reactor to air reactor, helium was injected at point A. The steady-state helium concentration in the middle section of the air reactor was measured. The measurement was repeated ten times during the half hour, and the average value of the last five stable concentrations was used to calculate the average gas leakage. 60 Figure 5.1 Possible routes for gas leakage from fuel reactor to air reactor The detection of helium in the air reactor confirmed the leakage through the two possible ways as mentioned above. If UA2 is equal to 0, no particle is transported in the loop 2 resulting in no helium leaking into air reactor from fuel reactor through loop-seal 2. Therefore, comparing the gas leakages from fuel reactor to air reactor when UA2=0 to that for UA2≠ 0 can determine the leakage pathway. This will be discussed further in Section 5.1.4. 5.1.2 Effect of pressure drop across loop-seal 1 Figure 5.2 shows the effect of pressure drop across loop-seal 1, i.e., the pressure difference between point D3 and B2 (PD3-PB2) in Figure 3.6, on the gas leakage from the fuel reactor to the air reactor. At Ua0=2.5 m/s, the gas leakage gradually decreases from 3.0% to 0.8% with increasing 61 pressure drop across loop-seal 1. This is a consequence of higher positive pressure gradient between solids outlet and solids inlet of loop-seal 1 which reduces the gas flow carried by the particles flowing at the opposite direction, i.e., from solids inlet to solids outlet of loop-seal 1. Similar phenomena can be observed when Ua0 varies at 3.0, 4.0 and 5.0 m/s, as shown in Figures D.1~D.3 in Appendix D. 2 2.5 3 3.5 4 0 0.5 1 1.5 2 2.5 3 3.5 4 Pressure drop across loopseal1, P D3 - P B2 (kPa) G as le ak ag e fro m fu el re ac to rt o ai rr ea ct or (% ) U A1 =1U mf UA1=2Umf U A1 =4U mf U A1 =6U mf Figure 5.2 Leakage from fuel reactor to air reactor versus the pressure drop across loop-seal 1, PD3-PB2, Ua0=2.5 m/s, Uf0=0.029 m/s (The standard deviation is shown in Table F.1 in Appendix F) After hours of continuous experimental run, the air reactor pressure started to increase due to particles accumulating in the filter, resulting in increasing pressure drop across the loop-seal 1, thus decreasing the gas leakage from fuel reactor to air reactor as shown in Figure 5.3. The system was run continuously for 8 hours at Ua0 = 4.0 m/s, while measuring the gas leakage from the fuel reactor to the air reactor for 12 operating conditions, i.e., UA1 varied at 1Umf, 2Umf, 4Umf and 6Umf, and UA2 varied at 0Umf , 1Umf and 1.5Umf . A comparison, as shown in Figure 5.3, was made for the gas leakage after running the system for 100 hours. It was observed that the gas leakage from fuel reactor to air reactor through loop-seal 1 had decreased as the pressure drop across loop-seal 1 increased over time. Thus, the pressure drop across the loop-seal 1 is the most important factor influencing the gas leakage from fuel reactor to air reactor through loop-seal 1. 62 2 2.5 3 3.5 4 4.5 0 0.5 1 1.5 2 2.5 3 Pressure drop across loopseal1, P D3 - P B2 (kPa) G a s le a ka ge fro m fu el re ac to rt o ai r re ac to r (% ) Running after 8 hours Running after 100 hours Figure 5.3 Leakage from fuel reactor to air reactor versus the pressure drop across loop-seal 1, PD3-PB2 (Ua0=4.0 m/s, Uf0=0.046 m/s, UA1=1Umf ~6Umf, UA2=0Umf ~1.5Umf) 5.1.3 Effect of aeration velocity in loop-seal 1 Figure 5.4 presents typical curves of gas leakage from fuel reactor to air reactor as a function of UA1, for Ua0 = 5.0 m/s, indicating that the variation of UA1 is not significantly affecting the degree of gas leakage. It is expected that more gas from the fuel reactor follows the particles flowing through loop-seal 1 at higher solids circulation flux leaking into the air reactor. Thus, higher solids circulation flux, which is created by higher UA1, will result in higher gas leakage. However, at the same time, the more rapid aeration gas flowing upward in loop-seal 1 reduces this gas leakage. As the result of these counter acting trends, UA1 did not significantly affect the gas leakage. 5.1.4 Effect of aeration velocity in loop-seal 2 Figure 5.5 shows the effect of UA2 on the gas leakage from fuel reactor to air reactor, when Ua0 = 2.5 m/s. The influences of UA2 for Ua0 = 3.0, 4.0 and 5.0 m/s are shown in the Figures D.4~D6 in Appendix D. Comparing the results measured at UA2=0 with that obtained at UA2≠ 0, while Ua0 and UA1 were held constant, the gas leakage becomes less or increases slightly. This suggests that the gas leakage through the route of fuel reactor - lower downcomer - upper downcomer - primary downcomer - loop-seal 2 - air reactor (in Figure 5.1) is very small and can be negligible. Most of the gases leaked from fuel reactor to air reactor actually are entrained by the solids flowing through loop-seal 1. 63 0 1 2 3 4 5 6 7 0 0.5 1 1.5 2 2.5 3 3.5 Aeration velocity for loopseal1, U A1 /U mf (-) G as le ak ag e fro m fu el re ac to rt o ai rr ea ct or (% ) UA2=1Umf UA2=1Umf UA2=1.5Um f Figure 5.4 Leakage from fuel reactor to air reactor versus aeration velocity in loop-seal 1, Ua0=5.0 m/s, Uf0=0.058 m/s (The standard deviation is shown in Table F.2 in Appendix F) -0.5 0 0.5 1 1.5 2 0 0.5 1 1.5 2 2.5 3 3.5 Aeration velocity in loopseal2, U A2 /U mf (-) G a s le ak ag e fro m fu el re ac to rt o ai rr ea ct or (% ) UA1=1Umf UA1=2Umf UA1=4Umf U A1 =6U mf Figure 5.5 Leakage from fuel reactor to air reactor versus aeration velocity for loop-seal 2, Ua0=2.5 m/s, Uf0=0.029 m/s (The standard deviation is shown in Table F.3 in Appendix F) As discussed in Chapter 4, the solids circulation flux in loop 1 decreases with increasing UA2. As a result, the gas entrained by solids flowing through loop-seal 1 from fuel reactor to air reactor should also be reduced. However, the discussion in Sections 5.1.2 and 5.1.3 indicate that less positive pressure drop across the loop-seal 1 results in higher gas leakage, while more rapid upward 64 gas flowing in the loop-seal 1 resulting from higher UA1 is helpful to reduce the gas leakage. Thus, gas leakage from the fuel reactor to the air reactor actually is a consequence of a combination of solids flowing through loop-seal 1, pressure drop across loop-seal 1 and aeration gas in loop-seal 1. 5.1.5 Summary for the gas leakage from fuel reactor to air reactor The gas leakage from the fuel reactor to air reactor results from a combination of pressure drop across loop-seal 1, solids flowing through loop-seal 1 and aeration gas in loop-seal 1. Higher positive pressure drop between solids outlet and inlet of loop-seal 1 reduces the gas leakage. On the other hand, the solids with higher solids circulation flux through loop-seal 1 could carry more gas into the air reactor from the fuel reactor. In the meanwhile, the more rapid upward gas flow in loop- seal 1 resulting from higher aeration velocity is helpful to reduce gas leakage. There are there mechanisms for gas to leak from one reactor to another, i.e., through gas pressure difference, gas dragged by particles movement and gas adsorption on the particles. As discussed in Section 5.3, the maximum gas leakage from the air reactor to fuel reactor is below 0.57%. Therefore, the maximum possible effect of helium adsorption on the spent FCC on the total gas leakage from one reactor to the other one should be less than 0.57%. However, the experiments present the gas leakage from the fuel reactor to another varied over a range of 0.8% ~ 3.0%. Thus, it can be concluded that most of the gas leaked from the fuel reactor to the air reactor is entrained by the particles stream flowing through loop-seal 1 since the pressure in the air reactor is higher than in the fuel reactor. 5.2 Gas leakage from fuel reactor to cyclone 5.2.1 Approaches to analyze gas leakage from fuel reactor to the cyclone As shown in Figure 5.6, the possible route for gas leaking from the fuel reactor to the cyclone is the route of fuel reactor - lower downcomer - upper downcomer - upper section of primary downcomer - cyclone. Helium was injected into the inlet gas of the fuel reactor operated at steady- state while the helium concentration at the exit of the cyclone was measured. As discussed in Section 5.1, helium can leak into the air reactor through loop-seal 1. Thus, the helium gas out of the cyclone can come through the air reactor and the fuel reactor. In order to determine the amount of helium leaking from the fuel reactor to the cyclone through the route shown in Figure 5.6, simultaneous helium concentration measurement at the middle section of air reactor is required. The comparison between the helium concentrations in the air reactor to at the cyclone outlet can indicate 65 the amount of helium which has leaked into the cyclone through the route of fuel reactor - lower downcomer –upper downcomer - upper section of downocomer 2 - cyclone. Figure 5.6 Possible route for gas leakage from fuel reactor to cyclone 5.2.2 Helium concentration fluctuation in air reactor and after-cyclones Figure 5.7 shows the typical fluctuations of helium concentration in the air reactor and at the exit of cyclone in 15 minutes time span at Ua0 = 2.5 m/s, where the difference in concentration is minimal. This indicates that very limited helium is transferred from the fuel reactor to the cyclone through the route of fuel reactor - lower downcomer - upper downcomer - upper section of primary downcomer - cyclone. As described in Chapter 4, the bottom end of the lower downcomer in the fuel reactor has been modified as shown in Figure 4.3, resulting in the lower downcomer working as an internal loop-seal to prevent the gas leakage. Figure 5.8 presents the helium concentration fluctuation in the air reactor and at the exit of cyclone prior to the modification of the lower downcomer, at Ua0 = 2.5 m/s. The high helium concentration difference between the two locations reveals that some gas leaked from the fuel 66 reactor into the cyclone through the lower downcomer. Therefore, the internal loop-seal in fuel reactor resulting from the sealing of the bottom end of the lower downcomer is very important to avoid gas leakage from the fuel reactor to the cyclone. 0 2 4 6 8 10 12 14 16 18 0 50 100 150 Time (minute) H el iu m co nc en tra tio n (p pm ) Helium concentration in air reactor Helium concentration at exit of cyclone Figure 5.7 Helium concentration fluctuation in air reactor and at the gas outlet of cyclone (Ua0=2.5 m/s, U f0=0.029 m/s, UA1=2Umf, UA2=1Umf) 0 3 6 9 12 15 18 0 50 100 150 200 250 300 350 Time (minute) H e liu m co n ce nt ra tio n (p pm ) Helium concentration in air reactor Helium concentration at exit of cyclone Figure 5.8 Helium concentration fluctuation in air reactor and at the gas outlet of cyclone when the bottom end of the lower downcomer is opened completely (Ua0=2.5 m/s, Uf0=0.029 m/s, UA1=2Umf, UA2=1 Umf) 67 5.3 Gas leakage from air reactor to fuel reactor 5.3.1 Approaches to analyze gas leakage The possible routes for gas leakage from air reactor to fuel reactor are shown in Figure 5.9. The first route is the gas in air reactor to leak into the fuel reactor through loop-seal 1, and the second route is the gas in the air reactor to follow the solids stream from the air reactor to the fuel reactor flowing through the route of air reactor –cyclone - upper downcomer - lower downcomer - fuel reactor to reach the fuel reactor. Figure 5.9 Possible routes for gas leakage from air reactor to fuel reactor In order to determine the gas leakage through one specific route, as indicated in Figure 5.9, helium was added to the air reactor at two locations: gas inlet (A); and the middle section of the air reactor (B), while simultaneously measuring the helium concentration in the fuel reactor. When 68 helium is injected at point B (Figure 5.9), the helium detected in the fuel reactor is only from the gas flowing through the cyclone. If helium is added at point A (Figure 5.9), helium in the fuel reactor would be the sum of the helium leaked through both of the possible routes mentioned above. Thus, comparing the helium concentrations obtained for the above meanings of injecting helium each other, the specific route for gas leaking from air reactor to fuel reactor could be determined. 5.3.2 Gas leakage through loop-seal 1 Figure 5.10 shows the typical helium concentration changes in the fuel reactor when the helium is injected into the air reactor at points A and B, separately. Very little difference in helium detection is shown indicating that there is minimal gas leakage from the air reactor to the fuel reactor through loop-seal 1. Thus, the gas from the air reactor leaks into the fuel reactor only through the route of air reactor –cyclone - upper downcomer - lower downcomer - fuel reactor. 2 4 6 8 10 12 14 16 0 50 100 150 Time (minute) H e liu m co n ce nt ra tio n in fu el re ac to r( p pm ) Inject Helium at A point Inject Helium at B point Figure 5.10 Comparison of helium concentrations in fuel reactor when inject helium at A and B (Ua0=3.0 m/s, Uf0=0.035 m/s, UA1=4Umf, UA2=1Umf) 5.3.3 Effect of solids circulation flux between air reactor and fuel reactor Figure 5.11 presents the typical curve for the effect of solids circulation flux between air reactor and fuel reactor on the gas leakage from the air to fuel reactors for Ua0 = 2.5 m/s. The plots for Ua0 = 3.0, 4.0 and 5.0 m/s are shown in Figures D.7 ~ D.9 in Appendix D. For the cases where the gas leakage is less than 0.57%, the gas leakage from the air to fuel reactors is related linearly with the solids circulation flux between air reactor and fuel reactor for a given Ua0. However, after 69 the gas leakage increased to a maximum value of 0.57%, it did not increase further with increasing of solids circulation flux between air reactor and fuel reactor. Thus, the gas leakage from air reactor to fuel reactor is a consequence of the particles carrying gas into fuel reactor as they flow down into fuel reactor through the lower downcomer, but the maximum gas leakage is only up to 0.57%. The gas leakage measurement for all of the operating conditions varied in the range of 0.1% ~ 0.57%, with less fluctuation at higher superficial gas velocity. Increasing the superficial gas velocity from 2.5 m/s to 5.0 m/s results in decrease of fluctuation range from 0.1 ~ 0.57% to 0.25 ~ 0.38%. 5 10 15 20 0 0.1 0.2 0.3 0.4 0.5 0.6 Solids circulation flux between air reactor and fuel reactor (kg/m2 .s) G as le ak ag e fro m a ir re ac to rt o fu el re ac to r (% ) UA1=1Um f UA1=2Um f U A1 =4U m f U A1 =6U m f Figure 5.11 Gas leakage from air reactor to fuel reactor versus solids circulation flux between air reactor and fuel reactor (Ua0=2.5 m/s, U f0=0.029 m/s) 5.3.4 Effect of aeration velocity in loop-seal 2 The typical curves reflecting the effects of UA2 on the gas leakage from air reactor to fuel reactor for the superficial gas velocity of 3.0 m/s are shown in Figures 5.12. For a given Ua0 , gas leakage from air reactor to fuel reactor decreases with the increasing of UA2 if UA1 is fixed. This is due to the decreasing solids circulation flux between the air reactor and fuel reactor with increasing UA2. When the solids circulation flux is decreased, the gas entrained by the solids circulation flux into the fuel reactor decreases correspondingly. 70 -0.5 0 0.5 1 1.5 2 0 0.1 0.2 0.3 0.4 0.5 Aeration velocity in loopseal2, U A2 /U mf (-)) G as le ak ag e fro m ai rr ea ct or to fu el re ac to r (% ) UA1=1Umf U A1 =2U mf UA1=4Umf U A1 =6U mf Figure 5.12 Leakage from fuel reactor to air reactor versus aeration velocity in loop-seal 2, UA2 (Ua0=3.0 m/s, Uf0=0.035 m/s) 5.3.5 Effect of aeration velocity in loop-seal 1 As represented in Figures 5.13, increasing UA1 increases the gas leakage from air reactor to fuel reactor at a constant Ua0. This is due to the increase in the solids circulation flux between the air reactor and fuel reactor with the increase in the UA1. 0 1 2 3 4 5 6 7 0 0.1 0.2 0.3 0.4 0.5 Aeration velocity for loopseal1, U A1 /U mf (-) G as le ak ag e fro m a ir re ac to r to fu e lr e ac to r (% ) UA2=1Um f U A2 =1U m f U A2 =1.5U mf Figure 5.13 Leakage from fuel reactor to air reactor versus aeration velocity in loop-seal 1, UA1 (Ua0=3.0 m/s, Uf0=0.035 m/s) 71 5.3.6 Effect of filter blockage As shown in Figure 5.14, there are two routes for the gas to go after the primary cyclone coming out of the air reactor. It is expected that most of the gas goes through the secondary cyclone and the filter before being released to the atmosphere; however, some small amount of gas can follow the particles flowing downward from primary cyclone to the fuel reactor. This amount is shown to be < 0.6% as discussed above. Figure 5.14 Effect of blockage in filter on the gas leakage from air reactor to fuel reactor However, over the course of long experimental runs, the fine particles accumulating in the filter affect the pressure balance in the system. The pressure in the pipe connected to the filter (shown by the dash arrow) has increased while the pressure in the downcomers connected to the fuel reactor (shown by the solid arrow) has not because the internal cyclone in the fuel reactor is open to 72 the atmosphere. As a result, more gas from the air reactor will leak into the fuel reactor along the route of upper section of primary downcomer - upper downcomer - lower downcomer - fuel reactor. As shown in Figure 5.15, the gas leakage measurement for 100 hours after the filter clean-up (> 4%) compared to 8 hours (< 0.5%) is extremely large. Therefore, the effect of particles accumulated in the filter is very significant. In order to avoid significant gas leakage from the air reactor to the fuel reactor through upper downcomer and the lower downcomer, the particles in the filter should be removed after the experimental system run for 8 hours to make sure there is no blockage in the filter. 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 Aeration velocity for loopseal1, U A1 /U mf (-) G as le a ka g e fro m a ir re a ct o rt o fu el re a ct o r (% ) 8 hours 100 hours Figure 5.15 Comparison of gas leakage from air reactor to fuel reactor between no blockage in filter and blockage in filter (Ua0=4.0 m/s, Uf0=0.046 m/s, UA2=1.5Umf) 5.4 Gas leakage from aeration gas of loop-seal 1 to fuel reactor 5.4.1 Approaches in analyzing gas leakage As shown in Figure 5.16, there are four ports to inject the aeration gas into the loop-seal 1 for the fluidization of solids in loop-seal 1, i.e., F1, F2, F3 and F4. In order to investigate the possible gas leakage from each of the port to the fuel reactor, helium tracer gas was added into the loop-seal 1 as a part of aeration gas through F1, F2, F3 and F4, respectively, while the helium concentration in fuel reactor was measured to determine the fraction of helium escaping to the fuel reactor. 73 Figure 5.16 Helium injection ports for measurement of gas leakage from loop-seal 1 to fuel reactor 5.4.2 From F1 and F2 ports to fuel reactor The typical curves for the gas leakage from F1 and F2 ports to fuel reactor when Ua0 = 2.5 m/s are shown in Figures 5.17 and 5.18, respectively. The measurement results for Ua0 of 3.0-5.0 m/s are shown in the Figures D.10 ~ D.15 in Appendix D. The figures indicate that when UA1 = 1Umf, the gas leakage from port F1 or F2 remains < 0.5%. This is explained by the aeration gas fixed at minimum fluidization velocity. When the aeration velocity is equal to the minimum fluidization velocity, the vertical section of loop-seal 1 connected with fuel reactor was filled with particles. The particles accumulated in this vertical section with higher height prevent the gas from diffusing into the fuel reactor. Furthermore, for UA1 > 2Umf, the gas leakage from both of the ports to the fuel reactor 74 increases greatly. This is due to having a high enough aeration velocity in the vertical section of loop-seal 1 causing fluidization gas to escape to the fuel reactor. Because the directions of solids stream and aeration gas flow in the vertical section of loop- seal 1 connected with fuel reactor are totally opposite to each other, the higher solids circulation flux can help prevent the aeration gas from leaking into the fuel reactor. However, when UA1 is held constant, the solids circulation flux between the fuel reactor and air reactor through loop-seal 1 is reduced with increasing UA2. Thus, as shown in Figures 5.17 and 5.18, the gas leakage from both F1 and F2 will increase with increasing UA2. Comparing the Figures 5.17 with 5.18, it can be seen that the gas leakage from F1 is more than from F2 at the same operating condition, because the pressure drop between port F1 and fuel reactor is less than that between port F2 and fuel reactor. -0.5 0 0.5 1 1.5 2 0 10 20 30 40 50 60 70 80 90 100 Aeration velocity in loopseal2, U A2 /U mf (-)) G as le ak ag e fro m F 1 po rt to fu el re ac to r (% ) U A1 =1U mf U A1 =2U mf U A1 =4U mf U A1 =6U mf Figure 5.17 Aeration velocity in loop-seal 2, UA2, versus gas leakage from F1 port to fuel reactor (Ua0=2.5 m/s, Uf0=0.029 m/s) 75 -0.5 0 0.5 1 1.5 2 0 10 20 30 40 50 60 70 80 90 100 Aeration velocity in loopseal2, UA2 /Umf (-)) G as le ak ag e fro m F1 po rt to fu el re ac to r (% ) UA1=1Umf U A1 =2U mf U A1 =4U mf U A1 =6U mf Figure 5.18 Aeration velocity in loop-seal 2, UA2, versus gas leakage from F2 port to fuel reactor (Ua0=2.5 m/s, Uf0=0.029 m/s) 5.4.3 From F3 and F4 ports to fuel reactor No measurable leakage can be detected in the fuel reactor when helium was injected into the loop-seal 1 through F3 and F4 ports. This is because the horizontal section of loop-seal 1 separates the aeration gas added from F3 and F4 from leaking into the fuel reactor. This result also confirms that there is no gas leakage from the air reactor to fuel reactor through loop-seal 1. Since helium injected at F3 and F4 in loop-seal 1 is not transferred into the flue reactor, it is not possible for helium in the air reactor to leak into fuel reactor through loop-seal 1. Thus for the CLC hot model, steam can be used as aeration gas for F1 and F2 instead of air to avoid the leakage of air from loop-seal 1 to fuel reactor. But for F3 and F4, air can be still used as the aeration gas. 5.5 Gas leakage from aeration gas of loop-seal 2 to fuel reactor As shown in Figure 5.19, ports F5 and F6 are used to inject aeration gas for loop-seal 2. In order to investigate the possible gas leakage from these ports to the fuel reactor, helium is injected through the ports into the loop-seal 2, while measuring the concentration in the fuel reactor. The experiments showed that no helium can be detected in the fuel reactor. The only possible route for 76 gas leakage from ports F5 and F6 is the route of primary downcomer - upper downcomer - lower downcomer. However, particles collected in the primary downcomer can prevent the helium from leaking into the fuel reactor completely. Thus, no gas leakage from aeration gas of loop-seal 2 to fuel reactor is expected. Figure 5.19 Helium injection ports for measurement of gas leakage from loop-seal 1 to fuel reactor 5.6 Summary  Gas leakage from the fuel reactor to the air reactor The gas leakage from the fuel reactor to air reactor was found to be between 0.8% and 3.0% resulting from a combination of the pressure drop across the loop-seal 1, the solids stream through loop-seal 1 and the aeration gas in loop-seal 1. Higher positive pressure drop between the solids outlet and inlet of the loop-seal 1 and higher aeration velocity would reduce the gas leakage; 77 whereas, the solids stream with higher solids circualtion flux through the loop-seal 1 would carry more gas into the air reactor from the fuel reactor.  Gas leakage from the fuel reactor to the cyclone No measurable gas leakage from the fuel reactor to the cyclone could be detected, because the sealing of the bottom end of the lower downcomer prevented gas from leaking into the lower downcomer from the fuel reactor.  Gas leakage from the air reactor to the fuel reactor The gas leakage from the air reactor to the fuel reactor was 0.1% ~ 0.57%.  Gas leakage from the loop-seal 1 to the fuel reactor There were four ports for feeding aeration gas to the loop-seal 1. The gas leakage from the two ports near the fuel reactor was considerable, while no gas leakage could be detected from the other two ports near the air reactor.  Good operating conditions for minimum gas leakage in the cold-flow model As discussed above, the main gas leakage occurred between the fuel reactor and the air reactor through the loop-seal 1. According to Figures 5.5 and D4 to D6 in Appendix D, with UA1 = 4Umf or 6Umf and UA2 = 1.5Umf, the gas leakage between the fuel reactor and the air reactor were minimum at 0.65%, 1.25% and 1.5% for Ua0 varying at 2.5, 3.0 and 4.0 m/s, respectively. Thus, the good operating conditions for minimizing gas leakage could be chosen as follows:  UA1 = 4Umf or 6Umf  UA2 = 1.5Umf  Ua0 = 2.5, 3.0 and 4.0 m/s  Uf0 = 0.028, 0.034 and 0.045 m/s 78 CHAPTER 6 SCALING CONSIDERATIONS In order to achieve hydrodynamic similarity between a hot unit and the cold unit of CLC reactor, certain scaling rules are applied. Glicksman (1984) presented a full set of scaling relationships for fluidized beds. Based on the governing equations of bubble and interstitial gas dynamics, Horio et al. (1986) developed a simpler similarity rule. Glicksman (1988) showed that the parameters of Horio et al. (1986) are equivalent to his reduced governing parameters under the condition of viscous limit. Horio et al. (1989) also developed another scaling law for circulating fluidized beds shown in Set 6.1. 0 0 2 0 ,,, U G U U gD U s s tf s   (6.1) Glicksman et al. (1993) further explored a new set of simplified scaling laws shown in Set 6.2. ,,,,,, 0 0 2 0    D H U G U U gD U s s mff s bed geometry, particle size distribution (6.2) And the simplified scaling law was verified through hydrodynamic scaling experiments in a CFB. The advantage of the simplified scaling laws is the increased flexibility in the design of a model to simulate a chemical reactor owing to the less restrictive conditions. However, the experiments by Horio et al. (1988) indicated that the flow in the top half of the riser is not affected significantly by a change of height for a riser with higher H/D. Furthermore according to van der Meer et al. (1999) the most important aspect of geometrical similarity is the design of the exit section. This implied that the requirement of H/D may be relaxed. Thus, following dimensionless groups shown in Set 6.3 will be used in this study as scaling law.   ,,,, 0 0 2 0 U G U U gD U s s mff s , geometrical similarity, particle size distribution (6.3) 6.1 Equations for scaling In the cold unit experiments, spent FCC particles were fluidized by a gas mixture of helium and air with different helium concentrations. The gases were recycled in the system shown in Figure 3.5, with the gas temperature in the reactor system raised to around 40°C due to heat generated from 79 the blower. From the operating conditions, the properties of gas and particles, i.e., ρfc, μfc, ρsc, dpc, in the cold unit were determined. As a scaling law, the dimensionless parameters listed in set 6.3 must have the same values for both cold and hot units. By matching these parameters, it is hoped that the hydrodynamics of the hot unit can be simulated by the cold unit experiments. For the simplified scaling relationships the gas-to-solid density ratio, ρs/ρf, in the hot and cold units must be kept identical as shown by Eq. 6.4 [Glicksman et al. 1998]. hs f cs f            (6.4) The particle density for the hot unit is determined based on the currently available oxygen carrier for the chemical looping combustion. On the other hand, the spent FCC particles were used in the cold-flow experiments. Thus, the gas density ratio between hot unit and cold unit can be determined by Eq. 6.5. sc sh fc fh      (6.5) So the gas density, ρfc, and pressure Ph , in the hot unit can be calculated by Eqs. 6.6 and 6.7 with applying the ideal gas law, respectively. fc sc sh fh    (6.6) 1 2730  h fh fh h TP   (6.7) where ρfh0 is the density of gas in the hot unit at standard status (1bar and 273K) and Th is the temperature in the hot unit. In general, minimum fluidization velocity, Umf, is a function of the properties of particles and gas. According to Yang (2003), for 310Ar , Umf can be calculated:   2)( 00075.0 pfsmf gd U  (6.8) Since the change in fluidizing gas density results in a change in Archimedes numbers ranging from 4.31 to 19.35, the ratio of Umf for the cold and hot units is expressed: h c pc ph sc sh h c pc ph fcsc fhsh mfc mfh d d d d U U         22 )()(     (6.9) According to Set 6.3, the Froude number and ratio of U0 to Umf for both units must also be constant: 80 2 1 0 0 2 0 2 0               c h c h hc D D U U gD U gD U (6.10)                mfc mfh c h hmfcmf U U U U U U U U 0 000 (6.11) Combining Eqs. 6.9, 6.10 and 6.11 and rearranging: 422             pc ph h c sc sh c h d d D D     (6.12) Thus, there is one free parameter: particle size in hot unit or diameter of hot unit reactor, i.e., dph or Dh . Three possible solutions can be applied for Eq. 6.12 to determine the value of dph or Dh, i.e.,  Solution A: Given a particle size in hot unit, dph , then calculate the needed diameter of hot unit reactor, Dh, can be calculated from Eq. 6.12.  