UNIVERSITY OF CINCINNATI

July 28 01 _____________ , 20 _____

Antoinette Weil Stein I,______________________________________________, hereby submit this as part of the requirements for the degree of:

Doctorate of Philosophy (Ph.D) ________________________________________________

in:
Environmental Engineering ________________________________________________

It is entitled:
Investigation of the Chemical Pathway for Gaseous Nitrogen ________________________________________________
Dioxide Formation during Flue Gas Desulfurization with Dry ________________________________________________

Sodium Bicarbonate Injection ________________________________________________

________________________________________________

Approved by: Dr. Tim Keener ________________________ Dr. Soon-Jai Khang ________________________ Dr. Riley N. Kinman ________________________ Dr.Paul Bishop ________________________ ________________________

INVESTIGATION OF THE CHEMICAL PATHWAY OF GASEOUS NITROGEN DIOXIDE FORMATION DURING FLUE GAS DESULFURIZATION WITH DRY SODIUM BICARBONATE INJECTION A dissertation submitted to the Division of Research and Advanced Studies Of the University of Cincinnati In partial fulfillment of the Requirements for the degree of DOCTOR OF PHILOSOPHY (PH.D) In the Department of Civil and Environmental Engineering Of the College of Engineering 2001 by Antoinette Weil Stein B.S., University of Wisconsin, Madison, 1983 M.S.E, Milwaukee School of Engineering, MSOE, 1988

Committee Chair: Dr. Timothy C. Keener

ABSTRACT The chemical reaction pathway for the viable flue gas desulfurization process, dry sodium bicarbonate injection, was investigated to mitigate undesirable plume discoloration. Based on a foundation of past findings, a simplified three-step reaction pathway was hypothesized for the formation of the plume-discoloring constituent, NO2. As the first step, it was hypothesized that sodium sulfite formed by sodium bicarbonate reaction with flue gas SO2. As the second step, it was hypothesized that sodium nitrate formed by sodium sulfite reaction with flue gas NO. And as the third step, it was hypothesized that NO2 and sodium sulfate formed by sodium nitrate reaction with SO2. The second and third hypothesized steps were experimentally investigated using an isothermal fixed bed reactor. As reported in the past, technical grade sodium sulfite was found to be un-reactive with NO and O2. Freshly prepared sodium sulfite, maintained unexposed to moist air, was shown to react with NO and O2 resulting in a mixture of sodium nitrite and sodium nitrate together with a significant temperature rise. This reaction was found to proceed only when oxygen was present in the flue gas. As reported in the past, technical grade sodium nitrate was shown to be un-reactive with SO 2. But freshly formed sodium nitrate kept unexposed to humidity was found to be reactive with SO2 and O2 resulting in the formation of NO2 and sodium sulfate polymorphic Form I. The NO 2 formation by this reaction was shown to be temperature dependant with maximum formation at 175C.

Plume mitigation methods were studied based on the validated three-step reaction pathway. Mitigation of NO2 was exhibited by limiting oxygen concentration in the flue gas to a level below 5%. It was also shown that significant NO2 mitigation was achieved by operating below 110C or above 250C. An innovative NO2 mitigation method was patented as a result of the findings of this study. The patented process incorporated a process step of sodium sulfite injection to remove flue gas NO prior to sodium bicarbonate injection.

Copyright 2001 Antoinette Weil Stein

ACKNOWLEDGEMENTS I wish to thank my advisor Dr. Keener for his support to accomplish this research. It was a special honor and opportunity for me to have worked under Dr. Keener on this project particularly since Dr. Keener was an early developer of sodium bicarbonate sorbent systems. I would also like to thank Dr. Keener for obtaining sufficient research funding from the Electric Power Research Institute and from the Department of Education (DOE) Fellowship that he encouraged. I wish to thank my thesis committee members, Professor Soon-Jai Khang, Professor Paul Bishop, Professor Riley Kinman and Professor Boerio for their participation and support. I wish to thank Barbara Toole ONeil from the Electric Power Research Institute for her project leadership in funding this project. I also would like to thank her for her mentorship and technical writing guidance. The reference books that she suggested were very helpful. I would like to thank a number of key people who technically contributed to this work. From the University of Cincinnati, I would like to thank a number of others in our interdepartmental energy and pollution control research group including Guang Li for his patience in teaching me the use of many laboratory pieces of equipment including the BET and the total sulfur analyzer and his willingness to explain and share ideas on achieving product conversion from elemental analysis data. Also I thank him for maintaining a professional and positive attitude and being very supportive on a day to day basis in the laboratory. Additionally I would like to thank Kasemsan Manomaiphiboon for his TGA and wet chemistry analysis of the sodium pyrosulfite decomposition samples. I would like to thank Dr. Wayne Bresser from the Electrical and Computer

Engineering and Computer Science Department for the extensive X-Ray diffraction analysis that his lab provided. I would like to thank Elvin McCorvey for his electrical, mechanical and computer expertise in setting up and maintaining suitable operation of the extensive network of equipment used for laboratory experimentation. I would like to thank Professor Boerio from the Material Science and Engineering Laboratory and his laboratory students for their XPS, and IR material analysis used in reaction product identification. I would like to thank Ernie Clark from the SEM laboratory in the Material Science and Engineering Department for his expert skill in providing SEM analysis and photography included in this dissertation. I wish to thank Jim Layden, the librarian in the geology and physics library for his help in obtaining excellent literature on the structural transformations of sodium sulfate with sodium carbonate. I wish to thank the librarian in the Chemistry Department library, and in the Geophysics library for their technical expertise in running successful data-base literature searches that found key background information used in this study. Additionally I wish to thank a number of people from outside the University of Cincinnati for their technical contributions. I would like to thank Malcolm W. Chase from the National Institutes of Science and Technology, editor of the Third edition of the JANF Thermochemical Tables, for his assistance in obtaining hard to find thermo-data. I would like to thank Pat McKeown and John Moulder of Physical Electronics Corporation for their generous donated hours of technical services in running SIMS and ESCA material analysis for product identification that was key evidence in the findings for this research. I wish to thank the technical support from the Porter instrument team, in particular, John Nealy, Art Schell, and Mike Marcio, for their generous donated time, concern, and quick turn around time in resolving maintenance

issues that arose with their manufactured mass flow controllers operated in the extremely corrosive conditions. I wish to thank Ted Jennings from Witco Surfactants group for their input on sodium nitrate additives used in manufacturing. And I wish to thank M.W. Hott Company, including Duane Lorenz for the help constructing the test reactor with their specialized components. Finally I would like to thank my family for their support. In particular I would like to thank my sister and husband for their encouragement in undertaking these studies; my husband for his patience and interest in this subject to partake in engaging discussions related to my research findings. It was through these discussions that many of my research findings were realized. Also, I would like to thank Morris Brown, Mary Kenney and Joe Pasqua and my husband for the technical support in the converting this document from a Word file to a PDF format. A special thank you to my son Adam and my daughter Mia, for their daily inspiration and energy. It was the contributions of all of the above persons that made it possible for me to persevere and complete these studies.

TABLE OF CONTENTS

CHAPTER 1...................................................................................................................... 22 INTRODUCTION............................................................................................................ 22 1.1 1.2 1.3 1.4 1.5 Background ........................................................................................................... 22 The Problem .......................................................................................................... 23 Prior study of this problem.................................................................................... 25 Goals and objectives of this study......................................................................... 26 Approach and scope .............................................................................................. 26

CHAPTER 2...................................................................................................................... 31 LITERATURE REVIEW................................................................................................. 31 2.1 2.1.1 2.1.2 Literature review of plume discoloration .............................................................. 31 History of plume discoloration...................................................................... 31 Reported observations of plume discoloration during sodium sorbent flue gas desulfurization trials. ....................................................................... 35 2.1.3 The discovery that NO2 was the plume discoloring agent for sodium bicarbonate bag-house injection.................................................................... 37 2.2 2.2.1 2.2.2 Literature review of sodium bicarbonate reaction chemistry................................ 38 Literature review of sodium bicarbonate decomposition.............................. 39 Literature review of sodium bicarbonate reaction with SO2 flue gas without NOx................................................................................................... 41 2.2.2.1 Review of Erdos investigations of sodium bicarbonate reaction with SO2 flue gas without NOx.............................................................................. 41 1

2.2.2.2

Review of Howatson et al. (1980)[8] investigations of sodium bicarbonate reaction with SO2 flue gas without NOx. ................................... 43

2.2.2.3

Review of Keener (1982)[9] investigations of sodium bicarbonate reaction with flue gas with SO2 but without NOx. ........................................ 46

2.2.3

Literature review of sodium bicarbonate and sodium carbonate reaction with NOx without SO2. .................................................................... 47

2.2.4

Literature review of sodium bicarbonate reaction with flue gas with SO2 and NOx.................................................................................................. 49

2.2.4.1

Review of Blands (1990) [16] investigations of the reaction of sodium bicarbonate with SO2 flue gas with NO......................................................... 50

2.2.4.2

Review of Verlaeten et al.s (1991)[20] investigations of the reaction of sodium bicarbonate with flue gas with SO2 and NO. ............................... 54

2.2.4.3

Review of Bortzs (1994)[45] investigations of the reaction of sodium bicarbonate with flue gas with SO2 and NO. ................................................ 55

2.2.5 2.2.5.1 2.2.5.2 2.2.5.3 2.2.5.4 2.2.6

Past Hypotheses for NO2 formation and NOx removal reactions.................. 57 Review of Millers (1986)[19] hypothesis for NO2 formation........................ 57 Review of Blands (1990)[16] hypothesis for NO2 formation. ....................... 60 Review of Verlaeten et al.s (1991)[20] hypothesis for NOx removal. ........... 65 Review of Bortzs (1994)[45] hypothesis for NO2 formation......................... 69 Review of experimental validation findings for the hypothesized reactions. ....................................................................................................... 72

2.2.6.1 2.2.6.1.1

Review of sodium sulfite reaction findings................................................... 72 Millers (1986)[19] sodium sulfite reaction findings. ..................................... 72

2.2.6.1.2 2.2.6.1.3 2.2.6.1.4 2.2.6.2 2.2.6.2.1

Blands (1990)[16] sodium sulfite reaction findings....................................... 74 Verlaeten et al.s (1991)[20] sodium sulfite reaction findings........................ 74 Lai and Yesavage (1994)[21] sodium sulfite reaction findings. ..................... 75 Review of sodium pyrosulfite experimental findings. .................................. 76 Blands (1990)[16] sodium bisulfite and sodium pyrosulfite reaction findings.......................................................................................................... 76

2.2.6.2.2 2.2.6.2.3 2.2.6.3 2.2.6.4 2.2.6.5 2.3

Verlaeten et al.s (1991)[20] sodium pyrosulfite reaction findings. ............... 77 Lai and Yesavage (1994)[21] sodium pyrosulfite reaction findings. .............. 77 Review of sodium sulfate reaction findings. ................................................. 78 Review of sodium nitrite and sodium nitrate reaction findings. ................... 79 Review of gas phase interactions of SO2 and NOx........................................ 80

Conclusions ........................................................................................................... 82

CHAPTER 3...................................................................................................................... 85 THEORETICAL DEVELOPMENT................................................................................ 85 3.1 3.2 3.3 3.3.1 3.3.2 3.3.3 3.4 3.4.1 Introduction to theoretical development ............................................................... 85 Summary of key undisputed reactions and end products...................................... 86 The criteria for assessment.................................................................................... 91 Criteria for SO2 removal ............................................................................... 91 Criteria for NOx removal. ............................................................................. 92 Criteria for NO2 formation. ........................................................................... 97 Conformity of each of the past hypothesized pathways with the criteria. ............ 99 Conformity of Millers hypothesized reaction pathway to the criteria for SO2 removal, NO2 formation, and NOx removal................................... 100

3.4.1.1 3.4.1.2

Conformity of Millers pathway to the SO2 removal criteria...................... 100 Conformity of Millers (1986)[19] pathway to the NOx removal criteria.......................................................................................................... 101

3.4.1.3

Conformity of Millers (1986)[19] pathway to the NO2 formation criteria.......................................................................................................... 102

3.4.1.4 3.4.2

Validity of Millers hypothesized pathway. ................................................ 103 Conformity of Blands (1990) [16] original hypothesized reaction pathway to the criteria for SO2 removal, NOx removal and NO2 formation. .................................................................................................... 104

3.4.2.1

Conformity of Blands (1990) [16] original pathway to the SO2 removal criteria. .......................................................................................... 105

3.4.2.2

Conformity of Blands (1990) [16] original pathway to the NOx removal criteria. .......................................................................................... 106

3.4.2.3

Conformity of Blands (1990) [16] original pathway to the NO2 formation criteria......................................................................................... 108

3.4.2.4

Validity and value of Blands (1990) [16] original hypothesized pathway. ...................................................................................................... 110

3.4.3

Conformity of Blands (1990) [16] final hypothesized reaction pathway to the criteria for SO2 removal, NOx removal and NO2 formation. ............ 110

3.4.3.1

Conformity of Blands (1990) [16] final pathway to the SO2 removal criteria.......................................................................................................... 111

3.4.3.2

Conformity of Blands (1990) [16] final pathway to the NOx removal criteria.......................................................................................................... 112

3.4.3.3

Conformity of Blands (1990) [16] final pathway to the NO2 formation criteria.......................................................................................................... 115

3.4.3.4 3.4.4

Validity and value of Blands (1990) [16] final hypothesized pathway. ....... 116 Conformity of Verlaeten et al.s (1991)[20] hypothesized reaction pathway to the criteria for SO2 removal, NO2 formation, and NOx removal........................................................................................................ 118

3.4.4.1

Conformity of Verlaeten et al.s (1991)[20] pathway to the SO2 removal criteria. .......................................................................................... 118

3.4.4.2

Conformity of Verlaeten et al.s (1991)[20] pathway to the NOx removal criteria. .......................................................................................... 120

3.4.4.3

Conformity of Verlaeten et al.s (1991)[20] pathway to the NO2 formation criteria......................................................................................... 122

3.4.4.4 3.4.5

Validity and value of Verlaeten et al.s hypothesized pathway. ................. 122 Conformity of Bortzs hypothesized reaction pathway to the criteria for SO2 removal, NO2 formation, and NOx removal................................... 126

3.4.5.1 3.4.5.2 3.4.5.3 3.4.5.4 3.4.6 3.5

Conformity of Bortzs pathway to the SO2 removal criteria....................... 126 Conformity of Bortzs pathway to the NOx removal criteria. ..................... 127 Conformity of Bortzs pathway to the NO2 formation criteria. .................. 129 Validity and value of Bortzs hypothesized pathway.................................. 131 Assessment of Lai and Yesavages experimental findings. ........................ 131 Summary of conformity of all of the hypothesized reaction pathways to the criteria for SO2 removal, NOx removal, and NO2 formation................. 133

3.6

Summary of key controversial reactions and end products................................. 134

3.7 3.7.1 3.7.1.1

Development of a hypothesized reaction pathway.............................................. 137 Conformity of the foundation of the hypothesized reaction pathway. ........ 137 Conformity of the foundation of the hypothesized reaction pathway to the SO2 removal criteria. ............................................................................. 138

3.7.1.2

Conformity of the foundation of the hypothesized reaction pathway to the NOx removal criteria. ............................................................................ 140

3.7.1.3

Conformity of the foundation of the hypothesized reaction pathway to the NO2 formation criteria........................................................................... 145

3.7.2 3.7.2.1

Assembling the overall hypothesized reaction pathway ............................. 146 Subsets of the overall hypothesized reaction pathway with respect to temperature.................................................................................................. 147

3.7.2.1.1 3.7.2.1.2 3.7.2.1.3 3.7.2.1.4 3.7.2.1.5 3.8 3.9

Overall hypothesized reaction pathway Subset for T< 50C. ..................... 148 Overall hypothesized reaction pathway, Subset for 50C<T<120C......... 149 Overall hypothesized reaction pathway Subset for 120C<T<150. ......... 151 Overall hypothesized reaction pathway Subset for 150C<T<250C........ 154 Hypothesized reaction pathway Subset for T > 250C. .............................. 158

Thermodynamic Evaluation of overall hypothesized reaction pathway. ............ 159 Conclusions of theoretical development. ............................................................ 163

CHAPTER 4.................................................................................................................... 165 METHODOLOGY......................................................................................................... 165 4.1 4.2 4.2.1 Introducton .......................................................................................................... 165 Experimental approach and scope of study......................................................... 165 The validation status of the reactions in the overall reaction pathway. ...... 165 6

4.2.2 4.2.3 4.3 4.3.1 4.3.2 4.3.2.1 4.3.2.2 4.3.2.3 4.3.2.4 4.3.2.5 4.3.2.6 4.3.2.7 4.3.3 4.4 4.4.1 4.4.2 4.4.3 4.4.4 4.4.5 4.4.6

The simplified three-step reaction pathway for NO2 formation.................. 170 Objective and scope of study ...................................................................... 170 Apparatus ............................................................................................................ 173 Thermal gravimetric analyzer. .................................................................... 173 Isothermal fixed bed reactor........................................................................ 174 The flue gas mixing station ......................................................................... 176 The gas pre-heater. ...................................................................................... 181 Powder injection module............................................................................. 182 The main isothermal reaction zone. ............................................................ 183 The quick-quench transfer zone for sample removal. ................................. 186 The gas analyzer station for SO2 and NOx monitoring. .............................. 188 The data acquisition computer system. ....................................................... 189 Precision and accuracy ................................................................................ 190 Analytical methods.............................................................................................. 190 Elemental analysis....................................................................................... 190 X-ray diffraction analysis............................................................................ 193 BET surface area measurements ................................................................. 194 Scanning electron microscope microstructure analysis .............................. 195 Time-of-flight static secondary ion mass spectrometry .............................. 195 X-ray photoelectron spectroscopy /electron spectroscopy for chemical analysis......................................................................................... 197

4.5 4.5.1

Test Procedures ................................................................................................... 198 Procedures for preliminary tests.................................................................. 198

4.5.1.1

Test procedure to investigate the effect of temperature on NO2 formation from the reaction of sodium bicarbonate and SO2 flue gas with NO. ...................................................................................................... 198

4.5.1.2 4.5.1.2.1

Test procedures for material characterization studies. ................................ 199 Conversion tracking test procedures for reactions related to sodium bicarbonate decomposition inhibition. ........................................................ 199

4.5.1.2.2

Conversion tracking test procedures for the study of the finite time period of sodium nitrate accumulation........................................................ 201

4.5.1.2.3

Procedure to study surface area changes for the reaction of sodium bicarbonate with SO2 flue gas with NO at 150 and 250C. ........................ 205

4.5.2 4.5.2.1

Procedures for reaction validation tests. ..................................................... 206 Test procedure to validate the reaction of sodium sulfite with NO and O2................................................................................................................. 206

4.5.2.1.1 4.5.2.1.2 4.5.2.2

Special preparations of sodium sulfite for reactivity tests. ......................... 208 Method of producing reaction product for material analysis. ..................... 209 Test procedure to validate the reaction of sodium nitrate with SO2 and O2................................................................................................................. 210

4.5.3 4.5.3.1

Procedures for mitigation study tests. ......................................................... 211 Test procedure to investigate the effect of oxygen percent on NO2 formation from the reaction of sodium bicarbonate and SO2 flue gas with NO. ...................................................................................................... 211

CHAPTER 5.................................................................................................................... 213 PRELIMINARY TESTS................................................................................................ 213

5.1 5.2

Introduction ......................................................................................................... 213 Test for proper operation of apparatus.- Measurements of NO2 versus temperature for the reaction of sodium bicarbonate with SO2 flue gas with NO. ...................................................................................................... 214

5.3 5.3.1

Material characterization tests............................................................................. 217 Sodium bicarbonate decomposition inhibition in SO2 flue gas with NO. .............................................................................................................. 217

5.3.2

Sodium nitrate accumulation by sodium bicarbonate reaction with SO2 flue gas with NO. ................................................................................. 236

5.3.2.1

The effect of flue gas water vapor on sodium nitrate accumulation for sodium bicarbonate reaction with SO2 flue gas with NO............................ 262

5.3.3

Sodium bicarbonate surface area and microstructure changes that occur during reaction with SO2 flue gas with NO....................................... 271

5.3.4

Sulfation product analysis for sodium bicarbonate reaction in SO2 flue gas without NO. ................................................................................... 276

5.3.5 5.4 5.5

Sodium bisulfite and sodium pyrosulfite decomposition products. ............ 282 Conclusions ......................................................................................................... 284 Follow on tests .................................................................................................... 288

CHAPTER 6.................................................................................................................... 290 VALIDATION OF THE REACTION OF SODIUM SULFITE WITH NO AND O2................................................................................................................. 290 6.1 Introduction ......................................................................................................... 290

6.1.1

Examination of past studies of the reaction of sodium sulfite with NO and O2. ......................................................................................................... 291

6.2 6.3 6.3.1 6.3.2

Experimental approach........................................................................................ 295 Experimental results of the reaction tests............................................................ 296 Experimental result of blank test................................................................. 297 Experimental results of technical grade sodium sulfite reaction with NO and O2. .................................................................................................. 298

6.3.3

Experimental results of freshly made sodium sulfite reaction with NO and O2. ......................................................................................................... 298

6.3.3.1

Analysis of reaction product formed by specially prepared freshly made sodium sulfite with NO and O2.......................................................... 300

6.3.3.1.1 6.3.3.1.2 6.3.3.1.3 6.4 6.5

Elemental Analysis...................................................................................... 302 Time of flight static secondary ion mass spectrometry analysis. ................ 303 Electron Spectroscopy for chemical analysis.............................................. 309

