CHEMICAL LOOPING GASIFICATION PROCESSES
DISSERTATION
Presented in Partial Fulfillment of the
Requirements for the Degree Doctor of Philosophy in the Graduate
School of The Ohio State University
By
Fanxing Li, M.S.
*****
The Ohio State University
2009
Dissertation Committee:
Approved by
Dr. Liang-Shih Fan, Adviser
Dr. Winston Ho
_______________________________
Dr. Michael Paulaitis
Adviser
Dr. Wenzhi Luo
Graduate Program in Chemical
Engineering
-1-
-I-
ABSTRACT
Through the assistance of oxygen carrier particles, the chemical looping processes
convert carbonaceous fuels while producing a sequestration ready CO2 stream. Two
chemical looping gasification processes, the syngas chemical looping (SCL) process and
the coal direct chemical looping (CDCL) process, are developed for hydrogen and
electricity co-production from carbonaceous fuels. Both processes involve the reduction
of a metal oxide with a fuel followed by regeneration of the reduced metal oxide with
steam and air in a cyclic manner. The syngas chemical looping process converts gaseous
fuels such as syngas and methane while the coal direct chemical looping process converts
solid fuels such as coal and biomass.
A novel iron oxide based composite oxygen carrier particle is currently being
developed for the aforementioned chemical looping gasification process. The physical
and chemical properties of the particles including compressive strength, attrition rate,
reactivity and recyclability are tested. Reduction of the particles with syngas in an
integral bed reactor is performed followed by oxidizing the reduced particles in the same
reactor with steam and then air. More than 99.7% syngas is converted during the
reduction step. During the regeneration step, hydrogen with an average purity of 99.8% is
-II-
produced. The results indicate that the particle is suitable for the chemical looping
gasification processes.
In the SCL process, the oxygen carrier particle is first reduced by a gaseous fuel
in a first reactor, the reducer. Next, the reduced oxygen carrier is partially regenerated
with steam to produce hydrogen in a second reactor, the oxidizer. The partially
regenerated particle is further oxidized to its original oxidation state by air in a third unit,
the combustor. The SCL process is extensively studied both analytically and
experimentally. Thermodynamic analysis shows that a countercurrent moving bed design
is suitable for both the reducer and the oxidizer. ASPEN Plus® simulation further
suggests the optimum operating conditions and pollutant control strategies for the SCL
process. Experiments are carried out in a bench scale (2.5 KWth) moving bed reactor to
validate the reducer and oxidizer operations. A quartz fixed bed reactor and TGA are
used to mimic the combustor operations. The experimental results match well with the
simulation outcomes. More than 99.5% of the syngas is converted during the reducer test.
The hydrogen generated during the subsequent oxidizer test has an average purity in
excess of 99.95%. The particles can also sustain the high operating temperature of
combustor without losing its reactivity and recyclability. ASPEN Plus® simulation
shows that the SCL process can improve the efficiency of the current coal to hydrogen
process by 4 – 10% with 100% CO2 capture. When integrated with the indirect coal-toliquid process, “the Chemical Looping system proposed by OSU has the potential to
significantly (~10%) increase the yield of the conventional cobalt based F-T process and
-III-
allow more efficient heat recovery and much lower (~19%) carbon emissions.”
The CDCL process has a fuel conversion scheme similar to that of the SCL
process. However, the CDCL process faces additional challenges resulting from direct
solid fuel conversion. These challenges include ash and pollutant handling, solid fuel
conversion enhancement, and heat management and integration. A solid fuel conversion
enhancement scheme is proposed and tested in the bench scale moving bed reactor. More
than 90% conversions for various types of coal chars are achieved. ASPEN Plus®
simulation is used to both analyze the fate of the pollutants involved and the optimization
of energy integration. The process simulation using ASPEN Plus® shows that the
hydrogen production efficiency for the CDCL process can reach nearly 80%.
A 25 KWth sub-pilot scale chemical looping demonstration unit is designed and
constructed. It is capable of demonstrating the chemical looping gasification processes in
an integrated, continuous manner. The unit, which is comprised of six sub-systems, has
been fully assembled. Preliminary tests including reactor leakage tests, solid flow
calibration, particle hydrodynamic studies, and integrated reactor operations are
performed. The test results show that the sub-pilot unit meets the design standard and is
ready for the SCL process demonstration. It can also be used for CDCL process
demonstration with minor modifications.
-IV-
To my family and friends whom I love.
-V-
ACKNOWLEDGMENTS
This thesis would not have been possible without the support of many people.
First and foremost, I would like to thank my adviser, Dr. Liang-Shih Fan, for his
intellectual supervision, continuous support, endless patience, motivation and
encouragement throughout my graduate studies. I am indebted for the time he spent on
helping me with every aspect of research, for his invaluable suggestions and guidance,
and for all his experiences that he shared with me, not only pertaining to research but also
about various other facets of life. I am fortunate to have Dr. Fan, an extremely
knowledgeable and enthusiastic professor, as my adviser. Working with him exposed me
to many invaluable experiences that I will deeply cherish for the rest of my life.
I would like to thank my committee members: Dr. Winston Ho, Dr. Michael
Paulaitis, and Dr. David Tomasko. I am grateful for the guidance they provided me. I
benefited enormously from their comments and suggestions. I would also like to thank
my M.S. advisors at Tsinghua University, Dr. Yong Jin, Dr. Dezheng Wang, Dr. Xiaolin
Wang, and Dr. Fei Wei for their patience and support.
-VI-
Appreciation goes to my fellow lab mates: Deepak Sridhar, Ray Kim, Liang Zeng,
Andrew Tong, and Fei Wang. It is the collaboration and tremendous team efforts that
made the process development go this far.
I am very thankful for my seniors, Dr. Puneet Gupta and Dr. Luis G. VelazquezVargas, for helping me getting started and for their unusual patience towards the many
doubts and questions. I have greatly benefited from the advices and suggestions from Dr.
Alissa Park. I also would like to thank my group mates, Zhao Yu, Dr. Songgeng Li, Dr.
Qussai Marashdeh, Fuchen Yu, William Wang, Shwetha Ramkumar, Zhenchao Sun, and
Orin Hemminger for their constructive suggestions and assistance.
I am also very grateful to Ms. Amy Dudley, Ms. Lynn Flanagan, Ms. Susan
Tesfai, and Ms. Kari Uhl for their administrative help and support. I would also like to
express my thanks to all of my friends and colleagues in the Department of Chemical and
Biomolecular Engineering for their encouragement and invaluable scientific discussions.
Finally, my special thanks go to my family. I am deeply and forever indebted to
my parents for their everlasting love, support and encouragement, for giving of
themselves beyond the call of duty. Especially, I want to thank my father for his
unconditional love.
-VII-
VITA
April 8th, 1980……………………………..
Born —
China
September 1997 – August 2001…………...
B. S. Chemical Engineering
Tsinghua University, Beijing, China
September 2001 – August 2004…………...
M. S. Chemical Engineering
Tsinghua University, Beijing, China
September 2004 – present…………………
Graduate Research Associate, The Ohio
State University, Columbus, OH, U.S.A.
Zhengzhou, Henan Province,
PUBLICATIONS
Li, F.; Fan, L.-S., Clean Coal Conversion Processes - Progress and Challenges. Energy
and Environmental Science, 2008, 1: 248 - 267.
Fan, L.-S.; Li, F.; Ramkumar, S., Utilization of Chemical Looping Strategy in Coal
Gasification Processes. Particuology, 2008, 6(3): 131-142.
Fan, L.-S.; Li, F, Clean coal. Physics World, 2007, 20(7): 37-41.
Li, F.; Wang, Y.; Wang, D.; Wei F., Characterization of single-wall carbon nanotubes by
N2 adsorption, Carbon, 2004, 42: 2375-2383.
Wang, D.; Li, F.; Zhao, X., Diffusion Limitation in Fast Transient Experiments,
Chemical Engineering Science, 2004, 59: 5615-5622.
Zhao, X.; Li, F.; Wang, D., Comparison of microkinetics and Langmuir -Hinshelwood
models of the partial oxidation of methane to synthesis gas, Studies in Surface Sciences
and Catalysis, 2004, 147: 235-240.
-VIII-
FIELDS OF STUDY
Major Field: Chemical Engineering
Minor Field: Applied Statistics
-IX-
TABLE OF CONTENTS
Page
ABSTRACT........................................................................................................................II
ACKNOWLEDGMENTS ................................................................................................ VI
VITA .............................................................................................................................. VIII
LIST OF TABLES......................................................................................................... XIII
LIST OF FIGURES .........................................................................................................XV
CHAPTER 1 ........................................................................................................................1
INTRODUCTION TO CLEAN COAL CONVERSION PROCESSES .............................1
1.1 Background............................................................................................................ 1
1.2 Coal Combustion Processes................................................................................... 3
1.2.1 Energy Efficiency Improvement..................................................................... 4
1.2.2 Flue Gas Pollutant Control Methods............................................................... 6
1.2.3 CO2 Capture Systems...................................................................................... 8
1.3 Coal Gasification Processes................................................................................. 11
1.3.1 Overview....................................................................................................... 12
1.3.2 ASPEN Analysis on IGCC System with CO2 Capture – A Case Study....... 15
1.4 Advanced Coal Conversion Processes ................................................................ 18
1.4.1 Membrane Based Gasification Systems........................................................ 18
1.4.2 Chemical Looping Based Gasification Systems ........................................... 25
1.4.3 Direct Coal Chemical Looping Processes..................................................... 38
1.5 Concluding Remarks ........................................................................................... 43
CHAPTER 2 ......................................................................................................................67
OXYGEN CARRIER PARTICLE FOR CHEMICAL LOOPING GAISIFICATION.....67
2.1 Syngas Chemical Looping Process Overview..................................................... 67
2.2 Oxygen Carrier Selection .................................................................................... 73
2.3 Oxygen Carrier Performance............................................................................... 74
2.3.1 Experimental ................................................................................................. 74
2.3.2 Results and Discussions................................................................................ 79
2.4. Conclusions ........................................................................................................ 84
CHAPTER 3 ....................................................................................................................102
SYNGAS CHEMICAL LOOPING GASIFICATION PROCESS..................................102
3.1 Thermodynamic Analyses of SCL Reactor Behavior ....................................... 103
3.1.1 Reactor Thermodynamic Analysis Based on Analytical Method............... 107
3.1.2 ASPEN Plus® Simulation of SCL Reactor Systems ................................... 114
3.2 Syngas Chemical Looping (SCL) Process Testing ........................................... 121
3.2.1 Experimental ............................................................................................... 122
3.2.2 Results and Discussions.............................................................................. 127
-X-
3.3 Process Simulation of the Traditional Gasification Processes and the Syngas
Chemical Looping Processes................................................................................... 130
3.3.1 Common Assumptions and Model Setup ................................................... 131
3.3.2 Description of Various Systems.................................................................. 133
3.3.3 ASPEN Plus® Simulation, Results, and Analyses ...................................... 135
3.4 Conclusions ....................................................................................................... 136
CHAPTER 4 ....................................................................................................................167
COAL DIRECT CHEMICAL LOOPING PROCESS ....................................................167
4.1 Coal Direct Chemical Looping (CDCL) Process Overview ............................. 167
4.1.1 Coal Direct Chemical Looping Process - Configuration I.......................... 168
4.1.2 Coal Direct Chemical Looping Process - Configuration II ........................ 171
4.1.3 Comments on the Coal Direct Chemical Looping Process......................... 172
4.2 Challenges to the Coal Direct Chemical Looping Processes and Strategy for
Improvements .......................................................................................................... 173
4.2.1 Oxygen Carrier Particle Reactivity and Char Reaction Enhancement ....... 173
4.2.2 Configurations and Conversions of the Reducer ........................................ 176
4.2.3 Performance of the Oxidizer and the Combustor ....................................... 186
4.2.4 Fates of Pollutants and Ash......................................................................... 189
4.2.5 Energy Management, Heat Integration, and General Comments ............... 192
4.3 Process Simulations on the Coal Direct Chemical Looping Process ................ 196
4.3.1 ASPEN Model Setup .................................................................................. 197
4.3.2 Simulation Results ...................................................................................... 198
4.4 Concluding Remarks ......................................................................................... 200
CHAPTER 5 ....................................................................................................................219
SUB-PILOT SCALE CHEMICAL LOOPING SYSTEM ..............................................219
5.1 Introduction ....................................................................................................... 219
5.2 Sub-Pilot Reactor System Design...................................................................... 220
5.2.1 Arrangement of the Overall Reactor System .............................................. 220
5.2.2 Gas Storage Assembly ................................................................................ 222
5.2.3 Gas Mixing and Delivery Panel.................................................................. 223
5.2.4 Reactor Assembly ....................................................................................... 224
5.2.5 Automation System..................................................................................... 229
5.2.6 Steam Generator and Air Compressors....................................................... 229
5.3 Preliminary Reactor Tests ................................................................................. 230
5.3.1 Reactor Leak Test ....................................................................................... 230
5.3.2 Rotary Solid Feeder Test and Solid Flow Calibration ................................ 232
5.3.3 Combustor and Particle Attrition Test ........................................................ 232
5.3.4 Test of the Integrated Unit .......................................................................... 236
5.4 Improvements and Future Tests of the Sub-Pilot Unit ...................................... 237
5.4.1 Valve System Design.................................................................................. 237
5.4.2 Bed Height Control ..................................................................................... 238
5.4.3 Future Test Plan .......................................................................................... 239
5.4.4 Coal Direct Chemical Looping Applications.............................................. 239
5.5 Concluding Remarks ......................................................................................... 239
CHAPTER 6 ....................................................................................................................256
-XI-
NOVEL APPLICATIONS OF CHEMICAL LOOPING, CONCLUSIONS AND
RECOMMENDATIONS.................................................................................................256
6.1 Novel SCL Applications – A Coal-to-Liquids Configuration........................... 256
6.1.1 Process Overview........................................................................................ 256
6.1.2 Mass/Energy Balance and Economic Evaluation ....................................... 258
6.2 Concluding Remarks ......................................................................................... 260
6.3 Recommendations ............................................................................................. 262
BIBLIOGRAPHY............................................................................................................266
-XII-
LIST OF TABLES
Page
Table 1.1
Energy Conversion Efficiencies (HHV) of Various Coal Combustion
Technologies and Energy Penalty for CO2 Capture Using MEA …………………….... 46
Table 1.2
ASPEN models for the key units in the IGCC process ……………………47
Table 1.3
Power Balance in a 1000 MWth IGCC plant with CO2 capture…………...48
Table 1.4
Performances of H2-Selective Membranes………………………………...49
Table 1.5
Reducer Performances Using Different Reactor Designs………………....50
Table 2.1
Reactor Type, Main Reactions, and Operating Conditions for the SCL
Reactors………………………………………………………………………………….86
Table
2.2
Comparisons
of
the
Key
Properties
of
Different
Metal
Oxide
Candidates…………………………………………………………………………….....87
Table 2.3
Inlet Gas Composition during the Reduction Experiment…………………88
Table 2.4 Reactivity Comparisons between Iron Ore Powders and OSU Composite
Particles………………………………………………………………………………….89
Table 2.5 Gas and Solid Conversions in the Reduction Stage of the Fixed Bed
Experiment………………………………………………………………………………90
Table 2.6
Gas and Solid Conversions in the Oxidation Stage of the Fixed Bed
Experiment……………………………………………………………………………....91
-XIII-
Table 3.1 Equilibrium Gas Compositions with Different Oxidization States of Iron at
850ºC……………………………………………………………………………………137
Table 3.2 Parameters for the ASPEN Plus® Model…...……………………………….138
Table 3.3 Components List in Reducer Simulation…………………………………...139
Table 3.4 Parameters in the ORGANIC Databank in ASPEN Plus®………………....140
Table 3.5 Gas flow to the bench scale unit……………………………………………141
Table 3.6 Solid conversions and carbon depositions along the reactor……………....141
Table 3.7 Gas composition at the reactor outlet over three hours…………………....141
Table 3.8 Physical and Chemical Properties of Pittsburgh #6 Coal………………….142
Table 3.9 ASPEN Models for the Key Units in the IGCC Process.…………………..143
Table 3.10 Comparisons of the Process Analysis Results……………………………..144
Table 4.1 Reducer Mass Balance Based on the ASPEN Plus® Model at 900 ºC…….202
Table 4.2 Summary of the Reducer Demonstration Results using Coal, Coal char, and
Volatile…………………………………………………………………………………203
Table 4.3 Overall Input-Output Diagram for the Mass Flow of the CDCL Process…..204
Table 4.4 Overall Input-Output Diagram for the Energy Flow of the CDCL Process...204
Table 4.5 Heat and Energy Requirements in the CDCL Process……………………..204
Table 4.6 Power Balance in the CDCL Process………………………………………204
Table 6.1 Overall Energy Input/Output for the SCL-CTL Process……………………264
-XIV-
LIST OF FIGURES
Page
Figure 1.1 a. Three Different (Long Term) World Oil Price Scenarios Predicted by EIA;
b. World Energy Consumption in 2030 Based on Energy Sources……………………...51
Figure 1.2
Simplified Schematic Diagram of a Pulverized Coal (PC) Combustion
Process for Power Generation……………………………………………………………52
Figure 1.3
Conceptual Schematic of a. the MEA Scrubbing Technology for CO2
Separation; b. the Chilled Ammonia Technology for CO2 Separation ………………….53
Figure 1.4 Conceptual Schematic of Carbonation-Calcination Reaction (CCR) Process
Integration in a 300 MWe Coal Fired Power Plant Depicting Heat Integration
Strategies…..………………………………………………………………………….….54
Figure 1.5 Schematic Diagram of Coal Gasification Processes………………………..55
Figure 1.6 IGCC Process with CO2 Capture……………………………………….…...56
Figure 1.7 Multiple Stage Membrane System for CO2 Recovery ………………….….57
Figure 1.8 a) Schematic of H2-Selective Membrane Enhanced IGCC Process; b)
Schematic of CO2-Selective Membrane Enhanced IGCC Process……………………...58
Figure 1.9 Integrated Membrane Separation with Gasifier…………………………....59
Figure 1.10 Schematic Flow Diagram of Syngas Chemical Looping Combustion
Processes…………………………………………………………………………………60
Figure 1.11 Simplified Schematic of the Syngas Chemical Looping Process…………61
-XV-
Figure 1.12 Syngas Chemical Looping Enhanced Coal-to-Liquids (SCL-CTL)
Process…………………………………………………………………………………...62
Figure 1.13 Schematic Flow Diagram of the Calcium Looping Process……………....63
Figure 1.14 Schematic Diagram of Coal-Direct Chemical Looping Process…………..64
Figure 1.15 Schematic Diagram of HyPr-RING Process………………………………65
Figure 1.16 Efficiency Comparisons among Various Coal Conversion Technologies with
> 90% CO2 Capture……………………………………………………………………..66
Figure 2.1 Simplified Schematic of the Syngas Chemical Looping Process for Hydrogen
Production from Coal……………………………………………………………………92
Figure 2.2 Schematic of the Experimental Setup for the Particle Reactivity and
Recyclability Studies…………………………………………………………………....93
Figure 2.3 Schematic of the Entrained Bed Setup for Particle Attrition Studies………94
Figure 2.4 Schematic of the Fixed Bed Reactor Setup………………………………...95
Figure 2.5 a. Pure Fe2O3 Particle Recyclability Test;
b. Composite Particle
Recyclability Test; c. Time (min) Required for 80% Reduction and Oxidation at Various
Cycles for the Composite Particle……………………………………………………….97
Figure 2.6 Crushing Strength Test of OSU Composite Pellets…………………………98
Figure 2.7 Attrition Rate of the Composite Pellet in an Entrained Flow Reactor………99
Figure 2.8 Composition of the Exhaust Gas Stream from the Fixed Bed Reactor during
the Reduction of the Fe2O3 Composite Pellets (Dry Basis)…………………………….100
Figure 2.9 Composition of the Exhaust Gas Stream from the Fixed Bed Reactor during
the Oxidation of the Reduced Fe2O3 Composite Pellet using Steam (dry basis)………101
Figure 3.1 Schematic Flow Diagram of Iron Based Chemical Looping Processes……145
-XVI-
Figure 3.2 Equilibrium Phase Diagrams of a) Iron-Carbon-Oxygen System b) IronHydrogen-Oxygen System……………………………………………………….……..146
Figure 3.3 Gas-Solid Contacting Pattern of the Reducer Using a) a Fluidized Bed Design;
b) a Moving Bed Design…………………………………………………………….….147
Figure 3.4 Operating Curves for a Fluidized Bed Reactor………………………….…148
Figure 3.5 Operating Lines in a Countercurrent Moving Bed Reactor…………….…..149
Figure 3.6 Operating Curve of a Fluidized Bed Reducer……………………………...150
Figure 3.7 Arrangement of Multi-Stage Fluidized Bed System for the Simulation of a
Moving Bed……………………………………………………………………….……151
Figure 3.8 5-Stage RGIBBS Model for Moving Bed Simulations……………….……152
Figure 3.9 The Gas and the Solids Conversions in a Countercurrent Moving Bed Reactor
in Case 1……………………………………………………………………………..….153
Figure 3.10 Relationship between the Gas and Solids Conversions and Solid to Gas
Molar Flow Rate Ratio..……………………………………………………………..….154
Figure 3.11
Gas and Solid Conversion Profiles in a Countercurrent Moving Bed
Reactor…………………………………………………………………………….……155
Figure 3.12 Effect of Temperature on the Conversions of Syngas, CO, and H2 in a
Countercurrent Moving Bed Reactor..……………………………………………….…156
Figure 3.13 Relationship between the Fe0.877S Formation and the Syngas H2S Level in a
Countercurrent Moving Bed Reactor……………………………………………….….157
Figure 3.14 Bench Scale Demonstration Unit for SCL Process…………………….…158
Figure 3.15 Gas and Solid Conversions in the Reducer Experiment..............................159
-XVII-
Figure 3.16 Reduction of Fe2O3 Composite Particles in a Bench Unit Using Methane as
Reducing Gas…………………………………………………………………………..160
Figure 3.17 Hydrogen Production Using Reduced Fe2O3 Particles in a Bench Unit…..161
Figure 3.18 IGCC Process with CO2 Capture…………………………………………162
Figure 3.19 Conventional Gasification–Water Gas Shift Coal to Hydrogen Process.....163
Figure 3.20 The Syngas Chemical Looping Process……………………………….…..164
Figure 3.21 ASPEN Simulation Flow Sheet for: a. IGCC System Using GE- High
Efficiency Quench (HEQ) Gasifier; b. Conventional Coal to Hydrogen System Using
Shell Gasifier; c. SCL System Using Shell Gasifier……………………………….…...166
Figure 4.1 A Simplified Flow Diagram for Coal Direct Chemical Looping Process –
Configuration I…………………………………………………………………….……205
Figure 4.2 A Simplified Flow Diagram for Coal Direct Chemical Looping Process –
Configuration II………………………………………………………………………...206
Figure 4.3 Char Reaction Enhancement Schemes: a. Using Recycled Hydrogen from the
Oxidizer; b. Using Recycled CO2 from the Reducer Exhaust…………………………207
Figure 4.4 Gas-solid Contacting Pattern of the Reducer……………………………….208
Figure 4.5 ASPEN Plus® Model Setup for a. Fluidized Bed and b. Moving Bed……...209
Figure 4.6 Concentration of CO with Respect to Different Fe2O3/Carbon ratios at 900 ºC
and 30 atm for a. a Reducer with Perfect Mixing; b. a Countercurrent Moving Bed
Reactor………………………………………………………………………………….210
Figure 4.7 Effect of Temperature on Carbon Conversions in Coal at 30 atm with an Fe2O3
to Coal ratio of 8.94:1 by Weight………………………………………………………211
-XVIII-
Figure 4.8 Effect of Pressure on Coal Conversion at 850 C with an Fe2O3 to Coal Ratio
of 8.94:1 by Weight…………………………………………………...………………..212
Figure 4.9 Effect of Steam and CO2 on the Fe2O3 Conversion at 900 ºC and 30 atm with
an Fe2O3 to Coal Ratio of 8.94:1 by Weight……………………………………….…..213
Figure 4.10 Reducer Test Results Using Anthracite Coal……………………….…….214
Figure 4.11 Steam to Hydrogen Conversion for: a. Countercurrent Moving Bed Oxidizer;
b. Fluidized Bed Oxidizer. Reactor Operating Conditions: 700 ºC, 30 atm……………215
Figure 4.12 Relationship between the Fe/FeO Composition and the Steam to Hydrogen
Conversion in a Countercurrent Moving Bed Oxidizer Operated at 30 atm and 700
ºC……………………………………………………………………………………….216
Figure 4.13 Material Flow and Energy Flow in a CDCL Process……………………..217
Figure 4.14. Process Flow Diagram of the ASPEN Plus® Model for the CDCL Process
Optimized for Hydrogen Production…………………………………………………..218
Figure 5.1 Sub-Pilot Scale Demonstration Unit for SCL Process a. Schematic Flow
Diagram; b. Photograph………………………………………………………………...241
Figure 5.2 Overall Arrangement of the Sub-Pilot SCL System………………………..242
Figure 5.3 Physical Locations of the Various Sub-Systems in the Sub-Pilot Scale Unit at
the OSU West Campus Demonstration Site……………………………………………243
Figure 5.4 a. Schematic Diagram of the Gas Mixing Panel Design; b. Photograph of the
Gas Mixing Panel ………………………………………………………………………244
Figure 5.5 Design of the Reducer and Oxidizer……………………………………….245
Figure 5.6 Design of the Gas Sampling Ports………………………………………….246
Figure 5.7 Bed Height Control System………………………………………………...247
-XIX-
Figure 5.8 a. Original Design of the Rotary Disk Solid Feeder; b. Updated Design of the
Rotary Disk Solid Feeder; c. Photograph of the Assembled Solid Feeder with Servo
Motor Installed……………………………………………………………………….....248
Figure 5.9 Modifications to the Rotary Solid Feeder for Smoother Solid Flow……….249
Figure 5.10 The Schematic of the Valve System……………………………………….250
Figure 5.11 User Interface of the Control Sequence……………………………………251
Figure 5.12 Steam Generator…………………………………………………………..252
Figure 5.13 Air Compressors…………………………………………………………..253
Figure 5.14. Attrition Test Results Using Type A Particles with 5 mm Diameter and 1.5
mm Thickness…………………………………………………………………………..254
Figure 5.15. Attrition Test Results Using Type B Particles with 5 mm Diameter and 4.5
mm Thickness…………………………………………………………………………..255
Figure 6.1 Syngas Chemical Looping enhanced Coal-to-Liquids (SCL-CTL)
process………………………………………………………………………………….265
-XX-
CHAPTER 1
INTRODUCTION TO CLEAN COAL CONVERSION PROCESSES
1.1 Background
Energy and global warming are two intertwined issues of significant magnitude in
the modern era. With oil prices rising above $120/barrel and atmospheric CO2 levels
increasing at a rate greater than 1.5 ppm each year
1-3
, an urgent need exists for
development of clean and cost effective energy conversion processes.
Renewable energy sources such as hydro, wind, solar, geothermal, and biomass
will help reduce anthropogenic CO2 emissions by mitigating fossil fuel consumption.
However, with the high cost, geological constraints, and intermittency issues, renewable
energy is not likely to contribute to a significant share of the total energy demands in the
foreseeable future
4, 5
. Similarly, concerns over plant safety and radioactive waste
disposal will impede the wide utilization of nuclear power 6. Thus, despite high crude oil
and natural gas prices, fossil fuels will continue to provide more than 85% of the overall
world energy consumption for the next several decades 7. The USDOE studies indicated
that the consumption of coal as an energy resource is more responsive to crude oil price
fluctuations than renewable energy sources in the near term, and coal could regain its role
as a major energy source by 2030 7. Figure 1.1 shows the impact of oil prices on the
-1-
consumption of coal and other energy sources. The attractiveness of coal lies in its
abundant reserves and stable prices when compared to both oil and natural gas.
Without the implementation of pollution control, enhanced coal usage will result
in serious environmental impacts since coal contains various contaminants and is the
most carbon-intensive energy source. Of major global concern is the fact that the
combustion of fossil fuels releases 27 gigatons of CO2 each year 7, 8. With increasing coal
consumption, the anthropogenic CO2 emission rate may reach well over 40 gigatons per
year within the next two decades in the absence of effective CO2 mitigation techniques 7, 8.
Therefore, modern coal conversion technologies need to be able to efficiently convert
coal into useful products while controlling the CO2 emission. Unlike crude oil, which is
primarily used as transportation fuels, coal is primarily used as a stationary source for
electricity generation. Thus, CO2 capture from coal can be more readily implemented.
In order provide the readers a better picture of the challenges and the exciting
progresses in the field of clean coal conversion technologies, this chapter addresses clean
coal conversion technologies from the process viewpoint. Coal combustion processes are
first discussed along with the various options for pollutant control and CO2 capture. It is
then followed by an overview of coal gasification processes. Advanced membrane and
chemical looping based systems using gaseous feedstock as well as advanced direct coal
chemical looping systems are illustrated. These advanced technologies that yield high
energy conversion efficiencies are at various stages of development and are potentially
deployable in the near or intermediate term.
-2-
1.2 Coal Combustion Processes
Archeological evidence indicates that humans have been burning coal for at least
4,000 years
9-11
. Throughout history, coal has been used to generate heat and to smelt
metals. However, it was not until the 18th century that coal started to play an
indispensible role in the economy. As an important fuel that propelled the industrial
revolution 12, 13, coal has been widely used since the 1700s to drive steam engines, in the
operation of blast furnaces for metal production, in the production of cement, and in the
generation of town gas for lighting and cooking. Since the late 19th century, coal has been
used to power utility boilers for electricity generation
14
. Although its dominance as an
energy source was replaced by crude oil in the 1950s, coal is still the single most
important fuel for electricity generation today, accounting for 40% of the electricity
generated worldwide 7. The dominance of coal in electricity generation is expected to
continue well into the 21st century.
Presently, pulverized coal (PC) fired power plants account for more than 90% of
the electricity generated from coal 15. The schematic flow diagram of a PC power plant is
illustrated in Figure 1.2.
In a PC power plant, coal is first pulverized into fine powder with over 70% of the
particles smaller than 74 µm (200 mesh). The pulverized coal powder is then combusted
in the boiler with the presence of ~20% excess air 14. The heat of combustion is used to
-3-
generate high pressure, high temperature steam that drives the steam turbine system
based on a regenerative Rankine cycle for electricity generation. Although the underlying
concept is quite simple, the following challenges need to be addressed for modern PC
power plants: enhancement of energy conversion efficiency; effective control of
hazardous pollutants emission; and CO2 capture (and sequestration).
1.2.1 Energy Efficiency Improvement
An increase in the combustion process efficiency leads to a reduced coal
consumption and hence, a potential cost reduction for electricity generation. The first
generation coal fired power plants constructed in the early 1900s converted only 8% of
the chemical energy in coal into electricity (based on the higher heating value, HHV)16.
Since then, a significant improvement in plant efficiencies has been made.
Thermodynamic principles require higher steam pressures and temperatures for a higher
plant efficiency. The corrosion resistance of the materials for boiler tubes, however,
constrains the maximum pressure and temperature of the steam. Most of the PC power
plants currently under operation utilize sub-critical PC (Sub-CPC) boilers which produce
steam with pressures up to 22 MPa and temperatures around 550 °C. The energy
conversion efficiencies of traditional sub-critical PC power plants typically range from
33% to 37% (HHV)17. With an increase in the steam pressure, supercritical PC (SCPC)
power plants were first introduced in the early 1960s in the U.S.
16
. Supercritical power
plants involve steam with a typical pressure of ~24.3 MPa and temperatures up to 565°C,
leading to a plant efficiency of 37 to 40%17. Many supercritical power plants were
constructed in the 1960s and 70s in the U.S. However, due to the low reliability of the
-4-
boiler materials, the further application of the SCPC technology was essentially halted in
the U.S. in the early 1980s. The development of high performance super alloys coupled
with increasing environmental concerns and the rising cost of coal during the last two
decades has stimulated the revival of supercritical technology, especially in Europe and
Japan, leading to the reduction of subcritical boilers in newly installed fleets.
Recent advancements in coal combustion technologies are highlighted by the
generation of “ultra-supercritical” (USCPC) steam conditions that can provide even
higher process efficiencies. The ultra-supercritical condition refers to the “operating
steam cycle conditions above 565° C (>1050° F)” 17. The pressure and temperature of the
steam generated from existing ultra-supercritical power plants can reach 32 MPa and
610 °C, corresponding to an energy conversion efficiency of over 43%
17, 18
. The global
on-going R&D activities on PC boilers focus on the development of super alloys that can
sustain steam pressures up to 38.5 MPa and temperatures as high as 720 °C. It is expected
that a plant efficiency of over 46% can be achieved under such conditions
17-19
. Other
efforts in ultra-supercritical technology include minimizing the usage of super alloys,
improving the welding technique, and optimizing the boiler structure design to minimize
the steam line to steam turbine 18.
Besides PC boilers, Fluidized Bed Combustors (FBC) using either turbulent
fluidized beds or circulating fluidized beds are also being used for steam and power
generation world wide. In these processes, limestone is often injected to capture SOx
formed during coal combustion. Compared to PC boilers, the FBC has lower SOx and
-5-
NOx emissions
20, 21
. Furthermore, it has superior fuel flexibility
22
. Most commercial
FBC plants operate under atmospheric pressures, with energy conversion efficiencies
similar to subcritical PC power plants. Higher efficiencies can be achieved by operating
the FBC at elevated pressures
22-24
. The Pressurized Fluidized Bed Combustor (PFBC)
generates a high temperature, high pressure exhaust gas stream which drives a gas turbine
– steam turbine combined cycle system for power generation. In an advanced PFBC
configuration, fuel gas is generated from coal via particle oxidation and pyrolysis. The
fuel gas is combusted to drive a gas turbine (topping cycle). Such a process has the
potential to achieve an energy conversion efficiency of over 46% 23. To date, the PFBC
demonstrations have shown relatively low plant availability. In addition, the capital
investment for PFBC is higher than PC power plants with a similar efficiency
25
. Other
potential challenges to the PFBC technology include scale-up, high temperature
particulates/alkali/sulfur removal for gas turbine operation, and mercury removal from
the flue gas
22, 26
. Table 1.1 compares the performance of different coal combustion
technologies.
1.2.2 Flue Gas Pollutant Control Methods
Modern coal combustion power plants need to be able to capture environmentally
hazardous pollutants released from coal combustion. Such pollutants include sulfur
oxides, nitrogen oxides, fine particulates, and trace heavy metals such as mercury,
selenium, and arsenic. Methods for capturing these contaminants from the flue gas
streams abound. The challenges, however, lie in the efficient and cost effective removal
of these contaminants.
-6-
The traditional method for SOx removal utilizes wet scrubbers with alkaline
slurries. The wet scrubber is effective; however, it is costly and yields wet scrubbing
wastes that must be disposed of. Alternative methods have included more cost effective
lime spray drying and dry-sorbent duct-injection. The lime spray drying method employs
slurry alkaline spray yielding scrubbing wastes in solid form, easing the waste handling.
The dry-sorbent duct-injection employs a dry alkaline sorbent for direct in-duct injection,
circumventing the use of the scrubber. The recent pilot testing using re-engineered
limestone sorbents of high reactivity yields a sorbent sulfation efficiency of over 90%,
compared to under 70% with ordinary limestone sorbent, indicating a viability of the drysorbent duct-injection method with very active sorbents
27-29
. The NOx is commonly
removed by selective catalytic reduction (SCR). Other methods that can be employed
include low NOx burner and O3 oxidation. The recent pilot testing of the CARBONOx
process using coal char impregnated with alkaline metal revealed a high NOx removal
efficiency at low flue gas temperatures
30
. The trace heavy metals such as mercury,
selenium, and arsenic can be removed by calcium based sorbent and/or activated carbon
27, 31
.
The techniques to control the flue gas pollutants indicated above are welldeveloped. An effective capture (and sequestration) of CO2, an important green house gas
(GHG) that accounts for 64% of the enhanced green house effect
challenging task.
-7-
32
is, however, a
1.2.3 CO2 Capture Systems
Coal-fired power plants are responsible for nearly a third of all anthropogenic
CO2 emissions
33
. Therefore, cost-effective carbon capture technologies for these plants
play an important role in CO2 mitigation.
CO2 represents ~15% of the atmospheric pressure flue gas stream from coal
combustion power plant (dry basis). Low CO2 partial pressures combined with the
extremely high flue gas generation rate make the CO2 capture from PC power plants an
energy consuming step. An ideal CO2 capture technology would incorporate effective
process integration schemes while minimizing the parasitic energy requirement for CO2
separation.
The existing CO2 capture techniques from PC power plants include the wellestablished monoethanolamine (MEA) scrubbing technology. Figure 1.3a shows the
schematic diagram of the MEA scrubbing process which indicates the key stream
conditions of the process 34-36. In this process, the flue gas is first cooled down to ~40 °C
before entering the absorber where fresh amine is used to absorb CO2 in the flue gas
stream. The spent amine solution with a high CO2 concentration is then regenerated in the
stripper under a higher temperature (100 – 150 °C), and CO2 is then recovered at low
pressure (1 – 2 atm). A large amount of high temperature steam is required to strip the
CO2 in the regeneration step
36, 37
. Therefore, a significant amount of energy will be
consumed for steam generation and subsequent CO2 compression step. It is estimated that
the CO2 capture (separation and compression) using the amine scrubbing will reduce the
-8-
power generated from the entire plant by as much as 42% 38, which amounts to ~70-80%
of the total cost in the overall three-fold carbon management steps, i.e. carbon capture,
transportation, and sequestration
39, 40
. As a result, a process that can reduce the energy
consumption in the CO2 capture step will be vital for CO2 management in coal fired
power plants.
The chilled ammonia process, illustrated in Figure 1.3b, is another solvent based
CO2 capture technology where ammonia carbonates and bicarbonate slurries are used to
capture the CO2 in the flue gas stream at 0 – 10 °C and atmospheric pressure. The CO2
rich solvent is then regenerated at 110 – 125 °C and 20 – 40 atm. The capability to
regenerate CO2 at elevated pressures reduces the energy consumption for CO2
compression. Based on the studies by Electric Power Research Institute (EPRI) and
ALSTOM, the overall energy penalty for CO2 capture is estimated to be lower than 16%
when the chilled ammonia process is used
41, 42
. A 5 MWth (mega watts thermo)
equivalent chilled ammonia process demonstration plant, jointly supported by ALSTOM
and EPRI, is currently under construction at We Energies’ Pleasant Prairie Power Plant in
Wisconsin 43. American Electric Power (AEP) is also planning to demonstrate the chilled
ammonia process at the 20 MWe (mega watts electricity) scale, starting in 2009, before
building a 200 MWe commercial level chilled ammonia retrofit system in 2012 44.
Similar to solvent based CO2 scrubbing techniques, high temperature sorbents
such as limestone, potassium carbonates, lithium silicates, and sodium carbonates can be
used to capture CO2 in the flue gas at elevated temperatures
-9-
45, 46
. With better heat
integration, these strategies can potentially decrease the energy consumption in the CO2
separation step. One scheme for heat integration is based on the calcium based
carbonation-calcination reaction (CCR) process which uses hydrated lime, and natural or
re-engineered limestone sorbents at 600-700oC for CO2 separation
47
. Figure 1.4
delineates the heat integration strategies for retrofitting the CCR process to an existing
PC power plant.
In the CCR process, both CO2 and SO2 in the flue gas are captured by the CaO
sorbent in the carbonator operated at ~ 650 oC, forming CaCO3 and CaSO3/CaSO4 The
carbonated sorbent, CaCO3, is then regenerated to calcium oxide (CaO) sorbent in the
calciner at 850 - 900oC, yielding a pure CO2 stream. The sulfated sorbent and fly ashes
are removed from the system by means of a purge stream. Due to an optimized energy
management scheme, the CCR process consumes 15 – 22% of the energy generated in
the plant 48, 49. The process is being demonstrated in a 120 KWth (kilo watts thermo) pilot
plant located at The Ohio State University (OSU). A similar process is being
demonstrated at CANMET Energy Technology Center in Canada 50.
In addition to the absorption/adsorption based technologies, oxy-fuel combustion
technology provides another means for carbon management in coal fired power plants. In
this technology, pure oxygen instead of air is used for coal combustion. As a result, a
concentrated CO2 stream is generated, avoiding the need for CO2 separation. However,
the energy-consuming cryogenic air separation step will reduce the overall plant
efficiency by 20 – 35%
40, 51, 52
. This process has been successfully demonstrated by the
-10-
Babcock & Wilcox Company on a 1.5 MWth pilot scale PC unit. Demonstration on a 30
MWth unit is currently under way. The on-going pilot scale studies on oxy-fuel
combustion include those carried out by ALSTOM, Foster-Wheeler, CANMET Energy
Technology Center, Vattenfall, and Ishikawajima-Harima Heavy Industries (IHI) 53.
To generalize, a number of retrofit systems under different stages of development
can be used to capture CO2 from existing power plants. Since PC power plants will
continue to provide a significant portion of the electricity needs well into the 21st century
7
, these CO2 capture systems are essential to mitigate the environmental impact from coal
burning. In general, however, CO2 capture and compression from a coal combustion flue
gas is costly and energy intensive. A more promising approach to reduce the overall
carbon footprint of a coal based plant is to adopt coal conversion processes that are
intrinsically advantageous from a carbon management and energy conversion standpoint.