Solution B: Given a diameter of hot unit reactor, Dh, then calculated the needed particle size, dph, can be calculated from Eq. 6.12.  Solution C: Apply the scaling law [Glicksman1984, Glicksman et al. 1990] with one extra dimensionless group, i.e., D d p , and keeping h ph c pc D d D d  for the hot and cold units, then Eq. 6.12 can be rewritten as: 3 2     c h sh sc pc ph c h d d D D     (6.13) Next, the relationship between the superficial gas velocities in the hot and cold units can be determined by Eq. 6.10, and the solids circulation flux can be calculated by following Eq. 6.14.        hs s cs s U G U G 00            c h sc sh sc sh U U G G 0 0   (6.14) In the following discussions, Solutions A and B will be used for scaling of cold unit with fluidizing gas of mixture of 96 vol% helium and 4 vol% air, while Solution C will be used for scaling the cold unit with fluidizing gas of mixture of air and helium with concentration varying from 0 vol% to 96 vol%. Based on the research on the different types of oxygen carriers (Table A.1), the particle density in the hot unit is set to be 2800kg/m3, and the temperature in the air and fuel reactors are set at 950 and 900°C, respectively. In addition both the cold and hot units should have identical dimensionless particle size distributions and sphericity. 81 6.2 Scaling of cold unit for a given particle size According to Table A.1, the particle size of the typical oxygen carriers range between 100 to 150 µm. Therefore, the particle size of 125 µm was chosen in the scaling studies. Table 6.1 gives the values of design and operating parameters of a hot unit, which can be simulated by the cold unit fluidized with gas mixture of 96 vol% helium and 4 vol% air when dph is pre-determined as 125 µm. In order to maintain the ρs/ρf constant, the hot unit should be run at pressure of 1.17 bar, and the diameter of the hot unit should be 3.62 times that of the cold unit. Table 6.2 shows the comparison of the dimensionless groups between cold unit and hot unit. Note that the additional dimensionless group, i.e., ratio of reactor diameter to particle size, is also shown in the tables, even though it is not included in the scaling law (Set 6.3). Table 6.2 shows that four dimensionless groups, except D/dp in Set 6.3 can be matched, since the particle size has been pre-determined. The gas in the hot fuel reactor is a mixture of 25 vol% CH4, 25 vol% CO2 and 50 vol% H2O based on the assumption that 50% of CH4 has been converted to CO2 and H2O. The gas species in the fuel reactor are different from that in the air reactor, and parameters such as gas density, viscosity and minimum fluidization velocity which are related with the gas species will change between the reactors, while the particles and pressure in the fuel reactor have to be kept the same as that in air reactor. Thus, as shown in Table 6.2, the dimensionless parameters for the fuel reactor between cold and hot unit cannot be completely matched. However, the ratios of the dimensionless parameters are reasonably close to 1.0. Therefore, it is expected that the hydrodynamic similarity of fuel reactor between the cold and the proposed hot units can approximately be achieved. 82 Table 6.1 Hot unit modeled by a cold unit fluidized with 96 vol% helium and 4 vol% air at temperature of 40 oC and atmosphere pressure (based on a pre-determined particle size in hot unit) Air Reactor Given Parameter Cold Unit Hot Unit Gas type 96 vol% helium + 4 vol% Air Air Temperature (oC) 40 950 Viscosity of gas (Pa.s) 2.04×10-5 4.94×10-5 Particle density s (kg/m3) 1560 2800* Particle diameter pd ( m ) 78 125 Gas density f (kg/m3) fc (0.19) fc79.1 (0.34) Pressure (bar) cP (1) cP17.1 (1.17) Bed diameter D (mm) cD (102) cD62.3 (369) Superficial gas velocity cU 0 cU090.1 Solids circulation flux scG scG41.3 Minimum fluidization velocity mfU (m/s) 0.0034 0.0065 Fuel Reactor Gas type 96 vol% helium + 4 vol% Air 25 vol% CH4+ 25 vol%CO2 +50 vol% H2O Temperature (oC) 40 900 Viscosity of gas (Pa.s) 2.04×10-5 4.03×10-5 Particle density s (kg/m3) 1560 2800* Pressure (bar) 1 1.17 Gas density f (kg/m3) 0.19 0.29 Particle diameter pd ( m ) 78 125 Bed diameter D (mm) 286 1035 Superficial gas velocity cU 0 cU 090.1 Solids circulation flux scG scG41.3 Minimum fluidization velocity mfU (m/s) 0.0034 0.0080 * Referred to Gayán et al. 2009 83 Table 6.2 Dimensionless group comparison for operation conditions in Table 6.1 Air reactor Dimensionless parameter Cold Unit Hot Unit Ratio between cold unit and hot unit gD U 20 2000.1 cU 2 000.1 cU 1.00 f s   8210 8235 1.00 mfU U 0 cU 0294 cU 0292 1.00 0U G s s  c sc U G 0 4-106.41 c sc U G 0 4-106.41 1.00 pd D 1308 2952 0.44 Fuel reactor gD U 20 2 036.0 cU 2 036.0 cU 1.00 f s   8210 9655 0.85 mfU U 0 cU 0294 cU 0238 1.24 0U G s s  c sc U G 0 4-106.41 c sc U G 0 4-106.41 1.00 pd D 3667 8280 0.44 6.3 Scaling of cold unit for a given bed diameter Table 6.3 shows the values of design and operating parameters of a hot unit which can be simulated by the cold unit fluidized with gas mixture of 96 vol% helium and 4 vol% air when the bed diameter in the hot unit is pre-set. As shown in Tables 6.3 the pre-determined bed diameters of air reactor and fuel reactor in the hot unit are 41 and 115 mm, respectively. Maintaining a constant ρs/ρf will still result in a pressure of 1.17 bars in the hot unit. However, the particle size of oxygen carrier has to be less than that of particles in the cold unit. Tables 6.4 compare the dimensionless groups between the cold unit and hot unit. Similarly, the dimensionless groups shown in Set 6.3 are satisfied well for the air reactor, but there are 84 discrepancies on the ρs/ρf, and U0/Umf for the fuel reactor, because the gas species in the fuel reactor are different from that in the air reactor. Table 6.3 Hot unit modeled by a cold unit fluidized with 96% helium and 4% air at temperature of 40 oC and atmosphere pressure (based on a given bed diameter in hot unit) Air Reactor Given Parameter Cold Unit Hot Unit Gas type 96% helium + 4% Air Air Temperature (oC) 40 950 Viscosity of gas (Pa.s) 2.04×10-5 4.94×10-5 Particle density s (kg/m3) 1560 2800 Bed diameter D (mm) 102 41 Gas density f (kg/m3) fc (0.19) fc79.1 (0.34) Pressure (bar) cP (1) cP17.1 (1.17) Particle diameter pd ( m ) pcd (78) pcd92.0 (72) Superficial gas velocity cU 0 cU 063.0 Solids circulation flux scG scG13.1 Minimum fluidization velocity mfU (m/s) 0.0034 0.0022 Fuel Reactor Gas type 96% helium + 4% Air 25%CH4+ 25%CO2 +50% H2O Temperature (oC) 40 900 Viscosity of gas (Pa.s) 2.04×10-5 4.03×10-5 Particle density s (kg/m3) 1560 2800 Pressure (bar) 1 1.17 Gas density f (kg/m3) 0.19 0.29 Particle diameter pd ( m ) 78 72 Bed diameter D (mm) 286 115 Superficial gas velocity cU 0 cU 063.0 Solids circulation flux scG scG13.1 Minimum fluidization velocity mfU (m/s) 0.0034 0.0026 85 Table 6.4 Dimensionless group comparison for operation conditions in Table 6.3 Air reactor Dimensionless parameter Cold Unit Hot Unit Ratio between cold unit and hot unit gD U 20 2 000.1 cU 2 099.0 cU 1.01 f s   8210 8235 1.00 mfU U 0 cU 0294 cU 0286 1.03 0U G s s  c sc U G 0 4-106.41 c sc U G 0 4-106.41 1.00 pd D 1308 569 2.30 Fuel reactor gD U 20 2 036.0 cU 2 035.0 cU 1.03 f s   8210 9655 0.85 mfU U 0 cU 0294 cU 0242 1.21 0U G s s  c sc U G 0 4-106.41 c sc U G 0 4-106.41 1.00 pd D 3667 1597 2.30 6.4 Scaling of cold unit for identical ratio of particle size to reactor size Although a good match of parameters between the cold and hot units may be obtained with the four dimensionless groups in Set 6.3, inclusion of an extra dimensionless group of D/dp provides a better hydrodynamic similarity between cold unit and hot unit [Glicksman 1984, Glicksman et al. 1990 and 2003]. Thus, for the cold unit fluidized by the gas mixture of air and helium with concentration varying from 0 vol% to 96 vol%, a set of scaling law comprised of five dimensionless groups is recommended:    ,,,,, 0 0 2 0 ps s mff s d D U G U U gD U , geometrical similarity, particle size distribution (6.15) Based on the discussions in Section 6.2, the linear scaling factor is determined by: 86 3 2     c h sh sc pc ph c h d d D D factorscaleLinear     (6.16) Tables 6.5 shows that the operating conditions of the hot unit which can be simulated by the cold unit with fluidizing gas of 96 vol% helium and 4 vol% air. Correspondingly, Tables 6.6 represents the comparison of dimensionless groups in Set 6.15. Similar calculation for helium concentrations at 80, 60, 40, 20 and 0 vol% are given by Table D.3~d.12 in Appendix D. Table 6.5 Hot unit modeled by a cold unit fluidized with 96 vol% helium and 4 vol% air at temperature of 40 oC and atmosphere pressure (based on identical ratio of particle size to reactor size) Air Reactor Parameter Cold Unit Hot Unit Gas type 96 vol% helium + 4 vol% Air Air Temperature (oC) 40 950 Viscosity of gas (Pa.s) 2.04×10-5 4.94×10-5 Particle density s (kg/m3) 1560 2800 Gas density f (kg/m3) fc (0.19) fc79.1 (0.34) Pressure (bar) cP (1) cP17.1 (1.17) Particle diameter pd ( m ) pcd (78) pcd22.1 (95) Bed diameter (mm) cD (102) cD22.1 (124) Superficial gas velocity cU 0 cU 010.1 Solids circulation flux scG scG98.1 Minimum fluidization velocity mfU (m/s) 0.0034 0.0038 Fuel Reactor Parameter Cold Unit Hot Unit Gas type 96 vol% helium + 4 vol% Air 25 vol%CH4+ 25 vol%CO2 +50 vol% H2O Temperature (oC) 40 900 Viscosity of gas (Pa.s) 2.04×10-5 4.03×10-5 Particle density s (kg/m3) 1560 2800 Pressure (bar) 1 1.17 Gas density f (kg/m3) 0.19 0.29 Particle diameter pd ( m ) 78 95 Bed diameter (mm) cD (286) cD22.1 (349) Superficial gas velocity cU 0 cU 010.1 Solids circulation flux scG scG98.1 Minimum fluidization velocity mfU (m/s) 0.0034 0.0046 87 Table 6.6 shows that the dimensionless parameters provided in Set 6.15 can be matched very well for the air reactor. But for the fuel reactor, because the gas species in the hot unit differ from that of the air reactor, there is about a 20% discrepancy on the two dimensionless groups, i.e., ρs/ρf, and U0/Umf, between the cold and hot units. This means the hydrodynamic similarity of the fuel reactor between cold and hot units cannot be achieved completely. The bubble size and bubble velocity in the cold unit are bigger than in the hot unit. Table 6.6 Dimensionless group comparison for operation conditions in Table 6.5 Air Reactor Dimensionless parameter Cold Unit Hot Unit Ratio between cold unit and hot unit gD U 20 2 000.1 cU 2 099.0 cU 1.01 f s   8210 8235 1.00 mfU U 0 cU 0294 cU 0289 1.02 0U G s s  c sc U G 0 4-106.41 c sc U G 0 4-1040.6  1.00 pd D 1308 1305 1.00 Fuel Reactor gD U 20 2 036.0 cU 2 035.0 cU 1.03 f s   8210 9655 0.85 mfU U 0 cU 0294 cU 0239 1.23 0U G s s  c sc U G 0 4-106.41 c sc U G 0 4-106.41 1.00 pd D 3667 3674 1.00 88 6.5 Summary Using the scaling law, the cold-flow unit operated with a fluidizing gas mixture of air and helium, while the helium concentration is varied from 96 vol% to 0 vol%, can be used to simulate the hydrodynamics of the hot unit with the pressure varying from 1.17 bar to 6.97 bar correspondingly. The analysis focuses on the scaling of a cold unit operated with a fluidizing gas mixture of 96 vol% helium and 4 vol% air, leading to the following conclusions:  If the particle size of oxygen carrier in hot unit is pre-determined as 125 µm, the cold unit can simulate the hydrodynamics of the hot unit with diameter being 3.62 times that of the cold unit, while the four groups of dimensionless parameters in Set 6.3 can be matched well.  When the bed diameters of air reactor and fuel reactor in the hot unit are pre-set as 41 and 115 mm, respectively, the particle size of the oxygen carrier in hot unit should be 72 µm to maintain the constant dimensionless parameters in Set 6.3 between cold unit and hot unit.  In order to provide a better hydrodynamic similarity between cold unit and hot unit, a set of scaling laws comprised of five dimensionless groups shown in Set 6.15 has been introduced for scaling up. It is expected that the hydrodynamics of the hot unit whose dimension is 1.22 times that of the cold unit, can be simulated by the cold unit, while the particle size of the oxygen carrier in hot unit has to be chosen at 97 µm. 89 CHAPTER 7 HYDRODYNAMIC STUDY WITH FLUIDIZING GAS MIXTURE OF HELIUM AND AIR As discussed in Chapter 6, the scaling laws emphasized the importance of keeping ρs/ρf identical for both the cold unit and the hot unit in order to avoid significant discrepancies in the hydrodynamic similarity [Horio et al. 1989, Glicksman et al. 1993, 1998]. In this study, consideration was given to keep the ρs/ρf similar between the cold and the hot units. However, the value of ρs/ρf for an atmospheric CLC hot unit is relatively large: the density of currently popular oxygen carrier (Ni-based) is around ρs=2800~4500 kg/m3 (refer to Table A.1) with typical ρf=0.29~0.36 kg/m3 giving ρs/ρf at 10000~15500. On the other hand, for a cold model using spent FCC with air, ρs/ρf remains around 1281. Thus, it is very difficult to use air and particles with density of less than 10,000 kg/m3 in cold-flow model to simulate the hydrodynamics of an atmospheric hot unit. In recently published hydrodynamic studies, the CLC cold-flow model experiments were conducted with air at ambient temperature and pressure [Johansson et al. 2002, Pröll et al. 2009a]. For keeping the density ratio constant, their results can only be scaled to predict a pressurized hot unit. Other studies were conducted using a mixture of helium and nitrogen as the fluidization gas for keeping the density ratio comparable with the hot CLC unit [Kronberger et al. 2004, Kronberger et al. 2005] in a small-scale CLC cold model (refer to Table A.2) with Ua0=0.26~1.5 m/s and solids circulation of 1 ~ 29 g/s, which is far from the typical operating conditions in a hot CLC unit. In this study, various concentrations of helium were added to the fluidizing gas to run the CLC cold unit for obtaining hydrodynamic data relevant to scaling the hot CLC unit. 7.1 Experimental procedure The schematic of the experimental system is shown in Figure 3.5. The required maximum flow rate of helium is around 136.4 Nm3/hr and one helium cylinder can last only for a few minutes at the above flow rate for an open system. Thus, the unit was modified to recycle the fluidizing gas in a closed experimental system through a blower as shown in Figure 3.5. However, there was still some gas leakage between the gas blower and atmosphere. In order to compensate for the gas loss from the blower, fresh helium was added to the recycle stream. A Micro-GC (Varian Inc.) was used to constantly monitor the helium concentration at the gas inlets to the reactor. The cold-flow model 90 was operated with spent FCC particles with density of 1560 kg/m3 and total solid inventory of 92 kg at atmosphere pressure and temperature of 40oC. The temperature in the reactors was higher than the room temperature because of the heat generated by the blower. The helium concentration in the gas mixture was maintained at 0, 20, 40, 60, 80 and 96 vol%. The superficial gas velocity in the air reactor: Ua0 varied from 2.5 to 8 m/s; and the corresponding superficial gas velocity in the fuel reactor: Uf0 was adjusted from 0.028 to 0.090 m/s in order to achieve a constant input volumetric flow rate ratio of gases to the air reactor and fuel reactor. For this study, the ratio was fixed at 11 simulating the excess air coefficient of 1.1. As shown in Figure 3.6, pressure transducers were used to measure the pressure drop along the loops in order to obtain the solids hold-up and the axial distribution of solids in the air reactor. The solids circulation flux was measured by closing the butterfly valves indicated in Figure 3.6. The aeration velocity of loop-seal 1, UA1, was fixed at 4Umf, and the aeration velocity of loop-seal 2, UA2, varied between 0Umf, 1Umf, and 1.5Umf. 7.2 Effect of helium concentration on the pressure According to Stearns et al. (1951), the volumetric gas flow rate measured by an orifice plate flowmeter can be calculated by f PDKQ   2)4/( 200 (7.1) where ΔP is the pressure drop across the orifice, K0 is the orifice discharge coefficient, D0 is the diameter of orifice, ρf is the gas density. Thus, the volumetric flow rate of the mixture of helium and air in the experiments can be shown as: )()1( 2 )4/( 200 YY P DKQ heliumair     (7.2) and the superficial gas velocity can be shown as: S Q U 0 (7.3) where ρair and ρhelium are the densities of air and helium, respectively, Y is the helium concentration (molar fraction), and S is the cross-sectional area of one column. Figure 7.1 shows the pressure variations at the bottom of the air reactor, before the orifice, and the pressure drop across the orifice for varying helium concentrations in the gas mixture while the volumetric flow rate of gas was kept at around 136.4 Nm3/hr. Note that the pressures and pressure 91 drop are decreased with increasing helium concentration, even when the volumetric flow rate was kept constant. 0 10 20 30 40 50 60 70 80 90 0 1 2 3 4 5 6 7 8 9 Helium concentration (mol%) P re ss ur e (K P a) gauge pressure at the bottom of the air reactor gauge pressure before the orifice pressure drop across the orifice Figure 7.1 Pressures at the bottom of the air reactor, before the orifice and across the orifice with increasing the helium concentration in the gas mixture, gas volumetric flow rate = 136.4 Nm3/hr 7.3 Minimum superficial gas velocity for stable particle transportation For a specific UA2, there is a minimum superficial gas velocity in the air reactor, Umstable, to achieve stable particle transportation in the system when helium concentration is fixed. If Ua0 < Umstable, then the solids mass flow rate from the primary downcomer to the air reactor is higher than the maximum carrying capacity of the gas-solid suspension. This means increasing the amount of particles accumulated in the air reactor until the pressure balance in the whole cold-flow model cannot be achieved. Table 7.1 show the values of Umstable for different helium concentrations and UA2. It is found that Umstable increases with increasing helium concentration. Thus, in the following experiments, Ua0 was varied in the range of 2.5 ~ 8 m/s (shown in the Table 7.1) for different helium concentrations. 92 Table 7.1 Summary of Umstable and operating conditions of the experiments Helium concentration (mol%) Umstable ( when UA2=1Umf) Umstable ( when UA2=1.5Umf) Ua0 in the experiments 0% 1.81 m/s 2.27 m/s 2.5, 3, 4, 5 m/s 20% 1.99 m/s 2.50 m/s 3, 4, 5 m/s 40% 2.24 m/s 2.80 m/s 3, 4, 5 m/s 60% 2.60 m/s 3.27 m/s 4, 5 m/s 80% 3.25 m/s 4.07 m/s 5, 6, 7 m/s 96% 4.36 m/s 5.46 m/s 6, 7, 8 m/s 7.4 Pressure profile As discussed in Section 4.2, there are two loops in the cold-flow model: loop 1 of D2-A2-B9- B8-B7-B6-B2-D6-D3-D2, and loop 2 of A2-B9-B8-B7-B6-B5-D5-D4-A2 (shown Figure 3.6). The typical pressure profile of loop 1, with the helium concentration varied from 0 vol% to 80 vol% (Ua0=5 m/s, UA1=4Umf, UA2=0Umf) is shown in Figure 7.2. Figures D.16~D19 in the Appendix D show additional pressure profiles of loop 1 and loop 2 as the function of helium concentration when Ua0=5 m/s, UA1=4Umf, UA2=1~1.5Umf. The figures indicate a decrease in pressure in the cold-flow model with increasing helium concentration. 1 2 3 4 5 6 7 8 9 10 11 0 1000 2000 3000 4000 5000 6000 Pressure P-Patm osphere (kPa) H ei gh ta bo ve th e ga s di st rib ut or of ai rr ea ct or (m m ) Helium concentration=0% Helium concentraton=20% Helium concentration=40% Helium concentration=60% Helium concentration=80% B9 B8 B7 B2 B6 A2 D3 D2 D6 Figure 7.2 Pressure loop 1 versus helium concentration, Ua0=5.0 m/s, UA1=4Umf, UA2=0Umf (The positions of the measurement points are shown in Figure 3.6) 93 7.5 Axial solids hold-up in air reactor Figure 7.3 shows the typical axial solids hold-up distributions in the air reactor for different helium concentrations, with Ua0=5.0 m/s, UA1=4Umf andUA2=1.5Umf. Figures D.20 ~ 21 in Appendix D show additional axial solids hold-up distributions in the air reactor for Ua0 =5.0 m/s, UA1=4Umf and UA2=0~1Umf. Because lower part of the air reactor (Z=0~500 mm) is connected with loop-seal 1 and loop-seal 2, there is a dense zone with strong backmixing and splashing of solids caused by the input particles from loop-seal 1 and loop-seal 2. Above the dense zone, the solids hold-up decreases along the height of the air reactor. These figures indicate that the solids hold-up in the air reactor increased with increasing helium concentrations. This is because an increase in helium concentration decreases the density of the fluidizing gas while proportionately reducing the drag force on the particles, and the maximum carrying capacity of gas-solid suspension in the air reactor is also decreased. Thus, more particles accumulate in the air reactor. The accumulation of particles was observed at Z= 0~500 mm, when the loop-seal 2 transferred the particles into the air reactor. This means that there are more bed materials in the air reactor to react with oxygen, leading to higher oxygen conversion efficiency. 0 5 10 15 20 25 0 1000 2000 3000 4000 5000 6000 Cross sectional average solids hold-up (%) H ei gh ta bo ve th e ga s di st rib ut or of ai rr ea ct or (m m ) Helium concentration=0% Helium concentration=20% Helium concentration=40% Helium concentration=60% Helium concentration=80% Figure 7.3 Solids hold-up distribution in air reactor at different helium concentration (Ua0=5.0 m/s, UA1=4Umf, UA2=1.5Umf) 94 7.6 Solids circulation flux As discussed in Section 7.3, in order to achieve stable particle transportation in the system, Ua0 must be more than Umstable for each helium concentration. Thus, Ua0 varied for different helium concentrations, as shown in the Table 7.1, while UA1=4Umf and UA2=0~1.5Umf. 7.6.1 Effect of helium concentration on solids circulation flux in loop 1 The solids flow between the fuel reactor and the air reactor in loop 1 has been shown in Figure 4.3, the solids are transferred in loop 1 through lower downcomer - fuel reactor - loop-seal 1 - air reactor. The achieved solids circulation flux is determined by the combination of maximum carrying capacity of gas-solid suspension of the gas in the air reactor, pressure head of the loop 1 (pressure drop between D6 and D3 as shown in Figure 4.3) and the solids flowing through the hole (referred to 10 in Figure 2.13) in the lower downcomer. The maximum carrying capacity of the gas-solid suspension is determined by the Ua0 and gas density, i.e., higher Ua0 and gas density result in higher maximum carrying capacity of gas-solid suspension in the air reactor. However, higher Ua0 and gas density will increase the pressure at point D3 and decrease pressure head of the loop. The pressure at point D6 is mainly decided by the height of accumulated particles in the loop-seal 1, i.e., H1 (shown in Figure 4.3), and these accumulated particles are from the solids stream which is transferred from the lower downcomer to the fuel reactor. On the other hand, as shown in Figure 4.3, in order to decrease the gas leakage from fuel reactor to the lower downcomer, the bottom end of the lower downcomer is sealed, and a hole with diameter of 0.1 m on the side for the solid flow. As a result, the lower downcomer and fuel reactor can work as a loop-seal. The solids mass flow rate from the lower downcomer to the fuel reactor is mainly dependent on the fluidization in the fuel reactor. Better fluidization caused by higher Uf0 and higher gas density in the fuel reactor provides more solids flowing from the lower downcomer to the fuel reactor. It is noted that when U f0>5.0 m/s, the gas density is not the main controlling parameter on the fluidization in fuel reactor, meaning a smaller gas density can still result in better fluidization. Therefore, when Ua0 is fixed, because of the effect of the hole in the lower downcomer, the solids circulation flux in loop 1, Gs1, is not increased monotonically with increasing helium concentration, as shown in Figure 7.4, when UA1=4Umf, UA2=1.5Umf . However, as a basic tendency, Gs1 is shown to increase with increasing of helium concentration. Figures D.22 and D23 in Appendix D show the similar trend of effect of helium concentration on Gs1 when UA1=4Umf, UA2=0~1Umf. 95 0 20 40 60 80 100 0 5 10 15 20 25 Helium concentration (mol%) S ol id ci rc ul at io n flu x in lo op 1 (k g /m2 .s ) Ua0=2.5m/s U a0 =3.0m/s U a0 =4.0m/s Ua0=5.0m/s U a0 =6.0m/s U a0 =7.0m/s Ua0=8.0m/s Figure 7.4 Solids circulation flux in loop 1 versus helium concentration in gas mixture for different superficial gas velocities in air reactor, Ua0 (UA1=4Umf, UA2=1.5Umf) 7.6.2 Effect of helium concentration on solids circulation flux in loop 2 The solids flow between air reactor and primary downcomer in loop 2 has been shown in Figure 4.6. The solids are transferred in the loop 2 through loop-seal 2 and the primary downcomer is acting as a standpipe to transfer solids between the cyclone and air reactor. Similarly, the solids circulation flux in loop 2, Gs2, is determined by the combination of Ua0, pressure head (pressure drop between D5 and D4 shown in Figure 4.6) of loop 2, and UA2. However, because the height of accumulated particles in primary downcomer, i.e., H2 (shown in Figure 4.6), is very high, the pressure at point D5 is large. This means that the pressure drop between D5 and D4 (shown in Figure 4.6) is the main controlling parameter for Gs2. Thus, as shown in Figure 7.5, Gs2 is increased monotonically with increasing helium concentration for a given Ua0 when UA1=4Umf, and UA2=1.5Umf. Meanwhile, the similar phenomenon can be observed in Figure D.24 in Appendix D which shows the effect of helium concentration on the Gs2 when UA1=4Umf, and UA2=1Umf. In particular, comparison between Figure 7.5 and Figure D.24 shows that when UA2=1.5Umf, the increasing degree of Gs2 with increasing of helium concentration is more significant. 96 0 20 40 60 80 100 0 5 10 15 20 25 30 35 40 Helium concentration (mol%) S ol id ci rc ul at io n flu x in lo op 2 (k g/ m2 .s ) U a0 =2.5m/s Ua0=3.0m/s U a0 =4.0m/s Ua0=5.0m/s Ua0=6.0m/s Ua0=7.0m/s Ua0=8.0m/s Figure 7.5 Solids circulation flux in loop 2 versus helium concentration in gas mixture for different superficial gas velocities in air reactor, Ua0 (UA1=4Umf, UA2=1.5Umf) Note that the effect of helium concentration on Gs1 and Gs2 is different for a given Ua0. Figure 7.4 indicates that Gs1 actually does not increase monotonically with increasing helium concentration for a given Ua0. However, according to Figure 7.5, for a given Ua0, Gs2 increases monotonically with increasing helium concentration. This phenomenon can be explained by the difference between loop 1 and loop 2. As shown in Figure 4.3, there are actually two loop-seals in loop 1, i.e., loop-seal 1, and internal loop-seal comprised of the lower downcomer and the fuel reactor. Thus, when UA1 and UA2 are held constant, the solids circulation flux through loop 1 are controlled not only by the pressure head in loop-seal 1, but also by the solids stream which can flow through the hole opened on the wall of the lower downcomer. On the other hand, because there is only one loop-seal, i.e., loop-seal 2, in loop 2, the solids circulation flux through loop 2 is determined only by the pressure head in loop-seal 2. Since higher helium concentration results in higher pressure heads across the loop-seal 1 and loop-seal 2, Gs2 increases monotonically with increasing helium concentration. But this phenomena of monotonic increasing of solids circulation flux with increasing helium concentration was not observed for loop 1 since Gs1 is controlled by both pressure head in loop-seal 1 and the solids stream through the hole opened on the wall of the lower downcomer. 97 7.6.3 Effect of helium concentration on total solids circulation flux Figure 7.6 present the total solids circulation flux, Gst , as a function of helium concentration when UA1=4Umf, and UA2=1.5Umf, and the similar curve is presented by the Figure D.25 in Appendix D for UA1=4Umf, and UA2=1.0Umf. Gst is the sum of the net solids circulation fluxes circulated in loop 1 and loop 2. It represents the solids stream flowing through air reactor and reacting with oxygen in air reactor. Figure 7.6 shows that for a specific superficial gas velocity, the total solids circulation flux increases with increasing helium concentration, as a result of decreasing pressure in the air reactor and the increasing pressure head of the both of the loops. This means more particles can be transferred from the primary downcomer and the fuel reactor to the air reactor. 