Conclusions ......................................................................................................... 311 Areas for further studies...................................................................................... 314

CHAPTER 7.................................................................................................................... 316 VALIDATION OF THE REACTION OF SODIUM NITRITE WITH SO2................. 316 7.1 7.1.1 Introduction ......................................................................................................... 316 Examination of past studies of the reaction of sodium nitrate with SO2. ............................................................................................................. 332 7.2 7.3 Experimental approach........................................................................................ 333 Experimental results of reaction tests.................................................................. 335

10

7.3.1 7.3.2

Experimental results of blank test. .............................................................. 336 Experimental results of SO2 and O2 reaction with technical grade sodium nitrate. ............................................................................................. 336

7.3.3

Experimental results of SO2 and O2 reaction with specially prepared sodium nitrate prepared fresh by reaction of sodium carbonate with NO2.............................................................................................................. 336

7.3.3.1

The effect of temperature on NO2 formation from the reaction of the specially prepared sodium nitrate and SO2 and O2. .................................... 337

7.3.4

Experimental results of SO2 and O2 reaction with specially prepared sodium nitrate prepared fresh by reaction of sodium sulfite with NO and O2. ......................................................................................................... 339

7.3.4.1

The effect of temperature on NO2 formation by the SO2 reaction with specially prepared sodium nitrate prepared fresh by reaction of sodium sulfite with NO and O2. .................................................................. 341

7.3.5

Experimental results of SO2 and O2 reaction with specially prepared sodium nitrate prepared fresh by the reaction of the three-step product with NO and O2. .......................................................................................... 343

7.4

X-ray diffraction analysis of sulfate reaction products from specially prepared sodium nitrate reaction with SO2 and O2. .................................... 343

7.5 7.6

Conclusions ......................................................................................................... 352 Areas for further studies...................................................................................... 355

CHAPTER 8.................................................................................................................... 356

11

MITIGATION OF PLUME DISCOLORATION FOR SODIUM BICARBONATE INJECTION IN SO2 FLUE GAS WITH NO ................ 356 8.1 8.2 8.2.1 Introduction ......................................................................................................... 356 Plume discoloration mitigation options............................................................... 358 Plume discoloration mitigation by limiting oxygen concentration in the flue gas. ................................................................................................. 358 8.2.2 Plume discoloration mitigation by adjusting the injection flue gas temperature.................................................................................................. 360 8.2.3 8.2.4 Plume discoloration mitigation by injecting flue gas additives. ................. 362 Plume discoloration mitigation by removing NO from the flue gas prior to sodium bicarbonate injection.......................................................... 363 8.2.5 Plume discoloration mitigation by injecting sodium carbonate after the sodium bicarbonate flue gas desulfurization injection step................... 364 8.2.6 Conclusions ................................................................................................. 365

CHAPTER 9.................................................................................................................... 367 CONCLUSIONS............................................................................................................ 367 9.1 9.1.1 9.1.2 9.1.3 9.1.4 9.1.5 9.1.6 Chapter Summaries ............................................................................................. 367 Summary of Chapter 1 ................................................................................ 367 Summary of Chapter 2 ................................................................................ 367 Summary of Chapter 3 ................................................................................ 371 Summary of Chapter 4 ................................................................................ 372 Summary of Chapter 5 ................................................................................ 375 Summary of Chapter 6 ................................................................................ 379

12

9.1.7 9.1.8 9.2

Summary of Chapter 7 ................................................................................ 382 Summary of Chapter 8 ................................................................................ 385 Overall Conclusions ............................................................................................ 387

BIBLIOGRAPHY .......................................................................................................... 388 APPENDIX A ................................................................................................................ 394 A.1 A.2 A.3 A.4 A.5 Introduction ......................................................................................................... 394 Gibbs free energy equation ................................................................................. 394 Reactions thermodynamically evaluated............................................................. 396 Equilibrium calculations from 27 to 627C for each of the reactions................. 396 Summary of all equilibrium calculations from 27 to 627 C................................ 406

APPENDIX B CALCULATIONS OF CONVERSION FOR SODIUM BICARBONATE AND SODIUM CARBONATE TRACKING REACTION EXPERIMENTS .................................................................... 410 B.1 B.2 B.3 Equations for molar percentage of each compound in the product..................... 410 Equations for elemental weight percent of hydrogen, nitrogen and sulfur. ........ 412 Determining the weight fractions of the respective compounds in the product......................................................................................................... 414 B.4 Calculating the moles of each compound in the product. ................................... 417

APPENDIX C UNITED STATES PATENT OF NITROGEN DIOXIDE MITIGATION PROCESS........................................................................... 418

13

LIST OF FIGURES

Figure 1-1 Dry sorbent baghouse injection method for coal burning utility applications. ................................................................................................... 22 Figure 1-2 Development history of the dry sodium bicarbonate bag-house injection method............................................................................................. 24 Figure 2-1 Spectra of NO2 absorbency, base e as a function of wavelength. .................. 32 Figure 2-2 Schematic of the sodium bicarbonate decomposition reaction. ..................... 40 Figure 2-3 Chemical structure of sodium carbonate. ....................................................... 41 Figure 2-4 Schematic of the reaction pathway for of sodium carbonate with SO2.......... 42 Figure 2-5 Schematic of Erdos postulated reaction pathway of sodium bicarbonate with flue gas with SO2................................................................ 43 Figure 2-6 Sulfation products formed for the reaction of sodium bicarbonate with SO2 without NOx............................................................................................ 44 Figure 2-7 Total weight percent of sodium sulfite versus temperature............................ 45 Figure 2-8 Schematic of Keener and Khang.s (1982,1993)[9,52]reaction pathway for SO2 removal by sodium bicarbonate. ....................................................... 47 Figure 2-9 Schematic of Knights (1977)[18] observed reaction of sodium carbonate with NO2........................................................................................ 48 Figure 2-10 Weight percent sodium sulfite, Na2SO3, and weight percent sodium sulfate, Na2SO4, in product as a function of temperature. ............................. 51 Figure 2-11 NO2 formation as a function of reactor temperature for sodium bicarbonate. .................................................................................................... 52

14

Figure 2-12 Schematic of Millers (1986)[19] Hypothesized Reaction Pathway (a) for sodium carbonate with flue gas with SO2 and NO, and (b) for sodium hydroxide with flue gas with SO2 and NO........................................ 59 Figure 2-13 Schematic of Blands(1990)[16] first hypothesized reaction pathway for NO2 formation and NOx removal. ............................................................ 62 Figure 2-14 Schematic of Blands (1990)[16] final theory on the reaction pathway for SO2 and NO removal and NO2 formation.................................. 65 Figure 2-15 Schematic of Verlaeten et al.s (1991)[20] hypothesis on SO2 and NOx removal........................................................................................................... 68 Figure 2-16 Schematic of schematic of Bortzs (1994)[45] hypothesis of SO2 and NOx removal and NO2 formation................................................................... 71 Figure 2-17 Schematic of mixed SOx/NOx compounds formed in the chamber process for the manufacture of sulfuric acid.................................................. 82 Figure 3-1 Compilation of essential reaction pathway segments from past researchers. ................................................................................................... 137 Figure 3-2 The reaction pathway of sodium carbonate with SO2 flue gas with NO reduced from the foundation of the hypothesized reaction pathway. .......... 142 Figure 3-3 The overall hypothesized reaction path........................................................ 147 Figure 3-4 Schematic of overall hypothesized reaction pathway subset for T< 50C ............................................................................................................. 149 Figure 3-5 Schematic of overall hypothesized reaction pathway, Subset for 50C<T<120C ............................................................................................ 151

15

Figure 3-6 Schematic of overall hypothesized reaction pathway, for 120C<T 150C ........................................................................................................... 154 Figure 3-7 Schematic diagram of simplified three-step reaction for NO2 formation...................................................................................................... 157 Figure 3-8 Schematic of overall hypothesized reaction pathway for T > 250C............ 159 Figure 3-9 Compilation of essential reaction pathway segments from past researchers. ................................................................................................... 160 Figure 4-1 Apparatus...................................................................................................... 175 Figure 4-2 Nitrogen rotometer calibration curve ........................................................... 177 Figure 4-3 Carbon dioxide rotometer calibration curve................................................. 177 Figure 4-4 Oxygen rotometer calibration curve............................................................. 178 Figure 4-5 SO2 Mass flow controller calibration curve ................................................. 179 Figure 4-6 NO mass flow controller calibration curve .................................................. 179 Figure 4-7 Calibration curve of heater controller 1........................................................ 185 Figure 4-8 Calibration curve of heater controller 2........................................................ 186 Figure 4-9 Calibration curves for elemental analysis. ................................................... 192 Figure 5-1 Validation test results of peak NO2 formation versus temperature. ............. 215 Figure 5-2 Snapshot of ash layers from reaction sequence............................................ 216 Figure 5-3 Conversion of sodium bicarbonate in inert nitrogen gas at 111C............... 220 Figure 5-4 Conversion of sodium bicarbonate in SO2 flue gas without NO at 111C. .......................................................................................................... 222 Figure 5-5 Conversion of sodium bicarbonate in SO2 flue gas with NO at 111C........ 224

16

Figure 5-6 Conversion of sodium bicarbonate in nitrogen gas with NO, O2 at 111C. .......................................................................................................... 227 Figure 5-7 Conversion of sodium bicarbonate in nitrogen gas with NO2 at 111C....... 229 Figure 5-8 Conversion of sodium bicarbonate in SO2 flue gas with NO2 at 111C. ..... 231 Figure 5-9 Conversion of sodium carbonate in nitrogen gas with NO2 at 111C. ......... 233 Figure 5-10 Conversion of sodium carbonate in nitrogen gas with NO and O2 at 111C. .......................................................................................................... 235 Figure 5-11 Conversion of sodium bicarbonate in inert nitrogen at 150C. .................. 239 Figure 5-12 Conversion of sodium bicarbonate in SO2 flue gas with NO at 150C. .......................................................................................................... 241 Figure 5-13 Conversion of sodium bicarbonate in nitrogen with NO and O2 at 150C. .......................................................................................................... 247 Figure 5-14 Conversion of sodium bicarbonate in nitrogen with NO2 at 150C for 120 minutes. ................................................................................................. 250 Figure 5-15 Conversion of sodium bicarbonate in SO2 flue gas with NO2 at 150C. .......................................................................................................... 253 Figure 5-16 Conversion of sodium carbonate with NO2 at 150C................................ 257 Figure 5-17 Conversion of sodium carbonate in SO2 flue gas with NO at 150C. ........ 259 Figure 5-18 Conversion of sodium carbonate in SO2 flue gas with NO2 at 150C. ....... 260 Figure 5-19 Conversion of sodium bicarbonate in humidified inert nitrogen at 150C. .......................................................................................................... 263 Figure 5-20 Conversion of sodium bicarbonate in humidified inert nitrogen with 5% H2O at 250C. ........................................................................................ 265 17

Figure 5-21 Conversion of sodium bicarbonate in humidified SO2 flue gas with NO at 150C................................................................................................. 267 Figure 5-22 Conversion of sodium bicarbonate in SO2 flue gas with NO at 250C. .......................................................................................................... 270 Figure 5-23 BET surface area measurements versus time for the reaction products of sodium bicarbonate in inert nitrogen. ...................................................... 272 Figure 5-24 Photo Micrographs of product particle surface from the reaction for sodium bicarbonate with SO2 flue gas with NO at 150C (a) after 10 minutes, (b) after 300 minutes. .................................................................... 273 Figure 5-25 BET surface area measurements versus time for the reaction products of sodium bicarbonate in inert nitrogen and in SO2 flue gas with NO at 250C. ...................................................................................................... 274 Figure 5-26 Photo Micrographs of product particle surface from the reaction for sodium bicarbonate with SO2 flue gas with NO at 250C (a) after 10 minutes, (b) after 60 minutes. ...................................................................... 275 Figure 5-27 Phase diagram[80] of sodium nitrate and sodium sulfate............................. 276 Figure 5-28 X-ray diffraction analysis of product from reaction of sodium bicarbonate with SO2 flue gas without NO at 150C. The scan in (a) is matched with the standard files for sodium sulfite in (b), and sodium bicarbonate in (c). Identical product was found for the reaction in SO2 flue gas without NO at 111C. ............................................................. 279 Figure 5-29 X-ray diffraction analysis of product from reaction of sodium bicarbonate with SO2 flue gas without NO at 250C................................... 281 18

Figure 5-30 Thermal gravimetric analysis of sodium pyrosulfite.................................. 283 Figure 6-1 NOx analysis for blank run........................................................................ 297 Figure 6-2 NOx versus time for the reaction of freshly made sodium sulfite with NO and O2 at 175C..................................................................................... 299 Figure 6-3 Diagram of reacted particle from experiment .............................................. 301 Figure 6-4 SIMS Positive ion scan of high intensity peak............................................. 305 Figure 6-5 SIMS positive ion scan of high intensity peak ............................................. 306 Figure 6-6 SIMS negative ion scans of high intensity peaks ......................................... 307 Figure 6-7 SIMS negative ion scans of high intensity peaks ......................................... 308 Figure 6-8 ESCA scan of product from the reaction of sodium sulfite with NO and O2........................................................................................................... 310 Figure 7-1 NO2 formation versus temperature for the reaction of NaNO3 with SO2 and O2. .................................................................................................. 338 Figure 7-2 NO formation versus temperature for the three-step reaction sequence. ..... 340 Figure 7-3 NO2 formation from the three step reaction sequence versus temperature................................................................................................... 342 Figure 7-4 X-ray diffraction analysis of product from reaction of sodium bicarbonate with SO2 flue gas with NO at 111C........................................ 349 Figure 8-1 NO2 Formation versus O2 percent in flue gas. ............................................. 360 Figure APPENDIX A-1 Summary of Gibbs free energy versus temperature for all of the reaction equations. ............................................................................. 408 Figure APPENDIX B-1 Details of numerical computations for weight fractions.......... 416

19

LIST OF SYMBOLS

f NaHCO3 f Na2 CO3

f Na2 SO3 f Na2 SO4

weight fraction of the product that is sodium bicarbonate weight fraction of the product that is sodium carbonate weight fraction of the product that is sodium sulfite weight fraction of the product that is sodium sulfate weight fraction of the product that is sodium nitrate Gibbs free energy of the reaction equilibrium constant equilibrium constant equilibrium constant for each reactant and product substances moles of sodium bicarbonate per 100 grams of product (mol/100g) moles of sodium carbonate per 100 grams of product (mol/100g) moles of sodium sulfite or sulfate per 100 grams of product (mol/100g),

f NaNO3 Gr Kr Kr Kf( i ) mNaHCO3 mNa2 CO3 mNa2 SO3/ 4

mNaNO3 MWH MWN MWS MWNaHCO3 MWNa2 CO3

moles of sodium nitrate per 100 grams of product (mol/100g) molecular weight of hydrogen (grams/mole) molecular weight of nitrogen (grams/mole) molecular weight of sulfur (grams/mole) molecular weight of sodium bicarbonate (grams/mole) molecular weight of sodium carbonate (grams/mole)

20

MW Na2 SO3 MWNa2 SO4

molecular weight of sodium sulfite (grams/mole) molecular weight of sodium sulfate (grams/mole) molecular weight of sodium nitrate (grams/mole) universal gas constant temperature is the stoichiometric coefficient of substance i weight percent hydrogen weight percent nitrogen weight percent sulfur molar percent of Na in the form of sodium bicarbonate in the product molar percent of Na in the form of sodium carbonate in the product molar percent of Na in the form of sodium sulfite or sulfate in the product molar percent of Na in the form of sodium nitrate in the product

MWNaNO3 R T i wt%(H) wt%(N) wt%(S) X NaHCO3 X Na2 CO3 X Na2 SO3 X NaNO3

21

CHAPTER 1 INTRODUCTION 1.1 Background

The Environmental Protection Agency (EPA) through the Acid Deposition Programs of Title IV of the Clean Air Act has promulgated stringent regulations limiting sulfur dioxide, SO2, and nitrogen oxide, NOx, emissions from US electric utilities.[1] In the year 2000, more than 1000 utilities will be required to comply with Phase II requirements of the Acid Rain Rule. There are several options available to meet these regulations including an installation of end-of-pipe SO2 and NOx removal device.[2] Dry sorbent baghouse injection is an attractive end-of-pipe SO2 removal option. This method is economically desirable since it involves low capital investment, is very simple to operate, and can be easily retrofitted into existing hardware configurations.[2] Figure 1.1 shows a representative diagram of the scrubbing method.
Cleaned Flue Gas Flue Gas: N2, CO2,O2, H2O, SO2,NO Dry Sorbent Injection Air Preheater Coal and Preheated Air Combustion Zone Baghouse Particulate Collector Stack

Ash

Fly Ash

Spent Sorbent and Fly Ash for recycle

Figure 1-1 Dry sorbent baghouse injection method for coal burning utility applications.

22

In this method, dry pulverized sorbent is injected into the coal combustion flue gas duct downstream of the air preheater. The sorbent is then collected in a baghouse collector. The injected sorbent reacts with gaseous SO2 in the flue gas during injection flight and within the baghouse. The solid reaction product that is captured on the sorbent is either regenerated for reuse or disposed of. Lime based dry alkali sorbents are commonly used for scrubbing in the eastern regions of the US since they are readily available in these regions. Sodium based dry alkali sorbents extracted from the Four Corners region of the western US, Wyoming, Idaho, Utah, and Colorado, may potentially be used for scrubbing in the western regions of the US.[3] 1.2 The Problem Dry sodium bicarbonate sorbent injection has been under development since the early 1960s in the US and Europe.[4] It has not yet been fully implemented as a viable technology option for use by electric utilities because of issues surrounding plume discoloration that were discovered during full-scale demonstration tests.[5] The plume discoloration has not yet been resolved for this scrubbing method. Figure 1-2 traces the development history of the dry sodium bicarbonate baghouse injection method.

23

Initial Laboratory

Full Scale Utility Demonstration Tests 1970-1986 Determine Reaction Pathway of NO2 2001

Enable Full Scale Implementation to Electric Utilities

Plume Discoloration Observed 1986

Devise Method to Abate NO2 2001

Figure 1-2 Development history of the dry sodium bicarbonate bag-house injection method. The dashed line represents the steps that are made within this study. The circle with the back slash indicates the route that was not taken; instead the detour drawn was taken. Note that the ultimate goal of the development of dry sodium bicarbonate baghouse injection is full-scale implementation of the technology into the marketplace. Figure 1.2 indicates that initial laboratory research and development occurred between 1960 and the mid-1980s.[4,6,7,8.9.10,11] Primarily, these laboratory investigations established the fundamentals of the reaction of sodium bicarbonate with SO2. Pilot and full-scale coal combustion demonstration tests were conducted between 1970 to 1986.[12] From 1984 to 1986 the Electric Power Research Institute (EPRI) successfully demonstrated in full scale SO2 removal rates of 70 to 90% for compliance with New Source Performance Standards.[13,14 ] It was during these 1986 full-scale demonstration tests that plume discoloration was noticed.[14] The undesirable reddish-brown discoloration of the exit stack gas was observed when sodium bicarbonate was injected into the duct during demonstration tests. Although the discolored exit gas did not violate stack opacity regulations and was fully compliant with all SO2 and NOx emission

24

regulations, it poses a serious public relations problem since discoloration of any kind is typically aesthetically unacceptable and a nuisance to local communities. As a result to date the technology has not been fully implemented into the marketplace. Instead research and development of the process has continued in the laboratory to determine the cause and the exact reaction pathway responsible for the plume discoloration. The reaction pathway in an effort to devise an abatement method[15] to ultimately allow the technology to reach full-scale implementation for use by utilities. Very shortly after the discoloration was first noticed investigators determined that the discoloration was caused by the formation of a small concentration of NO2.[16] It was common knowledge that coal combustion flue gas contained colorless NO[17]. But it was unclear how NO2 was formed from the flue gas NO during sodium bicarbonate injection. The cause of the NO2 formation has not to date been determined. Prior to learning of plume discoloration by NO2 formation, NO in flue gas was assumed to be nonparticipatory in desulfurization reactions based on studies by Knight and others that found NO to be unreactive with sodium bicarbonate alone.[18] 1.3 Prior study of this problem. As a result prior to 1986, the reaction of sodium bicarbonate with SO2 flue gas was experimentally studied in the absence of NO, under the assumption that NO had no role in desulfurization. After the discoloration findings in 1986, NO was included in the SO2 flue gas mixture for these sodium bicarbonate reaction studies. The role of NO in the reactions began to be closely examined in an attempt to reveal that there was in fact a complex interaction of NO in the reaction pathway. NO2 formation was found to occur by the reaction but little more of this complex interaction was known. 25

One theory hypothesized that NO2 was formed by the reaction of NO with a sulfation reaction product or sulfation product intermediate compound.[12] Between 1986 and 1994 numerous researchers investigated possible reactions of NO with various desulfurization products and intermediate product compounds.[19,20,21,22] But none of these studies successfully revealed any sulfation compound to be responsible for the NO2 formation. 1.4 Goals and objectives of this study The goal of this study was to enable full-scale implementation of the dry sodium bicarbonate bag-house injection method for SO2 removal allowing this technology to enter the marketplace for utilities to use for Phase II Acid Rain regulations. The specific objectives were to: 1) Determine the reaction pathway for NO2 formation during dry sodium bicarbonate injection. 2) Devise a method of abatement for plume discoloration during dry sodium bicarbonate injection based on knowledge of the NO2 formation reaction pathway. 1.5 Approach and scope Chapter 2 presents relevant literature on plume discoloration and reaction chemistry of sodium bicarbonate flue gas desulfurization. The plume discoloration section includes a general historical understanding of plume discoloration and plume discoloration reports from sodium bicarbonate and sodium sorbent flue gas desulfurization field trials. Also Chapter 2 provides a description of how NO2 was found to be the discoloring agent. The reaction chemistry section of Chapter 2 includes a review