Among the various options, coal gasification described below offers such attraction.
1.3 Coal Gasification Processes
For years, the commercial efforts on clean coal processes have been centered on
coal combustion for power generation. However, new process developments with a focus
on higher energy conversion efficiencies for electricity generation as well as variability in
product formation have generated considerable interest. Coal gasification schemes can
provide a variety of products – e.g. hydrogen, liquid fuels and chemicals – besides
electricity. Further, gasification is a preferred scheme from a pollutant and carbon
management viewpoint.
-11-
1.3.1 Overview
Compared to combustion, coal gasification is relatively new. Commercial
gasification processes date back to the late 18th century when coal was converted into
town gas for lighting and cooking. Since the 1920s, the gasification process has been
used to produce chemicals and fuels
54
. Unlike traditional combustion processes which
fully oxidize carbonaceous fuels to generate heat, modern coal gasifiers convert coal into
syngas via partial oxidation reactions with oxygen or with steam and oxygen under
elevated pressures 14, 54. The high pressure syngas stream, undiluted by N2 in the air, has a
much lower volumetric flow rate when compared to that of the flue gas from coal-fired
power plants. As a result, the partial pressure of the contaminants is significantly
increased. For instance, the volumetric flow rate of syngas generated from a dry feed,
oxygen blown gasifier can be two orders of magnitudes lower than that from a PC boiler
with similar coal processing capacity (dry basis). Meanwhile, the partial pressure of CO2
in the syngas after the water gas shift (WGS) reaction can be 80 times higher than that in
the PC boiler flue gas (dry basis). The significantly reduced gas flow rate and increased
gas partial pressures make the pollutant and CO2 control an easier task for gasification
processes when compared to coal combustion processes.
Figure 1.5 shows the modern coal gasification process that generates a variety of
products. In the coal gasification process, coal first reacts with oxygen (and steam) to
produce raw syngas. The raw syngas, with pollutants such as particulates, H2S, COS, HCl,
ammonia, and mercury, is purified before it is sent to a gas turbine-steam turbine
-12-
combined cycle system for electricity generation. This syngas route is known as the
Integrated Gasification Combined Cycle (IGCC). The electricity generation efficiency of
the IGCC process can be higher than 45% without CO2 capture
54, 55
. In a carbon
constrained scenario, however, the CO in the syngas stream will be further converted to
H2 through the water-gas shift (WGS) reaction:
CO + H2O → CO2 + H2
(1.1)
Thus, the resulting gas stream contains a high CO2 concentration (up to ~40% by volume
on the dry basis). The CO2 (and H2S) can be captured using either chemical absorption
based
acid
gas
removal
processes
such
as
monoethanolamine
(MEA)
or
methyldiethanolamine (MDEA) described in Section 1.2.3 or physical absorption based
processes such as Selexol and Rectisol, yielding concentrated H2 56. The H2 can be used
to generate electricity through a combined cycle system with minimal carbon emissions.
Alternatively, the H2 stream can be further purified using pressure swing
adsorption (PSA) units. The resulting high-purity H2 can be used for fuel cell applications.
Besides electricity and H2 generation, syngas can also be converted to chemicals and
liquid fuels such as diesel and naphtha through the Fischer-Tropsch (F-T) reactions,
which can be represented by 57-59:
(2n+1)H2 + nCO → CnH2n+2 + nH2O
(1.2)
Despite the advantages in product versatility and pollutant controllability
compared to combustion, gasification is more capital intensive. A study conducted in
-13-
2001 indicated that an IGCC system required 6-10% more capital investment when
60
compared to an ultra-supercritical PC plant
. Both plants have similar energy
conversion efficiencies. Although the CO2 capture from the gasification process is easier
compared to the PC plant, the CO2 capture, nevertheless, represents an energy and capital
intensive step of the process. The CO2 capture can derate the energy conversion
efficiency of the IGCC system by 13 – 24%, increasing the cost of electricity by 25% 45%
61-65
. Other issues related to gasification include large steam consumption in the
WGS step due to the need for the excessive steam as well as the temperature and pressure
swing requirement in the process for sulfur and mercury removal.
Gasification, like other technologies, has undergone evolution since its inception.
Over the years, different types of gasifiers have been developed which provide a higher
carbon conversion, cold gas and thermal efficiencies, and flexibility in the type of fuel
used. These gasifier types include the fixed/moving bed gasifier, fluidized-bed gasifier,
entrained-flow gasifier, and transport gasifier
54
. Most of the modern gasifiers adopt an
entrained-flow design due to better fuel flexibility, carbon conversions, and syngas
quality
66
. Other ongoing research activities include the use of an Ion Transport
Membrane (ITM) instead of the cryogenic separation technique to reduce the energy
consumption of the air separation unit (ASU)
67, 68
, the increase in the gas turbine inlet
temperature to increase the combined cycle efficiency, and the development of a warm
and hot gas clean up system to efficiently remove pollutants such as particulates, sulfur
and mercury 69-71.
-14-
As there are a large degree of the operational variations in individual units and in
an integrated process system, optimization of gasification process requires elaborate
consideration of all the viable process configurations. For this purpose, simulation
software such as ASPEN Plus® is often used to aid in the analysis of the process
configurations under various process variables. In the following section, a case study is
presented which illustrates the energy conversion efficiency for an IGCC system with
CO2 capture through simulation using the ASPEN® plus software.
1.3.2 ASPEN Analysis on IGCC System with CO2 Capture – A Case Study
Aspen Plus® has been widely used to simulate energy conversion systems [41, 7580]. Based on appropriate assumptions and relevant experimental data of the individual
units, the ASPEN Plus® software can assist in the evaluation of the process performance,
and in the optimization of the process configuration. The IGCC system illustrated in this
case study uses a GE/Texaco slurry-feed, entrained flow gasifier with total water quench
syngas cooler. The flow diagram of the process is shown in Figure 1.6.
In this process, coal is first pulverized and mixed with water to form coal slurry.
The coal slurry is then pressurized and introduced to the gasifier to be partially oxidized
at 1500 °C and 30 atm. The high temperature raw syngas after gasification is then
quenched to 250 °C with water. The quenching step solidifies the ash. Moreover, most of
the NH3 and HCl in the syngas are removed during this step. After quenching, the syngas
is sent to a Venturi scrubber for further particulate removal. The particulate-free syngas,
saturated with steam, is then introduced to the sour WGS unit. The syngas exiting the
-15-
WGS unit contains mainly of H2 and CO2 with small amount of CO, H2S, and mercury.
This gas stream is then cooled down to 40 °C and passed through an activated carbon bed
for mercury removal. The CO2 and H2S in the syngas are then removed using an MDEA
unit, resulting in a concentrated hydrogen stream with small amounts of CO2 and CO.
The hydrogen rich gas stream is then compressed, preheated, and combusted in a
combined cycle system for power generation. The combined cycle system consists of a
gas turbine with an inlet firing temperature of 1430 °C and a two stage steam turbine
working at 550 °C and 35 atm. The CO2 obtained from the MDEA unit is compressed to
150 atm for sequestration.
ASPEN modeling on coal conversion systems has been extensively discussed in
various literatures [41, 75, 77-80]. The following section briefly recapitulates the key
steps to set up an ASPEN simulation model on the IGCC system described above.
Prior to the simulation, a representative process flow sheet that contains all the major
units is developed (Figure 1.6). The appropriate assumptions for the simulation are then
determined. The key assumptions are listed as follows:
ƒ
132.9 tonne/hr of Illinois #6 coal is fed into the system (1000 MW in HHV)
ƒ
Energy consumed for units such as acid gas removal are simulated based on
performance data of the commercial units
ƒ
The GE slurry feed gasifier has a carbon conversion of 99%, heat loss in the
gasifier is 0.6% of the HHV of coal
-16-
ƒ
A GE 7H gas turbine combined cycle system is used, all the exhaust gas is cooled
down to 130 °C before exiting the Heat Recovery Steam Generator (HRSG)
ƒ
At least 90% of the CO2 generated needs to be captured and compressed to 150
atm for sequestration
ƒ
The mechanical efficiency of pressure changers is 1, whereas the isentropic
efficiency is 0.8~0.9
In order to accurately simulate the individual unit in the flow sheet, appropriate ASPEN
Plus model(s) for each unit is determined. These models are listed in Table 1.2.
Aspen Plus® has a comprehensive physical property database. Therefore, most of
the chemical species involved in the process can be selected directly from the build-in
database. The nonconventional components such as coal and ash can be specified
conveniently using general coal enthalpy modulus embedded in ASPEN software. After
the chemical species in the process are defined, the related physical property methods for
are selected according to the simulator’s category. In this simulation, the global property
method is PR-BM, whereas local property methods are specified whenever necessary.
The ASPEN model is finalized by establishing detailed operating parameters
based on the operating conditions and design specifications of the individual unit. The
units are then connected in the same arrangement as shown in the flow sheet. An
appropriate convergence setting is determined to ensure accurate simulation results.
Table 1.3 generalizes the simulation results of the IGCC system described above.
-17-
The results shown in Table 3.2 can replicate the performance of existing IGCC
power plants reported by Higman
54
. The ASPEN simulation can be effective for
evaluating the performance of various coal conversion systems based on a common set of
assumptions.
1.4 Advanced Coal Conversion Processes
Although with various improvements discussed in Section 1.3, the efficiency of
the conventional gasification systems is still limited due to the elaborate steps such as
syngas cleaning and conversion, and gas separation and compression. The advanced coal
conversion processes, which adopt novel process intensification strategies, streamline the
conversion processes, thereby yielding high energy conversion efficiency. Such
techniques, which are currently at various stages of demonstrations, encompass the
membrane based approach and chemical looping based approach. Both approaches can
process syngas derived from coal or any other carbonaceous feedstock. The chemical
looping approach can also process coal or other carbonaceous feedstock directly. These
approaches are elaborated below.
1.4.1 Membrane Based Gasification Systems
A membrane is a selective barrier between two phases. The molecules or small
particles can transport from one phase to the other through the membrane. A H2 or CO2
selective membrane can be utilized in gasification processes to reduce the energy penalty
for CO2 capture and to enhance the hydrogen/electricity generation.
-18-
The selective nature of a membrane can be attributed to one or more of the
following mechanisms: a. Knudson diffusion; b. surface diffusion; c. capillary
condensation; d. molecular sieving; e. solution diffusion; f. facilitate transport 72. As the
smallest diatomic molecule, hydrogen can be separated from other gaseous species
involved in the coal gasification process based on all the mechanisms stated above. On
the other hand, most CO2 selective membranes are based on either solution diffusion or
facilitate transport mechanism since the CO2 molecule is significantly larger. An amine
based carrier is often used to facilitate the transportation of CO2 from the retentate side to
the permeate side
72, 73
. As a result, a hydrogen selective membrane can be made of
metallic, inorganic (ceramic), porous carbon, polymer, or hybrid materials while most of
the CO2 selective membranes for separating CO2 from hydrogen are polymeric.
The desirable features of a membrane include good permeability, selectivity,
reliability, and tolerance to contaminants. For commercial applications in gasification
processes, it should also be affordable, thermally stable, and durable. Of all the H2selective membranes, metallic membranes and ceramic membranes are the most
extensively studied 74-79.
The metallic H2-selective membranes generally have a very high selectivity and
thermal stabilities. The potential candidates include palladium, platinum, tantalum,
niobium, and vanadium
72, 74, 77
. Among these metals, Pd-based membranes, although
relatively costly, have demonstrated the highest selectivity and good permeability and
-19-
thermal stability. However, the presence of hydrogen at below 300 °C can cause the
embrittlement of the Pd-based membrane due to the Pd-H phase transition. In order to
reduce the membrane degradation as well as to reduce the cost, Pd-based membranes are
often alloyed with Ag, Au, Y, Cu, or Se. These alloys are processed into a layer as thin as
blow 1 µm and then doped on top of a porous ceramic or metallic support
78, 80
. By
alloying and supporting, the usage of Pd is minimized with increased physical strength of
the membrane
78
. One major challenge to Pd-based membranes is that the presence of
sulfur compounds such as H2S and COS under elevated temperatures can poison the Pdbased membranes. Recent studies indicate that alloying can increase the sulfur tolerance
of the membrane81. However, a high sulfur content that is close to or beyond the
thermodynamic limit for the formation of stable sulfides will nevertheless deactivate the
membrane76. In addition, when ceramic support is used in the Pd-based metallic
membrane, it will need to resolve such issues as the mechanical strength of the support
and the large difference in thermal expansion coefficients between the metallic
membrane and the ceramic support. For metallic support, the challenge lies in the
stability of the crystal structure due to inter-metallic diffusion. Therefore, desirable
improvements in the Pd-based membrane for gasification applications include further
reduction in cost coupled with increased durability, sulfur tolerance, and H2 flux. Besides
the Pd-Based metallic membranes, non Pd-based alloys 82 and amorphous metals 77, 83 are
also under investigation with the prospect of developing less costly metallic membranes
with satisfactory performance.
-20-
Ceramic H2-selective membranes such as porous silica- and zeolite-based
membranes represent another category of promising hydrogen separation materials
84-86
.
Both membranes are micro-porous inorganic membranes comprised of a membrane layer,
an intermediate layer, and a support. These membranes have several advantages when
compared to metallic membranes including low cost, ease of fabrication, and less
susceptibility to H2 embrittlement. Moreover, very high hydrogen permeability can be
achieved using an ultra-thin amorphous silica membrane. However, improvements that
need to be made in these membranes include selectivity, defect reduction,
thermochemical stability and operational stability. Table 1.4 generalizes the
performances of existing H2-Selective membranes as compared to the 2010 performance
target set by the USDOE.
Although zeolite-based membranes can be used to selectively remove CO2 from
other gases such as N2 and CH4 based on adsorption preference 87, 88, very limited studies
have been performed on the separation of CO2 from H2 using such membranes51. Other
attempts include those performed by Air Products and Chemical Inc. that use nanoporous
carbon-based membranes to separate CO2 from the tail gas of the Pressure Swing
Adsorption (PSA) unit
89, 90
. However, these membranes have relatively low CO2
selectivity over H2 89-91. To date, most CO2-selective membranes for separating CO2 from
H2 are polymeric membranes based on either solution diffusion mechanism or facilitate
transport mechanism
73, 92-95
. The challenges to the polymeric CO2-selective membranes
include limited operating temperature and relatively low CO2/H2 selectivity and flux.
-21-
As mentioned in Section 1.3, the WGS reactor(s) and the CO2 separation units
consume a significant amount of parasitic energy for the coal to hydrogen process and
IGCC process with CO2 capture. The applications of the H2- or CO2-selective membranes
in coal gasification systems for the intensification of the CO shift and hydrogen
purification steps have been extensively studied during the last decade.
Several different configurations using different types of membranes have been
investigated, exhibiting promising results. Figure 1.7 shows a multi-stage membrane
system that recovers CO2 from a shifted syngas stream proposed by Kaldis et al 96. In this
process, the clean syngas stream resulting from the coal gasifier and gas cleanup units is
first shifted in a series of WGS units, resulting in a gaseous mixture consisting mainly of
H2, CO2, CO, and N2. The mixed gas is then introduced to a series of H2-selective
membranes to recover a concentrated CO2 stream on the retentate side. The permeate side,
with concentrated hydrogen, is combusted in the gas turbine for electricity generation.
The performances of both polymer and ceramic membranes are investigated using
ASPEN Plus® simulations. The results indicated that the CO2 emission can be reduced by
over 50% using the multi-stage membrane system but with 17 – 28% parasitic energy
consumptions 96.
More advanced membrane systems integrate the function of both WGS and CO2
separation using either H2- or CO2-selective membranes. Such configurations are shown
in Figures 1.8a and 1.8b
76, 97-103
. Figure 1.8a illustrates a specific configuration when a
H2-selective membrane is used 76, 97, 99-102. In such a configuration, a conventional gasifier
-22-
and a gas clean up system is used to produce clean syngas. The clean syngas is then sent
to the membrane-WGS reactor. The membrane-WGS reactor has two compartments, i.e.
reaction side and product side. The two compartments are segregated by a semipermeable membrane that is selective to hydrogen. In the reaction side, the CO in the
syngas is converted to H2 and CO2 via WGS reaction. The H2 produced in the reaction
side is continuously permeated through the membrane to the product side. As a result, a
high purity H2 product can be obtained without engaging traditional separation
techniques. Such hydrogen can either be used as a product or combusted with air for
power generation. In addition, due to the removal of the hydrogen product, the WGS
reaction, which is limited by thermodynamic equilibrium, can be enhanced. The tail gas
from the reaction side, with a high CO2 concentration mixed with residual CO and H2, is
combusted in a combined cycle system with O2 to generate electricity. The resulting CO2
is then sequestered.
The underlying principle for the membrane-based system shown in Figure 1.8b 98,
101
is similar to that of the system shown in Figure 1.8a. The only difference lies in the
type of membrane used for separating the shifted syngas. In this configuration, a CO2selective membrane is used to divide the reaction side and the product side in the
membrane reactor. As a result, the CO2 rather than the H2 will be transferred from the
reaction side to the product side. The simultaneous removal of CO2, which is another
product of the WGS reaction, can also enhance the reaction. The CO2 stream in the
product side, swept by steam, can be directly sequestered while the H2-rich stream in the
-23-
reaction side can either be purified to obtain a hydrogen product or combusted with air
for power generation.
Extensive studies have been performed to analyze the performance of gasification
processes integrated with membrane systems. Chiesa et al. (2007)
102
indicated that
although a significant energy penalty has to be paid for CO2 capture, a Pd-based
membrane system such as that shown in Figure 1.8a is thermodynamically advantageous
when compared to commercial WGS-CO2 capture systems. A process analysis carried out
by Amelio et al. (2007)
97
indicated that if integrated with an IGCC system using a GE
gasifier, an energy penalty around 17.5% (46.0% before capture to 39.3% HHV after
capture) will incur when a Pd-based H2-selective membrane system is used to capture
90% of the CO2. Grainger et al. (2008)
98
studied the performance of a CO2-selective
polyvinylamine membrane in an IGCC system identical to the Puertollano plant. The
results revealed a 22.9% energy penalty for 85% CO2 capture. Carbo et al.
101
compared
the performance of a H2-selective membrane system to that of a CO2-selective membrane
in an IGCC process with an entrained flow, oxygen blown gasifier. The results indicated
that the energy penalty is merely 11.2% when a H2-selective membrane is used for 100%
CO2 capture. In contrast, a 19.4% energy penalty will incur when a CO2-selective
membrane is used for 90% CO2 capture. The selectivity of both membranes was assumed
to be infinity in this study.
A more advanced approach integrates a H2-selective membrane into the gasifier
for H2 generation (Figure 1.9) 104. In this case, a membrane is installed in the coal gasifier
-24-
to separate out the hydrogen generated. The rest of the syngas is combusted with oxygen
for power generation. Such a process, although potentially more efficient, requires a
membrane that tolerates ultrahigh temperatures and various contaminants. The
development of such high performance membranes may not be feasible in the near future.
To generalize, although the membrane systems can not eliminate the energy
penalty for CO2 capture in gasification plants, they have the potential to reduce such a
penalty when compared to the traditional approach. The parasitic energy consumed for
CO2 capture using a membrane-based system lies in the need for gas compression, and in
some cases, the generation of extra oxygen to combust the CO2 rich tail gas and the need
for extra steam as sweep gas. It is also worth noting that from the economic standpoint,
the membrane-WGS reactor can replace both the shift unit and CO2 separation unit.
Therefore, notable cost reduction can be realized provided that a membrane with good
reliability and durability can be mass-produced at a reasonable cost.
1.4.2 Chemical Looping Based Gasification Systems
As discussed in Section 1.4.1, membrane-based systems intensify the syngas
conversion scheme by integrating the CO shift and the CO2 removal step. Due to the
limited tolerance of membranes towards pollutants such as sulfur and halogen
compounds, the raw syngas from the gasifier needs to be extensively cleaned before
entering the membrane system. Chemical looping based systems have the potential to
simplify the syngas cleaning procedures. Moreover, the pressure drop due to the
membrane separation can be reduced in chemical looping systems.
-25-
The chemical looping strategy that generates the end products with the aid of
chemical intermediates through a series of reaction schemes was proposed many years
ago. One example is the steam-iron process used for commercial hydrogen production
from coal derived producer gas in the early 20th century
105, 106
. Another example is the
CO2 generation, reported a half-century ago, for the beverage industry using the chemical
looping process with the oxides of copper or iron as the looping particles 107, 108. Although
the adoption of the chemical looping strategy in the early years was mainly prompted by
the lack of effective chemical conversion/separation techniques in the product generation,
modern applications of chemical looping processes are prompted by the need of
developing an optimized reaction scheme that minimizes the exergy loss involved in the
chemical/energy conversion system
109-111
. Also driven by the envisaged CO2 emission
control, the recent development in chemical looping systems have focused on the
efficient conversion of gaseous carbonaceous fuels such as natural gas and coal derived
syngas
48, 111-114
, and solid fuels such as petroleum coke and coal
115, 116
while separating
CO2 readily through the looping reaction scheme. In this section, chemical looping
systems using coal derived syngas will be discussed. Looping systems that directly
convert coal will be presented in Section 1.4.3.
In this section, two types of chemical looping based approaches that enhance the
performance of the coal gasification processes are given. Type A chemical looping such
as the Syngas Chemical Looping Combustion (Syngas-CLC) processes and the Syngas
Chemical Looping (SCL) process use oxygen carrier particles, typically metal oxides, to
-26-
convert coal derived syngas, whereas Type B chemical looping such as the Calcium
Looping Process (CLP) and the Thermal Swing Sorption Enhanced Reaction (TSSER)
process utilize solid CO2 sorbents to enhance the syngas conversions.
1.4.2.1 Type A Chemical Looping
Based on the type of the end product, the Type A chemical looping processes can
be divided into two sub-categories, i.e., chemical looping combustion117-120 where the
chemical intermediate is first reduced and then combusted with air to generate heat, and
chemical looping gasification114, 121-123 where fuel gas such as hydrogen is produced.
Syngas Chemical Looping Combustion
Figure 1.10 shows a typical chemical looping combustion process using coal
derived syngas as feedstock. As can be seen, coal is first gasified into raw syngas. A set
of gas cleanup units is then used to remove the contaminants to a level below the
tolerance limit of the oxygen carrier particle used in the process. The cleaned syngas then
reacts with the oxygen carrier particles in the first reactor which is noted as the reducer or
the fuel reactor. The main reactions in this reactor are:
MeO + H2 Æ Me + H2O
(1.3)
MeO + CO Æ Me + CO2
(1.4)
As can be seen from reaction 1.3 and 1.4, the syngas is oxidized to CO2 and steam
by the metal oxide particles before exiting the reducer. A concentrated CO2 stream can
-27-
then be readily obtained by condensing out the steam in the reducer. The CO2 stream can
be further pressurized and transported for sequestration. Meanwhile, the reduced metal
oxide particles will be introduced to the second reactor, i.e. the combustor or the air
reactor, to react with air:
Me + O2 (Air) Æ MeO
(1.5)
The oxidization reaction in the combustor is highly exothermic. As a result, a high
temperature, high pressure, oxygen depleted exhaust gas stream is generated from the
combustor. Such an exhaust gas stream is used to drive a combined cycle system for
electricity generation. Meanwhile, the particles, fully regenerated by air, are recycled to
the reducer for another redox (reduction – oxidation) cycle.
In the CLC process, the coal derived syngas is combusted with air indirectly
through the looping particles, i.e., metal oxide. Hence, the fuel combustion products, i.e.,
CO2 and steam are not diluted by nitrogen in the air, and the CO2 separation from
nitrogen is, therefore, avoided. Moreover, the syngas cleanup steps can potentially be
simplified since the metal oxide particles can be more robust towards contaminants when
compared to membranes 124. As a result, the acceptable level of the contamination in the
syngas for chemical looping processes can be higher than the membrane based systems.
An additional advantage for the looping system is that the difference between the
pressure of the concentrated CO2 exhaust and that of the syngas feedstock is merely the
pressure drop of the reducer, which can be significantly lower than the pressure drop in
the solvent-based and membrane-based CO2 separation system.
-28-
The focal areas of the research and development activities on the CLC processes
are on the oxygen carrier particle design and synthesis, looping reactor design and
operation, and looping process analysis and demonstration. Various types of oxygen
carrier particles, including the oxides of Ni, Fe, Mn, Cu, and Co, have been investigated
for syngas chemical looping combustion
112, 125-127
. Most of the studies focus on
developing particles that maintain good reactivity for multiple redox (reductionoxidation) cycles. Other factors being considered include particle strength improvements
and carbon deposition reduction. In order to obtain particles with the desirable properties,
ceramic materials are often used to support the oxygen carrier. These supporting
materials include Alumina, MgAl2O4, Yttria-Stabilized Zirconia (YSZ), TiO2, Bentonite,
and barium-hexaaluminate (BHA). Metal oxide particles that can sustain multiple redox
cycles in atmospheric reactor systems have been successfully synthesized. Important
areas that need to be further explored include the pollutant tolerance of the particles and
particle reactivity under elevated pressures. For instance, experiments in a high pressure
TGA indicated that an increase in total pressure may have negative effects on the
reduction rates of Cu, Ni, and Fe based oxygen carriers
128
. This finding, however, was
inconsistent with that obtained by Siriwardane et al. (2007) using NiO supported on
Bentonite
126
. Jin and Ishida (2004) studied the pressure effect on the reactivity of NiO
supported on MgAl2O4 under 1- 9 atmospheres using a fixed bed reactor 112. They found
that an increased carbon deposition under elevated pressures, which was consistent with
that predicted from thermodynamic principles
124
. An increased oxidation reaction rate
was also observed under higher pressure by Jin and Ishida (2004)
pressure effect on the reduction reaction rate was not reported.
-29-
112
; however, the
The syngas CLC process was tested in a 300 Wth (watts thermo) circulating
fluidized bed chemical looping combustor at Chalmers University in Sweden
129-131
Different types of oxygen carrier particles including NiO supported on MgAl2O4
Fe2O3 supported on Al2O3132, and Mn3O4 supported on Mg stabilized ZrO2
133, 134
.
129
,
have
been used, yielding 99% or higher syngas conversions. Other CLC testing facilities
include the 10 kWth circulating fluidized bed unit at Chalmers University 113, the 50 kWth
circulating fluidized bed unit at Korea Institute of Energy Research (KIER)
135
120 kWth circulating fluidized bed unit at Vienna University of Technology
, and the
136
. The
published experimental results obtained from these testing facilities focus on the
conversion of methane.
Both thermodynamic and ASPEN® plus simulations have been performed for the
chemical looping combustion systems with syngas as feedstock. The exergy analysis
conducted by Anheden and Syedberg (1998) indicated that when a CLC system with a
Fe2O3-based oxygen carrier particle is used to a retrofit IGCC plant, a 7.8% increase in
exegetic efficiency compared to a base case can be realized (from 45.19 to 48.72%)
109
.
In their study, however, the energy for CO2 compression was not considered. The
ASPEN® simulation conducted by Xiang et al. (2008) indicated that the gasificationCLC system has the potential to achieve 43.2% (LHV) efficiency for electricity
generation with 99% CO2 captured 117.
-30-
The performance of the syngas chemical looping combustion processes is
dependent on two closely related factors, i.e. the oxygen carrier particle performance and
the reactor design. Many research efforts on the CLC system have focused on the
development of reactive and recyclable particles, given that fluidized bed reactor are to
be used as the looping reactors. In fact, various factors need to be considered in selecting
a particle, i.e., particle oxygen carrying capacity, reactivity, recyclability, cost, physical
strength, oxygen carrying capacity, contaminant tolerance, melting points, and
environmental effects. On the looping reactor, the use of fluidized bed reactor is
evidenced by extensive on-going studies of high density circulating fluidized bed systems
in which the riser serves as the combustor and the downer in bubbling or turbulent mode
of operation serves as the reducer in chemical looping combustion applications
137
39, 129, 134,
. It should be noted, however, that reactor design can have a significant effect on
particle conversion, and hence the process efficiency. Table 1.5 illustrates the effect of
the flow pattern, i.e., fluidized bed or countercurrent moving bed in the fuel reactor on
the solid particle conversion when a Fe based oxygen carrier particle is used. The results
given in the table are based on the thermodynamic analysis and the assumptions
presented in the table.
It is seen that the theoretical solid conversion in the moving bed is nearly five
times higher than that in the fluidized bed, resulting in significantly reduced solid
circulation rate for the moving bed design and hence minimized reactor volume. Thus,
for a successful CLC system operation, flow pattern consideration for the reactor is
deemed important.
-31-
Syngas Chemical Looping Gasification
Compared to the CLC processes, the Syngas Chemical Looping (SCL) process
has the flexibility to co-produce hydrogen and electricity
114, 121-123
. Figure 1.11 shows a
simplified block diagram of the SCL process developed at the Ohio State University.
The SCL process can convert syngas with moderate levels of HCl, NH3, sulfur,
and mercury; therefore, existing hot gas cleanup units (HGCU) will be adequate for raw
syngas cleaning. The raw syngas exiting the HGCU will be introduced to the reducer,
which is a moving bed of specially tailored iron oxide composite particles operated under
a pressure similar to that of the syngas. In this reactor, the syngas is completely converted
into carbon dioxide and water while the iron oxide composite particles are reduced to a
mixture of Fe and FeO under 750 - 900 °C:
Fe2O3 +CO → 2FeO + CO2
(1.6)
FeO + CO → Fe + CO2
(1.7)
Fe2O3 + H2 → 2FeO + H2O
(1.8)
FeO + H2 → Fe +H2O
(1.9)
Similar to the CLC processes, an exhaust stream with concentrated CO2 can be
obtained from the reducer. The contaminants in the syngas will also exit the reducer with
the CO2 stream without attaching to the particle. These contaminants can be compressed
and sequestered along with CO2 if allowed by regulation. As a result, the gas cleaning
procedures are greatly simplified.
-32-
The Fe/FeO particles leaving the reducer are then introduced into the oxidizer
which is operated at 500 – 750 °C. In the oxidizer, the reduced particles react with steam
to produce a gas stream that contains solely H2 and unconverted steam. The steam can be
easily condensed out to obtain a high purity H2 stream. The reactions involved in the
oxidizer include:
Fe + H2O (g) → FeO + H2
(1.10)
3FeO + H2O (g) → Fe3O4 + H2
(1.11)
The steam used in the oxidizer is produced from the heat released from syngas cooling
and reducer/oxidizer exhaust gas cooling. In the SCL process, the oxidizer is slightly
exothermic while the reducer can either be slightly exothermic or slightly endothermic
depending on the syngas composition. Therefore, both reducer and oxidizer are operated
under the adiabatic conditions. Heat is provided to or removed from the reactors by the
oxygen carrier particles and the exhaust gas. The Fe3O4 formed in the reducer reactor is
regenerated to Fe2O3 in an entrained flow combustor which also transports solid particles
discharged from the oxidizer to the reducer. A portion of the heat produced from the
oxidation of Fe3O4 to Fe2O3 can be transferred to the reducer through the particles:
4 Fe3O4 + O2 → 6 Fe2O3
(12)
The high pressure, high temperature, spent air produced from the combustor can be used
to drive a gas turbine - steam turbine combined cycle system to generate electricity for
parasitic energy consumptions. In yet another configuration, a fraction or all of the
reduced particles from the reducer can bypass the oxidizer and be introduced directly to
-33-
the combustor if more heat or electricity is desired. Hence, both chemical-looping
reforming and chemical-looping combustion concepts are applied in the SCL system,
rendering it a versatile technology for H2 and electricity co-production.
The SCL process has been tested at Ohio State University (OSU) in a 2.5 kWth
bench scale moving bed unit for a combined operating time of > 100 hours
138
. Current
testing results indicate > 99.9% syngas conversion in the reducer and > 99.95% purity
hydrogen stream from the oxidizer. Nearly full conversion of gaseous hydrocarbons such
as CH4 was also obtained. A 25 kWth SCL demonstration unit is being constructed at
OSU. The process analysis based on the bench scale testing results indicated that the
overall efficiency for the SCL process can exceed 64% (HHV) with 100% CO2. For
comparison, the efficiency of a traditional coal-to-hydrogen process with 90% CO2
capture is estimated to be 57% (HHV) 139.
Besides serving as a stand alone hydrogen/electricity producer, the SCL process
can be integrated into other processes to improve the overall energy conversion scheme.
Figure 1.12 exemplifies the integration of the SCL process to the sate-of-the-art Coal-toLiquids (CTL) process
140
. In this configuration, the SCL system converts the C1 – C4
products from the Fischer-Tropsch (FT) reactor into H2 and recycles it to the F-T reactor
as feedstock, resulting in a 10% increase in the liquid fuel yield and a 19% reduction in
CO2 emissions 141.
-34-
Oxygen carriers other than iron oxide such as NiO were also explored for
hydrogen generation from syngas. The experiments carried out in a 20 mm I.D. fixed bed
reactor, however, indicated that Fe is a more favorable choice than Ni 142,143. Svodoba et
al. (2007,2008)
144, 145
also examined, using thermodynamic principles, the feasibility of
using Fe, Mn, Ni, Cr, and Co based particles for hydrogen production. They concluded
that Fe – Fe3O4 is more suitable for chemical looping gasification compared to other
particles; however, they further stated that Fe3O4 is more difficult to reduce based on a
fluidized bed design. Xiang et al. (2007) performed ASPEN Plus® simulation on an ironbased looping system for hydrogen generation
146
. In their system, reduced iron oxide is
only regenerated to Fe3O4 rather than Fe2O3. As a result, a significant amount of syngas
will leave the reducer unconverted. Based on the simulation results, the system has an
energy conversion efficiency as high as 58.33% (LHV).
1.4.2.2 Type B Chemical Looping Systems
The Type B chemical looping system uses a CO2 sorbent to enhance the WGS
reaction of syngas by simultaneous removal of the CO2 generated during the shift
reaction. The sorbents include CaO, which is used in the Calcium Looping Process (CLP),
and K2CO3 promoted hydrotalcite and Na2O promoted alumina, both used in the thermal
swing sorption-enhanced reaction (TSSER) process.
Calcium Looping Process (CLP)
Figure 1.13 shows the schematic integration of the calcium looping process in a
typical coal gasification system for the production of hydrogen 47, 140, 147, 148. As shown in
-35-
Figure 1.13, the calcium looping process comprises two reactors: the carbonation reactor
(carbonator), which produces high purity hydrogen while removing contaminants, and the
calciner, where the calcium sorbent is regenerated and a concentrated CO2 stream is
produced. The carbonator is operated at 550 - 650 °C and 20 - 30 atm. In the carbonator,
the CO2 generated by the WGS reaction is simultaneously removed by a CaO sorbent.
The mesoporous, Precipitated Calcium Carbonate (PCC-CaO) sorbent has much higher
reactivity and CO2 capture capacity (40 – 36 weight percent for 50th – 100th cycles) when
compared to most of the high temperature sorbents reported in the literature. Moreover, it
is capable of capturing the sulfur and halides in the raw syngas stream. Hence, the high
performance PCC-CaO sorbent captures the pollutants in the syngas while driving the
thermodynamic equilibrium of the WGS reaction towards the formation of hydrogen until
100% of the CO is consumed. As a result, high purity hydrogen with very low
concentration of H2S, COS, and HCl can be produced with drastically reduced steam
consumption (H2O: CO = ~1:1). The reactions occurring in the carbonator are as follows:
CO + H2O → H2 + CO2
(1.13)
CaO + CO2 → CaCO3
(1.14)
CaO + H2S → CaS + H2O
(1.15)
CaO + COS → CaS + CO2
(1.16)
CaO + 2HCl → CaCl2 + H2O
(1.17)
-36-
The spent sorbent, consisting mainly of CaCO3, is then recycled to the calciner,
where heat is provided to regenerate the carbonated sorbent. The calciner operates at 800
– 1000 °C and ambient pressure.
CaCO3 → CaO + CO2
(1.18)
A mixture of CO2 and steam will be produced from the calciner. After condensing the
steam, CO2 can be compressed and transported for sequestration.
Hence, this technology provides an efficient “one box” mode of operation for the
production of high purity hydrogen with CO2, sulfur and chloride capture that integrates
the WGSR, CO2 capture, sulfur removal and hydrogen separation in one consolidated
unit. High purity hydrogen (> 99.9%) was produced from a lab scale testing unit.
ASPEN® Plus simulations showed that the overall efficiency of the process for hydrogen
production is 63% (HHV)
140
. The large amount of heat required for the calcination
reactor and the sorbent reactivity after regeneration under an elevated temperature
represents a major challenge to the CLP.
Thermal Swing Sorption-Enhanced Reaction (TSSER) Process
The TSSER process also uses CO2 sorbent to enhance the WGS reaction and H2
production. The differences between the TSSER process and the CLP lie in the sorbent
properties, reactor system design, and operating conditions. The sorbents used in the
TSSER process such as K2CO3 promoted hydrotalcite and Na2O promoted alumina
cannot capture pollutants from syngas; therefore, the contaminants in the syngas need to
-37-
be removed before entering the sorbent bed. Moreover, the TSSER is composed of
multiple (fixed) sorbent beds operating in a sequential manner, which is similar to the
operations of the PSA system. The TSSER process is currently under the lab scale testing.
Fuel cell grade hydrogen has been produced from a 17.3 mm diameter fixed bed reactor
149, 150
. The potential challenges to the TSSER process include a relatively low CO2
capture capacity of the sorbent (< 4.4 w.t. %)
150
and constant temperature and pressure
swings in fixed beds under relatively high temperature (300 – 550 ºC).
Sorbent regeneration represents a crucial step to Type B chemical looping
processes. Since significant amount of heat is required for sorbent regeneration, an
optimized energy integration scheme is necessary in order to achieve high energy
conversion efficiency. Moreover, regeneration conditions can have notable effects on the
sorbent recyclability.
1.4.3 Direct Coal Chemical Looping Processes
Both the membrane and the syngas chemical looping approaches discussed in the
previous sections enhance the conventional coal gasification processes by integrating the
WGS and CO2 removal steps into the looping scheme. The advanced coal gasification
processes discussed in this section incorporate, not only the WGS and CO2 removal steps,
but also the coal gasification step. As a result, the coal conversion process is further
simplified.
-38-
1.4.3.1 Type A Coal Chemical Looping Processes
Type A chemical looping processes react coal directly with oxygen carrier
particles, resulting in reduced particles along with an exhaust gas stream with
concentrated CO2. Therefore, particle reduction and coal gasification are performed in the
same unit. Compared to the chemical looping processes discussed in Section 1.4.2, a
dedicated coal gasifier is avoided. The Type A chemical looping gasification processes
can be divided into two sub-categories, i.e. the Chemical Looping Combustion of Coal
(CLCC) process, and the Coal Direct Chemical Looping (CDCL) process.
Chemical Looping Combustion of Coal
Compared to syngas, coal is more difficult to react. The contaminants in coal that
may react with oxygen carrier particles make a direct oxidation of coal an even more
challenging task. Zhao et al. (2008) proposed to use NiO based oxygen carrier particles
(NiO 60% by weight) obtained from sol-gel technique to convert coal char
118
. A TGA
experiment indicted noticeable coal char/NiO conversion over a period of 120 minutes.
As noted, the major challenge associated with NiO based looping processes is the high
oxygen carrier cost, which is especially the case when the elaborate sol-gel technique is
used for synthesizing the particle. Further, the slow reaction kinetics between NiO and
char indicates the necessity for a char gasification promoter.
Yang et al. (2007) proposed to use Fe2O3 as the oxygen carrier particles to covert
coal 151. Fixed bed studies were performed which indicate that Fe2O3 can be converted to
Fe3O4 using coal volatiles and gasified coal gas (CO and H2). However, reduction from
-39-
Fe2O3 to Fe3O4 only utilizes 11.13% of the maximum oxygen carrying capacity of Fe2O3.
Scott et al. (2006) also utilized Fe2O3 as an oxygen carrier particle 152 to convert char in a
small fluidized bed. In their experiments, char was fed into a small fluidized bed. With
the presence of CO2, char was gasified and then reacted with Fe2O3. It was found that
Fe2O3 can only be reduced to Fe3O4 in the fluidized bed due to thermodynamic
limitations. Cao et al. (2006) proposed to use CuO as an oxygen carrier particle to
combust coal
153, 154
in a circulating fluidized bed with only the available data obtained
from TGA. It is noted that the low melting point of Cu/CuO can be a serious issue in its
applications. Studies carried out by others indicate that copper based particles will
deactivate beyond 800 ºC
155
. The low operating temperature will lead to significantly
reduced energy conversion efficiency.
The direct coal CLC processes are at the early stage of development and further
studies in particle development, process design, and analysis are necessary in order to
assess the technical feasibility and the commercial readiness for these processes.
Chemical Looping Gasification of Coal – Coal Direct Chemical Looping Process
The coal direct chemical looping (CDCL) process, illustrated in Figure 1.14, is
capable of converting coal into hydrogen and/or electricity at any relative proportions 121,
122, 156
. In the CDCL process, composite Fe2O3 particles are introduced into the reducer to
react with pulverized coal. With a desired gas-solid contacting pattern, coal is gasified insitu and reacted with Fe2O3 particles. Thus, a mixture of Fe and FeO is produced along
with a flue gas stream composed of CO2, H2O, and contaminants such as H2S and
-40-
elemental mercury. After condensing out the steam, the flue gas can be compressed and
sequestrated. A portion of the reduced Fe/FeO particle from the reducer will enter the
oxidizer to react with steam to form hydrogen. The resulting Fe3O4 exiting the oxidizer
along with the remaining portion of the reduced Fe/FeO particle will be combusted with
air in the entrained flow combustor. The combustor conveys the particle back to the
reducer pneumatically while regenerating the particle to its original oxidized form. Part
of the heat released in the combustor will be carried to the reducer by the hot particles to
compensate the endothermic heat required in the reducer. The remaining heat released in
the combustor heats up the exhaust gas, which can be used for steam or electricity
generation.