0 20 40 60 80 100 0 5 10 15 20 25 30 35 40 45 50 55 Helium concentration (mol%) T ot al so lid ci rc ul at io n flu x (k g/ m2 .s ) U a0 =2.5m/s U a0 =3.0m/s Ua0=4.0m/s Ua0=5.0m/s Ua0=6.0m/s U a0 =7.0m/s Ua0=8.0m/s Figure 7.6 Total solids circulation flux versus helium concentration in gas mixture for different superficial gas velocities in air reactor, Ua0 (UA1=4Umf, UA2=1.5Umf) 7.7 Summary With higher helium concentration in the fluidizing gas, the gas density decreases and ρs/ρf increases, leading to the following conclusions from the experimental investigation: 98  Minimum superficial gas velocity for stable particle transportation For a specific UA2, there is a minimum superficial gas velocity in the air reactor to achieve stable particle transportation, Umstable, in the system. It is found that the Umstable increases with increasing ρs/ρf.  Pressure profile The pressure in the cold-flow model decreases as the helium concentration increases.  Vertical solids hold-up distribution The solids hold-up in the air reactor increases with increasing ρs/ρf . For Z=0~500 mm, accumulation of particles is observed for higher density ratios. This can be beneficial to achieve higher oxygen conversion efficiency in the air reactor.  Solids circulation flux Gs1 depends on the combination of the maximum carrying capacity of gas-solid suspension in the air reactor, the pressure head of loop 1 (pressure drop between D6 and D3) and solids stream flowing through the hole (referred to Figure 4.3) in the lower downcomer when UA1 is fixed. Higher ρs/ρf decreases the maximum carrying capacity of the gas-solid suspension in the air reactor, resulting in a higher pressure head. It is found that the mass flow rate of solids stream through the hole in the lower downcomer depends on the fluidization in the fuel reactor. For higher ρs/ρf , higher Uf0 is required to achieve good fluidization in the fuel reactor of the cold unit, because in the cold unit with temperature fixed at 40 oC, higher ρs/ρf,means the increasing of Umf and decreasing of the excess gas velocity. Gs2 increases monotonically with increasing helium concentration, since higher helium concentration results in higher pressure heads across loop-seal 2.  Operation conditions for an atmospheric hot unit According to discussion in Chapter 6, the cold-flow model operated with 96 vol% helium and 4 vol% air can be used to simulate the hydrodynamics of an atmospheric hot unit. Because spent FCC is fluidized by a mixture of helium and air in the cold unit, and a high density oxygen carrier is fluidized by air at high temperature and atmospheric pressure in a hot unit, the solids circulation flux between the hot and cold units can be correlated by a dimensionless group in scaling law, i.e., Gs/ (ρsU0). Thus, as shown in Table 6.5, the solids circulation flux in the hot unit should equal to 1.98 Gsc of the cold unit. 99 The cold-flow model experiments showed that Gs1 was 19.8 ~ 24.1 kg/m 2.s, and Gs2 was 21.5 ~ 25.9 kg/m2.s when Ua0, UA1 and UA2 were 6 ~ 8 m/s, 4Umf and 1Umf, respectively. According to the relationships of the parameters between cold unit and hot unit in Table 6.5, if Ua0 = 6.6 ~ 8.8 m/s in the hot unit, the solids circulation flux in the hot unit can be achieved at: Gs1= 39.2 ~ 47.7 kg/m2.s, Gs2 = 42.4 ~ 51.2 kg/m2.s. and the solids circulation flux in the two loops will be: Fs1= 0.46 ~ 0.61 kg/s, Fs2 = 0.50 ~0.66 kg/s. Based on the 0.107 kg of oxygen per kg of a Ni-based oxygen carrier (Zafar et al. 2007b), the maximum oxygen, which can be transferred from air reactor to fuel reactor by the solids circulation in loop1 of the hot unit, could be up to 49 ~ 65 g/s. On the other hand, when the temperature in the atmospheric hot unit is 950oC and Ua0=6.6 ~ 8.8 m/s, the oxygen in the fluidizing air to the air reactor with excess air coefficient 1.1 is 5.67 ~ 7.56 g/s which is about 10% of the oxygen transport capacity of the CLC hot unit. Thus, there will be sufficient oxygen carrier circulating in the hot CLC unit to achieve high fuel conversion. This result will also be validated by the modelling simulation in Section 8.3.3 of Chapter 8. Moreover, the corresponding input mass flow rate of the CH4 will be 1.42 ~ 1.89 g/s in the hot unit. This means that the cold unit in this study can simulate a hot unit with fuel capacity of 80 ~ 105 kW. 100 CHAPTER 8 MODELING ANALYSIS OF METHANE COMBUSTION The ultimate goal of the CLC is to utilize the energy from fossil fuels while capturing most or all of the CO2 with low energy penalty and high-energy utilization efficiency. Many factors affect fluidized bed reactor performance, including hydrodynamics, heat and mass transfer of interparticles and intraparticles, and complexities of reaction kinetics [Jiang et al. 2003]. The design process of a CLC fluidized bed reactor can be described by considering various factors, as illustrated in Figure 8.1. Thus, the purpose of this chapter is to develop a model for predicting the performance of the CLC reactor, providing valuable data for reactor design and operation, and evaluating the effect of the operating conditions and the change in particle reactivity. Figure 8.1 Schematic diagram of CLC fluidized bed reactor development 8.1 Modeling development In general, if a gas-solid reaction occurring inside a fluidized bed is described mathematically, three subtopics should be considered separately: the flow structure, the solids mixing, and the gas phase dispersion coefficients and chemical reaction kinetics [Luecke et al. 2004]. Furthermore, the particle population balance model is believed to be the best way to describe the particles distribution in a circulating fluidized bed [Jiang et al. 1995, Ramkrishna et al. 2002, Adánez et al. 2003]. In this study, the CLC reactor system is composed of a fast fluidized bed air reactor and a bubbling bed fuel reactor, thus, a suitable model for a CLC interconnected fluidized bed reactor can be developed by 101 combining the flow structure and mixing of the solids and gas phase with the particle population balance for the calculation of particle conversion considering the chemical reaction of each particle. 8.1.1 Particle properties As discussed in Chapter 1, each particle needs two properties to be defined, i.e., oxidization degree, X and particle size, R. In the CLC reactor system, the properties of particles have significant influence on the reactor performance (i.e., air conversion and fuel conversion). Since the particles with different properties will behave differently in the reactors, it is necessary to divide the particles into a number of groups with different oxidization degree and particle sizes. The particle mass distribution function is defined as P(X,R). Therefore the mass fraction of the particles with conversion between X and X+dX and size between R and R+dR can be described by P(X,R)dXdR. Based on the particle mass distribution function, a particle population balance model in the reactor can be developed. Particles for CLC must possess a high mechanical strength to minimize particle breakage, attrition and agglomeration through continuous re-dox reactions [Adánez et al. 2004b]. As the first approximation, the particle size is set to be constant during the operation. At the same time, because the particle has a porous structure and gas can penetrate into the particles through the particle grain to react with the active metal oxide (or metal) distributed evenly in space, we can further assume that the apparent volume of one particle remains constant, and only the particle apparent density changes with oxidization degree. Note that this assumption is only suitable for the supported oxygen carrier. Based on the above assumption, the particle apparent density is described by Eq. 8.1, according to the definition of oxidization degree X in Chapter 1. ))(1( 0100100   XX (8.1) where ρ100 is the apparent density of particle with oxidation degree X of 100%, ρ0 is the apparent density of particle with oxidation degree X of 0%, ρx is the apparent density of particle with oxidation degree X . Differentiating both sides of Eq. 8.1, we have: dXd X )( 0100   (8.2) 8.1.2 General particle mass balance equation Particle population balance models have been used to predict solid conversion in a fluidized beds for gas-solid reactions [Jiang et al. 1995, Kunii et al. 1991, Adánez et al. 2003, and Wang et al. 2003], and received considerable attention because of theirs wide applicability to a variety of particulate processes affected by the particle size. As shown in Figure 8.2, if a group of particles 102 with mass of W in a fixed space volume at steady state can be described by a particle mass distribution function Pb(X,R), a solid flow F0 with a mass distribution function P0(X, R) is inputted into the volume, and a solid flow F1 with a mass distribution function P1(X, R) moves out from the volume, a particle mass population balance can be built. According to Levenspiel et al (1968), a mass balance on the particles with oxidization degree between X and X+dX and size between R and R+dR in a unit time is given: (solids entering from input flow) –(solids leaving in output flow) + (solids growing into the interval because of reaction) – (solids growing out of the interval because of reaction) + (generation of mass within interval because of reaction) =0 (8.3) Figure 8.2 Solids population balance for a controlled volume With an assumption that the distribution of particles in one range [dX, dR] is homogeneous, the mass of particles which will leave from the range [dX, dR] in unit time because of reaction can be shown as ] )( [ ),( dt dXdRd dXdR dXdRRXWPb . Because particle size has been assumed to be constant, we can find: dR dt dXdX dt dRdR dt dX dt dXdRd )( (8.4) The mass of particles which will leave from the range [dX, dR] in a unit time because of the reaction can then be described by: dRCRXRXWP dt dXdRRXWP dt dXdRd dXdR dXdRRXWP bb b ),,(),(),(])([),(  (8.5) where dt dXCRX ),,( is the reaction rate for given pressure and temperature, C is the gas reactant concentration. Similarly the mass of particles which will grow into the range [dX, dR] in a unit time because of reaction can be expressed as: 103 dRCRdXXRdXXWPb ),,(),(   (8.6) The mass generation within the interval because of the reaction in a unit time is given by: dt d dXdRRXPW dt dV dt Vd dt dm XX b XX   ),( )(  (8.7) where V is the total apparent volume of the group of particles. Substituting ρx, dρx into Eq. 8.7 with Eqs.8.1 and 8.2, we have )1( ),,(),( )1( )( 0100 100 0100 100 X dXdRCRXRXPW dt dX X dXdRXPW dt dm bb          (8.8) Let a  0100 100   , we can get Eq. 8.9: )1( ),,(),( Xa dXdRCRXRXPW dt dm b    (8.9) On the other hand, the mass of particles entering from input flow can be expressed as ),(00 RXPF (8.10) and the mass of particles leaving into output flow can be represented by ),(11 RXPF (8.11) Then with substituting Eqs. 8.5, 8.6, 8.9, 8.10 and 8.11 into Eq. 8.3, we have 0 )1( ),,(),( ),,(),( ),,(),(),(),( 1100     Xa dXdRCRXRXWP dRCRXRXWP dRCRdXXRdXXWPdXdRRXPFdXdRRXPF b b b   (8.12) Rearranging the above Eq. 8.12, the general particle mass balance equation can be shown as 0 )1( ),,(),()],,(),([),(),( 1100   Xa CRXRXWP dX CRXRXPdWRXPFRXPF bb  (8.13) 8.1.3 Fuel reactor model 8.1.3.1 Hydrodynamics of fuel reactor The fuel reactor model is based on the assumption that it is operated as a bubbling fluidized bed. Two-phase model of Davidson et al. (1963) is applied for the fuel reactor. Hydrodynamic behaviour in the reactor can be described by the two-phase model with following assumptions: (i) the bed is consist of two phase: a bubble phase with variable bubble size and bubble rising velocity, and an emulsion phase under minimum fluidization conditions; (ii) the bed is isothermal with 104 constant temperature; (iii) solids in the emulsion phase are well mixed and the bubble phase is plug flow [Jiang et al. 1995, Adánez et al. 2003]; (iv) the system is at steady-state; (v) no consideration of freeboard or heat transfer; (vi) chemical reaction only occurs in the emulsion phase; (vii) the average particle density and particle distribution function are assumed to be the same in the bed and the outflow; (viii) there is no attrition and particle size is constant; and (ix) no particle is lost due to entrainment. Therefore, the flow structure can be simplified as shown in Figure 8.3, where Ws,fuel is the total weight of the solids in the bed, Pb,fuel(X,R) is the particle distribution function, fuels, is the average particle density. The gas mass transfer coefficient between the bubble and emulsion phase, Kbe, can be calculated by the equation given in Table 8.1. The concentrations of methane in the bubble and emulsion phases are bCHC ,4 and eCHC ,4 , respectively. The superficial gas velocity of inlet gas for fuel reactor is U f0. The expressions for the hydrodynamics used in the two-phase bubbling bed model are listed in Table 8.1. Figure 8.3 Flow structure and solid flows in fuel reactor 105 Table 8.1 Expressions used in bubbling bed model for fuel reactor Parameter Expression Reference Bubble size 8.04.02.0 )4()(54.0)( Nor A ZUUgZD mfb   Darton et al. 1977 Bubble surface area per unit volume fuelbfuelb Da ,, /6 Sit and Grace. 1981 Relative rise velocity of bubble )(711.0)( ZgDZU bbr  Davidson and Harrison 1963 Absolute rise velocity of bubble )()( ZUUUZU brmfabs  Davidson and Harrison 1963 Expanded bed height  H abs mfmf ZU dZ UUHH 0 )( )( Davidson and Harrison 1963 Bubble-to-emulsion gas mass transfer coefficient b absmfABmf be D UDU K  4 3  Sit and Grace. 1981 Fraction of bed occupied by bubbles abs mf U UU  0 Davidson and Harrison 1963 8.1.3.2 Mole balance of gas phase The mole balances for the CH4 were developed for the fuel reactor modeled as bubbling bed. For the methane mole balance in the bubble phase )( )( ,,,, ,, 44 4 bCHeCHfuelbfuelbe bCHfuelabs CCaK dZ CUd  (8.14) For the methane mole balance in the emulsion phase    RandXall eCHCHfuelb fuelbfuelfuelmf eCHbCHfuelbefuelbfuel eCH fuelmffuelmffuel Xa dXdRCRXRXP CCKa dZ dC U )1( ),,(),( )1)(1( 64 1000 )()1( ,, ,, ,,,, , ,, 44 44 4   (8.15) The second term on the right hand side is the methane disappearance because of the gas-solid reaction (methane combustion) which depends on the individual particle reaction rate, ),,( ,44 eCHCH CRX defined below in Eq. 8.53 (shown in Section 8.1.5). The boundary condition at Z = 0 for the gas phase: 4,0,,4,,4 CHineCHinbCH CCC  (8.16) 106 8.1.3.3 Particle population balance Figure 8.3 also shows the solid flows in the bubbling bed. Fs0,fuel and Fs,out,fuel, are the feeding solids flow rate and the outflow solids flow rate because of overflow, respectively. P0,fuel(X,R) and Pout,fuel(X,R) are the particle distribution functions in the feed and in the outflow, respectively. Based on the above assumption that the particle size is constant during the reaction, Eq. 8.13 can be rewritten for a group of particles with oxidization degree between X and X+dX and size between R and R+dR in the fuel reactor: 0 )1( ),,(),( )],,(),([ ),(),( ,,, ,, ,,,,0,0 44 44     Xa CRXRXPW dX CRXRXPd WRXPFRXPF eCHCHfuelbfuels eCHCHfuelb fuelsfueloutfuelsoutfuelfuels   (8.17) In addition, for the particle distribution function, we have following restriction equation:   RandXall dXdRRXP 1),( (8.18) Therefore, integrating Eq. 8.17 for all of the particles, the mass balance for all of the particles can be given as:   RandXall eCHCHfuelbfuels fuelsoutfuels Xa dXdRCRXRXPW FF )1( ),,(),( ,,, ,,0 44  (8.19) and the bed mass in fuel reactor can be shown as:   H fuelmffuelfuelfuelbfuels dZAW 0 ,,, )1)(1(  (8.20) where fuelb, is the average density of particles in fuel reactor, Afuel is the cross-sectional area of the fuel reactor. The average particle density and particle distribution function are assumed to be the same in the bed and the outflow, ),(),( ,, RXPRXP fueloutfuelb  (8.21) 8.1.4 Air reactor model As shown in Figure 8.4, the air reactor is divided into two sections: the dense bottom zone and the dilute transport zone based on the cold-flow experimental results. 107 Figure 8.4 Flow structure in air reactor 8.1.4.1 Dense bottom zone The dense zone is assumed to operate in the bubbling bed regime. The average voidage along the height of the air reactor under different operating conditions was obtained from the cold-flow experiments. The height of the dense zone, i.e., Hdense, shown in Figure 8.4, can be determined by the gradient of average voidage since the gradient of average voidage at the point between the dense zone and the dilute transport zone will decrease significantly. According to the cold-flow experiments, the height of the dense zone is around 0.5 m. The solids mass balance in the air reactor is also given in Figure 8.4. Ws,dense is the inventory of the bed materials in the dense zone, Pb,dense is the particle distribution function and, denseb, is the average density of the particles in the dense zone. Fs0,air, Fsd, and Fst are the solids mass flow rates of 108 feed, reflux into the dense zone from the dilute transport zone, and flux entrained into the dilute transport zone from the dense zone, respectively, and P0air, Pd and Pt are the corresponding particles distribution function of these solids streams, respectively. F0,O2 andC0,O2 are the mass flow rate and concentration of oxygen in the feed to the air reactor, respectively. Ua0 is the superficial gas velocity of the air reactor. 8.1.4.1.1 Mole balance of gas phase Similar to the gas phase mole balance in the bubbling bed of the fuel reactor, the oxygen mole balance in the bottom dense zone of the air reactor can be developed. The oxygen mole balance in the bubble phase of the bottom dense zone is given by )( )( ,2,2,, ,2, bOeOdensebdensebe bOdenseabs CCaK dZ CUd  (8.22) The oxygen mole balance in the emulsion phase of the bottom dense zone is    RandXall eOOdenseb densebdenseairmf eObOdensebedensebdense eO airmfairmfdense Xa dXdRCRXRXP CCKa dZ dC U )1( ),,(),( )1)(1( 32 1000 )()1( ,22, ,, ,2,2,, ,2 ,,   (8.23) The second term on the right hand side is the oxygen disappearance because of the gas-solid reaction which is dependent on the individual particle reaction rate, ),,( ,22 eOO CRX , as defined in Eq. 8.52 in Section 8.1.5. The boundary condition at Z = 0 for oxygen is: 2,0,,2,,2 OineOinbO CCC  (8.24) From the cold-flow experiment, the radial average voidage in the bottom dense zone, dense , can be measured, and we have airmfdensedensedense ,)1(   (8.25) The bubble fraction in the bottom dense zone can then be calculated from airmf airmfdense dense , , 1       (8.26) At the same time, according to the two-phase bubbling bed model, the bubble fraction can be shown as the function of bubble rise velocity, i.e., denseabs airmfair dense U UU , ,,0  (8.27) Combining Eqs. 8.26 and 8.27, the bubble rise velocity in the bottom dense zone can be shown to be: 109 airmfdense airmfairairmf denseabs UU U , ,,0, , ))(1(      (8.28) The bubble size, Db,dense, and the bubble surface area per unit volume, ab,dense, can be calculated by equations listed in Table 8.1. 8.1.4.1.2 Particle population balance Applying Eq. 8.13 to the bottom dense zone of the air reactor for the population balance of a group of particles with oxidization degree between X and X+dX and size between R and R+dR, the solid mass balance can be given as: 0 )1( ),,(),( )],,(),([ ),(),(),( ,22,, ,22, ,,0,0     Xa CRXRXPW dX CRXRXPd WRXPFRXPFRXPF eOOdensebdenses eOOdenseb densesststsdsdairairs   (8.29) Integrating Eq. 8.29 for all of the particles, the mass balance for all of the particles can be given as:   RandXall eOOdensebdenses sesdairs Xa dXdRCRXRXPW FFF )1( ),,(),( ,22,, ,0  (8.30) and the bed mass in the bottom dense zone of the air reactor is given by:   denseH airmfdenseairdensebdenses dZAW 0 ,,, )1)(1(  (8.31) where denseb, is the average density of particles in the bottom dense zone of air reactor, Aair is the cross-sectional area of the air reactor, 8.1.4.2 Dilute transport zone The core-annulus flow structure is widely accepted for the hydrodynamic modeling of CFB [e.g., Talukdar et al. 1993, Patience et al. 1993, Puchyr et al. 1997, Davidson et al. 2000, Vandewalle et al. 2002, Teplitsky et al. 2002, Teplitsky et al. 2003, Lőffler et al. 2003]. Many studies have applied this flow structure to predict the performance of CFB riser with chemical reaction and given good agreement with the experiment data [e.g., Bi et al. 1992, Puchyr et al. 1996, Fakeeha et al. 2000, Mukadai et al. 2000, Mao et al. 2001, Mao et al. 2002, Wang et al. 2003, Hua et al. 2004]. Thus, for the dilute transport zone of the air reactor, the core-annulus flow structure is integrated with the particle population balance model to predict the performance of the air reactor. As shown in the Figure 8.5, the dilute transport zone in the air reactor is characterized by a core-annulus structure: particles are transported upward in the core region and downward in the 110 annulus region, gas in the annulus is stagnant. There is no radial gradient in fluidization properties within the core and annulus. Mass transfer of both gas and solids between core and annulus is characterized by mass exchange coefficients. 8.1.4.2.1 Hydrodynamics in dilute transport zone The hydrodynamic properties for the core-annulus structure in the dilute transport zone can be obtained by combining the experimental data from the cold-flow experiments and some empirical equations which are most popularly used for the core-annulus flow structure. A review of the different core-annulus flow models was given in Löffler et al. (2003). The equations used in this study are presented in Table 8.2. Figure 8.5 Core-annulus structure in the dilute transport zone of air reactor The radial average voidage in the dilute transport zone, dilute , can be measured from the cold- flow experiment, and is given by the following equation, with the assumption that voidage in the annulus is the voidage at minimum fluidization conditions, independent of height, i.e., εmf,air, we have 111 )1(, air c airmf air c c air a a air c cdilute A A A A A A A A   (8.32) εc and εa are the radial average voidage in the core and annulus, respectively. Ac, Aa and Aair are the cross-sectional area of the core, the annulus and the air reactor, respectively. Table 8.2 Expressions used for the core-annulus flow structure in the air reactor Parameter Expression Reference Core area Dependent * N/A Voidage in the annulus mfa  Berruti et al. 1989 Voidage in the core Dependent * N/A Gas velocity in the annulus Stagnant Adánez et al. 1995 Gas velocity in the core cc g gc A V U   Adánez et al. 1995 Solid velocity in the annulus 794.1 0055.0 UUsa  Namkung et al. 1998 Solid velocity in the core tgcsc UUU  Harris et al. 1993 Solid mass transfer coefficient from core to annulus t ca UU D   0 14.0 Adánez et al. 1995 Solid mass transfer coefficient from annulus to core Dependent * N/A Gas mass transfer coefficient between core and annulus smKca /03.0 Grace et al. 2003 Dependent*: calculated based on the mass balance On the other hand, the internal upward solid flow rate in the core can be shown as: ccpscsc AUF )1(   (8.33) and the internal downward solid flow rate in the annulus is aapsasa AUF )1(   (8.34) Both Fsc and Fsa can be linked by the solids circulation flux Gs as airssasc AGFF  (8.35) Combining Eqs. 8.33, 8.34, 8.35 and the expressions of velocities in Table 8.2 and rearranging, p s air c airmft air c ct cc g G A A U A A U A V   )1)(1()1)(( , (8.36) 112 The cross sectional area and the voidage of the core, i.e., Ac and εc, can be calculated as a function of radial average voidage in the dilute transport zone along the height of the reactor by combining Eqs. 8.32 and 8.36. Solid mass balance of the internal upward solids stream change in the core and the internal downward solid flow flux change in the annulus are related to the solid mass transfer between the core and the annulus [Vandewalle et al. 2002] as follows: )1(2)1(2 ])1([])1([ acacccca aasaccsc rDrD dZ AUd dZ AUd   (8.37) Thus, the solid mass transfer coefficient from annulus to core, Dac, can be calculated through Eq. 8.37 based on the solid mass transfer coefficient from the core to annulus, Dca, given in Table 8.2. 8.1.4.2.2 Particle population balance If the dilute transport zone is divided into a series of discrete elements along the height, the mass balance of solids and oxygen in one specific discrete element can be expressed as shown in Figure 8.6. In each discrete element, there are core and annulus regions composed of gases and solids. Wcb ,i and Wab,i represent the mass of particles with average density icb, in the core and the mass of particles with average density iab , in the annulus, respectively. Pcb,i and Pab,i are their corresponding mass distribution functions of particles. Fcs,in,i , Fcsout,i, Fas,in,i and Fas,out,i are input solids mass flow rates of the core, output solids mass flow rate of the core, input solids mass flow rates of the annulus and output solids mass flow rate of the annulus respectively. Pcs,in,i , Pcs,out,i, Pas,in,i and Pas,out,i are the mass distribution functions of particles in the input solids flow of the core, output solids flow of the core, input solids flow of the annulus, output solids flow of the annulus respectively. Similarly, iincs ,, , ioutcs ,, , iinas ,, and ioutas ,, are the average solid density of input solids flow of the core, output solids flow of the core, and input solids of the annulus, output solids flow of the annulus respectively. Fc,O2,in,i and Fc,O2,out,i are input oxygen mass flow rate of the core with concentration of Cc,O2,in,i and output oxygen mass flow rate from the core with concentration of Cc,O2,out,i, respectively. Oxygen concentrations in the core and the annulus are Cc,O2,b,i and Ca,O2,in,i,respectively. Ss,i and Sg,i are the mass flow rates of solid and molar flow rate of oxygen which are transferred from the core to the annulus, respectively. 113 iinas ,, iba ,, ioutas ,, iincs ,, ibc ,, ioutcs ,, Figure 8.6 Mole balance of gas mass balance of solids in one discrete element Applying Eq. 8.13 to the particle population balance in one specific discrete element shown in Figure 8.6, the mass balance of particles with oxidization degree between X and X+dX and size between R and R+dR in the core can be written as 0),( )1( ),,(),( )],,(),([ ),(),( , ,,2,2,, ,,2,2, ,,,,,,,,,     RXS Xa CRXRXPW dX CRXRXPd WRXPFRXPF is ibOcoicbicb ibOcOicb icbioutcsioutcsiincsiincs   (8.39) where: icbicbicicb dZAW ,,,, )1(  dZrRXPDRXPDXS iciabiaiabacicbicicbcais ,,,,,,,, 2)],()1(),()1([)(   Similarly, we can write the mass balance of particles with oxidization degree between X and X+dX and size between R and R+dR in the annulus as follows: 0),( )1( ),,(),( )],,(),([ ),(),( , ,,,, ,,2,2, ,,,,,,,,,   RXS Xa CRXRXPW dX CRXRXPd WRXPFRXPF is ibagiabiab ibOaOiab iabioutasioutasiinasiinas   (8.40) where: iabiabiaiab dZAW ,,,, )1(  114 It is noted that the output solids stream of one discrete element actually is the input solids stream of the next element. Thus, for the discrete elements in the core, ioutcsiincs FF ,,1,,  (8.41) ioutcsiincs PP ,,1,,  (8.42) and for the discrete elements in the annulus, 1,,,,  ioutasiinas FF (8.43) 1,,,,  ioutasiinas PP (8.44) The input solids stream of the core is actually the entrainment solids stream from the dense zone at the bottom of the dilute transport zone, stbottomincs FF ,, (8.45) tbottomincs PP ,, (8.46) The output solids stream from the bottom of annulus corresponds to the return solids stream from the dilute transport zone into the dense zone, sdbottomoutas FF ,, (8.45) dbottomoutas PP ,, (8.46) On the other hand, it is noted that for the annulus of the top discrete element, there is no input solids stream, i.e., 0,, topinasF (8.47) 8.1.4.2.3 Mole balance of gas phase The oxygen mole balance in the core is ig RandXall ibOcOicbicb iinOcioutOc SXa dXdRCRXrRXPW FF , ,,2,2,, ,,2,,,2, ))1( ),,(),( ( 32 1000     (8.48) where: dZrCCKS icibOaibOccaig ,,,2,,,2,, 2)(  The oxygen mole balance of oxygen in the annulus,   RandXall ibOaOiabiab ig Xa dXdRCRXRXPW S )1( ),,(),( 32 1000 ,,2,2,, ,  (8.49) Because the input gas flows of the bottom discrete element of the dilute transport zone are the gas outflows from the dense zone, so the boundary condition of gas for the dilute transport zone is: 2,,,2, OtbottominOc FF  (8.50) 2,,,2, Otbottominoc CC  (8.51) 115 8.1.5 Reaction rate of single particle Chapter 1 presents the literature review on the reaction kinetic model of a single particle. The chosen reaction kinetic model should consider the effect of temperature, particle size, oxidation degree of particle, and gas concentration. The reduction reactant should be CH4 in the experiments for obtaining the reaction kinetic parameters. Thus, the kinetic model adopted by Zafar et al. (2007a,b) is chosen in this study to describe the reaction rate of a single particle. For oxidation of a single particle 3/22 )1( 0026.0 3 X R Ckb dt dX m n Ooxox ox   where )/( 0 TRE oxox universaloxekk  (8.52) For reduction of a single particle 3/24 0026.0 3 X R Ckb dt dX m n CHrere re  where )/( 0 TRE rere universalreekk  (8.53) The parameters in Eqs.8.52 and 8.53 are given in Zafar et al. (2007a) and summarized in the following Table 8.3. Table 8.3 Kinetic parameters of single oxygen carrier (refer to Zafar et al. 2007a) Parameter Values Stoichometric factor for oxidation, oxb (-) 2 Stoichometric factor for reduction, reb (-) 4 Reaction order for oxidation, oxn (-) 1.0 Reaction order for reduction, ren (-) 0.4 Pre-exponential factor of the chemical reaction rate constant for oxidation, oxk0 (m/s) 5.43×10-3 Pre-exponential factor of the chemical reaction rate constant for oxidation, rek 0 (mol 0.6m-0.8/s) 2.74 Activation energy for oxidation, oxE (kJ/mol) 40 Activation energy for reduction, reE (kJ/mol) 114 Molar density of the reacting material, m (mol/m3) 47712 Universal gas constant, universalR (J/mol.k) 8.3145 116 8.1.6 Energy balance As a simplification, temperature distribution in the air reactor and the fuel reactor are assumed uniform at T1 and T2, respectively. Temperatures of the gas and solids streams out of each reactor are also assumed to be T1 and T2, respectively. Thus, temperature is dependent on the combination of the heat carried into and out of the reactor by the gas and solids flows, reaction heat, and heat removed from each reactor. Figure 8.7 presents the gas and solids streams in the reactor system. The oxidation reaction in the air reactor is exothermic and releases at heat rate, Qox, and the reduction reaction in the fuel reactor is endothermic and needs to absorb at heat rate, Qre. In addition, the air reactor and fuel reactor have some heat removed by the cooling system and heat transfer between reactors and environment, i.e. Qremove,air and Qremove,fuel. The solids flow rate, Fs,out,air, from the air reactor will be divided into two streams i.e., Fs1 and Fs2, one inputted into the fuel reactor, and the other returned to the air reactor directly through the bypass. Figure 8.8 show the energy balance for the whole system. The energy balance for the air reactor can be written as: airremove output ox input QhFQhF ,  (8.54) Energy balance for the fuel reactor can be written as   output fuelremovere input QQhFhF , (8.55) where h is the enthalpy of gas species and solids at their temperature. 