26

of reaction studies of sodium bicarbonate decomposition, sodium bicarbonate reaction with flue gas containing SO2 without NO, sodium bicarbonate and sodium carbonate reaction with NO and NO2 but without SO2, and sodium bicarbonate reaction with flue gas containing SO2 and NO. Additionally past investigators hypotheses on NO2 formation and NOx removal are reviewed. Illustrative schematic diagrams of the past hypothesized reaction paths are provided wherever possible to assist in pathway visualization. Experimental validation findings for these hypotheses are also reviewed including findings on reactions of sodium sulfite, sodium pyrosulfite, and sodium sulfate with NO and O2, reactions of sodium nitrite and sodium nitrate with SO2 and O2, and gas phase interactions of SO2 and NOx. Chapter 3 provides the theoretical development of the overall reaction pathway for NO2 formation and NOx removal. Criteria for SO2 removal, NOx removal, and NO2 formation are established based on past undisputed experimental findings reported in Chapter 2. The past proposed hypothesized reaction pathways from past researchers described in Chapter 2 are assessed in Chapter 3 for conformity with the established reaction criteria. The conforming portions of the past researchers hypothesized reaction pathways then serve as a foundation for a new hypothesized reaction pathway constructed for investigation in this study. This new hypothesized reaction pathway is checked for thermodynamic feasibility at the end of Chapter 3 with the supporting calculations described in Appendix A. Chapter 4 includes a description of the experimental methodology used to verify and validate the new hypothesized reaction pathway proposed for study. The experimental approach and objectives of the study are included. The laboratory

27

apparatus is also described in Chapter 4. In addition the special methods used in the laboratory for reaction reagent preparation and handling are described. The particular experimental test procedures are explained. The specific analytical test methods used to characterize and determine the reaction products are given. And the quality assurance steps employed are also included in Chapter 4. Chapter 5 presents the results of the preliminary tests. In particular the reaction of sodium bicarbonate with flue gas with SO2 and NO is investigated with respect to temperature in the reactor setup to show that the apparatus produces NO2 formation comparable to NO2 formation previously documented by past investigators. Chapter 5 also includes several preliminary material characterization studies. Sodium bicarbonate decomposition inhibition in SO2 flue gas with NO is studied by conversion tracking reactions. Sodium nitrate accumulation during sodium bicarbonate reaction in SO2 flue gas with NO is studied including the effect of added humidity by conversion tracking reactions. A description of the conversion tracking calculations is given in Appendix B. Sodium bicarbonate surface area and microstructure changes during reaction with SO2 flue gas with NO are also studied. Sulfation product formation by the reaction of sodium bicarbonate and SO2 flue gas without NO is also studied. And sodium bisulfite and sodium pyrosulfite decomposition is studied including the decomposition product identification. In Chapter 6 the key un confirmed reaction of sodium sulfite with NO and O2that is in the hypothesized reaction pathway is experimentally validated. This reaction, which is one of the three key reactions in the hypothesized reaction for NO2 formation, is validated. Past studies of this reaction are closely examined in Chapter 6 to determine the

28

reason why different researchers obtained different experimental results in the past. The effect of sodium sulfite freshness on the reaction is experimentally studied. In particular, various different sodium sulfite preparations are experimentally tested to demonstrate the experimental importance of sodium sulfite freshness on the reaction. Analytical results of product formation from the reaction of freshly made sodium sulfite with NO and O2 are presented including elemental analysis, X-ray diffraction (XRD), infrared (IR), X-ray photoelectron spectroscopy, Time-of-Flight Secondary Ion Mass Spectrometry (TOFSIMS), and Electron Spectroscopy for Chemical Analysis (ESCA). Additionally in Chapter 7 the reaction of sodium nitrate with SO2 and O2, another key reaction in the hypothesized reaction pathway for NO2 formation is validated. Various different sodium nitrate preparations are studied to demonstrate the importance of sodium nitrate freshness in the reaction for NO2 formation. Chapter 7 provides X-ray diffraction (XRD) analysis of the reaction product formed by the reaction of freshly made sodium nitrate with SO2 and O2. Chapter 8 investigates optional mitigation methods for plume discoloration during sodium bicarbonate injection into SO2 flue gas with NO. Changes in flue gas oxygen concentration are experimentally investigated for its effect on NO2 formation. Adjusting the duct temperature is contemplated, as well as injecting flue gas additives to avoid NO2 formation. And finally, removing NO from the flue gas prior to sodium bicarbonate injection is considered as a mitigation option. A novel process using sodium bicarbonate reaction chemistry was devised to remove NO from the flue gas prior to the desulfurization step of sodium bicarbonate injection and is described in Appendix C.

29

And finally Chapter 9 is a summary of the main findings from the previous chapters.

30

CHAPTER 2

LITERATURE REVIEW This chapter reviews relevant literature on plume discoloration and sodium bicarbonate reaction chemistry. 2.1 Literature review of plume discoloration

The following areas of plume discoloration are reviewed: 1) General historical understanding of plume discoloration. 2) Plume discoloration reports from sodium bicarbonate and sodium sorbent flue gas desulfurization trials. 3) A description of how NO2 was found to be the discoloring agent during sodium bicarbonate injection demonstration tests. 2.1.1 History of plume discoloration Around the early part of the 20th century plume coloration was studied by the original air pollution scientists and engineers in the smoke and soot era of the industrial revolution in Europe and in the US. In the early 1900s ordinances prohibiting dense black smoke prompted study of smoke and methods of prevention. Soon smoke abatement legislation emerged with adopted definition of smoke color intensity. According to Stern (1977)[23] smoke darker then a #3 on a Ringelmann scale, equivalent to 60% opacity, was prohibited in most communities. As smoke prevention measures were put into place by the 1950s, such as improved fuel utilization methods, plumes of black smoke largely disappeared and plumes took on a new appearance.

31

The U.S. Edison Electric Institute and the Public Health Services and the Central Electricity Generating Board in Great Britain conducted fundamental opacity studies of non-black emission plumes in the 1960s.[24] By the 1980s the contributions of NO2 absorption and particulate matter (aerosol sulfates or fly ash) scattering on visibility were well established by Charlson and Ahlquist (1969)[25], Horvath (1971)[26], Husar and White (1976)[27], and White and Patterson (1981)[24]. Fundamentally, light attenuation by NO2 within the visible range was reported as early as the 1800s by Brewster (1834)[28]. The NO2 absorbency spectra in the visible light range occurs from wavelength 250 nanometers (nm) to 650 nm with its peak near 400 nm according to Davidson et al.(1988)[29] and others[30,31,32]. Figure 2.1 shows this light absorbency spectrum.
2.4

2.0

1.6

Absorbance

1.2

0.8

0.4

0.0 250 300 350 400 450 500 550 600 650

Wavelength (nm)

Figure 2-1 Spectra of NO2 absorbency, base e as a function of wavelength. Measurement from Davidson et al.(1988)[29] at 9.7C (50F), [NO2] = 7.20 x1014 molecules/cm3, path length equal to 48.6 meters.

32

Under fixed meteorological and stack conditions Lindau (1991) showed that NO2 followed the Beer-Lambert rule in which the light absorption was proportional to the NO2 concentration.[33] White and Patterson (1981)[24] report that Middleton in 1952 found a visual threshold, above which plume contrast was perceptible to the eye, between 2 to 5 %. Likewise, Hardison (1970)[34] found a threshold of an NO2 laden plume from a nitric acid processing plant to be approximately 2%. Hardison (1970)[34] formulated a correlation relation equation for threshold with respect to NO2 concentration of plume visibility and stack diameter:
60.9 =T D

(2.1)

where D is stack diameter in meters T is the threshold NO2 concentration for plume visibility in ppm According to this equation a plume is visible to the eye from a 2 meter stack when NO2 concentrations exceeded 50 ppm. White and Patterson (1981)[24] additionally showed that plume visibility depended on other factors besides NO2 concentration. For example, plume visibility was shown to depend also on background sky contrast to the stack. Melo and Stevens (1981)[35] explained that NOx in flue gas from fossil-fired power plant plumes is produced as an unwanted side effect of fuel combustion where the nitrogen from the air and the fuel is oxidized in the flame by excess oxygen and temperature. They showed that NOx formed in the combustion flame was in the form of colorless NO. Kircher and Hougen (1957)[36] showed that in some cases the colorless NO species

33

converted to the reddish brown colored NO2 species by the following equilibrium trimolecular oxidation reaction with oxygen in the atmosphere, 2NO (g) + O2(g) 2NO2(g) (2.2)

According to Duecker and West (1959)[37] the tri-molecular gas phase conversion of NO to NO2 was non-catalytic and possessed a negative temperature coefficient. At typical stack emitted temperature the NO to NO2 conversion is very slow and does not occur instantaneously but requires a relatively long time period. Duecker and West (1959)[37] provided estimates of the steady state conversion time for NO to convert to NO2 based on the findings of Stevenius-Nielsen. They report that over 1 hour was required to convert 50% of the NO to NO2 at an exit temperature of 141C (286F) for a gas mixture of 500 ppm NO and 6.5% O2 at atmospheric pressure. England and Corcoran (1975) also reported very slow reaction rates for the conversion of NO to NO2. Exit flue gas from coal combustion sources such as electric utilities generally contains small quantities of NO due to combustion. Because the reaction rate of steady state conversion of NO to NO2 is slow at stack gas temperatures, electric utility plumes at or near the stack are typically colorless. However with sufficient time the emitted NO is converted to NO2 according to the steady state equilibrium reaction (equation 2.2). As a result, colorless exit flue gas under stable meteorological conditions discolors at a distance away from the stack. These unattached visible plumes may be seen when plumes drift distances without significant lateral mixing (i.e. within an inversion). Under these conditions the NO in the maintained plume converts to NO2 a distance away from the stack. Melo and Stevens (1981)[35] reported the occurrence of unattached reddish brown NO2 plumes at distances of 0.5 to 5 km from respective power plant stacks. These

34

unattached plumes stretched in length from 10 to 100 km depending on the amount of mixing from meteorological conditions. 2.1.2 Reported observations of plume discoloration during sodium sorbent flue gas desulfurization trials. To reiterate, coal combustion flue gas is generally colorless as it exits the stack because the NOx in the exit flue gas is in the form of colorless NO. Since SO2 scrubbing of coal combustion flue gas by dry sodium bicarbonate injection was not initially known or anticipated to have any effect on the colorless NO in the flue gas, plume discoloration and NO interactions were not a focus during early development of the injection process. Although some reports of plume discoloration during dry sodium bicarbonate injection exist as far back as the 1960s through the mid-1980, the discoloration was never attributed to the sodium bicarbonate injection process or any interaction of NO with desulfurization by sodium bicarbonate. Reexamining the record however, reveals that there were some early indications that plume discoloration was directly associated with dry sodium bicarbonate injection. For example, yellow discoloration of scrubbed exit gas was observed during dry sodium bicarbonate bag-house injection trials at the Air Preheater Company for the National Air Pollution Control Administration as early as 1969 by Liu and Chaffee (1969)[38]. Reports of plume coloration varied from trial to trial. In some cases no color was observed and in other cases faint yellow color was observed and in other cases a deep reddish brown color was observed. But because the discoloration occurrences were sporadic and inconsistent they were initially overlooked as a concern. For example Muzio et al. (1985)[13] reported no plume discoloration for the pilot scale 22MW demonstration tests at Cameo Station. Confounding the situation 35

further was the fact that discoloration was reported to change its color intensity during demonstration testing even though combustion conditions and sodium bicarbonate injection conditions were held constant.[14,15,16] Because of these inconsistencies sodium bicarbonate injection was not immediately charged as cause of the discoloration. The cause of plume discoloration was instead initially attributed to other factors besides sodium bicarbonate injection such as atmospheric humidity or temperature. It is pertinent to note that plume discoloration was reported in the literature to occur for other sodium related flue gas desulfurization processes besides dry sodium bicarbonate injection. Felsvang et al. (1983)[39], for example, reported plume discoloration at Riverside Station during lime spray dryer, Niro SOx/NOx ,trials. Importantly, these trials employed sodium-based additives. Markussen et al. (1986)[40] at Pittsburgh Energy Technology Center reported plume discoloration during lime spray dryer laboratory trials when caustic, NaOH, additive was used. Similarly NO2 formation was found by Argonne National Laboratory when caustic additions were used in a lime spray dryer trials.[41] Also, in the Oxley (1984)[42] interviews, observations of plume discoloration were said to have occurred at Coyote Station during wet sodium carbonate spray dryer trials. The Coyote Station trials involved burning of high sodium lignite coal which may have also contributed to the plume discoloration. Plume discoloration was noted at Antelope Valley Station which burned high sodium lignite coal but did not use any sodium based additives in its lime spray dryer flue gas desulfurization system.[42]

36

2.1.3

The discovery that NO2 was the plume discoloring agent for sodium bicarbonate bag-house injection. The connection between plume discoloration and dry sodium bicarbonate bag-

house injection was definitively made during Phase II of the full scale demonstration tests at R.D. Nixon Station in 1986 when it was observed that plume discoloration terminated when sodium bicarbonate injection was terminated.[15] Although it was not initially known what specific compound caused the plume discoloration, Fuchs et al.(1990)[14] and Bland (1990)[15] hypothesized that NO2 formation was the agent responsible for the plume coloration based on a number of factors: 1) The reddish brown coloration matched the typical discoloration from NO2. A clue that NO2 was the responsible agent for plume discoloration was the distinguishing reddish brown plume color of the plume. This reddish brown color was the same color observed in detached downwind NO2 plumes discussed in the previous section (Section 2.1.2). This reddish brown color corresponded to the familiar color of smog where a build up of NO2 forms. The reddish brown color also corresponded to the color of the NO2 fumes evolved from laboratory beakers of fuming nitric acid. 2) Other possible candidates such as sulfates and fly ash were not present as a result of the desulfurization and particulate collection process in place. NO2 was hypothesized as the responsible agent for the discoloration, partially by process of elimination, since it was known that the sulfates were not forming since the sodium bicarbonate flue gas desulfurization process removed 70 to 90% of the flue gas SO2. Additionally, fly ash was eliminated as a possible

37

coloration candidate since the bag-house filter assured that fly ash was removed. The hypothesis that NO2 was the plume-coloring agent was substantiated by quantitative NOx measurements within multiple investigations [14,15,43,44] including those at the Arapaho Test Facility. NO2 concentration prior to dry sodium bicarbonate injection was compared with the NO2 concentration after injection and collection and found to form by the sodium bicarbonate injection. However the exact chemistry and reaction path responsible for the NO2 formation was not defined. A complex interaction of the flue gas NO with the sodium bicarbonate injection process was thought to be responsible for the NO2 formation. NO2 formation was thought to be due to an associated reaction. It seemed plausible that sodium bicarbonate or a related by-product from the sodium bicarbonate desulfurization process was related to the NO oxidation to NO2. 2.2 Literature review of sodium bicarbonate reaction chemistry.

As a foundation for understanding the reaction chemistry of the reaction of sodium bicarbonate and flue gas with SO2 and NO, the following is a review of the literature on: 1) Sodium bicarbonate decomposition. 2) Sodium bicarbonate reaction with flue gas with SO2 but without NOx including the studies by Mocek and Wald (1988)[4], Howatson et al. (1980)[8], and Keener (1982)[9]. 3) Sodium bicarbonate reaction with NOx without SO2 including the study by Knight (1980)[18].

38

4) Sodium bicarbonate reaction with flue gas with SO2 and NOx including the studies of Bland (1990)[16], Verlaeten et al. (1991)[20] and Bortz (1994)[45]. 5) Hypothesis for NO2 formation and NOx removal reactions suggested by different investigators including Miller (1986)[19], Bland (1990) [16], Verlaeten et al. (1991)[20] and Bortz (1994)[45]. Illustrative schematic diagrams of the hypothesized reaction chemistry are provided wherever possible to assist in pathway visualization. 6) Experimental validation findings of hypothesized reactions of: a) Sodium sulfite as conducted by Miller (1986)[19], Bland (1990) [16], Verlaeten et al. (1991)[20], and Lai and Yesavage (1994)[21]. b) Sodium pyrosulfite as conducted by Bland (1990) [16], Verlaeten et al. (1991)[20], and Lai and Yesavage (1994)[21]. c) Sodium sulfate as conducted by Bland (1990) [16]. d) Sodium nitrite and sodium nitrate conducted by Kozlowski et al (1970)[22] and Bland (1990) [16]. e) Gas phase interactions of SO2 and NOx. 2.2.1 Literature review of sodium bicarbonate decomposition. According to Barrall and Rogers (1966)[46], the decomposition reaction of sodium bicarbonate was first discovered by Lescoeur in the late 1800s. The following reaction equation was established, 2NaHCO3 Na2CO3 + H2O + CO2 (2.3)

Because water vapor is a by-product in the reaction, the reaction is often referred to as a dehydration reaction rather than a decomposition reaction. For the remainder of 39

discussion in this work, equation 2.3 will be referred to as a decomposition reaction. Besides Lescoeur, many other researchers including Barrall (1966)[46], Subramanian et al. (1972)[47], Templeton (1978)[48], Keener et al. (1985)[49] and Hu et al. (1986)[50] studied this decomposition reaction has been studied by The schematic diagram of the decomposition reaction, equation 2.3, is shown in Figure 2.2 as a visual aid for reaction pathway development.

H2 O H2 O NaHCO3
T>67C

Na2CO3

Figure 2-2 Schematic of the sodium bicarbonate decomposition reaction. The past studies of this reaction revealed that the onset of fast sodium bicarbonate decomposition ranged from 67C to 125C depending upon the partial pressure of CO2 and H2O. These studies also revealed that decomposition dramatically changed the material microstructure. Theoretically, a mole of sodium bicarbonate with molar volume of 38.9 cm3, is decomposed to a half a mole of sodium carbonate with molar volume of 20.7 cm3. A porous popcorn-like microstructure with high surface area forms from the sodium bicarbonate decomposition because of the molar volume decrease. Figure 2.3 shows the chemical structure of sodium carbonate.

40

O O C

Na

Na

Figure 2-3 Chemical structure of sodium carbonate. Surface area measurements have shown an increase in surface area from less than 1 m2/g before decomposition to 7-10 m2/gram after decomposition.[51] 2.2.2 Literature review of sodium bicarbonate reaction with SO2 flue gas without NOx. This section reviews the early sodium bicarbonate reaction studies with SO2 flue gas without NOx. These studies were primarily conducted to establish the reactivity of sodium bicarbonate with SO2 for flue gas desulfurization processes. At the time NO was not included in the flue gas for these trials because at that time NO was not known to have any role in the reaction chemistry of SO2 removal. These studies were conducted prior to the discovery of plume discoloration and the realization that NO had a role in the process. The findings of Mocek and Wald (1988)[4], Howatson et al.(1980)[8], and Keener (1982)[9] are reviewed. 2.2.2.1 Review of Erdos investigations of sodium bicarbonate reaction with SO2 flue gas without NOx. According to Mocek and Wald (1988)[26], as early as 1964 Erdos at the Czechoslovakia Academy of Sciences was the first to establish that sodium carbonate was a reactive alkali with acidic gas concentrations of SO2: 41

Na2CO3 + SO2 + O2 Na2SO4 + CO2 Figure 2.4 shows a schematic diagram of the reaction pathway.

(2.4)

CO2 Na2CO3
SO2,O2

Na2SO4

Figure 2-4 Schematic of the reaction pathway for of sodium carbonate with SO2. Erdos discovered that SO2 was more reactive with sodium bicarbonate compared to sodium carbonate. Erdos attributed the improved reactivity to activation from the decomposition of sodium bicarbonate. Erdos conceptualized that the reaction between sodium bicarbonate and SO2 occurred in two steps: 1) Erdos postulated that decomposition first activated the freshly formed sodium carbonate (activation is indicated by the ! symbol), 2NaHCO3 Na2CO3 + H2O + CO2 2) Erdos postulated that SO2 then reacted with activated sodium carbonate to form sodium sulfate, Na2CO3 + SO2 + O2 Na2SO4 + CO2 Note that the sum of these two steps, equations 2.5 and 2.6, makes up Erdos overall reaction equation, 2NaHCO3 + SO2 + O2 Na2SO4 + H2O + 2CO2 42 (2.7) (2.6) (2.5)

As a visual aid, Figure 2.5 shows a schematic diagram of Erdos two-step reaction pathway.

CO2 CO2 SO2,O2 NaHCO3

H2O

T>67 C

Na2CO3!