The CDCL process testing has been carried out in the 2.5 kWt bench scale
moving bed unit at OSU. Different feedstock such as coal volatiles (simulated), lignite
coal char, bituminous coal char, and anthracite coal have been tested. Coal/coal char
conversion of as high as 95.5% has been obtained. The CO2 concentration in the exhaust
stream was > 97% (dry basis) in all cases. Moreover, the reactivity of the particles was
maintained after three redox cycles in which coal was used as the reducing agent. ASPEN
Plus® simulation showed that the energy conversion efficiency of the CDCL process was
higher 80% (HHV) for hydrogen production and over 50% for electricity generation with
zero carbon emissions 157.
-41-
1.4.3.2 Type B Coal Chemical Looping Process
The Type B coal chemical looping process utilizes high temperature CO2 sorbents
such as calcium oxide to enhance the coal gasification and hydrogen production.
HyPr-Ring Process
The HyPr-Ring process developed in Japan involves coal gasification using pure
oxygen and steam158-161. Figure 1.15 illustrates the HyPr-Ring process. In this process,
coal is fed to the gasifier along with calcium oxide, steam and oxygen. The presence of
excessive steam and the in-situ CO2 removal by calcium oxide drives the equilibrium in
the gasifier towards the formation of H2. As a result, a product gas stream of up to 90%
H2 mixed with methane, other hydrocarbons, and sulfur and nitrogen based contaminants
is generated
161
. The solids from the gasifier consist mostly of saturated CaO sorbents
(CaCO3) and unconverted carbon that is to be introduced to a regenerator along with
oxygen. The heat generated by combusting the unreacted carbon by oxygen allows the
calcination reaction to be carried out for CaO regeneration while producing high purity
CO2 for sequestration. The challenges for the HyPr-Ring process include the deactivation
of CaO in the presence of coal ash
159
and relatively low purity hydrogen product from
the gasifier. The HyPr-Ring process is currently under demonstration in a pilot scale unit
with a coal processing rate of 3.5 kg/hour. Process analysis showed that a 77% cold gas
efficiency (HHV) can be achieved when CO2 compression was not taken into account 158.
-42-
Different from either Type A or Type B chemical looping systems, the GE FuelFlexible Process
162, 163
and the ALSTOM Hybrid Combustion-Gasification Chemical
Looping process 164, 165 employ two separate looping particles, i.e., an oxygen carrier and
a CO2 sorbent, to carry out the coal conversion. Thus, there are two separate looping
schemes in each of these two looping processes. In these processes, the CO2 sorbent is
used to enhance the hydrogen generation while the oxygen carrier is either used for
indirect combustion of fuel to provide the heat needed for spent sorbent reactivation or
used for coal gasification. These processes can convert coal into a variety of products
with in-situ carbon dioxide capture. The mixing between the oxygen carrier particles and
sorbent particles as well as significantly large solid handling requirements render these
processes more difficult to operate as compared to chemical looping processes that
involve single chemical reaction loop.
1.5 Concluding Remarks
Coal will remain to be an important energy source well into the 21st century. With
a strong demand for an affordable energy supply which is compounded by the urgent
needs for CO2 emission control, the clean and efficient utilization of coal represents both
a major opportunity and challenge to current global R&D efforts in this area.
The coal conversion processes of the future prospect are plotted in Figure 1.16
along with the current or demonstrated processes for electricity and/or H2 production.
These efficiencies are given considering a CO2 controlled environment. The processes
considered in the figure include sub-critical and ultra supercritical PC processes retrofit
-43-
with either MEA or chilled ammonia system for CO2 capture, coal gasification processes
using the SELEXOL system for CO2 capture, the H2-selective membrane based
gasification process, syngas chemical looping processes, and the coal direct chemical
looping process (CDCL).
It is seen that in terms of electricity or H2 generation, the efficiencies for syngas
chemical looping processes and H2-selective membrane process are comparable and
could be considered as near term retrofit technology for current coal gasification
processes. Of particular noteworthiness is the CDCL process, which shows considerably
higher energy conversion efficiency than all the other processes considered. The direct
coal chemical looping processes can emerge as attractive clean coal conversion systems
for the intermediate term.
In this thesis, the details of chemical looping gasification technologies developed
at the Ohio State University (OSU) are presented. The key feature of the chemical
looping gasification processes in their ability to generate a sequestration ready CO2
stream is thoroughly discussed. As the looping media employed in these processes are
mainly in solid form and the success of the chemical looping technology applications
depends strongly on the performance of the particles, Chapter 2 is devoted entirely to the
subjects of solid particle design, composition, properties, and reactive characteristics in
the context of the syngas chemical looping (SCL) gasification process, which converts
gaseous fuel such as syngas and light hydrocarbons. Chapter 3 further discusses the
reactor and process design and optimizations of the SCL process using both
-44-
thermodynamic analysis and ASPEN Plus® simulations. Bench scale experimental
results are also presents to substantiate the simulation results. When the conversion of
solid fuel such as coal is desirable, the chemical looping gasification scheme faces
notably different challenges. The solid fuel chemical looping gasification process, i.e. the
coal direct chemical looping (CDCL) gasification process, is discussed in detail in
Chapter 4. Chapter 5 discusses the design, shakedown, and preliminary operation of a 25
KWth sub-pilot scale chemical looping unit. Novel chemical looping applications such as
integration with indirect coal liquefaction, conclusion, and future recommendations are
presented in Chapter 6.
-45-
Technology
Sub-CPC SCPC
USCPC AFBC
Base Case Efficiency (%) HHV
33~37
37~40
40~45
34~38
MEA Retrofit Derating (%)a
30 - 42
24 - 34
21 - 30
~ 35b
a
Percentage decrease in energy conversion efficiency when a retrofit MEA
used to capture 90% of the CO2 in the flue gas.
b
Estimated based on ASPEN simulation by authors
PFBC
38~45
~ 30b
system is
Table 1.1 Energy Conversion Efficiencies (HHV) of Various Coal Combustion
Technologies and Energy Penalty for CO2 Capture Using MEA 17, 38, 52, 62, 65, 166, 167
-46-
Unit Operation
Air Separation Unit
Aspen Plus Model
Sep
Coal Decomposition
Ryield
Coal Gasification
Quench
WGS
Rgibbs
Flash2
Rstoic or Rgibbs
MDEA
Sep or Radfrac
Burner
HRSG
Rgibbs or Rstoic
MHeatX
Gas Compressors
Heater and Cooler
Compr or Mcompr
Heater
Turbine
Compr
Comments / Specifications
Energy consumption of the ASU is based on
specifications of commercial ASU/compressors load.
Virtually decompose coal into various components
(Pre-requisite step for gasification modeling)
Thermodynamic modeling of gasification
Phase equilibrium calculation for cooling
To simulate conversion of WGS reaction based on
either WGS design specifications or thermodynamics
Simulation of acid gas removal based on design
specifications
Modeling of H2/syngas combustion step
Modeling of heat exchanging among multiple
streams
Determines power consumption for gas compression
Simulates heat exchange for syngas cooling and
preheating
Calculates power produced from gas turbine and
steam turbine
Table 1.2 ASPEN models for the key units in the IGCC process
-47-
Thermal
Energy Input
(MWth)
Parasitic Energy Consumptions Power
(MWe)
(MWe)
Generation
Net
Power
Steam
(MWe)
CO2
Turbine
CO2
Gas
Coal ASU Removal Compression Turbine
IP
LP
1000 39.4
9.4
17.7 -249.1 -86.4
-79.1 -348.1
Table 1.3 Power Balance in a 1000 MWth IGCC plant with CO2 capture
-48-
DOE 2010
Membrane Type
Metallic
Porous Ceramic Target
T (°C)
300 - 900
300-700
300-600
Opertating ΔP (MPa)
0.69
0.4
2.75
Selectivity
> 1000
5 - 139
N/A
2
Maximum Flux (SCFH/ft )
60 - 300
60-300
200
Sulfur Tolerance (ppm)
Low
> Metallic
20
>1500 (PtCost (USD/ft2)
Based)
~400
100
Table 1.4 Performances of H2-Selective Membranes72, 76, 85, 168, 169
-49-
Reactor Type
Countercurrent
Moving Bed
Fluidized
Bed/CSTRa
Oxygen carrier
Fe2O3
Fe2O3
51.5
11.27
Maximum metal oxide conversionb (%)
c
Effective oxygen carrying capacity (wt %)
10.82
2.37
CO + H2 concentration in gas exhaust (%)
0.005
0.005
a.
To account for back mixing in the fluidized bed reducer, the fluidized bed reactor is
considered as CSTR
b.
Maximum metal oxide reduction at 850 ºC when more than 99.9% syngas (CO:H2 =
2:1) is converted. Results were obtained based on thermodynamic analysis 170,
c.
Effective oxygen carrying capacity = Maximum oxygen carrying capacity × Metal
oxide loading × Maximum theoretical metal oxide conversion (metal oxide loading is
70% in this case).
Table 1.5 Reducer Performances Using Different Reactor Designs
-50-
a
b
Figure 1.1a Three Different (Long Term) World Oil Price Scenarios Predicted by EIA; b.
World Energy Consumption in 2030 Based on Energy Sources 7
-51-
Figure 1.2 Simplified Schematic Diagram of a Pulverized Coal (PC) Combustion Process
for Power Generation
-52-
a
b
Figure 1.3 Conceptual Schematic of a. the MEA Scrubbing Technology for CO2
Separation; b. the Chilled Ammonia Technology for CO2 Separation
-53-
Figure 1.4 Conceptual Schematic of Carbonation-Calcination Reaction (CCR) Process
Integration in a 300 MWe Coal Fired Power Plant Depicting Heat Integration Strategies
-54-
F-T
Shift reactor
Particulates
Fuels
Sulfur
CO2
Sequestration
Hydrogen
Separation
Syngas
Particulate Sulfur
removal removal
O2
Gasifier
Hydrogen
Clean syngas
Combustor
Fuel cell
Electricity
Compressed
Air
Generator
Coal
Biomass
Air
Compressor
Gas turbine
Steam
Slag
Heat recovery
steam generator
Generator
Steam turbine
Figure 1.5 Schematic Diagram of Coal Gasification Processes 100
-55-
Stack
Figure 1.6 IGCC Process with CO2 Capture
-56-
Figure 1.7 Multiple Stage Membrane System for CO2 Recovery 96
-57-
a.
b.
Figure 1.8 a) Schematic of H2-Selective Membrane Enhanced IGCC Process101; b)
Schematic of CO2-Selective Membrane Enhanced IGCC Process101
-58-
Figure 1.9 Integrated Membrane Separation with Gasifier104
-59-
Figure 1.10 Schematic Flow Diagram of Syngas Chemical Looping Combustion
Processes
-60-
Figure 1.11 Simplified Schematic of the Syngas Chemical Looping Process
-61-
Figure 1.12 Syngas Chemical Looping Enhanced Coal-to-Liquids (SCL-CTL) Process
-62-
Figure 1.13 Schematic Flow Diagram of the Calcium Looping Process
-63-
Figure 1.14 Schematic Diagram of Coal-Direct Chemical Looping Process
-64-
Figure 1.15 Schematic Diagram of HyPr-RING Process 158
-65-
Figure 1.16 Efficiency Comparisons among Various Coal Conversion Technologies with
> 90% CO2 Capture*
* Key Assumptions for Figure 1.16:
Illinois #6 coal is used in all cases;
For SCL, Syngas-CLC, IGCC-Selexol, and Gasification-WGS, a GE quench gasifier is used. A GE 7H
gas turbine combined cycle system is used to generate electricity;
Sub-critical plant operates at 17.5 MPa/538°C/538°C, ultra-supercritical plant operates at 26 MPa/600°C/
600°C;
CO2 is compressed to 15.20 MPa (150 atm) for sequestration.
-66-
CHAPTER 2
OXYGEN CARRIER PARTICLE FOR CHEMICAL LOOPING
GAISIFICATION
2.1 Syngas Chemical Looping Process Overview
The syngas chemical looping process (SCL) has been extensively tested at the
Ohio State University. The process was briefly covered in Chapter 1. This chapter
focuses on the selection and characterization of the oxygen carrier, which is the key to the
SCL process. In the following sections, the SCL process is first elaborated. Different
types of metal oxides were then evaluated based on factors such as oxygen capacity,
reaction kinetics, thermodynamics, physical strength, melting points, and health and
environmental impact. It is determined that an iron oxide-based oxygen carrier can
potentially deliver better performance for hydrogen production. The performance of a
composite iron oxide particle developed at OSU is then reported. The performance
parameters reported include reactivity, recyclability, crushing and attrition strength, and
fixed bed reducing and oxidizing characteristics. It was found that the composite particle
possesses superior reactivity and recyclability when compared to pure iron oxide. The
composite particle is suitable oxygen carrier candidate for the SCL process.
The Syngas Chemical Looping (SCL) process co-produces hydrogen and
electricity from syngas. The SCL process is based on the cyclic reduction and oxidation
-67-
of specially tailored metal oxide composite particles
27, 171
. This process produces a pure
hydrogen stream and a concentrated carbon dioxide stream in two separate reactors.
Therefore, an additional CO2 separation step is avoided. The SCL process consists of five
major components: an Air Separation Unit (ASU); a gasifier; a gas clean-up system; a
reducer and an oxidizer. Figure 2.1 shows a simplified block diagram of the SCL process.
In the SCL process, a high-purity oxygen stream (oxygen concentration > 95%)
from an ASU is sent to the gasifier. The gasifier utilizes oxygen from the ASU and steam
to partially oxidize coal, forming high temperature raw syngas with contaminants such as
particulates, sulfur, mercury, and halogen compounds. After gas quench and particulates
removal, a hot gas cleanup unit (HGCU) is used to remove most of the sulfur in the raw
syngas. As can be seen from Figure 2.1, the SCL process utilizes existing syngas
generation and cleanup systems. The difference between the SCL process and the
conventional coal to hydrogen process lies in the manner in which H2 is generated. For
the coal-to-hydrogen conversion, the SCL process can carry out such functions,
performed in the traditional gasification process, as water-gas-shift reactions, CO2
separation, and pressure swing adsorption for H2 purification (purity > 99.95%) by
utilizing two key looping reactors, i.e. the reducer and the oxidizer. Thus, the overall
energy conversion scheme in the SCL process is significantly simplified over the
traditional process. There are fundamentally three different operating modes for looping
unit operation: fluidized bed, moving bed, and fixed bed. The following discussion
emphasizes more the moving bed mode of operation recognizing that looping particle
recycling often involves dense or dilute pneumatic transport and thus, in essence, the
-68-
looping system in discussion is mostly a moving bed-fluidized bed system. The rationale
behind the selection of such a reactor system is elaborated in Chapter 3. The syngas is
converted in a looping system to hydrogen and electricity in three steps that involve a
reducer, an oxidizer, and a combustor.
Reducer
The purified syngas from the gas cleanup units is introduced to the reducer, which
is a moving bed of iron oxide composite particles operated at 750 - 900 ºC and 30 atm. In
this reactor, the syngas is completely converted to carbon dioxide and water while the
iron oxide composite particles are reduced to a mixture of Fe and FeO (Reaction 2.1 –
2.4).
Fe2O3 +CO Æ 2FeO + CO2
(2.1)
FeO + CO Æ Fe + CO2
(2.2)
Fe2O3 + H2 Æ 2FeO + H2O
(2.3)
FeO + H2 Æ Fe +H2O
(2.4)
The overall reaction in the reducer can be either slightly endothermic or slightly
exothermic depending on the syngas composition, reaction temperature, as well as the
particle reduction rate. The mild endothermic to mild exothermic nature of the reducer
simplifies the heat integration scheme of the reducer reactor, since heat can be readily
carried in or out of the reactor by the chemical looping particles.
-69-
In the reducer, the syngas is nearly completely oxidized by the Fe2O3 composite
particles. Thus, the exhaust gas from the reducer contains mainly CO2 and steam. Steam
can be condensed out by extracting the heat from the high temperature exhaust gas,
resulting in a concentrated high pressure CO2 stream that can be transported for
sequestration.
Oxidizer
After being reduced in the reducer, a portion of the particles are introduced to the
oxidizer. In the oxidizer, the reduced particles react with steam to produce a gas stream
that contains solely H2 and unconverted steam. Once the steam is condensed out from the
gas stream, an H2 stream of very high purity (> 99.95%) can be obtained. The reactions
involved in the reducer reactor include:
Fe + H2O (g) Æ FeO + H2
(2.5)
3FeO + H2O (g) Æ Fe3O4 + H2
(2.6)
The steam used in the oxidizer is produced from the syngas cooling units and combustor.
The oxidizer reactor operates at 30 atm and 500 – 750 ºC. By introducing the low
temperature steam, the oxidizer is adjusted to be heat neutral. The heat released from the
oxidation of the particles to Fe3O4 is used in the same reactor to heat the feed water/steam.
The lower operation temperature of the oxidizer favors the steam to hydrogen conversion.
The oxidizer can also be operated to generate syngas by introducing CO2 produced from
the reducer along with steam. By changing the ratio between the CO2 and steam, the CO
-70-
to H2 ratio in the syngas produced can be altered. The reactions between CO2 and reduced
composite particles are:
Fe + CO2 (g) Æ FeO + CO
(2.7)
3FeO + CO2 (g) Æ Fe3O4 + CO
(2.8)
The CO and H2 mixture can be used for chemical and liquid fuel synthesis. For example,
a syngas product with H2 to CO ratio of 2:1 can be directly converted to liquid fuels in a
F-T reactor.
In the SCL process, hydrogen and/or CO are produced using the chemical looping
reforming concept, where syngas is indirectly converted to hydrogen with the assistance
of iron oxide particles. This is fundamentally different from the traditional coal-tohydrogen process where the syngas is shifted to make hydrogen and followed by a CO2
removal operation. By using the syngas chemical looping process, CO2 is produced from
a reactor different from where hydrogen is produced and hence, eliminating the energy
consuming CO2 separation step. Since a significant portion of the carbon management
cost is associated with the separation and compression of CO2, the SCL process offers a
major advantage over the traditional coal–to-hydrogen process.
Combustor
Fe3O4 formed in the reducer is regenerated to Fe2O3 in a combustor.
The
combustor is a riser that conveys particles to the reducer with pressurized air. The
-71-
combustor also serves as a heat generator since a significant amount of heat is produced
during the combustion of Fe3O4 to Fe2O3:
4 Fe3O4 + O2 → 6 Fe2O3
(2.9)
The high pressure, high temperature gas produced from the combustor can be used for
electricity generation to compensate the parasitic energy consumptions. In yet another
configuration, part or all of the reduced particles from the reducer reactor can be directly
sent to combustor without reacting with steam in the reducer reactor. By doing so, more
heat would be available for electricity generation at the expense of decreased hydrogen
production.
The step of combusting the reduced particles characterizes chemical-looping
combustion, which differs from the chemical-looping reforming step in which the particle
is regenerated using steam instead of oxygen. Hence, the utilization of both chemicallooping reforming and chemical-looping combustion concepts in the SCL system allows
it to be flexible in adjusting the product yield between H2 and electricity. Moreover, only
one type of particles with high recyclibility and reactivity is circulating among the three
units, rendering the process highly efficient. The overall efficiency for the SCL process is
estimated to be equivalent 1 to ~ 67% (HHV) for hydrogen generation. A process concept
similar to that of the SCL is being explored by ENI S.p.A.
172
. In the ENI process,
countercurrent multi-stage fluidized bed reactors are used for both the reducer and the
1
Electric energy of 1 joule is considered to be equivalent to H2 with heat content of 1.6 joules (HHV)
-72-
oxidizer 173. The reduction of Fe2O3 to a mixture of FeO and Fe3O4 was reached in their
lab scale reactors.
2.2 Oxygen Carrier Selection
Table 2.1 illustrates the importance of the oxygen carrier particles, which
circulate inside the chemical looping system and participate in all the major chemical
reactions. The functions of the oxygen carrier particles include the following: oxidation
of the syngas in the reducer, production of hydrogen in the oxidizer, and generation of
heat to produce power in the combustor. Therefore, particle performance is crucial to the
process. This section discusses the criteria for oxygen carrier selection.
A number of factors need to be considered to obtain the optimum oxygen carrier
particle. Such factors include thermodynamic properties, reactivity, recyclability, cost,
melting temperatures, physical strength, and health and environmental impacts. Among
these factors, thermodynamic properties underlie the functionality of the particle.
Specifically, the oxidized particle must fully oxidize syngas and the reduced particle must
convert a significant portion of steam into hydrogen. More detailed information on the
thermodynamic feasibilities of various particles has been discussed in various literature
114, 144, 145, 174
. These studies identified the oxides of Fe, Mn, Ni, Cu and Co as potential
candidates for hydrogen generation. Rather than repeating the thermodynamic analysis,
the present study approaches the particle selection using a systematic comparison. Table
2 generalizes the various factors that affect the particle performance in the looping
process. As can be seen in Table 2.2, although the iron-based oxygen carrier particle has
-73-
relatively slow reduction kinetics, it possesses favorable thermodynamic properties and is
less costly. Moreover, iron-based particles have good physical strength, high melting
temperatures, and fewer environmental concerns. Therefore, iron oxide is a favorable
choice for the SCL process. The performance of the iron-based particles is further
discussed in the following sections.
2.3 Oxygen Carrier Performance
2.3.1 Experimental
Particle and Pellet Preparation
Fe2O3 composite particles were prepared in forms of both powder and pellets
using methodology similar to that described in Gupta et al.114. The composite particle
contains up to 70 wt.% Fe2O3. The composite pellets are cylindrical with a 5 mm
diameter and 1.5 – 4.5 mm in height.
Particle Reactivity and Recyclability
A Perkin Elmer Pyris 1 thermogravimetric analyzer (TGA) is used to characterize
the reactivity and recyclability of different particles. The schematic of the experimental
setup is shown in Figure 2.2. Powder samples are directly used in the TGA whereas pellet
samples are broken in a mortar then sieved into different size ranges before being loaded
into the TGA. Unless otherwise mentioned, broken pellet samples with sizes ranging
between 710 μm and 1 mm are used in the TGA. Before each experiment, around 20 mg
of particle are loaded into a quartz crucible. Next, the TGA is purged with N2 to introduce
-74-
an inert atmosphere. The crucible is then heated to the desired reacting temperature, 900
ºC.
To compare the reactivity of various Fe2O3-based samples, about 160 ml/min of
reducing gas, comprised of 37.5% H2 balanced with N2, is introduced to the TGA. The
weight change of the sample is recorded as a function of time, and the experiment is
stopped after the sample weight stabilizes. This point corresponds to the maximum
reduction achievable under the given conditions. Since the rate of weight change (oxygen
carrying capacity change) is proportional to the rate of the reduction reaction, the
reactivity can be quantified using the TGA curve. In the present study, two values, i.e. the
maximum rate of weight change (wt % / min) and the time required to reach 80%
conversion are used to compare the reactivity of the particles during reduction, with a
similar analysis used to obtain oxidation reactivity. Following the complete reduction of
the previous sample, the TGA is purged with N2. Then, 200 ml/min of oxidization gas,
consisting of 45% air balanced with N2, is fed into the TGA. To simulate redox cycles,
the reducing and oxidizing gases are alternately introduced to the TGA with 15 minutes
N2 flushing in between. The change in reactivity is then monitored across cycles to
calculate recylability.
Particle (Pellet) Strength
The crushing strength of the cylindrical shaped particles/pellets is determined
using a modified version of the ASTM D4179 standard using a hydraulic press installed
with a digital pressure transducer. During testing preparation, a pellet is loaded into the
-75-
hydraulic press. Next, the manually-operated press compresses the pellet in an axial
direction. Then, a computer records the pressure at which the pellet is crushed. In order to
obtain a reliable mean crushing strength and distribution of the pellet crushing strength,
more than 50 pellets are tested for each sample composition.
The attrition rate of the particles/pellets was tested in an entrained flow reactor
that simulates the combustor operation, as shown above in Figure 2.3. The reactor is 2.7
meters tall with an outer diameter (O.D.) of 2.54 cm and inner diameter (I.D.) of 1.91 cm.
Gas can be introduced to the bottom of the reactor through the distributor.
328.5 g cylindrical composite pellets that have been reduced and oxidized for two
cycles are used as the fresh sample. Before each experiment, the composite pellets are
loaded to the bottom of the reactor. The valve is then opened to send air to the reactor at
5.43 liter/second. Such an air flow rate corresponds to a superficial gas velocity of 18.96
m/s, which is higher than the terminal velocity of the pellets. As a result, the air
pneumatically conveys the pellets to the top of the reactor,through the U bend, and
eventually to the funnel shaped pellet collector. The particles, after being collected, are
sieved into four different size ranges, i.e. >2.8 mm, 2.8 – 1.98 mm, 0.71 – 1.98 mm, <
0.71 mm. The weight of particles in each size range is then recorded. After recording the
weight, the particles are mixed together and reloaded into the reactor to test the attrition
rate for another cycle. This operation is repeated for more than 10 cycles.
Fixed Bed Studies
-76-
A fixed bed reactor system, shown above in Figure 2.4, is used to study the redox
reactions involved in the SCL process. The fixed bed reactor consists of a quartz tube
with 16mm OD and 12mm ID surrounded by a 20 mm ID external electric oven capable
of reaching 1000 ºC. This oven heats 60 cm of the quartz tube. The inlet of the reactor at
the top of the quartz tube consists of a water inlet and a gas inlet connected with a tee.
Water is delivered from a Harvard PHD 2000 syringe pump. A capillary tube, with an ID
of 0.5 mm and an OD of 0.32 mm feeds the water into the heated zone of the reactor,
which vaporizes the water into steam. This capillary tube steam nozzle is located inside a
large diameter tube (6 mm ID) connected to the gas inlet. Thus, the reactant gases flow
into the reactor outside of the capillary tube. In an experiment, 8 – 30 grams of solid
sample is first loaded into the quartz tube. Next, the reactant gases and/or steam are then
introduced from the reactor inlet (top). Finally, the exhaust gas exiting the outlet (bottom)
of the reactor is passed through a DRIERITE desiccant bed and characterized using a
Varian CP 4900 MicroGC with with a Molarsieve 5A column and a Poraplot U column,
both with TCD detectors.
Particle Reduction using Syngas
In the particle reduction experiment, 22.1 g of fresh Fe2O3 composite pellets (60%
Fe2O3 with 40% support) were used. The total flow rate of the reducing gas was 505.6
ml/min (STP), and its composition is shown in Table 3. Before the experiment, N2 was
introduced to the reactor at a flow rate of 500 ml/min while the reactor was gradually
heated up to a reaction temperature of 830 ºC. Then, the reactant gas with aforementioned
the composition was introduced. The Varian CP 4900 MicroGC was used to monitor the
exhaust gas composition until steady state was achieved. After the experiment, the reactor
-77-
was cooled down using an N2 purge. To analyze carbon content, four grams of solid
sample (~ 40 pellets) were taken out and grinded into powder. Then, a CM-120 (UIC
Inc.) total carbon analyzer was used to characterize the amount of carbon in the
pulverized sample. The pulverized sample was also oxidized in the TGA with air to
characterize the conversion of the sample. The solid conversion can be calculated using
the following equation:
x = (W2 − W1 + W1c) 0.3W2θ
(2.10)
Here, x denotes the solid conversion, which is defined as the ratio between the
remaining reducible oxygen content (by weight) in the solid sample and the total
reducible oxygen content of the sample in its fully oxidized form. W1 is the weight of the
sample loaded into the TGA, W2 is the weight of the same sample after being oxidized in
the TGA with air, c is the carbon content (wt %) in the original sample, and θ is the Fe2O3
content (wt %) in the composite particle.
Hydrogen Production and Particle Oxidation
After the reduction experiment, water was injected into the reactor by the syringe
pump at a rate of 0.1553 ml/min, or 0.00863 mol/min, to oxidize the remaining pellets.
Nitrogen was also introduced at a flow rate of 773.2 ml/min (STP) or 0.0345 mol/min.
Therefore, the molar concentration of water, and thus the steam inside the reactor, was
kept at 20%. The reaction temperature was kept at 830 ºC, and the hydrogen
concentration in the exhaust gas stream was monitored using the Varian CP4900
MicroGC until the hydrogen concentration in the exhaust gas stream decreased to zero.
After the experiment, the fixed bed reactor was cooled to room temperature with nitrogen.
-78-
Solid sample was taken and characterized using the same methodology used in the
reduction experiment. The remaining particles were then further oxidized using air to
complete the redox loop of the SCL process. Recyclability of the oxidized sample was
tested in the TGA using the cyclic reduction-oxidation method described earlier.
2.3.2 Results and Discussions
Oxygen Carrier Reactivity and Recyclability
A desirable oxygen carrier particle for the SCL process possesses good reactivity
and recyclability. Increased reactivity reduces the reactor size while increased
recyclability minimizes the particle makeup rate. The reactivity and recyclability of a lab
grade Fe2O3 powder (NOAH Tech. Co.) was first tested. As can be seen in Figure 2.5a,
the reactivity and the oxygen carrying capacity of the Fe2O3 decreased significantly after
the first redox cycle. Following the initial reduction, the iron oxide particle never fully reoxidized to Fe2O3 form. Therefore, pure Fe2O3 is proved unsuitable for the SCL process.
In comparison, the recyclability test results obtained from crushed Fe2O3 composite
pellets (850 – 1000 μm) developed at OSU are shown in Figure 2.5b. Figure 2.5c
superimposes the reduction curves at different redox cycles for the composite particles.
As can be seen in Figure 2.5b, the composite particles can be fully reduced and
oxidized for 100 redox cycles. Figure 2.5c further illustrates time required to achieve
80% reduction and oxidation at various redox cycles showing that the reactivity of the
particle was maintained during the 100 redox cycle test. For example, the 80% reduction
time for particles at the 100th cycle was merely 3% longer than that for particles at the 2nd
-79-
cycle. Notable increase in the oxidation reactivity with increasing number of cycles can
also be observed in Figure 6c. The key results from the TGA studies are generalized in
Table 2.4. In conclusion, the composite particles display distinctively higher reactivity
and recyclability when compared to the pure iron oxide powder. Thus, the composite
particles can be suitable for SCL operations.
Pellet Strength
To avoid fluidization of the particles in the moving bed reducer and oxidizer, the
SCL particles are in pellet form. Desirable pellets must maintain their physical integrity
for extended number of redox cycles. Thus, the strength of the pellet is an important
factor that affects the particle purge rate. Calculation of reactor hydrodynamics for a
commercial SCL system indicates that composite particle (pellets) must be larger than
700 μm for use in the SCL process. In the following discussions, particles larger than 710
μm are considered to be usable for the SCL operations. Particles smaller than 710 μm are
purged and captured for reprocessing.
Pellet Crushing Strength
Three types of pellets, i.e. commercial water gas shift (WGS) catalyst pellets,
fresh Fe2O3 composite pellets, and Fe2O3 composite pellets after two redox cycles, were
studied. All three pellets were cylindrical shape with 5 mm diameter and 1.5 – 4.5 mm
height. The histogram of the composite pellet crushing strengths is shown in Figure 2.6
along with the mean crushing strength of the commercial WGS catalyst pellets. As shown
in Figure 2.6, the crushing strength of the composite pellets nearly doubled after two
redox cycles. The dramatic increase in strength may have resulted from the change in
-80-
physical properties during the redox reactions at high temperature. The mean crushing
strength of the composite pellet was over 20 MPa after two redox cycles, three times
higher than the WGS catalyst pellet. The high crushing strength makes the composite
pellets suitable for commercial SCL units.
Pellet Attrition Rate
Due to the turbulent gas-solid interactions in the entrained bed reactor, the particle
attrition in the combustor is more severe than that in the reducer or the oxidizer. The
attrition tests assessed the pellet attrition rate in combustor operations. Composite pellets
after two redox cycles were tested in the entrained flow reactor. Ten consecutive
pneumatic conveying tests were conducted, and the weight of particles in each particle
size range was recorded after each experiment. The attrition test results are plotted in
Figure 2.7.
Since pellets smaller than 710 μm are deemed as fine and will be purged, 710 μm
is used as the critical value to determine the pellet attrition rate. The pellet attrition rate is
defined as the incremental weight fraction of the pellets smaller than 710 μm after each
attrition test. As shown by the solid curve in Figure 8, the attrition rate remained stable
during the 10 pneumatic conveying tests at a value of 0.57% (by weight). The dotted
curve in Figure 2.7 illustrates the amount of pellets (>710 μm) in the system as a function
of the conveying cycles given the addition of 0.57% (by weight) fresh makeup pellets
each cycle. With 0.57% fresh pellet makeup rate, the total weight of pellets circulating
-81-
inside the SCL process remained stable. Thus, the attrition test results can be used as an
estimate of the pellet attrition/purge rate during commercial SCL operations.
Fixed Bed Reactor Experiments
Pellet Reduction Experiment
The purpose of reduction experiment is to validate the reducer concept proposed
in the SCL process. The composition of the exhaust gas (dry basis) during the reduction
experiment is plotted in Figure 9. The concentration N2 carrier gas is not shown in Figure
2.8. The CO and H2 conversions before breakthrough as well as the carbon content and
Fe2O3 conversion in the particle after the completion of the experiment are listed in Table
2.5.
Before breakthrough, both CO and H2 were almost completely oxidized given the
rather short gas residence time in the reactor (~ 6 ms). This is mainly due to the presence
of Fe2O3, which, as shown in Table 2.2, is capable of oxidizing nearly 100% of the
syngas. As the reaction proceeded, the Fe2O3 phase disappeared due to particle reduction
and the breakthrough subsequently took place. The reduction of Fe2O3 to lower oxidation
states such as Fe3O4, FeO, or Fe led to significantly decreased CO and H2 conversion.
This phenomenon agrees with predictions based on the thermodynamic properties of the
metal oxides in Table 2. Solid analysis showed that the particles were reduced by 94.6%
with small amounts of carbon deposition on the surface of particles (0.02 wt. %). The
carbon deposition resulted from the reverse Boudard Reaction:
-82-
2CO → C +CO2
Methods that minimize carbon deposition have been further discussed by Gupta et al 114.
The reduction experiment in the fixed bed reactor validates that the presence of Fe2O3 can
oxidize syngas into an exhaust stream of CO2 and steam. Therefore, a ready-to-sequester
CO2 stream can be obtained by condensing out the steam from the reducer effluent gas.
Moreover, Fe2O3 is reduced by syngas to its metallic form with minimal carbon
deposition.
Pellet Oxidation Experiment
The reduced particles from previous experiment were oxidized using 20% steam
balanced with N2 in the same fixed bed reactor. Figure 10 below shows the concentration
of H2 and CO (dry basis) exiting from the fixed bed reactor.
Although water is injected at a constant rate, it evaporates in the capillary tube in a “batch
mode”. The instability in steam flow rate leads to the fluctuation in hydrogen
concentration at the outlet. Table 2.6 shows the average steam conversion before
breakthrough, the average and the lowest hydrogen purity (normalized to N2 and moisture
free basis), and the compositions of the composite particle after the experiment.
As can be seen from Figure 2.9 and Table 2.6, before the breakthrough, nearly
80% of the steam is converted due to the presence of metallic iron. After the complete
oxidation of the iron phase to higher oxidation states, the breakthrough occurs,
characterized by a sharp decrease in steam conversion. The average H2 purity is 99.8%
-83-
with CO as the only impurity. The CO is formed due to steam gasification of the carbon
deposited during the syngas reduction stage, as shown below:
C + H2O Æ CO + H2
Characterization of the solid sample after the steam oxidation experiments
showed that the carbon content in the solid sample remained undetectable. This suggests
that all the carbon formed during reduction stage was gasified during the steam oxidation
stage. Also, the particle was almost completely regenerated to Fe3O4. After the steam
oxidation experiment, the pellets in the fixed bed were fully re-oxidized with air to Fe2O3
in this step, releasing heat. TGA recyclability tests performed on the re-oxidized pellets
showed that the pellets were recyclable with reactivity comparable to fresh pellets.
2.4. Conclusions
The studies presented in this chapter show that an iron oxide-based oxygen carrier
particle a suitable for the novel syngas chemical looping (SCL) process. Adding supports
to the iron oxide drastically increased the reactivity and recyclability of the oxygen
carrier. The TGA experiments showed that the composite particle can maintain
recyclability for more than 100 cycles. The pelletized particle showed good crushing
strength (> 20 MPa) and a low attrition rate (< 0.57 %).
The composite iron oxide-based oxygen carrier particle was used to perform a
complete syngas reduction – steam oxidation – air regeneration cycle in a fixed bed
reactor. During the reduction stage, more 99.75% syngas was converted into steam and
-84-
CO2. Meanwhile, the oxygen carrier particle was reduced by nearly 95% with minimal
carbon deposition. During the steam oxidation stage, an average hydrogen purity of
99.8% (dry, N2 free basis) was obtained. The oxygen carrier was partially regenerated to
Fe3O4 during this step. It was then fully regenerated with air. The purity of the hydrogen
product can be further increased by minimizing carbon deposition during the particle
reduction stage.
In conclusion, the composite iron oxide particle shows the potential to be a viable
oxygen carrier for the SCL process for hydrogen and electricity production. In fact, such
a particle can also be applied to the coal direct chemical looping (CDCL) process, which
is elaborated in Chapter 4. Through the TGA and fixed integral bed experiments, the key
concepts involved in the SCL process are proved to be feasible.
-85-
Name
Reducer
Oxidizer
Combustor
Reactor Type
Counter
Current
Moving Bed
Counter
Current
Moving Bed
Entrained Bed
Reactions
Temperature
Pressure
Fe2O3 +CO/H2 Æ 2FeO + CO2/H2O (g)
FeO + CO/H2 Æ Fe + CO2/H2O (g)
750 - 900 ºC
3.0 MPa
Fe + H2O (g) Æ FeO + H2
3FeO + H2O (g) Æ Fe3O4 + H2
500 – 750 ºC
3.0 MPa
4 Fe3O4 + O2 → 6 Fe2O3
950 – 1150 ºC 3.2 MPa
Table 2.1 Reactor Type, Main Reactions, and Operating Conditions for SCL Reactors
-86-
NiO
CuO
Mn3O4
CoO
Fe2O3
1
Cost
+
–
~
–
–
Oxygen Capacity2 (wt %) 30
21
20
20
21
Thermodynamics: Syngas Fe2O3/Fe3O4/FeO 99.3
CuO/Cu2O Mn3O4/MnO CoO
Conversion3
100/83.2/42.5
100/100
100/0
96.3
Thermodynamics: Steam Fe/FeO/Fe3O4
–
–
–
Co
Conversion4
55.9/15.8/0.05
3.5
Reduction
~
+
+
~
–
Kinetics/Reactivity5
Melting Points6
1275 (pure iron) 1452 1026
1260 (Mn)
1480
Strength
+
–
–
~
~
Environmental& Health ~
–
~
–
–
Impacts
1.
+: positive; –: negative; ~neutral
2.
Maximum possible oxygen carrying capacity by weight percent, pure basis; achievable
using excess fuel (actual)
3.
Maximum theoretical conversion of a syngas (66.6% CO and 33.3% H2) to CO2 and
H2O with the presence of the given metal oxide at 850 ºC (calculated by Aspen Plus)
4.
Maximum theoretical conversion of steam with the presence of given metal oxide at
different oxidation states at 850 ºC (calculated by Aspen Plus)
5.
Reactivity refers to the rates of the reactions between metal oxides and syngas (CO and
H2)
6.
Lowest melting points of the metal/metal oxides under various oxidation states (ºC),
Co3O4 and Co2O3 is not considered in this case since it hard to be oxidized;
Table 2.2 Comparisons of the Key Properties of Different Metal Oxide Candidates 114, 175177
-87-
Type of Gas
Flow Rate (ml/min, STP)
Concentration (%)
CO2
13.0
2.6
H2
127.1
25.1
CO
252.2
49.9
N2
113.3
22.4
Table 2.3 Inlet Gas Composition during the Reduction Experiment
-88-
Total
505.6
100.0
Maximum
Reduciton
Rate
(w.t.%/min)
5.89
80%
Reduction
Time (min)
Maximum
80%
Oxidation
Oxidation
Rate
Time
(w.t.%/min)
(min)
5.78
42.9
Pure Fe2O31
38.0
Composite
11.11
6.9
6.13
5.1
Particle1
1
Data from the second redox cycle is used for pure iron oxide (since it is not recyclable)
and data from the fifth redox cycle is used for OSU composite pellet;
Table 2.4 Reactivity Comparisons between Iron Ore Powders and OSU Composite
Particles
-89-
CO conversion before H2 conversion before Carbon content after Particle conversion
breakthrough (%)
experiment (%)
after experiment (%)
breakthrough (%)
99.76%
99.75%
0.02
94.6
Table 2.5 Gas and Solid Conversions in the Reduction Stage of the Fixed Bed
Experiment
-90-
Average steam conversion Average H2 purity in Lowest H2 purity in Fe3O4 content after
the experiment (%)
the experiment (%) the experiment (%)1
before breakthrough (%)
79.10
99.80
99.66
99
1.