8.2 Computational procedure Based on the energy balance calculation, if both reactors are regarded as adiabatic (no heat removed from the reactor system), the temperatures in the reactor system could exceed 2000oC, which is the adiabatic flame temperature of CH4 when air is the oxidizer. However, in the actual situation, temperatures in the reactors will never be so high because of the heat removed by the solids flow and heat exchangers. Because the reduction in fuel reactor is endothermic, the heat for reaction is provided by the solids stream from the air reactor to the fuel reactor, and the temperature difference between air reactor and fuel reactor is an inversely proportional function of the solids circulation flux between the two reactors. Thus, as shown in Figure 8.9, the temperatures in both reactors can be initially assumed equal to the temperature in the experiments investigating the reaction kinetics of particles [Zafar et al. 2007a,b]. Using the computational procedure for mass balance shown in Figures 8.10 and 8.11, the 117 airoutX , airoutX , airoutfuelin XX ,,  fueloutX , airinX , mass flow rates of solids and gas in the reactor system can be obtained for a specific reactor temperature. Considering the removed heat because of the cooling system and the heat transfer between the reactors and environment, new temperatures in reactors can be determined from the energy balance equations given by Eqs. 8.54 and 8.55. This cycle of calculation is repeated until the difference between the temperatures calculated at the last cycle and the former cycle is less than the convergent limit. The simulation flow charts for the mass balance are shown in Figures 8.10 and 8.11. The computational procedure can calculate the local gas concentrations and particles distribution in the reactors, and the concentrations and mass flow rate of oxygen and methane in the outlet gas from the reactors. The iteration is repeated until all parameters in the reactors satisfy the convergent limits. Figure 8.7 Gas-solids flow in reactor system 118 Figure 8.8 Energy balance in reactor system 119 Figure 8.9 Flow chart for determining temperatures of the air and fuel reactors Heat balance for air reactor: Heat carried into by solids + Reaction heat =Heat carried out by solids +Heat carried out by gases +Heat removal Heat balance for fuel reactor: Heat carried into by solids - Reaction heat=Heat carried out by solids +Heat carried out by gases + Heat removal Using flow chart in Figures 8.10 and 8.11 for mass balance Assuming the temperatures of air and fuel reactors as T1 and T2, respectively New T1 and T2 Input solids stream and gas stream Output solids stream and gas stream Old T1 and T2 from last calculation ΔT1 >0.001 or ΔT2 >0.001 ΔT1 < 0.001 and ΔT2 < 0.001 Stop 120 Figure 8.10 Flow chart of the mass balance for the air reactor Air Reactor Bottom Dense Zone Particle population balance Mole balance of gas phase Dilute Transport Zone Particle population balance Mole balance of gas phase Geometry of air reactor Temperature and pressure in air reactor T1 , P Reaction kinetic of single particle (Oxidation) Flow structure Bubble-emulsion phase model Empirical equations and experimental data Flow structure Core-annulus Empirical equations and experimental data Input gas of air reactor (T0) Gas flow rate concentration O2 and N2 Input particles of air reactor Solids flow rate Particle distribution Output gas of air reactor (T1) Gas flow rate Concentration Solids stream Solids stream Gas stream Local gas concentrati on and particle distribution function Output particles from air reactor (T1) Solids flow rate Particle distribution function 121 Figure 8.11 Flow chart of the mass balance for the fuel reactor Fuel reactor Reaction kinetic of single particle (Reduction) Flow structure Bubble-emulsion phase model Empirical equations and experimental data Particle population balance Mole balance of gas phase Geometry of fuel reactor Temperature and pressure in fuel reactor T2, P Input gas of Fuel reactor (T0) Gas flow rate concentration CH4 Output particles from Fuel reactor (T2) Solids flow rate Particle distribution function Output particles from air reactor (T1) Solids flow rate Particle distribution function Particles mixing Local gas concentration and particle distribution function Input particles of fuel reactor (T1) Solids flow rate Particle distribution functionReturn particlesfrom air reactor (T1) Solids flow rate Particle distribution function Output gas from fuel reactor (T2) Gas flow rate Concentration Input particles of air reactor Solids flow rate Particle distribution 122 8.3 Results and discussion 8.3.1 Analysis on the fuel reactor With an assumption that the temperature of the fuel reactor can be controlled at a specific T2 through a cooling or heating system, the effect of the temperature of the fuel reactor on the fuel reactor performance can be investigated. Figure 8.12 shows the effect of temperature, T2, on the performance of the fuel reactor. Because the reaction rate of a single particle has a strong dependency on the temperature [García-Labiano et al. 2006, Zafar et al. 2007a,b], the CH4 conversion increases with increasing temperature of the fuel reactor. As shown in Figure 8.12, increasing the fuel reactor temperature from 873 to 1173 K increases the CH4 conversion from 38% to 95%. Higher CH4 concentration in the flue gas from the fuel reactor resulting from lower CH4 conversion will reduce the CO2 separation efficiency. Thus, the temperature in the fuel reactor is set to a higher value of 1173 K to achieve a good fuel conversion for all simulation runs. 850 900 950 1000 1050 1100 1150 1200 30 40 50 60 70 80 90 100 CH4 conversion CH4 concentration Temperature in fuel reactor (K) C H 4 co n ve rs io n (% ) 0 2 4 6 8 10 12 C H 4 co ncen tra tion in flue g as (m ol/m 3) Figure 8.12 Effect of temperature of fuel reactor, T2 , on the fuel reactor performance F s0, fuel=0.12 kg/s, Ws,fuel=19 kg, F0,CH4=1.42 g/s, fuelinX , =40% Based on the reaction kinetics of oxygen carrier [Zafar et al. 2007a,b], temperature in the air reactor is set at 1223 K. Since the reaction in the air reactor is exothermic, heat required for the endothermic reaction in the fuel reactor eventually is provided by the solids stream from the air reactor to the fuel reactor. If the temperature in the air reactor is set at 1223 K, and the fuel reactor is 123 assumed as adiabatic, the minimum input solids circulation flux and solids circulation rate for maintaining the temperature of fuel reactor at 1173 K is proportional to the mass flow rate of input CH4 based on the energy balance in fuel reactor as shown in Figure 8.13. 0.08 0.10 0.12 0.14 0.16 0.18 30 35 40 45 50 55 60 65 70 75 Solids circulation flux Solids circulation rate Mass flow rate of input CH4 for fuel reactor (mol/s)M in im um so lid s ci rc ul at io n flu x be tw ee n ai r re ac to r an d fu el re ac to r( kg /m 2 . s) 0.25 0.30 0.35 0.40 0.45 0.50 0.55 0.60 M inim um solids circulation rate betw een air reactorand fuelreactor(kg/s) Figure 8.13 Minimum solids circulation flux and solids circulation rate to achieve temperature in fuel reactor at 1173 K (T1 = 1223 k) Figure 8.14 presents the effect of input solids mass flow rate on the performance of the fuel reactor. The solids mass flow rate, Fs0,fuel, directly influences the amount of oxygen transferred from the air reactor to the fuel reactor. Thus, the fuel conversion increases with increasing solids mass flow rate. Another parameter which directly influences the amount of oxygen transferred from the air reactor to the fuel reactor is the average oxidation degree of particles, fuelinX , . For a constant input solids mass flow rate, higher fuelinX , means more oxygen can be transferred into the fuel reactor in a unit time by the solids flow. In addition, the experimental study on the reaction kinetics of oxygen carrier also shows that the oxygen carrier with higher oxidation degree has higher reduction rate. Figure 8.15 shows the effect of fuelinX , on the fuel conversion. This figure indicates that when fuelinX , < 40%, increasing fuelinX , can result in obvious improvement on the performance of the fuel reactor, and after fuelinX , > 40%, the improvement is minimum. 124 0.0 0.2 0.4 0.6 0.8 1.0 20 30 40 50 60 70 80 90 100 CH4 conversion CH4 concentration Input solids mass flow rate of fuel reactor (kg/s) C H 4 co nv er si on (% ) 0 2 4 6 8 10 C H 4 concentration flue gas (m ol/m 3) Figure 8.14 Effect of input solids mass flow rate of fuel reactor, Fs0,fuel, on fuel reactor performance Ws,fuel =18 kg, F0,CH4 =1.42 g/s, fuelinX , =10%, T2 =1173 K 0 20 40 60 80 100 80 90 100 CH4 conversion CH4 concentration Average oxidation degree of input particles of fuel reactor (%) C H 4 co nv e rs io n (% ) 0.0 0.5 1.0 1.5 2.0 C H 4 con cen tratio n in flu e g as (m ol/m 3 ) Figure 8.15 Effect of average oxidation degree of input particles, fuelinX , , on fuel reactor performance Fs0,fuel =0.48 kg/s, Ws,fuel =18 kg, F0,CH4 =1.42 g/s, T2=1173 K In order to investigate the influence of superficial gas velocity in the fuel reactor, Uf0, on the fuel reactor performance, U f0 was varied between 0.076 m/s and 0.152 m/s, as a result, the mass flow rates of CH4, F0,CH4 , changed from 1.42 g/s to 2.84 g/s correspondingly, since the temperature and pressure in the fuel reactor were set at fixed values at 1173 K and 1 bar, respectively. The fuel 125 conversion, as shown in Figure 8.16, is decreased linearly with increasing Uf0, since the residence time of gas in the fuel reactor decreases with increasing Uf 0. 0.07 0.08 0.09 0.10 0.11 0.12 0.13 0.14 0.15 0.16 82 84 86 88 90 92 94 96 98 100 CH4 conversion CH4 concentration Superficial gas velocity in fuel reactor (m/s) C H 4 co nv er si on (% ) 0 2 C H 4 con ce ntration in flue gas (m ol/m 3) Figure 8.16 Effect of superficial gas velocity of fuel reactor, Uf0, on fuel reactor performance Fs0,fuel=0.48 kg/s, Ws,fuel =18 kg, fuelinX , =40%, T2=1173 K The effect of bed materials in the fuel reactor, Ws,fuel, is shown in Figure 8.17. For a constant input solids mass flow rate, increasing of bed materials means longer residence time of particles and fuel in the fuel reactor. Thus, the fuel conversion is improved. As discussed above, for an atmospheric fuel reactor at a specific temperature, fuel conversion actually can be shown as a function of the following parameters: Fs0,fuel, Ws,fuel, fuelinX , , and F0,CH4 (Uf0) (8.56) In order to show the combined effect of the above parameters, a dimensionless parameter, which represents the mass ratio of oxidized input solids to the input fuel in unit time, is introduced. 4,0 ,,0 CH fuelinfuels fuel F XF Mr  (8.57) Simulation runs were carried out for the fuel reactor at different combination of operation parameters listed in Eq. 8.56 with the Fs0,fuel, Ws,fuel, fuelinX , , and F0,CH4 varying at 0.12~0.96 kg/s, 18~72 kg, 10%~100% and 1.42~2.84 g/s, respectively. The specific values of the parameters are listed in Table D.13 in Appendix D. Figure 8.18 presents the effect of Mrfuel on the fuel conversion with the variations of F s0,fuel, Ws,fuel, fuelinX , , and F0,CH4 126 10 20 30 40 50 60 70 80 80 85 90 95 100 CH4 conversion CH4 concentration Bed materials in fuel reactor (kg) C H 4 co n ve rs io n (% ) 0.0 0.5 1.0 1.5 2.0 C H 4 con cen tra tion in flue g a s (m ol/m 3 ) Figure 8.17 Effect of bed materials in fuel reactor, Ws,fuel, on fuel reactor performance Fs0,fuel =0.48 kg/s, F0,CH4 =1.42 g/s, fuelinX , =10%, T2 =1173 K 0 50 100 150 200 250 300 350 20 30 40 50 60 70 80 90 100 Dimensionless paramter Mrfuel (-) C H 4 co nv er si o n (% ) Ws,fuel=18kg Ws,fuel=36kg Ws,fuel=72kg Figure 8.18 Effect of dimensionless parameter fuelMr (T2=1173 K) Figure 8.18 indicates that Mrfuel has the biggest influence on the performance of fuel reactor. In order to achieve higher fuel conversion, it is necessary that the Mrfuel > 50 for the oxygen carrier 127 used in this model analysis. This would be very helpful to determine the suitable solids circulation rate from the air reactor to the fuel reactor for a given mass flow rate of input fuel. On the other hand, when the fuel conversion is above 90%, the effect of increasing Mrfuel will not be obvious. Figure 8.19 is the partial enlargement of Figure 8.18. It shows that the bed materials must be increased in order to improve the performance of fuel reactor when the fuel conversion is more than 90%. 0 50 100 150 200 250 300 350 80 82 84 86 88 90 92 94 96 98 100 Dimensionless paramter Mr fuel (-) C H 4 co nv e rs io n (% ) Ws ,fuel=18kg Ws ,fuel=36kg Ws ,fuel=72kg Figure 8.19 Effect of dimensionless parameter fuelMr (partial enlargement of Figure 8.18) 8.3.2 Analysis on the air reactor Figure 8.20 shows the effect of input solids mass flow rate of the air reactor, F s0,air, on the oxygen conversion. Similarly, increasing the input solids mass flow rate can result in higher oxygen conversion since more particles carry oxygen in unit time. The experimental studies [García-Labiano et al. 2006, Abad et al. 2007b, Zafar et al. 2007a,b] also indicate that the oxidation rate of oxygen carrier is reduced with increasing degree of oxidation even when the temperature and gas reactant concentration are kept constant. The strong dependency of the oxygen conversion on the average oxidation degree of input particles, airinX , , is shown in 128 Figure 8.21. When airinX , is increased from 0% to 90%, the oxygen conversion is dramatically reduced from 85% to 40%. 0.5 0.6 0.7 0.8 0.9 1.0 1.1 60 70 80 90 100 O2 conversion O2 concenration Input solids mass flow rate of air reactor (kg/s) O 2 co n ve rs io n (% ) 0.3 0.4 0.5 0.6 0.7 0.8 O 2 co nce ntra tion in th e flue g as (m o l/m 3 ) Figure 8.20 Effect of solids mass flow rate of input particles of air reactor, Fs0,air, on the performance of air reactor Ua0=6.6 m/s, F0,O2=7.02 g/s, airinX , =60%, T1=1223 K 0 20 40 60 80 100 40 50 60 70 80 90 100 O2 conversion O2 concentration Average oxidation degree of input particles of air reactor O 2 co n ve rs io n (% ) 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 O 2 con ce ntra tio in th e flue gas (m ol/m 3) Figure 8.21 Effect of average oxidation degree of input particles for air reactor, airinX , , on air reactor performance Ua0 =6.6 m/s, Fs0,air =0.49 kg/s, F0,O2 =7.02 g/s, T1 =1223 K 129 The effect of superficial gas velocity of the air reactor is shown in Figure 8.22. The value of Ua0 was varied between 6.6 and 8.8 m/s, as a result, the mass flow rates of O2,F0,O2, changed from 7.02 to 9.36 g/s correspondingly, since the temperature and pressure in the fuel reactor were set at fixed values at 12233 K and 1 bar, respectively. Since higher superficial gas velocity results in less residence time of oxygen and more mass flow rate of oxygen, when Fs0,air , airinX , and T2 are set at fixed values, the oxygen conversion decreases with decreasing Ua0. 6 7 8 9 90 92 94 96 98 O2 conversion O2 cencentration Superficial gas velocity in air reactor, U0a (m/s) O 2 co nv er si on (% ) 0.06 0.08 0.10 0.12 0.14 0.16 0.18 0.20 0.22 0.24 0.26 O 2 co ncentration in flue gas o fair rea ctor Figure 8.22 Effect of superficial gas velocity in air reactor, Ua0, on air reactor performance F s0,air =1.15 kg/s, airinX , =0%, T1 =1223 K A dimensionless parameter representing the mass ratio of reduced input particles to the input oxygen in a unit time can describe the combination effect of different operation parameters, i.e., Fs0,air , airinX , , Ua0 and F0,O2, when the air reactor is run under the condition of a specific temperature and atmospheric pressure. 2,0 ,,0 )1( O airinairs air F XF Mr  (8.58) Similarly, simulation runs were carried out for the air reactor at different combination of Fs0,air, airinX , , and F0,O2 varying at 0.52~1.03 kg/s, 0%~90% and 7.02~9.36 g/s, respectively. The specific 130 values of the parameters are listed in Table D.14 in Appendix D. Figure 8.23 presents the effect of Mrair on the oxygen conversion with the variations of F s0,air, airinX , , and F0,O2. 0 50 100 150 30 40 50 60 70 80 90 100 Dimensionless parameter Mrair O xy ge n co nv er si on (% ) U a0 =6.6m/s Ua0=7.7m/s U a0 =8.8m/s Figure 8.23 Effect of dimensionless parameter airMr on the oxygen conversion Figure 8.23 indicates that for the oxygen carrier used in this modeling analysis the dimensionless parameter, Mrair, should be more than 60 to achieve oxygen conversion of 85%, and in order to increase the oxygen conversion to be more than 90%, the value of Mrair has to be more than 140. This result implies the necessity of higher excess air coefficient. Since the oxygen conversion is about 85% at the lower number of Mrair, the excess air coefficient, λ, should be larger than 1.1 in order to transfer enough oxygen to the fuel reactor for complete conversion of fuel, 8.3.3 Linkage of air reactor and fuel reactor According to the discussion on the scaling consideration in Chapter 6, the cold-flow experiments with fluidization gas of mixture of helium (96 vol%) and air (4 vol%) can be used to simulate the hydrodynamics in a hot model running at around the atmospheric pressure. Based on the calculation in Chapter 6, the relationship of the parameters between the cold unit and hot unit can be shown by Table 8.4. 131 Table 8.4 Relationship of the parameters between cold unit and hot unit Parameter Cold Unit Hot Unit Pressure (bar) cP (1) cP17.1 (1.17) Particle diameter pd ( m ) pcd (78) pcd22.1 (95) Reactor diameter (mm) cD cD22.1 Superficial gas velocity cU 0 cU 010.1 Solids circulation flux scG scG98.1 In order to apply the hydrodynamic data obtained from the cold-unit experiments to the hot unit, the proposed hot unit has to operate at the conditions shown in Table 8.5 to satisfy the relationships listed in Table 8.4. Table 8.5 Main operating parameters of the hot unit Parameter Air reactor Fuel reactor Pressure in reactors (bar) 1.17 1.17 Temperature in reactors T (K) 1223 1173 Particle diameter pd ( m ) 95 95 Reactor diameter D (m) 0.12 0.35 Reactor height H (m) 5.60 1.90 excess air coefficient   λ (-) 1.1 N/A As discussed in Section 7.7, when the cold-flow was operated with the fluidizing gas mixture of 96 vol% helium and 4 vol% air, the cold-flow model experiments showed that Gs1 was 19.8 ~ 24.1 kg/m2.s, and Gs2 was 21.5 ~ 25.9 kg/m 2.s when Ua0, UA1 and UA2 were 6 ~ 8 m/s, 4Umf and 1Umf, respectively. According to the relationships of the parameters between cold unit and hot unit in Table 8.4, if Ua0 = 6.6 ~ 8.8 m/s in hot unit, the solids circulation flux in hot unit can be achieved at: Gs1= 39.2 ~ 47.7 kg/m2.s, Gs2 = 42.4 ~ 51.2 kg/m2.s. Therefore, the model can simulate the linkage of the air reactor and the fuel reactor for the reference cases shown in Table 8.6. The values of the performance parameters, which are illustrated by the Figure 8.7, are shown in Tables 8.7 and 8.8. 132 Table 8.6 Specific operation conditions for reference cases Cases Ua0 (m/s) U f0 (m/s) λ   (-) Ws,fuel (kg) Fs1 (kg/s) Fs2 (kg/s) Case 1 6.6 0.076 1.1 72 0.4585 0.4967 Case 2 7.7 0.089 1.1 72 0.5797 0.6280 Case 3 8.8 0.101 1.1 72 0.6090 0.6597 Table 8.7 Performance parameters for the reference cases (Fuel reactor) Cases F0,CH4 (g/s) Fs0,fuel (kg/s) fuelinX , (%) Fs,out,fuel (kg/s) fueloutX , (%) Fuel conversion (%) fuelMr (-) Case 1 1.42 0.4585 42.5 0.4529 27.6 98.3 137 Case 2 1.66 0.5797 37.1 0.5732 23.4 97.5 130 Case 3 1.89 0.6090 12.8 0.6017 3.4 96.4 41 Table 8.8 Performance parameters for the reference cases (Air reactor) Cases F0,O2 (g/s) Fs0,air (kg/s) airinX , (%) Fs,out,air (kg/s) airoutX , (%) Oxygen conversion (%) airMr (-) Case 1 6.24 0.9496 33.8 0.9552 42.5 89.5 101 Case 2 7.28 1.2013 29.1 1.2077 37.1 88.1 117 Case 3 8.32 1.2614 5.6 1.2687 12.8 87.0 143 The profile of oxygen conversion along the height of the air reactor is shown in Figure 8.24. The lower section with height of 0.5 m is the dense zone. The oxygen conversion of 80% has been achieved in the section with height of less than 3.5 m. 133 0 20 40 60 80 100 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 Oxygen conversion (%) H ei gh ta bo ve th e di st rib u to ro ft he ai rr ea ct or (m ) Case1 Case2 Case3 Dense zone Figure 8.24 Oxygen conversion along air reactor Figure 8.25 presents the profile of fuel conversion along the height of fuel reactor. 80% of the input fuel can react with oxygen in the particles in the section with height of less than 0.15 m. However, additional height of 0.15 ~ 0.7 m is required to complete the conversion of remaining 20% of fuel. Most of the fuel conversion is completed in the lower section because of the higher concentrations of gas and solids reactants, and many small bubbles at the lower section resulting in higher reaction rate. However, with increasing reactor height, the concentration of the reactants and number of bubbles are decreased significantly while bubble size becomes bigger, resulting in dramatic decrease of reaction rate and gas-solids contact. 134 0 10 20 30 40 50 60 70 80 90 100 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 Fuel conversion (%) H ei gh ta bo ve th e di st rib ut or of fu el re ac to r( m ) Case1 Case2 Case3 Figure 8.25 Fuel conversion along fuel reactor It is noted that in the real situation, there is an increase in the number of moles of gas due to reaction in the fuel reactor: one mole CH4 reacts with the oxygen from the oxygen carrier to produce one mole CO2 and two moles H2O. Thus, the superficial gas velocity in the fuel reactor increases along the height of the fuel reactor resulting in larger bubble size, higher bubble rise velocity, and increased bubble fraction. As a consequence, the fuel conversion in the fuel reactor is expected to become less due to less desirable contact between the oxygen carriers and the reacting gas. Figure 8.26 presents the effect of gas volume expansion from on the fuel conversion. The model analysis indicates that the gas volume expansion will decrease in the fuel conversion. 135 0 10 20 30 40 50 60 70 80 90 100 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 Fuel conversion (%) H e ig ht ab o ve th e di st ri bu to r of fu el re a ct or (m ) Gas volume expansion Constant gas volume Figure 8.26 Fuel conversion along fuel reactor (effect of gas volume expansion) Figures 8.27 and 8.28 show the particle distribution functions of input particles and output particles for the fuel reactor which was running at the conditions shown by the reference Case 1 in Table 8.6. The comparison between the figures indicates that the movement of the particles from the higher oxidation degree to the lower oxidation degree. 136 0 50 100 0.10.20.30.40.50.60.70.80.91 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 Particle radius (m) Oxididation degree of particles (-) P ar tic le d is tr ib ut io n fu nc tio n P (1 /  m ) Figure 8.27 Particle distribution of the input particles of fuel reactor (Case1) 0 50 100 0.10.20.30.40.50.60.70.80.91 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 Particle radius (m)Oxidiation degree of particles (-) P ar tic le di st ri bu tio n fu nc tio n P (1 / m ) Figure 8.28 Particle distribution of the output particles of fuel reactor (Case1) 137 8.4 Procedure of model optimization of atmospheric CLC reactor Based on the parameters illustrated in Figure 8.7 and flow chart in Figure 8.28, the optimization procedure for the atmospheric CLC reactor using model can be summarized as follows:  Choose the temperature in the air reactor and the fuel reactor at 1223 K and 1173 K, respectively, for a given CLC reactor with known geometry size.  Determine the mass flow rate of oxygen, F0,O2, based on the excess air coefficient, λ, of 1.1~1.2 for a given mass flow rate of CH4, F0,CH4.  Determine the minimum solids circulation rate from the air reactor to the fuel reactor, Fs0,fuel , based on the energy balance with assumption that the fuel reactor is adiabatic.  Determine the minimum average oxidation degree of input solids stream of the fuel reactor, fuelinX , , according to 50 4,0 ,,0  CH fuelinfuels fuel F XF Mr ,  Use the model of fuel reactor to calculate the output solids stream from the fuel reactor, Fs,out,fuel, and their average oxidation degree, fueloutX , , with a assumption that the fuel conversion is 100%.  Determine a minimum solids stream through bypass, F s2, according to 90 )1( 2,0 ,,0    O airinairs air F XF Mr , because the input solids mass flow rate of the air reactor and their average oxidation degree, i.e. Fs0,air and airinX , , can be determined by the solids mixing of Fs,out,fuel and Fs2.  Use the model of air reactor to calculate the output solids mass flow rate from the air reactor, Fsout,air,and their average oxidation degree, airoutX , .  Check fuelinairout XX ,,  , if not, increasing Fs0,fuel and repeat the above steps.  Determine the minimum bed materials in fuel reactor, Ws,fuel, using model of the fuel reactor for the pre-determined Fs0,fuel ,F0,CH4 and fuelinX , .  Link the fuel reactor and the air reactor to check the whole performance of the CLC reactor system and especially make sure the fuel conversion is very close to 100%. 138 F or a given F 0,CH 4 M inim um F s0 ,fue l M inimum f uelinX , 50f u elM r F sou t,fuel and fu elo u tX , C hoose m inim um F s2 ( f u eli nX , ) S ol ids m ixing F s0,a ir and a irinX , F0,O 2 λ= 1.1 ~ 1.2 Sat isfy 90ai rM r M odel of fuel reactor M odel of ai r reactor F sou t ,air and ai ro utX , fu eli nai ro ut XX ,,  ? M odel of fuel r eactor In crease F s0 ,fuel M inim um W fue l Linkage of air reactor and fuel reactor T 1 fixed at 1223 K T 2 fixed at 1173K NO Y Figure 8.29 Flow chart for optimization with using model 139 8.5 Comparison of simulation results and experimental data The model predictions were compared with the experimental data given by Pröll et al. (2009b). The reactor geometry and typical operation conditions of CLC reactor from the experiment and model simulation are shown in Table 8.9. The simulation was based on the hot unit parameters modeled by a cold unit fluidized with 96 vol% helium and 4 vol% air at temperature of 40oC and atmospheric pressure given in Table 6.5. As shown in Table 8.9, the particle properties and operating conditions are similar, but the geometries of both reactors are quite different. Table 8.9 Reactor geometry and typical operation conditions for CLC reactor Parameter Experiments in Pröll et al. (2009b) Simulation in hot model Air reactor diameter 0.150 m 0.124 m Air reactor height 4.1 m 6.0 m Fuel reactor diameter 0.159 m 0.349m Fuel reactor height 3.0 m 1.0 m Oxygen carrier Ni-based Ni-based Mean particle size 120 μm 95μm Solid inventory 65 kg 72 kg Apparent density 3200 kg/m3 2800 kg/m3 Temperature in air reactor 1213 K 1223 K Temperature in fuel reactor 1173 K 1173 K Fuel CH4 CH4 Fuel capacity 60 ~140 kW 80 ~105 kW Excess air coefficient 1.2 1.1 Table 8.10 shows the dimensionless group comparison between the hot simulation model and the hot unit in Pröll et al. (2009b). It is showed that the dimensionless groups cannot be matched well. This means that the hydrodynamic data from the cold-flow experiments would not apply well to the hot unit in Pröll et al. (2009b). 