Na2SO4

Figure 2-5 Schematic of Erdos postulated reaction pathway of sodium bicarbonate with flue gas with SO2. Other researchers including, Knight (1977)[18], Howatson et al. (1977)[6], Stern (1978)[31], Carson (1980)[7] accepted and used Erdos conceptual model for theoretical development of flue gas desulfurization reactions. 2.2.2.2 Review of Howatson et al. (1980)[8] investigations of sodium bicarbonate reaction with SO2 flue gas without NOx. Howatson et al. (1980)[8] conducted experiments measuring the composition of reaction products from the sodium bicarbonate reaction with SO2 flue gas without NO (the mixture contained 14%CO2, 5%O2, 4% H2O, 1000ppm SO2) from 90C to 300C. In particular Howatson et al.(1980)[8] found that the reaction product formed depended on the reaction temperature: 1) Below 150C, sodium sulfite was found to be the main product formed with very little sodium sulfate product formation. 2) Above 250C, the reverse was found; sodium sulfate was found to be the primary product formed with very little sodium sulfite product. 43

3) Between 150 and 250C, the conversion product was found to depend on temperature, where sodium sulfite was formed at the lower end of the temperature scale and sodium sulfate at the higher end of the temperature scale. Howatson et al.(1980)[8] did not postulate the chemical reactions involved in the reaction pathway. The product formations were graphically presented and are reproduced in Figure 2.6.

Weight Percent Conversion in Product Vs. Temperature for SO2 flue gas
90 Weight percent conversion (%) 80 70 60 50 40 30 20 10 0 75 125 175 225 275 325 sodium sulfate sodium sulfite

Tem perature (C)

Figure 2-6 Sulfation products formed for the reaction of sodium bicarbonate with SO2 without NOx. Weight percent sodium sulfite, Na2SO3, and weight percent sodium sulfate, Na2SO4, found in the product as a function of temperature. The data markers with solid dots are sodium sulfate formation, and the solid triangle data markers represent the sodium sulfite formation. Reproduced with permission, Howatson et al. (1980)[8] 44

Howatson et al. (1980)[8] presented total conversion results for this reaction, by combining the results of conversion to sodium sulfite with those of conversion to sodium sulfate, as shown in Figure 2.7.
Total Weight Percent Conversion versus Temperature
100 Total wt. percent conversion (%) 90 80 70 60 50 40 30 20 10 0 75 125 175 225 275 325 Total sulfation

Tem perature (C)

Figure 2-7 Total weight percent of sodium sulfite versus temperature. Na2SO3, combined with sodium sulfate, Na2SO4, in product as a function of temperature. Reaction was for nahcolite with sulfur dioxide in flue gas without NOx. Reproduced with permission from Howatson et al.(1980)[8] Note that these results show high total conversion primarily in two temperature ranges, from 100C to 175C, and above 250C. And these results show poor total conversion from 175C to 250C. As a result, Howatson et al. (1980)[8] recommended that the baghouse be operated at temperatures either less than 150C or greater than 300C (572F) for maximum efficiency and suggested avoiding operation at 175C to 250C. Bland (1990) [16] later experimentally repeated Howatson et al.s (1980)[8] reaction experiments from 120C to 550C and confirmed the trends that Howatson et al. (1980)[8]

45

had previously reported in Figures 2.6 and 2.7 showing that sodium sulfite and sodium sulfate formation were temperature dependent. 2.2.2.3 Review of Keener (1982)[9] investigations of sodium bicarbonate reaction with flue gas with SO2 but without NOx. Keener (1982)[9] demystified Erdos sodium carbonate activation model [4] providing evidence that increased reactivity of decomposed sodium bicarbonate correlated to increased surface area produced by decomposition of sodium bicarbonate. As a result Keener and Khang (1993)[52] postulated the sulfation chemical reaction from equation 2.6 where Erdos speculated that the sodium carbonate was activated to nonactivated reaction: Na2CO3 + SO2 + O2 Na2SO4 + CO2 (2.8)

According to Keener and Khang (1993)[52] the sodium sulfate reaction was a two-step process involving the formation of sodium sulfite reaction intermediate that subsequently oxidized to sodium sulfate: Na2CO3 + SO2 Na2SO3 + CO2 Na2SO3 + O2 Na2SO4 (2.9) (2.10)

Keener and Khang (1993)[52] also experimentally reported that SO2 flue gas without NO reacted directly with non-decomposed sodium bicarbonate (at T<67C (153F)) according to the following equation: 2NaHCO3 +SO2 Na2SO3 + 2CO2 + H2O (2.11)

Figure 2.8 shows a schematic diagram of Keener and Khang (1982,1993)[9,52] overall reaction pathway.

46

CO2 H2 O NaHCO3 Na2CO3 SO2 CO2 Na2SO3 O2 Na2SO4

H2 O CO2 SO2, O2 Na2SO3

Figure 2-8 Schematic of Keener and Khang.s [9,52] reaction pathway for SO2 removal by (1982,1993) sodium bicarbonate. Includes discovery of low temperature SO2 reaction. Keener and Khangs (1982,1993)[9,52] reaction pathway shown in Figure 2.8 coincides with the findings of Howatson et al. (1980)[8] described in Section 2.2.2.2. 2.2.3 Literature review of sodium bicarbonate and sodium carbonate reaction with NOx without SO2. In addition to observing and studying the reactivity of sodium bicarbonate with SO2 flue gas without NO, the reactivity of sodium bicarbonate with NOx without SO2 was also studied. Knight (1977)[18] experimentally investigated the reaction of sodium bicarbonate and sodium carbonate with NOx in the laboratory. Knight (1977)[18] found that neither sodium bicarbonate nor sodium carbonate reacted with NO, with or without the presence of O2. Knight (1977)[18] however revealed that sodium carbonate was

47

reactive with NO2 in the presence of oxygen and postulated the following reaction chemistry, Na2CO3 + NO2 2NaNO3 + NO + CO2 Although oxygen was not part of Knights (1977)[18] proposed balanced reaction equation, Knight (1977)[18] noted that it was necessary for the reaction to occur and that without oxygen very little NO2 was found to react with sodium carbonate. As a visual aid Figure 2.9 shows a schematic diagram of Knights (1977)[18] reaction pathway. (2.12)

CO2 H2O NaHCO3 Na2CO3 NO2,O2

NO CO2 NaNO3

Figure 2-9 Schematic of Knights (1977)[18] observed reaction of sodium carbonate with NO2. Note that Knights (1977)[18] postulated NOx reaction was independent of SO2. Knights (1977)[18] reactivity reaction of sodium carbonate with NO2 was later repeated and verified by Bland (1990) [16]. Blands (1990) [16] results corroborated with Knights (1977)[18] results where a significant amount of sodium nitrate was found to form by the reaction of sodium carbonate with NO2. Bland (1990) [16] verified Knights (1977)[18] postulated reaction stoichiometry: 3NO2 + Na2CO3 2NaNO3 + NO + CO2 Bland (1990) [16] found as Knight (1977)[18] had reported that this sodium nitrate formation increased with surface area and decreased with increased temperature. Bland (1990) [16] however made no mention of the necessity of oxygen in the reaction gas as 48 (2.13)

Knight (1977)[18] had previously noted. In fact Blands (1990) [16] reactions were run without oxygen. 2.2.4 Literature review of sodium bicarbonate reaction with flue gas with SO2 and NOx. The early laboratory investigations reviewed in Section 2.2.2 focused upon the reaction pathways for SO2 removal without considering the presence of NO and any possible interaction that NO might have on the overall reaction. NO was not included in the gas mixture for the past experimental SO2 reaction testing because in the past it was not thought to participate in the reaction chemistry. Since plume discoloration observations in 1986 was it realized that NO had a role in the overall pathway and that it interacted in the process. After plume discoloration observations in 1986 studies were needed to understand how SO2 and NOx interacted with each other during sodium bicarbonate flue gas desulfurization reactions. Investigators sought answers to the following key questions regarding the role of NO in the reaction process: 1) Did the presence of NO effect sodium bicarbonate decomposition? 2) Did the presence of NO effect the reaction of sodium bicarbonate with SO2? 3) How was NO2 formed from the reaction of sodium bicarbonate with flue gas with SO2 and NO? 4) How was NOx removed during reaction of sodium bicarbonate with flue gas with SO2 and NO? To answer the above questions, additional laboratory studies were needed of sodium bicarbonate reactions with SO2 flue gas that contained NO. The SO2 reaction pathways 49

previously reported for reactions of sodium bicarbonate with SO2 flue gas without NO (as reviewed in Section 2.2.2) were not assumed to be the reaction pathways for the reaction of sodium bicarbonate with flue gas with both SO2 and NO. This section reviews past investigations of sodium bicarbonate with SO2 flue gas with NO. Commentary and implications of each of the past studies is held for later discussion in the theoretical development section of Chapter 3. 2.2.4.1 Review of Blands (1990) [16] investigations of the reaction of sodium bicarbonate with SO2 flue gas with NO. Bland (1990) conducted the largest most comprehensive investigation into the reaction of sodium bicarbonate with SO2 flue gas with NO.[15,16] Bland (1990) [16] studied the reaction pathway by which NO2 was formed so that a mitigation method could be identified and the technology could be fully implemented. Bland (1990) [16] simulated in the laboratory the sodium bicarbonate flue gas desulfurization process that produced the plume discoloration in the field demonstration test. Using simulated flue gas mixtures with both SO2 and NO, Bland (1990) [16] showed: 1) NO2 was indeed formed in his bench scale fixed bed reactor when sodium bicarbonate was reacted with SO2 flue gas with NO. 2) NO2 formation was linked with active SO2 removal. NO2 formation was found to abruptly stop with the SO2 concentration in the simulated flue gas mixture was abruptly turned off. 3) The level of NO2 formed by the reaction of sodium bicarbonate depended on the inlet SO2 / NO ratio in the flue gas mixture. Holding NO constant, Bland (1990)

50

[16]

observed that NO2 formation increased with increasing SO2 concentration in

the mixture. 4) NO2 formation terminated when the oxygen in the flue gas mixture was terminated even though SO2 removal (sodium sulfite formation) continued. 5) Decomposition was inhibited from 121C to 162C. 6) At all temperatures only sodium sulfate product was formed, no sodium sulfite product was found to form by the reaction. Figure 2.10 shows Blands results for conversion to sodium sulfite and sodium sulfate.
Weight Percent Conversion in Product vs. Temperature for SO2 flue gas with NO
45 Weight percent conversion (%) 40 35 30 25 20 15 10 5 0 75 125 175 225 275 325 sodium sulfate sodium sulfite

Tem perature (C)

Figure 2-10 Weight percent sodium sulfite, Na2SO3, and weight percent sodium sulfate, Na2SO4, in product as a function of temperature. Reaction was for sodium bicarbonate with SO2 (540 ppm) in flue gas with NO (270ppm). The data markers with solid dots are sodium sulfate formation, and the triangle data markers represent the sodium sulfite formation. Data obtained from Bland (1990)[16]. Compiling total conversion, from the sum of the sodium sulfite and sodium sulfate conversion found by Bland (1990)[16] for the reaction of sodium bicarbonate with SO2 51

flue gas with NO in Figure 2.10, reveals that total conversion is equal to the sodium sulfate conversion. Note that there is no sodium sulfite formed. This total conversion is lowest at low temperature and increases linearly with increasing temperature. Conversion was high at high temperature. As a result bag-house operation for maximum efficiency is at high temperature. These results are very different from the results reported for SO2 flue gas without NO. 7) NO2 formation by the reaction was found to be temperature dependent. NO2 formation was found to peak at approximately 175C, and decrease above 175C as shown in Figure 2.11.

200

NO2 Formation (ppm)

150

NO2 Formation

100

50

0 0 50 100 150 200 250 300 350 400

Temperature (C)

Figure 2-11 NO2 formation as a function of reactor temperature for sodium bicarbonate. SO2/NO 2 and inlet SO2 concentration =540 ppm. Bland (1990)[16] 8) NOx removal as mentioned in Section 2.2.4.1 was also found to occur for a short finite period beginning at the start of the reaction. Sodium nitrate was found to be the product formed during this period. Because Bland(1990)[16] found that NO2 was not 52

formed during this short finite NOx removal period, it was referred to as the NO2 lag period. Bland (1990)[16] found that SO2 removal proceeded during this NO2 lag period. The amount of NOx removal or sodium nitrate formed during the lag period was found to be temperature dependent. NOx removal was found to increase up to a maximum at 150C and decrease sharply above 150C with increasing temperature with no NOx removal above 275C. In addition to testing the reaction of sodium bicarbonate with flue gas with SO2 and NO, Bland (1990)[16] tested the reaction of sodium carbonate with flue gas with SO2 and NO and found similar results: 1) NO2 was formed as SO2 was removed. 2) Only sodium sulfate product accumulated, no sodium sulfite product was found to form. However Bland (1990)[16] found that NOx was not removed by the reaction. Bland (1990)[16] did not find any sodium nitrate product to be formed by the reaction of sodium carbonate with flue gas with SO2 and NO. For fast and easy reference in later discussion, Blands (1990)[16] most important findings have been compiled into Table 2.1.

53

Table 2-1 Summary of findings from Blands (1990)[16] experimental studies 1) Sodium bicarbonate decomposition was found to be inhibited from 121C to 162C when NO was present in the flue gas. 2) Sodium bicarbonate was found to react with SO2 flue gas with NO. This reaction was found to produce sodium sulfate and sodium nitrate and NO2. No sodium sulfite was found to form at all reaction temperatures. 3) NOx removal, in the form of sodium nitrate, was found to occur for a short finite time period immediately after sodium bicarbonate injection. The amount of sodium nitrate formed was found to depend on temperature. Maximum nitrate formation was found to occur at low temperature above the decomposition temperature and decreased with increasing temperature. NO2 formation was not found to occur during the NOx removal period. 4) Sodium carbonate was found to react with SO2 flue gas with NO to form sodium sulfate and NO2. Sodium nitrate was not found to form by the reaction with sodium carbonate. 5) NO2 formation was found to form only when sodium bicarbonate or sodium carbonate were actively removing SO2. NO2 formation was found to terminate when sodium bicarbonate or sodium carbonate injection was stopped or when the SO2 flow was stopped. In addition the amount of NO2 formed was found to depend upon the SO2 concentration, with greater SO2 concentration, greater NO2 was found to form. 6) NO2 formation was found to be temperature dependent. NO2 formation was found to be very small at low temperature, T<150C ; peak NO2 formation was found to be at 175C; and NO2 formation at high temperatures above 300C were found to be very small. 7) NO2 formation was found to increase when the ratio of inlet SO2 to NO concentration was increased. 8) Without oxygen in the gas mixture, NO2, sodium nitrate and sodium sulfate were not found to form instead only sodium sulfite was found to form. 9) Compared to the reaction of sodium bicarbonate and SO2 flue gas without NO, a significantly smaller amount of total sulfation was found for the reaction of SO2 flue gas with NO. 2.2.4.2 Review of Verlaeten et al.s (1991)[20] investigations of the reaction of sodium bicarbonate with flue gas with SO2 and NO. As in the US, European demonstration tests of sodium bicarbonate flue gas desulfurization were conducted. According to Verlaeten et al. (1991)[20] both 90% SO2 removal and 60% NOx removal were demonstrated at the NATO Air Base in Geilenkirchen in 1990 for an inlet SO2/NOx ratio of 5 and a 1.5 normalized stoichiometric ratio (NSR) for sodium bicarbonate injection. No plume discoloration was 54

reported at this test site. Through fluidized bed studies of the reaction of sodium bicarbonate with flue gas with SO2 and NO, Verlaeten et al.s (1991)[20], at the Institute Francais du Petrole, focused on ways to improve NOx removal yield. Similar to Bland (1990)[16], Verlaeten et al. (1991) found: 1) A NOx removal window existed from 127C to 227C. NOx removal increased from 127C to 160C and then decreased above 160C. 2) NOx removal increased as the SO2/NO ratio increased. For the reaction of sodium carbonate with SO2 flue gas with NO, Verlaeten et al. (1991)[20] importantly found, as Bland (1990)[16] had also reported that NOx was not removed by sodium carbonate at any temperature. These findings served as the basis for Verlaeten et al.s (1991)[20] hypothesis on the reaction pathway for NOx removal described in Section 2.2.5.3. 2.2.4.3 Review of Bortzs (1994)[45] investigations of the reaction of sodium bicarbonate with flue gas with SO2 and NO. Using a pilot plant, Bortz (1994)[45] experimentally investigated the sodium bicarbonate reaction with SO2 flue gas with NO. Bortz (1994)[45] made the following findings: 1) NOx removal is significantly greater using an SO2/NO flue gas mixture than an SO2/NO2 flue gas mixture. 2) As Bland (1990)[16] had previously reported, NO2 formation lagged behind in the sodium bicarbonate reaction relative to injection and SO2 removal. NO2 formation lagged for several minutes before gradually reaching steady state. SO2 and NOx removal did not lag, both SO2 and NOx removal began immediately after 55

injection. Bortz (1994)[45] showed that temperature and humidity both affected the NO2 lag period length. He found that the lag period was shortened as temperature was increased and lengthened as humidity was increased. Zero NO2 lag was found at 177C with 0% humidity. At 121C with 9% humidity the NO2 lag was observed to be 6 minutes long. At 121C with 15% humidity the NO2 lag was 9 minutes long. At 93C with 15% humidity the lag was observed to be over 35 minutes long. 3) Flue gas humidity was also found to effect NO2 formation especially at low temperatures. Bortz (1994)[45] found that twice as much NO2 was formed at 9% humidity (100 ppm) compared to at 15% humidity (50 ppm), (inlet NO= 500ppm, and SO2/NO=3.6 and T= 121C. 4) Bortz (1994)[45] evaluated the equilibrium reaction chemistry for the system. In particular the McBride and Gordons 1972 NASA Lewis equilibrium computer program was employed to predict reaction products based on thermodynamic equilibrium equations. The computer program had the capability of predicting the reaction products based on minimum Gibbs free energy calculations. Equilibrium predictions were made for the reactions of sodium bicarbonate with SO2 flue gas NO. All of the pertinent reactants and possible products along with their respective thermodynamic constants were input into the program. Bortz (1994)[45] found from the NASA equilibrium program that the reaction of sodium bicarbonate with SO2 flue gas with NO would according to equilibrium convert flue gas NO to N2. Because the equilibrium calculations did not corroborate with

56

the actual observed products, NO2 and NaNO3, it was concluded that the actual reactions were non-equilibrium in nature. 2.2.5 Past Hypotheses for NO2 formation and NOx removal reactions. A number of hypotheses were formulated by past investigators in an effort to understand how NO2 was formed and how NOx was removed by the reaction of sodium bicarbonate with SO2 and NO. A description of Miller (1986)[19], Bland (1990)[16] and Bortzs (1994)[45] reaction pathway hypothesis for NO2 formation are described in this section. Also a description of Millers (1986)[19], Blands (1990)[16], Verlaeten et al.s (1991)[20] and Bortzs (1994)[45] reaction pathway hypothesis for NOx removal are described below. Schematic diagrams are included to aid in visualization. 2.2.5.1 Review of Millers (1986)[19] hypothesis for NO2 formation. Miller (1986)[19] at the University of North Dakota Energy Research Center (UNDERC) aimed to study the cause of the plume discoloration observed at Coyote Station for a sodium carbonate spray dryer flue gas desulfurization application. Although no clear reaction pathway for NO2 formation was overtly evident, Miller (1986)[19] formulated a reaction pathway hypothesis based on the aqueous findings of Ackermann and Powell (1967).[53] According to Miller (1986)[19], Ackermann and Powell (1967)[53], found aqueous sulfite to react with NO and form a N-nitrosohyroxyamine-N-sufonate, (NHAS) product, SO3-2(aq) + 2NO(gas) SO3(NO)2-2(aq) (2.14)

Under acidic conditions in the absence of oxygen it was found that NHAS reacted to form sulfate and nitrous oxide:

57

acidic conditions SO3(NO)2-2(aq) SO4-2(aq) + N2O(gas) (2.15)

The NHAS was also shown to form a stable solid salt with potassium, K2SO3(NO)2. Other researchers including Martin et al. (1981)[54], Ito et al. (1984)[55], Takeuchi et al. (1977)[56], Weisweiler and Blumhofer (1984)[57], Sada et al. (1988)[58], and Littlejohn et al. (1986)[59] studied the aqueous reactions for the formation of NHAS. Stepwise, for the spray dryer application of interest, Miller (1986)[19] hypothesized the following steps: 1) Sodium carbonate or sodium hydroxide was reacted with SO2 to produce sodium sulfite, Na2CO3 + SO2 Na2SO3 + CO2 or 2NaOH + SO2 Na2SO3 + H2O (2.17) (2.16)

Equation 2.16, was consistent with the low temperature reaction products reported by Howatson et als (1980)[8]. Equation 2.17, was previously suggested by Takeuchi and Yamanaka (1978).[60] when sodium sulfite was found to form by the rapid reaction between SO2 and the hydroxyl ion. 2) Sodium sulfite, formed in step 1, was reacted with NO and acted as a catalytic site for gathering NO molecules. This reaction formed the intermediate compound, sodium-nitrogen-sulfonate, Na2SO3(NO)2: Na2SO3 + 2NO Na2SO3(NO)2 3) The intermediate compound formed in step 2, when reacted with oxygen produced NO2 : (2.18)

58

Na2SO3(NO)2 + O2 Na2SO3 + 2NO2

(2.19)

Sodium sulfite was recycled out as an end product by this reaction. Figure 2.12 shows a schematic diagram of all of the steps of Millers (1986)[19] hypothesized reaction pathway for sodium carbonate and caustic sodium hydroxide reaction with flue gas with SO2 and NO. a. CO2
SO2 82C<T<125C NO O2

NO2 Na2SO3(NO)2 Na2SO3

Na2CO3

Na2SO3

b.