The product from the steam oxidation experiment is a mixture of FeO and Fe3O4.
Percentage of Fe3O4 denotes the percentage (by weight) of Fe3O4 in the solid mixture.
Estimated based on the average oxidation state of iron in solid sample taken after the
fixed bed experiment.
Table 2.6 Gas and Solid Conversions in the Oxidation Stage of the Fixed Bed Experiment
-91-
Figure 2.1 Simplified Schematic of the Syngas Chemical Looping Process for Hydrogen
Production from Coal
-92-
Balance
Computer
T
N2 Purge
Reactant Gas
Figure 2.2 Schematic of the Experimental Setup for the Particle Reactivity and
Recyclability Studies
-93-
0.3 m
2.0 m
Downer
Riser
2.7 m
Superficial Gas
Velocity: v=18 m/s
ID: 1.91 cm
OD: 2.5 cm
Gas
Outlet
Filter
Collector
0.5 m
Pellets
Valve
Distributor
Gas Inlet Gas Flow
Meter
Figure 2.3 Schematic of the Entrained Bed Setup for Particle Attrition Studies
-94-
Figure 2.4 Schematic of the Fixed Bed Reactor Setup
-95-
a
100
Weight (%)
95
90
85
80
75
70
0
100
200
300
400
Time (min)
500
600
700
800
b
100
98
96
Weight %
94
92
90
88
86
84
82
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
Time (min)
Figure 2.5 a. Pure Fe2O3 Particle Recyclability Test;
b. Composite Particle
Recyclability Test; c. Time (min) Required for 80% Reduction and Oxidation at Various
Cycles for the Composite Particle. (To be Continued)
-96-
Figure 2.5: Continued
c
80% Reduction and Oxidation Time (min)
10
Oxidation
Reduction
9
8
7
6
5
4
3
2
1
2
5
10
20
30
40
50
Number of Cycles
-97-
60
70
80
90
100
0.4
Fresh Pellet
0.35
Pellet after Two
Redox Cycles
Frequency
0.3
Commercial WGS
Catalyst Pellet
0.25
0.2
0.15
0.1
0.05
0
3.0
6.8
10.5
14.3
18.0
21.8
25.5
Crushing Strength (MPa)
Figure 2.6 Crushing Strength Test of OSU Composite Pellets
-98-
5.1
Percent of Pellets with Desirable Size
100
99
98
97
96
95
Pellets >0.71 mm
94
Pellets >0.71 mm with 0.57%
fresh pellet makeup
93
92
0
1
2
3
4
5
6
7
8
9
Number of Pneumatic Conveying Cycles
10
Figure 2.7 Attrition Rate of the Composite Pellet in an Entrained Flow Reactor
-99-
70
Outlet Gas Concentration (%)
H2
60
CO
CO2
50
40
30
20
10
0
0
20
40
60
80
Time (min)
Figure 2.8 Composition of the Exhaust Gas Stream from the Fixed Bed Reactor during
the Reduction of the Fe2O3 Composite Pellets (Dry Basis)
-100-
Outlet Gas Concentration (%)
20
H2
CO
15
10
5
0
0
10
20
30
40
Time (min)
50
60
70
Figure 2.9 Composition of the Exhaust Gas Stream from the Fixed Bed Reactor during
the Oxidation of the Reduced Fe2O3 Composite Pellet using Steam (dry basis)
-101-
CHAPTER 3
SYNGAS CHEMICAL LOOPING GASIFICATION PROCESS
With the development of a suitable oxygen carrier particle, this chapter discusses
the modeling and optimization of the key SCL reactors, the bench scale experiment
results for the key SCL reactor operations, and the SCL process simulations. The
thermodynamic analysis based on both analytical method and ASPEN Plus® simulations
show that a countercurrent moving bed design should be used for the SCL reducer and
oxidizer in order to achieve better gas and solid conversions. The experimental results
performed in the bench scale moving bed corroborate well with the reactor simulation
outcomes. SCL process simulation using ASPEN Plus® is also performed, taking into
account the results obtained from both reactor simulations and bench scale experiments.
The process simulation shows a 5 – 7% efficiency increase for the SCL process when
compared to the current coal to hydrogen process based on coal gasification – water gas
shift route. The material presented in this chapter provides further evidences that support
the feasibility as well as the technical advantages of the SCL process. Information
regarding the on-going sub-pilot scale SCL demonstration is provided in Chapter 5.
Novel applications of the SCL concept are discussed in Chapter 6.
-102-
3.1 Thermodynamic Analyses of SCL Reactor Behavior
The thermodynamic properties of iron as well as other metals are briefly covered
in Chapter 2 in the context of the primary metal/metal oxide selection. Here, the
thermodynamic properties of iron are examined in relation to its reactions with gaseous
reactants and reactor operations. Figure 3.1 illustrates a simplified iron oxide conversion
scheme in the SCL process. As can be seen in the figure, the SCL process produces
hydrogen and/or heat through the reduction and oxidation of iron (oxide) particles in a
cyclic manner. The maximum gas and solid conversions, which are determined by
thermodynamics, are closely related to the performance of the chemical looping process.
Higher gas and solid conversions will lead to lower particle circulation rate, higher
product purity, lower parasitic energy requirements and hence improved energy
conversion efficiency.
Thermodynamic Properties of Iron Oxides
Iron has four oxidation states: metallic iron, wustite, magnetite, and hematite. In
chemical looping processes, iron can swing between any two of the four oxidation states
or the mixtures thereof. For example, in the conventional steam-iron process, iron swings
between Fe3O4 and a mixture of FeO and Fe3O4. It can be illustrated using the
thermodynamic diagrams, such as those in Figure 3.2, that the two oxidation states
between which iron swings determine the extents of the gas and the solid conversions.
Figure 3.2 shows the equilibrium gas compositions of both iron-carbon-oxygen
system and iron-hydrogen-oxygen system at different temperatures 178, 179. It is noted that
-103-
FeO is used here to represent Wustite, whose exact formula varies with temperature.
From the phase diagrams, the equilibrium gas concentrations for different oxidation states
of iron may vary significantly at any given temperatures. The equilibrium gas
compositions in both iron-carbon-oxygen and iron-hydrogen-oxygen systems at 850ºC
are shown in Table 3.1.
The implications of the equilibrium gas concentrations are two fold: From the gas
conversion viewpoint, the fact that 99.9955% CO2 equilibrates with Fe2O3/Fe3O4
suggests that at 850 ºC, the presence of excessive Fe2O3 will lead to 99.9955%
conversion of CO to CO2 for an Fe-C-O system; From the solids conversion viewpoint,
provided that CO concentration is higher than 45 ppm, part or all of the Fe2O3 in the
system will be reduced to a lower oxidation state. To compare, the CO concentration
must be higher than 62% in order to reduce FeO to Fe. Similar cases can be observed for
an Fe-H-O system. Therefore, iron at higher oxidation states is more effective in
oxidizing H2/CO to H2O/CO2 while that at lower oxidation states is more favorable for
the conversion of H2O/CO2 to H2/CO.
In the iron based chemical looping processes generalized in Figure 3.2, the iron/
iron oxide reduced by the reducer is used in the oxidizer to convert steam into hydrogen.
From its thermodynamic properties, iron under high oxidation states is a desirable
feedstock for the reducer in order to achieve high syngas conversions. Meanwhile, the
iron produced from the reducer should have low oxidation states in order to maximize the
steam to hydrogen conversion in the oxidizer. Therefore, a reducer that maximizes solids
-104-
and gas conversions (x - y in Figure 3.1) has the potential to maximize the overall energy
conversion efficiency of the process. As illustrated in the next section, the reactor design
plays an important role in maximizing the gas and solids conversions in chemical looping
processes.
Reactor Design and Gas-Solid Contacting Patterns
A key challenge to the SCL process lies in the design of the reducer and the
oxidizer. Unlike the oxidation reaction in the combustor which is intrinsically fast and is
thermodynamically favored, the reactions in the reducer and the oxidizer are limited by
the thermodynamic equilibrium and are relatively slow. It is, therefore, desirable to
utilize an optimal reducer and oxidizer design that is reliable and less capital intensive
while maximizing the solid and gas conversions via a thermodynamically favored gassolid contacting scheme. In this section, different designs for the reducer are analyzed.
The oxidizer design analysis can be carried out in a similar manner.
Three reactor operating modes, i.e. fixed bed, moving bed, and fluidized bed, are
investigated. A fixed bed design similar to that employed in Lane’s Steam iron process
eliminates the particle movement; however, such a design involves constant switching of
reducing and oxidizing gases in an intermittent manner. The needs for gas switching
under high temperatures and high pressures make it challenging for the design of the
valve system. In addition, the reduced iron oxide particle tends to catalyze the reverse
Boudard reaction giving rise to carbon deposition on the particle 27. Therefore, excessive
carbon deposition may occur at the inlet of the fixed bed during the reducer operation.
-105-
The carbon deposition will reduce the purity of the hydrogen product. Moreover, it can
affect the reactivity of the particle. Another challenge to a fixed bed design is the
effective heat removal, which is required during the combustion step. To compare, the
aforementioned problems can be either avoided or minimized when a fluidized bed or a
moving bed design is employed due to the continuous movement of both gas and solids.
Therefore, a fluidized bed or a moving bed design is preferred for both the reducer and
oxidizer.
In order to compare the maximum gas and solid conversions using a fluidized bed
design to that using a moving bed design, a thermodynamic analysis was performed on a
fluidized bed reducer and a moving bed reducer. It should be noted that the
thermodynamic analysis predicts the gas and solid conversions when a thermodynamic
equilibrium among the reactants and the products is reached. Such the equilibrium can be
easily reached under the condition when the reaction is sufficiently fast and/or the gassolid contacting time in the reactor is sufficiently long. The reactions conducted in the
chemical looping processes are under this condition and therefore, the thermodynamic
analysis can project the performance of the looping reactors with reasonable accuracy.
For illustration purpose, pure H2 and pure Fe2O3 are used as the feedstock for the
reducer, which is operated at 850 ºC. The ratio between the solid and gas molar flow rate
is set to be s. The conversions of H2 and Fe2O3 are denoted as x and y respectively. The
conversion of Fe2O3 is defined as the percentage of oxygen depleted from pure Fe2O3 as
given by
-106-
y=
nˆ O nˆ Fe − nO n Fe
× 100%
nO n Fe
(3.1)
Here, nO n Fe corresponds to the molar ratio between the oxygen atom and the iron atom
in Fe2O3 while nˆ O nˆ Fe corresponds to the molar ratio between the oxygen atom and the
iron atom in the reduced solid product, i.e. FeOx (0 < x < 1.5). For instance, the reduction
of Fe2O3 to Fe3O4 corresponds to a solid conversion of (3/2-4/3)/(3/2)×100% = 11.11%,
FeO corresponds to a conversion of 33.33% and Fe corresponds to 100% solid
conversion. When H2 is the only reducing gas, the possible reactions in the reducer
include:
K1= PH2O/PH2=1.92×104
@850ºC
(3.2)
H2 + Fe3O4 ÅÆ H2O + 3FeO
K2= PH2O/PH2=3.5454
@850 ºC
(3.3)
H2 + FeO ÅÆ H2O + Fe
K3= PH2O/PH2=0.5344
@850 ºC
(3.4)
H2 + 3Fe2O3 ÅÆ H2O + 2Fe3O4
Here, K1-K3 are the equilibrium constants which can be readily derived from Figure 3.2.
3.1.1 Reactor Thermodynamic Analysis Based on Analytical Method
Fluidized Bed Reducer
In a fluidized bed reactor such as a dense-phase fluidized bed, significant mixing
for the gas and the solid in the reactor occurs. Thus, in a fluidized bed reducer (Figure
3.3a), the fresh syngas feedstock will be diluted by the gaseous product which is rich in
H2O and CO2. From Figure 3.2, the dilution by CO2 and H2O decreases the reducing
-107-
capability of the syngas. Similarly, the mixing of solids in a fluidized bed results in the
discharge of low conversion solids from the fluidized bed. Therefore, the gas and solid
conversions in the fluidized bed reducer is constrained. A similar constraint applies when
a fluidized bed reactor is used as the oxidizer.
To illustrate the effect of mixing, it is assumed that both the solid and the gas are well
mixed in a fluidized bed reactor in the following analysis. The oxygen mass balance on
the reactor can be given as:
x = 3sy
(3.5)
This equation indicates that the oxygen depleted from the solid is transferred to the gas
through the formation of steam or CO2. Meanwhile, the thermodynamic equilibrium
gives:
K n = x /(1 − x)
(3.6)
Here, n is 1, 2, or 3 depending on the phase of iron in the reducer. According to Figure
4.4.2, when excessive Fe2O3 is present (0 ≤ y < 11.11%), the equilibrium constant will
follow K1; when Fe3O4 and FeO mixture is present (11.11% ≤ y < 33.33%), the
equilibrium gas composition is determined by K2; when FeO and Fe are co-existing
(33.33% ≤ y < 100%), the equilibrium constant will follow K3.
-108-
Equation (3.5) and (3.6) can be solved together to arrive at a relationship between the gas
and the solid conversions (x, y) and the ratio of the solid to gas flow rates. It can be
shown that x is a step function with respect to s as:
x=
K1/(K1+1)
s > 3K1/(K1+1)
s/3
3K1/(K1+1) ≥ s > 3K2/(K2+1)
K2/(K2+1)
3K2/(K2+1) ≥ s > K2/(K2+1)
s
K2/(K2+1) ≥ s > K3/(K3+1)
K3/(K3+1)
K3/(K3+1) ≥ s > K3/3(K3+1)
3s
s ≤ K3/3(K3+1)
(3.7)
y can be obtained from x using Equation (3.7):
y = x / 3s
(3.8)
For a fluidized bed reducer operated at 850 ºC, the solid and gas conversions can be
obtained by substituting the values of K1 – K3 in Equation (3.2) – (3.4) to equation (3.7)
and (3.8):
x=
1
s > 3.0
0.3333/s
s > 3.0
s/3
3.0 ≥ s > 2.34
0.1111
3.0 ≥ s > 2.34
0.78
2.34 ≥ s > 0.78
0.26/s
2.34 ≥ s >0.78
s
0.78 ≥ s > 0.348
0.3333
0.78 ≥ s > 0.348
y=
0.348
0.348 ≥ s > 0.116
0.116/s
0.348 ≥ s > 0.116
3s
s ≤ 0.116
1
s ≤ 0.116
Figure 3.4a shows the relationship between the gas and the solid conversions and the
ratio between the solid and the gas molar flow rates. Figure 3.4a’ shows the relationship
-109-
between the gas and the solid conversions. Since, for a fluidized bed reducer, the gas and
solid conversions at any steady state operation conditions will fall on the curves shown in
Figure 3.4a and Figure 3.4a’, these curves are called operating curves 180. As can be seen,
the conversions of gas and solid are inversely correlated, i.e., a higher solid conversion
corresponds to a lower gas conversion and vice versa. In actual reactor operation, a full
conversion of fuel gas is crucial. Figure 3.4a’ shows a gas conversion of 100%
corresponding to a solids conversion of less than 11.11%.
Aside of the analytical approach, the operating curve for the fluidized bed can be
derived directly from the thermodynamic phase diagram. The rationale being that since
the solid and gas are well mixed and are at the thermodynamic equilibrium, the gas and
solid concentrations should be at the equilibrium concentrations described by the
thermodynamic phase diagram. Thus, the operating curve should coincide with the gassolid equilibrium curve.
The translation of the equilibrium phase diagram into the fluidized bed operating
curve is illustrated in Figure 3.4b. The vertical dashed line describes the relationship
between the phases of iron and the equilibrium gas concentration at a certain temperature
(850 ºC as in the figure). For a fluidized bed reducer, the equilibrium steam concentration
is identical to the hydrogen conversion x in the operating curve. The phases of iron that
equilibrate with such a gas concentration determine the solid conversion y defined by
Equation (3.1). For instance, a solid conversion of 100% corresponds to pure iron while a
solids conversion of 33.33% corresponds to pure FeO. It is noted that the operating curve
-110-
in the ideal case, which is also the equilibrium line, divides the graph into two regions.
Since the gas and the solid conversions will always be lower than or equal to those at
equilibrium in practical reactor operations, the gas and solid conversions will approach
the equilibrium line from its left hand side as the gas and solid contacting time increases.
The thermodynamics dictates that these conversions will not cross the equilibrium line.
Countercurrent Moving Bed Reactor
Contrary to the fluidized bed reactor, a moving bed reactor has the minimal axial
mixing of the gas and solid phases. When a moving bed reactor with a countercurrent
gas-solid contacting pattern is used as the reducer (Figure 3.3b), a fresh syngas feed with
high H2 and CO concentrations will react with iron at lower oxidation states. Meanwhile,
the partially converted syngas with low H2 and CO concentrations will meet iron at
higher oxidation states. Based on the thermodynamic diagram shown in Figure 3.2, such
a contact pattern will maximize both solid and gas conversions. A similar case can also
be expected when a moving bed reactor is used as the oxidizer. For simplicity, with the
mixing for both the solid and the gas phases neglected in the moving bed reactor analysis,
both the gas and solid compositions will vary with the axial position of the moving bed
reactor. From Figure 3.3, the mass balance of oxygen in an infinitesimal layer between z
and z + Δz of the bed reactor at a steady state can be written as
3s( y z + Δz − y z ) = ( x z − x z + Δz )
(3.9)
This is equivalent to
-111-
dx / dy = −3s
(3.10)
Therefore, the relationship between the solid conversion y and the gas conversion x under
a certain solid/gas molar flow rate ratio s, is a straight line with a slope of -3s. Such a line
represents the operating line. The operating line is only restricted by thermodynamic
equilibrium. Specifically, at any point of the operating line, the ratio of the concentration
between steam and hydrogen should not be higher than the equilibrium constant K:
x/(1-x) ≤ K1
for 0 ≤ y < 0.1111
x/(1-x) ≤ K2
for 0.1111 ≤ y < 0.3333
x/(1-x) ≤ K3
for 0.3333 ≤ y < 1
The above equations are equivalent to:
x ≤ K1/(1+K1) for 0 ≤ y < 0.1111
x ≤ K2/(1+K2) for 0.1111 ≤ y < 0.3333
x ≤ K3/(1+K3) for 0.3333 ≤ y < 1
The above restrictions are in conformity with the earlier discussion which concludes that
a practical reducer operating line should locate at the left hand side of the equilibrium
line and should not cross it. Thus, the feasible operating lines can be determined based on
-112-
the mass balance and thermodynamic phase diagram. Two possible operating lines are
shown in Figure 3.5.
It can be shown that each point on the operating line corresponds to the gas/solid
conversions on a certain axial position of the moving bed reactor. Therefore, the
theoretical gas and solid conversions can be obtained from the intercept of the operating
line with the x and y axes. The intercept of the operating line with the y axis corresponds
to the solid conversion at the solids outlet located at the bottom of the reactor where the
highest possible solid conversion is achieved. Similarly, the intercept of the operating line
with x axis is the gas conversion at the gas outlet located at the top of the reactor. For
example, the solid line in Figure 3.5 corresponds to a solid/gas molar flow ratio of 2:3.
Under this operating condition, H2 will be 100% converted and the solid will be reduced
by ~50%. Since the operating line is restricted to the left hand side of the equilibrium
curve, it is not possible to achieve 100% conversions simultaneously for the gas and the
solid. In fact, the solid line corresponds to the maximum achievable solids conversion
when H2 is fully converted. Similarly, to achieve a full solids conversion, at least 92%
excessive H2 over the stoichiometric requirement needs to be introduced to the reactor,
yielding a maximum gas conversion of ~52%.
With Figure 3.5, the optimum gas and solid flow rates for the SCL reducer can be
determined. For example, a full conversion of syngas is essential for the reducer since
incomplete syngas gas conversion will lead to a reduced energy conversion efficiency of
the process. Therefore, the optimum operating line is the solid line in Figure 3.5, which
-113-
corresponds to a solids conversion of ~50%. It can also be shown that multiple-stage
interconnected fluidized bed reactors with countercurrent gas and solid contact pattern
can achieve a conversion similar to that of the moving bed. In contrast, a single-stage
fluidized bed reducer can achieve merely an 11.11% solid conversion under the same
operating conditions. When utilized as an oxidizer, a moving bed is also anticipated to
achieve higher conversions. To generalize, significantly improved gas and solid
conversions can be achieved when a countercurrent moving bed is used as the reducer or
the oxidizer.
3.1.2 ASPEN Plus® Simulation of SCL Reactor Systems
Although the operating lines, as given in Figure 3.5, obtained from the
thermodynamic phase diagrams and the mass balance are useful, they are rather difficult
to be constructed and applied to analyzing the gas and solid conversions, especially when
a gas mixture is involved and the temperature along the axial positions in the reactor is
not constant. Therefore, an alternative method is desired. One viable method is to use
computer simulation method based on such software as the Advanced System for Process
Engineering software, or ASPEN Plus®. The comprehensive physical and thermodynamic
property data banks built in the ASPEN Plus® software render it suitable for reactor and
process simulation. With appropriate modeling parameters, the ASPEN Plus® software is
able to simulate simultaneously the flows of mass, heat, and work in process units and the
process.
-114-
Selection of ASPEN Plus® Modeling Parameters
Before employing the simulation model to analyze a fluidized bed or a moving
bed, a set of common parameters need to be determined. This section describes the
necessary procedures for setting up these parameters. A build-in module in ASPEN,
RGIBBS, is used to determine the equilibrium condition among the reactants and the
various possible products. Other parameters that need to be selected include physical and
thermodynamic property data banks and property methods, stream classes, chemical
components, and calculation algorisms. Selection of appropriate parameters is essential
for the accurate simulation results. Tables 3.2 to 3.4 list the key parameters selected for
the simulations. It is noted that modifications to the physical property data and physical
property methods for the solids are often necessary in order to obtain consistent results
from the literature and from the ASPEN Plus® simulation.
The INORGANIC databank in the ASPEN Plus® software, which determines the
physical and thermodynamic properties of the solids at various conditions, uses Barin
Equation (Equation 3.11) and its CPSXP (a-h) coefficients 181 to obtain the Gibbs Energy
(G), Enthalpy (H), Entropy (S), and Heat Capacity (Cp) for the solids:
G = a + bT + cT ln T + dT 2 + eT 3 + fT 4 + gT −1 + hT −2
H = a − cT − dT 2 − 2eT 3 − 3 fT 4 + 2 gT −1 + 3hT −2
S = −b − c(1 + ln T ) − 2dT − 3eT 2 − 4 fT 3 + gT −2 + 2hT −3
C p = −c − 2dT − 6dT 2 − 12 fT 3 − 2 gT −2 − 6hT −3
-115-
(3.11)
Table 3.4 lists the CPSXP coefficients for iron and its oxides in the INORGANIC
databank in ASPEN Plus®. The HSC® chemistry 5.1 and other literature sources were
used to verify the values of these coefficients
177, 182-184
. In doing so, the reference states
in the literature sources were adjusted to be identical to that specified in ASPEN Plus®,
i.e. 25˚C and 1atm. Minor differences were found in coefficients a, b, and c for Fe3O4 and
FeO (Fe0.947O). The differences amount to ~1% of the original values in the ASPEN
databank. Although the differences are rather small, the simulation results can be
significant varied, particularly with respect to the phase transition conditions of various
iron states.
Fluidized Bed Reactor Model Setup and Simulations
When a fluidized bed reactor is approximated by a continuous stirred tank reactor
(CSTR), a simple RGIBBS module can simulate the fluidized bed operated under the
equilibrium conditions. With the modeling parameter selected, the fluidized bed model
can be set up by connecting reactants and products streams to the RGIBBS module and
by inserting the operating conditions and inlet compositions into the ASPEN flow sheet.
The thermodynamic simulation is performed on the fluidized bed reactor to corroborate
the theoretical analysis results obtained in Section 3.1. To obtain data comparable to
those shown in Figure 3.4, a sensitivity analysis is performed to determine the
relationship between the gas and solid conversions in a fluidized bed by varying the mole
flow rate ratio between Fe2O3 and H2. The operating temperature (850 ˚C) for the
simulation is identical to that in the theoretical analysis. The ASPEN simulation results
-116-
are given in Figure 3.6. As can be seen from the figure, the simulation results are almost
identical to those obtained from theoretical analysis in Figure 3.4a.
Moving Bed Reactor Model Setup and Simulations
A.
Model Setup
The simulation of a countercurrent moving bed reactor is more complex than
fluidized bed simulation. Since no ASPEN Plus® simulator module is available for the
simulation of a countercurrent moving bed reactor operated under equilibrium, a series of
interconnected CSTR reactor based on RIGIBBS module is used to simulate the moving
bed reactor. The model configuration is shown in Figure 3.7. As can be seen in the figure,
the solid entering stage k is the solid product discharged from stage k+1 while the gas
entering stage k is the gaseous product of stage k-1. It can be shown that such a model
configuration satisfies the mass balance and thermodynamic restrictions imposed on
countercurrent moving bed reactor. With a large number of RGIBBS blocks, the
countercurrent moving bed reactor can be approximated.
The simulation of a moving bed system using an infinite number of RGIBBS
blocks is not feasible. However, the simulation results show asymptotic behavior with
increasing number of RGIBBS blocks. It is found, based on numerous case studies, that a
5-stage model configuration can simulate the countercurrent moving bed with good
accuracy. This is verified by comparing results obtained from a 5-stage model with those
obtained from a 6–stage model. The comparisons indicate that the two models are
-117-
identical in all cases. Therefore, the 5-stage RGIBBS model shown in Figure 3.8 is used
to simulate the countercurrent moving bed reactor.
B. Case 1: Moving Bed with 100% H2 Conversion and Maximized Solids Conversion
In this case, pure H2 is the reducing gas. The goal for the case is to validate the
(solid) operating line shown in Figure 3.5. The Fe2O3 to H2 molar flow rate ratio s is set
to be 2:3. The reactor is operated at 850 ˚C. These conditions are identical to those
denoted on the solid operating line in Figure 3.5, which shows a maximum solid
conversion with near complete conversion of gas. Figure 3.9 presents an accumulative
gas/solid conversion along the five reaction stages of the reactor. The results indicate that
the solids conversion is 49.98% and hydrogen conversion is 99.95%, which are consistent
with the outcomes from the theoretical analysis in Section 3.1. The simulation results also
indicate that the gas and the solid conversions are irrelevant to the operating pressure of
the reactor.
C.
Case 2: Moving bed with 100% Syngas Conversion
The goal for the syngas chemical looping process is to convert 100% gaseous fuel
to hydrogen and/or electricity. The effect of varying the iron oxide flow rate over the
syngas gas (H2:CO = 1:2) conversion in a reducer operated at 30 atm and 900˚C is
investigated. Figure 3.10 shows the syngas and solids conversions under different solid
gas molar flow rate ratios with and without Fe3C formation. As can be seen from the
figure, without Fe3C formation, the solid to gas molar flow rate ratio should be more than
-118-
0.66 in order to convert all the syngas. This result corresponds to a solid conversion of
~50.0%. Figure 3.11 shows the gas and solid conversions under such reaction conditions.
When Fe3C formation is considered, the minimum solid/gas ratio, s, could drop to
0.58 and the maximum solids conversion is 46.7%. However, due to the slow reaction
kinetics the formation of Fe3C is seldom observed, especially with the presence of steam
185, 186
. The simulation results were corroborated by the experimental results.
Further applications
Based on the multistage ASPEN modeling, additional simulation conditions,
results and analyses are presented below.
A. Effect of Temperature
From the equilibrium considerations, the higher temperature favors the
endothermic reaction and the lower temperature favors the exothermic reaction. The
reaction between Fe2O3 and CO is exothermic while the reaction between Fe2O3 and H2 is
endothermic. Figure 3.12 shows the effect of temperature on the syngas conversion in a
countercurrent moving bed reactor. The reactor is operated at 30 atm with stoichimetric
amount of gaseous and solid reactants. This figure indicates that the equilibrium
conversions of CO and H2 show opposite trends. Except for a significant increase at
~580˚C, the overall syngas conversion slowly decreases when the temperature increases.
The inflexion at ~580˚C is caused by the emergence of Wustite phase, which does not
exist below 550˚C. Since the decrease in overall syngas conversion is relatively slow at
-119-
temperatures above 600˚C, the optimum operating temperature range for the reducer is
determined to be 700 to 900˚C due to the fast reaction kinetics at higher temperatures.
B.
Fates of Sulfur and Mercury
The pollutant control is essential for any coal conversion processes. H2S, COS
and mercury are important pollutants that present in coal derived syngas. The multistage
model can assist in determining the fates of these pollutants. The ASPEN simulation is
conducted to examine the relationship between the sulfur content in the syngas and the
formation of iron-sulfur compounds. The potential compounds considered include S,
COS, SO2, H2S, FeS, Fe0.877S. At 900˚C, 30 atm with s of 0.66 (2:3), simulation results
indicate that Fe0.877S is the only sulfur compound that may form in the reducer. As shown
in Figure 3.13, H2S will exit from the moving bed reducer without reacting with the
Fe2O3 particles unless the H2S level in syngas is higher than 600 ppm. Similarly, unless
the COS level exceeds 650 ppm, COS will exit from the moving bed reducer without
reacting with the Fe2O3 particles. The practical implication of these simulation results is
significant: with the absence of sulfur attachment to the solids, the hydrogen product
stream from the oxidizer will be sulfur free. Thus, the sulfur control strategy for the SCL
process is simplified. Specifically, a hot gas cleanup unit (HGCU) is installed at the
upstream of the reducer for the bulk sulfur removal. Since the available high temperature
sorbent can reduce the sulfur level in raw syngas to below 50 ppm with ease
187
, the
remaining sulfur in the syngas will exit from the reducer along with CO2 and H2O. After
condensing the steam, the sulfur containing CO2 will be ready for geological
sequestration
188
. Such a process arrangement avoids the energy intensive solvent based
-120-
H2S stripping process. Cooling and reheating of the syngas can also be avoided,
rendering a more efficient sulfur control scheme.
Elemental mercury, the major form of mercury that presents in the raw syngas
derived from coal, is usually captured using activated carbon bed under a low
temperature in conventional coal gasification processes. From the ASPEN simulation,
mercury will not react with any substances that are present in the reducer. Thus, all the
mercury in the syngas stream will exit from the reducer flue gas and will not be present in
the hydrogen stream from the oxidizer. The mercury containing flue gas from the
oxidizer can be treated before sequestration using activated carbon. The mercury
separation in this manner is more efficient than does the traditional process which
involves cooling and reheating of the syngas.
Thermodynamic
analyses
exemplified
in
this
section
indicate
that
a
countercurrent moving bed delivers best overall performance for both the reducer and the
oxidizer operations. The thermodynamic models constructed using the ASPEN
simulation are used to conveniently determine the optimum operating conditions for the
SCL reactors as well as pollutant control strategies.
3.2 Syngas Chemical Looping (SCL) Process Testing
As noted, a countercurrent moving bed with solid particles flowing downward and
gas flowing upward represents an effective gas-solid contact mode for the operations of
both the SCL reducer and the oxidizer. The concept of the countercurrent moving bed can
also be realized using a series of fluidized beds with countercurrent flows of the gas and
-121-
solids
189
. For illustration, the moving bed configuration is used to characterize the
looping reactor operation. This section highlights the experiments that aim to validate the
proposed SCL concept.
3.2.1 Experimental
Bench Scale Reactor
A bench moving bed reactor setup a maximum capacity of 2.5 kWth (kilo-watts
thermal) was used to test the SCL reactor operations 138. The reactor setup is illustrated in
Figure 3.14. The moving bed is a scale up version of that described by Gupta et al
138
.
The reactor assembly consists of heated reaction zone, solids holding funnels, screw
feeder and solids flow controller, bed height control system, solids sampling ports, gas
flow panels, gas sampling ports, gas delivery and handling system, gas analysis system,
and computer control system 190. The material used for construction is primarily stainless
steel type SS-304, which provides good inert characteristics for high temperature
reactions. The reactor was designed to handle solid flow rates up to 82 g/min (4.9 kg/hr)
and gas flow rates up to 200 ml/s (12 l/min). The heated reaction section of the reactor
has an I.D. (inner diameter) of 1.6 inches and a bed height of 40 inches. In a typical
experiment, solid reactants such as oxygen carrier particle are first loaded to the top
funnel and then moved downwards steadily by the screw conveyor system with the bed
height is maintained by the bed height sensor system. Reactant gases are introduced from
the gas inlet located at the bottom of the moving bed to react with the solids in a
countercurrent manner. The gas composition along the axial positions of the reactor is
-122-
constantly monitored using a Varian CP-4900 micro GC. The solid samples can be taken
from the solid sampling ports after the experiment for further characterization. The
moving bed setup mimics the gas-solid contacting pattern in the proposed reducer and
oxidizer operation. The bench scale system can be operated as either the reducer or the
oxidizer in a “semi-continuous” mode.
Reducer Experiment Procedure
A. Reducer Experiment with Syngas
The goal of this experiment is to demonstrate the continuous operation of the
reducer. Before the experiment, 14 kg of composite pellets with 60% Fe2O3 and 40%
inert support was loaded to the top funnel. The solid flow rate was then adjusted and
calibrated. The solid flow rate in this experiment was 12.87 g/min. After solid flow rate
calibration, the reactor was sealed and flushed overnight using N2 at a flow rate of 500
ml/min.
Prior to heating up the reactor, gas samples from various sections of the reactor
were analyzed using a Varian CP-4900 MicroGC to confirm that the oxygen content in
the reactor is low. Pressurized helium leak test was also conducted to examine potential
gas leakage. After the oxygen test and leak test, the solid transport system was activated
to continuously move the particles down towards the bottom funnel. Meanwhile, the
reactor was gradually heated up to around 900 °C. Once steady solid flow was achieved,
simulated syngas that consists of CO, H2, and CO2 was introduced into the bottom of the
reactor. The composition of the simulated syngas is similar to that from a Shell gasifier.
-123-
In this experiment, N2 was also introduced along with the simulated syngas as an internal
standard. The flow rates of the gas mixture used in the experiment are shown in Table 1.
After the reactant gas was injected, the gas compositions at the various positions
of the reactor were constantly monitored and sampled using the MicroGC until the
establishment of steady state operation. The steady state operation is usually achieved
after one solid residence time and can be determined by the stabilization of gas
composition, i.e. the composition of the gas sample at a certain position of the reactor no
longer fluctuates over time. After reaching the steady state, the gas composition along the
reactor was recorded. The steady state was maintained for more than 13 hours while the
reactor was operated under hot condition for more than 15 hours. During the course of the
experiment, slight increase in the syngas flow rate was attempted; however, notable
increase in CO and H2 concentrations were identified at the reactor outlet. Such
phenomena imply that the current solid flow rate of 12.87 g/min will not be able to fully
convert syngas with a flow rate higher than that shown in Table 3.5.
After the experiment was finished, the reactant gas, the solid transport system,
and the heating elements of the reactor were simultaneously switched off. The reactor
was flushed with N2 until it was cooled down to room temperature. The solid samples at
various positions along the reactor were taken from the solid sampling ports. A portion of
these samples were then characterized using a CM-120 total carbon analyzer (UIC Inc.)
for carbon content. The remaining solid samples were oxidized in a TGA using air. The
-124-
solid conversions were then determined using the weight change obtained in the TGA
coupled with the carbon content characterized by the carbon analyzer.
Since the Fe2O3 composite particle contains support in this case, its conversion (y)
is defined as:
y=
nˆ O nˆ Fe − nO n Fe
× 100%
nO n Fe
(3.12)
Here nO n Fe corresponds to the molar ratio between oxygen atom and iron atom in Fe2O3
and nˆ O nˆ Fe corresponds to the molar ratio between oxygen atom and iron atom in the
converted solid product, i.e. FeOx (0 < x < 1.5). The support material is treated as an inert.
Therefore, the oxygen atoms associated with inert ingredient are not included in Equation
3.12. For instance, the reduction of Fe2O3 to Fe3O4 corresponds to a particle conversion
of (3/2-4/3)/(3/2)×100% = 11.11%, FeO corresponds to a particle conversion of 33.33%
and Fe corresponds to 100% particle conversion.
The particle conversion defined in equation 1 can be calculated through the
following equation:
y=
W1 − W0 (1 − c%)
× 100%
0.3 × W1 × L%
(3.13)
-125-
Here W0 denotes the weight of the converted sample before oxidized by air (in TGA). W1
represents the weight of the sample after air oxidation. c designates the percentage of
carbon content in the converted sample obtained from the carbon analyzer. L represents
the percentage of iron oxide in the composite particle.
B. Reducer Experiment with Methane
Iron oxide is active in cracking the coal volatiles to methane
hydrocarbon up to 1030ºC
192
191
, the most stable
. Experiments carried out in a fixed bed reactor confirmed
that the Fe2O3 composite particles can convert more than 90% of the ethylene and
propane to methane with the presence of steam and hydrogen. Therefore, if the reducer
can fully convert methane to CO2 and steam using the composite particle, other
hydrocarbons can also be fully converted with ease. The experimental conditions for
methane conversions are similar to those for syngas conversions. The Fe2O3 composite
particle flow rate was 11.08 g/min and the methane flow rate was 360 ml/min. H2 was
also introduced to the reactor at a flow rate of 133 ml/min to enhance the methane
conversion. The average reactor temperature was 930 ºC. Continuous operation was
maintained for more than 10 hours.
Oxidizer Experiment Procedure
Particles reduced by syngas are circulated to the oxidizer where they react with
steam to generate hydrogen. Specifically, the particles were introduced from the top of
-126-
the oxidizer, which is of a same geometric dimension as the reducer, at a rate of 13 g/min.
Steam was injected at the bottom of the oxidizer at a rate of 1.355 g/min. Continuous
operation was maintained for more than 10 hours.
Combustor Performance
The design of an entrained flow combustor is less challenging when compared to
the reducer and the oxidizer due to much favored kinetics and thermodynamics for this
oxidation reaction. The thermal stability of the composite particles under high operating
temperatures in the combustor (~ 1200 ºC), however, needs to be ascertained. An
externally heated quartz tube was used to demonstrate the combustor performance. The
quartz tube has an outer diameter of 20 mm and an inner diameter of 18 mm. 20 grams of
composite particles from the oxidizer experiment was loaded into the quartz tube. A type
K thermocouple was inserted into the particles to monitor the temperature. Nitrogen was
then introduced to the tube at a flow rate of 200 ml/min before the tube and the particles
were gradually heated up to 1100 ºC. Once the temperature of the particles stabilizes at
1100 ºC, oxygen was introduced to the tube at 50 ml/min to mimic a total air flow rate of
250 ml/min. The reactor temperature was monitored throughout the experiment. The
oxygen injection was stopped after the particle temperature stabilizes. After the reactor is
cooled down under nitrogen environment, the conversion, reactivity, and recyclability of
the particle were characterized using the methods discussed in Chapter 2.
3.2.2 Results and Discussions
Reducer Experiments
A. Syngas Experiment
-127-
Figure 3.15 shows the gas and solid conversion along the axial positions of the
reactor. The carbon depositions along the reactor are shown in Table 3.6 while the outlet
gas compositions are shown in Table 3.7.
As can be seen in Table 3.6 and Figure 3.15, solid was steadily converted as it
moves towards the solid outlet located at the bottom of the reactor. The iron oxide
conversion at the reactor outlet was 49.5%, which corresponds to a mixture of FeO
(74.6%) and Fe (25.4%). Such a conversion match well to the simulation results obtained
from ASPEN Plus® model reported in the previous section. Throughout the reactor,
carbon deposition was not significant owing to the relatively high operating temperature
and the presence of CO2 114. The carbon deposition can be further inhibited by increasing
the CO2 concentration in the syngas.
The syngas conversion profile shows an opposite trend when compare to the solid
conversion profile due to the countercurrent contacting pattern between solid and gas. As
can be seen in Table 3.7 and Figure 3.15, the syngas was nearly completely converted
before exiting from the moving bed. The outlet gas is comprised mainly of CO2 (dry, N2
free basis). Another trend that can be observed from the moving bed experiment is that
the gas and solid conversions are faster at the gas outlet/solid inlet when compared to the
gas inlet/solid outlet. This is due to the faster reduction kinetics for the conversion of
Fe2O3 to Fe3O4 when compared to the conversion of Fe3O4 to FeO and FeO to Fe.
Moreover, little or no change in gas and solid conversions are observed in the middle part
of the reactor. This is mainly due to the gas and solid compositions in this intermediate
-128-
region are close to equilibrium. Such a phenomenon suggests that the solid residence time
can be further shortened without affecting the reducer performance. The conversion
profile of the gas is generally consistent to that of the solid. The minor fluctuation in solid
conversion along the axial position of the reactor may result from the randomness of solid
sampling since the amount of sample that can be extracted from the reactor is relatively
small when compared to the size of the moving bed.
B. Syngas Experiment
Figure 3.16 shows the gas and solids conversion profiles under the steady state
operation.. It is seen that > 99.8% methane is converted to CO2 and H2O and the Fe2O3
particle is reduced by 49%. The ability of SCL process in the methane conversion renders
its capability in converting a variety of other hydrocarbon fuels since, as mentioned
previously, methane is a most stable hydrocarbon.
Oxidizer Experiments
The gas and solid analysis shows that more than 99.3% of the reduced particle
was converted into Fe3O4 in the small preheating zone located above the reaction zone.
Since the preheating zone was maintained at ~580 ºC, the results indicate that the reaction
between steam and reduced Fe2O3 composite particles is fast even under a relatively low
temperature.