140 Table 8.10 Comparison of dimensionless groups between the simulation hot model and the hot unit in Pröll et al. 2009b Air Reactor Dimensionless parameter Simulation hot Model Hot unit in Pröll et al. (2009b) Ratio between two hot units gD U 20 36 16 2.25 f s   8235 9411 0.88 mfU U 0 1764 856 2.06 0U G s s  0.0044 0.0024 1.83 pd D 1305 1111 1.17 Fuel Reactor gD U 20 0.0017 0.1357 0.01 f s   9655 11034 0.88 mfU U 0 16.5 69.7 0.24 0u G s s  0.18 0.03 6.00 pd D 3674 1178 3.12 On the other hand, because the published experimental data on the CLC reactor hot unit is limited, the comparison can still provide somewhat reference for the validation of the model. The most important criterion to measure the performance of a CLC reactor is whether it provides sufficient oxygen to achieve high fuel conversion. Thus, as shown in Figure 8.30 the comparison will only focus on the fuel conversion as a function of fuel capacity of the reactor. 141 60 70 80 90 100 110 120 130 140 70 75 80 85 90 95 100 Fuel capacity (kW) C H 4 co n ve rs io n (% ) Experiments Model Figure 8.30 Comparison between simulation result and experimental data The detailed discussion on the simulations shown in the Figure 8.30 has been carried out in Section 8.3.3. The model showed that when the fuel capacity increases from 80 to 105 kW, the CH4 conversion will decrease slightly from 98.3% to 96.4% because of the reduced residence times of CH4 at higher superficial gas velocity. Meanwhile, the experiments showed that CH4 conversion increased slightly from 95% to 97% with increasing of fuel capacity because the increased gas-solid contact at higher superficial gas velocity over-compensated the decreasing residence times in the experiments [Pröll et al. 2009b]. As shown in Table 8.9, since the height of fuel reactor in the model is much less than that in the experiment, the effect of residence times would probably be more significant for the fuel reactor performance in the model. However, it can be stated that both the model analysis and the experiment in Pröll et al. (2009b) show that similar fuel conversion can be achieved for a same fuel capacity and sufficient oxygen for the fuel conversion can be transferred from the air reactor and the fuel reactor. However, it should be pointed out that the comparison is not the complete validation of the model because of the limited published experimental data. Further tests of the model should be carried out based on the experimental data from the proposed hot unit at UBC in the future. 142 8.6 Optimum operating condition for an atmospheric CLC reactor Because incomplete fuel conversion will reduce the energy utilization efficiency, and additional energy and equipment are needed to separate the unconverted fuel from CO2 in the exhaust gas from the fuel reactor, therefore, ideally fuel conversion should be 100% as a good CLC reactor. However, this is not possible for the real situation, so the fuel conversion should be as high as possible, preferably very close to 100%. The model simulation indicated that 98.3% of CH4 conversion can be achieved when the fuel capacity is 80 kW. On other hand, the result of model analysis shows that excess air in the air reactor is needed because the maximum oxygen conversion in the air reactor cannot be 100%. The excess air coefficient of 1.1 ~ 1.2 is recommended, because at a higher excess air coefficient the residence times of oxygen carrier would be too short to transfer sufficient oxygen from the air reactor to the fuel reactor. Thus, the oxygen conversion in air reactor should be above 80% for a good performance of CLC reactor. Based on the above the discussion, the optimum operating condition of an atmospheric CLC reactor hot unit (with dimensions shown in Table 8.9) should be chosen as follows: fuel capacity is 80 kW, Ua0=6.6 m/s, U f0= 0.076 m/s, UA1=4Umf, UA2=1Umf, and the temperature in air the reactor is 1223 K and in the fuel reactor is 1173 K. 143 CHAPTER 9 CONCLUSIONS AND RECOMMENDATION 9.1 Conclusions from this study A novel interconnected fluidized bed (IFB) reactor cold-flow model with a bypass line for CLC was designed and constructed to carry out comprehensive hydrodynamic study. The detailed mapping of the operating conditions for the reactor system and the influence of the operating conditions on the gas leakage were investigated. For scaling consideration, the cold-flow model was operated using gas mixture of helium and air as fluidizing gas to simulate the hydrodynamics of a hot model. The connection between the cold and the hot models is achieved by applying the hydrodynamic scaling law. A comprehensive mathematical model for investigating the reactor performance has been developed by combining fluidization properties and particle population balance for calculating the bed particle conversion while considering the chemical reaction of single particle. The following conclusions can be drawn from this study:  The interconnected fluidized bed CLC reactor system has been successfully operated under the following fluidization regimes: air reactor and fuel reactor working under fast fluidization and bubbling fluidization, respectively. A bypass line was added to transport some particles out of the air reactor to return the particles directly back to the air reactor without reduction. In this manner, higher solid conversion together with the higher gas velocity in the riser and solids circulation flux can be achieved.  The solids circulation flux, pressure loop and cross-sectional average solids hold-up in the air reactor were studied under different of operating conditions by changing the superficial gas velocities and aeration velocities in both loop-seals. The maximum solids circulation flux occurs at Ua0 = 4.0 m/s, when the fluidizing gas is air.  The pressure in the reactor is increased with increasing superficial gas velocity. There is a dense zone in the lower part of the air reactor with strong backmixing and splashing of solids caused by the particles returned from loop-seal 1 and loop-seal 2. Above the dense zone, the solids hold-up decreases along the height of the air reactor. The solids hold-up 144 in the air reactor decreased with increasing superficial gas velocity. Meanwhile, higher UA1 and UA2 provide higher solids hold-up in the air reactor.  For the cold-flow model operated with fluidizing gas of air, UA2 should be chosen at a 1.5Umf for Ua0=3.0 or 4.0 m/s to maintain the ratio of solids circulation fluxes between loop 1 and loop 2 at around 1:1. For a specific aeration velocity in loop-seal 2, i.e., UA2, there is a minimum superficial gas velocity in the air reactor, Umstable , to achieve a stable particle transport in the system. The recommended operating conditions are: Ua0 = 3.0 or 4.0 m/s, U f0 = 0.035 or 0.046 m/s, UA1 = 4Umf or 6Umf and UA2 =1.5Umf.  The gas leakage from the fuel reactor to air reactor was found to be between 0.8% and 3.0% resulting from a combination of pressure drop across loop-seal 1, solids stream through loop-seal 1 and aeration gas in loop-seal 1. Higher positive pressure drop between solids outlet and inlet of loop-seal 1 and higher aeration velocity would reduce the gas leakage; whereas, the solids stream with higher solids flow rate through loop-seal 1 would carry more gas into the air reactor from the fuel reactor.  No measurable gas leakage from the fuel reactor to the cyclone could be detected. The gas leakage from the air reactor to the fuel reactor was very small and could be negligible. There were four ports for feeding aeration gas in loopseal 1. The gas leakage from the two ports near fuel reactor was considerable, and no gas leakage could be detected from the other two ports near air reactor.  For the system under investigation, the good operating conditions for minimum gas leakage could be chosen as: Ua0 = 2.5, 3.0 and 4.0 m/s, U f0 = 0.029, 0.035 and 0.046 m/s, UA1 = 4Umf or 6Umf, UA2 = 1.5Umf.  In order to apply the scaling law, the cold-flow model was operated with fluidizing gas mixture of helium and air, while varying the helium concentration from 96 vol% to 0 vol%. This was used to simulate the hydrodynamics of a hot model under the pressure varying from 1.17 to 6.97 bar.  Hydrodynamic study using fluidizing gas mixture of helium and air was carried out on the cold-flow model to investigate the influence of density ratio of solids to gas, i.e., ρs/ρf. 145 The minimum superficial gas velocity in the air reactor for stable particle transportation, Umstable increases with increasing ρs/ρf. The pressure in the cold-flow model decreases with increasing ρs/ρf while the solids hold-up in the air reactor increases with increasing ρs/ρf.  The solids circulation flux in loop 1, Gs1, depends on the combination of the maximum carrying capacity of gas-solid suspension in the air reactor, the pressure head of loop 1 and the solids stream through the hole (referred to Figure 4.3) in the lower downcomer, which depends on the fluidization in the fuel reactor. For higher ρs/ρf , higher U f0 is required to achieve good fluidization in fuel reactor of cold-flow model. The solids circulation flux in loop 2, Gs2, increases monotonically with increasing ρs/ρf .  The cold-flow model operated with fluidizing gas mixture of 96 vol% helium and 4 vol% can simulate the hot model with fuel capacity of 80 ~ 105 kW.  A CLC interconnected fluidized bed reactor has been modeled by combining the fluidization properties and the particle population balance. A procedure for optimizing the performance of atmospheric CLC reactor with using the mathematical model has been developed.  For the fuel and oxygen conversions, the modeling analysis indicated that the optimum operating condition of an atmospheric CLC reactor hot unit should be chosen as follows: fuel capacity is 80 kW, Ua0=6.6 m/s, Uf0= 0.076 m/s, UA1=4Umf, UA2=1Umf, and the temperature in air reactor is 1223 K and in fuel reactor is 1173 K. 9.2 Recommendations for future work If the problem of sealing for the gas blower is solved and the experimental system can be operated at high helium concentration while no helium is leaked into atmosphere, further investigation can reveal the behaviour in the cold-flow model of CLC:  Because the density and particle size of oxygen carrier which is currently available are 2800 ~ 4500 kg/m3 and 180 μm, respectively (refer to Table A.1), the bronze powder with 180 μm particle size can be fluidized by the mixture of air and helium in the cold-flow 146 model to simulate the hydrodynamics of the atmospheric CLC reactor which is operated by the oxygen carrier with density of 3000 ~ 4500 kg/m3.  More measurements on the local flow properties are needed in the cold-flow model, including the radial distributions of the particles velocity and voidage.  In order to decrease the gas leakage from the fuel reactor to the cyclones, the bottom end of the lower downcomer was sealed and a hole (referred to 10 in Figure 2.13) of 0.1 m in diameter was opened on the side. The experiments showed that the hole limited the solids circulation flux in loop 1, Gsl. The effect of the open area produced by the holes on the solids circulation flux and gas leakage need to be further investigated.  Heat transfer in the reactor has not been covered in the model. The consideration of heat transfer should be combined with the mass balance and the hydrodynamics to calculate the temperature distribution instead of the assumption of uniform distribution of temperature.  The effect of the reaction in the bubble phase and freeboard should be further investigated in the model of the fuel reactor. 147 REFERENCES Abad, A., T. Mattisson, A. Lyngfelt and M. Rydén (2006a). Chemical-looping combustion in a 300W continuously operating reactor system using a manganese-based oxygen carrier. Fuel, 85:1174–1185. Abad, A., J. Adánez, F. García-Labiano, L. F. de Diego and P. Gayán and J. Celaya (2006b). Mapping of the range of operational conditions for Cu-, Fe-, and Ni-based oxygen carriers in chemical-looping combustion. Chemical Engineering Science, 62(1-2): 533-549. Abad, A., T. Mattisson, A. Lyngfelt and M. Johansson (2007a).The use of iron oxide as oxygen carrier in a chemical-looping reactor. Fuel, 86(7-8):1021-1035. Abad, A., F. García-Labiano L. F. de Diego, P. Gayán and J. Adánez (2007b). Reduction kinetics of Cu-, Ni-, and Fe-based oxygen carriers using syngas (CO + H2) for chemical-looping combustion. Energy & Fuels, 21(4):1843-1853. Abellon, R. D., Z. I. Kolar, W. den Hollander, J. J. M. de Goeij, J. C. Schouten and C. M. Van Den Bleek (1997). A single radiotracer particle method for the determination of solid circulation flux in interconnected fluidized beds. Powder Technology, 92: 53-60. Adánez, J., L. F. de Diego, P. Gayán, L. Armesto and A. Cabanillas (1995). A model for prediction of carbon combustion in circulating fluidized bed combustor. Fuel, 74(7): 1049-1056. Adánez, J., P. Gayán and L. F. de Diego (1996). Modeling and simulation of sulfur retention in circulating fluidized bed combustors. Chemical Engineering Science, 51(11): 3077-3082. Adánez, J., P. Gayán, L. F. de Diego and F. García-Labiano (2003). Combustion of wood chips in a CFBC. Modeling and validation. Industrial & Engineering Chemistry Research. 42(5): 987-999. Adánez, J., L. F. de Diego, F. García-Labiano, P. Gayán, A. Abad and J. M. Palacios (2004a). Selection of oxygen carriers for chemical-looping combustion. Energy & Fuels, 18(2): 371-377. Adánez, J., F. García-Labiano, L. F. de Diego, P. Gayán, J. Celaya and A. Abad (2004b). Characterization of oxygen carriers for chemical-looping combustion. Proceedings of 7th international conference on greenhouse gas control technologies, Convetion centre, Vancouver, Canada. Adánez, J., P. Gayán, J. Celaya, L. F. de Diego, F. García-Labiano and A. Abad (2006a). Chemical looping combustion in a 10 kWth prototype using a CuO/Al2O3 oxygen carrier: effect of operating conditions on methane combustion. Industrial & Engineering Chemistry Research,45(17):6075- 6080. Adánez, J., F. García-Labiano, L. F. de Diego, P. Gayán, J. Celaya and A. Abad (2006b). Nickel- copper oxygen carriers to reach zero CO and H2 emissions in chemical-looping combustion. Industrial & Engineering Chemistry Research, 45(8): 2617-2625. 148 Adánez, J., C. Dueso, L. F. de Diego, F. García-Labiano, P. Gayán and A. Abad (2009a). Methane combustion in a 500Wth chemical-looping combustion system using an impregnated Ni-based oxygen carrier. Energy & Fuels, 23:130–142 Adánez, J., C. Dueso, L. F. de Diego, F. García-Labiano, P. Gayán and A. Abad (2009b). Effect of fuel gas composition in chemical-looping combustion with Ni-based oxygen carriers. 2. fate of light hydrocarbons. Industrial & Engineering Chemistry Research, 48(5):2509-2518. Berguerand, N. and A. Lyngfelt (2008). Design and operation of a 10 kWth chemical-looping combustor for solid fuels –testing with South African coal. Fuel,87: 2713–2726. Berruti,F., and N. Kalogerakis (1989). Modeling the internal flow structure of circulating fluidized beds. Canadian Journal of Chemical Engineering, 67: 1010-1014. Bi, H., P. Jiang, R. Jean and L. S. Fan (1992). "Coarse-particle effects in a multisolid circulating fluidized bed for catalytic reactions." Chemical Engineering Science, 47(12): 3113-3124. Brandvoll, Ø., L. Kolbeinsen, O. Niklas and O. Bolland (2003). Chemical looping combustion- reduction of nickel oxide/nickel aluminate with hydrogen. The 6 th italian conference on chemical and process engineering, Pisa, Italy. Brandvoll, Ø. (2004). Inherent CO2 capture using chemical looping combustion in a natural gas fired power cycle. Journal of Engineering for gas turbines and power (Transactions of the ASME), 126: 316-321. Chandel, M. K., A. Hoteit and A. delebarre (2009). Experimental investigation of some metal oxides for chemical looping combustion in a fluidized bed reactor. Fuel, 88(5): 898-908. Chen, Z. (2007). Hydrodynamics, stability and scale-up of slot-rectangular spouted beds. Ph.D. thesis, The University of British Columbia. Cho, P., T. Mattisson and A. Lyngfelt (2002). Reactivity of iron oxide with methane in a laboratory fluidized bed-application of chemical-looping combustion. Proceedings of the 7th International Conference on Circulating Fluidized Beds, Niagara Falls, Ontario, Canada. Cho, P., T. Mattisson and A. Lyngfelt (2004). Comparison of iron-, nickel-, copper- and manganese- based oxygen carriers for chemical-looping combustion. Fuel, 83: 1215-1225. Cho, P., T. Mattisson and A. Lyngfelt (2005). Carbon formation on nickel and iron oxide-containing oxygen carriers for chemical-looping combustion. Industrial & Engineering Chemistry Research, 44(4): 668-676. Cho, P., T. Mattisson and A. Lyngfelt (2006). Defluidization Conditions for a Fluidized Bed of Iron Oxide-, Nickel Oxide-, and Manganese Oxide-Containing Oxygen Carriers for Chemical-Looping Combustion. Industrial & Engineering Chemistry Research, 45(3): 968-977. Corbella, B. M., L. F. de Diego, F. García-Labiano, J. Adánez, and J. M. Palacios (2005a). The performance in a fixed bed reactor of copper-based oxides on titania as axygen carriers for chemical looping combustion of methane. Energy & Fuels, 19(2): 433-441. 149 Corbella, B. M., L. F. de Diego, F. García-Labiano, J. Adánez, and J. M. Palacios (2005b). Characterization Study and Five-Cycle Tests in a Fixed-Bed Reactor of Titania-Supported Nickel Oxide as Oxygen Carriers for the Chemical-Looping Combustion of Methane. Environmental Science and Technology, 39(15): 5796-5803. Corbella, B. M. and L. F. de Diego, F. García-Labiano, J. Adánez, and J. M. Palacios (2006a). Performance in a fixed-bed reactor of titania-supported nickel oxide as oxygen carriers for the chemical-looping combustion of methane in multicycle Tests. Industrial & Engineering Chemistry Research, 45(1): 157-165. Corbella, B. M. and L. F. de Diego, F. García-Labiano, J. Adánez, and J. M. Palacios (2006b). Characterization and performance in a multicycle Test in a fixed-bed reactor of silica-supported copper oxide as oxygen carrier for chemical-looping combustion of methane. Energy & Fuels, 20(1): 148-154. Corbella, B. M. and J. M. Palacios (2007). Titania-supported iron oxide as oxygen carrier for chemical-looping combustion of methane. Fuel, 86(1-2): 113-122. Chuang, S. Y., J. S. Dennis, A. N. Hayhurst and S. A. Scott (2008). Development and performance of Cu-based oxygen carriers for chemical-looping combustion. Combustion and Flame, 154(1/2): 109-121. Darton, R.C, R. D. LaNauze, J. F. Daridson and D. Hairison (1977). Bubble growth due to coalescence in fluidized beds. Transactions of the Institution of Chemical Engineers, 55:274-280. Davidson, J. F. (2000). Circulating fluidised bed hydrodynamics. Powder Technology, 113: 249-260. Davidson, J. F. and D. Harrison (1963). Fluidized particles, Cambridge: Cambridge University Press. de Diego, L. F., F. García-Labiano, J. Adánez, P. Gayán and A. Abad (2004). Development of Cu- based oxygen carriers for chemical-looping combustion. Fuel, 83: 1749-1757. de Diego, L. F., P. Gayán, F. García-Labiano, J. Celayn and A. Abad (2005). Impregnated CuO/Al2O3 oxygen carriers for chemical-looping combustion: avoiding fluidized bed agglomeration. Energy & Fuels, 19(5):1850-1856. de Diego, L. F., F. García-Labiano, P. Gayán, J. Celayn, J. M. Palacios and J. Adánez (2007). Operation of a 10 kWth chemical-looping combustor during 200 h with a CuO-Al2O3 oxygen carrier. Fuel, 86(7-8):1036-1045. de Diego, L. F., M. Ortiz, J. Adánez, F. García-Labiano, A. Abad and P. Gayán (2008). Synthesis gas generation by chemical-looping reforming in a batch fluidized bed reactor using Ni-based oxygen carriers. Chemical Engineering Journal (Amsterdam, Netherlands), 144(2): 289-298. Ellis, N. (2003). Hydrodynamics of gas-solid turbulent fluidized beds. Ph.D. thesis, The University of British Columbia. Erri, P. and A. Varma (2007a). Spinel-Supported Oxygen Carriers for Inherent CO2 Separation during Power generation. Industrial & Engineering Chemistry Research, 46, 8597-8601 . 150 Erri, P. and A. Varma (2007b). Solution combustion synthesized oxygen carriers for chemical looping combustion. Chemical Engineering Science, 62(18-20): 5682-5687. Erri, P. and A. Varma. (2009). Diffusional effects in nickel oxide reduction kinetics. Industrial & Engineering Chemistry Research, 48(1): 4-6. Fakeeha, A. H., M. A. Soliman and A. A. Lbrahim (2000). Modeling of a circulating fluidized bed for ammoxidation of propane to acrylonitrile. Chemical Engineering and Processing, 39: 161-170. Fox, D., Y. Molodtsof and J. F. Large (1989). Control mechanisms of fluidized solids circulation between adjacent vessels. AIChE Journal, 35(12): 1933-1940. García-Labiano, F., L. F. de Diego, J. Adánez, A. Abad and P. Gayán (2004). Reduction and oxidation kinetics of a copper-based oxygen carrier prepared by impregnation for chemical-looping combustion. Industrial & Engineering Chemistry Research, 43(26): 8168-8177. García-Labiano, F., J. Adánez, L. F. Diego, P. Gayán and A. Abad (2006). Effect of Pressure on the Behavior of Copper-, Iron-, and Nickel-Based Oxygen Carriers for Chemical-Looping Combustion. Energy & Fuels, 20(1): 26-33. García-Labiano, F., L. F. de Diego, P. Gayán, J. Adánez, A. Abad and C. Dueso (2009). Effect of fuel gas composition in chemical-looping combustion with Ni-based oxygen carriers. 1. fate of sulfur. Industrial & Engineering Chemistry Research, 48(5): 2499-2508. Gayán, P., L. F. de Diego, F. García-Labiano, J. Adánez, A. Abad and C. Dueso (2008). Effect of support on reactivity and selectivity of Ni-based oxygen carriers for chemical-looping combustion. Fuel, 87(12): 2641-2650. Gayán, P., C. Dueso, A. Abad, J. Adánez, L. F. de Diego and F. García-Labiano (2009). NiO/Al2O3 oxygen carriers for chemical-looping combustion prepared by impregnation and deposition- precipitation methods. Fuel, 88(6): 1016-1023. Glicksman, L. R. (1984). Scaling relationships for fluidized beds. Chemical Engineering Science, 39(9): 1373-1379. Glicksman, L. R. (1988). "Scaling relationships for fluidized beds." Chemical Engineering Science, 43(6): 1419-1412. Glicksman, L.R., D. Westphalen, C. Brereton and J. R. Grace (1990). Verification of the scaling laws for circulating fluidized beds. Proceedings of the 3 th international conference on circulating fluidized beds, Nagoya, Japan. Glicksman, L. R., M. Hyre and K. Woloshun (1993). Simplified scaling relationships for fluidized beds. Powder Technology, 77: 177-199. Glicksman, L. R., M. Hyre and P. A. Farrell (1998). Importance of the solid-to-gas density ratio for scaling fluidized bed hydrodynamics. Process of the 9th International Conference on Fluidization, Fluidization IX. Glicksman, L. R., (2003). Fluidized bed scaleup. Handbook of fluidization and fluid-particle systems: 343-378. Editor: W. C. Yang, U.S.A (2003), publisher: Marcel Dekker, Inc., New York. 151 Grace, J. R., H. Bi and M. Golriz (2003) Circulating fluidized beds. Handbook of fluidization and fluid-particle systems: 485-544. Editor: W. C. Yang, U.S.A (2003), publisher: Marcel Dekker, Inc., New York. Grönkvist, S. (1995). Modeling two phase gas-solid reactors for a chemical looping combustion system. Dept. of Chemical engineering & Technology, Royal Institute of Technology. Harris, B. J. and J. F. Davidson (1993). Modeling options for circulating fluidized beds: a core/annuls deposition model. Proceedings of the 4th international conference on circulating fluidized beds. Harvey, S. P. and H. Richter (1994). A high-efficiency gas turbine power generation cycle with solid oxide fuel cell technology and chemical looping fuel combustion. Proceedings 1994 international mechanical engineering congress and exposition, Chicago, Illinois. He, F., Y. Wei, H. Li and H. Wang (2009). Synthesis Gas Generation by Chemical-Looping Reforming Using Ce-Based Oxygen Carriers Modified with Fe, Cu, and Mn Oxides. Energy & Fuels, 23(4): 2095-2102. Herzog, H., E. Drake and E. Adams (1997). CO2 capture, reuse, and storage technologies for mitigating global climate change, A white paper (final report) from Energy Laboratory, Massachusetts Institute of Technology, Cambridge, MA, DOE Order No. DE-AF22 96PC01257. Horio, M., H. Ishii and Y. Kobukai and N. Yamanishi (1989). A scaling law for circulating fluidized beds. Journal of Chemical Engineering of Japan, 22(6): 587-592. Horio, M., A. Nonaka, Y. Sawa and I. Muchi (1986). A new similarity rule for fluidized bed scale- up. AIChE Journal, 32(9): 1466-1482. Hossain, M., K. E. Sedor and H. I. de Lasa (2007a). Co-Ni/Al2O3 oxygen carrier for fluidized bed chemical-looping combustion: desorption kinetics and metal-support interaction. Chemical Engineering Science, 62(18-20):5464-5472. Hossain, M. and H. I. de Lasa (2007b). Reactivity and stability of Co-Ni/Al2O3 oxygen carrier in multicycle CLC. AIChE Journal , 53(7):1817-1829. Hossain, M. M., D. Lopez, J. Herrera and H. I. de Lasa (2009). Nickel on lanthanum-modified γ- Al2O3 oxygen carrier for CLC: reactivity and stability. Catalysis Today,143(1-2): 179-186. Hoteit, A., M. K. Chandel and A. Delebarre (2009). Nickel- and copper-based oxygen carriers for chemical looping combustion. Chemical Engineering & Technology , 32(3): 443-449. Hua, Y., G. Flamant, J. Lu and D. Gauthier (2004). Modeling of axial and radial solid segregation in a CFB boiler. Chemical Engineering and Processing, 43(8): 971-978. Ishida, M. and H. Jin (1994). A novel combustor based on chemical-looping reactions and its reaction kinetics. Journal of chemical engineering of Japan, 27(3): 296-301. Ishida, M., H. Jin and T. Okamoto (1996a). A fundamental study of a new kind of medium material for chemical-looping combustion. Energy & Fuels, 10(4): 958-963. 152 Ishida, M. and H. Jin (1996b). A novel chemical-looping combustor without NOx formation. Industrial & Engineering Chemistry Research, 35(7): 2469-2472. Ishida, M. and H. Jin (1997). CO2 recovery in a power plant with chemical looping combustion. Energy Conversion and Management, 38(Suppl.): 187-192. Ishida, M., H. Jin and T. Okamoto (1998). Kinetic bahavior of solid particle in chemical-looping combustion: suppressing carbon deposition in reduction. Energy & Fuels, 12(2): 223-229. Ishida, M. and H. Jin (2001). Fundamental study on a novel gas turbine cycle. Journal of Energy Resource Technology (Transactions of the ASME), 123: 10-14. Ishida, M., M. Yamamoto and T. Ohba (2002). Experimental results of chemical-looping combustion with NiO/NiAl2O4 particle circulation at 1200 oC. Energy Conversion and Management, 43: 1469-1478. Ishida, M., K. Takeshita, K. Suzuk and T. Ohba (2005). Application of Fe2O3-Al2O3 composite particles as solid looping material of the chemical-loop combustor. Energy & Fuels, 19(6): 2514- 2518. Jerndal, E., T. Mattission and A. Lyngfelt (2009). Investigation of different NiO/NiAl2O4 particles as oxygen carriers for chemical-looping combustion. Energy & Fuels, 23(2): 665-676. Jiang, M. X., T. C. Keener and S. J. Khang (1995). The use of a circulating fluidized bed absorber for the control of sulfur dioxide emissions by calcium oxide sorben via in situ hydration. Powder Technology, 85: 115-126. Jiang, P., F. Wei and L. Fan (2002). General approaches to reactor design. Handbook of fluidization and fluid-particle systems: 309-420. Editor: W. C. Yang, U.S.A (2003), publisher: Marcel Dekker, Inc., New York. Jin, H., T. Okamoto and M. Ishida (1998). Development of a novel chemical-looping combustion: Synthesis of a looping material with a double metal oxide of CoO-NiO. Energy & Fuels, 12(6): 1272-1277. Jin, H., T. Okamoto, M. Ishida (1999). Development of a novel chemical-looping combustion: Synthesis of a solid looping material of NiO/NiAl2O4. Industrial & Engineering Chemistry Research, 38(1): 126-132. Jin, H. and M. Ishida (2001). Reactivity study on a novel hydrogen fueled chemical-looping combustion. International Journal of Hydrogen Energy, 26: 889-894. Jin, H. and M. Ishida (2002). Reactivity study on natural-gas-fueled chemical-looping combustion by a fixed-bed reactor. Industrial & Engineering Chemistry Research, 41(16): 4004-4007. Jin, H. and M. Ishida (2004). A new type of coal gas fueled chemical-looping combustion. Fuel, 83: 2411-2417. 