CO2
SO2 82C<T<125C NO O2

NO2 Na2SO3(NO)2 Na2SO3

NaOH

Na2SO3

Figure 2-12 Schematic of Millers (1986)[19] Hypothesized Reaction Pathway (a) for sodium carbonate with flue gas with SO2 and NO, and (b) for sodium hydroxide with flue gas with SO2 and NO. To prove that Millers (1986)[19] hypothesized reactions were the actual reactions responsible for NO2 formation during flue gas desulfurization, Millers (1986)[19] reaction steps 2 and 3 (equations 2.18 and 2.19) needed to be validated experimentally. Experimental validation was needed to prove that the reaction of sodium sulfite with NO produced sodium-nitrogen-sulfonate, Na2SO3(NO)2. Also, experimental validation was 59

needed to show that NO2 was released by the reaction of sodium nitrogen sulfonate with oxygen. Section 2.2.6.1.1 presents Millers (1986)[19] unsuccessful validation attempts of these reactions. 2.2.5.2 Review of Blands (1990)[16] hypothesis for NO2 formation. In Blands (1990)[16] work two different reaction path hypotheses were discussed. First, Bland (1990)[16] hypothesized that NO2 was formed by heterogeneous catalysis of flue gas NO by a sodium sulfur-oxide reaction intermediate. The sodium sulfur-oxide reaction intermediate was hypothesized to form through the reaction of sodium bicarbonate with SO2. This hypothesis incorporated Keener and Khangs (1993)[52] and Howatson et al.s (1980)[8] sulfation reactions described by Figures 2.7 and 2.8. Bland (1990)[16] however, hypothesized that a sodium sulfur-oxide reaction intermediate was formed within these sodium sulfite and sodium sulfate formation reactions. Accordingly, this hypothesized reaction pathway proceeded with the following steps: 1) A sodium sulfur-oxide intermediate compound was formed during the reaction of sodium bicarbonate with SO2. This sodium sulfur-oxide compound was formed as an intermediate step prior to the formation of the sulfation products sodium sulfite or sodium sulfate (depending on temperature): a) Below 120C (248F), the sodium-sulfur oxide intermediate compound (Na/SOx intermediate) was formed by reaction of sodium bicarbonate with SO2 without decomposition of the sodium bicarbonate, 2NaHCO3 + SO2 Na/SOx intermediate I + CO2 + H2O (2.20)

The sodium-sulfur oxide intermediate compound was eventually converted to a final sodium sulfite product, 60

Na/SOx intermediate Na2SO3

(2.21)

b) Above 120C, because decomposition of sodium bicarbonate is relatively fast, decomposition occurred simultaneous to sulfation of sodium bicarbonate, 2NaHCO3 Na2CO3 + CO2 + H2O (2.22)

For this temperature range, the sodium-sulfur oxide intermediate compound was formed by the reaction of decomposed sodium bicarbonate (sodium carbonate) with SO2, Na2CO3 + SO2 Na/SOx intermediate + CO2 Bland (1990)[16] suggested a number of sodium sulfur-oxide intermediate compounds that may have formed including: sodium sulfite, sodium pyrosulfite, Na2S2O5, and sodium bisulfite, NaHSO3 2) The sodium sulfur-oxide intermediate formed in step 1 heterogeneously catalyzed the oxidation of NO to NO2, Sodium-sulfur oxide intermediate acts as a catalyst (2.23)

NO + O2 NO2

(2.24)

3) The sodium sulfur-oxide intermediate compound eventually converted to sodium sulfite or sodium sulfate (depending on temperature): Na/SOx intermediate Na2SO3 at low temperature (2.25)

or to sodium sulfate (depending on temperature), Na2SO3 + O2 Na2SO4 at high temperature (2.26)

61

Note that the final end products in equations 2.25 and 2.26 are consistent with Howatson et al.s (1980)[8] sulfation product findings for flue gas without NO (Section 2.2.2.2) 4) The NO2, formed by step 2, was reacted with sodium carbonate, according to Knights (1977)[18] previously described NOx removal reaction, Na2CO3 + NO2 2NaNO3 + NO + CO2 (2.27)

A schematic diagram of this multi-step hypothesized reaction pathway is shown in Figure 2.13.

H2O NaHCO3 Na2CO3

SO2

CO2 Na/SOx Intermediate I Na2SO3

O2

Na2SO4

acts as catalyst

NO, O2

NO2

NO CO2
NO2

NaNO3

SO2

Na/SOx Intermediate I

Na2SO3

acts as catalyst

NO, O2

NO2

Figure 2-13 Schematic of Blands(1990)[16] first hypothesized reaction pathway for NO2 formation and NOx removal.

62

To prove that Blands (1990)[16] 1st hypothesized reaction pathway shown in Figure 2.12 was the reaction pathway actually responsible for NO2 formation during flue gas desulfurization, experimental validation of reaction equation 2.24 was needed. In particular, it was necessary to determine a reaction intermediate and show that NO2 was produced when it was reacted with NO. All known or possible products of the reaction of sodium bicarbonate and flue gas with SO2 and NO needed to be screened for this reaction. Section 2.2.5.2 details that Bland (1990)[16] was unable to experimentally show any catalysis producing NO2 formation by any reaction intermediates tested including: sodium sulfite, sodium pyrosulfite, sodium sulfate, sodium nitrite, or sodium nitrate. Because Bland (1990)[16] was unable to validate the heterogeneous catalysis hypothesis original hypothesis Bland (1990)[16] formulated an alternate reaction path hypothesis to explain how NO2 was formed. Instead of hypothesizing that NO2 was heterogeneously catalyzed by a sodium sulfur-oxide reaction intermediate, Bland (1990)[16] alternately hypothesized that NO2 was formed by gas phase catalysis with the following reaction steps: 1) Some SO2 reacts with sodium bicarbonate or sodium carbonate to form sodium sulfite and sodium sulfate as described by Howatson et al.s (1980)[8] product findings (Section 2.3.2.1) and Keener and Khangs (1993)[52] reaction equations 2.9 through 2.11 (Section 2.3.2.2). 2) Some SO2 is catalyzed to SO3 due to the presence of NO, NO catalyst SO2 + O2 SO3 (2.28)

63

3) Simultaneously, some flue gas NO is catalyzed to NO2. Bland (1990)[16] did not offer any detailed explanation however on how the NO oxidizes to NO2. One can surmise, possibly, that Bland (1990)[16] believed that some free radicals of oxygen were generated from the catalysis of SO2 to SO3, NO catalyst SO2 + O2 SO3 + O and that these free radicals oxidize NO to NO2, NO + O NO2 (2.30) (2.29)

4) Any SO3 formed from step 2 reacts with available fresh sodium carbonate to form sodium sulfate, SO3 + Na2CO3 Na2SO4 + CO2 5) Some of the NO2 that forms by step 3 reacts with sodium carbonate to form sodium nitrate according to by Knights(1977)[18] postulated reaction path, equation 2.27. Bland (1990)[16] clarified, however that this NO2 reaction occurs only during decomposition of sodium bicarbonate when NO2 is able to diffuse to fresh sodium carbonate beneath the sodium sulfate ash layer, since this ash layer is mechanically broken up during decomposition. Bland (1990)[16] viewed sodium carbonate as the substrate where these gas-phaseconstituents congregated for reactions together. Figure 2.14 shows a schematic diagram of Blands (1990)[16] alternate reaction path hypothesis. (2.31)

64

catalyst

SO3 O NO NO2 CO2 NO NO2 NaNO3

O2 SO2

NO

H2 O NaHCO3 Na2CO3 SO2 Na2SO3 O2 Na2SO4

SO3

Na2SO4

SO3

Na2SO4

SO2

Na2SO3

Figure 2-14 Schematic of Blands (1990)[16] final theory on the reaction pathway for SO2 and NO removal and NO2 formation. To prove that Blands (1990)[16] alternate reaction pathway hypothesis shown in Figure 2.14 was the reaction pathway truly responsible for NO2 formation and NOx removal during flue gas desulfurization, equations 2.29 and 2.31 needed to be experimentally validated. In particular, it was necessary to demonstrate that mixing NO with SO2 resulted in the formation of SO3, free radicals of oxygen, and NO2. Section 2.2.6.5 discusses Blands (1990)[16] experimental attempts to validate these gas phase reactions. 2.2.5.3 Review of Verlaeten et al.s (1991)[20] hypothesis for NOx removal. There was interest in Europe in developing dry sodium bicarbonate injection process for SO2 removal as a low cost alternative to wet scrubbing in the early 1990s. 65

There was special interest in this process because in addition to SO2 removal this process was found to remove NOx and perhaps serve as an attractive alternative to the costly NOx removal process, selective catalytic reduction (SCR). As discussed in Section 2.3.4.2, Verlaeten et al. (1991)[20] discovered that NOx was removed by the reaction of sodium bicarbonate with flue gas with SO2 and NO but was not removed by the reaction of sodium carbonate with flue gas with SO2 and NO. This important finding served as the basis for Verlaeten et al.s (1991)[20] NOx removal hypothesis. Specifically Verlaeten et al. (1991)[20] hypothesized NOx removal stemmed from a reaction with sodium bicarbonate not sodium carbonate. The hypothesis included the following reaction steps: 1) Some sodium bicarbonate reacts with SO2 to form a sodium sulfur-oxide reaction intermediate, sodium bisulfite, NaHCO3 + SO2 NaHSO3 + CO2 (2.32)

Note that the sodium bisulfite reaction intermediate does not, however, form by the reaction of sodium carbonate with SO2. Instead, the reaction of sodium carbonate with SO2 is described in step 6. 2) The sodium bisulfite reaction intermediate formed in step 1 dehydrates to sodium pyrosulfite, 2NaHSO3 Na2S2O5 + H2O (2.33)

3) Some of the sodium pyrosulfite formed in step 2, reacts with NO and oxygen to form the sodium nitrite and sodium nitrate, Na2S2O5 + 2NO + O2 NaNO2 + NaNO3 + 2SO2 Note that SO2 revolves out as a product in this reaction. (2.34)

66

4) The remainder of sodium pyrosulfite that forms in step 2 decomposes to sodium sulfite, Na2S2O5 Na2SO3 + SO2 Note that SO2 also revolves out as a product from this reaction. 5) The remainder of sodium bicarbonate that does not react with SO2 according to step 1 decomposes to sodium carbonate, 2NaHCO3 Na2CO3 + CO2 + H2O 6) Some of the sodium carbonate formed in step 5 reacts with SO2 to form sodium sulfite and sodium sulfate depending on temperature, Na2CO3 + SO2 Na2SO3 + CO2 Na2SO3 + O2 Na2SO4 (2.37) (2.38) (2.36) (2.35)

Note that sodium pyrosulfite does not form by the reaction of sodium carbonate with SO2. By combining the two key elementary reactions, equations 2.33 and 2.34, from steps 2 and 3, the following overall reaction results: 2NaHSO3 + 2NO + O2 NaNO2 + NaNO3 + 2SO2 + H2O (2.39)

Note that in Verlaeten et al.s (1991)[20] hypothesis only sodium pyrosulfite reacts with NO to form nitration compounds of NOx removal. Sodium sulfite, on the other hand, does not react with NO to form any nitration compounds of NOx removal. Figure 2.15 shows the schematic diagram of Verlaeten et al.s (1991)[20] theory on NOx removal.

67

H2 O NaHCO3 SO2

CO2 O2 Na2SO4

Na2CO3

Na2SO3

CO2 SO2 NaHSO3

H2 O Na2S2O5

SO2 Na2SO3

SO2 NO, O2 NaNO2 NaNO3

Figure 2-15 Schematic of Verlaeten et al.s (1991)[20] hypothesis on SO2 and NOx removal. To prove that Verlaeten et al.s (1991)[20] reaction pathway hypothesis shown in Figure 2.15 was truly the reaction pathway responsible for NOx removal during flue gas desulfurization, reaction equations 2.32 and 2.34 needed to be experimentally validated. Also it was necessary to experimentally verify that sodium sulfite did not react with NO to form any nitration compounds. In particular, it was necessary to demonstrate that sodium bisulfite was a reaction intermediate that formed by the reaction of sodium bicarbonate and SO2. Also it was necessary to demonstrate that sodium pyrosulfite reacted with NO and oxygen to form the nitration reaction products sodium nitrite and sodium nitrate.

68

2.2.5.4

Review of Bortzs (1994)[45] hypothesis for NO2 formation.

The reaction path hypothesis for NO2 formation and NOx removal formulated by Bortz (1994)[45] differed intrinsically from Blands (1990)[16] hypotheses. Bortz (1994)[45] assumed a different NOx removal pathway based on the experimental findings that Bortzs (1994)[45] reported for the reaction of sodium bicarbonate and flue gas with SO2 and NO as described in Section 2.3.4.2. In particular Bortz (1994)[45] reported that significantly more NOx was removed by the reaction of sodium bicarbonate and flue gas with SO2 and NO than was removed by the reaction of sodium bicarbonate and flue gas with SO2 and NO2. Therefore, Bortzs (1994)[45] reaction pathway hypothesis included the following reaction steps: 1) Bortz (1994)[45] maintained that NOx removal did not first require the conversion of NO to NO2. Instead of assuming as Bland (1990)[16] had, that the primary NOx removal reaction was Knights (1977)[18] reaction of sodium carbonate with NO2 (equation 2.27), Na2CO3 + NO2 2NaNO3 + NO + CO2 (2.40)

Bortz (1994)[45] argued that the primary NOx removal reaction occurred either by direct reaction of NO with sodium bicarbonate: 4NaHCO3 + SO2 + 2O2 + 2NO Na2SO4 + 2NaNO3 + 4CO2+ 2H2O or by direct reaction of NO with sodium carbonate 2NaHCO3 Na2CO3 + CO2 + H2O 2Na2CO3 + SO2 + 2O2 + 2NO Na2SO4 + 2NaNO3 + 4CO2 (2.42) (2.43) (2.41)

depending on temperature. Note that Bortz (1994)[45] did not refute the validity of Knights (1977)[18] postulated NO2 reaction. Bortz (1994)[45] instead acknowledged 69

Knights (1977)[18] postulated NO2 reaction but maintained that it was not the primary reaction pathway for NOx removal. 2) The sodium nitrate formed in step 1 reacts with SO2 resulting in the release of NO2. 2NaNO3 + SO2 Na2SO4 + 2NO2 (2.44)

Bortz (1994)[45] maintained that the release of NO2 occurred after a period of sodium nitrate formation not simultaneously. 3) Sodium bicarbonate and sodium carbonate react with SO2 without oxygen and form sodium sulfite as the sulfation product. NaHCO3 + SO2 Na2SO3 + 2CO2 + H2O Na2CO3 + SO2 Na2SO3 + CO2 + H2O Figure 2.16 shows the schematic of Bortzs (1994)[45] hypothesis of SO2 and NOx removal and NO2 formation. (2.45) (2.46)

70

H2 O CO2 NaHCO3 Na2CO3 SO2, O2, NO Na2SO4 NaNO3 NO2 NaNO3 CO2 SO2 Na2SO3 SO2 NO2 Na2SO4

H2 O CO2 SO2, O2, NO Na2SO4 NaNO3 CO2 SO2 Na2SO3 H2 O SO2 NO2 Na2SO4

Figure 2-16 Schematic of schematic of Bortzs (1994)[45] hypothesis of SO2 and NOx removal and NO2 formation. To prove that Bortzs (1994)[45] reaction pathway hypothesis shown in Figure 2.16, was truly the reaction pathway responsible for NOx removal during flue gas desulfurization, experimental validation was needed to show that sodium nitrate was in fact formed by: 1) The reaction of sodium bicarbonate and flue gas with SO2 and NO (equation 2.41). 2) The reaction of sodium carbonate and flue gas with SO2 and NO (equation 2.43). 71

Validation of the sodium carbonate reaction with flue gas with SO2 and NO was of special interest since both Blands (1990) [16] and Verlaeten et al.s (1991)[20] experimental findings reported that sodium nitrate was formed only by the reaction of sodium bicarbonate (not sodium carbonate) and flue gas with SO2 and NO. Additionally experimental validation that NO2 was released by the reaction of sodium nitrate with SO2 was needed to prove Bortzs (1994)[45] hypothesis. Section 2.3.6.4 presents Blands (1990)[16] unsuccessful experimental validation attempts of this reaction. 2.2.6 Review of experimental validation findings for the hypothesized reactions. In attempts to prove the hypotheses described in Section 2.3.5 true, the respective investigators and others conducted validation reaction experiments of the unproven reactions branches in the pathways. In particular there was interest by various investigators to determine if either NOx was removed or NO2 was formed by the reaction of NO with sodium sulfite, sodium pyrosulfite, sodium sulfate, sodium nitrite and sodium nitrate 2.2.6.1 2.2.6.1.1 Review of sodium sulfite reaction findings. Millers (1986)[19] sodium sulfite reaction findings.

Millers (1986)[19] hypothesis on the reaction pathway of NO2 formation was described in Section 2.3.5.1, and shown in Figure 2.12. Recall that Miller (1986)[19] hypothesized the following two reactions, equations 2.47 and 2.48: Na2SO3 + 2NO Na2SO3(NO)2 Na2SO3(NO)2 + O2 Na2SO3 + 2NO2 The overall reaction equation of these two reactions is as follows: 72 (2.47) (2.48)

Na2SO3 + 2NO + O2 Na2SO3 + 2NO2

(2.49)

Miller (1986)[19] attempted to validate the above overall reaction in the laboratory. Miller (1986)[19] reacted sodium sulfite with flue gas with SO2 and NO at 152C and checked for NO2 formation. Miller (1986)[19] had anticipated that the sodium sulfite reaction would produce an NO2 peak comparable to the NO2 peak produced by the reaction of sodium carbonate with flue gas with SO2 and NO. The sodium carbonate reaction however produced a much larger more pronounced NO2 peak measuring 70 ppm and lasting over 25 minutes while the sodium sulfite reaction produced a small NO2 peak measuring 10 ppm for less than 3 minutes. Because the NO2 peak produced by the sodium sulfite reaction was very small, Miller (1986)[19] thought that the sodium sulfite reaction was not adequately validated. Miller (1986)[19] thought that additional reaction experiments were needed that incorporated special attention to high quality sodium sulfite preparations. Miller (1986)[19] suggested that the failure to validate the reaction was a result of poor experimental sodium sulfite quality. According to Miller (1986)[19], the reason that NO2 was not produced by the sodium sulfite reaction was that the sodium sulfite used in the experiments had poor surface quality. Miller (1986)[19] explained that it was common knowledge that sodium sulfite oxidized rapidly when exposed to moisture and air. Mellor (1947)[61] for example, published in his handbook that sodium sulfite is not changed in air at 100 C, but is rapidly oxidized at 15 C in moist air. As a result Miller (1986)[19] presumed the sodium sulfite validation reactions were unsuccessful experiments because the sodium sulfite used in the experiments was surface oxidized. The sodium sulfite used in the experiments was a technical grade sodium sulfite. It was likely that this powder

73

surface was oxidized since this grade was unavoidably exposed to air and moisture during laboratory and manufacturing storage and handling. Miller (1986)[19] recommended future re-testing of sodium sulfite using a special high purity grade of sodium sulfite that was not surface oxidized from moisture and air exposure. Miller (1986)[19] also recommended that future screening tests be made in particular on the reactivity of Na2SO3(NO)2 with oxygen for NO2 formation in an effort to validate reaction equation 2.19. 2.2.6.1.2 Blands (1990)[16] sodium sulfite reaction findings.

In an effort to test Blands (1990)[16] heterogeneous catalysis hypothesis Bland (1990)[16] conducted a series of packed bed screening experiments in search of identifying a sodium sulfur-oxide intermediate compound responsible for NO2 formation. Sodium sulfite was one of the candidate compounds that Bland (1990)[16] tested the reactivity with NO and O2 from 21C to 188C for NO2 formation. Bland (1990)[16] reported no NO2 formation or NOx removal for any of these tests. Technical grade sodium sulfite was used in these experiments; no mention of concern for surface oxidation of the sodium sulfite was made. 2.2.6.1.3 Verlaeten et al.s (1991)[20] sodium sulfite reaction findings.

In an effort to substantiate Verlaeten et al.s (1991)[20] reaction pathway sodium sulfite reaction tests with NO and O2 from 27C to 727C were conducted to check for NOx removal. Verlaeten et al. (1991)[20] like Bland (1990)[16] found no NO2 formation and no NOx removal by the reaction at any temperature tested. Reagent grade sodium sulfite

74

was used for Verlaeten et al.s (1991)[20] tests; no mention of any special handling of sodium sulfite was made to avoid surface oxidation of the sodium sulfite. 2.2.6.1.4 Lai and Yesavage (1994)[21] sodium sulfite reaction findings.