The purity of the hydrogen produced at the outlet of the oxidizer indicates a trace
of CO. From Figure 3.17, it is seen that the normalized H2 concentration (moisture and
-129-
N2 free basis) exceeds 99.95%. Further increase in the hydrogen purity can be realized by
inhibiting carbon deposition during the reducer operation. The results suggest the purity
of hydrogen produced from SCL process is such that it can be directly used for ammonia
synthesis and oil refining applications. Clearly, when hydrogen is to be used in the PEM
fuel cell application, further purification of the hydrogen product stream from the
oxidizer using purification techniques such as pressure swing adsorption (PSA) will be
necessary.
Combustor Experiments
After the injection of oxygen, a dramatic increase in particle temperature was
identified. The maximum particle temperature slightly exceeds 1200 ºC. The particle was
found to be fully oxidized after the high temperature oxidation; however, the color of the
high Fe2O3 composite particle changed from red to light gray. The reactivity and
recyclability of the particle after high temperature oxidation was found to be similar to
that of the fresh particles. The particle performance after extended high temperature
combustion operation, however, needs to be further assured.
As can be seen from the discussions in this section, experimental studies in the
bench reactor corroborate the reducer and the oxidizer performances predicted by the
thermodynamic analysis, validating the proposed SCL process concepts.
3.3 Process Simulation of the Traditional Gasification Processes and the
-130-
Syngas Chemical Looping Processes
In this section, ASPEN simulation models are used to evaluate the overall
performance of the energy conversion processes. The processes exemplified in this
section include IGCC process using the GE High Efficiency Quench (GE-HEQ) gasifier,
conventional coal to hydrogen process using the Shell SCGP gasifier, and the SCL
process using Shell SCGP gasifier.
3.3.1 Common Assumptions and Model Setup
In order to compare the performance of the conventional processes and the SCL
process, a common set of assumptions and modeling parameters is defined as given
below:
1) The CO2 capture in the process is considered to be at least 90%.
2) The ambient temperature is 25 °C and the ambient pressure is 1 bar.
3) A feeding rate of 132.9 ton/hr of Illinois #6 coal is used (approximately 1000 MW
in HHV) and the properties of the Illinois #6 coal is shown in Table 3.8.
4) The GE-HEQ gasifier is used for the IGCC process and the Shell gasifier with gas
quench configuration is used for both the conventional coal to hydrogen process
and the SCL process.
5) Air consists of 21% O2 and 79% N2 by volume.
-131-
6) The solids circulating in SCL consist of 70% Fe2O3, 15% TiO2, and 15% Al2O3,
by weight.
7) The H2 product is compressed to 3 MPa for subsequent transportation.
8) CO2 is compressed to 15 MPa for sequestration.
9) The carbon conversion in the gasifier is 99% and the heat loss in the gasifier is
0.6% of the HHV of coal.
10) The pressure level in a steam cycle is 124/30/2 MPa, and the HP and IP steam is
superheated up to 550 °C, while the temperature of the flue gas in the stack is
130°C.
11) All the compressors are designed using four stages and the outlet temperature of
the intercooler is 40°C.
12) The temperature is constant during the individual unit operation.
13) The mechanical efficiency of pressure changers such as compressors and
expanders is 1 while their isentropic efficiency is 0.8~0.9.
14) The gas and solid conversions in the SCL system are based on experimental
results.
15) The CO2 capture cost is 106 kWh per ton of CO2 for traditional processes (using
SELEXOL process)193.
16) The inlet firing temperature of the gas turbine is 1250 ºC.
To accurately simulate the individual unit in the flow sheet, appropriate ASPEN Plus®
model(s) for each unit is determined. These models are listed in Table 3.9.
-132-
3.3.2 Description of Various Systems
IGCC Process using GE-HEQ Gasifier
The IGCC system illustrated in this case study uses a GE/Texaco slurry-feed,
entrained flow gasifier with total water quench syngas cooler. The flow diagram of the
process is shown in Figure 3.18, which is similar to that described in Chapter 1. In this
process, coal is first pulverized and mixed with water to form coal slurry. The coal slurry
is then pressurized and introduced to the gasifier to be partially oxidized at 1500 °C and
30 atm. The high temperature raw syngas after gasification is then quenched to 250 °C
with water. The quenching step solidifies the ash. Moreover, most of the NH3 and HCl in
the syngas are removed during this step. After quenching, the syngas is sent to a venturi
scrubber for further particulate removal. The particulate-free syngas, saturated with steam,
is then introduced to the sour WGS unit. The syngas exiting from the WGS unit contains
mainly of H2 and CO2 with small amount of CO, H2S, and mercury. This gas stream is
then cooled down to 40 °C and passed through an activated carbon bed for mercury
removal. The CO2 and H2S in the syngas are then removed using an MDEA unit,
resulting in a concentrated hydrogen stream with small amounts of CO2 and CO. The
hydrogen rich gas stream is then compressed, preheated, and combusted in a combined
cycle system for power generation. The combined cycle system consists of a gas turbine
and a two stage steam turbine. The CO2 obtained from the MDEA unit is compressed to
150 atm for sequestration.
Conventional Coal to Hydrogen Process
-133-
The coal to hydrogen system illustrated in this case study uses a Shell SCGP dryfeed, entrained flow gasifier with gas quench configuration. The flow diagram of the
process is shown in Figure 3.19. In this process, coal is first pulverized and dried. The
coal powder is then pressurized in the lock hopper and introduced to the gasifier to be
partially oxidized at 1500 °C and 30 atm. The high temperature raw syngas after
gasification is then quenched to 900 °C with low temperature syngas from particulate
removal unit. The quenched syngas is then introduced to a syngas cooler. After being
further cooled to ~300°C, the syngas is introduced to a particulate removal unit where the
entrained solids are removed from the syngas. The syngas is then saturated with steam
and enters the sour WGS unit. In the WGS unit nearly 96% of the CO is converted to H2.
This gas stream then flows to an activated carbon bed for mercury removal. Subsequently,
acid gases, H2S and CO2, in the raw hydrogen stream are removed using such solvents as
MDEA. Further purification of hydrogen is made via PSA unit. The tail gas from PSA is
combusted for electricity generation.
Syngas Chemical Looping Process
In order to obtain data directly comparable to conventional processes, the SCL
system illustrated here also uses a Shell SCGP dry-feed, entrained flow gasifier with gas
quench configuration. The flow diagram of the process is given in Figure 3.20. The SCL
system shown in Figure 3.20 uses a syngas generation, quenching and cooling system
identical to that used in the convention coal to hydrogen case. However, the syngas is
only cooled down to 550 °C. The cooled syngas is introduced to a hot gas cleanup unit to
-134-
strip the sulfur level down to 50 ppm. The low sulfur syngas is then introduced to the
SCL system for hydrogen and power co-generation.
3.3.3 ASPEN Plus® Simulation, Results, and Analyses
Based on the parameters and process configurations described in Sections 3.3.1
and 3.3.2, the ASPEN simulation is conducted based on the ASPEN flow sheets shown in
Figure 3.21 for the three energy conversion processes described above. The Aspen Plus®
has a comprehensive physical property database. Therefore, most of the chemical species
involved in the process can be selected directly from this database. The nonconventional
components such as coal and ash are specified conveniently using general coal enthalpy
modulus embedded in ASPEN software. After the chemical species in the process are
defined, the related physical property methods are selected among various simulator’s
choices. In this simulation, the global property method used is PR-BM, the local property
methods are specified whenever required.
The ASPEN model is finalized by establishing detailed operating parameters
based on the operating conditions and design specifications of the individual unit. The
units are then connected in the same arrangement as shown in the flow sheet. An
appropriate convergence setting is determined to ensure accurate simulation results.
Table 3.10 compares the simulation results among the three systems. The case when all
-135-
the hydrogen generated in the SCL process is used for electricity generation is also
investigated.
As can be seen in Table 3.10, the syngas chemical looping process is significantly
more efficient when compared to conventional processes, especially under a carbon
constrained situation. The advantage in the SCL process results from improved energy
conversion scheme coupled with the integrated CO2 capture capability.
3.4 Conclusions
The syngas chemical looping (SCL) process generates H2 from syngas through
reduction and oxidation reactions in a reducer, oxidizer and combustor. The
countercurrent moving bed for gas-solid contact is revealed to offer the optimum reactor
operation mode for the reducer and the oxidizer in light of the reactant and product
thermodynamic properties. The operation of the key SCL reactors has been validated to
be feasible under bench scale. The experimental results match well to the simulation
outcomes. The thermodynamic and process simulations further show that the overall
energy conversion scheme and control strategy of the pollutants such as H2S, COS, and
CO2 in the SCL process can be significantly and effectively simplified over those in the
traditional process, resulting in notably increased energy conversion efficiency.
-136-
Iron Phase
Fe2O3/Fe3O4
mixture
Fe3O4/FeO
mixture
FeO/Fe mixture
Iron Phase
Fe2O3/Fe3O4
mixture
Fe3O4/FeO
mixture
FeO/Fe mixture
Equilibrium Concentration of Equilibrium Concentration
of CO
CO2
99.9955%
45 ppmv
80.3%
19.7%
38.0%
62.0%
Equilibrium Concentration of Equilibrium Concentration
of H2
H2O
99.995%
50 ppmv
78.0%
22.0%
34.8%
65.2%
Table 3.1 Equilibrium Gas Compositions with Different Oxidization States of Iron at
850ºC
-137-
Name of the Parameter
Parameter Setting
Reactor Module
RGIBBS
Physical and Thermodynamic Databanks
COMBUST, INORGANIC, SOLIDS and PURE
Stream Class
MIXCISLD
Chemical Components
Listed in Table 3.3
Property Method (for Gas and Liquid)
PR-BM*
Calculation Algorism
Sequential Modular (SM)
* Property methods for solids are discussed separately and some correlative parameters
are presented in Table 3.4
Table 3.2 Parameters for the ASPEN Plus® Model
-138-
Component ID
Stream Type
Component name
Formula
CO
CONV
CARBON-MONOXIDE
CO
CO2
CONV
CARBON-DIOXIDE
CO2
H2
CONV
HYDROGEN
H2
H2O
CONV
WATER
H2O
FE2O3
SOLID
HEMATITE
FE2O3
FE3O4
SOLID
MAGNETITE
FE3O4
FEO
SOLID
WUESTITE
FEO
FE0.947O
SOLID
WUESTITE
FE0.947O
FE
SOLID
IRON
FE
C
SOLID
CARBON-GRAPHITE
C
FE3C
SOLID
TRIIRON-CARBIDE
FE3C
HG
CONV
MERCURY
HG
HGS
CONV
MERCURY-SULFIDE-RED
HGS
S
CONV
SULFUR
S
H2S
CONV
HYDROGEN-SULFIDE
H2S
FES
SOLID
IRON-MONOSULFIDE
FES
FE0.877S
SOLID
PYRRHOTITE
FE0.877S
#
Species such as FeS2, HgO do not exist in the interested temperature range; therefore,
they are not included.
Table 3.3 Components List in Reducer Simulation#
-139-
Components
Temperature
units
Property units
T1
T2
a
a’
b
b’
c
c’
d
e
f
g
h
’ The revised values
FE2O3
FE3O4
FE
FE0.947O
˚C
˚C
˚C
˚C
J/kmol
25
686.85
-9.28E+08
-9.28E+08
1.98E+06
1.98E+06
-2.58E+05
1.98E+06
165.486384
0.066806967
1.17E-05
7.66E+09
-3.76E+11
J/kmol
576.8500000
1596.850000
-9.7072850E+8
-9.5672850E+8
5.27383876E+5
5.355839E+05
-50171.18100
-5.089700E+04
-35.96733770
J/kmol
25
626.85
3.78E+07
3.78E+07
-6.54E+05
-6.54E+05
1.09E+05
-6.54E+05
-214.129205
J/kmol
25.00000000
1376.850000
-2.8212753E+8
-2.81844E+8
4.01635664E+5
4.029657E+05
-4.878544E+04
-4.860400E+04
-4.184000020
-6.0151695E-5
0.084705631
0.0
6.12900216E-9
-4.277784E+10
5.46763727E+9
-1.95E-05
-4.01E+09
1.98E+11
0.0
1.40164001E+8
0.0
Table 3.4 Parameters in the ORGANIC Databank in ASPEN Plus®
-140-
Gas Type
Flow Rate (ml/min)
Composition (mol %)
CO
990.0
43.77
N2
498.1
22.02
H2
660.3
29.19
CO2
113.6
5.02
Table 3.5 Gas flow to the bench scale unit
Distance from the
Reactor Bottom (cm)
Carbon percentage
(wt %)
Solid Conversion (%)
88.75
81.25
0 0.0283
0 25.460
63.75
46.25
28.75
0.0750 0.092
22.859 28.364
0.024
25.202
Table 3.6 Solid conversions and carbon depositions along the reactor
Gas Type
Percentage (%)
CO
0 - 0.41
H2
0.001 – 0.1
Table 3.7 Gas composition at the reactor outlet over three hours
-141-
CO2
99.5 – 99.999
11.25
0
0.114 0.061
32.11 49.50
Proximate
Analysis
Moisture
Fixed Carbon
Volatiles
Ash
HHV (MJ/kg)
Wt%
(AsReceived) Wt%, dry
11.12
44.19
49.72
34.99
39.37
9.7
10.91
100
100
27.13511
Wt%
(AsWt%,
Received) dry
Ultimate
Moisture
11.12
ASH
9.7
10.91
CARBON
63.75
71.72
HYDROGEN
4.5
5.06
NITROGEN
1.25
1.41
CHLORINE
0.29
0.33
2.51
2.82
30.53107 SULFUR
OXYGEN
6.88
7.75
Table 3.8 Physical and Chemical Properties of Pittsburgh #6 Coal 194
-142-
Unit Operation
Air Separation Unit
Coal Decomposition
Coal Gasification
Quench
WGS
MDEA
Burner
HRSG
Gas Compressors
Heater and Cooler
Turbine
Aspen Plus® Model Comments / Specifications
Sep
Energy consumption of the ASU is based on
specifications of commercial ASU/compressors load.
Ryield
Virtually decompose coal to various components
(Pre-requisite step for gasification modeling)
Rgibbs
Thermodynamic modeling of gasification
Flash2
Phase equilibrium calculation for cooling
Rstoic or Rgibbs
Simulation of conversion of WGS reaction based on
either WGS design specifications or thermodynamics
Sep or Radfrac
Simulation of acid gas removal based on design
specifications
Rgibbs or Rstoic
Modeling of H2/syngas combustion step
MHeatX
Modeling of heat exchanging among multiple
streams
Compr or Mcompr Evaluation of power consumption for gas
compression
Heater
Simulation of heat exchange for syngas cooling and
preheating
Compr
Calculation of power produced from gas turbine and
steam turbine
Table 3.9 ASPEN Models for the Key Units in the IGCC Process
-143-
Coal feed(ton/hr)
Carbon
Capture(%)
Hydrogen(ton/hr)
Net Power(MW)
Efficiency(%HHV)
IGCC
Process
SCL Process
Electricity
132.9
132.9
Conventional Coal
to Hydrogen
Process
132.9
90
100
90
100
0
348.1
34.8
0
422.0
42.2
14.4
57.6
62.7
15.6
57.4
66.5
Table 3.10 Comparisons of the Process Analysis Results
-144-
SCL Process
132.9
CO2, H2O, CO, H2
H2, H2O
H2O
FeOy (y<x)
Oxidizer
FeOx
Reducer
CO, H2
FeOz (z≤x)
Spent Air, Heat
FeOx
Combustor
(Optional)
FeOz (z≤x)
Air
Figure 3.1 Schematic Flow Diagram of Iron Based Chemical Looping Processes
-145-
1
0
0.8
1
0.2
0.8
0.4
0.6
0
Fe
0.2
0.6
Fe3O4
0.2
0.8
0.4
0.2
Fe2O3
0
1
200
400
600
800
1000
0.6
Fe3O4
0.8
Fe2O3
0
1200
1
200
Temperature, oC
a
0.4
FeO
H 2O %
FeO
0.4
H2 %
0.6
CO2 %
CO %
Fe
400
600
800
1000
1200
Temperature, oC
b
Figure 3.2 Equilibrium Phase Diagrams of a) Iron-Carbon-Oxygen System b) IronHydrogen-Oxygen System
-146-
FeOx
CO2/H2O
FeOx
CO2/H2O
z
z+Δz
FeOy
a
(x>y)
FeOy
CO/H2
b
(x>y)
CO/H2
Figure 3.3 Gas-Solid Contacting Pattern of the Reducer Using a) a Fluidized Bed
Design; b) a Moving Bed Design
-147-
1
b
0.8
0.7
S o lid C o n v e r s io n
C onversion
0.8
1
a
Solid Conversion
Gas Conversion
0.9
0.6
0.6
0.5
0.4
0.4
0.3
0.2
0.2
0
0.1
0
0
0.5
1
1.5
2
2.5
3
3.5
Ratio of Solid/Gas Molar Flow Rate
1
FeO
0.4
0
0.1
0.2
1
0.3
0.8
0.4
0.6
0.5
0.4
0.6
0.2
0.7
0
0.8
1
0
0.9
Fe3O4
0.2
o
Temperature, C
Fe3O4
0.6
200 300 400 500 600 700 800 900 1000 1100 1200
1
0.9
0.8
0.7
0.6
0.5
0.4
0.2
0.1
0.3
FeO
Fe
H2 %
S o lid C o n vers io n
0.8
0.8
Gas Conversion
aI
Fe
0.6
H2O Content
0
1
0.4
H2O %
0
0.2
H2 Content
Gas Conversion
Figure 3.4 Operating Curves for a Fluidized Bed Reactor: a. Solid and Gas Conversion
Versus Solid/Gas Molar Flow Rate Ratio; a’. Relationship between Solid Conversion and
Gas Conversion; b. Derivation of Operating Curve from Thermodynamic Phase Diagram
-148-
1
0.9
S o lid C o n versio n
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0
0.2
0.4
0.6
0.8
1
Gas Conversion
Figure 3.5 Operating Lines in a Countercurrent Moving Bed Reactor under 850ºC
-149-
Figure 3.6 Operating Curve of a Fluidized Bed Reducer at 850 oC
-150-
Figure 3.7 Arrangement of Multi-Stage Fluidized Bed System for the Simulation of a
Moving Bed
-151-
Figure 3.8 5-Stage RGIBBS Model for Moving Bed Simulations
-152-
Figure 3.9 The Gas and the Solids Conversions in a Countercurrent Moving Bed Reactor
as Described in Case 1. Operating Conditions: Temperature 850 ºC; Pressure 1 – 30 atm;
Reducing Gas H2
-153-
Figure 3.10 Relationship between the Gas and Solids Conversions and Solid to Gas
Molar Flow Rate Ratio with (Gas1 and Solid1) and without (Gas and Solid)
Considering Fe3C Formation in a Countercurrent Moving Bed Reactor as Described
in Case 2. Operating Conditions: Temperature 900 ºC; Pressure 30 atm; Syngas
Composition CO 66.6%, H2 33.3%
-154-
Figure 3.11 Gas and Solid Conversion Profiles in a Countercurrent Moving Bed Reactor
Operating Conditions: Temperature 900 C; Pressure 30 atm; Solid to Gas Molar Flow
Rate Ratio 0.66; Syngas Composition CO 66.6%, H2 33.3%
-155-
Figure 3.12 Effect of Temperature on the Conversions of Syngas, CO, and H2 in a
Countercurrent Moving Bed Reactor. Operating Conditions: Pressure 30 atm, Solid to
Gas Molar Flow Rate Ratio 0.33
-156-
Figure 3.13 Relationship between the Fe0.877S Formation and the Syngas H2S Level in a
Countercurrent Moving Bed Reactor. Operating Conditions: Temperature 900 ºC;
Pressure 30 atm; Solid to Gas Molar Flow Rate Ratio 0.66
-157-
a
Figure 3.14 Bench Scale Demonstration Unit for SCL Process a. Schematic Flow
Diagram of the Unit; b. Picture of the Unit. Reactor Parts Shown in a: (1) CO
Pretreatment (2) 3-Way Safety Valve (3) Cocurrent/Countercurrent Flow Selector
Valve (4) Reactor Gas-Out Miniature Vacuum Pump (5) Gas Sample Miniature
Vacuum Pump (6)Sample Port Selector Valve (7) Needle Valve and Bubble Flow
Meter (8) Light Source (9) Top Solid Holding Bin (10) Light Photocell (11)
Reactor Gas Out Port/Line (12) Gas and Solid Sample Ports (13) Flanges (14)
Heating Coils (15) Thermocouples (16) Reactor Gas in Port/Line (17) Bottom
Screw Conveyor (18) Bottoms Solid Collection Container (19) Computer Control
and Data Logging (20) Emergency Shut Off Valve
-158-
b
100
50
Solid Conversion (%)
H2
CO
90
40
80
35
70
30
60
25
50
20
40
15
30
10
20
5
10
0
Gas Conversions (%)
Solid
45
0
0
10
20
30
40
50
60
Axil Position (cm)
70
80
90
Figure 3.15 Gas and Solid Conversions in the Reducer Experiment (0 cm Corresponds to
Gas Inlet/Solid Outlet)
-159-
1.0
Solid (meassured)
Solid (Calculated)
CH4
0.45
Solid Conversion
0.40
0.9
0.8
0.35
0.7
0.30
0.6
0.25
0.5
0.20
0.4
0.15
0.3
0.10
0.2
0.05
0.1
0.00
CH4 Conversion
0.50
0.0
-2 0
2
4
6
8 10 12 14 16 18 20 22 24 26 28 30 32
Distance from the bottom, inches
Figure 3.16 Reduction of Fe2O3 Composite Particles in a Bench Unit Using Methane as
Reducing Gas
-160-
Nomalized H 2 Concentration (%)
100
99.9
99.8
99.7
99.6
99.5
99.4
99.3
99.2
99.1
99
0.00
20.00
40.00
60.00
Time (min)
80.00
100.00
120.00
Figure 3.17 Hydrogen Production Using Reduced Fe2O3 Particles in a Bench Unit
-161-
Figure 3.18 IGCC Process with CO2 Capture
-162-
Figure 3.19 Conventional Gasification – Water Gas Shift Coal to Hydrogen Process
-163-
Figure 3.20 The Syngas Chemical Looping Process
-164-
a
b
Figure 3.21 ASPEN Simulation Flow Sheet for: a. IGCC System Using GE- High
Efficiency Quench (HEQ) Gasifier; b. Conventional Coal to Hydrogen System Using
Shell Gasifier; c. SCL System Using Shell Gasifier (To be Continued).
-165-
Figure 3.21: Continued
c
-166-
CHAPTER 4
COAL DIRECT CHEMICAL LOOPING PROCESS
4.1 Coal Direct Chemical Looping (CDCL) Process Overview
Iron based chemical looping gasification processes convert carbonaceous fuels
through the steam iron reactions:
FeOx + R Æ FeOx-1 + RO
FeOx-1 + H2O Æ FeOx
where x can vary between 0 (Fe) and 1.5 (Fe2O3) and R can be gaseous fuels including
syngas and methane and solid fuels including coal and biomass. The analysis conducted
in Chapter 2 shows that iron oxide is a suitable oxygen carrier. Chapter 3 further
discusses the iron based syngas chemical looping (SCL) processes using gaseous fuels. In
this chapter, the iron based coal direct chemical looping (CDCL) process which converts
coal with the gas-solid contact pattern similar to that in the SCL process is discussed.
Challenges for the CDCL process such as solid fuel conversion enhancement, ash
separation, and pollutants control evolved in-situ from solid fuels are addressed. It
should be noted that a similar process concept can be extended to the use of other solid
fuels such as petroleum coke and cellulosic biomass. The CDCL process can be
-167-
configured differently depending on the heat integration schemes in the reactor system.
Two representative configurations are presented in this section.
4.1.1 Coal Direct Chemical Looping Process - Configuration I
Figure 4.1 shows the simplified flow diagram of Configuration I for the CDCL
process. Similar to the SCL process, the CDCL process is also comprised of three
reactors, i.e., the reducer (Unit 1), the oxidizer (Unit 2), and the combustor (Unit 3). The
reducer converts carbonaceous fuels to CO2 while reducing Fe2O3 to a mixture of Fe and
FeO; the oxidizer oxidizes the reduced Fe/FeO particles to Fe3O4 using steam, producing
a H2 rich gas stream; the combustor pneumatically transports the Fe3O4 particles from the
H2 reactor outlet to the reducer inlet using air, the Fe3O4 particles are re-oxidized to
Fe2O3 during the conveying process.
Reducer
The reducer is a countercurrent gas-solid reactor operated at 750 – 950 ºC and 1 –
30 atm. The countercurrent operational mode is intended to maximize the solids and gas
conversions. The solids flow can be in a moving bed or in a series of fluidized beds. It is
noted that the moving bed contact mode is highlighted in this design as it represents the
fundamental countercurrent solids contact pattern with gases that is preferred in this
reactor system. The desirable reaction in the reducer is:
C11H10O (coal) + 8.67 Fe2O3 Æ 11 CO2 + 5 H2O + 17.34 Fe
-168-
The coal exemplified here is Pittsburgh #8 and is represented as C11H10O given the
elemental composition
14
. The reaction is highly endothermic with the heat of reaction
equal to 1,794 kJ/mol at 900oC. Therefore, a significant amount of heat needs to be
provided to the reducer.
One option for balancing the heat is to partially combust coal in-situ by sending a
sub-stoichiometric amount of O2 into the reducer. The overall reaction is:
C11H10O + 6.44 Fe2O3 + 3.34 O2 Æ 11CO2 + 5 H2O + 12.88 Fe
with zero heat of reaction at 900 oC. Since the amount of oxygen required for the above
reaction is significantly less than that for the coal gasification reactions, the size of the air
separation unit (ASU) is smaller than those in traditional gasification processes. The
oxygen reduction leads to savings in both operating cost and capital investment of the
coal to hydrogen plant. Another option for heat balance in the reducer is to combust a
portion of reduced particles and uses them as the heat source. This option is similar to the
steam methane reforming process where heat is provided by combusting a portion of the
methane outside the reforming tubular reactors. Such an option will be further discussed
in Sections 4.1.2, 4.2, and 4.3.
Oxidizer
The oxidizer is a moving bed reactor operating at 500 – 850 ºC and 1 – 30 atm. In
the oxidizer, the Fe and FeO mixture from the reducer reacts with steam countercurrently.
-169-
The reactions in the oxidizer are as follows:
Fe + H2O Æ FeO + H2
3FeO + H2O Æ Fe3O4 + H2
This reaction is slightly exothermic. To maintain adiabatic operation, steam at a
moderately low temperature is introduced into the oxidizer to modulate the reactor
temperature.
Combustor
Fe3O4 from the oxidizer is fully regenerated in the combustor. The combustor is
an adiabatic entrained bed reactor operated at a pressure similar to the reducer and the
oxidizer. Air is used to pneumatically convey the Fe3O4 particles from the oxidizer outlet
to the reducer inlet while fully regenerating the particles by the following reaction:
4Fe3O4 + O2 Æ 6Fe2O3
This exothermic reaction heats up the solids as well as the gas. The hot solids are
subsequently fed into the reducer to partially compensate for the heat needed for the coal
conversion. The hot gas is then be used for power generation. The coal ash, which is
significantly smaller in size than the Fe2O3 composite particles, is separated out from the
cyclone on the top of the reducer along with the fine particles. Fresh makeup particles are
also introduced to the reducer to maintain reactivity.
-170-
4.1.2 Coal Direct Chemical Looping Process - Configuration II
Configuration II of the coal direct chemical looping process is briefly presented in
this section with details given in Sections 4.2 and 4.3 in the context of process analysis
and ASPEN Plus® simulation.
There are also three major units involved in CDCL Configuration II, i.e. the
reducer, the oxidizer, and the combustor. A simplified diagram for the CDCL process is
shown in Figure 4.2. The key difference between Configurations I and II lies in the heat
integration strategy. As indicated in Section 4.1.1, the reaction between coal and iron
oxide in the reducer is highly endothermic. Unlike Configuration I where the heat
requirement of the reducer is met by partial oxidation of coal, in Configuration II,
reduced iron oxide particles are combusted to compensate for the heat deficit of the
reducer.
In Configuration II, composite Fe2O3 particles with an inert support are used to
oxidize coal in the reducer while being reduced to a mixture of metallic iron and FeO.
The gaseous product of this reactor is mainly CO2 mixed with a small amount of H2O.
The reduced Fe/FeO particles from the reducer are split into two streams. The first stream,
comprising most of the reduced Fe/FeO particles, is sent to the oxidizer to perform the
steam-iron reaction. The oxidizer produces H2 while oxidizing the reduced Fe particles to
Fe3O4 with steam. The rest of the reduced particles from the reducer, along with the
Fe3O4 particles discharged from the oxidizer, are burned in the combustor with air. As a
result, high temperature solid and gas streams are generated from the combustor. The
-171-
sensible heat carried by the high temperature solids is used to support the heat
requirement in the reducer. By increasing the amount of particles being combusted, the
excess heat can also be produced from combustor for electricity generation at the expense
of the hydrogen yield.
4.1.3 Comments on the Coal Direct Chemical Looping Process
The goal of the CDCL processes is to efficiently produce hydrogen or syngas
from coal in a cost effective and environmentally friendly manner. The underlying coal
gasification-conversion strategies between the conventional coal gasification–water gas
shift and the CDCL processes, however, are quite different. Unlike the conventional
process where coal is directly gasified, the CDCL process gasifies coal indirectly.
Therefore, the water gas shift reaction which commonly appears in the carbonaceous fuel
conversion processes is not used in the iron based chemical looping gasification
processes. The successive CO2 and H2 separation step can, thus, be avoided.
The iron based chemical looping gasification processes are characterized by the
unique properties of the iron based particles (see Chapters 2 and 3). Iron has multiple
oxidation states. These oxidation states are inter-convertible. For example, iron oxides
can be reduced to a lower oxidation state by various types of fuels. Further, the reduced
iron oxide can be fully or partially regenerated with oxygen/air or steam. Various
reduction and oxidation reactions along with different kinetic/thermodynamic behavior
that is associated with these reactions allow numerous schemes possible for the heat and
energy management when the iron based chemical looping medium is used. These
schemes, once optimized, can dramatically improve the energy conversion efficiencies of
-172-
the coal gasification processes. The design criteria, practical issues, and the energy
management schemes of the chemical looping gasification processes will be discussed in
the following sections.
4.2 Challenges to the Coal Direct Chemical Looping Processes and
Strategy for Improvements
This section illustrates several critical issues challenging the coal direct chemical
looping process and the strategies that can be adopted to overcome them. These issues
include oxygen carrier particle reactivity, char reaction enhancement, gas and solid
conversions, fates of pollutants and ash, and heat management and integration. Results
from both experiments and theoretical analysis using ASPEN Plus® simulations are
presented.
4.2.1 Oxygen Carrier Particle Reactivity and Char Reaction Enhancement
As noted in Chapter 1, the reactivity of the oxygen carrier particle is the most
widely examined subject in the research and development of chemical looping processes
employing gaseous fuels such as methane and/or syngas. For chemical looping processes
employing solid fuels, however, additional issues must be considered that are associated
with the oxygen carrier interaction in-situ with gaseous and solid pollutants and coal ash.
Particle Reactivity
-173-
NiO is often considered as an attractive oxygen carrier because of its higher reactivity
than do the other oxygen carriers 118. However, it is noted that for the solid fuel chemical
looping reactor, the reaction rate between metal oxide and solid fuel, e.g. coal char, is
controlled by the rate of gasification of char
152
. Specifically, the solid-solid reaction
between coal char and metal oxide is very slow. Therefore, the reactions of oxygen
carrier and the char are mainly through the following solid-gas reactions in the presence
of CO2 or H2O:
H2O/CO2 + C Æ CO + H2/CO
(4.1)
MeO + H2/CO Æ Me + H2O/CO2
(4.2)
The overall reaction of reactions 5.1 and 5.2 is:
2MeO + C Æ 2Me + CO2
(4.3)
Thus, CO2 and H2O act as char reaction enhancers that promote the overall reaction in the
reducer. For most metal oxide based oxygen carriers such as NiO and Fe2O3, Reaction 4.2
is significantly faster than Reaction 4.1. Thus, in the presence of CO2/H2O, the overall
rate of reaction, i.e., Reaction 4.3, between coal char and most metal oxide based oxygen
carriers will be comparable even though there is varied metal oxides reaction kinetics
with CO/H2. For solid carbonaceous fuel conversions, there are advantages for using an
iron based oxygen carrier, which include low raw material cost, favorable
thermodynamic properties, high oxygen carrying capacity, high melting points for all
-174-
oxidation states, large particle strength, and low environmental and health concerns (see
Chapter 2), rendering them an ideal looping particles for solid carbonaceous fuel
applications.
Coal Char Reaction Enhancement
Char reaction enhancement schemes that improve the extent of the conversion for
coal char and oxygen carrier particles are illustrated in Figure 4.3. Figure 4.3a shows
scheme A, when hydrogen is used as char reaction enhancer. Under such a scheme, a
small portion of the hydrogen (< 5%) produced in the oxidizer is recycled and introduced
to the bottom of the reducer. The hydrogen introduced into the reducer flows
countercurrent to the solids flow. As shown in Figure 4.3a, the hydrogen reacts with iron
oxide to form steam (Reaction 4.2) as it enters the reactor. The steam will then react with
carbon in coal char to form hydrogen and carbon monoxide via the steam carbon reaction
(Reaction 4.1). Since one mole of steam will generate two moles of reducing gases, i.e.
CO and H2, more iron oxides will be reduced, producing steam and CO2. As a result, the
amount of char reaction enhancer, which is represented by compounds that can enhance
char gasification, is doubled. Thus, introducing a small amount of hydrogen into the
reducer initiates a “chain reaction”, producing a large amount of steam and CO2 that
enhance the conversions of coal char and metal oxide particles.
As illustrated in Figure 4.3b, scheme B introduces CO2 or steam at the bottom of
the reducer to enhance the char conversion in a manner similar to that explained in
scheme A for the iron oxide conversion.
The CO2 can be obtained from the reducer
-175-
exhaust which is then recycled to the reducer inlet. Although CO can also trigger a
similar effect as H2, CO2, or steam does, it is not attractive because: 1) CO is not
produced in the standard CDCL configuration and hence, not readily available; and 2)
high CO concentration at the reducer inlet can cause carbon deposition.
Schemes A and B have their respective advantages and disadvantages. From the
thermodynamic analysis using ASPEN Plus®, scheme A leads to a slightly higher iron
oxide conversion in the reducer than does scheme B; however, a portion of the valuable
hydrogen product must be recycled under scheme A. On the other hand, although CO2
and steam may negatively affect the iron oxide conversion in the reducer, they are
byproducts with low economic value. Clearly, the choice of one reaction enhancement
scheme over the other will depend on the process economics, which is affected by a
number of factors including hydrogen output and price, coal/coal char properties, and the
performance of the enhancing agent under the specific reducer design and operating
conditions. Simulations and bench scale tests reported in Section 4.2.2 further illustrate
the effects of the addition of a char reaction enhancer.
4.2.2 Configurations and Conversions of the Reducer
From the discussion in Section 4.1, it is noted that the design and operation of the
oxidizer and the combustor in the CDCL process are similar to that in the SCL process.
The key difference lies in the reducer, where coal and metal oxide are converted.
-176-
Coal is a combustible sedimentary rock made up primarily of carbon, but also
containing hydrogen, oxygen, nitrogen, sulfur, ash, and trace amounts of heavy metals.
At the reducer operating temperature, which is between 750 and 950ºC, coal is
decomposed to volatiles and char. In order to fully utilize the chemical energy in coal and
to generate sequestration ready CO2, both volatiles and coal char need to be fully
oxidized in the reducer. A higher conversion of the oxygen carrier is also desirable in
order to increase the steam to hydrogen conversion in the oxidizer and to reduce the solid
circulation rate. Therefore, a well conceived reducer design and gas-solid contacting
pattern is essential.
Overview of the Reducer Configurations
Figure 4.4 exemplifies the reducer design for CDCL Configuration II. In this
configuration, fresh Fe2O3 composite particles are fed from the top of the reducer while
pulverized coal is pneumatically conveyed to the middle section of the reactor using CO2.
A small amount of CO2 or steam is also introduced from the bottom of the reducer to
enhance the char conversion. The coal injection port divides the reducer into two sections.
The function of the upper section (Stage I) is to ensure full conversion of gaseous species
to CO2 and H2O whereas the lower section (Stage II) is used to maximize the char and
iron oxide conversions.
Due to the high reducer operating temperature, coal will be devolatilized in the
pneumatic injection zone. The coal volatiles will move upwards along with other gases
such as CO2, H2O, CO, and H2. Fresh iron oxide particles which enter from the top of the
-177-
reducer will interact with these gases in a countercurrent manner. The countercurrent
interaction between the coal volatiles and iron oxide particles ensures the complete
conversion of gaseous carbonaceous fuels. As can be seen from Figure 4.4, an annular
region is present for the coal flow in the reducer around the internal hopper in Stage I.
The hopper is designed to allow a countercurrent gaseous flow through it when iron oxide
solids are being discharged. Coal char and iron oxide particle mixing is initiated in the
annular region where coal devolatization begins. The mixing of the devolatilized coal
char and partially reduced iron oxide particles continue to occur as particles descend from
Stage I to Stage II. In Stage II, the ascending gaseous species, which contain mainly of
H2O, CO2, CO, and H2, will react with the descending solids. During this contact, the
coal char is progressively gasified by CO2 and H2O formed at the lower portion of the
reducer. Provided that an adequate residence time is given, which is estimated at 30 - 90
minutes, coal char can be fully converted. Further, the Fe2O3 particles can be reduced to a
mixture of metallic Fe and FeO. Coal ash will exit from the bottom of the reducer along
with the reduced particles. Thus, reaction product streams from the reducer include a
solid particles stream which exits from the bottom of the reducer, containing Fe, FeO,
and coal-ash, and an exhaust gas stream which exits from the top of the reducer,
containing mainly CO2 and H2O. By condensing out the H2O in the exhaust gas stream, a
pressurized CO2 stream can be obtained from the reducer. The following section further
illustrates the reducer performance using ASPEN Plus® simulation.
-178-
ASPEN Plus® Simulation on the CDCL Reducer
A. Model Setup
The ASPEN Plus® thermodynamic models described in Chapter 3 are shown to
predict the theoretical performance of a reducer that converts gaseous fuels. To extend
the models for simulating the performance of a CDCL reducer that converts solid fuels,
additional model simulation information need to be provided. They include definition of
the solid fuel, change in physical property method, and designation of solid unit
operations. In this simulation, solid fuels are handled with ASPEN Plus® built-in methods
195
. The choice of a suitable method is dependent upon the fuel type. For example, simple
solid fuels such as pure carbon are defined as a conventional solid whereas complicated
mixtures such as coal and biomass are defined as a nonconventional solid. Since the
analysis on conventional solids directly utilizes the built-in physical/chemical property
databank in ASPEN Plus®, simulation of pure carbon is rather straight forward. When a
nonconventional solid fuel is used, however, the physical and chemical property
information of the solid is necessary for accurate simulation. Such information includes
proximate analysis, ultimate analysis, and heat of combustion.
In the following sections, Illinois #6 coal with properties identical to that given in
Chapter 3 is used as the fuel unless otherwise noted. Since the moisture in coal can affect
the oxygen carrier conversion, an extra coal drying step using high temperature nitrogen
gas is incorporated prior to coal entering the reducer. The drying step is modeled using an
RStoic block and a Flash2 block. Since coal is defined as a nonconventional solid, it can
not directly “react” with other reactants in an ASPEN modulus. Instead, an RYield block
-179-
with a FORTRAN subroutine is used to decompose coal into water, inert ash and
elements such as oxygen, hydrogen and carbon. The decomposed components, all of
which are conventional components except ash, are then sent to an RGibbs block to
perform the desirable chemical reactions. The heat of reaction for coal conversion is
equal to the sum of the heat of conversions of the RYield (decomposition) block and the
RGibbs (reaction) block. If carbon conversion is not fully completed according to
experimental results, unreacted carbon shall be split out from the interblock stream
between RYield and RGibbs, and then mixed with the solid stream after the reducer.
Figure 4.5 illustrates the block diagram of the ASPEN Plus® models for both the
fluidized bed and the moving bed reducers. Similar to the models used for the looping
process with gaseous fuels given in Chapter 3, a model of one RGibbs block is used to
represent a fluidized bed reducer as shown in Figure 4.5a, while a model of five RGibbs
blocks in series is used to represent a moving bed reducer as shown in Figure 4.5b. For
the moving bed model, dry coal is injected into the middle block (Block 3) in the 5-block
moving bed model to mimic the reducer design given in Figure 4.4. In addition, char
reaction enhancers such as H2, CO2, and/or steam can be introduced from the bottom
block (Block 1). In case when Fe2O3 is in excess, coal can be fully gasified in Block 3 of
the moving bed model. Thus, the syngas from coal gasification will then move upwards
and be converted in Blocks 4 and 5. In this case, no reaction will take place in Blocks 1
and 2. In case when coal can not be fully gasified in Block 3, further conversion of coal
char will take place in Blocks 1 and 2.