153 Johansson, E., A. Lyngfelt, T. Mattisson and F. Johnsson (2002). A circulating fluidized bed combustor system with inherent CO2 separation-application of chemical looping combustion. Proceedings of the 7th International Conference on Circulating Fluidized Beds, Niagara Falls, Ontario, Canada. Johansson, E., A. Lyngfelt, T. Mattisson and F. Johnsson (2003). Gas leakage measurements in a cold model of an interconnected fluidized bed for chemical-looping combustion. Powder Technology, 134: 210-217. Johansson, M., T. Mattisson and A. Lyngfelt (2004). Investigation of Fe2O3 with MgAl2O4 for chemical-looping combustion. Industrial & Engineering Chemistry Research, 43(22): 6978-6987. Johansson, E., T. Mattisson, A. Lyngfelt and H. Thunman (2006a). Combustion of syngas and natural gas in a 300W chemical-looping combustor. Chemical Engineering Research and Design, 84(A9):819–827. Johansson, E., T. Mattisson, A. Lyngfelt and H. Thunman (2006b). A 300W laboratory reactor system for chemical-looping combustion with particle circulation. Fuel, 85: 1428–1438. Johansson, M., T. Mattisson and A. Lyngfelt (2006c). Use of NiO/NiAl2O4 particles in a 10 kW chemical-looping combustor. Industrial & Engineering Chemistry Research, 45(17):5911-5919. Johansson, M., T. Mattisson and A. Lyngfelt (2006d). Investigation of Mn3O4 with stabilized ZrO2 for chemical-looping combustion. Chemical Engineering Research and Design, 84(A9):807-818. Johansson, M., T. Mattisson and A. Lyngfelt (2006e). Creating a synergy effect by using Mixed oxides of iron- and nickel oxides in the combustion of methane in a chemical-looping combustion reactor. Energy & Fuels, 20(6):2399-2407. Johansson, M., T. Mattisson, A. Lyngfelt and A. Abad (2008). Using continuous and pulse experiments to compare two promising nickel-based oxygen carriers for use in chemical-looping technologies. Fuel, 87(6): 988-1001. Jung, J. and I. K. Gamwo. (2008). Multiphase CFD-based models for chemical looping combustion process: Fuel reactor modeling. Powder Technology, 183:401–409. Kolbitsch, P., T. Pröll, J. Bolhàr-Nordenkampf and H. Hofbauer (2009a). Design of a chemical looping combustor using a dual circulating fluidized bed reactor system. Chemical Engineering & Technology, 32(3):398–403. Kolbitsch, P., T. Pröll, J. Bolhàr-Nordenkampf and H. Hofbauer (2009b). Characterization of chemical looping pilot plant performance via experimental determination of solids conversion. Energy & Fuels, 23: 1450–1455. Kolbitsch, P., T. Pröll and H. Hofbauer (2009c). Modeling of a 120kW chemical looping combustion reactor system using a Ni-based oxygen carrier. Chemical Engineering Science, 64:99- 108. Kolbitsch, P, J. Bolhàr-Nordenkampf, T. Pröll and H. Hofbauer (2009d). Comparison of two Ni- based oxygen carriers for chemical looping combustion of natural gas in 140 kW continuous looping operation. Industrial & Engineering Chemistry Research, 48: 5542–5547 154 Kronberger, B., E. Johansson, G. Löffler, T. Mattisson, A. Lyngfelt and H. Hofbauer (2004). A two- compartment fluidized bed reactor for CO2 capture by chemical-looping combustion." Chemical Engineering & Technology, 27(12): 1318-1326. Kronberger, B., A. Lyngfelt, G. Löffler and H. Hofbauer (2005). Design and fluid dynamic analysis of a bench-scale combustion system with CO2 separaation-chemical-looping combustion. Industrial & Engineering Chemistry Research, 44(3): 546-556. Kunii, D. and O. Levenspiel (1991). The design of noncatalystic gas-solid reactors. Fluidization Engineering, Butterworth-Heinemann: 449-479. Kuramoto, M., T. Furusawa and D. Kunii (1985). Development of a new system for circulating fluidized particles within a single vessel. Powder Technology, 44: 77-84. Lee, J., C. Park, S. Choi, Y. Song, Y. Kim and H. Yang (2005). Redox characteristics of various kinds of oxygen carriers for hydrogen fueled chemical-looping combustion. Journal of Industrial and Engineering Chemistry (Seoul, Republic of Korea), 11(1): 96-102. Leion, H., A. Lyngfelt, M. Johansson, E. Jerndal and T. Mattisson (2008). The use of ilmenite as an oxygen carrier in chemical-looping combustion. Chemical Engineering Research and Design, 86(9):1017-1026. Leion, H., T. Mattisson and A. Lyngfelt (2009). Use of ores and industrial Products as oxygen carriers in chemical-looping combustion. Energy & Fuels, 23(4): 2307-2315. Levenspiel, O., D. Kunii and T. Fitzgerald (1968). The processing of solids of changing size in bubbling fluidized beds. Powder Technology, 2:87-96. Levenspiel, O. (1972). Fluid-particle reactors. Chemical Reaction Engineering. New York, Johan Wiley & Sons, Inc: 357-377 Linderholm, C., A. Abad, T. Mattisson and A. Lyngfelt (2008). 160 h of chemical-looping combustion in a 10kW reactor system with a NiO-based oxygen carrier. International Journal of Greenhouse Gas Control, 2(4):520-530. Linderholm, C., T. Mattisson and A. Lyngfelt (2009). Long-term integrity testing of spray-dried particles in a 10-kW chemical-looping combustor using natural gas as fuel. Fuel, 88(11):2083-2096. Lőffler, G., S. Kaiser, K. Bosch and H. Hofbauer (2003). Hydrodynamics of a dual fluidized-bed gasifier---Part I: simulation of a riser with gas injection and diffuser. Chemical Engineering Science, 58(4197-4213). Luecke, K., E.-U. Hartge and J. Werther (2004). A 3D model of combustion in large-scale circulating fluidzed bed boilers. International Journal of Chemical Reactor Engineering 2(Article A11): http://www.bepress.com/ijcre/vol2/A11. Lyngfelt, A., B. Kronberger, J. Adánez, J. Morin and P. Hurst (2004). The grace project. Devlopment of oxygen carrier particles for chemical-looping combustion. Design and operation of a 10KW chemical-looping combustor. Proceedings of 7th international conference on greenhouse gas control technologies, Convetion centre, Vancouver, Canada. 155 Lyngfelt, A., B. Leckner and T. Mattisson (2001). A fluidized-bed combustion process with inherent CO2 separation; application of chemical-looping combustion. Chemical Engineering Science, 56: 3101-3113. Malecka, B., E. Crozdz-Ciesla E and A. Malecki (2002). Non-isothermal studies on mechanism and kinetics of thermal decomposition of cobalt (II) oxalate dehydrate. Journal of Thermal Analysis and Calorimetry, 68(3): 819–831. Malecka, B., E. Crozdz-Ciesla E and A. Malecki (2004). Mechanism and kinetics of thermal decomposition of zinc oxalate. Thermochimica Acta, 423(1-2):13–18. Mao, D., J. R. Edwards, A. V. Kuznetsov and R. Srivastava (2001). Numerical simulation of particle flow, mixing and chemical reaction in the circulating fluidized bed absorbers. Proceedings of 2001 ASME international mechanical engineering congress and exposition, New York. Mao, D., J. R. Edwards, A. V. Kuznetsov and R. Srivastava (2002). Particle flow, mixing, and chemical reaction in circulating fluidized bed absorbers. Chemical Engineering Science, 57(3107- 3117). Mattisson, T., A. Lyngfelt, P. Cho (2000). Possibility of using iron oxide as an oxygen carrier for combustion of methane with removal of CO2-application of chemical-looping combustion. Proceedings of the 5th International Conference on Greenhouse Gas Control Technologies, Cairns, Australia. Mattisson, T., A. Lyngfelt and P. Cho (2001a). The use of iron oxide as oxygen carrier in chemical- looping combustion of methane with inherent separation of CO2. Fuel, 80: 1953-1962. Mattisson, T. and A. Lyngfelt (2001b). Application of chemical-looping combustion with capture of CO2. The 2nd Nordic Minisymposium on Carbon Dioxide Capture and Storage, Goteborg. Mattisson, T., A. Järdnäs and A. Lyngfelt (2003). Reactivity of some metal oxides supported on alumina with alternatiing methane and oxygen-application for chemical-looping combustion. Energy & Fuels, 17(3): 643-651. Mattisson, T., M. Johansson and A. Lyngfelt (2006). The use of NiO as an oxygen carrier in chemical-looping combustion. Fuel, 85(5-6): 736-747. Mattisson, T., F. García-Labiano, B. Kronberger, A. Lyngfelt, J. Adánez and H. Hofbauer (2007). Chemical-looping combustion using syngas as fuel. International Journal of Greenhouse Gas Control, 1(2): 158-169. Mattisson, T., M. Johansson, E. Jernadal and A. Lyngfelt (2008). The reaction of NiO/NiAl2O4 particles with alternating methane and oxygen. Canadian Journal of Chemical Engineering, 86(4): 756-767. Mukadi, L., C. Guy and R. Legros (2000). Predictions of gas emissions in an internally circulating fluidized bed combustor for treatment of industrial solid wastes. Fuel, 79: 1125-1136. Namkung, W., and Kim, S.D. (1998). Gas backmixing in a circulating fluidized bed. Powder Technology, 99:70-78. 156 Noorman, S., M. van S. Annaland and H. Kuipers (2007). Packed Bed Reactor Technology for Chemical-Looping Combustion. Industrial & Engineering Chemistry Research, 46(12):4212-4220. Patience, G. S. and J. Chaouki (1993). Gas phase hydrodynamics in the riser of a circulating fluidized bed. Chemical Engineering Science, 48(18): 3195-3205. Pröll, T., K. Rupanovits, P. Kolbitsch and J. Bolhàr-Nordenkampf (2009a). Cold flow model study on a dual circulating fluidized bed system for chemical looping processes. Chemical Engineering & Technology, 32(3):418-424. Pröll, T., P. Kolbitsch, J. Bolhàr-Nordenkampf and H. Hofbauer (2009b), A novel dual circulating fluidized bed system for chemical looping process. AIChE Journal (Environmental and Energy Engineering), 55(12):3255-3266. Puchyr, D. M., A. K. Mehrotra and L. A. Behie (1997). Modelling a criculating fluidized bed riser reactor with gas-solids downflow at the wall. Canadian Journal of Chemical Engineering, 75(2): 317-326. Ramkrishna, D. and A. W. Mahoney. (2002). Population balance modeling: promise for the future. Chemical Engineering Science, 57(4):595-606. Readman, J. E., A. Olafsen, J. B. Smith and R. Blom (2006). Chemical looping combustion using NiO/NiAl2O4: mechanisms and kinetics of reduction-oxidation (red-ox) reactions from in situ powder X-ray diffraction and thermogravimetry experiments. Energy & Fuels, 20(4): 1382-1387. Rydén, M. and A. Lyngfelt (2004). Hydrogen and power production with integrated carbon dioxide capture by chemical-looping reforming. Proceedings of 7 th international conference on greenhouse gas control technologies, Convetion centre, Vancouver, Canada. Rydén, M., A. Lyngfelt and T. Mattisson (2008a). Chemical-looping combustion and chemical- looping reforming in a circulating fluidized-bed reactor using Ni-based oxygen carriers. Energy & Fuels, 22(4):2585-2597. Rydén, M., A. Lyngfelt, T. Mattisson, D. Chen and A. Holmen (2008b). Novel oxygen-carrier materials for chemical-looping combustion and chemical-looping reforming: LaxSr1-xFeyCo1-yO3-δ perovskites and mixed-metal oxides of NiO, Fe2O3 and Mn3O4. International Journal of Greenhouse Gas Control, 2(1):21-36. Ryu, H., D.-H. Bae, K.-H. Ham, S.-Y. Lee, G.-T. Jin and J.-H. Choi (2001). Oxidation and reduction characteristics of oxygen carrier particles and reaction kinetics by unreacted core model. Korean Journal of Chemical Engineering, 18(6): 831-837. Ryu, H.-J., N.-Y. Lim, D.-H. Bae and G.-T. Jin (2003a). Carbon deposition characteristics and regenerative ability of oxygen carrier particles for chemical-looping combustion. Korean Journal of Chemical Engineering, 20(1): 157-162. Ryu, H., D.-H. Bae and G.-T. Jin (2003b). Effect of temperature on reduction reactivity of oxygen carrier particles in a fixed bed chemical-looping combustor. Korean Journal of Chemical Engineering, 20(5): 960-966. 157 Ryu, H., S.-Y. Lee, Y. C. Park and M.-H. Park (2008a). Solid circulation flux and gas leakage measurements in an interconnected bubbling fluidized beds. International Journal of Applied Science, Engineering and Technology, 4(2):113~118. Ryu, H., Y.-C. Park, S.-H. Jo and M.-H. Park (2008b). Development of novel two-interconnected fluidized bed system. Korean Journal of Chemical Engineering, 25(5):1178-1183. Ryu, H., D. Shun, D.-H. Bae and M.-H. Park (2009). Syngas combustion characteristics of four oxygen carrier particles for chemical-looping combustion in a batch fluidized bed reactor. Korean Journal of Chemical Engineering, 26(2): 523-527. Sedor, K. E., M. M. Hossain and H. I. de Lasa (2008a). Reduction kinetics of a fluidizable nickel- alumina oxygen carrier for chemical-looping combustion. Canadian Journal of Chemical Engineering, 86(3):323-334. Sedor, K. E., M. M. Hossain and H. I. de Lasa (2008b). Reactivity and stability of Ni/Al2O3 oxygen carrier for chemical-looping combustion (CLC). Chemical Engineering Science, 63(11):2994-3007. Shen, L., M. Zheng, J. Xiao and R. Xiao (2008). A mechanistic investigation of a calcium-based oxygen carrier for chemical looping combustion. Combustion and Flame, 154(3): 489-506. Shulman, A., C. Linderholm, T. Mattisson and A. Lyngfelt (2009). High reactivity and mechanical durability of NiO/NiAl2O4 and NiO/NiAl2O4/MgAl2O4 oxygen carrier particles used for more than 1000 h in a 10 kW CLC reactor. Industrial & Engineering Chemistry Research, 48(15): 7400-7405. Siriwardane, R., J. Poston, K. Chaudhari, A. Zinn, T. Simony and C. Robinson (2007). Chemical- looping combustion of simulated synthesis gas using nickel oxide oxygen carrier supported on bentonite. Energy & Fuels, 21 (3): 1582-1591. Sit, S. P. and J. R. Grace. (1981). Effect of bubble interaction on interphase mass transfer in gas fluidized beds. Chemical Engineering Science, 36(2):327-35. Snieders, F. F., A. C. Hoffmann, D. Cheesman, J. G. Yates, M. Stein and J. P. K. Seville (1999). The dynamics of large particles in a four-compartment interconnected fluidized bed. Powder Technology, 101: 229-239. Snip, O. C., M. Woods, R. Korbee, J. C. Schouten and C. M. Van Den Bleek (1996). Regenerative removal of SO2 and NOx for a 150 MWe power plant in an interconnected fluidized bed facility. Chemical Engineering Science, 51(10): 2021-2029. Son, S. R. and S. D. Kim (2006). Chemical-Looping Combustion with NiO and Fe2O3 in a Thermobalance and Circulating Fluidized Bed Reactor with Double Loops. Industrial & Engineering Chemistry Research, 45(8):2689-2696. Song, K. S., Y. S. Seo, H. K. Yoon and S. J. Cho (2003). Characteristics of the NiO/Hexaaluminate for chemical looping combustion. Korean Journal of Chemical Engineering, 20(3): 471-475. Song, Y., J. B. Lee, C. S. Park, G. J. Hwang, S. Choi, H. S. Yang and Y. H. Kim (2006). Synthesis and redox properties of NiO/NiAl2O4 oxygen carriers for hydrogen-fueled chemical-looping combustion. Journal of Industrial and Engineering Chemistry (Seoul, Republic of Korea) , 12(2): 255-260. 158 Song, Q., R. Xiao, Z. Deng, W. Zheng, L. Shen and J. Xiao (2008a). Multicycle study on chemical- looping combustion of simulated coal gas with a CaSO4 oxygen carrier in a fluidized bed reactor. Energy & Fuels, 22(6): 3661-3672. Song, Q., R. Xiao, Z. Deng, L. Shen, J. Xiao and M. Zhang (2008b). Effect of temperature on reduction of CaSO4 oxygen carrier in chemical-looping combustion of simulated coal gas in a fluidized bed reactor. Industrial & Engineering Chemistry Research, 47(21): 8148-8159. Song, Q., R Xiao, Z. Deng, H. Zhang, L. Shen, J. Xiao and M. Zhang (2008c). Chemical-looping combustion of methane with CaSO4 oxygen carrier in a fixed bed reactor. Energy Conversion and Management, 49(11): 3178-3187. Song, Q., R. Xiao, Z. Deng, L. Shen and M. Zhang (2009). Reactivity of a CaSO4-oxygen carrier in chemical-looping combustion of methane in a fixed bed reactor. Korean Journal of Chemical Engineering, 26(2): 592-602. Stearns, R. F., R. R. Johanson, R. M. Jackson and C. A. Larson (1951). Flow measurement with orifice meters. D.Van Nostrand Company, Inc., Toronto. Talukdar, J., P. Basu and E. Joos (1993). Sensitivity analysis of a performance predictive model for circulating fluidized bed boiler furnaces. Proceedings of the 4th international conference on circulating fluidized beds. Talukdar, J., and Basu, P. (1995). Modelling of nitric oxide emission from a circulating fluidized bed combustor. Fluidization VIII, 829-837. Teplitskii, Y. S., V. A. Borodulya and E. F. Nogotov (2003). Axial solids mixing in a circulating fluidized bed. International Journal of Heat and Mass Transfer, 46: 4335-4343. Teplitskii, Y. S. and E. F. Nogotov (2002). Mixing of particles in a criculating fluidized bed. Journal of Engineering Physics and Thermophysics, 75(3): 529-539. Tian, H., Q. Guo, J. Chang (2008a). Investigation into decomposition behavior of CaSO4 in Chemical-Looping Combustion. Energy & Fuels, 22(6): 3915-3921. Tian, H., K. Chaudhari, T. Simonyi, J. Poston, T. Liu, T. Sanders, G. Veser, R. Siriwardane (2008b). Chemical-looping combustion of coal-derived synthesis gas over copper oxide oxygen carriers. Energy & Fuels, 22(6): 3744-3755. Vandewalle, C., J. Baeyens and A. Claerbout (2002). Hydrodynamic model for the gas flow in circulating fluidized bed reactors. Industrial & Engineering Chemistry Research, 41(24): 5983- 5989. van der Meer, E.H., P. B. Thorpe and J. F. Davidson (1999). Dimensionless groups for practicable similarity of circulating fluidized beds. Chemical Engineering Science, 54:5369-5376 Villa, R., C. Cristiani, G. Groppi, L. Lietti, P. Forzatti, U. Cornaro and S. Rossini (2003). Ni based mixed oxide materals for CH4 oxidation under redox cycle conditions. Journal of Molecular catalysis A: chemical, (204-205): 637-646. 159 Wang, Q., Z. Luo, M. Ni and K. Cen (2003). Particle population balance model for a circulating fluidized bed boiler. Chemical Engineering Journal, 93(121-133). Wang, Z. (2006). Experimental studies and CFD simulations of conical spouted bed hydrodynamics. Ph.D. thesis, The University of British Columbia. Yang, W. (2002). Particle characterization and dynamics. Handbook of fluidization and fluid- particle systems: 1-27. Editor: W. C. Yang, U.S.A (2003), publisher: Marcel Dekker, Inc., New York. Zafar, Q., T. Mattisson and B. Gevert (2006). Redox investigation of some oxides of transition-state metals Ni, Cu, Fe, and Mn supported on SiO2 and MgAl2O4. Energy & Fuels, 20(1): 34-44. Zafar, Q., A. Abad, T. Mattisson, B. Gevert and M. Strand (2007a). Reduction and oxidation kinetics of Mn3O4/Mg-ZrO2 oxygen carrier particles for chemical-looping combustion. Chemical Engineering Science, 62(23):6556-6567 Zafar, Q., A. Abad, T. Mattisson and B. Gevert (2007b). Reaction kinetics of freeze-Granulated NiO/MgAl2O4 oxygen carrier particles for chemical-looping combustion. Energy & Fuels, 21(2):610-618. 160 APPENDIX A SUMMARY OF PREVIOUS WORKS ON OXYGEN CARRIERS AND COLD-MODEL HYDRODYNAMICS 161 Table A.1. Summary of investigation on oxygen carrier (metal oxide) for Chemical-looping combustion* Authors Particle Experimental conditions Remarks Leion et al. 2009 Sweden Fe- and Mn-based ores and industrial products pd : 125~180 m Reactor: Fluidized bed Gases: 50% H2+50% CO at 450 mLn/min (reduction) 5%O2+N2 at 1000 mLn/min (oxidation). Temperature: 935–945 o C All materials had a high reactivity with syngas. Some materials such as the Mn-based and the Fe-based had a high reactivity with methane making them possible candidates for CLC with gaseous fuel. Some of the materials, especially the Mn-based ones, showed poor mechanical stability and poor fluidizing properties. Linderholm et al. 2009 Sweden NiO(40%) + NiAl2O4 pd : 125~180 m Apparent density: 3250~3800 kg/m3 Reactor: Circulating fluidized bed Gases: air (oxidation) natural gas (90% CH4 for reduction) Temperature: 625~950 o C The performance of the oxygen carrier in the CFB is test and the oxygen carrier is suitable for the CLC process using natural gas as fuel. Jerndal et al. 2009 Sweden NiO+NiAl2O4 pd : 125~180 m Apparent density: 1470~4460 kg/m3 Reactor: Fluidized bed Gases: 5% O2 + N2 (oxidation) CH4 (reduction) Flow rates: 8~28Umf (oxidation) 3~9Umf (reduction) Temperature: 950 o C Two of the investigated particles displayed a combination of high reactivity and strength as well as excellent fluidization behavior and should be feasible for use in a chemical-looping combustion unit Shulman et al. 2009 Sweden NiO(40%)+NiAl2O4 NiO(40%)+NiAl2O4(42%)+ MgAl2O4(18%) pd : 125~180 m Reactor: Fluidized bed Gases: 5%O2+N2 (oxidization) CH4 (reduction) Temperature: 850~950 o C Both of the oxygen carriers demonstrate high reactivity and mechanical durability after having been used for more than 1000 h. * The references are listed according to the country and for the same country the references are listed chronologically. For gases, unit of “%”means “vol%); for solids, unit of “%”means “wt%”. 162 Table A.1 continued Authors Particle Experimental conditions Remarks Mattisson et al. 2008 Sweden NiO(20%)+NiAl2O4 Reactor: TGA Gases: 10%CH4+10% H2O+5% CO2 + 75% N2 (reduction) 5%O2 + N2 (oxidation) Flow rate: 40 mL/min Temperature: 750–950 o C Reactor: Fluidized bed Gases: 50%CH4+50% H2O (reduction) 5%O2 + N2 (oxidation). Flow rate: 1.8~3.8Umf Temperature: 950 o C A simple reactor model using kinetic data from a previous study predicted the gas yield during the reduction in the fluidized bed experiments with reasonable accuracy. The oxygen carrier system investigated in this work shows high promise for use in a real CLC system. Rydén et al. 2008a Sweden NiO(20%)+MgAl2O4 NiO(18%~20%) + Al2O3 pd : 90~250 m Reactor: Circulating fluidized bed Natural gas at 0.9~1.6 Ln/min Air at 3~10 Ln/min air Temperature: 800~950 o C There was significant formation of solid carbon in the fuel reactor. Adding 30% steam or CO2 to the fuel removed or decreased the carbon formation. The particles retained their physical and chemical structure. Rydén et al. 2008b Sweden La0.5Sr0.5Fe0.5Co0.5O3-δ LaFeO3-δ , La0.8Sr0.2FeO3-δ La0.5Sr0.5FeO3-δ 60% NiO on MgAl2O4, 40% Fe2O3 on MgAl2O4, 40% Mn3O4 on Mg–ZrO2 pd : 90~250 m Reactor: Fixed bed Gases: 60mL/min CH4 (reduction) 60mL/min Air (oxidation) Temperature: 900 o C LaxSr1-xFeCo0.5O3-δwas found to be well suited for chemical-looping reforming applications Leion et al. 2008 Sweden FeTiO3 pd :0.125–0.180 mm Reactor: Fluidized bed 50%CO+50%H2 450 mLn/min 5%O2+N2 1000 mLn/min Temperature: 970~980 o C The ilmenite particles showed no decrease in reactivity in the experiments after 37 cycles of oxidation and reduction. Equilibrium calculations indicate the reduced ilmenite is in the form FeTiO3 and the oxidized carrier is in the form Fe2TiO5+TiO2. 163 Table A.1 continued Authors Particle Experimental conditions Remarks Zafar et al. 2007a Sweden Mn3O4(40%)+Mg-ZrO2 pd : 90~250 m Reactor: TGA 5~25% CH4+20%H2O+N2 3~15% O2+N2 Flow rates: 100~300 mL/min Temperature: 800~950 o C The particles showed very high reactivity during both reduction and oxidation. Zafar et al. 2007b Sweden NiO(60%)+NiAl2O4 pd : 125~180 m Density: 3200 kg/m3 Reactor: TGA 5~20% CH4+20%H2O+N2 3~15% O2+N2 Flow rates: 300 mL/min Temperature: 800~1000 o C Detailed equations for the reaction kinetic were installed. Cho et al. 2006 Sweden Fe2O3(60%)+Al2O3 NiO(60%)+NiAl2O4 Mg3O4(40%)+Mg-ZrO2 pd : 125~180 m Reactor: Fluidized bed U/Umf = 3.0-5.4 for the reducing period and 9.8- 17.2 for the oxidizing period 1000mL/min 5%O2+N2 for oxidization 450mL/min 100% CH4 for reduction Temperature: 950 o C For nickel oxide, the defluidization was dependent on the sintering temperature with no defluidization in experiments conducted with particles sintered at 1300 and 1400oC. The defluidization of the bed leads to agglomeration for the iron oxide particles, but not for the particles of nickel oxide Mattisson et al. 2006 Sweden NiO(40%)+NiAl2O4 NiO(40%)+NiAl2O4(48%)+ kaolinclay NiO(40%)+ MgAl2O4 Reactor: fluidized bed 50%CH4+50%H2O (reduction) 5%O2+N2 (oxidization) Temperature: 950 o C Some minor amounts of CO are released because of the thermodynamic limitations. A small amount of CH4 was released from the reactor at high degrees of oxidation of the NiAl2O4 and MgAl2O4-based carriers. Steam reforming of CH4 to CO and H2 became considerable, with Ni catalyzing this reaction. 164 Table A.1 continued Authors Particle Experimental conditions Remarks Zafar et al. 2006 Sweden NiO+MgAl2O4 Fe2O3+MgAl2O4 Mn2O3+MgAl2O4 CuO+MgAl2O4 NiO+SiO2 Fe2O3+ SiO2 Mn2O3+ SiO2 CuO+ SiO2 pd : 180~250 m Reactor : TGA Gases: 10% CH4, 10% H2O, 5% CO2, and 75% N2 (reduction) 5% O2 +N2 (oxidation) Flow rate: 40 mL/min Temperature: 800~1000 o C NiO and CuO supported on both SiO2 and MgAl2O4 showed very high reactivity. SiO2-supported Mn and Fe oxides may not be feasible oxygen carriers. Iron and manganese oxide supported on MgAl2O4 showed a rather high reactivity Johansson et al. 2006c Sweden NiO(40%)+NiAl2O4 Density: 3447 kg/m3 pd : 90~125 m Reactor: Fluidized bed Gases: Gases: 50% CH4+50%H2O (reduction) 4.5% O2+N2 (oxidization) Flow rate: 900 mL/min (reduction) 1000 mL/min (oxidation) Temperature: 950 o C Particles showed good performance for the long-time running. Johansson et al. 2006d Sweden Mn3O4(40%)+(Ca/Mg/Ce)ZrO3 pd : 125~180 m Reactor: fluidized-bed reactor of quartz (Column with length of 820 mm, diameter of 19~30 mm) Gases: 50% CH4+50% H2O (reduction), 5% O2+N2 (oxidization) Flow rate: 900 mL/min (reduction) 1000 mL/min (oxidation) Temperature: 950 o C The reactivity was inversely proportional to the sintering temperature and the strength of the particles. Johansson et al. 2006e Sweden 60% NiO/40% MgAl2O4 60% Fe2O3/40% MgAl2O4 pd : 125~180 m Reactor: fluidized-bed reactor of quartz (Column with length of 820 mm, diameter of 19~30 mm) Gases: Gases: 50% CH4+50% H2O (reduction), 5% O2+N2 (oxidization) Flow rate: 900mL/min (reduction) 1000mL/min (oxidation) Temperature: 950 o C The mixed-oxide system of 3% nickel oxides in 97% iron oxides produced significantly more CO2 than the sum of the metal oxides run separately, thus giving evidence of the synergy in using nickel oxide together with iron oxide. 165 Table A.1 continued Authors Particle Experimental conditions Remarks Cho et al. 2005 Sweden NiO(60%)+ NiAl2O4 (40%) Fe2O3 (60%)+Al2O3 (40%) pd : 125~180 m Reactor: fluidized-bed reactor of quartz (Column with length of 820 mm, diameter of 19~30 mm) Gases: Gases: 50% CH4+50% H2O (reduction), 5% O2+N2 (oxidization) Flow rate: 8.2~10.3 Umf (Oxidation) 2.5~27.0 Umf (reduction) Temperature: 750 ~ 950 o C For the oxygen carrier based on nickel oxide, carbon formation was strongly dependent on the availability of oxygen. For the particles with iron oxide, no or very little carbon was formed Johansson et al. 2004 Sweden Fe2O3 (40%~80%) +MgAl2O3 sinT : 950~1300 o C p :1400~3500 kg/m3 pd : 90~250 m Reactor: fluidized-bed reactor of quartz (Column with length of 820 mm, diameter of 19~30 mm) Gases: 50% CH4+50% H2O (reduction), 5% O2(oxidation) Velocity: 2~8 mfU (reduction), 5~11 mfU (oxidation) Temperature: 650~950 o C Pressure: atmospheric pressure Bed materials: 10 g, 15 g 60%Fe2O3 on 40% MgAl2O3 sintered at 1100 o C is the most suitable. The amount of material needed in the fuel reactor would be in the order of 150 MW Tobias et al. 2004 Sweden Fe2O3(60%)+Al2O3, Al2O3(32%) + kaolin(8%), MgAl2O4, ZrO2,TiO2 sinT : 950~1400 o C p :1100~4200 kg/m3 pd : 125~180 m Reactor: fluidized-bed reactor of quartz (Column with length of 820mm, diameter of 19~30mm) Gases: 50% CH4+50% H2O (reduction), 5% O2(oxidation) Flow rate: 600 mLn3/min 2~8 mfU (reduction), 600 or 1000 mLn3/min 5~12 mfU (oxidation) Temperature: 950 o C Pressure: atmospheric pressure Oxygen carriers that are based on iron oxide have been investigated. Fe2O3+ MgAl2O4 sintered at 950 oC, Fe2O3+ZrO2 sintered at 1100oC and Fe2O3+Al2O3 sintered at 1300 o C are good oxygen carriers. 