Lai and Yesavage (1994)[21] apparently recognized the importance of the quality of sodium sulfite used in experimentation with NO and O2. Lai and Yesavage (1994)[21] employed two special preparation procedures for sodium sulfite to avoid surface oxidation. Lai and Yesavage (1994)[21] examined the reaction of sodium sulfite with NO and O2 in a batch reactor at 125C (257F) to determine if either Blands (1990)[16] or Verlaeten et al.s (1991)[20] hypothesized reaction pathways were valid. These investigations sought to determine if either NO2 was formed by the reaction as Bland (1990)[16] hypothesized or if NOx was removed (sodium nitrite and sodium nitrate formation) by the reaction as Verlaeten et al. (1991)[20] hypothesized. Lai and Yesavages (1994)[21] investigations avoided sodium sulfite surface oxidation problems by employing special sodium sulfite preparations. Two special sodium sulfite preparation methods were used: 1) A special method involving crushing and storing the sodium sulfite prior to test was used. This method mechanically crushed (manual ground the powder with a mortar and pestle) to remove surface oxidation that may have formed on the crystalline technical grade sodium sulfite powder during its shelf life. This processing aimed to expose fresh un-oxidized sodium sulfite from the particle bulk beneath the surface. Additionally, after crushing, the sodium sulfite was stored in a special desiccator to maintain dryness and prevent subsequent oxidation. 75

2) A special method involving fresh preparation of the sodium sulfite in the reactor was used. For this method, sodium sulfite was formed fresh in the reactor by reacting sodium bicarbonate with SO2 (without NO) at 125C. Recall that Howatson et al. (1980)[8] had demonstrated and already established that sodium sulfite formed under these conditions. The fresh sodium sulfite was not removed from the reactor, or exposed in any way to the atmosphere prior to test but was instead kept under nitrogen to prevent surface oxidation. Lai and Yesavage (1994)[21] used the two special sodium sulfite preparation methods described above in preparing the sodium sulfite for reaction with NO and O2. Lai and Yesavage (1994)[21] reported that these specially prepared sodium sulfite samples reacted with NO and O2 and resulted in NOx removal. The NOx removal from these reactions was reported to form sodium nitrite and sodium nitrate product. Note that SO2 formation was not detected during either of these NOx removal reactions involving the specially prepared sodium sulfite. 2.2.6.2 2.2.6.2.1 Review of sodium pyrosulfite experimental findings. Blands (1990)[16] sodium bisulfite and sodium pyrosulfite reaction findings. In addition to sodium sulfite, Bland (1990)[16] screened sodium bisulfite, NaHSO3, and sodium pyrosulfite, Na2S2O5, to determine if it caused heterogeneous catalysis according to Blands (1990) [16] hypothesis. Sodium bisulfite and sodium pyrosulfite were tested because they were both sodium sulfur-oxide intermediate compounds that may have formed during sulfation. Bland (1990)[16] tested the reactivity of these two

76

compounds with NO and O2 at 163C for NO2 formation. Bland (1990)[16] reported no NO2 formation or NOx removal for either of these tests. 2.2.6.2.2 Verlaeten et al.s (1991)[20] sodium pyrosulfite reaction findings.

As described in Section 2.3.6.1 Verlaeten et al. (1991)[20] and Bland (1990)[16] were in agreement regarding sodium sulfite reactivity with NO and O2. Verlaeten et al. (1991)[20] and Bland (1990)[16] both believed that sodium sulfite was non-reactive with NO and O2. Verlaeten et al. (1991)[20] and Bland (1990)[16] were not in agreement regarding the reactivity of sodium bisulfite and sodium pyrosulfite with NO and O2. Verlaeten et al. (1991)[20] tested reagent grade sodium pyrosulfite and reagent grade sodium sulfite with NO and O2 from 27C to 727C. As hypothesized Verlaeten et al. (1991)[20] reported that sodium pyrosulfite reacted with NO and O2 between 127C and 227C. NOx removal was reportedly measured, and sodium nitrite and sodium nitrate were reportedly found as products. It is unclear why Bland (1990)[16] found no NOx removal by NaHSO3 and Na2S2O5 at 162C. 2.2.6.2.3 Lai and Yesavage (1994)[21] sodium pyrosulfite reaction findings.

To resolve the discrepancy in the sodium pyrosulfite findings between Bland (1990)[16] and Verlaeten et al. (1991)[20], Lai and Yesavage (1994)[21] re-tested the reactivity of sodium pyrosulfite with NO and O2. They speculated that the discrepant results may have resulted because Blands (1990)[16] sodium pyrosulfite powder many have been surface oxidized or contaminated in some way. As a result they ran their reactivity tests using specially prepared sodium pyrosulfite to avoid the possibility of surface oxidation. In particular, they mechanically crushed (manual ground the powder 77

with a mortar and pestle) the crystalline technical grade sodium pyrosulfite powder prior to test to remove possible surface oxidation that may have formed during its shelf life. This processing aimed to expose fresh un-oxidized sodium pyrosulfite from the particle bulk beneath the surface. Additionally, after crushing, the sodium pyrosulfite was stored in a special desiccator to maintain dryness and prevent subsequent oxidation. The storage of this material was maintained separate and away from other materials to avoid it contaminating other materials, since Lai and Yesavage (1994)[21] understood that over time SO2 was slowly evolved from sodium pyrosulfite due to its equilibrium decomposition to sodium sulfite. The reactivity test of sodium pyrosulfite with the NO flue gas mixture resulted in NOx removal and SO2 evolution as Verlaeten et al. (1991)[20] had reported and hypothesized. The mole ratio of SO2 evolution to NOx removal measured by Lai and Yesavage (1994)[21] matched the stoichiometry of Verlaeten et al.s (1991)[20] reaction equation 2.39. It should be mentioned that Lai and Yesavages (1994)[21] tests experienced NOx loss from leakage around the reactor rotating paddle stem. However for the 5 minute reactivity tests these loses were shown to be less then 1% of the total inlet NO which was small compared to the NOx removal percentages measured for the sodium pyrosulfite. 2.2.6.3 Review of sodium sulfate reaction findings. In addition to screening sodium sulfite, sodium bisulfite, and sodium pyrosulfite Bland (1990)[16] also screened sodium sulfate, Na2SO4 to determine if it caused heterogeneous catalysis of NO to NO2 according to Blands (1990)[16] hypothesis. Sodium sulfate was tested because it was a sodium sulfur-oxide compound that was 78

known to form by the sulfation reactions. Bland (1990) [16] tested its reactivity with NO and O2 at 121C to 188C for NO2 formation. Bland (1990)[16] found that no NO2 was formed by the reaction and that no NOx was removed by the reaction for any of the test temperatures. 2.2.6.4 Review of sodium nitrite and sodium nitrate reaction findings. According to Kozlowski et al. (1970)[22] as early as 1892, Hodgkinson and Young reported that reactions between SO2 and solid nitrates resulted in complete conversion of nitrate to sulfate along with the evolution of nitrous fumes.[22] Also Kozlowski et al.(1970)[22] experimentally found from 310C-400C that molten sodium nitrate, NaNO3(liquid), reacted with SO2 to form solid sodium sulfate and NO2 gas. And that molten sodium nitrite reacted with SO2 to form solid sodium sulfate, NO, NO2 and N2O. Kozlowski et al.s (1970)[22] sodium nitrite and sodium nitrate tests were conducted only at molten conditions. Bland (1990)[16] tested the reaction of sodium nitrite and sodium nitrate with SO2 and O2 between 21C and 177C to determine if either NO or NO2 was evolved or SO2 was removed. Bland (1990)[16] found for the reaction of sodium nitrite with SO2 and O2 that NO and NO2 were both evolved at 116C. According to Bland (1990)[16] sodium nitrite could not be formed at 177C. It is unclear by what reaction Bland (1990)[16] was referring to that sodium nitrite could not be formed at 177C since sodium nitrite as a compound is stable at 177C. According to the handbook it has a melt point value of 271C.[62]

79

Bland (1990)[16] found for the reaction of sodium nitrate and SO2 and O2 that neither NO nor NO2 was evolved at any temperature tested. According to Bland (1990)[16] sodium nitrate could be formed at 177C. Blands (1990)[16] results for sodium nitrate conflict with the results of Kozlowski et al. (1970)[22] that found NO2 formation for the reaction. This discrepancy may be accounted for by either the test temperatures investigated or possible surface contamination of the sodium nitrate used by Bland (1990)[16]. Bland (1990)[16] used technical grade sodium nitrate with no special surface preparation. The lack of reactivity of the technical grade sodium nitrate used by Bland (1990)[16] may have been due to sodium nitrate surface contamination from sodium mono-dimethyl napthalene sulfonate, an anti-caking additive [63] commonly added as a surfactant during sodium nitrate manufacture. The lack of reactivity may also have been due to water vapor surface coverage from deliquescence with the atmosphere during shelf life and handling. 2.2.6.5 Review of gas phase interactions of SO2 and NOx. The gas phase reactions between sulfur dioxide, and nitrogen oxide, nitrogen dioxide, oxygen, and water vapor have been the subject of many investigations related to the chamber processes for the manufacture or sulfuric acid. According to Duecker and West (1959)[37], Saleski and Lunge studied the gas phase oxidation of SO2 to SO3 and found that NO2, not NO oxidized SO2 to SO3 in the presence of water vapor from 50100C according to the following reaction: H2O SO2 + NO2 SO3 + NO (2.47)

80

Although water vapor does not appear as a reactant in the reaction equation, it is required according to the researchers for the reaction to proceed and is considered a catalyst. In the absence of water vapor Stopperka et al.(1968)[64] searched for catalysts that oxidized SO2 to SO3, catalyst SO2 + O2 SO3 and found only the solid V2O5 to catalyze the reaction. In the absence of water, Stopperka et al.(1968)[64] indicated that SO2, NO2, and NO reacted to form a solid from 225 to 230C, 2SO2 + NO2 + NO (NO)2S2O7 (2.49) (2.48)

With water vapor present a number of researchers have shown that nitrosyl hydrogen sulfate, NOHSO4, also known as nitrosylsulfuric acid, SO5NH, form, 3NO2 + 2SO2 + H2O 2NOHSO4 + NO (2.50)

Figure 2.17 shows the schematic representation of these two mixed SOx/NOx compounds.

81

SO2, NO2, T>225C

(NO)2S2O7

NO SO2, NO2, 25C< T NOHSO4

Figure 2-17 Schematic of mixed SOx/NOx compounds formed in the chamber process for the manufacture of sulfuric acid. 2.3 Conclusions This chapter provided a brief review of plumes. The history of plumes was presented along with descriptions of reported observations of plume discoloration during dry sodium bicarbonate injection and from other sodium related flue gas desulfurization processes. A description of how NO2 was hypothesized and validated as the agent responsible for the discoloration was given. Then past investigations on the reaction pathway for discoloration were discussed. First the early studies made when SO2 and NOx reactions were considered independent of one another were discussed. Second the tests made after it was realized that NO2 formation was linked with SO2 removal were discussed. The reaction pathways of Miller (1986)[19], Bland (1990)[16], Verlaeten et al. (1991)[20], and Bortz (1994)[45] were all described including schematic representations of each hypothesized reaction pathway. 82

By reviewing the literature on sodium bicarbonate reaction chemistry it is evident that the reaction of sodium bicarbonate with flue gas with SO2 and NO is complex. The reaction is complicated by a number of factors including, decomposition and temperature. From the literature review a number of fundamental reaction findings have been well established and serve as a foundation from which the reaction pathway for NO2 formation and SO2 and NOx removal may be determined. Additionally, a number of discrepancies in reaction findings have been noted for further understanding: 1) Sodium bicarbonate decomposition reactions have been well established and serve as a fundamental building block. It is still however unclear how the presence of NO affects acts to inhibit decomposition at low temperatures. 2) Sodium bicarbonate sulfation reactions have been well established and serve as a fundamental building block. It is still however, unclear how the presence of NO acts to affect sulfation where only sodium sulfate is produced. 3) It is well established that NOx removal occurs for the reaction of sodium bicarbonate with flue gas with SO2 and NO during a finite time period after initial exposure, however the exact reaction path is not understood. 4) It is well established that NO2 formation occurs for the reaction of sodium bicarbonate with flue gas with SO2 and NO only after the finite time period where NOx is removed is complete. 5) It is well established that at low temperature water vapor affects the amount of NOx removal by affecting the finite reaction time period.

83

6)

It is well established that NOx is removed by the reaction of sodium carbonate with NO2, but it is unclear why NOx is not removed by the reaction of sodium carbonate with SO2 and NO when NO2 is formed by the reaction.

7)

Discrepant results exist on the reaction of sodium sulfite with NO and O2. It is unclear if NOx is removed by the reaction since conflicting results have been reported. Surface purity issues have been raised by past researchers.

8)

Discrepant results exist on the reaction of sodium pyrosulfite with NO and O2. It is unclear if NOx is removed by the reaction since conflicting results have been reported. Also it is unclear if sodium pyrosulfite actually forms as a sulfation reaction intermediate compound.

9)

It has been well established that the reaction of sodium nitrite with SO2 and O2 results in NO, and NO2 formation. But it is unclear how sodium nitrite is formed within the reaction of sodium bicarbonate with flue gas with SO2 and NO.

10)

Discrepant results exist on the reaction of sodium nitrate with SO2 and O2. It is unclear if NO2 is formed by the reaction since conflicting results have been reported. Additionally it is unclear how sodium nitrate is formed within the reaction of sodium bicarbonate with flue gas with SO2 and NO. As already stated discrepant results on sodium nitrate formation exist for the reactions of sodium sulfite and sodium pyrosulfite with NO and O2; and the formation of sodium pyrosulfite as a reaction intermediate has not yet been validated.

The next chapter will discuss the theoretical implications of the findings enumerated in this chapter by the past investigators. These implications will then serve as the technical basis for the theoretical development of the hypothesis of study.

84

CHAPTER 3 THEORETICAL DEVELOPMENT 3.1 Introduction to theoretical development

Despite numerous past investigations of NO2 formation during the reaction of sodium bicarbonate and SO2 flue gas with NO, as described in Chapter 2, the overall reaction pathway of this reaction has not yet been found by any investigators to date. The investigated findings to date have however, revealed a number of key undisputed reactions and end products. For this chapter the relevant undisputed reactions and corresponding end products from past investigations are compiled together for theoretical development of a new pathway. Knowledge of the individual related reactions is useful in determining the overall reaction sequence for sodium bicarbonate combined with SO2 flue gas with NO. A nested combination of these known particular reactions and end products is used as the framework for the development of the overall reaction pathway. For example, the reaction of sodium bicarbonate with SO2 flue gas without NO serves as a segment of the framework of the overall reaction pathway for the reaction of sodium bicarbonate with SO2 flue gas with NO. Also in this chapter a list of reaction pathway criteria are established from the set of past relevant undisputed reactions and corresponding end products. These reaction criteria are used to evaluate the overall validity of past and present hypothesized reaction pathways for NO2 formation. In particular the established reaction criteria serve to evaluate hypothesized reaction pathways of Miller (1986), Bland (1990), Verlaeten et al.

85

(1991), and Bortz (1994) and the work of Lai and Yesavage (1994)[21]. Recall that these past pathways were described in Chapter 2 Sections 2.2.4.1-2.2.4.3 respectively. Through these evaluations the shortcomings of each of the past hypothesized reaction pathways is identified. Also through these evaluations controversial reactions where there is little agreement among the past investigators on the reaction end products are identified. The summary set of relevant undisputed reactions and relevant controversial reactions are then compiled together for the development of a foundation for the new hypothesized reaction pathway. 3.2 Summary of key undisputed reactions and end products The following is a summary list of relevant key reactions and their respective end products compiled from past investigations as described in detail in Chapter 2. These reactions and end products are undisputed and have been validated by past investigators: 1) The decomposition reaction of sodium bicarbonate as given in equation 3.1 was revealed by past investigators[46,47,48,49,50] as discussed in Section 2.2.1. Sodium carbonate was found to form as the solid phase end product for this decomposition reaction depending on temperature. 2NaHCO3 Na2CO3 + H2O + CO2 (3.1)

2) The reaction of sodium bicarbonate with flue gas with SO2 (without NO) was shown by Howatson et al.(1980) [8] to produce a temperature dependent mixture of sodium sulfite and sodium sulfate as described in Section 2.2.2.2. Note that the flue gas for this reaction includes nitrogen, oxygen, CO2, and SO2 but does not include NO. Keener and Khang (1982, 1993)[9,52] postulated the NaHCO3 reactions with SO2 and 86

O2 without NO as listed in equations 3.2-3.5. Sodium sulfite oxidizes to sodium sulfate with temperature and oxygen. 2NaHCO3 + SO2 Na2SO3 + 2CO2 + H2O 2NaHCO3 Na2CO3 + H2O + CO2 Na2CO3 + SO2 Na2SO3 + CO2 Na2SO3 + O2 Na2SO4 (3.2) (3.3) (3.4) (3.5)

3) The reaction of sodium bicarbonate with NO and O2 as well as the reaction of sodium carbonate with NO and O2 were found to produce no reaction or reaction end products. These findings were first reported by Knight (1977)[18] as discussed in Section 2.2.3. 4) The reaction of sodium carbonate with NO2 (in the absence of SO2 ) was shown to produce sodium nitrate as listed in equation 3.6. These findings were first reported by Knight (1977)[18] as discussed in Section 2.2.3. Na2CO3 + NO2 2NaNO3 + NO + CO2 (3.6)

5) The reaction of sodium bicarbonate with oxygen free flue gas with SO2 was shown by Bland (1990)[16] to produce sodium sulfite regardless of temperature as listed in equations 3.7-3.9 and as discussed in Section 2.2.4.1. Similarly the reaction of sodium carbonate with oxygen free flue gas with SO2 was shown by Bland (1990)[16] to produce sodium sulfite regardless of temperature as equation 3.7 and 3.9 show, and as discussed in Section 2.2.4.1. Note that the flue gas for these reactions was absent of NO. 2NaHCO3 + SO2 Na2SO3 + 2CO2 + H2O 2NaHCO3 Na2CO3 + H2O + CO2 87 (3.7) (3.8)

Na2CO3 + SO2 Na2SO3 + CO2

(3.9)

6) The reaction of sodium sulfate with NO and O2 was found to produce no reaction or end product according to Bland (1990)[16] as discussed in Section 2.2.4.1. 7) The reaction of sodium nitrite with SO2 and O2 was found by Bland (1990)[16] to produce gaseous NO, NO2 and solid sodium sulfate as discussed in Section 2.2.4.1. Bland (1990)[16] did not identify the exact stoichiometry, or the resultant ratio of NO to NO2 with respect to temperature. The speculated reactions are listed in equations 3.10-3.12. 2NaNO2 + SO2 2NO + Na2SO4 2NaNO2 + SO2 + O2 2NO2 + Na2SO4 2NaNO2 + SO2 + O2 NO + NO2 + Na2SO4 (3.10) (3.11) (3.12)

8) The reaction of sodium bicarbonate and flue gas with SO2 and NO was found to produce the following end products according to Bland (1990)[16] as described in Section 2.2.4.1: a) solid sodium sulfate b) sodium nitrate during a finite initial time period. c) gaseous NO2 after the initial finite time period. It should be noted that Verlaeten et al.[20] reported that in addition to sodium nitrate, sodium nitrite was also formed by the nitration portion of the reaction. 9) The reaction of sodium carbonate and flue gas with SO2 and NO was found by Bland (1990)[16] as described in Section 2.2.4.1 to produce the following end products: a) solid sodium sulfate b) gaseous NO2.

88

Bland (1990)[16] noted that this reaction was unlike the reaction with sodium bicarbonate since sodium nitrate was not produced. Similarly Verlaeten et al.(1991)[20] as discussed in Section 2.2.4.2 reported that this reaction did not produce either sodium nitrate nor sodium nitrite. The exact reaction equations were not postulated by either of these investigators. 10) The reaction of sodium bicarbonate or sodium carbonate with oxygen free flue gas with SO2 and NO was found by Bland (1990)[16]. The oxygen free reaction was unlike the reaction with oxygen. With oxygen sodium sulfate, sodium nitrate, sodium nitrite and NO2 were found to form; without oxygen only sodium sulfite was found to form as described in Section 2.2.4.1. Neither sodium sulfate, sodium nitrate, sodium nitrite nor NO2 were reported to form when oxygen was absent. The exact reaction equation was not postulated. Table 3.1 summarizes the reactants and the corresponding end products discussed above.

89

Table 3-1 Summary of undisputed relevant reactions and end products. List No. 1 2 Rxn. Table 3.1 3.2 Solid Reactant NaHCO3 NaHCO3

Gas Reactants N2 SO2, O2

Solid End Products Na2CO3* Na2SO3, Na2SO4, Na2CO3* Na2SO3, Na2SO4 Na2CO3*, no rxn no rxn NaNO3, NO, CO2 Na2SO3, Na2CO3* Na2SO3 no rxn NO, NO2, Na2SOx NO2, Na2SO4, NaNO3, (NaNO2), Na2CO3* NO2, Na2SO4, (no NaNO3 or NaNO2)

2 3 3 4 5 5 6 7 8

3.2 ----3.3 3.4 3.4 --3.5 ---

Na2CO3 NaHCO3 Na2CO3 Na2CO3 NaHCO3 Na2CO3 Na2SO4 NaNO2 NaHCO3

SO2, O2 NO, O2 NO, O2 NO2 SO2 (no O2) SO2 (no O2) NO, O2 SO2, O2 SO2, NO, O2

Reference Barrall and Rogers[46] Howatson et al.[8], Keener and Khang[9,52] Bland[16] Knight[18] Knight[18] Knight[18] Bland[16] Bland[16] Bland[16] Bland[16] Bland[16], (Verleaten et al.[20]) Bland[16], (Verleaten et al.[20]) Bland[16]

---

Na2CO3

SO2, NO, O2

10

SO2, NO, (no Na2SO3, Na2CO3* O2) 10 --Na2CO3 SO2, NO, (no Bland[16] Na2SO3 O2) * from decomposition of sodium bicarbonate, no rxn no reaction products

---

NaHCO3

90

3.3

The criteria for assessment.