-180-
B. Coal and Oxygen Carrier Conversions
An ideal reducer should be configured so that coal is fully oxidized to CO2 and
H2O. Unconverted fuel can exit from the reducer either in the form of unconverted coal
char or partially converted CO and H2. The unconverted coal char will be carried over to
the oxidizer, resulting in a contaminated H2 stream. The partially converted CO and H2,
however, will lower the energy conversion efficiency of the process as well as
contaminate the CO2 stream from the reducer. Although a higher iron oxide conversion is
also desirable, optimization of the reducer should emphasize the coal conversion
efficiency rather than the iron oxide conversion.
The char gasification enhancer can enhance the reaction rate between the oxygen
carrier and coal; however, the extent of reaction can be limited by the thermodynamic
equilibriums. The thermodynamic analysis for syngas conversions in Chapter 3 shows
that above the stoichiometric amount of Fe2O3 is required in order to fully oxidize coal to
CO2 and H2O. Further, a countercurrent moving bed reducer will be more effective than a
fluidized bed reducer. In order to examine the reducer performance when solid fuels such
as coal are used, ASPEN Plus® simulation is performed. For illustration purposes, pure
carbon (graphite) is used to represent coal. Figure 4.6 shows the relationship between the
outlet CO concentration and the Fe2O3/carbon molar ratio using two different reducer
designs.
From Figure 4.6a, it is seen that in order to fully oxidize one mole of carbon to
CO2 in a fluidized bed reducer, more than 6 moles of Fe2O3 needs to be consumed. Based
-181-
on the same iron oxide conversion definition used in Chapter 3, the fluidized bed reducer
yields an Fe2O3 conversion of 11.11%. In comparison, a countercurrent moving bed
reducer requires merely 0.7 mole of Fe2O3 to convert 1 mole of carbon to CO2, which
corresponds to 95.2% Fe2O3 conversion to a reduced particle with 92.8% Fe and 7.2%
FeO. The dramatic increase in the iron oxide particle reduction illustrates a desirable gassolid flow pattern for using countercurrent moving beds over fluidized beds.
C. Fates of Pollutants in Reducer
The previous figure illustrates the effect of reducer design with coal approximated
by pure carbon. The actual composition of coal, however, is far more complex than pure
carbon. This section analyzes the fates of pollutants such as sulfur and mercury in coal.
Here, Illinois #6 coal with composition shown in Table 5.1 is used as the fuel. The
mercury content in coal is assumed to be 100 ppb (by weight). All the mercury in the coal
is assumed to be in elemental form. At 900 ºC and 30 atm, the maximum conversion of
Fe2O3 that ensures full conversion of coal to CO2 and H2O is calculated to be 73.8%.
Table 4.1 shows the mass balance of the reducer using the ASPEN Plus® model.
As can be seen from Table 4.1, all the mercury and chloride compounds will exit
from the reducer as a part of the gaseous stream. Although a small amount of sulfur will
escape from the reducer along with the exhaust gas in the form of H2S, 93.5% of the
sulfur in coal will react with iron oxide in the presence of reducing agents, forming
Fe0.877S. The solid sulfur compound will be carried over to the oxidizer along with the
reduced iron oxide particles.
-182-
D. Effect of Temperature
The sensitivity analysis is carried out using the ASPEN Plus® model. The result
of the analysis is shown in Figure 4.7. As can be seen, a higher reaction temperature
favors the endothermic coal-Fe2O3 reaction from both kinetic and thermodynamic
viewpoints. As shown in Figure 4.7, with a Fe2O3/coal ratio identical to the case shown in
Table 4.1, carbon can not be fully converted at temperatures below 850 °C. Therefore, the
reducer needs to be operated above 850°C to maximize the fuel conversion.
A practical concern on the reducer operating temperatures is its effect on ash
handling. An operating temperature significantly exceeding 900 ºC is expected to make
coal ash sticky and hence affect the solids flow in the reactor. Therefore, it is desirable to
operate a reducer at temperatures at ~900 ºC.
E. Effect of Pressure
The coal-Fe2O3 reaction in the CDCL reducer generates gaseous products. Thus, a
lower operating pressure favors the coal conversion as evidenced in Figure 4.8 for coal
reaction at 850 ºC. Considering both the kinetics and the equilibrium conversion, the
suitable operating pressure range for the CDCL reducer is determined to be 1 - 30 atm.
F. Effect of Steam and CO2
Although steam and CO2 can be used as char reaction enhancers, an excessive
amount of steam and/or CO2 can negatively affect the Fe2O3 conversion due to their
-183-
capability of oxidizing Fe/FeO. Figure 4.9 illustrates the effect of steam and CO2 on the
Fe2O3 conversion. The operating conditions of the reducer are identical to those shown in
Table 4.1. It is seen that an injection of a small amount of CO2 or steam into the reducer
will not lead to a drastic decrease in the Fe2O3 conversion.
Experimental Testing of Reducer Operations
Bench scale tests were carried out in the moving bed reactor at the Ohio State
University (OSU) based on the moving bed design and configuration described in
Chapter 4. Given below are representative test results obtained from the operational
configuration of a moving bed given in Figure 4.4 where the reducer can be divided into
two stages with Stage I conducting coal volatile conversion and Stage II performing char
gasification and iron oxide particle reduction.
A. Reducer Stage I Testing
Iron oxide is active in cracking coal volatiles into methane
191
, which is the most
stable hydrocarbon formed from the cracking of coal volatiles up to 1030ºC
192
. The
reactions between Pittsburgh #8 coal volatiles/tar and the composite Fe2O3 particles in a
fixed bed reveal that 87% of the volatiles was either cracked to methane or oxidized to
CO2/H2O by the composite particle within a gas residence time of 4.6 seconds at 850 ºC.
Thus, if the composite particles are capable of oxidizing methane, they will be able to
oxidize coal volatiles.
-184-
The methane conversion profile in a given countercurrent moving bed reactor is
shown in Figure 3.15 of Chapter 3. This methane conversion profile indicates the Stage I
reducer operation behavior. As can be seen in the figure, more than 99.8% of methane is
converted to CO2 and H2O. Thus, the results reflect that the iron oxide composite particle
is capable of fully oxidizing coal volatiles into CO2 and H2O.
B. Reducer Stage II Testing
The oxidation of coal chars obtained from both bituminous and lignite coal using
the composite particles reveals over 90% char conversion while that from low volatile
anthracite coal reveals 95% conversion as shown in Figure 4.10. During the testing, the
solids residence time in the bed is 60 – 100 minutes and char gasification enhancers such
as H2 and/or CO2 are present. The test results are given in Table 4.2.
As shown in Table 4.2, a countercurrent moving bed reducer with iron oxide
based oxygen carrier can convert 90–95% of coal char into concentrated CO2 and H2O.
The X-Ray Diffraction (XRD) analysis of the composite particles obtained after the
reducer testing reveals the formation of Fe0.877S, which is consistent with the
thermodynamic analysis. Although the coal and particle conversions are lower than those
predicted from the thermodynamic analysis, optimization of design and operating
conditions such as char reaction enhancer flow rate, operating temperature, and gas and
solid contact time can improve the reducer performance.
-185-
4.2.3 Performance of the Oxidizer and the Combustor
Oxidizer
The oxidizer in the CDCL process is similar to that in the SCL process, which has
been discussed in Chapter 3. Steam is sent to the oxidizer to convert Fe and FeO to Fe3O4
while producing hydrogen. Although the unconverted steam can be readily condensed, a
lower steam conversion will lead to larger energy consumption for steam generation.
Thus, it is essential to maximize the steam to hydrogen conversion. Similar to the reducer,
a countercurrent moving bed reactor is an effective design for the oxidizer. Figure 4.11
compares the theoretical steam to hydrogen conversion in a moving bed and a fluidized
bed operated at 700 °C and 30 atm using pure iron particles. At a given steam to iron
molar flow rate, a moving bed results in a significantly higher conversion.
Besides the gas-solid contacting mode, the composition of the Fe/FeO particle is
also an important factor in the steam to hydrogen conversion in the oxidizer. Figure 4.12
shows the effect of Fe content on the steam conversion in a countercurrent moving bed
oxidizer. As can be seen from the figure, the increased presence of Fe in the particle can
drastically improve the steam to hydrogen conversion. When pure FeO is used, only
26.4% of the steam can be converted to hydrogen; however, when the particle contains
more than 33% metallic iron (by mole), the steam conversion is increased to 62.2%.
Further increase in the iron content will not lead to higher steam conversions since the
gas composition has already reached the equilibrium point with Fe. Therefore, in order to
achieve optimum steam to hydrogen conversion, the reduced Fe2O3 from the reducer
-186-
should contain at least 33.5% Fe, which corresponds to a reduction rate of 49.0% or
higher. Thus, the performance of the reducer and that of the oxidizer are closely related
and contribute synergistically to the overall looping process efficiency. The importance
of such inter-relationship can further be illustrated by an opposite case evidenced by the
early steam-iron processes where the iron oxide reduction in the fluidized bed or twostage fluidized bed reducer was low, thereby leading to a low steam conversion in the
oxidizer.
The reaction temperature also has a significant effect on the steam to hydrogen
conversion in the oxidizer. A lower reaction temperature would thermodynamically favor
the exothermic reaction between steam and Fe/FeO. For instance, the reaction
temperature of 500 ºC would lead to a steam to hydrogen conversion of 82.2% as
compared to 62.2% conversion at 700 ºC. Further, a lower operating temperature will
reduce the capital cost of the oxidizer as lower cost materials can be utilized. However,
the low reactivity of the iron ore renders it challenging to perform the steam-iron reaction
below 600 ºC. To compare, the composite Fe2O3 particle obtained through particle
optimization can undergo the steam-iron reaction at 500 ºC or even lower with a
satisfactory reaction rate.
The fate of sulfur in oxidizer operation is another important issue that needs to be
accounted for. At 700 ºC and 30 atm, the oxidizer can convert all of Fe and FeO to Fe3O4,
with a steam conversion of 62%. Further, about 9.9% of Fe0.877S carried over to the
oxidizer will react with steam to produce Fe3O4 and H2S based on the thermodynamic
-187-
analysis. This reaction leads to an H2S concentration of 430 ppmv in the hydrogen
product (dry basis). The remaining Fe0.877S will be carried over to the combustor along
with Fe3O4 to react with air.
Combustor
The early steam iron processes did not encompass a combustor. However, it is
crucial for the SCL and the CDCL processes to include this combustor unit. The
combustor plays an important role in the fuel conversion and the energy management of
the process. The early steam-iron processes directly circulate the solid products from
oxidizer to the reducer inlet. Since steam oxidation can only regenerate iron to Fe3O4,
complete oxidation of fuels to CO2 and H2O cannot be achieved in these processes. For
example, when Fe3O4 is used in the reducer to convert coal at 900 ºC, the exhaust gas will
have a CO concentration of 11.2% at minimum. To compare, nearly 100% fuel
conversion can be achieved when Fe2O3 is used. The CDCL process can also be operated
by sending part of the reduced particles from the reducer outlet directly to the combustor
for heat generation, while the remaining particles are processed through the oxidizer for
hydrogen generation. In this manner, the CDCL system can generate any combination of
hydrogen and electricity products from coal. Further information regarding the
combustion of reduced iron oxide particles are given in Chapters 3.
NOx and SOx are the main contaminants from the combustor. All the Fe0.877S
introduced to the combustor will be fully oxidized to Fe2O3, SO2 and SO3. NOx will also
be formed at a high operating temperature of the combustor. For example, when the
-188-
combustor is operated at 30-32 atm and 1150 ºC, the NOx level is estimated to be at ~300
ppm and the concentration of SO2 and SO3 are 1.4% and 500 ppm (by volume)
respectively. These compounds can be separated from flue gas using commercial flue gas
cleaning devices. The above estimations are made based on the assumption that
thermodynamic equilibriums among the reactants and potential products are reached. It is
likely that reactions with slow kinetics will not reach their equilibrium. For example, the
actual NOx level may well below 300 ppm due to the flameless combustion environment
and the absence of coal nitrogen during the combustor operation. Further account of
pollutants is given in the following section.
4.2.4 Fates of Pollutants and Ash
The following section discusses issues such as ash separation, pollutant control,
and particle tolerance to pollutants that are of importance to the CDCL process operation.
Ash Separation
Determining an appropriate size of iron oxide particles requires considering, in
addition to reactivity, the fluidization characteristics during the pneumatic conveying in
the entrained bed combustor and during the gas-solid countercurrent flow in the moving
bed reducer and oxidizer. With this consideration, the iron oxide particle sizes of 0.7 mm
to 8 mm are regarded appropriate. The particles in this size range are substantially larger
than those of pulverized coal (50–250 µm). With the operating conditions of the reducer,
oxidizer and combustor, most of the ash particles will be in fly ash form of sizes 0.5 –
100 µm, which are about an order of magnitude smaller than the iron particles. Thus,
-189-
most of the ash particles can be separated from the iron oxide particle based on the size
difference using devices like sieves and/or cyclones. To test the fly ash separation, fly ash
obtained from a pulverized coal combustion power plant is mixed with cylindrical Fe2O3
composite pellets of 5 mm in diameter and 3 mm in height. The ash concentration in the
mixture used for this test is 8.77% by weight. To simulate the reducer operation, the fly
ash and pellet mixture passed through the moving bed reactor at 1030 ºC against the
countercurrent flow of hydrogen. The pellet/ash mixture is then collected and regenerated
with air at 900 ºC. The high temperature used in the test is intended to determine whether
eutectic melting occurs between the pellets and the ash. After the particle/ash mixture is
fully regenerated with air, it is loaded into a 2.8 mm sieve and sieved for 15 seconds. The
resulting pellets and powder were characterized, respectively, for Fe2O3 and ash contents
after completing the test. It was determined that 75.8% of the ash can be separated from
the pellets with ease. Given 75.8% ash separation efficiency, the ash content that will
accumulate in the CDCL system is calculated to be 0.54% with respect to the solid
content in the reactor. Such low ash content is expected to have little effect on the CDCL
operations. The test results also imply that mechanical ash separation methods are
feasible.
Particle Tolerance to Contaminants
Understanding the blinding effect of the oxygen carrier with coal contaminants is
important to the assessment of the viability of oxygen carrier particles in the CDCL
process. The effect due to the blinding and the particle attrition directly affects the extent
of the fresh particle makeup, and hence, the process economics. As discussed in Chapter
-190-
2, the particle attrition rate is ~0.57%/cycle. To assess the blinding effects induced by the
coal contaminants, the same batch of iron oxide particles is used to react with different
types of coal, coal char and ash. The particles are regenerated with air after each
reduction experiment and the reactivity of the particles that have endured three redox
cycles are characterized in a TGA and a differential bed. No notable decrease in the
particle reactivity is evidenced from these tests. The blinding effect of the particles needs
to be further tested with more cycles to ascertain that the iron oxide particles are robust
enough for the CDCL process application. Optimum design of particles that increase the
tolerance towards contaminants in coal is an important area that requires extensive further
study.
Fates of Contaminants
The fates of the contaminants are discussed in Sections 4.2.2 and 4.2.3 in
connection with CDCL reactor operations. The pollutant control strategy for the CDCL
process from a process viewpoint is given in this section. The SOx, NOx, and mercury
compound generated from the CDCL process can be captured with ease using existing
pollutant control methods. Based on thermodynamic analysis, 93.5% of the sulfur in coal
can be “captured” by the composite particle in the form of Fe0.877S. The remaining sulfur
will be released from the reducer along with CO2 in the form of H2S. The H2S can be
sequestrated along with CO2
196
. Fe0.877S in the reducer will be carried over to the
oxidizer to react with steam. In the oxidizer, 9.9% of Fe0.877S will be oxidized with steam,
forming Fe3O4 and H2S. The H2S concentration in the oxidizer product gas stream is 430
ppmv, which can be removed using traditional scrubbing technique such as MDEA or
-191-
SELEXOL. Remaining Fe0.877S will be carried over to the combustor to form Fe2O3, SO2,
and SO3. The SOx concentration in the exhaust gas is estimated to be 1.5%, which can be
stripped using existing flue gas desulfurization (FGD) unit utilized in the pulverized coal
(PC) power plant. Due to the presence of nitrogen in air, the combustor also generates
NOx. The NOx concentration from the combustor exhaust gas stream is estimated to be
300ppm, which is much lower than that produced from a PC boiler. The existing selective
catalytic reduction (SCR) method can be used to capture NOx. The ASPEN Plus® process
simulation shows that 100% of mercury in coal will be emitted in elemental form from
the reducer. The concentration of mercury will be at ~43 ppb (wt) when Illinois #6 coal is
used. The CO2 stream containing mercury may possibly be directly sequestrated. If not,
an activated carbon bed can be used to remove mercury from the CO2 stream before it is
sequestrated.
At the present stage of the development of the CDCL process, the fates of the
contaminants are mainly determined based on the thermodynamic analysis. Experimental
data will be needed to substantiate such results.
4.2.5 Energy Management, Heat Integration, and General Comments
Stoichiometrically, one mole of carbon in coal can be converted to two moles of
hydrogen according to:
C + 2H2O (g) Æ CO2 + 2H2
ΔH = 178 kJ/mol @ 298.15 K
-192-
In the conversion process, a portion of the energy from coal needs to be used to support
the steam generation, air separation, hydrogen product separation and purification, and
endothermic steam-carbon reaction. A practical coal to hydrogen process will deliver a
thermal efficiency far lower than 100% also due to exergy degradation induced by the
less than perfect operating efficiency for such units as heat exchanging devices, and the
limitation on heat integration. Thus, to enhance the overall energy conversion efficiency,
the process intensification strategies that minimize the parasitic energy consumptions and
energy loss are required. This section discusses such strategies utilized in the CDCL
process and they are extendible to other coal conversion processes. These strategies
include: (1) minimization of air separation; (2) minimization of the energy consumption
for CO2 separation; (3) reduction of the steam usage; (4) optimization of the heat
integration scheme.
Air Separation
The traditional gasifier consumes a significant amount of oxygen. The cryogenic
distillation for air separation requires an extremely low temperature and extensive gas
compression and recompression for separation of O2 from N2, which is energy intensive.
Thus, oxidation using oxygen carriers would be a more economical approach than the use
of pure oxygen. Further, using one chemical looping and hence one looping particles as
in the coal direct chemical looping (CDCL) gasification processes will be more
economically feasible than using two chemical looping and hence two looping particles
as in the ALSTOM Hybrid Combustion-Gasification process. The heat required in the
-193-
CDCL process, and hence the oxygen need, can be completely compensated for via the
CDCL Configuration II discussed in 4.1.2.
CO2 Separation
The CO2 separation is another energy consuming step in traditional coal
conversion processes. Various CO2 separation methods for combustion flue gas and
syngas applications are available as discussed in Chapter 1. Basically, there are two
approaches: the high-temperature sorbent based and the low-temperature solvent based.
Both approaches are energy intensive. In the CDCL process, the chemical looping
combustion and chemical looping gasification are both adopted to provide process
versatility in product generation and CO2 separation. This gasification scheme can
circumvent the energy intensive CO2 separation approaches encountered in the traditional
processes. Moreover, the CO2 from the CDCL process is already at a high pressure and
thus, the energy consumption in CO2 compression can also be reduced.
Minimization of Steam Usage
The hydrogen is produced from coal by reacting with steam, either directly or
indirectly in a traditional process. The latent heat and the sensible heat for steam
generation represent a notable portion of the total energy consumption of the process.
Due to the equilibrium and kinetics limitations, the excessive steam beyond the
stoichiometric requirement is usually used to reach a desired H2 yield. Although the heat
in the excess steam can be partially recovered, a significant increase in the process
entropy occurs. For iron based chemical looping processes such as the CDCL process,
-194-
the extent of reduction of the iron oxide particle in the reducer is the key factor affecting
the steam requirement. Through the coal and iron oxide particle conversion enhancement
strategies discussed in Sections 4.2.1 and 4.2.2, the steam usage in the CDCL process is
minimized.
Optimization of the Heat Integration Scheme
Heat integration is of direct relevance to the efficiency of an energy conversion
process. Optimization of heat integration schemes is generally a complex issue. When an
exothermic reaction in a process system takes place at a temperature higher than another
reaction in the process system that is endothermic, the heat generated from the
exothermic reaction can then readily be used to support the endothermic reaction. In the
sense of the heat integration, the heat generated by the exothermic reaction is stored in
the reaction products generated from the endothermic reaction. Thus, in this integration,
the exergy in heat is recuperated. To further illustrate, the heat integration scheme for
Configuration II of the CDCL process is given for the overall reaction of the process as:
xC + yH2O + (x-y/2)O2 Æ xCO2 + yH2 + ΔH
For simplicity, the composition of coal is considered as pure carbon. Principally, the heat
of reaction can be adjusted by varying the ratio of x to y. Thus, at the condition when ∆H
is zero, all the chemical energy in the carbon is converted to H2 considering that the
enthalpy of devaluation for CO2, H2O, and O2 is negligible as compared to carbon and H2.
For a real process, however, heat loss is inevitable. Further, the parasitic energy in the
-195-
process needs to be consumed to generate excessive steam for hydrogen production, to
produce electricity for gas compression and particle circulation, and to generate heat for
pollutant separation. Therefore, an optimal heat integration strategy that maximizes the
H2 yield would minimize the absolute value of ∆H.
In the CDCL process, the combustion of fully or partially reduced iron oxide is
highly exothermic and can take place at temperatures that are much higher than those for
the endothermic coal oxidation reaction. Therefore, the heat released in the reduced iron
oxide combustion is used to compensate for the endothermic coal oxidation reaction.
Moreover, the heat generation from combustion of particles directly from the reducer and
from the oxidizer can easily be altered by adjusting the percentage of the particles from
the reducer and the oxidizer to the combustor, making the process flexible for hydrogen
and electricity co-production. The overall material and energy integration scheme for the
CDCL process is shown in Figure.4.13.
4.3 Process Simulations on the Coal Direct Chemical Looping Process
The following section presents a case study on the energy conversion efficiency
of the CDCL process system for H2 production using ASPEN Plus® process simulation.
The advantages of the aforementioned strategies for the energy management of the
looping processes are illustrated.
-196-
4.3.1 ASPEN Model Setup
Model Assumptions
The ASPEN simulation model is used to further illustrate the configuration II of
the coal direct chemical looping process (CDCL). Hydrogen is the desired product in this
case. Thus, the power generated from the system is only used to offset the parasitic
energy consumptions. The process flow diagram is shown in Figure 4.14. The
assumptions used in the simulation are identical to that given in Chapter 3, except that:
1) All the reactions reach their thermodynamic equilibrium.
2) Solids consist of 48.9% Fe2O3 and 51.1% supporting materials that comprise
mainly SiC (by weight).
3) Steam is utilized as the char reaction enhancer.
4) The heat loss in the CDCL system is 0.5% of the HHV of coal.
5) Coal is dried to 3% moisture prior to entering the reducer.
In setting up the ASPEN simulation, it is important to verify whether the correct physical
properties are represented for the system. Aspen Plus® retrieves parameters from different
databanks, and the COMBUST, INORGANIC, SOLIDS and PURE databanks are
selected for the CDCL. Revised physical property data as discussed in Chapter 4 are used
for FeO and Fe3O4. PR-BM is the property method used for the global system, whereas
STEAM-TA is the method used for steam generation and conversion units such as HRSG
and Steam Turbines.
-197-
4.3.2 Simulation Results
Mass and energy flows obtained from the ASPEN simulation for the CDCL
process are discussed below.
Reducer
Coal is first pulverized and dried to a moisture content of 3%. The coal powder is
then introduced to the middle section of the reducer at a rate of 132.9 ton/hr. The hot
Fe2O3 and SiC particles (1250 ºC) from the combustor enter the reducer from the top at a
rate of 2312.2 ton/hr (with 48.9% Fe2O3 and 51.1% SiC). The sensible heat of the particle
is used to partially compensate the heat needed in the reducer. In addition, the sensible
heat in combustor flue gas (1250 ºC, 373.4 ton/hr) is also transferred to the reducer
through the heat exchangers. The reducer is operated at 870ºC and 30atm. In the reducer,
Fe2O3 is reduced to a mixture of Fe (26.1% by weight according to the overall solids),
FeO (wuestite, 15.6% by weight), and FeS (Fe0.877S, 0.35% by weight). The overall exit
solids flow rate from the reducer is 2063.8 ton/hr. An exhaust gas stream of 370.8 ton/hr
with 68.7% CO2 and 30.6% steam is produced from the top of the reducer. The exhaust
gas also contains 0.58% N2 and 0.14% H2S. The hot gas is sent to HRSG for heat
recovery. It is then cleaned, condensed, and compressed to 150 atm for sequestration.
Oxidizer
Oxidizer is operated at 720 ºC. 70.0% of the reduced Fe/FeO particles are split for
hydrogen generation. The remainder is directly sent to the combustor. Steam at 240 ºC
and 32 atm is introduced at the bottom of the oxidizer at a flow rate of 286.0 ton/hr. The
-198-
excess heat is removed by generating high pressure steam. In this reaction, Fe and FeO
are oxidized to Fe3O4 while H2O is reduced to H2. A product gas stream with 19.9 ton/hr
H2 (62.1% by mole) is produced from the oxidizer. The product gas stream is first sent to
the HRSG and then an acid gas removal (AGR) system. The purified hydrogen is
compressed to 60 atm. The solid stream is sent to the combustor.
Combustor
Fe3O4 exiting from the oxidizer, together with 30.0% of the solids discharged
from the reducer, is pneumatically conveyed with air to the reducer. During the
conveying, all the reduced particles are re-oxidized to Fe2O3 and a significant amount of
heat is released. As a result, both the solids and the spent air are heated to ~1250 ºC. The
sensible heat carried by the particle and a portion of the hot spent air is used to meet the
reducer heat requirement. The remaining hot spent air is used for feedstock preheating
and steam generation. The spent air, after heat exchange, will go through SCR and FGD
scrubbers before vented to the atmosphere. Ash in the solid stream is separated from the
hot solids by a cyclone before entering the reducer.
Tables 4.3 to 4.6 illustrate the simulation results from the ASPEN model. As can
be seen, CDCL can produce 19.9 tons of H2 from 132.9 tons of coal while generating
sufficient electricity for parasitic energy consumptions. This energy conversion
corresponds to a theoretical efficiency of 78.3% (HHV) for hydrogen production. There
are alternative heat management schemes for the CDCL process. For example, an
expander can be used after the combustor for the power generation, resulting in a higher
efficiency. Further, the heat required by the reducer can be fully provided by the high
-199-
temperature solids from the combustor. The efficiency for this case is estimated at 79%
(HHV) with 72% hydrogen and 7% power.
Overall, the energy conversion efficiency for the CDCL process is nearly 20%
higher compared to the traditional coal gasification-water gas shift process for H2
production with CO2 capture discussed in Chapter 1. The improved efficiency for the
CDCL process results from the improved energy management of the process.
4.4 Concluding Remarks
The coal direct chemical looping (CDCL) process using iron based oxygen
carriers is carried out in a countercurrent gas-solid flow mode for the reducer and
oxidizer in a similar manner as the syngas chemical looping (SCL) process. The solids
flow can be in a moving bed or in a series of fluidized bed. The CDCL process and the
SCL process differ in heat loading requirement in the reducer and in the composition of
solid and gas pollutants in the reducer and oxidizer. The CDCL can be operated in two
configurations depending on the reducer heat integration scheme. Understanding the
blinding effect of the oxygen carrier with coal contaminants is critical to the accurate
assessment of the viability of the oxygen carrier particle in the CDCL process. The
blinding effect and particle attrition rate directly affect the recycleability of the particle,
and hence, the extent of the particle make-up requirement. The ASPEN Plus® simulation
projects the fates of the pollutants including SOx, NOx, mercury, and H2S in the looping
reactors. The simulation indicates that sulfur in coal can be bound by the iron oxide
-200-
particle in the form of Fe0.877S. The simulation also reveals a high conversion efficiency
of coal to hydrogen or to electricity in the CDCL process.
-201-
Species
CO
CO2
H2
H2O
HG
S
H2S
N2
O2
CL2
HCL
H3N
COS
CH4
NO
NO2
N2O
FE2O3
FE3O4
FE0.947O
FE0.877S
FE
C
ASH
Total Flow (kg/sec)
* The molar weight of ash is set to be 1.
Inlet Flow Rate
(mol/sec)
0
0
2.232276
0.617254
4.99E-07
0.078276
0
0.044621
0.215008
4.09E-03
0
0
0
0
0
0
0
5.6
0
0
0
0
5.307635
9.7
994.2763
Outlet Flow Rate
(mol/sec)
2.89E-03
5.304384
1.17E-03
2.839538
4.99E-07
1.25E-08
4.73E-03
0.044621
7.74E-11
4.64E-11
8.18E-03
6.94E-09
3.61E-04
1.61E-14
8.08E-10
5.50E-16
2.20E-14
0
0
4.395712
0.073184
6.973078
0
9.7
994.2763
Table 4.1 Reducer Mass Balance Based on the ASPEN Plus® Model at 900 ºC
-202-
Type of Fuel
Configuration Tested
Fuel Conversion (%)
CO2 Concentration in
Exhaust (% Dry Basis)
Conversion Enhancer Used
Coal Volatile
(CH4)
Both
99.8
Lignite
Char
I
94.9
Bituminous
Char
II
90.5
Anthracite
Coal
I
95.5
98.8
99.23
99.8
97.3
H2
CO2
O2 and CO2
H2
Table 4.2 Summary of the Reducer Demonstration Results using Coal, Coal char, and
Volatile
-203-
Input
Stream
Coal
Air
Water
Output
Recycly Solids
Off Gas
CO2
H2
Ash
other
Temperature ºC
25
1250
120
40
40
800
25
Pressure
1
30
16
135
60
30
1
2312.2
373.4
311.8
19.9
12.9
70.6
atm
Mass Flow ton/hr
Table 4.3
132.9
178
477.7
Overall Input-Output Diagram for the Mass Flow of the CDCL Process
Stream/Unit
MW
Coal
-1000
Air Compressor
-67.3
Steam Turbine
67.5
H2
782.9
Waste heat
217.3
Table 4.4 Overall Input-Output Diagram for the Energy Flow of the CDCL Process
-204-
Unit
MW
Reducer
50.4
Table 4.5
Combustor
0
HRSG
344.1
Heat and Energy Requirements in the CDCL Process
Output
Steam Turbine
Input
Compressor
Unit
Operations
Power
MW
Table 4.6
Oxidizer
-46.2
Net
Air
H2
CO2
HP
45.8
12.6
8.9
-17.2
IP
LP
-30.6 -19.7
Power Balance in the CDCL Process
-204-
-0.2
N2
Fe2O3
Coal +
O2/CO2
CO2
CO2 H2O
1
H2O
Fe, FeO,
Fe3O4
H2
3
Fe
Fe
H2O
2
H2O
H2
H2, H2O
Fe3O4
Air
Fe3O4 Æ Fe2O3
Figure 4.1. A Simplified Flow Diagram for Coal Direct Chemical Looping Process –
Configuration I
-205-
Figure 4.2 A Simplified Flow Diagram for Coal Direct Chemical Looping Process –
Configuration II
-206-
1H2
1CO2
H2 + FeO Æ Fe + H2O
1 CO2
1 H2O
0 CO2
CO2 + C Æ 2CO
H2O + C Æ H2 + CO
H2 + FeO Æ Fe + H2O
CO + FeO Æ Fe + CO2
H2O + C Æ H2 + CO
CO2 + C Æ 2 CO
H2 + FeO Æ Fe + H2O
1 H2O
2CO + 2FeOx Æ 2FeOx-1 + 2CO2
1 CO2
2 CO2
2CO2 + 2C Æ 4CO
1 H2O
3CO + 3FeO Æ 3Fe + 3CO2
4CO + 4FeOx Æ 4FeOx-1 + 4CO2
3 CO2
4 CO2
1 H2O
7 CO2
a
b
Figure 4.3. Char Reaction Enhancement Schemes: a. Using Recycled Hydrogen from the
Oxidizer; b. Using Recycled CO2 from the Reducer Exhaust.
-207-
Fe2O3
CO2 + H2O
Particle reduction :
CH4 + 4Fe2O3 → CO2 + 2H2O + 8FeO
Stage 1I
Coal devolatilization :
Coal
(Conveyed with CO2)
Coal → C + CH4
CO + FeO Æ Fe + CO2
C + CO2 Æ 2CO
Stage 2II
Fe/FeO
Char gasification and particle
reduction:
C + CO2 Æ 2CO
C + H2O Æ CO + H2
CO + FeO Æ Fe + CO2
H2 + FeO Æ Fe + H2O
CO2/H2O
Figure 4.4 Gas-solid Contacting Pattern of the Reducer
-208-
a
b
Figure 4.5 ASPEN Plus® Model Setup for a. Fluidized Bed and b. Moving Bed
-209-
a
b
Figure 4.6 Concentration of CO with Respect to Different Fe2O3/Carbon ratios at 900 ºC
and 30 atm for a) a Reducer with Perfect Mixing; b) a Countercurrent Moving Bed
Reactor.
-210-
Figure 4.7 Effect of Temperature on Carbon Conversions in Coal at 30 atm with an Fe2O3
to Coal ratio of 8.94:1 by Weight
-211-
Figure 4.8 Effect of Pressure on Coal Conversion at 850 C with an Fe2O3 to Coal Ratio of
8.94:1 by Weight
-212-
Figure 4.9 Effect of Steam and CO2 on the Fe2O3 Conversion at 900 ºC and 30 atm with
an Fe2O3 to Coal Ratio of 8.94:1 by Weight
-213-
98
90
97
Solid Conversion (%)
80
96
70
95
60
94
50
40
93
30
Coal Conversion
20
Particle Conversion
92
CO 2 Concentration (%)
100
91
10
Outlet CO2 Concentration
0
90
0
Solid Outlet
10
20
30
Axil Position from Solid Outlet (inch)
40
Gas Outlet
Figure 4.10 Reducer Test Results Using Anthracite Coal
-214-
a.
b.
Figure 4.11 Steam to Hydrogen Conversion for: a. Countercurrent Moving Bed Oxidizer; b.
Fluidized Bed Oxidizer. Reactor Operating Conditions: 700 ºC, 30 atm
-215-
Input stream
Figure 4.12 Relationship Between the Fe/FeO Composition and the Steam to Hydrogen
Conversion in a Countercurrent Moving Bed Oxidizer Operated at 30 atm and 700 ºC.
(X-Axis Denotes the Molar Percentage of Metallic Iron in the Fe/FeO Mixture)
-216-
Figure 4.13 Material Flow and Energy Flow in a CDCL Process
-217-
H1
AD
S0
1
S1
G5
RGI BBS
P
N2
CO21
SG52
RGI BBS
CH
GGGG
S5
COMP
OFF
AIRO
OX
F
SG1
RGI BBS
H4
COOL
OFFFF
G1
RGI BBS
Q
H1
5
4
C2
OFFF
CO22
SG
S4
3
RGI BBS
REN
OFFF1
S2
G3
2
G4
SE
D
MULT
MULT
SS5
RGI BBS
GT
REC
SS
REW2
COAL
SEP SG51
SG5
S1
CW
SS1
SS2
S2
SG4
RGI BBS
POWER
S3
S4
SS4
SG3
RGI BBS
RGI BBS
SG1
RGI BBS
S5
RGI BBS
STEAM
ST
SG
SH1
H3
ST1
H2O
ST1
PUMP
H2OP
ST2
Figure 4.14. Process Flow Diagram of the ASPEN Plus® Model for the CDCL Process
Optimized for Hydrogen Production
-218-
CHAPTER 5
SUB-PILOT SCALE CHEMICAL LOOPING SYSTEM
5.1 Introduction
As discussed in Chapter 3, the syngas chemical looping (SCL) concept has been
validated in a bench scale unit under a semi-continuous mode, i.e. the continuous
operation of the three key reactors are tested individually. Further, the capacity of the
bench scale unit, which is 2.5 KWth, is significantly smaller comparing to that of a typical
power plant. To perform the continuous looping operation and to study the scale up
effects of the looping units, a sub-pilot scale chemical looping unit that integrates the
reducer, oxidizer, and combustor is constructed. The integrated unit has a maximum fuel
processing capacity of 25 KWth and is designed to operate at pressures up to 2 atm and
temperatures up to 1000 oC. The unit can also be used for CDCL operation with minor
modifications.
The schematic and photograph of the sub-pilot unit are shown in Figure 5.1. The
unit demonstrates the reducer, the oxidizer, and the combustor operations in a
simultaneous and continuous manner. As can be seen in Figure 5.1, the reducer (heating
section A) and the oxidizer (heating section B) are scale up versions of the reacting zone
-219-
of the bench unit. The reduced oxygen carrier particle is transported from the reducer
outlet to the oxidizer inlet through a rotary solid feeder. The partially regenerated oxygen
carrier from the oxidizer is then introduced to the combustor through a slant pipe and
finally conveyed pneumatically to the reducer inlet to complete the redox cycle. In the
following sections, the design of the sub-pilot unit is discussed in detail. Next, the
preliminary operations of the unit are reported. Finally, the potential improvements on the
current unit as well as the future demonstration plan are provided.
5.2 Sub-Pilot Reactor System Design
5.2.1 Arrangement of the Overall Reactor System
The design and construction of the integrated sub-pilot unit require
comprehensive consideration of continuous and accurate gas and solid handling, data
acquisition and sampling, automation, and most importantly, safety. The following
criteria are taken into account in designing the sub-pilot system:
1. Safety and reliability
2.
Ease of installation
3.
Ease of servicing
4.
Flexibility in operation
5. Capability for high temperature operations
6. Capability for gas and solid profile monitoring
The arrangement of various sub-systems of sub-pilot unit and their physical locations are
shown in Figure 5.2 and Figure 5.3 respectively.
-220-
As can be seen in Figure 5.2, the sub-pilot scale system is comprised of 6 subsystems/assemblies, viz. reactor assembly, which includes the reducer, the oxidizer, and
the combustor, automation and control assembly, gas delivery assembly, steam
generation assembly, air compression assembly, and sample analysis assembly. Various
assemblies interact with each other to perform the desirable functions.
During the operation of the sub-pilot unit, the gas delivery assembly continuously
delivers reactant gas mixture to the reducer at a desirable composition and rate.
Meanwhile, the steam generation assembly delivers steam to the oxidizer and the air
compressor assembly delivers air to the combustor. As a result, the gaseous reactants
including syngas, steam, and air are introduced to the individual reactors in the reactor
assembly. Oxygen carrier particles are circulated at a specific rate through the
coordinated operation of 11 pneumatically actuated ball valves, 6 bed height monitors,
and 2 rotary solid feeders. The operations of the aforementioned assemblies are
controlled by the automation and control assembly through a central computer console
with a user friendly interface. During the reactor operation, the gas/solid sampling
assembly samples the compositions of gases at various locations of the looping reactors.
Solid samples are taken after the experiment.
The SCL process involves conversion and generation of hazardous gases such as
CO, H2, and CH4 at elevated temperature and pressure. For instance, the explosion limits
for hydrogen in air is 18.3 – 59%. Therefore, safety considerations play a vital role in the
design and operation of the SCL process systems. This is especially true for the testing
-221-
stages due to the relatively frequent start up and shut down tests involved compared to
the industrial operations. The safety features adopted in the sub-pilot unit include
automatic reactor flushing and preheating sequence during the start up stage for the safe
introduction of combustible gases; built-in high temperature oxygen sensors in the
reducer and the oxidizer to continuously monitor the oxygen level; pressure transducers
and pressure relief valves to prevent pressure build up; waste gas burners to prevent
leakage of combustible gases in case of reactor malfunction; ventilation fans to prevent
the build up of flue gas; and a fully automatic one-click shutdown sequence that
simultaneously turns off reactant gas, switches on the N2 flush gas, and decreases the
reactor temperature. Numerous hazardous/combustible gas leak detectors were also
installed in the vicinity of the reactor to alert the operator. These leak detectors were
linked to the central control console and will trigger the automatic shut down sequence
upon the detection of levels meeting or exceeding preset threshold levels.
The details of the sub-systems involved in the integrated unit are provided in the
following sections.
5.2.2 Gas Storage Assembly
To ensure the reactor gas supply, a set of changeovers (Model 8404) were
purchased from Scott Gas/Air Liquide for N2, CO, H2, and CH4 delivery. The changeover
system has two sides; each side holds 1 or 2 gas cylinders. During the operation, gas will
first be withdrawn from one side of the changeover until the cylinder pressure drops to a
preset value of 200 psi. Once the preset pressure is reached, the changeover considers the
-222-
tanks on the present side depleted and starts to withdraw gas from the cylinders installed
on the other side. Moreover, an alarm signal is sent to the control room to notify the
operator for gas tank replacement. The automatic gas switching function of the
changeovers ensures the gases are continuously delivered without interruption.
The design of the gas storage and delivery setup has multiple safety features for
unexpected system malfunctions. These features include flash arrestors to prevent back
flames, pressure relief valves to prevent pressure buildup in the gas delivery system, and
safety shut off valves to prevent leakage of gases at high flow rate. The changeovers
along with the gas cylinders, flash arrestors, pressure relief valves, and safety shutoff
valves are housed in a well ventilated 11 gauge steel gas cabinet (model 59-SC) to
prevent gas leakage to the room. Hazardous/combustible gas detectors are installed in the
gas storage room and gas cabinets to ensure safe operation.
5.2.3 Gas Mixing and Delivery Panel
The gas mixing panel is designed in collaboration with Air Products and
Chemicals Inc. The schematic diagram of the gas mixing panel is shown in Figure 5.4a,
with the photographs of the panel shown in Figure 5.4b.