166 Table A.1 continued Authors Particle Experimental conditions Remarks Cho et al. 2004 Sweden Fe2O3(60%)+Al2O3, Al2O3(32%) + kaolin(8%), NiO(60%)+NiAl2O4,CuO(60%)+CuAl2O4, Mn3O4(60%)+MnAl2O4 sinT : 1300 o C pd : 125~180 m Reactor: fluidized-bed reactor of quartz (Column with length of 820 mm, diameter of 19~30 mm) Gases: 50% CH4+50% H2O (reduction), 5% O2(oxidation) Flow rate: 600m Ln3/min 5~10 mfU (reduction), 1000 mLn3/min 10~20 mfU (oxidation) Temperature: 950 o C, 850 o C Oxygen carriers based on Fe, Ni, Cu showed high reactivity. CuO, Fe2O3+Al2O3 showed agglomeration Bed mass needed is 80~330 kg/MWth, and recirculation needed is 4~8 kg/s,MWth. Cho et al. 2002 Sweden Fe2O3 (40%~80%) +Al2O3 sinT : 1300 o C pd : 125~250 m Reactor: fluidized-bed reactor of quartz (Column with length of 820mm, diameter of 19~30mm) Gases: 100% CH4 (reduction), 5% O2(oxidation) Gas flows: 250, 500 NmL/min (reduction), 1000, 1000 NmL/min(oxidation) Temperature: 950 o C Pressure: atmospheric pressure The feasibility of using iron oxide as an oxygen carrier was investigated. Mattisson et al. 2003 Sweden NiO(33%)+Al2O3, Mn2O3(29.4%)+Al2O3, Co3O4(34.8%)+Al2O3, CuO(33.8%)+Al2O3 sinT : 550 o C Reactor: a fixed bed Gases: 10% CH4+10% H2O+5%CO2 +75%N2(reduction), 10%O2+N2 (oxidation) Flow rate: 80 mLn/min Temperature: 750~950 o C Pressure: atmospheric pressure Bed materials: 100 mg The particles of Ni and Cu showed high reactivity, with reduction rate up to 100%/min for CuO and 45%/min for Ni, oxidation rate up to 25%/min for them. Mn and Co aren’t suitable. It was estimated that 560~620 kg/MW oxygenic carrier would be needed The circulation flux is 1~8 kg/MW.S 167 Table A.1 continued Authors Particle Experimental conditions Remarks Mattisson et al. 2000 Sweden Fe2O3 (100%~58%) +Al2O3 sinT : 1300 o C pd : 120~500 m Reactor: a fixed-bed reactor of quartz Gases: 100% CH4 (reduction), Air(oxidation) Gas flows: 300 mL/min (reduction), 900 mL/min(oxidation) Temperature: 950 o C Pressure: atmospheric pressure Bed materials: 20 g, 60 g, 90 g redt dX )( 3~23%/min oxdt dX )( considerably faster, 20~90%/min Mattisson et al. 2001b Sweden Carajas (mainly Fe2O3) SiO2+Al2O3+Mn) pd : 180~250 m Reactor: Fixed bed Oxidising gas: Air 900 Nml/min, Reducing gas: CH4 300 Nml/min. Temperature: 950 o C Bed materials: 15~90 g (reduction), 1~2g (oxidation) Pressure: atmospheric pressure The reduction rate 1~8%/min and oxidation rate is considerably faster than the reduction, up to 90%/min Song et al. 2009 China CaSO4 pd : 0.15~0.2 mm Bulk density: 1500 kg/m3 Specific density: 2900 kg/m3 Reactor: Fixed bed Gases: 50 mL/min CH4 (reduction) 1000 mL/min Air (oxidation) Temperature:850~ 950 o C The mass-based reaction rates during the reduction and oxidation also demonstrated the variation of reactivity of CaSO4 oxygen carrier. Tian et al. 2008a China CaSO4 pd : 8.9 m Reactor: TGA Gases: 40% CO2+40% N2+20% CO (reduction) 21% O2+79%N2 (oxidation) Temperature: increase the temperature from 25 to 1355 o C at different heating rate Kinetic parameters of the decomposition reaction were achieved. Song et al. 2008a China CaSO4 pd : 0.15~0.2 mm Bulk density: 1500 kg/m3 Specific density: 2900 kg/m3 Reactor: fluidized bed Gases: 600 mL/min of 50%H2+25%CO+25%CO2 at 7.12-8.15 Umf (reduction) 1200 mL/min of 5%O2+N2 ) at 15.6 Umf (oxidation) Temperature: 950 o C The performance of in the cyclic experiments was tested. The oxygen carrier conversion after the reduction reaction decreased gradually in the cyclic test. 168 Table A.1 continued Authors Particle Experimental conditions Remarks Song et al. 2008b China CaSO4 pd : 0.15~0.2 mm Bulk density: 1500 kg/m3 Reactor: fluidized bed Gases: 600 mL/min 50%H2+25%CO+25%CO2 (reduction) Temperature: 890-950 o C The oxygen carrier conversion and mass-based reaction rates during the reduction at typical temperatures were compared. The apparent kinetic parameters were obtained. Song et al. 2008c China CaSO4 pd : 0.2~0.8 mm Specific density: 2900 kg/m3 Reactor: fixed bed Gases: 200 mL/min CH4 (reduction) 1000 mL/min Air (oxidation) Temperature: 1100 o C Temperature is the most important operating parameter on reduction reaction. SO2 content as high as 6% was produced Jin et al. 2004 China NiO(60%)+NiAl2O4 CoO-NiO(60%)+YSZ sinT : 1300 o C Size: pellet with 4.0 mm diameter and 1.5mm height Reactor: a fixed bed Gases: 33~40% CO, 17~20% H2, 20~33% H2O, 0~10% CO2, Ar (reduction), Air 100% (oxidation) Gas flow: 500~1200 ml/min Temperature: 600~1000 o C Pressure: 1~9 atm Bed materials: 50 g Identify the reaction kinetics of coal gas fueled chemical-looping combustion Jin et al. 2002 China NiO(42.6%)+YSZ, NiO(48.7%)+NiAl2O4 CoO-NiO(42.7%)/YSZ sinT : 1300 o C Size: pellet of 4.0 mm diameter and 1.5 mm height Reactor: a fixed bed, TGA Gases: CH4/H2O=1:2(reduction), 10% O2+N2 (oxidation) Flow rate: 300 mL/min (0.2 m/s) Temperature: 600~700 o C Pressure: 1~3 atm Bed materials: 50 g NiO/NiAl2O4 and CoO-NiO/YSZ are good candidates. Carbon deposition could be avoided completely by CoO- NiO/YSZ without addition of water and by NiO/NiAl2O4 at the ratio of H2O to CH4 of 2.0 Ishida et al. 2005 Japan Fe2O3 (25%)-Al2O3 (75%) pd :70 m Reactor: fixed bed A quantity of 1 mL of H2-Ar mixture (molar ratio 1:1) per pulse was supplied. Temperature: 900 °C The mechanical strength of particles was improved by increasing the content of corundum 169 Table A.1 continued Authors Particle Experimental conditions Remarks Ishida et al. 2002 Japan NiO (60%)+ NiAl2O4 sinT : 1300 oC pd : 97 m Reactor: TGA Gases: H2(reduction), Air(oxidation) Temperature: 900 o C Bed materials: 10mg Reactor: fluidized bed Gases: 67% H2+33% Ar (4.7 cm/s standard temp. o C) (reduction), Air (1.7cm/s standard temp. 0 o C) (oxidation) Temperature: 600, 900, 1200 o C Pressure: atmospheric pressure NiO (60%)+NiAl2O4 has good circulation properties, high reactivity and high mechanical strength. It could be used in the CLC circulation Jin et al. 2001 Japan NiO, NiO(60%)+YSZ, NiO(60%)+NiAl2O4 pd : 1800, 2100 m F =1.7e7 N/m2 (NiO+YSZ), 3.47e7 N/m2(NiO+NiAl2O4) Reactor: Fixed bed Oxidizing: 1000 o C Air Reducing 600 o C H2 (600 ml/min) H2 ((50%)+Ar(50%) 1000 ml/min) Pressure: 1~9 atm NiO+NiAl2O4 has excellent regenerability in cyclic use and fast rate. Jin et al. 1999 Japan NiO+Al2O3, NiO+TiO2, NiO+MgO; CoO+Al2O3, CoO+TiO2, CoO+MgO; Fe2O3+Al2O3, Fe2O3+TiO2, Fe2O3+MgO; NiO+Al2O3, NiO+NiAl2O4, NiO+YSZ Reactant: binder=60%:40% pd =1800, 2100 m sinT : 1300 o C Reactor: TGA Reduction : 600 o C, H2 100 mL/min (STP) 700 o C, CH4+H2O (1:2) Oxidation: 1000 o C, Air Pressure: 1~9 atm NiO+NiAl2O4 will provide an outstanding performance for CLC. H2O/CH4=2.0 could avoid the carbon deposition 170 Table A.1 continued Authors Particle Experimental conditions Remarks Jin et al. 1998 Japan NiO+YSZ, CoO+YSZ, Fe2O3+YSZ; CoO-NiO+YSZ ( Mole ratio of CoO to NiO is 1:1) Reactant: binder=60%:40% pd =1800, 2100 m sinT : 1300 o C Reactor: TGA Reduction : 600 o C, H2 100 mL/min (STP), CH4 Oxidation: 1000 o C, Air Pressure: 1~9 atm CoO-NiO+YSZ will provide an outstanding performance for CLC with good reactivity, significant regenerability and no carbon deposition. It is better than NiO+YSZ. Ishida et al. 1998 Japan NiO, NiO(20%~80%)+YSZ, NiO(60%)+Al2O3, NiO(60%)+TiO2, Fe2O3(60%)+YSZ, Fe2O3(60%)+ Al2O3, Fe2O3(60%)+ TiO2, sinT : 1300 o C F : 0.54~5.85e7 N/m2 Reactor: TGA Reduction: 50%H2+50%N2, 50%CO+50%N2, 600 o C, 100 mL/min (STP) Pressure: atmospheric pressure The condition that carbon deposition would be avoided was identified. NiO+YSZ is a good candidate. Ishida et al. 1996a Japan NiO (60%)+YSZ sinT : 1300 o C pd =1.0~3.2 mm p =3600~5800 kg/m3 F : 1.0~5.0e7 N/m2 Reactor: TGA Reduction: H2 600 o C, 800 o C, 1000 o C Oxidation: Air, 600 o C, 800 o C, 1000 o C Pressure: atmospheric pressure NiO/YSZ shows a good property. The reaction temperature, particle size, and gas composition strongly affect the reaction rate. Ishida et al. 1996b Japan NiO, NiO (60%)+YSZ sinT : 1300 o C pd =2.0mm Reactor: TGA Reduction: H2 600 o C, 350mL/s Oxidation: Air,1200 o C, 350mL/s Pressure: atmospheric pressure NiO/YSZ is a quite suitable, and no Nox formation Ishida et al. 1994 Japan NiO, NiO(60%,80%)+YSZ, Fe2O3(60%)+YSZ sinT : 1300 o C pd =1.3, 2.0, 2.8 mm Reactor: TGA Reduction: H2 550 ~ 950 o C Oxidation: Air 1000 o C Pressure: atmospheric pressure NiO/YSZ is suitable material, and the reaction temperature is the strongest factor in the reduction. 171 Table A.1 continued Authors Particle Experimental conditions Remarks Ryu et al. 2009 Korea NiO/bentonite, NiO/NiAl2O4, CoxOy/CoAl2O4, and OCN-650 Reactor: Fluidized bed Simulated syngas at 0.89 L/min +N2 at 1.11 L/min for reduction Air at 2.0 L/min for oxidation Temperature: 900 o C The inherent CO2 separation, NOx- free combustion, and long-term operation without reactivity decay of oxygen carrier particles are possible in a syngas fueled chemical-looping combustion system with the particles. However, CoxOy/CoAl2O4 represented slight decay of oxidation reactivity with the number of cycles increased and the oxidation rate slower than other particles. Son et al. 2006 Korea NiO/Fe2O3+TiO2/Al2O3/bentonite pd =106~150 m Reactor: thermobalance reactor and Circulating Fluidized Bed Reduction in thermobalance reactor : 50 mm/s at 650~950 o C 10%CH4, 10% H2O, 5% CO2, and 75% N2 Oxidation in thermobalance reactor : 50mm/s at 650~950 o C 10% O2, and 90% N2 Reduction in CFB : 25~100 mm/s at 650~950 oC CH4 Oxidation CFB: 1.5 m/s at 650~950 oC air The reactivity of NiO is higher than Fe2O3, and the particles supported on bentonite or Al2O3 produce higher reactivity than those on TiO2. The reactivity of the metal oxide particles increases with increasing temperature and the amount of NiO. Song et al. 2006 Korea NiO(26%~91%)+NiAl2O4 Reactor: TGA Reduction: H2 600 o C, 100 mL/min Oxidation: Air,1000 o C, 100 mL/min NiO/NiAl2O4 particles containing 57~83 wt% are good oxygen carrier. Lee et al. 2005 Korea Metal oxide (Ni, Fe, Co-based) mixed with ZrO2, YSZ, AlPO4, Al2O3 pd =2 mm Reactor : TGA Reduction: H2 600 o C, 100 mL/min Oxidation: Air,1000 o C, 100 mL/min NiO-based oxygen carrier has higher redox reaction rates and regeneration properties. Song et al. 2003 Korea NiO(20, 30, 40%)+hexaaluminate (NiO+LaAl11O18) (NiO+NiAl2O4) sinT : 1000 o C Reactor: TGA Gases: H2 (5.6%)+Ar (reduction), Air(oxidation) Showed good reduction and oxidation properties. 172 Table A.1 continued Authors Particle Experimental conditions Remarks Ryu et al. 2003a Korea NiO (59%)+ bentonite (Al3SiO2) sinT : 900 o C p :4038 kg/m3 b : 1319 kg/m3 : 0.69 pd : 91 m Reactor: TGA Gases: CH4 (5.04%) (reduction), Air(oxidation) Gas flows: 100 ml/min (reduction), 100 ml/min(oxidation) Temperature: 650~1000 o C Pressure: atmospheric pressure The carbon deposition, reduction kinetics and regenerative ability were examined. 900 oC is the appropriate for reduction and avoiding carbon deposition Ryu et al. 2003b Korea NiO(59%)+bentonite sinT : 900 oC p  :4080 kg/m3 b : 1407 kg/m3 : 0.59~0.714 pd : 400 m Reactor: A fixed bed Gases: 13% CH4 (reduction), 8.6% O2(oxidation) Gas flows: 2.3 L/min (reduction), 2.2 L/min(oxidation) Temperature: 500~1000 o C Pressure: atmospheric pressure Bed materials: 40 g Carbon deposition, inter and lump of NiO+bentonite will take place at 1000oC. 900oC is the most appropriate temperature for NiO+bentonite Adánez et al. 2009a Spain NiO(18%)+ α-Al2O3 Apparent density: 2470 kg/m3 pd : 0.1~0.3 mm Reactor: Circulating Fluidized bed Gases: 30~50% CH4 at 0.1 m/s (170 LN/h) (reduction) Primary air gas: 720 LN/h and the secondary air gas 150 LN/h (oxidation) Temperature: 800~880 o C The effect of operating conditions on the performance of the oxygen carrier in the CLC prototype was analyzed. During operation of the CLC prototype, no signs of agglomeration or carbon formation were detected and the main properties of particles did not vary. Adánez et al. 2009b Spain NiO(19%)+NiAl2O4 pd :0.1~0.3 mm density: 2500 kg/m3 Reactor: TGA (2-15%) of fuel gas (CH4, C2H6, or C3H8), 20 vol% H2O, and N2 balance (reduction) Air (oxidation) Tempeature: 700~950 o C Reactor: Circulated fluidized bed Gases: CH4 +N2 (reduction) Air (oxidation) Temperature: 870 o C No special measures should be taken in a CLC process with respect to the presence of Light Hydrocarbons in the fuel gas, e.g. refinery gas or crude natural gas. 173 Table A.1 continued Authors Particle Experimental conditions Remarks Gayán et al. 2009 Spain NiO(25~30%)+NiAl2O4 Apparent density: 1700~2800 kg/m3 Reactor: TGA Gases: 15% CH4 +20%H2O+N2 (reduction) Air (oxidation) Temperature: 950 o C Reactor: Fluidized bed Gases: 25% CH4 +N2 (reduction) 10~15%O2+ N2 (oxidation) Flow rate: 0.1 m/s Temperature: 950 o C The oxygen carrier prepared by at ambient (AI), and hot conditions (HI) and by deposition–precipitation (DP) methods on α-Al2O3 as support had appropriated characteristics to be used in the chemical-looping combustion process. García- Labiano et al. 2009 Spain NiO(19%)+NiAl2O4 pd :0.1~0.3mm density: 2500 kg/m3 Reactor: Circulated fluidized bed Gases: CH4 +N2 (reduction) Air (oxidation) Temperature: 870 o C Nickel sulfide, Ni3S2, was formed at all operating conditions in the fuel reactor, which produced an oxygen carrier deactivation and lower combustion efficiencies. However, the oxygen carrier recovered their initial reactivity after certain time without sulfur addition. The sulfides were transported to the air reactor where SO2 was produced as final gas product. Linderholm et al. 2008 Spain NiO(60%)+NiAl2O4 pd :125~180 m Apparent density: 4400 kg/m3 Reactor: Circulated fluidized bed Gases: natural gas at 15L/min (reduction) Air (oxidation) Temperature: 850~1000 o C Particles showed good performance in the experiments. Johansson et al. 2008 Spain NiO(40%)+NiAl2O4 NiO(60%)+MgAl2O4 pd :125~180 m Reactor: Fluidized bed 50%CH4+50%H2O+ (reduction) 5%O2+N2(oxidation) Temperature: 950 o C NiO(60%)+MgAl2O4 offers several advantages at elevated temperatures, i.e. higher methane conversion, higher selectivity to reforming and lesser tendency for carbon formation. 174 Table A.1 continued Authors Particle Experimental conditions Remarks de Diego et al. 2008 Spain NiO (11%~28%)+ Al2O3 Density: 1700~2600 kg/m3 Reactor: TGA 15%CH4+20%H2O+N2 (reduction) Air (oxidation) Temperature: up to 950 o C 10%H2+Ar at 75mL/min temperature: up to1000 o C at a rate of 20 o C /min Reactor: Fluidized bed 25%CH4+7.5~17.5%H2O+N2 (reduction) 10~15%O2+N2(oxidation) Temperature:800~950 o C Ni-based oxygen carriers prepared by dry impregnation are suitable for autothermal reforming of methane during long periods of time without carbon deposition. Gayán et al. 2008 Spain NiO+ Al2O3, Al2O3+ NiAl2O4 NiO+ Al2O3+ NiAl2O4, MgAl2O4+NiAl2O4 MgAl2O4+NiO, CaAl4O7+NiAl2O4 CaAl4O7+CaAl2O4+NiO Bulk density: 1400~3100 kg/m3 Reactor: TGA Gases: 15% CH4+20% H2O+N2(reduction) Air (oxidization) Temperature: 950 o C Reactor: fluidized bed Gases:25%CH4+N2 (reduction) 10~15% O2+N2 (oxidation) Temperature: 950 o C Ni-based oxygen carriers prepared by impregnation on α-Al2O3, CaAl2O4 and MgAl2O4 showed very high reactivity and high methane combustion selectivity to CO2 and H2O. In addition, these oxygen carriers had low attrition rates and did not show agglomeration problems during operation in fluidized beds. Abad et al. 2007a Spain Fe2O3(60%)+Al2O3 pd 90~212 m Density: 2125 kg/m3 Reactor: CFB Gas: Natural gas Flow rates: 2–18 Umf for fuel reactor, 75Umf for air reactor Temperature: 800~950 o C The performance of particles was tested in a CFB reactor. Abad et al. 2007b Spain CuO(10%)+Al2O3 NiO(40%)+Al2O3 Fe2O3(60%)+Al2O3 pd :0.15~0.2 mm Density: 4180~5380 kg/m3 Reactor: TGA Gas: 5~20%H2+5~30%CO+5~55%CO2+5~30%H2O Flow rates: 6 cm3/s Temperature: 600~950 o C Pressure: 0.1 MPa~3.0 MPa The oxygen carriers exhibited very high reactivity during the reduction reaction. 175 Table A.1 continued Authors Particle Experimental conditions Remarks de Diego et al. 2007 Spain CuO(14%)+Al2O3 pd 0.1~0.5 mm Reactor: CFB Gas: CH4 for reduction, Air for oxidation Operating temperature in the range of 700~800 o C and of the linear velocities in the range of 0.07~0.14 m/s. The experiments revealed a good performance of these CuO-based materials as oxygen carriers in a CLC process. Mattisson et al. 2007 Spain CuO (10%)+Al2O3 FeO(60%)+Al2O3 NiO(40%)+NiAl2O4 NiO(60%)+MgAl2O4 Mn3O4(40%)+ZrO2 Reactor: CI Thermobalance Gases: For reduction: fuel (CO, H2) 5~70%; H2O 0~48%; CO2 0~40%. For the oxidation reaction, oxygen 5 ~ 21% Temperature:500 to 950 oC The kinetic parameters for the particle reaction had been discussed Adánez et al. 2006b Spain NiO(6%~30%)+ Al2O3 NiO(1%~12%)+CuO(1%~15%)+ Al2O3 Density: 1500~2000 kg/m3 Reactor: TGA, Fixed bed, Fluidized bed Gases: 15%+20%H2O+N2 (reduction for TGA) Air (oxidation for TGA) 60~120L/h CH4, CO, or H2 at 20%~100% (reduction for fixed bed) 60~120L/h 4%O2+N2 (oxidation for fixed bed) 0.1m/s 25%CH4+N2 (reduction for fluidized bed) 0.1m/s 10~15% O2+N2 (oxidation for fluidized bed) Temperature: 950 o C NiO-Ral2O3 oxygen carriers prepared by dry impregnation showed very high reactivity. The presence of CuO in the Ni-Cu oxygen carriers allows the full conversion of CH4 to CO2 and H2O in the batch fluidized bed reactor during the initial part of the reduction time, and this time depended on the CuO content of the oxygen carrier. Abad et al. 2006b Spain CuO (10%)+Al2O3 FeO(45%)+Al2O3 NiO(40%)+Al2O3 pd :0.15~0.2 mm Reactor: thermobalance Gases: Fuel: 5~70%; H2O: 0~48%; CO2: 0~40% (reduction) 5 to 21% O2 (oxidization) Temperature: 500~950 o C Pressure: atmospheric pressure Analyze the possible operation conditions for oxygen carriers with different metal oxide content. 176 Table A.1 continued Authors Particle Experimental conditions Remarks García- Labiano et al. 2006 Spain CuO (10%)+Al2O3 FeO(45%)+Al2O3 NiO(40%)+Al2O3 pd : 0.15~0.2 mm Reactor: TGA Gases: H2/N2, CO/CO2/N2 (reduction) O2/N2 (oxidization) Gas flows: 83 mL/s Temperature: 800 o C Pressure: up to 30 atm An increase in total pressure has a negative effect on the reaction rates of all the oxygen carriers. Corbella et al. 2006a Spain NiO+TiO2 pd : 0.2~0.4 mm Reactor: fixed bed (ID 1.6 cm, Height: 34 cm) 80 cm3/min of either dilute CH4 (20% in N2) or pure CH4 for the reduction stage and pure air for the regeneration stage Temperature: 900 °C Reactions are fast, carbon is mostly deposited as uniform coatings on Ni catalyst particle. Corbella et al. 2006b Spain CuO(10%~45%)+Silica pd : 0.2~0.4 mm Reactor: fixed bed (ID 1.6 cm, Height: 34 cm) Gases: 13 cm3/min of CH4 Temperature: 800 °C Pressure: atmospheric pressure CuO might decompose into Cu2O at the reduction stage, but the decomposition rate is so low that it has no effect on the oxygen amount initially available for chemical- looping combustion. Copper does not promote the thermal decomposition of Methane. Neither performance decay nor mechanical degradation of the oxygen carrier has been observed. Corbella et al. 2005a Spain CuO(5%~34%)+TiO2 pd : 0.2~0.4 mm Reactor: fixed bed (ID 1.6 cm, Height: 34 cm) Gases: 10 cm3/min of CH4 Temperature: 800 and 900 °C Pressure: atmospheric pressure The copper –based oxygen carrier supported by titania exhibited a good performance. Corbella et al. 2005b Spain NiO (3%~16%)+NiTiO3 pd : 0.2~0.4 mm Reactor: fixed bed (ID 1.6 cm, Height: 34 cm) Gases: 9 cm3/min CH4 (reduction) and 9 cm3/min Air (oxidation) Temperature: 900 oC Pressure: atmospheric pressure The reactivity of these nickel-based oxygen carriers almost independent of the nickel loading. However, in the reduction stage, carbon deposition may impose some constraints to the efficiency of the overall chemical- looping combustion process in CO2 capture. 177 Table A.1 continued Authors Particle Experimental conditions Remarks de Diego et al. 2005 Spain CuO (10%~26%)+Al2O3 pd : 0.1~0.32 mm Reactor: Fluidized bed reactor (ID 54 mm, Height 0.5 m) Gases: 8% O2+N2 (oxidation) 25%CH4+N2 (reduction) Gas flows: 0.1 m/s Temperature: 800 o C The reactivity is high, but the agglomeration is a problem for the particles with CuO content of more than 20%. Adánez et al. 2004a Spain 40%~80% Cu, Fe, Mn, Ni oxides with Al2O3, sepiolite, SiO2, TiO2, ZrO as inert. sinT : 950~1300 o C p :1400~5000 kg/m3 : 0.1~0.77 Reactor: TGA Gases: 70% CH4+30% H2O (reduction), Air(oxidation) Gas flows: 25 nL/h Temperature: 800~950 o C Bed materials: 20~100 mg SiO2 or TiO2 best for Cu-based oxygen. Al2O3 and ZrO2 best for Fe- based carriers. ZrO2 best for Mn- based carriers. TiO2 best for Ni-based carriers Villa et al. 2003 Italy Ni-Al-O (Ni/Al 0.5~2.25) =NiO(6%~100%) Ni-Mg-Al-O (Ni/Mg=1, (Ni+Mg)/Al=1) sinT : 1000 o C H2(5%)+Ar, O2(2%)+He, CH4 (1.6%)+He Gas flows: 30 Ncc/min Bed materials: 120 mg Activity run (under constant temperature) Gases: 50%CH4 +He(reduction), 50%CH4 +H2O(reduction)20% O2+He(oxidation) Gas flows: 20 cm3/min (reduction), Temperature: 800 o C Pressure: atmospheric pressure Bed materials: 2 g Addition of Mg was found to stabilize Ni2+ in the cubic oxide and spinel phase, increasing the reduction temperature, markedly improves regenerability. CH4/H2O (1:1) could avoid coke formation in Ni-Mg-Al-O Ni-based systems are poorly selective to H2O and CO2, being CO and H2 if feeding CH4 Hoteit et al. 2009 France NiO(60%)+NiAl2O3 , Cu0.95Fe1.05AlO4 pd : 0.1~0.3 mm Bulk density: 2700 kg/m3 Reactor: Fluidized bed Gases: 100%CH4, 50%CH4+50%H2O at 0.05–0.15 m/s (reduction) Air at 0.15–0.3 m/s (oxidation) Temperatue:800~900 o C The particles were tested experimentally in a fluidized-bed reactor as a function of number of oxidation/reduction cycles, temperature and effect of steam mixed with methane as a fuel. 178 Table A.1 continued Authors Particle Experimental conditions Remarks Chandel et al.2009 France NiO(60%)+NiAl2O3 , Cu0.95Fe1.05AlO4 CuO(23.5%)+ Cu0.95Fe1.05AlO4 pd : 0.1~0.3 mm Bulk density: 2700 kg/m3 Reactor: Fluidized bed Gases: 100%CH4, 50%CH4+50%H2O at 0.05– 0.15 m/s (reduction) Air at 0.15–0.3 m/s (oxidation) Temperatue:800~900 o C Carbon formation could occur during the reduction phase but it decreased with an increase in temperature and inventory and could be completely avoided by mixing steam with the fuel. The reactivity of NiO/NiAl2O4 was higher than the Cu- and Fe-based oxygen carriers. Increasing the CuO fraction in the oxygen carrier led to defluidization of the bed during the reduction and oxidation phases. Hossain et al.2009 Canada Ni/La-Al2O3 Reactor: Fluidized bed Gas: CH4 10 mL/40 s (reduction) Air (oxidation)Temperature: 600~750 o C Multiple reduction/oxidation cycles demonstrate that the Ni/La-Al2O3 particles display excellent reactivity and stability. Sedor et al. 2008a Canada NiO(5~20%)+ Al2O3 Reactor: Fluidized bed Gas: CH4 10 mL/40 s (reduction) Air (oxidation) Temperature: 680 o C (reduction) 525 o C (oxidation) A kinetic model is developed to describe the reaction of particles. Sedor et al. 2008b Canada NiO(5~20%)+ Al2O3 Reactor: Fluidized bed Gas: CH4 10 mL/40 s (reduction) Air (oxidation) Temperature: 680 o C (reduction) 525 o C (oxidation) The oxygen carrier is stable under industrial relevant fluidized bed reaction conditions, converting 76% of methane to carbon dioxide and water vapor, the combustion products. Hossain et al.2007 a Canada Co-Ni+ Al2O3 Ni+ Al2O3 Reactor: Fluidized bed Gas: CH4 10 mL/40 s (reduction) Air (oxidation) Temperature: 650 o C (reduction) 525 o C (oxidation) Develop a kinetic model to describe the reduction of the particles 179 Table A.1 continued Authors Particle Experimental conditions Remarks Hossain et al.2007b Canada Co-Ni+Al2O3 pd : 60~100 m Density: 1500 kg/m3 Reactor: Fluidized bed Gas: CH4 10 mL/40 s (reduction) Air (oxidation) Temperature: 680 o C (reduction) 525 o C (oxidation) Co.Ni/Al2O3 particles displayed excellent reactivity and stability Co might promote the performance of Co.Ni/Al2O3 particles by reducing the metal support interactions Kolbitsch et al. 2009d Austria NiO(40%)+NiAl2O4 NiO(40%)+MgAl2O4 pd : 90~212 m Reactor: Circulating fluidized bed Gases: Natural gas (reduction) air (oxidation) Temperature: 800~950 o C No carbon formation was observed under any conditions and the results indicated the high reactivity of both of the particles. Readman et al. 2006 Norway NiO (60%)+NiAl2O4 pd : 90-210 m Reactor: TGA Gases: 10% H2 +Ar at 10mL/min (reduction) 16.7% CH4+Helium at 12mL/min (reduction) Air 5mL/min (oxidation) Temperature: heating rate of 15 o C /min from room temperature to 800 o C Rapid reduction is followed by a much slower reduction, where oxygen transport through the particle is expected to be ratelimiting. The fast reduction reaction is determined to be first order, with respect to H2, whereas an order slightly smaller than unity is observed when using CH4 as a reducing agent. Reoxidation is observed to be first order, with respect to O2. Brandvoll et al. 2003 Norway NiO+NiAl2O4 pd : 300~500 m , 600~1000 m , 1200~1700 m , 2000~3150 m , Reactor: fluidized/fixed-bed reactor Gases: H2 Temperature: 850~ 560 o C Pressure: atmospheric pressure The mass transfer mechanisms, i.e. particle-external and particle-internal diffusion, control the overall rat of reduction reaction. Erri et al. 2009 USA NiO+NiAl2O4 Reactor: TGA Gases: 20%H2+Ar Temperature: 1~45 o C /min The experiments demonstrate the presence of internal and external mass-transfer limitations during the reaction. 180 Table A.1 continued Authors Particle Experimental conditions Remarks Tian et al. 2008b USA CuO/bentonite pd : 74~840 m Reactor: TGA Gases: 12% CO2, 36% CO, 25% He, and 27% H2 (reduction) Air (oxidation) Flow rates: 45 cm3/min Temperature: 700~900 o C The CuO/bentonite prepared by solid- state mixing was found to be a good oxygen carrier for CLC of simulated synthesis gas without the impurities such as H2S for IGCC systems. Siriwardane et al. 2007 USA NiO(60%)+bentonite pd : 74~840 m Reactor: TGA and fixed bed Gases: 12% CO2, 36% CO, 25% He, and 27% H2 (reduction) Air (oxidation) Flow rates: 45 cm3/min Temperature: 700~900 o C The effect of temperature and particle size and pressure on the reaction rate was tested. Erri et al. 2007a USA NiO/NiAl2O4 (NiO)1−y (MgO)y/Ni(1−x)MgxAl2O4 Reactor: TGA Gases: 35%H2, 18% CO, 47% CO2 (reduction) 10% O2 +Ar (oxidation) Temperature: 750-1100 oC (NiO)0.79(MgO)0.21/Ni0.62Mg0.38Al2O4 exhibited excellent reactivity and stability. Erri et al. 2007b USA NiO/NiAl2O4 (NiO)1−y (MgO)y/Ni(1−x)MgxAl2O4 Fe2O3 Reactor: TGA Gases: 3%CH4+3%H2O+Ar (reduction) dilute O2 (oxidation) Temperature: 800~1200 o C (NiO)0.79(MgO)0.21/Ni0.62Mg0.38Al2O4 showed superior performance. Chuang et al.2008 UK CuO(20%~100%)+Al2O3 pd : 500~710 m Reactor: Fluidized bed Gases: 69.7 cm3/s 2.5% CO+N2 (reduction) 50.9 cm3/s air (oxidation) Temperature: 600~1000 oC Supporting CuO on Al2O3 enhances the ability of the resulting particles to withstand mechanical and thermal stresses in a fluidized bed. Only co- precipitation produces particles that have a high loading of copper and do not agglomerate at 800–900 ◦C. The performance of co-precipitated particles of CuO/Al2O3 at oxidizing CO to CO2 was significantly affected by the pH of the solution in which precipitation occurred. 181 Table A.2 Experiments about the hydrodynamics research on cold-flow model Reactor size Particle Superficial gas velocity (m/s) Author Air reactor Diameter ×Height Fuel reactor Diameter ×Height Diameter ( m ) Density (kg/m3) Gas Reactor type Air reactor Fuel reactor Solid inventory (kg) Solids flux or solids flow rate Johansson et al 2002 0.19 m × 1.9 m 0.19 m × 0.5 m 150 2600 Air IFB with combining riser and bubbling bed 0.7~1.2 0.12 5~10 0~8 kg/m2.s Johansson et al 2003 0.19 m × 1.9 m 0.19 m × 0.5 m 150 2600 Air IFB with combining riser and bubbling bed 0.75~1.15 0.09~0.31 N/A N/A Kronberger et al 2004 0.019 m × 0.019m 0.019 m ×0.027 m 70 1500 He+N2 Two- compartment IFB (1.3~2.1)Ut 18Umf 0.043~0.058 0.1~1.0 g/s Kronberger et al 2005 0.077 m ×0.29 m 0.14 m × 0.19 m 67.5 2550 He+N2 IFB with combining riser and bubbling bed (3~7.5)U t (4~15)Umf 1.1~2.2 0~29 g/s Ryu et. al 2008a,b 0.8 m × 0.15 m 0.8 m × 0.15 m 106~212 2575 Air IFB with two bubbling beds 0.034~0.072 0.034~0.072 0.5 m bed height 0~60g/s Pröll et al. 2009a 0.05 m × (?) 0.054 m ×(?) 54 8730 Air IFB with two risers 4.3 1.4 4~5 kg 20~100 kg/m2.s 182 APPENDIX B MEASUREMENT SYSTEM B.1 Pressure measurement Pressure transducers used in this study are shown in the Table B.1. The locations of the pressure transducers have been presented in Figure 3.6. Table B.1 Pressure transducers used in this study Pressure transducer Model number Pressure type / Pressure range A1~A15 PX164-010D5V Differential pressure /0~10 inch water B1~B9 PX142-001D5V Differential pressure / 0~1psi C1~C4 PX142-005D5V Differential pressure / 0~5psi D1~D6 D6 D7* D8* PX142-015D5V Differential pressure / 0~15psi E1* E2* PX142-030A5V Absolute pressure / 0~30psi * D7 and D8: pressure difference across the orifices. E1 and E2: absolute pressure before the orifice plate The pressure transducers were calibrated by the following system shown in Figure B.1. A series of pressures were created by adjusting the amount of the air. The pressures were measured by a U-tube manometer, and the pressure transducer produced the corresponding electrical signal for a specific pressure. Figures B.2~B.6 show the calibration results of the typical pressure transducers used in this study. The pressure is proportional to the magnitude of the electrical signal produced by the pressure transducers. And the relationship between pressure and electrical signal can be shown by following Eq. B.1 )( ptptpttp aVkP  (B.1) 183 Figure B.1 Pressure transducer calibration system 0 0.5 1 1.5 2 2.5 3 3.5 4 0 0.5 1 1.5 2 2.5 Vpt-apt (V) P (k P a) P=kpt(Vpt-apt) a pt =0.9936 kpt=0.5039 Figure B.2 Calibration results for pressure transducers (PX164-010D5V) 184 0 0.5 1 1.5 2 2.5 3 3.5 4 0 1 2 3 4 5 6 Vpt-apt (V) P (k P a) P=k pt (V pt -a pt ) apt=1.0159 kpt=1.3985 Figure B.3 Calibration results for pressure transducers (PX142-001D5V) 0 0.3 0.6 0.9 1.2 1.5 0 2 4 6 8 10 12 Vpt-apt (V) P (k P a) P=kpt(Vpt-apt) apt=1.0010 k pt =6.8787 Figure B.4 Calibration results for pressure transducers (PX142-005D5V) 185 0 0.1 0.2 0.3 0.4 0.5 0 2 4 6 8 10 12 V pt -a pt (V) P (k P a ) P=kpt(Vpt-apt) a pt =0.9994 kpt=21.225 Figure B.5 Calibration results for pressure transducers (PX142-015D5V) 2.4 2.45 2.5 2.55 2.6 2.65 2.7 100 102 104 106 108 110 112 114 V pt -a pt (V) P (k P a) P=k pt (V pt -a pt ) apt=0.9900 k pt =42.005 Figure B.6 Calibration results for pressure transducers (PX142-030A5V) 186 B.2 Gas velocity measurement The superficial gas velocities for air rector and fuel reactor were calculated based on the inlet gas volumetric flow rate. In this study, as shown in Figure 3.4, the gas volumetric flow rates were measured by the orifice meters which were mounted before the inlet of the reactors. The specific properties of the orifice plate located before the inlet of the reactors are shown in Figure B.7. If the pressure drop across the orifice plate and pressure at the upstream of the orifice are measured, the discharge coefficient of the orifice meter can be calculated based on the throat diameter of the orifice plate and the diameter of the pipe which is connected to the orifice plate using the usual orifice equation. Figure B.7 Location of the orifice meter Stearns et al. (1951) presented the following usual orifice equations for calculations of discharge coefficient and the equations have been verified by the experiments of Wang (2006) and Chen (2007). 