The criteria for assessment are based on the undisputed reactions obtained experimental findings made in past laboratory and demonstration tests from as listed in Section 3.2 discussed in further detail in Chapter 2. The criteria have been sub-divided into the following three groups: 1) Criteria related to SO2 removal 2) Criteria related to NOx removal 3) Criteria related to NO2 formation These criteria are made with the assumption that the normalized stoichiometric ratio (NSR) is greater then 1.5, and the inlet SO2 to NO ratio is greater then 2. 3.3.1 Criteria for SO2 removal

Table 3.2 lists a summary of all of the SO2 removal criteria. Table 3-2 Summary of SO2 removal criteria. Criteria ID SO2 CR 1 Requirement Without NO present sodium bicarbonate and sodium carbonate react with flue gas with SO2 (without NO) to form a temperature dependent mixture of sodium sulfite and sulfate in the product. For the reaction of sodium bicarbonate and sodium carbonate with flue gas with SO2 and NO, sodium sulfate accumulates as the SO2 removal product regardless of temperature (no sodium sulfite forms). For the reaction of sodium bicarbonate and sodium carbonate with flue gas with SO2 and NO but without O2, only sodium sulfite forms as a product. Basis Section: Howatson et al.[8] 2.3.2.2 Bland [16] 2.3.4.1 Bland [16] 2.3.4.1

SO2 CR 2

SO2 CR 3

Each of the SO2 removal criteria (CR) are discussed below including their experimental basis. 91

Recall that SO2 CR 1 criterion was established based on the findings of Howatson et al.(1980)[8] who showed that a temperature dependent mixture of sodium sulfite and sulfate formed when sodium bicarbonate and sodium carbonate were reacted with flue gas with SO2 (without NO) as described in Section 2.2.2.2. Specifically at low temperatures sodium sulfite was primarily found to form while at high temperatures sodium sulfate was primarily found to form (see Figure 2.6). Bland (1990)[16] additionally investigated and verified these findings of Howatson et al.(1980)[8] for the reaction of flue gas with SO2 (without NO). Recall that SO2 CR 2 criterion was established based on the findings of Bland (1990)[16]. Bland (1990)[16] showed that the sulfation product formed for this reaction was different than the temperature dependent mixture of sodium sulfite and sodium sulfate product formed for the reaction of sodium bicarbonate and sodium carbonate with flue gas with SO2 (without NO). Bland (1990)[16] showed that only sodium sulfate formed regardless of temperature when NO was present in the flue gas with SO2 as described in Section 2.2.4.1. Recall that SO2 CR 3 criterion was established by the laboratory experiments of Bland (1990)[16]. Bland (1990)[16] found sodium bicarbonate reacted with oxygen free flue gas containing SO2 and NO to form only sodium sulfite product as described in Section 2.2.4.1. 3.3.2 Criteria for NOx removal.

Table 3.3 lists a summary of all of the criteria for NOx removal for the reaction of sodium bicarbonate and sodium carbonate with SO2 flue gas with NO.

92

Table 3-3 Summary of NOx removal criteria. Criteria ID NOx CR 1 NOx CR 2 Requirement NOx is removed by the reaction of sodium carbonate with NO2. Sodium nitrate is formed by this reaction. NOx is not removed (sodium nitrate is not produced) by the reaction of sodium carbonate with SO2 flue gas with NO. NO2 and sodium sulfate are however produced by this reaction. For the reaction of sodium bicarbonate with SO2 flue gas with NO: 3a NOx is removed only during SO2 removal; NOx removal is directly linked to SO2 removal. 3b NOx is removed during a finite time period immediately following injection. The duration of this time period is temperature dependent. 3c NOx removal is dependent on flue gas humidity where the amount of NOx removed increases with increasing flue gas humidity. 3d NOx removal is dependent on the presence of oxygen. NOx is removed only when O2 is present in the flue gas. Basis Section: Knight[19] Section 2.3.3 Bland[16] 2.3.4.1 Verlaeten et al.[20] Section 2.3.4.2 Bland[16] Section 2.3.4.1 Bland[16] Section 2.3.4.1 Bortz[45] Section 2.3.4.3 Bland[16] Section 2.3.4.1 Bland[16] Section 2.3.4.1

NOx CR 3 NOx CR NOx CR

NOx CR

NOx CR

The NOx removal criteria are described below including mention of their experimental basis. Recall that NOx CR 1 criterion was established based on the findings of Knight (1977)[19]. Knight (1977)[19] found that sodium carbonate when reacted with NO2 produced sodium nitrate as described in Section 2.2.3. It should be noted that this criterion does not preclude the existence of other NOx removal reactions. NOx CR 2 criterion was established based on the findings of Verlaeten et al.(1990)[20] and Bland (1990)[16] which experimentally showed that sodium carbonate and SO2 flue gas with NO did not result in any sodium nitrate formation. Recall in particular that Verlaeten et al.s (1990)[20] work showed sodium nitrate was formed by the reaction of sodium bicarbonate with SO2 flue gas with NO while sodium nitrite was not formed by

93

the reaction of sodium carbonate and SO2 flue gas with NO. Verlaeten et al. (1990)[20] speculated that the reactions of sodium bicarbonate and sodium carbonate with SO2 flue gas with NO were different because the sodium bicarbonate reacted with the SO2 and formed sodium bisulfate when the sodium carbonate was unable to react to form the sodium bicarbonate. Blands (1990)[16] packed bed screening tests confirmed that sodium nitrate was not formed by the reaction of sodium carbonate and SO2 flue gas with NO. Although Verlaeten et al.s(1990)[20] and Blands (1990)[16] experimental work confirms that sodium nitrate is not formed by the reaction of sodium carbonate and SO2 flue gas with NO, theoretically it may initially seem odd since the reaction of sodium carbonate and SO2 flue gas with NO produces NO2. And it is well established that sodium carbonate reacts with NO2 to produce sodium nitrate. But with closer examination one realizes that no past findings have shown that sodium nitrate is formed when sodium carbonate is reacted with NO2 when SO2 is also present in the flue gas. There have been no past claims of sodium nitrate formation by this reaction in the presence of SO2. Evidently in the presence of SO2 there is an interaction that interferes with the formation of sodium nitrate. One possible interaction may be that the SO2 reacts with sodium nitrate destroying it at a rate comparable to the rate at which it is formed resulting in no net accumulation of sodium nitrate. In this way the formation of sodium nitrate may in a sense be transparent since it is converted immediately after it forms. SO2 in the flue gas may immediately react with this sodium nitrate causing it to breakdown into gaseous NO2 and sodium

94

sulfate. With SO2 present sodium nitrate does not accumulate but instead converts to sodium sulfate and NO2. Another possible reason that sodium nitrate does not form may be that forming an impermeable sulfation product layer around the core sodium carbonate forms as a result of the SO2 reaction with sodium carbonate from the SO2 flue gas with NO. This impermeable sulfation product layer may then inhibit NO2 gas from diffusing through to the sodium carbonate core for reaction. This possibility is however highly unlikely since an impermeable sulfation would also result in the cessation of further diffusion and reaction of SO2 with sodium carbonate since the molecular diameter of SO2 is larger than the molecular diameter of NO2. In reality SO2 diffusion and reaction with sodium carbonate does not cease but is instead ongoing. NOx removal criteria (3a-d) relate to the reaction of sodium bicarbonate with SO2 flue gas with NO. These NOx removal criteria all involve the formation of a solid nitration product such as sodium nitrate or sodium nitrite. NOx CR 3a criterion was established based on the findings of Bland (1990)[16]. Recall that Bland (1990)[16] experimentally showed that NOx removal occurred only during SO2 removal as described in Section 2.2.4.1. In particular, Bland specifically revealed the linkage between SO2 removal and NOx removal by showing that NOx removal ceased when SO2 was shut off during sodium bicarbonate injection. Bland (1990)[16] showed experimentally that NOx was not removed if SO2 was not present in the flue gas mixture. Bland reported that NOx was not removed when sodium bicarbonate was reacted with NO. Blands (1990)[16] work showed that active sulfation was needed for NOx removal to occur.

95

NOx CR 3b criterion was established based on the batch experiments of Bland (1990)[16] and Bortz (1994)[45]. Recall from Sections 2.2.4.1 and Section 2.2.4.3 that Bland (1990)[16] and Bortz (1994)[45] experimentally tracked NOx removal with respect to time after batch injection. Bland (1990)[16] showed for the batch reaction of sodium bicarbonate with SO2 flue gas with NO there was a finite time period of NOx removal immediately following sodium bicarbonate injection in which NOx was removed. Bortzs (1994)[45] work in particular showed in more detail that this finite time period of NOx removal was temperature dependent. The finite time period shortened when the reaction temperature was increased. Note that Bland (1990)[16] and Bortzs (1994)[45] experiments were conducted with temperatures close to the sodium bicarbonate decomposition temperature. Bortz (1994)[45] showed that the finite time period of NOx removal went to zero when the reaction temperature was raised well above the temperature of decomposition of sodium bicarbonate. This is of interest because it indicates that either diffusion mechanics or water vapor chemically interacts within the reactions. NOx CR 3c criterion was established based on Bortzs (1994)[45] findings as described in Section 2.2.4.3. Bortz (1994)[45] showed that the amount of NOx removed was dependent upon the flue gas humidity. Bortzs (1994)[45] studies showed 30% more NOx removal occurred for flue gas with 15% humidity compared to 9% humidity. This is of interest because it reveals that water vapor interacts within the reaction sequence. NOx CR 3d criterion was established based on the laboratory studies of Bland (1990)[16] as described in Section 2.2.4.1. Recall that Bland (1990)[16] ran experiments with and

96

without oxygen in the flue gas and showed that NOx was not removed when oxygen was not present. 3.3.3 Criteria for NO2 formation.

A summary of the criteria for NO2 formation is listed in Table 3.4. Table 3-4 Summary of NO2 formation criteria. Criteria ID NO2 CR 1 Requirement For the reaction of sodium bicarbonate/carbonate with SO2 flue gas with NO, NO2 formation occurs only during SO2 removal. After sodium bicarbonate is injected into SO2 flue gas with NO, there is a time lag before NO2 begins to form. NO2 formation lags behind NOx removal and SO2 removal. This lag period increases with decreasing temperature. At low temperatures NO2 formation is decreased by increasing flue gas humidity during sodium bicarbonate injection into SO2 flue gas with NO. For the reaction of sodium bicarbonate/carbonate with SO2 flue gas with NO, NO2 formation occurs only when O2 is present in the flue gas. Basis Section: Bland [16] Sec. 2.3.4.1 Bland [16] Sec. 2.3.4.1 Bortz [20] Sec. 2.3.4.3 Bortz [20] Sec. 2000000000 0000.3.4.3 Bland [16] 2.3.4.1

NO2 CR 2

NO2 CR 3

NO2 CR 4

The NO2 formation criteria are described below including mention of the experimental basis for each criterion. NO2 CR 1 criterion was established based on the demonstration test results and the laboratory findings of Bland (1990) [16]. Recall from Section 2.2.4.1 Bland (1990)[16] found that NO2 formation was linked directly to active SO2 removal. Bland (1990)[16] showed that NO2 formation occurred whenever SO2 was present. At the demonstration tests at R.D. Nixon, Bland (1990)[16] showed that plume discoloration (from NO2 formation) stopped whenever sodium bicarbonate injection was stopped despite the

97

contact of the flue gas with the spent sorbent (sodium sulfate). Bland (1990)[16] in the laboratory also showed that NO2 formation stopped whenever SO2 flow was shut off. NO2 CR 2 criterion was established based on the laboratory findings of Bland (1990) [16] and Bortz (1994) [20]. Concentration traces of SO2, NO, and NO2 for sodium bicarbonate injection revealed a lag in NO2 formation relative to the onset of NOx and SO2 removal with respect to time. These concentration traces revealed that NOx and SO2 are removed immediately following sodium bicarbonate injection but that NO2 is not formed immediately following sodium bicarbonate injection. Instead NO2 was shown to form after a finite time period occurred called a lag. It is important to note that this lag period for NO2 formation was shown to occur precisely when the finite time period of NOx removal occurred. Importantly, the finite time period NOx removal (discussed under the NOx CR 3b criterion) is shown to coincide with the NO2 lag- that is NOx is removed during the period that no NO2 is formed. Bortz (1994) [20] showed that the NO2 lag period depended on temperature where increased temperature decreased the lag period. The lag in NO2 formation is likely to be related to the evolution of water vapor from decomposition. Careful review of Bortzs (1994) [20] low temperature concentration traces of NO2 formation from a continuous sodium bicarbonate injection process provides some evidence that NO2 lag is associated with water vapor generation from decomposition. Importantly, Bortzs (1994) [20] traces of NO2 concentration at temperatures close to decomposition show that there is a delineated time period where no NO2 is formed followed by a period where a gradual increase in NO2 formation up to a steady state level of NO2 formation occurs. The NO2 lag time period coincides consistently with the time when water vapor is released from

98

decomposition of sodium bicarbonate. Correspondingly, as temperature increases, both the time period for water vapor release by decomposition and the time period of lag in NO2 formation decrease. Up to this point, the exact reaction sequence involving water vapor and NO2 formation has not been clearly explained. Counter diffusion of water vapor from decomposition can be ruled-out as the cause of the lag in NO2 formation since counter diffusion of water vapor would also affect SO2 and NOx removal. But Bland (1990) [16] and Bortz (1994)
[20]

showed that SO2 removal does not lag during the time period of water vapor release

by decomposition. SO2 removal instead has been shown to occur smoothly after injection during decomposition despite counter diffusion of water vapor. NO2 CR 3 criterion was established based on the humidity studies of Bortz (1994) [20]. Bortz (1994) [20] showed that NO2 formation decreased for sodium bicarbonate injection when the flue gas humidity was increased. NO2 formation was shown to decrease 50% when the flue gas humidity was raised 6% (from 9% humidity to 15% humidity). NO2 CR 4 criterion was established based on the laboratory findings of Bland [16]. Bland
[16]

showed for the reaction of sodium bicarbonate and SO2 flue gas with NO that NO2

formation stopped when oxygen was shut off from the flue gas mixture. 3.4 Conformity of each of the past hypothesized pathways with the criteria.

The criteria summarized in Table 3.2, 3.3, and 3.4 serve as a basis for assessing possible proposed reaction paths. If a hypothesized reaction pathway conforms to all of the criteria and if each of the reactions in the pathway can be individually validated then it is highly likely that a proposed pathway is the actual pathway. As a means of understanding why the past hypothesized reaction pathways of Miller (1986)[19], Bland 99

(1990) [16], Verlaeten et al. (1991)[20] and Bortz (1994)[45] ) are not the actual pathway, each will be assessed for conformity to the criteria. Assessing the past hypothesized reaction pathways is useful in determining the portions of each which do not conform to the criteria and thus may be discarded while maintaining the portions which conform to the criteria and may thus be used as a building block for a new reaction pathway hypothesis. 3.4.1 Conformity of Millers hypothesized reaction pathway to the criteria for SO2 removal, NO2 formation, and NOx removal. Millers (1986)[19] pathway was described in Section 2.2.5.1. A summary of the applicable pathway is shown in Figure 2.12a. 3.4.1.1 Conformity of Millers pathway to the SO2 removal criteria.

Table 3.5 lists the conformity of Millers (1986)[19] hypothesized reaction pathway to the SO2 removal criteria. Table 3-5 Conformity of Millers (1986)[19] hypothesized reaction pathway to the SO2 removal criteria
SO2 Criteria Type CR 1 Criteria Number Pathway Millers NC C-conforms, NC- not conform SO2 CR 2 SO2 CR 3

NC

NC

The conformance to each of the SO2 removal criteria is briefly discussed below.
Conformity to SO2 CR 1.

Millers (1986)[19] pathway does not conform to SO2 CR 1 for the broad range of temperatures under study. Miller constructed his pathway to describe only reactions below 125C where it was hypothesized that without NO only sodium sulfite formed by 100

reaction equation 2.16 and 2.17. Recall however that without NO in the flue gas Howatson et al.(1980)[8] had shown that a temperature dependent mix of sodium sulfite and sodium sulfate form. For Millers pathway to describe the entire temperature range it is necessary to add an additional oxidation reaction of sodium sulfite for sodium sulfate formation.
Conformity to SO2 CR 2.

Millers (1986)[19] pathway does not conform to SO2 CR 2 since Miller (1986)[19] hypothesized that sodium sulfite product accumulated when NO was present in the flue gas. Evidently Miller (1986)[19] was unaware of Blands (1990) [16] findings that only sodium sulfate formed when NO was present. Miller (1986)[19] thought that the sulfation product formed by the reaction of SO2 and sodium carbonate was the same sulfation product that formed by this reaction when NO was present.
Conformity to SO2 CR 3.

Millers (1986)[19] pathway does not conform to SO2 CR 3 since Miller (1986)[19] hypothesized that in the absence of O2 the sulfation compound Na2SO3(NO)2 formed (equations 2.18) in addition to sodium sulfite. 3.4.1.2 Conformity of Millers (1986)[19] pathway to the NOx removal criteria.

Table 3.6 lists the conformity of Millers (1986)[19] hypothesized reaction pathway to the NOx removal criteria.

101

Table 3-6 Conformity of Millers (1986)[19] pathway to the NOx removal criteria.
NOx NOx NOx NOx Criteria Type CR 1 CR 2 CR 3a CR 3b Criteria number Pathway Millers (1986) NA NA NA NA C-conform, NC- not conform, NA- not applicable NOx CR 3c NOx CR 3d

NA

NA

Millers (1986)[19] reaction pathway focused on NO2 formation without specifically considering NOx removal. As a result, the NOx CR 1-3d do not apply. Incidentally however in Millers (1986)[19] NO2 formation pathway it is possible for NOx removal to occur if the reaction rate for equation 2.18 is faster then the reaction rate for equation 2.19; accordingly the compound, Na2SO3(NO)2 would form. Miller implied in his hypothesis however that Na2SO3(NO)2, was an intermediate compound and did not accumulate (it was assumed that the reaction rate of equation 2.19 was equal to or faster then the reaction rate of equation 2.18). 3.4.1.3 Conformity of Millers (1986)[19] pathway to the NO2 formation criteria.

Table 3.7 lists the conformity of Millers (1986)[19] hypothesized reaction pathway to the NO2 formation criteria. Table 3-7 Conformity of Millers (1986)[19] pathway to the NO2 formation criteria
NO2 NO2 NO2 Criteria Type CR 1 CR 2 CR 3 Criteria Number Millers NC NA NC C-conform, NC- not conform, NA- not applicable NO2 CR 4

The conformance of Millers (1986)[19] pathway to each of the NO2 formation criteria will be briefly discussed below.

102

Conformity to NO2 CR 1.

Millers (1986)[19] reaction pathway does not conform to NO2 CR 1 since Miller (1986)[19] hypothesized that NO2 could form without SO2 removal. In Millers (1986)[19] hypothesis NO2 formation occurs after SO2 flow is terminated. In Millers (1986)[19] pathway sodium sulfite, is not destroyed by the NO2 formation reaction, instead, sodium sulfite is generated as an end product (reaction 2.19). After some finite amount of sodium sulfite is formed, NO2 formation would be anticipated to be perpetual since sodium sulfite formation is cyclic within Millers (1986)[19] pathway.
Conformity to NO2 CR 2.

NO2 CR 2 is not applicable to Millers (1986)[19] pathway since Millers (1986)[19] work did not involve the reaction of sodium bicarbonate instead the scope of Millers (1986)[19] pathway includes only the reaction of sodium carbonate with SO2 flue gas with NO.
Conformity to NO2 CR 3.

Millers (1986)[19] pathway does not conform to NO2 CR 3 since the scope of Millers (1986)[19] pathway does not include water as a reactant or a reaction parameter.
Conformity to NO2 CR 4.

Millers (1986)[19] pathway conforms to NO2 CR 4 since NO2 does not form by the pathway when O2 is absent. In particular in Millers (1986)[19] pathway NO2 is not released from the nitrogen-sulfur-oxide reaction intermediate, Na2SO3(NO)2 without the presence of oxygen as described by equation 2.19. 3.4.1.4 Validity of Millers hypothesized pathway.

Millers (1986)[19] pathway does not conform to all of the SO2 removal, NOx removal, and NO2 formation criteria as indicated in Tables 3.13, 3.14 and 3.15. As a result it is 103

unlikely that Millers (1986)[19] pathway is the actual pathway for NO2 formation. Nevertheless, Millers (1986)[19] reaction pathway and experimental findings do possess the following merit: 1) Miller (1986)[19] provided experimental evidence that sodium sulfite is reactive with SO2 flue gas with NO to form NO2. Lai and Yesavage (1994)[21] additionally reported sodium sulfite reactivity with NO. Although both Bland (1990) [16] and Verlaeten et al. (1991)[20], were unable to experimentally show any reactivity of sodium sulfite with NO and O2, Miller (1986)[19] importantly explained that sodium sulfite reactivity depended on its surface purity, suggesting that surface oxidation could inhibit its reactivity. Miller (1986)[19] suggested that sodium sulfite reactivity could only be obtained if special surface preparation of sodium sulfite were completed prior to screening tests. 2) Miller (1986)[19] purported that NO2 formation was due to a release reaction, where NO2 was released from a nitrogen containing reaction-intermediate compound. This idea of NO2 release shows some theoretical promise. Bortz (1994)[45] similarly theorized that the NO2 was released from a nitration intermediate compound formed. Blands (1990) [16] experiments showed the release of NO and NO2 from the nitration compound, sodium nitrite. Continued study of specially prepared sodium sulfite is needed to identify possible intermediate compounds for NO2 release. 3.4.2 Conformity of Blands (1990) [16] original hypothesized reaction pathway to the criteria for SO2 removal, NOx removal and NO2 formation. Blands (1990) [16] original hypothesized reaction pathway was described in Section 2.2.5.2. A summary of this pathway is shown in Figure 2.13. 104

3.4.2.1

Conformity of Blands (1990) [16] original pathway to the SO2 removal criteria.