As can be seen in Figure 5.4, the gas mixing panel regulates the flow rates of five
gases, i.e. H2, CO, CO2, CH4, and N2. A separate line of N2 flushing gas is also
incorporated in the gas mixing panel. One mass flow controller is installed on each gas
-223-
line to regulate the gas flow. A pneumatic solenoid valve is installed on each gas line.
Check valves are installed to avoid back flow of gases. The flow rate of individual gases
can either be altered by the touch screen installed on the front of the gas panel or be
modified on the central control consol. The gas mixing panel has the provision of sending
the gas either to the gas sampling system or the reactor system. Moreover, various built in
safety features are installed in the gas panel. These features include initial flushing
sequence to eliminate the oxygen in the reactor, reactor temperature monitor to prevent
the formation of explosive gaseous mixtures, reactor pressure monitor to prevent pressure
build up in the reactor, ventilation monitor to prevent the combustible gas flow in the
case of vent failure, and hazardous gas monitor to prevent the discharge of hazardous gas
to the environment.
5.2.4 Reactor Assembly
The reactor assembly is comprised of three reactors: the reducer, the oxidizer, and
the combustor. The three reactors are integrated and mounted on a three-storey support
structure. Solid circulation is enabled by the operation of valve system and rotary solid
feeders. The design of the reactor system is shown in Figure 5.1.
Reducer and Oxidizer
As discussed in Chapter 3, the reducer and the oxidizer are essential to the SCL
process. Both reactors adopt an identical reactor design, which is shown in Figure 5.5. As
can be seen in Figure 5.5, the reactor design is similar to that used in the bench scale
-224-
moving bed. There are a total of 16 3/8’’ ports installed on each reactor: 5 for gas
sampling, 7 for temperature measurement and solid sampling, 3 for bed height sensing,
and 1 for pressure monitoring. The design of the gas sampling ports is shown in Figure
5.6. The gas sampling port inlet, which is dent-like opening on the tip of a 1/8’’ SS304
tube, is positioned opposite to the direction of the particle flow to prevent plugging. A
particulate filter is also installed on the gas sampling port to filter out the particulates
entrained by the sample gas. A Teflon tube is used to send the sample gas to a 16-way
automatic valve system and then to a Varian CP-4900 MicroGC for gas analysis.
The bed height of the reactor is monitored using a resistance based technique.
Although iron oxide is not conductive at room temperature, its conductivity increases
significantly with temperatures. At 350 ºC and above, the conductivity of the particle will
reduce to mega Ohms level. Such a phenomenon is utilized to determine the bed height of
the reactor. Figure 5.7 shows the schematic of the system. The bed height control sensor
consists of a 1/16’’ SS304 probe installed inside a 1/8’’ quartz tube. The quartz tube is
then inserted into a ¼’’ tube, which is finally connected to the 3/8’’ port on the reactor.
By doing this, the probe is insulated from the reactor wall by the quartz tube. The tip of
the probe is located inside the reactor and can directly contact the particles.
In order to measure the bed height, a voltage (4 V) is applied to the probe via a 1
MΩ resistor. Since the reactor wall is grounded, when the probe is not in contact with the
particles, the voltage between the probe and the wall will be approximately to 4V. In the
case when the probe is connected to the wall via the hot particles, the voltage between the
-225-
wall and the tip of the probe will decrease notably. Therefore, the particle level in bed can
be easily determined based on the voltage between the probe and the reactor wall. Such
voltage signals are used to control the operation of the rotary disk and the valve system to
ensure that the bed height of the reactor is maintained at a desirable level.
Rotary Disk for Solid Metering
In order to control the flow of solids, a rotary disk solid feeder was designed and
constructed. A one inch thick ceramic disc is used to insulate the particles from the shaft.
A Pittman GM-9413-5 DC brush gear motor was originally used to drive the disc through
a shaft. The original technical drawing of the rotary disc is shown in Figure 5.8a. Further
information regarding the original design of the rotary disk can be found in the Ph.D
thesis of Dr. L.G. Velazquez-Vargas 197.
The rotary disc design also contains an upper matching plate with flanges to
prevent particles from falling off the ceramic disc, and a scraper to drive the particles
from one reactor to the other. This design has the advantages of minimal particle attrition
while transferring the solids and ease of particle metering by controlling the speed of the
motor.
The reactor along with the valves and rotary discs are assembled and initial solid
circulation was tested. It was found that particles, especially the broken particles, can fall
from the edge of the disc to its side or bottom. The particles in these interstitial spaces
between the rotating disk and the rotary feeder enclosure may cause the jamming of the
-226-
rotary disc. A large torque (up to 50 ft·lbf) is often required in order to operate the disc in
a continuous manner. It was further found that the original gear motor (Pittman GM9413-5) was incapable of delivering the required torque, resulting in unexpected failure
of the motor. Two modifications were attempted: 1. to reduce the torque requirement; 2.
to increase the maximum torque of the motor.
Several modifications to the original rotary disc design were performed: 1. the
gap between the side of the rotary disc and the enclosure is widened to at least three
particle characteristic lengths to allow better movement of fallen particles; 2. the metal
scraper above the rotary disc was shortened to allow better particle movement; 3. radial
grooves were cut on the bottom of the rotary plate to reduce the chance of particle
jamming below the plate, and the grooves would allow for the effective crushing of
particles in case of jamming. With these modifications, the maximum torque required to
operate the rotary disc was reduced to around 10 ft·lbf. These modifications are illustrated
in Figure 5.9. To increase the duty of the motor and to ensure better control of the rotary
disc, a new set of servo motors (JVL MAC800-D2) with 20:1 ratio gear heads was
purchased and installed. The motor, once installed with the gear head, is capable of
delivering a maximum continuous operating torque of 47 ft·lbf. Moreover, the precise
position, speed, and operating torque of the motor can be monitored on the central
computer console. Figures 5.8 and 5.9 show the updated schematic of the rotary disc
assembly.
Ball Valve Systems for Solid Transport
-227-
The valve system consists of three ball valves in series. The top valve receives the
solids, which are hot, and its function is just to hold the solids. The valve may be
permeable to the gas since the seals cannot sustain the high operating temperatures. In
order to transfer the solids to the next reactor valves 1 and 2 are opened simultaneously.
The solids will flow from the top to the bottom section, where they are held by valve 3
(bottom valve). The particle flow is fast and the amount of solids is small compared to
the size of valve 2. This consideration was made in order to preserve the seals of valve 2.
Valve 2 will prevent the mixing of the gases in the two reactors. Once the solids are in
the bottom section, valves 1 and 2 close and valve 3 opens. The solids then flow out of
the valve system and are transferred to the next reactor. Finally valve 3 is closed and the
cycle is repeated. Figure 5.10 illustrates the valve configurations and the operating
sequence of the valve system. As can be seen in Figure 5.10, the proposed valve system
operation is rather complicated. Simplifications to current valve system design include
using two high temperature valves operated like a lock hopper system or single rotary
solid valve to transfer solids while preventing gas leakage. These designs will be
discussed in Section 5.4.
As will be discussed in Section 5.3, the valve system was successfully operated
using the designed sequence with the assistance of the automation and control assembly.
Further, the continuous solid transport at room temperature was successfully tested by
operating the valve systems and rotary feeders simultaneously.
-228-
5.2.5 Automation System
Automation and control assembly is comprised of Labjack data acquisition and
control boards, relay boards, thermocouple boards, and a central computer console.
DAQ-Factory software is used to control the various equipments and to acquire data. One
control room was constructed to house the central computer console and data analysis
instruments. Two control boxes were mounted outside of the control room to house the
various computer controlled boards that interact among the computer and the equipments.
A fully automatic, user friendly control sequence for the reactor operation is developed.
Figure 5.11 shows the user interface of the control sequence of the central computer
console.
5.2.6 Steam Generator and Air Compressors
The oxidizer operation requires a constant flow of steam and for this purpose, a
Sussman MBA6 Steam generator was purchased. It has the capacity to maintain a
constant flow rate of 18 lbs/hr operating in the range of 85 – 100 psig. Figure 5.12 shows
the picture of the steam generator. The brake horse power (BHP) rating of the generator
is 0.6. The generator comes with an internal control circuit that can monitor the steam
flow rate independent of the heaters for steam generation.
The combustor serves the purpose to convert and convey the iron based looping
particles. To perform this operation, a large flow rate of air for a short period is required.
To meet this requirement, an Atlas Copco GA15FF air compressor was purchased from
Air Technologies. This has a maximum operating pressure of 100 psig and when coupled
-229-
with the already existing Atlas Copco GA7FF air compressor, the system provides an
ample supply of air for the combustor. The two air compressors are connected to a single
reservoir. The GA15FF compressor has high generation power while the GA7FF
compressor has a higher pressure limit (120 psig). The current connection aids utilization
of both the benefits. When a large air supply is required, the new compressor is in
standby till the pressure in the reservoir drops below 90psi. This configuration reduces
power consumption. Figure 5.13 shows the picture of this configuration.
The various aforementioned sub-systems have been integrated successfully. The
next section discusses preliminary results obtained from the sub-pilot unit.
5.3 Preliminary Reactor Tests
The reactor system was fully assembled in November 2008. Prior to continuous
operation, shakedown and preliminary reactor tests are performed.
5.3.1 Reactor Leak Test
Since hazardous and combustible gases are converted and generated in the SCL
unit, minimal gas leakage needs to be ensured. Extensive leak tests are performed on
both individual parts of the reactor and the integrated unit. Before the leak test, the part or
unit being tested is sealed and pressurized to 50 psig with helium. The helium gas is then
stopped and the time for the pressure to decrease to 45 psig is recorded. The gas leakage
rate can then be calculated by:
L = 5V /(t 50 − >45 × 14.7)
-230-
L is the gas leakage rate in ml/min, V is the unit volume in ml, and t50Æ
45
is the
depressurization time in minutes. Since leakage of carbon monoxide from the reducer is
the main concern to the safety of the operator, the maximum leakage rate is calculated
based on a rate above which the CO concentration within 5 ft of the reactor will exceed
the OSHA workplace standard of 50 ppm. Based on the aforementioned criteria, the
maximum leakage rate for the integrated unit should be less than 3 liter/min. In the case
when the unit tested has a leakage rate comparable to 3 liter/min, a Varian CP87610
helium leak detector and a Snoop liquid leak detector are used to identify the leak.
Actions are taken to either fix the leak or replace the unit.
During the initial leak test, significant leakage was found at the joints between the
valves. These joints are screw on type and high temperature anti-seize was used to
prevent damage to the threads on the screw. It was found that the anti-seize does not seal
the joints. High temperature sealant and nickel impregnated Teflon tapes were applied to
the joints for sealing. It was found that nickel impregnated Teflon tapes provide the best
sealing. After sealing the joints, the valve assemblies are tested. Several valves were
found to be defective and are being replaced. The leak test was then extended to the
rotary solid feeder. Minor leakage was identified at the bottom screws through which the
rotary disk shaft is fastened to the rotary solid feeder. Again, Teflon tape was applied to
the screws to fix the leak. The reactor was also tested separately and the leak was
undetectable. Finally, the valve assembly, the rotary solid feeder, and the reactor were
integrated and the leak rate was tested. It was found that the leakage rate was lower than
50 ml/min for the integrated unit. Such a leakage rate is deemed acceptable. Leakages of
-231-
the reactor before and after high temperature operation (up to 950 ºC) were compared and
no notable difference was identified.
5.3.2 Rotary Solid Feeder Test and Solid Flow Calibration
The solid flow rate determines the solid residence time in each reactor as well as
solid and gas conversions. Therefore, accurate solid flow rate control is key to the
successful operation of the SCL unit. Although the valve assembly can hold the solid for
a certain period of time, the solid flow rate is ultimately controlled by the rotary solid
feeder. Therefore, the solid flow rate calibration was performed on the rotary solid feeder.
A linear relationship between the motor speed and the solid flow rate was found. It was
determined that the rotary disk rotation speed should be maintained at around 0.5 rpm.
5.3.3 Combustor and Particle Attrition Test
Particle attrition has been tested in a small entrained bed reactor with a height of 2
meters. The results are reported in Chapter 2. The particle attrition test was repeated in
the combustor of the integrated SCL unit. The test serves three purposes: firstly, the
attrition of the particle in a significantly larger entrained bed can be determined and
compared to the results from the small unit; secondly, the SCL combustor operation can
be validated; and thirdly, the hydrodynamic properties of the particles can be determined.
Experimental
As can be seen in Figure 5.1, the particle enters the combustor through a slant
stainless steel tube where it is then pneumatically conveyed vertically back to the reducer
inlet to complete the redox/solid conveying cycle. The combustor, air inlet, and slant tube
-232-
are connected by a “Y” shaped tube. A programmed gate valve is mounted on the slant
pipe to control the particle feed to the vertical section of the combustor. A distributor
made of Hastelloy X is installed at the bottom of the combustor to ensure even
distribution of the air. The main section of the combustor is a vertical pipe with a 2 in ID
and a height of 20 ft. A 900 elbow and a horizontal pipe are connected to the top of the
vertical pipe with a cyclone connected on the other end of the horizontal pipe. The
cyclone allows the separation of the fines and ensures that larger particles will be reintroduced to the reducer. All the parts for the combustor are made of 304L stainless steel.
An air compressor, with a capacity of ~30 L/s at 180 psig, is connected to the bottom of
the distributor to provide air for both particle conveying and oxygen carrier combustion
purposes.
During the reactor operation, particles are loaded into the slant pipe from the
bottom of the oxidizer with the programmed ball valve. After the particles in the slanted
pipe have accumulated to a preset amount, all the particles are introduced to the bottom
section of the vertical combustor by opening the programmed gate valve for a short
period of time (< 10s). After the programmed ball valve is closed, the air source from the
compressor is introduced into the combustor to convey all the particles at the bottom of
the vertical pipe into the reducer inlet. Exhaust air coming out of the cyclone will then be
introduced to a powder separator for the separation of fine powders before ventilation.
The powder separator is not shown in Figure 5.1.
Results and Discussion
-233-
A. Pneumatic Conveying of Particles
The pneumatic conveying system consists three parts: (1) vertical flow through
the combustor pipe, (2) change in direction through a 900 elbow, and (3) horizontal flow.
The flow pattern in the gas-solid pipe is a dilute suspension flow with a very low particle
concentration. Several sets of particles are selected to test the combustor operation. The
particles are pellets processed from iron oxide and an inert support with a density of
2500 kg / m 3 . A 3kW rotary pelletizer (ZP-35) is used to fabricate pellets with a
production rate of up to 15 kg/hour. The pellets have a cylindrical shape with a diameter
of 5 mm and a thickness of 1-5 mm depending on the setting of the pelletizer. Before
testing, the pellets are sintered at 900
0
C for more than 20 hours to achieve better
physical and chemical properties.
It was determined through experiments that the terminal velocity for the pellets
with a diameter of 5 mm and a thickness of 1.5 mm, or type A particles, is around 9.3 m/s.
For this type of particles, the pneumatic conveying system needs to be operated at a gas
velocity ~1 m/s greater than its terminal velocity since horizontal pneumatic transport
requires higher gas velocity than vertical pneumatic transport. The greater gas velocity
requirement is due mainly to higher drag and frictional forces in the horizontal section of
the combustor. However, the effects of the drag forces and the frictional forces in the
horizontal pneumatic transport are not significant since the length of the horizontal
section is relatively short. It was also determined that the terminal velocity for the pellets
with a diameter of 5 mm and a thickness of 4.5 mm, or type B particles, is around 15 m/s.
For type B particles, the pneumatic conveying system can be operated at a gas velocity
-234-
slightly higher than its terminal velocity due to the better flow properties of the particle in
the horizontal pipe. The higher terminal velocity and better flow properties of type B
particles can be explained by it near spherical shape which allows for fluidizeability,
lower effects of drag forces, and less friction than type A particles.
B. Particle Attrition
In order to reduce the particle purging rate, it is desirable for the oxygen carrier
particles to maintain their chemical reactivity and physical integrity for multiple cycles.
Since particles are in vigorous motion in the combustor, the attrition rate of the particle is
an important parameter for the SCL combustor design and particle optimizations. Particle
attrition affects the flow performance and economics of the SCL process. A low attrition
rate of the particles is highly desirable. Due to the turbulent movements at high gas
velocities, the frequent particle-particle/ particle-wall collisions during the pneumatic
conveying step in the combustor and the particle separation step in the cyclone play an
important role for particle attrition, which is defined as the degradation of particles due to
mechanical stress 199. Abrasion and fragmentation of the particles are the main modes of
the attrition
200, 201
. The attrition effects due to the combustor and the cyclone are
measured in the SCL process. For the attrition test in combustor, particles are loaded at
the bottom of the vertical pipe and then conveyed out of the horizontal part of the
combustor. These particles are subsequently collected using a bucket. The particles are
then sieved and weighed according to different size fractions. All the particles are then
mixed and reloaded into the combustor for the next conveying cycle. To test the attrition
in both the combustor and the cyclone, a similar testing procedure is adopted. The only
-235-
difference is that the particles are collected after the cyclone. The attrition rates for both
type A and type B particles were tested through 20 runs.
Figure 5.14 and Figure 5.15 shows the combustor attrition test results for type A
particles and type B particles respectively. A superficial gas velocity of 10 m/s is used to
convey the type A particles. From Figure 5.14, the particle attrition rate is 0.56% per
cycle when the cutoff is set to be 600 μm , i.e., particles smaller than 600 μm will be
purged out from the system. A superficial gas velocity of 15 m/s is used to convey the
type B particles. From Figure 4, the particle attrition rate is 0.23% per cycle when the
cutoff is set to be 600 μm . The particle attrition rate is around 0.38% per cycle when the
cyclone is used for the test at the same experimental condition, which means an
additional attrition of 0.15% was introduced due to the addition of the cyclone. The type
B particles with 5 mm diameter and 4.5 mm thickness perform better on attrition than
type A particles due to its close-to-spherical shape. Such a shape leads to reduced effects
of the drag forces and the various particle-particle and particle-wall frictions. As can be
seen, both type A and type B particles possess excellent physical strength. Moreover, the
particle attrition in the sub-pilot SCL combustor is comparable to that in the small
entrained bed reactor.
5.3.4 Test of the Integrated Unit
Integrated unit testing was performed. Around 15 kg of Type B composite particle
was loaded to the reducer. The reactor was heated up to 750 ºC and was operated using
the pre-programmed control sequence. The solid transfer speed of the rotary disk was set
-236-
at 100 g/hour and the valve system was programmed at a sequence identical to that
explained in Section 5.2.4. The valve system, rotary solid feeder, and the reactors worked
harmoniously for more than 4 hours. This indicates that the control system and the
various parts of the reactor meet the design standard. The particles were able to be
transferred from the reducer to the oxidizer smoothly. However, the particles
occasionally got stuck in the slant pipe that transfers the particles from the oxidizer outlet
to the combustor inlet. An additional air inlet was installed on the slant pipe. Air is
introduced to help the particle flow.
5.4 Improvements and Future Tests of the Sub-Pilot Unit
The preliminary tests conducted on the sub-pilot unit provide valuable
information regarding the functionality of the various parts of the reactor. Several
modifications that can improve performance of the current unit are identified.
5.4.1 Valve System Design
The current valve system is comprised of three low temperature ball valves with
maximum operating temperature of 230 ºC. However, such a design is space consuming.
One set of the three-valve system requires a minimum of 30’’ of clearance, which is
comparable to the height of the reaction zone of the reducer. Moreover, the high
temperature particle circulation test shows that the particles entering the first ball valve
can reach temperatures as high as 500 ºC. The high temperature particles can potentially
damage the Teflon O-ring that seals the shaft of the low temperature ball valve, causing
the leakage between the inside of the valve and the environment outside. An improved
-237-
valve system design uses two high temperature ball valves sealed with graphite. Such ball
valves can withstand 550 ºC. The two ball valves can work in a stepwise mode similar to
a lock hopper system: Step 1, the bottom ball valve is kept closed, the top ball valve is
opened to dump the particle to the bottom ball valve; Step 2, the top ball valve is closed;
Step 3, the bottom ball valve is opened to discharge the particles to the next unit; Step 4,
the bottom ball valve is closed and the valve system is ready for the next operating cycle.
The two ball valve system simplifies both the valve system design and the valve
operating sequence. Moreover, the reliability of the system will increase as the valves are
always operated at temperatures lower than their maximum operating temperature.
5.4.2 Bed Height Control
The resistance based bed height control systems have been used in the bench unit
and worked well. However, the integrated unit involves simultaneous operation of three
units. Therefore, better monitoring of the bed height is desirable. The most reliable way
to monitor the particle movement is through visualization. It is proposed that a full view
sight flow indicator produced by Jacoby-Tarbox be installed between the bottom ball
valve and the inlet of the reactor. The sight flow indicator is a 3’’ pipe installed with
quartz side window. Therefore, particle flow towards the reactor can be visualized. With
the option of directly observing the solid flow, the operation of the sub-pilot unit will be
simplified.
-238-
5.4.3 Future Test Plan
Syngas and light hydrocarbons of various compositions will be tested in the subpilot unit. The unit will be continuous operated for 100 hours at minimum. Information
regarding composite particle stability and long term performance, gaseous fuel
conversions, hydrogen purity, and looping unit scale up effects will be obtained through
the tests. Such information will be incorporated to the ASPEN Plus® process simulation
to evaluate the process performance. It is expected that the sub-pilot scale tests will pave
a way for the 250 kWth – 1 MWth pilot scale demonstrations in the future.
5.4.4 Coal Direct Chemical Looping Applications
Since the CDCL process and the SCL process have identical reactor configuration
schemes, the current sub-pilot unit only needs slight modifications for CDCL process
tests. The modification mainly lies in the reducer in which an injection port for
pulverized coal needs to be installed. The reducer design is provided in Chapter 4.
5.5 Concluding Remarks
A sub-pilot scale chemical looping demonstration unit is designed, assembled, and tested.
The unit integrates the reducer, the oxidizer, and the combustor in a manner identical to
that proposed in the SCL and the CDCL process. Various parts of the reactor are tested
and determined to be fully functional. The unit is also successfully operated in an
integrate mode. Several improvements to the original design are made. The sub-pilot unit
is ready for tests under reaction conditions. It is expected that valuable data for further
scale up and commercialization of the novel chemical looping gasification processes can
-239-
be obtained from the continuous operations of the sub-pilot demonstration unit.
-240-
a
b
Figure 5.1 Sub-Pilot Scale Demonstration Unit for SCL Process a. Schematic Flow
Diagram; b. Photograph
-241-
Regulate Solid Flow
Automation
and Control
Setup
Safety Feature
• Leakage
• Pressure
• Temperature
• Vent Operation
Control Air Flow
Period and Rate
Gas Delivery
System
Set Solid Transfer Rate
Reactor Temperature control
Mix and Deliver Fuel gas to
Reducer
Steam
Generator
Deliver Steam to Oxidizer
Air
Compressor
Provide Air for Combustor
Sample
Analysis
System
Online gas analysis using
Gas Sampling Port
Offline solid analysis using
Solid Sampling Port
Figure 5.2 Overall Arrangement of the Sub-Pilot SCL System
-242-
Reactor System
Door
Door
Door
Door
Door
Air
Compressors
Control
room
Room 2
Tools
Steam
Generator
Phase I unit
Door
Door
Pelletizer
Door
Gas Panel
Door
First Aid Kit
Door
Gas Cabinets
Room 3
Door
Hall
Way
Room 1
Figure 5.3 Physical Locations of the Various Sub-Systems in the Sub-Pilot Scale Unit at
the OSU West Campus Demonstration Site
-243-
Gases From cylinders
CO
H2
CH4
CO2
N2
Automation
System
ng
hi
us
Fl
Solenoid valve
MFC MFC MFC MFC MFC
Check valve
Process gas
GC sampling
Signals from reactor
• Gas Monitor signal
• Pressure transducer signal
• Temperature signal
• Vent Flow signal
a
b
Figure 5.4 a. Schematic Diagram of the Gas Mixing Panel Design; b. Photograph of the
Gas Mixing Panel
-244-
Reducer/Oxidizer
Reactor with Heater
Pressure
Transducer
Bed Height
Control
Sensor ports
32
TC/Solid
Sampling
Ports
Gas
Sampling
ports
Figure 5.5 Design of the Reducer and Oxidizer
-245-
Installed Reactor
Figure 5.6 Design of the Gas Sampling Ports
-246-
3/8’’ SS Tube
1/4 ” Tube
Particulate Filter
3/8” Tube
2 um Filter
1/8 ” Tube
1/8’’ Teflon Tube
Reactor Wall
1/8’’ SS Tube
Measurement
Circuit Sensor
Reactor
Resistor
Schedule 40 pipe
Resistor
V
Inner 3 in. pipe
V
Particles
Reactor
Wall
Steel
Tubing
Probe
Particles
Figure 5.7 Bed Height Control System
-247-
From the
first
reactor
Electrical
Heaters
Heated Section
Scraper
Grooved Plate
Ceramic Disk
Motor
Shaft
To the
second
reactor
3 in. Pipe
Cross Section view
Servo Motor
b
a
c
Figure 5.8 a. Original Design of the Rotary Disk Solid Feeder; b. Updated Design of the
Rotary Disk Solid Feeder; c. Photograph of the Assembled Solid Feeder with Servo
Motor Installed
-248-
Exit to the Next
Reactor
Scraper Shortened to Allow
Smooth Solid Flow
Rotary Disk – Top View
Portion Removed from
the Original Design to
Ensure Smooth Solid
Flow
Rotary Disk – Bottom View
Groves with Sharp Edges,
Rotates Clockwise from this
View
Carbon Steel Wall of the
Rotary Disk Enclosure
Figure 5.9 Modifications to the Rotary Solid Feeder for Smoother Solid Flow
-249-
Valve Sequencing
Valve Layout
Valve
Valve 1
Pneumatic
Valve 2
Actuators
Valve 3
Step 1:
Solids on
hold
Step 2:
Solids In
Step 3:
Solids out
Figure 5.10 The Schematic of the Valve System
-250-
Installed Valves
Figure 5.11 User Interface of the Control Sequence
-251-
Figure 5.12 Steam Genertor
-252-
Figure 5.13 Air Compressors
-253-
Percent of Unbroken Pellets
100
95
90
>1.4 mm
>1.0 mm
>600 um
85
0
5
10
15
20
Numbers of Conveying Cycles
Figure 5.14. Attrition Test Results Using Type A Particles with 5 mm Diameter and 1.5
mm Thickness
-254-
Percent of Unbroken Pellets
100
99
98
97
>1.4 mm
>1 mm
96
>600 um
95
0
5
10
15
20
Numbers of Conveying Cycles
Figure 5.15. Attrition Test Results Using Type B Particles with 5 mm Diameter and 4.5
mm Thickness
-255-
CHAPTER 6
NOVEL APPLICATIONS OF CHEMICAL LOOPING,
CONCLUSIONS AND RECOMMENDATIONS
In this chapter, integration of the chemical looping gasification to conventional
coal-to-liquid process is exemplified. This is followed by concluding remarks and
recommendations for future work.
6.1 Novel SCL Applications – A Coal-to-Liquids Configuration
6.1.1 Process Overview
The SCL process can be integrated or retrofitted to existing processes to improve
their overall energy conversion efficiencies. This section exemplifies a novel
configuration that integrates the SCL process to the traditional coal-to-liquids (CTL)
process.
There are several different schemes to incorporate the syngas chemical looping (SCL)
process into the CTL process. In a conservative scheme, the SCL process is used as a
retrofit to the traditional CTL process in which the role of the SCL process is to generate
additional hydrogen for Fischer-Tropsch (F-T) synthesis. The schematic flow diagram for
such a scheme is illustrated in a process configuration as given in Figure 6.1.
-256-
In a conventional CTL plant using cobalt based catalysts, the syngas generated
from the gasifier has a hydrogen concentration (30–40%) which is significantly lower
than the required H2 concentration (67%) for the liquid fuel synthesis. This shortage in
hydrogen concentration is usually compensated in a traditional CTL process by additional
steps to partially shift the CO in the syngas stream. Meanwhile, the Fischer-Tropsch (FT) reactor converts only part of the syngas (60 –85%) to a wide variety of hydrocarbons
ranging from methane to hard wax. The gaseous hydrocarbon fuels and unconverted
syngas are considered as by-products and a significant portion of these gaseous
compounds are combusted to generate electricity.
In the SCL-CTL configuration, the unconverted syngas and gaseous products from
the Fischer-Tropsch reactor are introduced to the reducer of the SCL system. These
gaseous fuels are converted to carbon dioxide and water through the following reaction.
CxHyOz + (2x+y/2-z)MO → (2x+y/2-z)M + xCO2 + y/2H2O
(6.1)
Here, MO and M refer to the different iron oxide phases. Reaction (6.1) reduces the iron
oxide from higher oxidation states to lower oxidation states. The reduced iron particles
are then introduced to the oxidizer where they are reacted with steam to produce pure
hydrogen and regenerate the iron oxide (Reaction 6.2).
M + H2O → MO + H2
(6.2)
-257-
As can be seen, the major feed gases for the SCL reducer is the by-products from the F-T
reactor. Meanwhile, the large amount of the medium pressure (~25 atm) steam generated
by the low grade heat from the F-T reactor provides ample steam supply required for the
SCL oxidizer. Therefore, through the utilization of the SCL system, hydrogen, an
essential feedstock for the CTL, is generated from the by-products of the F-T synthesis.
The liquid fuel yield of the CTL process is thus improved. The integrated carbon capture
capability of the SCL renders the SCL-CTL configuration even more attractive under a
carbon constrained situation.
6.1.2 Mass/Energy Balance and Economic Evaluation
A system analysis based on ASPEN Plus® and flow sheet analysis is conducted on
the SCL-CTL configuration to evaluate the performance of the SCL-CTL process relative
to the conventional CTL process using assumptions identical to those in Chapter 3.
Several additional assumptions used in the CTL process analysis include:
1) Cobalt based catalysts can achieve 75% per pass conversion on syngas with a
H2/CO ratio of 2:1 while having an 85% selectivity for liquid phase hydrocarbons.
2) All the combustible gas in the traditional CTL process is used for electricity
generation.
3) The higher heating value of Naphtha product is 31.6 MJ/liter and the higher
heating value of jet fuel/diesel product is 35.5 MJ/liter.
-258-
4) Steam of 200 ºC, 20 atm is generated from the high temperature gas streams (200500 ºC) and the exothermic reactions that occur between 200 and 500 ºC.
The mass and energy balances on both traditional CTL and SCL-CTL process are
shown in Tables 6.1. The plant size is 1000 MWth.
Table 6.1 shows that the liquid fuel production efficiency for the SCL-CTL process is
in excess of 40%. This production efficiency is nearly 10% higher than the traditional
CTL process, which is at 37.7%. When the extra electricity generation is also taken into
account, the system efficiency for the SCL-CTL process would be at 41.3% compared to
38.2% for the conventional CTL process. Thus, the SCL enhanced CTL process yields a
significantly improved efficiency for the liquid fuel production over the traditional CTL
processes. The reasons can be summarized by:
1) Improved energy utilization scheme: The SCL-CTL process utilizes both the
byproducts (C1-C4) and low grade energy (steam) from the Fischer-Tropsch (F-T)
reactor and effectively converts them to H2, which is then used to synthesize the
desired
liquid
fuel
products.
Thus,
the
SCL-CTL
process
is
more
thermodynamically efficient compared to the traditional CTL process for liquid
fuel syntheses;
2) Simpler process scheme: The presence of the SCL system simplifies the
traditional CTL process in that the SCL system can perform multiple functions
and effectively replace such units in the traditional CTL process as the water gas
-259-
shift reactor, pressure swing adsorption unit, and multiple F-T product upgrader.
Further, the load of CO2 separation using such units as Selexol can be reduced by
66%. The SCL system can also realize the 100% carbon capture without
additional economic burden.
An economic analysis on a commercial scale coal to liquid plant using either the
conventional CTL or the SCL-CTL technology indicates that although the capital
requirement for a SCL-CTL plant is slightly higher than that for a traditional CTL plant
with identical coal processing capacity, the normalized capital requirement based on the
liquid fuel production capacity ($/daily barrel) for the SCL-CTL process is notably lower.
The normalized operating cost for the SCL-CTL process is also reduced resulting from
less coal input and decreased parasitic energy consumption. The increase in liquid fuel
yield as well as the decrease in operating cost renders the SCL-CTL process more
economical than the traditional CTL process. Thus, the integration of SCL to a
conventional CTL plant has the potential to significantly increase the profitability of the
plant. Such a configuration was explored by Noblis Systems and further information can
be obtained from their report for the USDOE 141.
6.2 Concluding Remarks
Two novel chemical looping gasification processes, the syngas chemical looping
(SCL) process and the coal direct chemical looping (CDCL) process, are developed for
hydrogen and electricity co-production from carbonaceous fuels. Through the assistance
-260-
of iron oxide based oxygen carrier particles, both processes are highly efficient with zero
carbon emissions.
The performance of the oxygen carrier particle is the key to both processes. A
novel iron oxide based composite particle is developed. The physical and chemical
properties of the particle including compressive strength, attrition rate, reactivity and
recyclability are tested. Reduction and regeneration of the particle in an integral bed is
performed. More than 99.7% syngas is converted during the reduction step. During the
regeneration step, hydrogen with an average purity of 99.8% is produced. The particle is
deemed suitable for the chemical looping gasification processes.
The SCL process is extensively studied both analytically and experimentally.
Thermodynamic analysis shows that a countercurrent moving bed design is suitable for
both the reducer and the oxidizer. ASPEN Plus® simulation further suggests the
optimum operating conditions and pollutant control techniques for the SCL process.
Experiments are carried out in a 2.5 kWth bench scale reactor. More than 99.5% of the
syngas is converted during the reducer test. The hydrogen generated during the oxidizer
test has an average purity higher than 99.95%. The particles can also sustain the high
operating temperature of combustor. ASPEN Plus® simulation shows that the SCL
process can improve the current coal to hydrogen by 4 – 10% while capturing 100% of
the carbon in coal. Significantly increase in liquid fuel yield can be realized when the
SCL process is retrofit to a conventional coal-to-liquid plant.
-261-
The CDCL process is also tested in the bench scale moving bed reactor. More
than 90% conversions for various types of coal char are achieved. ASPEN Plus®
simulation is used to analyze the fates of pollutants. It is also used for process
optimization. The process simulation using ASPEN Plus® shows that the hydrogen
production efficiency for the CDCL process can reach nearly 80%.
A 25 kWth sub-pilot scale chemical looping unit is designed and constructed. It
demonstrates the chemical looping gasification processes in an integrated, continuous
manner. The unit, which is comprised of six sub-systems, has been completely assembled.
Preliminary tests including reactor leakage testing, solid flow calibration, particle
hydrodynamic studies, and integrated reactor operations are performed. The test results
show that the sub-pilot unit meets the design standard. The unit is ready for the SCL
demonstration. It can also be used to demonstrate the CDCL process with minor
modifications.
6.3 Recommendations
The development of the SCL and CDCL processes evolved from a new idea to
successful bench scale test over the last six years. A 25 kWth sub-pilot unit has been
constructed and continuous tests are underway. ASPEN Plus® process simulation and
preliminary economic analysis show that the processes are highly attractive both
technically and economically. The future work should focus on further validating the
feasibility of the processes with commercialization as the ultimate goal. To achieve such
-262-
a goal, future R&D efforts should concentrate on both the process scale up and particle
development.
Following the successful operation of the current sub-pilot unit, the current
techno-economic analysis should be updated. Provided that the updated analysis confirms
the attractiveness of the chemical looping processes, a pilot scale chemical looping unit
with a capacity between 250 kWth to 1 MWth will be designed constructed. An outside
Architectural and Engineering firm will be hired to perform the unit design and
construction. The company will also perform economic analysis for a commercial
chemical looping unit. The pilot unit will be operated at elevated pressure. Moreover, the
proposed heat integration scheme will be demonstrated without providing external heat.
Heat loss for each reactor will be evaluated. The pilot scale demonstration will provide
convincing operational data for the design and construction of a small commercial
chemical looping plant.
In addition to scale up demonstrations, studies on the fuel-oxygen carrier particle
reaction mechanism, interactions among pollutants, ash, and particle, and particle
reactivity, recyclability, and physical stability improvements are of vital importance to
the success of the looping process. Optimization of the looping particle will provide
strong support to the demonstration and can significantly improve the looping process
economics.
-263-
Coal Feed In
Naphtha Out
Diesel Out
Net Electricity
Traditional CTL
3190 ton/day
2579 bbl/day
5511 bbl/day
5 MW
SCL-CTL
3190 ton/day
2811 bbl/day
6020 bbl/day
2 MW
37.7
38.2
41.1
41.3
Fuel Conversion Efficiency (%)
Overall System Efficiency (%)
Table 6.1 Overall Energy Input/Output for the SCL-CTL Process
-264-
Figure 6.1 Syngas Chemical Looping enhanced Coal-to-Liquids (SCL-CTL) process.
-265-
BIBLIOGRAPHY
1.
World Meteorological Organization, Atmospheric Carbon Dioxide Levels Highest
on Record. In ScienceDaily, 2006; Vol. 2006, November 4.
2.
Gains,
National Oceanic and Atmospheric Administration After Two Large Annual
Rate
of
Atmospheric
CO2
Increase
Returns
to
Average.
http://www.publicaffairs.noaa.gov/releases2005/mar05/noaa05-035.html (Accessed May,
1st, 2008),
3.
Hansen, J.; Sato, M., Greenhouse Gas Growth Rates. Proceedings of the National
Academy of Sciences of the United States of America 2004, 101, (46), 16109-16114.
4.
Jacobsson, S.; Johnson, A., The Diffusion of Renewable Energy Technology: an
Analytical Framework and Key Issues for Research. Energy Policy 2000, 28, (9), 625640.
5.
McVeigh, T.; Burtraw, D.; Darmstadter, J.; Palmer, K., Winner, Loser, or
Innocent Victim? Has Renewable Energy Performed as Expected? Solar Energy 2000, 68,
(3), 237-255.
6.
Ewing, R. C., Nuclear Fuel Cycle: Environmental Impact. Mrs Bulletin 2008, 33,
(4), 338-340.
-266-
7.
Energy Information Administration (EIA), International Energy Outlook 2007.
USDOE, 2007.
8.
International Panel for Climate Change (IPCC), Climate Change 2007: Synthesis
Report. IPCC, 2007.
9.
Ying, W.; Zhu, W., Around the Markets: Future for Coal Brighter. International
Herald Tribune April 16th, 2007.
10.
Britannica, Coal Mining. http://www.britannica.com/eb/article-9110682/coal-
mining (Accessed April 28, 2008),
11.
Britannica Coal. http://www.britannica.com/eb/article-9110442/coal (Accessed
April 28, 2008),
12.
Ashton, T. S., The Industrial Revolution, 1760-1830. Oxford University Press:
New York, 1997.
13.
Church, R., The History of the British Coal Industry, vol. 3. 1830-1913.
Clarendon Press: Oxford, 1986.
14.
Stultz, S. C.; Kitto, J. B.; Babcock & Wilcox Company., Steam, Its Generation
and Use. 40th ed.; Babcock & Wilcox: Barberton, Ohio, 1992.
15.
Miller, B. G.; NetLibrary Inc., Coal Energy Systems. Elsevier: Boston, 2005.
16.
Yeh, S.; Rubin, E. S., A Centurial History of Technological Change and Learning
Curves for Pulverized Coal-Fired Utility Boilers. Energy 2007, 32, (10), 1996-2005.
17.
Massachusetts Institute of Technology, The Future of Coal: Options for a
Carbon-Constrained World. Massachusetts Institute of Technology: Somerset, MA, 2007.
18.
Bugge, J.; Kjaer, S.; Blum, R., High-Efficiency Coal-Fired Power Plants
Development and Perspectives. Energy 2006, 31, (10-11), 1437-1445.
-267-
19.
Stringer, J.; Ruth, L. A., Future Perspectives Including Fuel Cells, Gas Turbines,
USC and HGCU the US Perspective. Materials at High Temperatures 2003, 20, (2), 233239.
20.
Bradley, M. J.; Jones, B. M., Reducing Global NOx Emissions: Developing
Advanced Energy and Transportation Technologies. Ambio 2002, 31, (2), 141-149.
21.
Banales-Lopez, S.; Norberg-Bohm, V., Public Policy for Energy Technology
Innovation - A Historical Analysis of Fluidized Bed Combustion Development in the
USA. Energy Policy 2002, 30, (13), 1173-1180.
22.
US
Department
of
Energy
Fluidized
Bed
Technology
-
Overview.
http://www.fossil.energy.gov/programs/powersystems/combustion/fluidizedbed_overvie
w.html (Accessed April 27th, 2008).
23.
Rockey, J. M.; Weinstein, R. E., Gas Turbines for Advanced Pressurized
Fluidized Bed Combustion Combined Cycles (APFBC). In 15th International Conference
on Fluidized Bed Combustion, ASME: Savannah, Georgia, 1999.
24.
U.S. Clean Coal Technology Demonstration Program., The JEA Large-Scale CFB
Combustion Demonstration Project: a Report on a Project Conducted Jointly under a
Cooperative Agreement Between the U.S. Department of Energy and JEA. In Topical
report ; no. 22., Clean Coal Technology: Washington, D.C., 2003.
25.
Watson, J., Advanced Cleaner Coal Technologies for Power Generation: Can
They Deliver? In British Insitute of Energy Economics(BIEE) Academic Conference, St.