1 0 0 D D (B.2) 1 0 530 D B  (B.3) )420090005000830( 0 3 0 2 0000 BDA   (B.4) 1 0 007.05993.0 D b  (B.5) 1 0 076.0 364.0 D n  (B.6) 187 4 000 nbK e  (B.7) 15 10 Re 0 6 D e  (B.8) f orf DU   0 0Re  (B.9) e e A A KK Re/1 Re/1 0 00 0   (B.10) where D0 is the diameter of throat of orifice plate, D1 is the internal diameter of the pipe connected with the orifice plate. β0 is the diameter ratio of throat-to-pipe. A0, B0, b0, n0 are the intermediate parameters. Re0 is the Reynolds number based on the gas velocity through the throat, Uor, and diameter of throat, D0. ρf is the gas density and μf is the viscosity of the gas. K0 is the orifice discharge coefficient, Ke is the orifice discharge coefficient when the Reynolds number is equal to Ree. It should be emphasized that the unit of D0 and D1 in the Eqs.B.2~B.10 is inch. Thus, the gas volumetric flow rate can be calculated based on the following Eq. B.11. f PDKQ    2)4/( 200 (B.11) In this study, two orifice plates provided by BG Controls with D0 of 41.4 mm and 17.8 mm were used for the measurement of inlet volumetric gas flow rates of air reactor and fuel reactor, respectively. The corresponding internal diameter, D1, of the pipes connected with the orifice plates are 102.3 mm and 77.9 mm, respectively. In the Eqs.B.10 and B.11, the unit of D0 is meter. B.3 Helium concentration measurement In this study, the helium concentration in gas mixture was measured by Micro Gas Chromatograph (GC) provided by Varian Co. When a gas stream containing some helium flows through the Micro GC, a specific curve shown in Figure B.8 will be produced by the instrument. At the same time the surface area under the curve representing the helium molar fraction in the gas mixture will be calculated by the professional software provided by the Varian Co. 188 Figure B.8 Typical curve produced by Micro GC for helium concentration In order to calibrate the Micro GC for helium concentration measurement, the following system shown in Figure B.9 was used. The gas flows of helium and air with pre-determined flow rates were mixed in a buffer and one sampling stream were carried into the Micro GC. The flow rates of helium and air were adjusted by the mass flow controllers. Thus, the molar fraction of the helium in the gas mixture had been known before they were carried into the Micro GC, and the corresponding surface area under the curve was calculated. The relationship between the helium molar fraction and the surface area under the curve can be shown in Figure B.10. Figure B.9 Experimental system for Micro GC calibration on helium concentration 189 0 20 40 60 80 100 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 x 10 6 Helium concentration (mol%) A re a (  V .S ec ) He concentration(mol%) =Area/ 17698 Figure B.10 Calibration result of Micro GC for helium concentration measurement It was found that the surface area under the curve is proportional to the helium concentration (mole fraction), and the linear relationship can be shown by the Eq. B.14. Helium concentration (mol%) = measured surface area / 17698 (B.14) 190 APPENDIX C DRAWINGS OF THE MAIN COMPONENTS OF THE COLD- FLOW MODEL Figure C.1 Distributor of air reactor 191 Figure C.2 Lower section of air reactor 192 Figure C.3 Middle section of air reactor 193 Figure C.4 Upper section of air reactor 194 Figure C.5 Lower section of fuel reactor 195 Figure C.6 Middle section of fuel reactor 196 Figure C.7 Upper section of fuel reactor 197 Figure C.8 Loop-seal 1 198 Figure C.9 Loop-sea1 2 199 Figure C.10 Primary cyclone 200 Figure C.11 Secondary cyclone 201 Figure C.12 Internal cyclone in fuel reactor 202 APPENDIX D ADDITIONAL TABLES AND FIGURES Table D.1 Positions of ports for pressure measurement Pressure transducer Pressure type Vertical distance relative to the distributor of air reactor (mm) B1 Gauge pressure 1219 B2 Gauge pressure 1422 B3 Gauge pressure 1955 B4 Gauge pressure 2261 B5 Gauge pressure 2811 B6 Gauge pressure 3945 B7 Gauge pressure 4401 B8 Gauge pressure 5226 B9 Gauge pressure 5588 C1 Gauge pressure 1422 C2 Gauge pressure 273 C3 Gauge pressure 915 C4 Gauge pressure 1016 D1 Gauge pressure 0 D2 Gauge pressure 102 D3 Gauge pressure 979 D4 Gauge pressure 1083 D5 Gauge pressure 284 D6 Gauge pressure 284 203 Table D.2 Combination of operating conditions Ua0 (m/s) UA1 (m/s) UA2 (m/s) 2.5 1Umf , 2Umf , 4Umf , 6Umf 0Umf ,1Umf ,1.5Umf 3 1Umf , 2Umf , 4Umf , 6Umf 0Umf ,1Umf ,1.5Umf 4 1Umf , 2Umf , 4Umf , 6Umf 0Umf ,1Umf ,1.5Umf 5 1Umf , 2Umf , 4Umf , 6Umf 0Umf ,1Umf ,1.5Umf Table D.3 Hot unit modeled by a cold unit fluidized with 80 vol% helium and 20 vol% air at temperature of 40 oC and atmosphere pressure (based on identical ratio of particle size to reactor size) Air Reactor Parameter Cold Unit Hot Unit Gas type 80 vol% helium + 20 vol% Air Air Temperature (oC) 40 950 Viscosity of gas (Pa.s) 2.03×10-5 4.94×10-5 Particle density s (kg/m3) 1560 2800 Gas density f (kg/m3) fc (0.35) fc79.1 (0.63) Pressure (bar) cP (1) cP17.2 (2.17) Particle diameter pd ( m ) pcd (78) pcd23.1 (96) Bed diameter (mm) cD (102) cD23.1 (125) Superficial gas velocity cU 0 cU 011.1 Solids circulation flux scG scG99.1 Minimum fluidization velocity mfU (m/s) 0.0034 0.0038 Fuel Reactor Parameter Cold Unit Hot Unit Gas type 80 vol% helium + 20 vol% Air 25% vol CH4+ 25 vol% CO2 +50 vol% H2O Temperature (oC) 40 900 Viscosity of gas (Pa.s) 2.03×10-5 4.03×10-5 Particle density s (kg/m3) 1560 2800 Pressure (bar) 1 2.17 Gas density f (kg/m3) 0.35 0.54 Particle diameter pd ( m ) 78 96 Bed diameter (mm) cD (286) cD23.1 (352) Superficial gas velocity cU 0 cU 011.1 Solids circulation flux scG scG99.1 Minimum fluidization velocity mfU (m/s) 0.0034 0.0047 204 Table D.4 Dimensionless group comparison for operation conditions in Table D.3 Air Reactor Dimensionless parameter Cold Unit Hot Unit Ratio between cold unit and hot unit gD U 20 2000.1 cU 2 000.1 cU 1.00 f s   4457 4444 1.00 mfU U 0 cU 0294 cU 0292 1.01 0U G s s  c sc U G 0 4-106.41 c sc U G 0 4-1040.6  1 pd D 1308 1302 1.00 Fuel Reactor Dimensionless parameter Cold Unit Hot Unit Ratio between cold unit and hot unit gD U 20 2 036.0 cU 2 036.0 cU 1.00 f s   4457 5185 0.86 mfU U 0 cU 0294 cU 0236 1.25 0U G s s  c sc U G 0 4-106.41 c sc U G 0 4-106.41 1 pd D 3667 3667 1.00 205 Table D.5 Hot unit modeled by a cold unit fluidized with 60 vol% helium and 40 vol% air at temperature of 40 oC and atmosphere pressure (based on identical ratio of particle size to reactor size) Air Reactor Parameter Cold Unit Hot Unit Gas type 60 vol% helium + 40 vol% Air Air Temperature (oC) 40 950 Viscosity of gas (Pa.s) 2.02×10-5 4.94×10-5 Particle density s (kg/m3) 1560 2800 Gas density f (kg/m3) fc (0.54) fc79.1 (0.97) Pressure (bar) cP (1) cP34.3 (3.34) Particle diameter pd ( m ) pcd (78) pcd23.1 (96) Bed diameter (mm) cD (102) cD23.1 (125) Superficial gas velocity cU 0 cU 011.1 Solids circulation flux scG scG99.1 Minimum fluidization velocity mfU (m/s) 0.0034 0.0038 Fuel Reactor Parameter Cold Unit Hot Unit Gas type 60 vol% helium + 40 vol% Air 25 vol% CH4+ 25 vol% CO2 +50 vol% H2O Temperature (oC) 40 900 Viscosity of gas (Pa.s) 2.02×10-5 4.03×10-5 Particle density s (kg/m3) 1560 2800 Pressure (bar) 1 3.34 Gas density f (kg/m3) 0.54 0.83 Particle diameter pd ( m ) 78 96 Bed diameter (mm) cD (286) cD23.1 (352) Superficial gas velocity cU 0 cU 011.1 Solids circulation flux scG scG99.1 Minimum fluidization velocity mfU (m/s) 0.0034 0.0047 206 Table D.6 Dimensionless group comparison for operation conditions in Table D.5 Air Reactor Dimensionless parameter Cold Unit Hot Unit Ratio between cold unit and hot unit gD U 20 2 000.1 cU 2 000.1 cU 1.00 f s   2889 2887 1.00 mfU U 0 cU 0286 cU 0292 0.98 0U G s s  c sc U G 0 4-106.41 c sc U G 0 4-1040.6  1 pd D 1308 1302 1.00 Fuel Reactor Dimensionless parameter Cold Unit Hot Unit Ratio between cold unit and hot unit gD U 20 2 036.0 cU 2 036.0 cU 1.00 f s   2889 3373 0.86 mfU U 0 cU 0286 cU 0236 1.21 0U G s s  c sc U G 0 4-106.41 c sc U G 0 4-106.41 1.00 pd D 3667 3667 1.00 207 Table D.7 Hot unit modeled by a cold unit fluidized with 40 vol% helium and 60 vol% air at temperature of 40 oC and atmosphere pressure (based on identical ratio of particle size to reactor size) Air Reactor Parameter Cold Unit Hot Unit Gas type 40 vol% helium + 60 vol% Air Air Temperature (oC) 40 950 Viscosity of gas (Pa.s) 2.00×10-5 4.94×10-5 Particle density s (kg/m3) 1560 2800 Gas density f (kg/m3) fc (0.74) fc79.1 (1.33) Pressure (bar) cP (1) cP59.4 (4.59) Particle diameter pd ( m ) pcd (78) pcd24.1 (97) Bed diameter (mm) cD (102) cD24.1 (126) Superficial gas velocity cU 0 cU 011.1 Solids circulation flux scG scG99.1 Minimum fluidization velocity mfU (m/s) 0.0034 0.0039 Fuel Reactor Parameter Cold Unit Hot Unit Gas type 40 vol% helium + 60 vol% Air 25 vol%CH4+ vol 25%CO2 +50 vol% H2O Temperature (oC) 40 900 Viscosity of gas (Pa.s) 2.00×10-5 4.03×10-5 Particle density s (kg/m3) 1560 2800 Pressure (bar) 1 4.59 Gas density f (kg/m3) 0.74 1.14 Particle diameter pd ( m ) 78 97 Bed diameter (mm) cD (286) cD24.1 (355) Superficial gas velocity cU 0 cU 011.1 Solids circulation flux scG scG99.1 Minimum fluidization velocity mfU (m/s) 0.0034 0.0048 208 Table D.8 Dimensionless group comparison for operation conditions in Table D.7 Air Reactor Dimensionless parameter Cold Unit Hot Unit Ratio between cold unit and hot unit gD U 20 2 000.1 cU 2 000.1 cU 1.00 f s   2108 2105 1.00 mfU U 0 cU 0286 cU 0285 1.00 0U G s s  c sc U G 0 4-106.41 c sc U G 0 4-1040.6  1.00 pd D 1308 1299 1.01 Fuel Reactor Dimensionless parameter Cold Unit Hot Unit Ratio between cold unit and hot unit gD U 20 2 036.0 cU 2 035.0 cU 1.03 f s   2108 2456 0.86 mfU U 0 cU 0286 cU 0231 1.24 0U G s s  c sc U G 0 4-106.41 c sc U G 0 4-106.41 1.00 pd D 3667 3660 1.00 209 Table D.9 Hot unit modeled by a cold unit fluidized with 20 vol% helium and 80 vol% air at temperature of 40 oC and atmosphere pressure (based on identical ratio of particle size to reactor size) Air Reactor Parameter Cold Unit Hot Unit Gas type 20 vol% helium + 80 vol% Air Air Temperature (oC) 40 950 Viscosity of gas (Pa.s) 1.99×10-5 4.94×10-5 Particle density s (kg/m3) 1560 2800 Gas density f (kg/m3) fc (0.93) fc79.1 (1.67) Pressure (bar) cP (1) cP76.5 (5.76) Particle diameter pd ( m ) pcd (78) pcd24.1 (97) Bed diameter (mm) cD (102) cD24.1 (126) Superficial gas velocity cU 0 cU 011.1 Solids circulation flux scG scG99.1 Minimum fluidization velocity mfU (m/s) 0.0034 0.0039 Fuel Reactor Parameter Cold Unit Hot Unit Gas type 20 vol% helium + 80 vol% Air 25 vol% CH4+ 25 vol% CO2 +50 vol% H2O Temperature (oC) 40 900 Viscosity of gas (Pa.s) 1.99×10-5 4.03×10-5 Particle density s (kg/m3) 1560 2800 Pressure (bar) 1 5.76 Gas density f (kg/m3) 0.93 1.43 Particle diameter pd ( m ) 78 97 Bed diameter (mm) cD (286) cD24.1 (355) Superficial gas velocity cU 0 cU 011.1 Solids circulation flux scG scG99.1 Minimum fluidization velocity mfU (m/s) 0.0034 0.0048 210 Table D.10 Dimensionless group comparison for operation conditions in Table D.9 Air Reactor Dimensionless parameter Cold Unit Hot Unit Ratio between cold unit and hot unit gD U 20 2 000.1 cU 2 000.1 cU 1.00 f s   1677 1677 1.00 mfU U 0 cU 0286 cU 0285 1.00 0U G s s  c sc U G 0 4-106.41 c sc U G 0 4-1040.6  1.00 pd D 1308 1299 1.01 Fuel Reactor Dimensionless parameter Cold Unit Hot Unit Ratio between cold unit and hot unit gD U 20 2 036.0 cU 2 035.0 cU 1.03 f s   1677 1958 0.86 mfU U 0 cU 0286 cU 0231 1.24 0U G s s  c sc U G 0 4-106.41 c sc U G 0 4-106.41 1.00 pd D 3667 3660 1.00 211 Table D.11 Hot unit modeled by a cold unit fluidized with 100 vol% air at temperature of 40 oC and atmosphere pressure (based on identical ratio of particle size to reactor size) Air Reactor Parameter Cold Unit Hot Unit Gas type Air Air Temperature (oC) 40 950 Viscosity of gas (Pa.s) 1.98×10-5 4.94×10-5 Particle density s (kg/m3) 1560 2800 Gas density f (kg/m3) fc (1.13) fc79.1 (2.02) Pressure (bar) cP (1) cP97.6 (6.97) Particle diameter pd ( m ) pcd (78) pcd25.1 (98) Bed diameter (mm) cD (102) cD25.1 (128) Superficial gas velocity cU 0 cU 012.1 Solids circulation flux scG scG00.2 Minimum fluidization velocity mfU (m/s) 0.0034 0.0039 Fuel Reactor Parameter Cold Unit Hot Unit Gas type Air 25 vol% CH4+ 25 vol% CO2 +50 vol% H2O Temperature (oC) 40 900 Viscosity of gas (Pa.s) 1.98×10-5 4.03×10-5 Particle density s (kg/m3) 1560 2800 Pressure (bar) 1 6.97 Gas density f (kg/m3) 1.13 1.73 Particle diameter pd ( m ) 78 98 Bed diameter (mm) cD (286) cD25.1 (358) Superficial gas velocity cU 0 cU 012.1 Solids circulation flux scG scG00.2 Minimum fluidization velocity mfU (m/s) 0.0034 0.0049 212 Table D.12 Dimensionless group comparison for operation conditions in Table D.11 Air Reactor Dimensionless parameter Cold Unit Hot Unit Ratio between cold unit and hot unit gD U 20 2 000.1 cU 2 000.1 cU 1.00 f s   1381 1386 1.00 mfU U 0 cU 0286 cU 0287 1.00 0U G s s  c sc U G 0 4-106.41 c sc U G 0 4-1040.6  1.00 pd D 1308 1306 1.00 Fuel Reactor Dimensionless parameter Cold Unit Hot Unit Ratio between cold unit and hot unit gD U 20 2 036.0 cU 2 036.0 cU 1.00 f s   1381 1618 0.85 mfU U 0 cU 0286 cU 0229 1.25 0U G s s  c sc U G 0 4-106.41 c sc U G 0 4-106.41 1.00 pd D 3667 3653 1.00 213 Table D.13 Operating conditions of the simulation runs for fuel reactor Cases Wfuel (kg) F0,CH4 (g/s) F s0,fuel (kg/s) fuelinX , (%) fuelMr (-) Fuel conversion (%) F1 18 1.42 0.48 40 135 95.3 F2 18 1.65 0.48 40 116 93.4 F3 18 1.89 0.48 40 101 91.4 F4 18 2.13 0.48 40 90 89.4 F5 18 2.36 0.48 40 81 87.4 F6 18 2.6 0.48 40 74 85.4 F7 18 2.84 0.48 40 67 83.5 F9 18 1.42 0.12 40 34 93.7 F10 18 1.42 0.24 40 67 95.2 F11 18 1.42 0.48 40 135 95.2 F12 18 1.42 0.96 40 270 95.3 F14 18 1.42 0.12 10 8 26.9 F15 18 1.42 0.24 10 17 52.3 F16 18 1.42 0.48 10 34 88.6 F17 18 1.42 0.96 10 67 94.2 F19 18 1.42 0.48 40 135 95.2 F20 18 1.42 0.48 70 236 95.3 F21 18 1.42 0.48 100 337 95.3 F23 36 1.42 0.48 10 34 93.7 F24 36 1.42 0.48 40 135 97.5 F25 36 1.42 0.48 70 236 97.5 F26 36 1.42 0.48 100 337 97.5 F28 72 1.42 0.48 10 34 96.3 F29 72 1.42 0.48 40 135 98.3 F30 72 1.42 0.48 70 236 98.3 F31 72 1.42 0.48 100 337 98.3 214 Table D.14 Operating conditions of the simulation runs for air reactor Cases F0,O2 (g/s) Fs0,airl (kg/s) airinX , (%) airMr (-) Oxygen conversion (%) A1 7.02 0.48 0 68 85.4 A2 7.02 0.5 30 49 79.9 A3 7.02 0.52 60 29 69.7 A4 7.02 0.54 90 8 40.8 A5 7.02 0.52 60 29 69.7 A6 7.02 0.97 60 55 86 A7 7.02 1.03 60 58 87 A8 7.02 0.48 0 68 85.4 A9 7.02 1.01 0 145 96.9 A10 7.02 0.96 0 136 96.3 A11 8.19 0.55 0 67 80.3 A12 8.19 1.15 0 141 94.4 A13 8.19 1.16 0 142 94.9 A14 9.36 0.96 0 102 86.1 A15 9.36 1.24 0 132 90.1 A16 9.36 1.24 0 133 90.9 2.8 3 3.2 3.4 3.6 3.8 4 4.2 0 0.5 1 1.5 2 2.5 3 Pressure drop across loopseal1 (kPa) G a s le a ka g e fro m fu el re a ct or to ai rr ea ct or (% ) UA1=1Um f UA1=2Um f UA1=4Um f UA1=6Um f Figure D.1 Leakage from fuel reactor to air reactor versus the pressure drop across loop-seal 1 (Ua0=3.0 m/s) 215 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 3 0 0.5 1 1.5 2 2.5 3 Pressure drop across loopseal1 (kPa) G as le ak ag e fro m fu el re ac to rt o ai rr ea ct or (% ) U A1 =1U mf UA1=2Umf UA1=4Umf U A1 =6U mf Figure D.2 Leakage from fuel reactor to air reactor versus the pressure drop across loop-seal 1 (Ua0=4.0 m/s) 3.3 3.4 3.5 3.6 3.7 3.8 3.9 4 4.1 0 0.5 1 1.5 2 2.5 3 Pressure drop across loopseal1 (kPa) G as le a ka ge fro m fu el re a ct or to ai r re ac to r (% ) UA1=1Umf UA1=2Umf UA1=4Umf U A1 =6U mf Figure D.3 Leakage from fuel reactor to air reactor versus the pressure drop across loop-seal 1 (Ua0=5.0 m/s) 216 -0.5 0 0.5 1 1.5 2 0 0.5 1 1.5 2 2.5 3 Aeration velocity in loopseal2, UA2 /Um f (-) G a s le ak ag e fr om fu el re ac to r to ai r re ac to r (% ) UA1=1Um f UA1=2Um f U A1 =4U m f U A1 =6U m f Figure D.4 Leakage from fuel reactor to air reactor versus aeration velocity for loop-seal 2, UA2 (Ua0=3.0 m/s) -0.5 0 0.5 1 1.5 2 0 0.5 1 1.5 2 2.5 3 Aeration velocity in loopseal2, UA2/Umf (-) G a s le a ka ge fr om fu el re ac to r to a ir re a ct o r (% ) U A1 =1U mf U A1 =2U mf U A1 =4U mf U A1 =6U mf Figure D.5 Leakage from fuel reactor to air reactor versus aeration velocity for loop-seal 2, UA2 (Ua0=4.0 m/s) 217 -0.5 0 0.5 1 1.5 2 0 0.5 1 1.5 2 2.5 3 Aeration velocity in loopseal2, UA2/Umf (-) G as le ak ag e fro m fu el re ac to rt o ai rr ea ct or (% ) UA1=1Umf U A1 =2U mf UA1=4Umf U A1 =6U mf Figure D.6 Leakage from fuel reactor to air reactor versus aeration velocity for loop-seal 2, UA2 (Ua0=5.0 m/s) 10 15 20 25 0 0.1 0.2 0.3 0.4 0.5 Solids circulation flux between air reactor and fuel reactor (kg/m2.s) G as le ak ag e fro m ai rr ea ct or to fu el re ac to r (% ) U A1 =1U mf UA1=2Umf UA1=4Umf U A1 =6U mf Figure D.7 Gas leakage from air reactor to fuel reactor versus solids circulation flux between air reactor and fuel reactor (Ua0=3.0 m/s) 218 8 10 12 14 16 18 20 22 24 0 0.1 0.2 0.3 0.4 0.5 Solids circulation flux between air reactor and fuel reactor (kg/m2 .s) G a s le a ka g e fr o m ai rr e ac to r to fu e lr e a ct or (% ) U A1 =1U mf U A1 =2U mf U A1 =4U mf U A1 =6U mf Figure D.8 Gas leakage from air reactor to fuel reactor versus solids circulation flux between air reactor and fuel reactor (Ua0=4.0 m/s) 6 8 10 12 14 16 0 0.1 0.2 0.3 0.4 0.5 Solids circulation flux between air reactor and fuel reactor (kg/m 2.s) G as le ak ag e fro m ai rr ea ct or to fu el re ac to r (% ) UA1=1Umf U A1 =2U mf UA1=4Umf U A1 =6U mf Figure D.9 Gas leakage from air reactor to fuel reactor versus solids circulation flux between air reactor and fuel reactor (Ua0=5.0 m/s) 219 -0.5 0 0.5 1 1.5 2 0 10 20 30 40 50 60 70 80 90 100 Aeration velocity in loopseal2, U A2 /U m f (-) G a s le ak a ge fro m F 1 po rt to fu el re a ct o r (% ) UA1=1Umf U A1 =2U mf U A1 =4U mf UA1=6Umf Figure D.10 Aeration velocity in loop-seal 2 versus gas leakage from F1 port to fuel reactor (Ua0=3.0 m/s) -0.5 0 0.5 1 1.5 2 0 10 20 30 40 50 60 70 80 90 100 Aeration velocity in loopseal2, UA2/Umf (-) G as le a ka g e fro m F 2 po rt to fu el re ac to r (% ) UA1 =1Um f UA1 =2Um f UA1 =4Um f UA1 =6Um f Figure D.11 Aeration velocity in loop-seal 2 versus gas leakage from F2 port to fuel reactor (Ua0=3.0 m/s) 220 -0.5 0 0.5 1 1.5 2 0 10 20 30 40 50 60 70 80 90 100 Aeration velocity in loopseal2, UA2/Um f (-) G as le a ka ge fro m F 1 po rt to fu el re a ct or (% ) U A1 =1U mf UA1=2Umf U A1 =4U mf UA1=6Umf Figure D.12 Aeration velocity in loop-seal 2 versus gas leakage from F1 port to fuel reactor (Ua0=4.0 m/s) -0.5 0 0.5 1 1.5 2 0 10 20 30 40 50 60 70 80 90 100 Aeration velocity in loopseal2, UA2/Umf (-) G as le a ka g e fro m F 2 po rt to fu el re ac to r (% ) U A1 =1U mf U A1 =2U mf UA1=4Umf UA1=6Umf Figure D.13 Aeration velocity in loop-seal 2 versus gas leakage from F2 port to fuel reactor (Ua0=4.0 m/s) 221 -0.5 0 0.5 1 1.5 2 0 10 20 30 40 50 60 70 80 90 100 Aeration velocity in loopseal2, UA2/Umf (-) G a s le a ka ge fr om F 1 po rt to fu el re ac to r (% ) UA1=1Umf U A1 =2U mf UA1=4Umf U A1 =6U mf Figure D.14 Aeration velocity in loop-seal 2 versus gas leakage from F1 port to fuel reactor (Ua0=5.0 m/s) -0.5 0 0.5 1 1.5 2 0 10 20 30 40 50 60 70 80 90 100 Aeration velocity in loopseal2, U A2 /U mf (-) G as le ak ag e fro m F 2 po rt to fu el re ac to r (% ) UA1=1Umf U A1 =2U mf U A1 =4U mf U A1 =6U mf Figure D.15 Aeration velocity in loop-seal 2 versus gas leakage from F2 port to fuel reactor (Ua0=5.0 m/s) 222 1 2 3 4 5 6 7 8 9 10 11 0 1000 2000 3000 4000 5000 6000 Pressure P-P atmosph ere (kPa) H ei gh ta b ov e th e ga s di st rib ut or of ai r re ac to r( m m ) Helium concentration=0% Helium concentration=20% Helium concentration=40% Helium concentration=60% Helium concentration=80% B9 B8 B7 A2 B2 B6 D2 D3 D6 Figure D.16 Pressure loop 1 versus helium concentration (Ua0=5.0 m/s, UA1=4Umf, UA2=1Umf) 2 4 6 8 10 12 0 1000 2000 3000 4000 5000 6000 Pressure P-Patmo sphere (kPa) H ei gh ta bo ve th e ga s di st rib ut o ro fa ir re ac to r( m m ) Helium concentration=0% Helium concentration=20% Helium concentration=40% Helium concentration=60% Helium concentration=80% B8 B9 B6 B7 A2 D4 B5 D5 Figure D.17 Pressure loop 2 versus helium concentration (U0=5.0 m/s, UA1=4Umf, UA2=1Umf) 223 0 2 4 6 8 10 12 0 1000 2000 3000 4000 5000 6000 Pressure P-P atmosphere (kPa) H ei gh ta bo ve th e ga s di st rib ut or of ai rr ea ct or (m m ) Helium concentration=0% Helium concentration=20% Helium concentration=40% Helium concentration=60% Helium concentration=80% B9 B8 B7 B6 A2 B2 D2 D3 D6 Figure D.18 Pressure loop 1 versus helium concentration (U0=5.0 m/s, UA1=4Umf, UA2=1.5Umf) 2 4 6 8 10 12 14 0 1000 2000 3000 4000 5000 6000 Pressure P-Patm osp here (kPa) H ei gh ta bo ve th e g as di st rib ut or of ai rr e ac to r (m m ) Helium concentration= 0% Helium concentration=20% Helium concentration=40% Helium concentration=60% Helium concentration=80% B9 B7 B8 B5 D4 B6 A2 D5 Figure D.19 Pressure loop 2 versus helium concentration (Ua0=5.0 m/s, UA1=4Umf, UA2=1.5Umf) 224 0 5 10 15 0 1000 2000 3000 4000 5000 6000 Cross sectional average solids hold-up (%) H e ig ht a bo ve th e g as di st rib ut or of a ir re a ct or (m m ) Helium concentration=0% Helium concentration=20% Helium concentration=40% Helium concentration=60% Helium concentration=80% Figure D.20 Solids fraction distribution in air reactor at different helium concentration (Ua0=5.0 m/s, UA1=4Umf, UA2=0Umf) 0 2 4 6 8 10 12 14 0 1000 2000 3000 4000 5000 6000 Cross sectional average solids hold-up (%) H ei gh ta bo ve th e ga s di st rib ut or of ai rr ea ct or (m m ) Helium concentration=0% Helium concentration=20% Helium concentration=40% Helium concentration=60% Helium concentration=80% Figure D.21 Solids fraction distribution in air reactor at different helium concentration (Ua0=5.0 m/s, UA1=4Umf, UA2=1Umf) 225 0 20 40 60 80 100 0 5 10 15 20 25 Helium concentration (mol%) S ol id ci rc ul at io in flu x in lo op 1 (k g/ m2 .s ) U a0 =2.5m/s U a0 =3.0m/s U a0 =4.0m/s U a0 =5.0m/s U a0 =6.0m/s U a0 =7.0m/s U a0 =8.0m/s Figure D.22 Solids circulation flux in loop 1 versus helium concentration in gas mixture for different actual superficial gas velocities (UA1=4Umf, UA2=0Umf) 0 20 40 60 80 100 0 5 10 15 20 25 Helium concentration (mol%) S o lid ci rc ul at io n flu x in lo o p1 (k g /m2 .s ) U a 0 =2.5m/s Ua 0=3.0m/s U a 0 =4.0m/s Ua 0=5.0m/s U a 0 =6.0m/s U a 0 =7m/s Ua 0=8.0m/s Figure D.23 Solids circulation flux in loop 1 versus helium concentration in gas mixture for different actual superficial gas velocities (UA1=4Umf, UA2=1Umf) 226 0 20 40 60 80 100 0 5 10 15 20 25 30 Helium concentration (mol%) S ol id ci rc ul at io n flu x in lo op 2 (k g/ m2 .s ) U a0 =2.5m/s U a0 =3.0m/s U a0 =4.0m/s Ua0=5.0m/s U a0 =6.0m/s U a0 =7.0m/s U a0 =8.0m/s Figure D.24 Solids circulation flux in loop 2 versus helium concentration in gas mixture for different superficial gas velocities in air reactor, Ua0 (UA1=4Umf, UA2=1Umf) 0 20 40 60 80 100 0 5 10 15 20 25 30 35 40 45 50 55 Helium concentration (mol%) To ta ls ol id ci rc ul at io n flu x (k g/ m2 .s ) U a0 =2.5m/s U a0 =3.0m/s Ua0=4.0m/s U a0 =5.0m/s U a0 =6.0m/s U a0 =7.0m/s U a0 =8m/s Figure D.25 Total solids circulation flux versus helium concentration in gas mixture for different actual superficial gas velocities. (UA1=4Umf, UA2=1Umf) 227 APPENDIX E MODIFICATION OF THE ORIGINAL CLC COLD-FLOW MODEL Figure E.1 shows the original schematic of the CLC cold-flow model. Experimental results in the original cold-flow model showed that the solids circulation flux in the loop 1 was relatively small. Because the solids circulation flux in the loop 1 is crucial for oxygen transportation and energy transfer between the air and fuel reactors, it is necessary to improve the solids circulation to achieve better performance of the CLC reactor. According to the experimental experience, the pressure head (i.e. pressure drop between D6 and D3 shown in Figures E.1 and E.2) of the loop1 has crucial effect on the solids circulation flux in loop1. Thus, increasing the elevation of fuel reactor to increase the pressure-head would produce higher solids circulation flux. Therefore, as shown in Figure E.2, following modifications had been made: a. Removed the top section of Fuel reactor; b. Moved the fuel reactor up; c. Extended the connection of loop-seal 1 to the fuel reactor. The height difference between the solids exit of fuel reactor and solids exit of loop-seal 1 (as shown in Figures E1 and E2) is increased from 0.2 m to 0.6 m, and the pressure head of the loop1 is increased by increasing the pressure at point D6 for a given Ua0. In addition, the bottom end of the lower downcomer in the fuel reactor was sealed and a hole of 0.1 m in diameter was opened on the side as shown in Figure E.2 to prevent the gas leakage from the fuel reactor to the cyclones through the lower downcomer. 228 Figure E.1 Schematic of the CLC cold-flow model Figure E.2 Schematic of the CLC cold-flow model CLC before-modification after-modification 229 APPENDIX F MEASUREMENT ERROR AND STANDARD DEVIATION 0 5 10 15 20 25 30 2.5 3 4 5 Superfic ial gas veloc ity for air reac tor (m /s ) S ol id s ci rc ul at io n flu x in lo op 1 (k g/ m 2 .s ) UA1 =1Um f UA1 =2Um f UA1 =4Um f UA1 =6Um f Figure F.1 Measurement error bar for Figure 4.1 0 2 4 6 8 10 12 14 16 18 20 2.5 3 4 5 Superfic ial gas veloc ity for air reactor (m /s) S ol id s ci rc ul at io n flu x in lo op 1 (k g/ m 2 . s) UA1=1Umf UA1=2Umf UA1=4Umf UA1=6Umf Figure F.2 Measurement error bar for Figure 4.4 230 0 2 4 6 8 10 12 14 16 2.5 3 4 5 Supe rfic ia l gas ve locity fo r a ir reacto r (m/s) S ol id s ci rc ul at io n flu x in lo op 1 (k g/ m 2 .s ) UA1 = 1U m f UA1 =2U m f UA1 =4U m f UA1 =6U m f Figure F.3 Measurement error bar for Figure 4.5 0 1 2 3 4 5 6 7 8 9 10 2.5 3 4 5 Superfic ial gas veloc ity for air reac tor (m /s ) S ol id s ci rc ul at io n flu x in lo op 1 (k g/ m 2 .s ) UA1 =1Umf UA1 =2Umf UA1 =4Umf UA1 =6Umf Figure F.4 Measurement error bar for Figure 4.7 231 0 5 10 15 20 25 2.5 3 4 5 S uperfic ia l gas veloc ity for air reac tor (m /s ) S ol id s ci rc ul at io n flu x in lo op 1 (k g/ m 2 .s ) UA1 =1Umf UA1 =2Umf UA1 =4Umf UA1 =6Umf Figure F.5 Measurement error bar for Figure 4.8 0 5 10 15 20 25 30 2.5 3 4 5 Superfic ial gas veloc ity for air reac tor (m /s ) S ol id s ci rc ul at io n flu x in lo op 1 (k g/ m 2 .s ) UA1=1Umf UA1=2Umf UA1=4Umf UA1=6Umf Figure F.6 Measurement error bar for Figure 4.9 232 0 5 10 15 20 25 30 35 2.5 3 4 5 Superfic ial gas veloc ity for air reac tor (m /s ) S ol id s ci rc ul at io n flu x in lo op 1 (k g/ m 2 .s ) UA1=1Um f UA1=2Um f UA1=4Um f UA1=6Um f Figure F.7 Measurement error bar for Figure 4.10 0 5 10 15 20 25 30 1 2 4 6 Aeration ve locity in loop-seal 1 UA1 /Um f (-) UA 2= 0Umf UA 2= 1Umf UA 2= 1.5Um f Figure F.8 Measurement error bar for Figure 4.11 233 0 5 10 15 20 1 2 4 6 Aeration velocity in loop-seal 1 UA1/Umf (-) S ol id s ci rc ul at io n flu x in lo op 2 (k g/ m 2 . s) UA2=1Umf UA2=1.5Umf Figure F.9 Measurement error bar for Figure 4.12 0 5 10 15 20 25 30 35 1 2 4 6 Aera tion ve locity in loop-sea l 1 UA1/U m f (-) S ol id s ci rc u la tio n flu x in lo o p 1 (k g/ m 2 .s ) U A2= 0U m f U A2 = 1U m f U A2 = 1.5 Um Figure F.10 Measurement error bar for Figure 4.13 234 Table F.1 Standard deviation of fluctuation of gas leakage from fuel reactor to air reactor for Figure 5.2 (Ua0=2.5m/s) Pressure drop (kPa) Average gas leakage (%) Standard deviation (%) 2.18 2.73 0.11 2.565 2.07 0.06 3.376 0.78 0.04 3.057 2.78 0.04 2.933 2.97 0.07 3.581 1.09 0.09 3.198 1.77 0.07 3.278 1.62 0.04 3.75 1.17 0.08 3.395 2.74 0.07 3.436 1.24 0.09 3.827 0.85 0.03 Table F.2 Standard deviation of fluctuation of gas leakage from fuel reactor to air reactor for Figure 5.4 (Ua0=5.0m/s) UA1 UA2 Average gas leakage (%) Standard deviation (%) 1Umf 0Umf 2.73 0.11 1Umf 1Umf 2.07 0.06 1Umf 1.5Umf 0.78 0.04 2Umf 0Umf 2.78 0.04 2Umf 1Umf 2.97 0.07 2Umf 1.5Umf 1.09 0.09 4Umf 0Umf 1.77 0.07 4Umf 1Umf 1.62 0.04 4Umf 1.5Umf 1.17 0.08 6Umf 0Umf 2.74 0.07 6Umf 1Umf 1.24 0.09 6Umf 1.5Umf 0.85 0.03 Table F.3 Standard deviation of fluctuation of gas leakage from fuel reactor to air reactor for Figure 5.5 (Ua0=2.5m/s) UA1 UA2 Average gas leakage (%) Standard deviation (%) 1Umf 0Umf 2.79 0.36 1Umf 1Umf 2.67 0.12 1Umf 1.5Umf 2.30 0.15 2Umf 0Umf 2.77 0.36 2Umf 1Umf 2.22 0.33 2Umf 1.5Umf 2.59 0.26 4Umf 0Umf 2.30 0.26 4Umf 1Umf 2.83 0.15 4Umf 1.5Umf 2.28 0.12 6Umf 0Umf 2.46 0.23 6Umf 1Umf 2.42 0.23 6Umf 1.5Umf 2.16 0.16

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0058888/manifest

Comment

Related Items