John's College, Oxford, 2005.
-268-
26.
Minchener, A. J., Fluidized Bed Combustion Systems for Power Generation and
Other Industrial Applications. Proceedings of the Institution of Mechanical Engineers
Part a-Journal of Power and Energy 2003, 217, (A1), 9-18.
27.
Gupta, H.; Thomas, T. J.; Park, A. H. A.; Iyer, M. V.; Gupta, P.; Agnihotri, R.;
Jadhav, R. A.; Walker, H. W.; Weavers, L. K.; Butalia, T.; Fan, L. S., Pilot-Scale
Demonstration of the OSCAR Process for High-Temperature Multipollutant Control of
Coal Combustion Flue Gas, using Carbonated Fly Ash and Mesoporous Calcium
Carbonate. Industrial & Engineering Chemistry Research 2007, 46, (14), 5051-5060.
28.
Fan, L. S.; Ghosh-Dastidar, A.; Mahuli, S. Calcium Carbonate Sorbent and
Methods of Making and Using Same. 1998.
29.
Fan, L. S.; Jadhav, R. A., Clean Coal Technologies: OSCAR and CARBONOX
Commercial Demonstrations. Aiche Journal 2002, 48, (10), 2115-2123.
30.
Gupta, H.; Benson, S. A.; Fan, L. S.; Laumb, J. D.; Olson, E. S.; Crocker, C. R.;
Sharma, R. K.; Knutson, R. Z.; Rokanuzzaman, A. S. M.; Tibbets, J. E., Pilot-scale
Studies of NOx Reduction by Activated High-Sodium Lignite Chars: A Demonstration of
the CARBONOX Process. Industrial & Engineering Chemistry Research 2004, 43, (18),
5820-5827.
31.
Taerakul, P.; Sun, P.; Golightly, D. W.; Walker, H. W.; Weavers, L. K.; Zand, B.;
Butalia, T.; Thomas, T. J.; Gupta, H.; Fan, L. S., Characterization and re-use Potential of
By-Products Generated from the Ohio State Carbonation and Ash Reactivation (OSCAR)
process. Fuel 2007, 86, (4), 541-553.
32.
Bryant, E. A., Climate process & change. Cambridge University Press: New York,
1997.
-269-
33.
Armor, J. N., Addressing the CO2 Dilemma. Catalysis Letters 2007, 114, (3-4),
115-121.
34.
Abu-Zahra, M. R. M.; Niederer, J. P. M.; Feron, P. H. M.; Versteeg, G. F., CO2
Capture from Power Plants Part II. A Parametric Study of the Economical Performance
Based on Mono-Ethanolamine. Interational Journal of Greenhouse Gas Control 2007, 1,
(1), 135-142.
35.
Abu-Zahra, M. R. M.; Schneiders, L. H. J.; Niederer, J. P. M.; Feron, P. H. M.;
Versteeg, G. F., CO2 Capture from Power Plants Part I. A Parametric Study of the
Technical Performance based on Monoethanolamine. International Journal of
Greenhouse Gas Control 2007, 1, (1), 37-46.
36.
Singh, D.; Croiset, E.; Douglas, P. L.; Douglas, M. A., Techno-Economic Study
of CO2 Capture from an Existing Coal-Fired Power Plant: MEA Scrubbing vs. O2/CO2
Recycle Combustion. Energy Conversion and Management 2003, 44, (19), 3073-3091.
37.
Rao, A. B.; Rubin, E. S., A Technical, Economic, and Environmental Assessment
of Amine-based CO2 Capture Technology for Power Plant Greenhouse Gas Control.
Environmental Science & Technology 2002, 36, (20), 4467-4475.
38.
Ramezan, M.; Skone, T. J.; Nasakala, N. Y.; Liljedahl, G. N. Carbon Dioxide
Capture from Existing Coal-Fired Power Plants; DOE/NETL-401/110907; NETL, US
Department of Energy, 2007.
39.
Mattisson, T.; Lyngfelt, A.; Cho, P., The use of Iron Oxide as an Oxygen Carrier
in Chemical-Looping Combustion of Methane with Inherent Separation of CO2. Fuel
2001, 80, (13), 1953-1962.
-270-
40.
Châtel-Pélage, F.; Varagani, R.; Pranda, P.; Perrin, N.; Farzan, H.; Vecci, S. J.; Lu,
Y.; Chen, S.; Rostam-Abadi, M.; Bose, A. C., Applications of Oxygen for NOx control
and CO2 Capture in Coal-Fired Power Plants. In Second International Conference on
Clean Coal Technologies for our Future, Sardinia, Italy 2005.
41.
Chilling
News
for
Carbon
Capture.
http://www.modernpower
systems.com/story.asp?story Code=2045485 (Accessed May 5th, 2008),
42.
Rhudy, R., Chilled-Ammonia Post Combustion CO2 Capture System - Laboratory
and Economic Evaluation Results. In Electric Power Research Institute (EPRI), 2006.
43.
Electric Power Research Institute, The Challenge of Carbon Capture. EPRI
Journal May 5th, 2008.
44.
Gary, M. A. Carbon Capture Demonstration Projects: AEP's Perspective.
http://www.westgov.org/wga/meetings/coal07/GRAY.ppt#256,1,CarbonCaptureDemonst
rationProjects:AEP’s Perspective Western Governors’ Association Workshop (Accessed
May, 4th, 2008),
45.
White, C. M.; Strazisar, B. R.; Granite, E. J.; Hoffman, J. S.; Pennline, H. W.,
Separation and Capture of CO2 from Large Stationary Sources and Sequestration in
Geological Formations - Coalbeds and Deep saline Aquifers. Journal of the Air & Waste
Management Association 2003, 53, (6), 645-715.
46.
Aaron, D.; Tsouris, C., Separation of CO2 from Flue Gas: A Review. Separation
Science and Technology 2005, 40, (1-3), 321-348.
47.
Iyer, M. V.; Gupta, H.; Sakadjian, B. B.; Fan, L. S., Multicyclic Study on the
Simultaneous Carbonation and Sulfation of High-Reactivity CaO. Industrial &
Engineering Chemistry Research 2004, 43, (14), 3939-3947.
-271-
48.
Fan, L. S.; Iyer, M., Coal cleans up its Act. The Chemical Engineers 2006, pp 36-
38, 2006.
49.
Sakadjian, B. B.; Wang, W. K.; Wong, D. M.; Iyer, M. V.; Ramkumar, S.; Li, S.;
Fan, L. S.; Statnick, R. M., Sub-Pilot Demonstration of the CCR Process: A Calcium
Oxide Based CO2 Capture Process for Coal Fired Power Plants. In The 33rd International
Conference of Coal Utilization and Fuel Systems, Clear Water, Florida, USA, 2008.
50.
Lu, D. Y.; Hughes, R. W.; Anthony, E. J., In-Situ CO2 Capture using Ca-Based
Sorbent Looping in Dual Fluidized Beds. In The 9th International Conference on
Circulating Fluidized Beds, Werther, J., Ed. Hamburg, Germany, 2008.
51.
Massachusetts Institute of Technology, The Future of Coal: Options for a Carbon-
Constrained World. In Massachusetts Institute of Technology: Somerset, MA, 2007.
52.
Lu, Y.; Chen, S.; Rostam-Abadi, M.; Varagani, R.; Chatel-Pelage, F.; Bose, A. C.,
Techno-Economic Study of the Oxy-Combustion Process for CO2 Capture from CoalFired Power Plants. In Annual International Pittsburgh Coal Conference, Pittsburgh, PA,
USA, 2005.
53.
Buhre, B. J. P.; Elliott, L. K.; Sheng, C. D.; Gupta, R. P.; Wall, T. F., Oxy-Fuel
Combustion Technology for Coal-Fired Power Generation. Progress in Energy and
Combustion Science 2005, 31, (4), 283-307.
54.
Higman, C., Gasification. 2ND ED. ed.; Gulf Professional: Boston, 2008.
55.
Shinada, O.; Yamada, A.; Koyama, Y., The Development of Advanced Energy
Technologies in Japan IGCC: A Key Technology for the 21st Century. Energy
Conversion and Management 2002, 43, (9-12), 1221-1233.
56.
Kohl, A. L.; Nielsen, R., Gas Purification. 5th ed.; Gulf Pub.: Houston, TX., 1997.
-272-
57.
Dry, M. E., Practical and Theoretical Aspects of the Catalytic Fischer-Tropsch
Process. Applied Catalysis a-General 1996, 138, (2), 319-344.
58.
Dry, M. E., Review of the Fischer-Tropsch Process: 1950-2000. Abstracts of
Papers of the American Chemical Society 2000, 219, U254-U254.
59.
Ma, W. P.; Zhao, Y. L.; Li, Y. W.; Xu, Y. Y.; Zhou, J. L., An Investigation of
Chain Growth Probability in Fischer-Tropsch Synthesis over an Industrial Fe-Cu-K
Catalyst. Reaction Kinetics and Catalysis Letters 1999, 66, (2), 217-223.
60.
Lewandowski, D.; Gray, D., Potential Market Penetration of IGCC in the East
Central North American Reliability Council Region of the U.S. In Gasification
Technologies Conference, 2001.
61.
Kanniche, M.; Bouallou, C., CO2 Capture Study in Advanced Integrated
Gasification Combined Cycle. Applied Thermal Engineering 2007, 27, (16), 2693-2702.
62.
Rubin, E. S.; Chen, C.; Rao, A. B., Cost and Performance of Fossil Fuel Power
Plants with CO2 Capture and Storage. Energy Policy 2007, 35, (9), 4444-4454.
63.
Damen, K.; van Troost, M.; Faaij, A.; Turkenburg, W., A Comparison of
Electricity and Hydrogen Production Systems with CO2 Capture and Storage. Part A:
Review and Selection of Promising Conversion and Capture Technologies. Progress in
Energy and Combustion Science 2006, 32, 215-246.
64.
National Energy Technology Laboratory Carbon Sequestration: CO2 Capture.
http://www.netl.doe.gov/technologies/carbon_seq/core_rd/co2capture.html
(Accessed
May 4th, 2008),
65.
Woods, M. C.; Capicotto, P. J.; Haslbeck, J. L.; Kuehn, N. J.; Matuszewski, M.;
Pinkerton, L. L.; Rutkowski, M. D.; Schoff, R. L.; Vaysman, V. Cost and Performance
-273-
Baseline for Fossil Energy Plants; DOE/NETL-2007/1281; NETL, US Department of
Energy: 2007.
66.
Hotchkiss, R., Coal gasification technologies. Proceedings of the Institution of
Mechanical Engineers Part a-Journal of Power and Energy 2003, 217, (A1), 27-33.
67.
Dyer, P. N.; Richards, R. E.; Russek, S. L.; Taylor, D. M., Ion Transport
Membrane Technology for Oxygen Separation and Syngas Production. Solid State Ionics
2000, 134, (1-2), 21-33.
68.
Baker, R. W., Membrane Separation Systems Recent Developments and Future
Directions. Noyes Data Corp.: Park Ridge, N.J., 1991.
69.
Park, N. K.; Lee, D. H.; Jun, J. H.; Lee, J. D.; Ryu, S. O.; Lee, T. J.; Kim, J. C.;
Chang, C. H., Two-Stage Desulfurization Process for Hot Gas Ultra Cleanup in IGCC.
Fuel 2006, 85, (2), 227-234.
70.
Poulston, S.; Granite, E. J.; Pennline, H. W.; Myers, C. R.; Stanko, D. P.;
Hamilton, H.; Rowsell, L.; Smith, A. W. J.; Ilkenhans, T.; Chu, W., Metal Sorbents for
High Temperature Mercury Capture from Fuel Gas. Fuel 2007, 86, (14), 2201-2203.
71.
Smid, J.; Hsiau, S. S.; Peng, C. Y.; Lee, H. T., Hot Gas Cleanup: Pilot Testing of
Moving Bed Filters. Filtration & Separation 2006, 43, (4), 21-24.
72.
Ockwig, N. W.; Nenoff, T. M., Membranes for Hydrogen Separation. Chemical
Reviews 2007, 107, (10), 4078-4110.
73.
Zou, J.; Ho, W. S. W., CO2-Selective Polymeric Membranes Containing Amines
in Crosslinked Poly(Vinyl Alcohol). Journal of Membrane Science 2006, 286, (1-2), 310321.
-274-
74.
Ayturk, A. E.; Mardilovich, I. P.; Engwall, E. E.; Ma, Y. H., Synthesis of
Composite Pd-Porous Stainless Steel (PSS) Membranes with a Pd/Ag Intermetallic
Diffusion Barrier. Journal of Membrane Science 2006, 285, (1-2), 385-394.
75.
Rothenberger, K. S.; Cugini, A. V.; Howard, B. H.; Killmeyer, R. P.; Ciocco, M.
V.; Morreale, B. D.; Enick, R. M.; Bustamante, F.; Mardilovich, I. P.; Ma, Y. H., High
Pressure Hydrogen Permeance of Porous Stainless Steel Coated with a Thin Palladium
Film via Electroless Plating. Journal of Membrane Science 2004, 244, (1-2), 55-68.
76.
Iyoha, O.; Enick, R.; Killmeyer, R.; Howard, B.; Ciocco, M.; Morreale, B., H2
production from simulated coal syngas containing H2S in multi-tubular Pd and 80 wt%
Pd-20 wt% Cu membrane reactors at 1173 K. Journal of Membrane Science 2007, 306,
(1-2), 103-115.
77.
Hara, S.; Sakaki, K.; Itoh, N.; Kimura, H. M.; Asami, K.; Inoue, A., An
amorphous alloy membrane without noble metals for gaseous hydrogen separation.
Journal of Membrane Science 2000, 164, (1-2), 289-294.
78.
Sholl, D. S.; Ma, Y. H., Dense metal membranes for the production of high-purity
hydrogen. Mrs Bulletin 2006, 31, (10), 770-773.
79.
Jordal, K.; Bredesen, R.; Kvamsdal, H. M.; Bolland, O., Integration of H-2-
separating membrane technology in gas turbine processes for CO2 capture. Energy 2004,
29, (9-10), 1269-1278.
80.
McCool, B. A.; Lin, Y. S., Nanostructured thin palladium-silver membranes:
Effects of grain size on gas permeation properties. Journal of Materials Science 2001, 36,
(13), 3221-3227.
-275-
81.
Kamakoti, P.; Morreale, B. D.; Ciocco, M. V.; Howard, B. H.; Killmeyer, R. P.;
Cugini, A. V.; Sholl, D. S., Prediction of hydrogen flux through sulfur-tolerant binary
alloy membranes. Science 2005, 307, (5709), 569-573.
82.
Yukawa, H.; Teshima, A.; Yamashita, D.; Ito, S.; Yamaguchi, S.; Morinaga, M.,
Alloying effects on the hydriding properties of vanadium at low hydrogen pressures.
Journal of Alloys and Compounds 2002, 337, 264-268.
83.
Hara, S.; Hatakeyama, N.; Itoh, N.; Kimura, H. M.; Inoue, A., Hydrogen
permeation through amorphous-Zr36-xHfxNi64-alloy membranes. Journal of Membrane
Science 2003, 211, (1), 149-156.
84.
Verweij, H., Ceramic membranes: Morphology and transport. Journal of
Materials Science 2003, 38, (23), 4677-4695.
85.
Verweij, H.; Lin, Y. S.; Dong, J. H., Microporous silica and zeolite membranes
for hydrogen purification. Mrs Bulletin 2006, 31, (10), 756-764.
86.
Mitchell, M. C.; Gallo, M.; Nenoff, T. M., Computer simulations of adsorption
and diffusion for binary mixtures of methane and hydrogen in titanosilicates. Journal of
Chemical Physics 2004, 121, (4), 1910-1916.
87.
Bernal, M. P.; Coronas, J.; Menendez, M.; Santamaria, J., Separation of CO2/N-2
mixtures using MFI-type zeolite membranes. Aiche Journal 2004, 50, (1), 127-135.
88.
Li, S. G.; Martinek, J. G.; Falconer, J. L.; Noble, R. D.; Gardner, T. Q., High-
pressure CO2/CH4 separation using SAPO-34 membranes. Industrial & Engineering
Chemistry Research 2005, 44, (9), 3220-3228.
89.
Sircar, S.; Rao, M.; Thaeron, C., Selective Surface Flow Membrane for Gas
Separation. Separation Science & Technology 1999, 34, (10), 2081.
-276-
90.
Sircar, S.; Waldron, W. E.; Rao, M. B.; Anand, M., Hydrogen production by
hybrid SMR-PSA-SSF membrane system. Separation and Purification Technology 1999,
17, (1), 11-20.
91.
Kapteijn, F.; Bakker, W. J. W.; Vandegraaf, J.; Zheng, G.; Poppe, J.; Moulijn, J.
A., Permeation and Separation Behavior of a Silicalite-1 Membrane. Catalysis Today
1995, 25, (3-4), 213-218.
92.
Lin, H. Q.; Freeman, B. D., Gas and vapor solubility in cross-linked poly(ethylene
glycol diacrylate). Macromolecules 2005, 38, (20), 8394-8407.
93.
Lin, H. Q.; Kai, T.; Freeman, B. D.; Kalakkunnath, S.; Kalika, D. S., The effect of
cross-linking on gas permeability in cross-linked poly(ethylene glycol diacrylate).
Macromolecules 2005, 38, (20), 8381-8393.
94.
Yamaguchi, T.; Koval, C. A.; Noble, R. D.; Bowman, C. N., Transport
mechanism of carbon dioxide through perfluorosulfonate ionomer membranes containing
an amine carrier. Chemical Engineering Science 1996, 51, (21), 4781-4789.
95.
Matsuyama, H.; Teramoto, M.; Sakakura, H.; Iwai, K., Facilitated transport of
CO2 through various ion exchange membranes prepared by plasma graft polymerization.
Journal of Membrane Science 1996, 117, (1-2), 251-260.
96.
Kaldis, S. P.; Skodras, G.; Sakellaropoulos, G. P., Energy and capital cost
analysis of CO2 capture in coal IGCC processes via gas separation membranes. Fuel
Processing Technology 2004, 85, (5), 337-346.
97.
Amelio, M.; Morrone, P.; Galucci, F.; Basile, A., Integrated gasification gas
combined cycle plant with membrane reactors: Technological and economical analysis.
Energy Conversion and Management 2007, 48, (10), 2680-2693.
-277-
98.
Grainger, D.; Hägg, M.-B., Techno-economic evaluation of a PVAm CO2-
selective membrane in an IGCC power plant with CO2 capture. Fuel 2008, 87, (1), 14-24.
99.
Ciocco, M. V.; Iyoha, O.; Enick, R. M.; Killmeyer, R. P., High-Temperature
Water-Gas Shift Membrane Reactor Study. 2007.
100.
Stiegel, G. J.; Ramezan, M., Hydrogen from coal gasification: An economical
pathway to a sustainable energy future. International Journal of Coal Geology 2006, 65,
(3-4), 173-190.
101.
Carbo, M. C.; Jansen, D.; Haije, W. G.; Verkooijen, A. H. M., Advanced
Membrane Reactors for Fuel Decarbonisation in IGCC: H2 or CO2 Separation? In Fifth
Annual Conference on Carbon Capture and Sequestration, Alexandria VA, U.S.A., 2006.
102.
Chiesa, P.; Kreutz, T. G.; Lozza, G. G., CO2 sequestration from IGCC power
plants by means of metallic membranes. Journal of Engineering for Gas Turbines and
Power-Transactions of the Asme 2007, 129, (1), 123-134.
103.
Huang, J.; Ho, W. S. W., Effects of system parameters on the performance of
CO2-selective WGS membrane reactor for fuel cells. Journal of the Chinese Institute of
Chemical Engineers 2008, 39, (2), 129-136.
104.
Shain, D.; Estela, O.; Mike, A.; Francis, L.; Mike, R. A Novel Membrane Reactor
for Direct Hydrogen Production From Coal; United States, 2006.
105.
Lane, H. Process for the Production of Hydrogen. US 1078686, 1913.
106.
Hurst, S., Production of Hydrogen by the Steam-Iron Method. Journal of the
American Oil Chemists' Society 1939, 16, (2), 29-36.
107.
Lewis, W. K.; Gilliland, E. R. Production of Pure Carbon Dioxide. US 26665972,
1954.
-278-
108.
Lewis, W. K.; Gilliland, E. R. Production of Pure Carbon Dioxide. US 2665973,
1954.
109.
Anheden, M.; Svedberg, G., Exergy analysis of chemical-looping combustion
systems. Energy Conversion and Management 1998, 39, (16-18), 1967-1980.
110.
Dewulf, J.; Van Langenhove, H.; Muys, B.; Bruers, S.; Bakshi, B. R.; Grubb, G.
F.; Paulus, D. M.; Sciubba, E., Exergy: Its potential and limitations in environmental
science and technology. Environmental Science & Technology 2008, 42, (7), 2221-2232.
111.
Ishida, M.; Zheng, D.; Akehata, T., Evaluation of a Chemical-Looping-
Combustion Power-Generation System by Graphic Exergy Analysis. Energy 1987, 12,
(2), 147-154.
112.
Jin, H. G.; Ishida, M., A new type of coal gas fueled chemical-looping
combustion. Fuel 2004, 83, (17-18), 2411-2417.
113.
Johansson, M.; Mattisson, T.; Lyngfelt, A., Use of NiO/NiAl2O4 particles in a 10
kW chemical-looping combustor. Industrial & Engineering Chemistry Research 2006, 45,
(17), 5911-5919.
114.
Gupta, P.; Velazquez-Vargas, L. G.; Fan, L. S., Syngas redox (SGR) process to
produce hydrogen from coal derived syngas. Energy & Fuels 2007, 21, (5), 2900-2908.
115.
Fan, L. S.; Li, F. X., Clean coal. Physics World 2007, 20, (7), 37-41.
116.
Leion, H.; Mattisson, T.; Lyngfelt, A., The use of petroleum coke as fuel in
chemical-looping combustion. Fuel 2007, 86, (12-13), 1947-1958.
117.
Xiang, W.; Wang, S.; Di, T. T., Investigation of gasification chemical looping
combustion combined cycle performance. Energy & Fuels 2008, 22, (2), 961-966.
-279-
118.
Zhao, H.; Liu, L.; Wang, B.; Xu, D.; Jiang, L.; Zheng, C., Sol–Gel-Derived
NiO/NiAl2O4 Oxygen Carriers for Chemical-Looping Combustion by Coal Char. Energy
& Fuels 2008, 22, (2), 898-905.
119.
de Diego, L. F.; Garcia-Labiano, F.; Adanez, J.; Gayan, P.; Abad, A.; Corbella, B.
M.; Palacios, J. M., Development of Cu-based oxygen carriers for chemical-looping
combustion. Fuel 2004, 83, (13), 1749-1757.
120.
Gabbrielli, R.; Singh, R., Thermodynamic performance analysis of new gas
turbine combined cycles with no emissions of carbon dioxide. Journal of Engineering for
Gas Turbines and Power-Transactions of the Asme 2003, 125, (4), 940-946.
121.
Fan, L.-S.; Gupta, P.; Velazquez-Vargas, L. G.; Li, F. Systems and methods of
converting fuels. PCT 2007/082089, 2007.
122.
Thomas, T.; Fan, L.-S.; Gupta, P.; Velazquez-Vargas, L. G. Combustion looping
using composite oxugen carriers. US 2005/0175533, 2004.
123.
Velazquez-Vargas, L. G.; Thomas, T.; Gupta, P.; Fan, L.-S. In Hydrogen
production via redox reaction of syngas with metal oxide composite particles,
Proceedings of AIChE Annual Technical Meeting, 2004.
124.
Wang, B.; Yan, R.; Lee, D. H.; Liang, D. T.; Zheng, Y.; Zhao, H.; Zheng, C.,
Thermodynamic investigation of carbon deposition and sulfur evolution in chemical
looping combustion with syngas. Energy & Fuels 2008, 22, (2), 1012-1020.
125.
Abad, A.; Adanez, J.; Garcia-Labiano, F.; de Diego, L. F.; Gayan, P.; Celaya, J.,
Mapping of the range of operational conditions for Cu-, Fe-, and Ni-based oxygen
carriers in chemical-looping combustion. Chemical Engineering Science 2007, 62, (1-2),
533-549.
-280-
126.
Siriwardane, R.; Poston, J.; Chaudhari, K.; Zinn, A.; Simonyi, T.; Robinson, C.,
Chemical-looping combustion of simulated synthesis gas using nickel oxide oxygen
carrier supported on bentonite. Energy & Fuels 2007, 21, (3), 1582-1591.
127.
Ryden, M.; Lyngfelt, A.; Mattisson, T.; Chen, D.; Holmen, A.; Bjorgum, E.,
Novel oxygen-carrier materials for chemical-looping combustion and chemical-looping
reforming; LaxSr1-xFeyCo1-yO3-z perovskites and mixed-metal oxides of NiO, Fe2O3 and
Mn3O4. International Journal of Greenhouse Gas Control 2008, (2), 21-36.
128.
Garcia-Labiano, F.; Adanez, J.; de Diego, L. F.; Gayan, P.; Abad, A., Effect of
pressure on the behavior of copper-, iron-, and nickel-based oxygen carriers for chemicallooping combustion. Energy & Fuels 2006, 20, (1), 26-33.
129.
Johansson, E.; Mattisson, T.; Lyngfelt, A.; Thunman, H., A 300 W laboratory
reactor system for chemical-looping combustion with particle circulation. Fuel 2006, 85,
(10-11), 1428-1438.
130.
Johansson, E.; Mattisson, T.; Lyngfelt, A.; Thunman, H., Combustion of syngas
and natural gas in a 300 W chemical-looping combustor. Chemical Engineering Research
& Design 2006, 84, (A9), 819-827.
131.
Mattisson, T.; Garc; iacute; a-Labiano, F.; Kronberger, B.; Lyngfelt, A.; Ad;
aacute; nez, J.; Hofbauer, H., Chemical-looping combustion using syngas as fuel.
International Journal of Greenhouse Gas Control 2007, 1, (2), 158-169.
132.
Abad, A.; Mattisson, T.; Lyngfelt, A.; Johansson, M., The use of iron oxide as
oxygen carrier in a chemical-looping reactor. Fuel 2007, 86, (7-8), 1021-1035.
-281-
133.
Abad, A.; Mattisson, T.; Lyngfelt, A.; Ryden, M., Chemical-looping combustion
in a 300 W continuously operating reactor system using a manganese-based oxygen
carrier. Fuel 2006, 85, (9), 1174-1185.
134.
Johansson, M.; Mattisson, T.; Lyngfelt, A., Investigation of Mn3O4 with
stabilized ZrO2 for chemical-looping combustion. Chemical Engineering Research &
Design 2006, 84, (A9), 807-818.
135.
Ryu, H. J.; Jin, G. T.; Bae, D. H.; Yi, C. K., Continuous Operation of a 50 kWth
Chemical-Looping Combustor: Long-Term Operation with Ni- and Co- Based Oxygen
Carrier Particles. http://lib.kier.re.kr/balpyo/clean5/13.pdf (Accessed May 6th, 2008).
136.
Kolbitsch, P.; Bolhàr-Nordenkampf, J.; Pröll, T.; Hofbauer, H., Design of a
Chemical Looping Combustor using a Dual Circulating Fluidized Bed (DCFB) Reactor
System. In The 9th International Conference on Circulating Fluidized Beds, Hamburg,
Germany, 2008.
137.
Ishida, M.; Jin, H. G., A New Advanced Power-Generation System Using
Chemical-Looping Combustion. Energy 1994, 19, (4), 415-422.
138.
Gupta, P.; Velazquez-Vargas, L. G.; Valentine, C.; Fan, L. S., Moving bed reactor
setup to study complex gas-solid reactions. Review of Scientific Instruments 2007, 78, (8)
139.
Fan, L. S.; Li, F.; Velazquez-Vargas, L. G.; Ramkumar, S., Chemical Looping
Gasification. In The 9th International Conference on Circulating Fluidized Beds,
Hamburg, Germany, 2008.
140.
Fan, L. S.; Li, F.; Ramkumar, S., Utilization of Chemical Looping Strategy in
Coal Gasification Processes. Particuology 2008, (in Press).
-282-
141.
Tomlinson, G.; Gray, D., Chemical Looping Process in a Coal-to-Liquids
Configuration. USDOE/NETL-2008/1307, 2007.
142.
Ryu, H. J.; An, G. T., Chemical-looping hydrogen generation system:
Performance estimation and process selection. Korean Journal of Chemical Engineering
2007, 24, (3), 527-531.
143.
Jin, G. T.; Ryu, H. J.; Jo, S. H.; Lee, S. Y.; Son, S. R.; Kim, S. D., Hydrogen
production in fluidized bed by chemical-looping cycle. Korean Journal of Chemical
Engineering 2007, 24, (3), 542-546.
144.
Svoboda, K.; Siewiorek, A.; Baxter, D.; Rogut, J.; Pohorely, M., Thermodynamic
possibilities and constraints for pure hydrogen production by a nickel and cobalt-based
chemical looping process at lower temperatures. Energy Conversion and Management
2008, 49, (2), 221-231.
145.
Svoboda, K.; Slowinski, G.; Rogut, J.; Baxter, D., Thermodynamic possibilities
and constraints for pure hydrogen production by iron based chemical looping process at
lower temperatures. Energy Conversion and Management 2007, 48, (12), 3063-3073.
146.
Xiang, W. U.; Chen, Y. Y., Hydrogen and electricity from coal with carbon
dioxide separation using chemical looping reactors. Energy & Fuels 2007, 21, (4), 22722277.
147.
Fan, L.-S.; Gupta, H.; Iyer, M. V. Separation of carbon dioxide (CO2) from gas
mixtures by calcium based reaction separation (CaRS-CO2) process. WO 2007002792.
US 2007.
148.
Iyer, M. V. High temperature reactive separation process for combined carbon
dioxide and sulfur dioxide capture from flue gas and enhanced hydrogen production with
-283-
insitu carbon dioxide capture using high reactivity calcium and biomineral sorbents.
Doctoral Dissertation, The Ohio State University, 2006.
149.
Lee, K. B.; Beaver, M. G.; Caram, H. S.; Sircar, S., Reversible chemisorption of
carbon dioxide: simultaneous production of fuel-cell grade H-2 and compressed CO2
from synthesis gas. Adsorption-Journal of the International Adsorption Society 2007, 13,
(3-4), 385-397.
150.
Lee, K. B.; Beaver, M. G.; Caram, H. S.; Sircar, S., Performance of Na2O
promoted alumina as CO2 chemisorbent in sorption-enhanced reaction process for
simultaneous production of fuel-cell grade H-2 and compressed CO2 from synthesis gas.
Journal of Power Sources 2008, 176, (1), 312-319.
151.
Yang, J.; Cai, N.; Li, Z., Reduction of Iron Oxide as an Oxygen Carrier by Coal
Pyrolysis and Steam Char Gasification Intermediate Products. Energy & Fuels 2007, 21,
(6), 3360-3368.
152.
Scott, S. A.; Dennis, J. S.; Hayhurst, A. N.; Brown, T., In situ gasification of a
solid fuel and CO2 separation using chemical looping. Aiche Journal 2006, 52, (9), 33253328.
153.
Cao, Y.; Casenas, B.; Pan, W. P., Investigation of chemical looping combustion
by solid fuels. 2. Redox reaction kinetics and product characterization with coal, biomass,
and solid waste as solid fuels and CuO as an oxygen carrier. Energy & Fuels 2006, 20,
(5), 1845-1854.
154.
Cao, Y.; Pan, W. P., Investigation of chemical looping combustion by solid fuels.
1. Process analysis. Energy & Fuels 2006, 20, (5), 1836-1844.
-284-
155.
Corbella, B. M.; Palacios, J. M., Titania-supported iron oxide as oxygen carrier
for chemical-looping combustion of methane. Fuel 2007, 86, (1-2), 113-122.
156.
Gupta, P.; Velazquez-Vargas, L. G.; Thomas, T.; Fan, L.-S. In Hydrogen
production from combustion looping (solids-coal), Proceedings of Clear Water Coal
Conference, 2004.
157.
Gupta, P.; Velazquez-Vargas, L. G.; Li, F.; Fan, L.-S. In Chemical looping
reforming process for the production of hydrogen from coa., Proceedings of 23rd Annual
International Pittsburg Coal Conference, 2006.
158.
Lin, S. Y.; Harada, M.; Suzuki, Y.; Hatano, H., Process analysis for hydrogen
production by reaction integrated novel gasification (HyPr-RING). Energy Conversion
and Management 2005, 46, (6), 869-880.
159.
Kuramoto, K.; Shibano, S.; Fujimoto, S.; Kimura, T.; Suzuki, Y.; Hatano, H.; Lin,
S. Y.; Harada, M.; Morishita, K.; Takarada, T., Deactivation of Ca-based sorbents by
coal-derived minerals during multicycle CO2 sorption under elevated pressure and
temperature. Industrial & Engineering Chemistry Research 2003, 42, (15), 3566-3570.
160.
Lin, S. Y.; Harada, M.; Suzuki, Y.; Hatano, H., Continuous experiment regarding
hydrogen production by Coal/CaO reaction with steam (II) solid formation. Fuel 2006, 85,
(7-8), 1143-1150.
161.
Lin, S. Y.; Harada, M.; Suzuki, Y.; Hatano, H., Continuous experiment regarding
hydrogen production by coal/CaO reaction with steam (I) gas products. Fuel 2004, 83, (78), 869-874.
-285-
162.
Rizeq, R. G.; Lyon, R. K.; Zamansky, V. M.; Das, K. In Fuel-flexible AGC
technology for production of H2 power, and sequestration-ready CO2, Proceedings of
the International Technical Conference on Coal Utilization & Fuel Systems, 2001.
163.
Rizeq, R. G.; West, J.; Frydman, A.; Subia, R.; Kumar, R.; Zamansky, V. Fuel-
flexible gasification-combustion technology for production of H2 and sequestration-ready
CO2; 2002.
164.
Andrus, H. A. E., Jr.; Chiu, J. H.; Liljedahl, G. N.; Stromberg, P. T.; Thibeault, P.
R.; Jain, S. C. In ALSTOM's hybrid combustion-gasification chemical looping technology
development, 22nd Annual International Pittsburgh Coal Conference, Pittsburgh, PA,
USA, Sep 2005, 2005; Pittsburgh, PA, USA, 2005.
165.
Andrus, H. E.; Burns, G.; Chiu, J. H.; Liljedahl, G. N.; Stromberg, P. T.;
Thibeault, P. R. Hybrid Combustion-Gasification Chemical Looping Power Technology
Development; 2006.
166.
Bohm, M. C. Capture-Ready Power Plants - Options, Technologies, and
Economics. Massachusetts Institute of Technology, Boston, 2006.
167.
Hutchinson, H., Old king coal. Mechanical Engineering 2002, 124, (8), 41-45.
168.
Yoshino, Y.; Suzuki, T.; Nair, B. N.; Taguchi, H.; Itoh, N., Development of
tubular substrates, silica based membranes and membrane modules for hydrogen
separation at high temperature. Journal of Membrane Science 2005, 267, (1-2), 8-17.
169.
Bredesen, R.; Jordal, K.; Bolland, O., High-Temperature Membranes in Power
Generation with CO2 Capture. Chemical Engineering and Processing 2004, 43, 11291158.
-286-
170.
Elliott, J. F.; Smailer, R. M.; Stephenson, R. L.; Iron and Steel Society of AIME.,
Direct reduced iron : technology and economics of production and use. Iron & Steel
Society of AIME: Warrendale, Pa., 1980.
171.
Velazquez-Vargas, L. G.; Thomas, T.; Gupta, P.; Fan, L.-S. In Hydrogen
production via redox reaction of syngas with metal oxide composite particles, AIChE
Annual Technical Meeting, 2004.
172.
Chiesa, P.; Lozza, G.; Malandrino, A.; Romano, M.; Piccolo, V., Three-reactors
chemical looping process for hydrogen production. International Journal of Hydrogen
Energy 2008, 33, (9), 2233-2245.
173.
Sanfilippo, D.; Mizia, F.; Malandrino, A.; Rossini, S. Process for the Production
of Hydrogen and the Co-Production of Carbon Dioxide. US 7404942, 2008.
174.
Svoboda, K.; Siewiorek, A.; Baxter, D.; Rogut, J.; Puncochar, M.,
Thermodynamic possibilities and constraints of pure hydrogen production by a chromium,
nickel, and manganese-based chemical looping process at lower temperatures. Chemical
Papers 2007, 61, (2), 110-120.
175.
Adanez, J.; de Diego, L. F.; Garcia-Labiano, F.; Gayan, P.; Abad, A.; Palacios, J.
M., Selection of Oxygen Carriers for Chemical-Looping Combustion. Energy & Fuels
2004, 18, (2), 371-377.
176.
Garcia-Labiano, F.; de Diego, L. F.; Adanez, J.; Abad, A.; Gayan, P.,
Temperature variations in the oxygen carrier particles during their reduction and
oxidation in a chemical-looping combustion system. Chemical Engineering Science 2005,
60, (3), 851-862.
-287-
177.
Perry, R. H.; Green, D. W., Perry's chemical engineers' handbook. 8th Edition ed.;
McGraw-Hill: New York, 2008.
178.
Alcock, C. B., Principles of pyrometallurgy. Academic Press Inc.: London ; New
York, 1976.
179.
Gaskell, D. R., Introduction to metallurgical thermodynamics. 2nd Edition ed.;
Hemisphere Publications: New York, 1981.
180.
Elliott, J. F.; Smailer, R. M.; Stephenson, R. L., Direct reduced iron : technology
and economics of production and use. Iron & Steel Society of AIME: Warrendale, Pa.,
1980.
181.
AspenTech, Aspen Physical Property System: Physical Property Methods and
Models. 2006.
182.
Barin, I., Thermochemical data of pure substances. VCH: Weinheim, Federal
Republic of Germany; New York, NY, 1989.
183.
Chase, M. W., NIST-JANAF thermochemical tables. 4th Edition ed.; American
Chemical Society: Washington, DC, 1998.
184.
Knacke, O.; Kubaschewski, O.; Hesselmann, K., Thermochemical properties of
inorganic substances. 2nd Edition ed.; Springer-Verlag: New York, 1991.
185.
Mondal, K.; Lorethova, H.; Hippo, E.; Wiltowski, T.; Lalvani, S. B., Reduction of
iron oxide in carbon monoxide atmosphere - reaction controlled kinetics. Fuel Processing
Technology 2004, 86, (1), 33-47.
186.
Ito, T.; Tsutsumi, A.; Akiyama, T., Coproduction of iron and hydrogen from iron
carbide. Journal of Chemical Engineering of Japan 2003, 36, (7), 881-886.
-288-
187.
Wu, S.; Uddin, M. A.; Sasaoka, E., Effect of Pore Size Distribution of Calcium
Oxide High-Temperature Desulfurization Sorbent on Its Sulfurization and Consecutive
Oxidative Decomposition. Energy & Fuels 2005, 19, (3), 864-868.
188.
Cachu, S.; Gunter, W. D., Acid Gas Injection in the Alberta Basin, Canada: a CO2
Storage Experience. Geological Society Special Publication 2004, 233, 225 - 234.
189.
Thomas, T.; Fan, L.-S.; Gupta, P.; Velazquez-Vargas, L. G. Combustion looping
using composite oxugen carriers. US11,010,648, 2004.
190.
Gupta, P.; Velazquez-Vargas, L. G.; Valentine, C.; Fan, L. S., Moving bed reactor
setup to study complex gas-solid reactions. Review of Scientific Instruments 2007, 78, (8),
0851061 - 7.
191.
Simell, P. A.; Leppalahti, J. K.; Bredenberg, J. B. S., Catalytic Purification of
Tarry Fuel Gas with Carbonate Rocks and Ferrous Materials. Fuel 1992, 71, (2), 211-218.
192.
Gueret, C.; Daroux, M.; Billaud, F., Methane Pyrolysis: Thermodynamics.
Chemical Engineering Science 1997, 52, (5), 815-827.
193.
Shah, V. A.; Huurdeman, T. L., Synthesis gas treating with physical solvent
process using Selexol process technology. Ammonia Plant Safety & Related Facilities
1990, 30, 216-24.
194.
Shelton, W.; Lyons, J., Shell Gasifier IGCC Base Cases. U. S. Department of
Energy, 2000.
195.
AspenTech, Getting Started Modeling Processes with Solids. 2006.
196.
Bachu, S.; Gunter, W. D., Acid Gas Injection in the Alberta Basin, Canada: a CO2
Storage Experience. Geological Society Special Publication 2004, (223), 225-234.
-289-
197.
Velazquez-Vargas, L. G. Development of Chemical Looping Gasification
Processes for the Production of Hydrogen from Coal the Ohio State University, Doctoral
Dissertation, The Ohio State University: Columbus, 2007.
198.
Fan, L.-S.; Zhu, C., Principles of Gas-Solids Flows. Cambridge University Press:
New York, 1998.
199.
Werther, J.; Reppenhagen, J., Attrition. In Handbook of Fluidization and Fluid-
Particle Systems, Yang, W.-C.; Dekker, M. Eds, 2003.
200.
Bliniche.V.N; Streltso.V.V; Lebedeva, E. S., An Investigation of Size Reduction
of Granular Materials during Their Processing in Fluidized Beds. International Chemical
Engineering 1968, 8, (4), 615-618.
201.
Pell, M., Gas fluidization. In Handbook of Powder Technology, Elsevier:
Amsterdam, 1990.
-290-