Table 3.8 lists the conformity of Blands (1990) [16] original hypothesized reaction pathway to the SO2 removal criteria. Table 3-8 Conformity of Blands (1990) Pathway to the SO2 removal criteria
[16]

original

SO2 SO2 SO2 Criteria Type CR 1 CR 2 CR 3 Criteria Number Pathway Blands original C NC C C-conforming NC- non conforming NA- not applicable

The conformance to each of the SO2 removal criteria will be briefly discussed below.
Conformity to SO2 CR 1.

Blands (1990) [16] original pathway conforms to SO2 CR 1 since within Blands (1990)
[16]

original pathway a mixture of sodium sulfite and sodium sulfate forms when NO is

not present.
Conformity to SO2 CR 2.

Blands (1990) [16] original pathway does not conform to SO2 CR 2 since both sodium sulfite and sodium sulfate are formed in Blands (1990) [16] original pathway when NO is present. Sodium sulfite and sodium sulfate are formed as an end product when NO is present by reaction equation 2.21 and 2.24 in Blands (1990) [16] original pathway. Bland (1990) [16] initially maintained that the sulfation product formed in flue gas with NO was the same sulfation product that formed in flue gas without NO; Bland in subsequent hypothesis later modified his view on this issue.

105

Conformity to SO2 CR 3.

Blands (1990) [16] original pathway conforms with SO2 CR 3 since only sodium sulfite is formed by Blands (1990) [16] pathway in the absence of oxygen. Specifically, in the absence of oxygen, sodium sulfate is not formed by reaction equation 2.25, only sodium sulfite is shown to form in the absence of O2. Note that the sodium sulfite formation reactions (equations 2.21 or 2.24) do not require O2 as a reactant. 3.4.2.2 Conformity of Blands (1990) [16] original pathway to the NOx removal criteria. Table 3.9 lists the conformity of Blands (1990) [16] original hypothesized reaction pathway to the NOx removal criteria. Table 3-9 Conformity of Blands (1990) Pathway to the NOx removal Criteria
[16]

original
NOx CR 3d

NOx NOx NOx NOx NOx Criteria Type CR 1 CR 2 CR 3a CR 3b CR 3c Criteria number Pathway Blands original C NC C NC NC C-conforming, NC- non conforming, NA- not applicable

Conformity to NOx CR 1.

Blands (1990) [16] original hypothesis conforms to NOx CR 1 since Blands (1990) [16] pathway incorporates the reaction of NO2 with sodium carbonate according to equation 2.27.
Conformity to NOx CR 2.

Blands (1990) [16] original hypothesis does not conform to NOx CR 2 since in Blands (1990) [16] original pathway NOx is removed by dry injected sodium carbonate. For the reaction of sodium carbonate in SO2 flue gas with NO, NOx is removed by the reaction sequence of equation 2.23 to equation 2.26 to equation 2.27. In Blands (1990) [16] 106

original pathway, NOx removal can occur from either sodium carbonate or from sodium bicarbonate at temperatures above the onset of decomposition.
Conformity to NOx CR 3a.

Blands (1990) [16] original hypothesis conforms to NOx CR 3a. In Blands (1990) [16] original pathway NOx is removed for the reaction of sodium bicarbonate with SO2 flue gas with NO as described in the reaction equation sequences: equation 2.20 to equation 2.26 to equation 2.27 or equation 2.22 to equation 2.23 to equation 2.26 to equation 2.27. For these paths, NOx removal occurs only when NO2 is formed and NO2 is formed only when SO2 is removed. Also SO2 removal is needed to produce the sodium sulfur oxide reaction intermediate that catalyzes NO2 formation in Blands (1990) [16] original pathway.
Conformity to NOx CR 3b.

Blands (1990) [16] original hypothesis does not conform to NOx CR 3b. As discussed in Section 3.1.3, to conform to NOx CR 3b the NOx removal reaction must either stem from sodium bicarbonate, not sodium carbonate, or it must involve water vapor as a reactant. In Blands (1990) [16] original pathway the NOx removal reaction stems from sodium carbonate, by equation 2.27, and does not occur directly from sodium bicarbonate. Additionally the NOx removal reaction sequence does not involve water vapor as a reactant.
Conformity to NOx CR 3c.

Blands (1990) [16] original pathway does not conform to NOx CR 3c. As discussed in Section 3.1.3 for NOx removal to increase when flue gas humidity is increased, the NOx removal reaction must either stem directly off of sodium or NOx removal must involve

107

water vapor as a reactant. In Blands (1990) [16] original pathway however NOx removal does not stem from sodium bicarbonate but instead stems from sodium carbonate and water vapor is not included as a reactant anywhere in Blands (1990) [16] sequence.
Conformity to NOx CR 3d.

Blands (1990) [16] original pathway conforms to NOx CR 3d since oxygen is required in Blands (1990) [16] original pathway for NO2 formation, equation 2.26. Without oxygen NO2 does not form in Blands (1990) [16] original pathway, and without NO2 formation, NOx is not removed according to equation 2.27. 3.4.2.3 Conformity of Blands (1990) [16] original pathway to the NO2 formation criteria. Table 3.10 lists the conformity of Blands (1990) [16] original hypothesized reaction pathway to the NO2 formation criteria. Table 3-10 Conformity of Blands (1990) pathway to the NO2 formation criteria.
[16]

original

NO2 NO2 NO2 NO2 CriteriaType CR 1 CR 2 CR 3 CR 4 Criteria Number Pathway Blands original C NC NC C C-conforming, NC- non conforming, NA- not applicable

The conformance of Blands (1990) [16] original hypothesis to each of the NO2 formation criteria will be briefly discussed below.
Conformity to NO2 CR 1.

Blands (1990) [16] original hypothesis conforms to NO2 CR 1 since NO2 is formed in Blands (1990) [16] original hypothesis by a catalytic reaction of a sodium sulfur oxide reaction intermediate formed during SO2 removal. If either SO2 or the sodium

108

bi/carbonate are discontinued then this sodium sulfur oxide reaction intermediate will cease to form and as a result NO2 formation will not be catalyzed.
Conformity to NO2 CR 2.

Blands (1990) [16] original hypothesis does not conform to NO2 CR 2 since the NO2 formation reaction in Blands (1990) [16] original hypothesis occurs whenever SO2 removal occurs. NO2 formation is not expected therefore to lag behind SO2 removal, instead NO2 formation is expected to occur simultaneous to SO2 removal when the sodium sulfur oxide reaction intermediate is formed. In addition in Blands (1990) [16] original hypothesis the sodium sulfur oxide reaction intermediate responsible for NO2 formation was expected to form by SO2 removal from either sodium bicarbonate or sodium carbonate. Therefore after injection, when decomposition is occurring and both sodium bicarbonate and carbonate are reacting with SO2, NO2 formation will be catalyzed by the sodium sulfur oxide intermediate formed by either equation 2.20 or 2.23.
Conformity to NO2 CR 3.

Blands (1990) [16] original hypothesis does not conform to NO2 CR 3 since Blands (1990) [16] original hypothesis does not incorporate water vapor as a reactant in any reactions in the pathway. Increased water vapor in the flue gas, besides its role as a reactant, may at low temperature effect the decomposition equilibrium and result in a slow down in the rate of decomposition. But slowing down decomposition in Blands (1990) [16] original pathway does not effect NO2 formation catalyzed by the sodium sulfur oxide reaction intermediate formed in SO2 removal.

109

Conformity to NO2 CR 4.

Blands (1990) [16] original hypothesis conforms to NO2 CR 4 since in Blands (1990) [16] original pathway NO2 does not form when O2 is absent. In particular Blands (1990) [16] NO2 formation reaction equation 2.26 requires oxygen as a reactant for the formation of NO2. 3.4.2.4 Validity and value of Blands (1990) [16] original hypothesized pathway.

Because Blands (1990) [16] original pathway does not conform to all of the criteria described in Sections 3.3.1-3.3.3 it is highly unlikely that it is the actual pathway for NO2 formation. After experimentally testing numerous sodium-sulfur and sodium-nitrogen compounds for their ability to form NO2 Bland (1990) [16] realized that there was no evidence of catalysis of any of the candidate sodium-sulfur oxide compounds for direct conversion of NO and O2 to NO2. Blands (1990) [16] experimental studies including the screening tests indicated that the reaction pathway was not due to a simple single step reaction including a heterogeneous catalysis reaction. Nonetheless Blands (1990) [16] experimental studies are a cornerstone of understanding of sodium bicarbonate reaction chemistry to date. 3.4.3 Conformity of Blands (1990) [16] final hypothesized reaction pathway to the criteria for SO2 removal, NOx removal and NO2 formation. Blands (1990) [16] final hypothesized reaction pathway was described in Section 2.2.5.2. A summary of his pathway is shown in Figure 2.13. Bland (1990) [16] abandoned the original hypothesis that NO2 was formed by heterogeneous catalysis because a suitable sodium sulfur-oxide compound that catalyzed NO to NO2 could not be found. The

110

following is a brief assessment of the conformity of Blands (1990) [16] final hypothesized reaction pathway with respect to the criteria of SO2 removal, NOx removal, and NO2 formation. 3.4.3.1 Conformity of Blands (1990) [16] final pathway to the SO2 removal criteria. Table 3.11 lists the conformity of Blands (1990) [16] final hypothesized reaction pathway to the SO2 removal criteria. Table 3-11 Conformity of Blands (1990) [16] final Pathway to the SO2 Removal Criteria SO2 SO2 SO2 SO2 Criteria Type CR 1 CR 2 CR 3 CR 4 Criteria Number Pathway Blands (1990) [16] final C C C C C-conforming, NC- non conforming, NA- not applicable The conformance to each of the SO2 removal criteria will be briefly discussed below.
Conformity to SO2 CR 1.

Blands (1990) [16] final pathway conforms to SO2 CR 1. In Blands (1990) [16] final hypothesis SO2 would not be catalyzed to SO3 without NO present. Instead SO2 and sodium bicarbonate would form the temperature dependent sodium sulfite and sodium sulfate as established by Howatson et al. (1980)[8] and described by Keener (1982)[9] in equations 2.9 and 2.10.
Conformity to SO2 CR 2.

Blands (1990) [16] final hypothesis conforms to SO2 CR 2 since only sodium sulfate is formed in the pathway when NO is present in the flue gas. Only sodium sulfate is formed because in the pathway all of the flue gas SO2 is catalyzed by NO to SO3; and SO3 reacts

111

with sodium bicarbonate and sodium carbonate to form sodium sulfate as stated by reaction equation 2.32.
Conformity to SO2 CR 3.

Blands (1990) [16] final pathway conforms to SO2 CR 3 since only sodium sulfite accumulates when oxygen is not present. Without oxygen SO2 is not catalyzed to SO3 by reaction equation 2.29. Sodium sulfate therefore is not formed without oxygen since reaction equation 2.32 requires SO3. Instead without oxygen SO2 reacts directly with sodium bicarbonate and sodium carbonate to form sodium sulfite. 3.4.3.2 Conformity of Blands (1990) [16] final pathway to the NOx removal criteria. Table 3.12 lists the conformity of Blands (1990) [16] final hypothesized reaction pathway to the NOx removal criteria. Table 3-12 Conformity of Blands (1990) [16] final Pathway to the NOx removal Criteria
NOx NOx NOx NOx NOx Criteria Type CR 1 CR 2 CR 3a CR 3b CR 3c Criteria number Pathway Blands final C NC C NC NC C-conforming, NC- non conforming, NA- not applicable NOx CR 3d

Conformity to NOx CR 1.

Blands (1990) [16] final hypothesis conforms to NOx CR 1 since Blands (1990) [16] final pathway includes Knights (1980)[18] reaction of NO2 with sodium carbonate according to equation 2.27. Bland (1990) [16] suggested that NOx removal would occur immediately after injection when NO2 would have the opportunity to react with exposed sodium carbonate. Bland (1990) [16] believed however that as soon as all of the exposed sodium

112

carbonate was reacted with NO2 or SO2 that NOx removal would dissipate since the sodium sulfate product layer would build up and prohibit NO2 penetration.
Conformity to NOx CR 2.

Blands (1990) [16] final hypothesis does not conform to NOx CR 2 since in Blands (1990) [16] final hypothesis the reaction of NO with sodium carbonate according to reaction equation 2.27 occurs for a finite time period after injection. Bland (1990) [16] hypothesized that after injection NO would react with exposed sodium carbonate until the sulfation ash layer covered the exposed surface and prohibited penetration by NO. Perhaps Bland (1990) [16] envisioned this finite time period to be negligible for sodium carbonate. Perhaps Bland (1990) [16] thought that the low surface area of sodium carbonate assured that the surface would cover over very quickly to prohibit NO2 penetration and reaction with sodium carbonate beneath. However NO2 penetration would be expected for the sulfation ash layer if SO2 penetration occurs since the NO2 molecule is smaller in radius and volume then the SO2 molecule. Therefore NO2 penetration and reaction would be expected to be ongoing when SO2 reaction proceeds. Sodium nitrate would then be expected to accumulate by this reaction. Blands (1990) [16] final pathway does not include any subsequent reaction converting sodium nitrate formed by this reaction such that it will not accumulate in the product.
Conformity to NOx CR 3a.

Blands (1990) [16] final hypothesis conforms to NOx CR 3a. According to Blands (1990)[16] final pathway sodium bicarbonate injected into flue gas decomposes to sodium carbonate and then reacts with NO2 for NOx removal. Although in Blands (1990) [16] final pathway NO2 formation does not depend on SO2 removal, the reaction of NO2 with

113

sodium carbonate does occur simultaneous with SO2 removal. If SO2 is shut off then NO2 will not form by equation 2.27 and NOx removal will stop.
Conformity to NOx CR 3b.

Blands (1990)[16] final hypothesis does not conform to NOx CR 3b. According to this pathway NOx removal occurs whenever NO2 reacts with sodium carbonate. Bland (1990)[16] explained that the NOx removal reaction would proceed for a finite time period when sodium carbonate was exposed to NO2 for reaction. Evidently Bland (1990)[16] thought that decomposition controlled this time period when fresh sodium carbonate was exposed by the increase in surface area that accompanied decomposition. Bland (1990)
[16]

believed that after the exposed sodium carbonate became covered with a layer of

sodium sulfate ash that NO2 penetration would be inhibited resulting in an end to NO2 reaction and the definition of the finite NOx removal reaction period. As discussed above however, inhibition of NO2 penetration would correspond to the inhibition of SO2 penetration since the NO2 molecule is smaller in radius and volume then the SO2 molecule. Since SO2 reaction proceeds through the ash layer build up with the sodium carbonate beneath, NO2 penetration according to Blands (1990) [16] reactions would be expected to proceed.
Conformity to NOx CR 3c.

Blands (1990) [16] final hypothesis does not conform to NOx CR 3c since the NOx removal reaction does not involve water. As explained for conformity to NOx CR 3b, Bland thought that NOx removal occurred only when sodium carbonate became exposed to NO2. Increasing flue gas humidity would have the effect of slowing down decomposition and extending the decomposition time when NOx removal occurs. This

114

would result in increased NOx removal due to the increased reaction time for NO2 and sodium carbonate reaction.
Conformity to NOx CR 3d.

Blands (1990) [16] final hypothesis conforms to NOx CR 3d since NOx removal in Blands (1990) [16] final hypothesis occurs when oxygen is present. Without oxygen, NO2 does not form since the catalysis of SO2 to SO3 by reaction equation 2.30 requires oxygen. Without oxygen, oxygen free radicals do not form to oxidize NO to NO2. 3.4.3.3 Conformity of Blands (1990) [16] final pathway to the NO2 formation criteria. Table 3.13 lists the conformity of Blands (1990) [16] final hypothesized reaction pathway to the NO2 formation criteria. Table 3-13 Conformity of Blands (1990) [16] final Pathway to the NO2 Formation Criteria
NO2 NO2 NO2 NO2 CriteriaType CR 1 CR 2 CR 3 CR 4 Criteria Number Pathway Blands final NC NC NC C C-conforming, NC- non conforming, NA- not applicable

The conformance of Blands (1990) [16] final hypothesis to each of the NO2 formation criteria is briefly discussed below.
Conformity to NO2 CR 1.

Blands (1990) [16] final hypothesis does not conform to NO2 CR 1 since NO2 formation does not require SO2 removal. In Blands (1990) [16] final pathway NO2 formation occurs in the gas phase by the generation of the oxygen free radical from the catalysis of SO2 to SO3 according to reaction equation 2.30. According to Blands (1990) [16] final pathway

115

it is possible for NO2 to form even when sodium bicarbonate and sodium carbonate are not injected.
Conformity to NO2 CR 2.

Blands (1990)[16] final hypothesis does not conform to NO2 CR 2 since NO2 formation is not expected to lag behind SO2 removal or NOx removal. As described, NO2 formation is hypothesized to occur in the gas phase and is expected to proceed as long as SO2, O2 and NO are available in the gas phase.
Conformity to NO2 CR 3.

Blands (1990) [16] final hypothesis does not conform to NO2 CR 3 since NO2 formation in Blands (1990) [16] final hypothesis is not linked to water vapor or decomposition in any way.
Conformity to NO2 CR 4.

Blands (1990) [16] final hypothesis conforms to NO2 CR 4 since NO2 forms only when O2 is present in the flue gas. In Blands (1990) [16] final hypothesis SO2 does not catalyze to SO3 without oxygen. Also in Blands (1990) [16] final hypothesis the oxygen free radical does not form without oxygen. And hence NO does not oxidize to NO2 without the oxygen free radical. Another problem with Blands (1990) [16] final hypothesis is that a large percentage of NO is expected to oxidize to NO2 rather then a small percentage since a free radical of oxygen is expected to form for every SO2 molecule removed. 3.4.3.4 Validity and value of Blands (1990) [16] final hypothesized pathway.

As shown in Tables 3.16, 3.17 and 3.18 Blands (1990) [16] final hypothesis does not conform to all of the criteria. No feasible modification of Blands (1990) [16] final

116

pathway is apparent to bring it into conformity. And no portions of this hypothesis appear valid for use in assembling the overall reaction pathway. It should be noted that if SO2 were catalyzed to SO3 in the gas phase by NO then SO3 would be detectable in the ducts if monitored. Past investigators however have not reported the presence of SO3 in the ducts. Muzio and Arand (1981)[65] measured SO3 in the ducts prior to the bag-house and at the bag-house exit and found no significant SO3 concentrations in either location. Additionally a number of plausible proportions are unclear. If SO2 were catalyzed to SO3 by NO then the catalysis would need to catalyze all of the SO2 to SO3 to assure that only sodium sulfate forms since only sodium sulfate is found to form according to criteria. Proportionally it is unclear how all of the SO2 could be catalyzed when the flue gas contains a three to one proportion of SO2 to NO concentration. If all of the SO2 were catalyzed to SO3 by the lower concentration of NO it still remains unclear why all of the NO would not likewise be converted to NO2 by free radicals. Recall that only a small portion of the NO is converted to NO2. Bland also believed that the sodium sulfate product layer was responsible for the finite NOx removal period. Bland believed that the sodium sulfate product layer prohibited NO2 diffusion resulting in stoppage of NOx removal. But there is evidence that sodium sulfate product layer does not prohibit NO2 diffusion. In particular SO2 diffusion and reaction with sodium carbonate proceed after the NOx removal stops. Since the molecular radius and volume of NO2 is smaller than SO2, NO2 diffusion would not therefore be prohibited when SO2 diffusion occurs.

117

3.4.4

Conformity of Verlaeten et al.s (1991)[20] hypothesized reaction pathway to the criteria for SO2 removal, NO2 formation, and NOx removal.

Verlaeten et al.s (1991)[20] hypothesized pathway was described in detail in Section 2.2.5.3. Recall that Verlaeten et al. (1991)[20] hypothesized the reaction pathway for NOx removal during SO2 removal. 3.4.4.1 Conformity of Verlaeten et al.s (1991)[20] pathway to the SO2 removal criteria. Table 3.14 lists the conformity of Verlaeten et al.s (1991)[20] hypothesized reaction pathway to the SO2 removal criteria. Table 3-14 Conformity of Verlaeten et al.s (1991)[20] pathway to the SO2 removal criteria. SO2 SO2 SO2 SO2 Criteria Type CR 1 CR 2 CR 3 CR 4 Criteria Number Pathway Verlaeten et al. C NC C NA C-conforming, NC- non conforming, NA- not applicable The removal of SO2 was not included in Verlaeten et al.s (1991)[20] hypothesis. Verlaeten et al.(1991)[20] was primarily concerned with NOx removal reactions. Nevertheless, SO2 removal can be reviewed as it relates to reaction intermediates formed for the NOx removal pathways. The conformance to each of the SO2 removal criteria will be briefly discussed below.
Conformity to SO2 CR 1.

Verlaeten et al.s (1991)[20] pathway conforms to SO2 CR 1 since Verlaeten et al. (1991)[20] included the formation of a temperature dependent mix of sodium sulfite and

118

sodium sulfate when NO was not present in the flue gas according to reaction equations 2.37 and 2.38.
Conformity to SO2 CR 2.

Verlaeten et al.s (1991)[20] pathway does not conform to SO2 CR 2. Sodium sulfate is not hypothesized in Verlaeten et al.s (1991)[20] pathway to be the only product that accumulates when NO is present in the flue gas. In Verlaeten et al.s (1991)[20] pathway sodium sulfite is hypothesized to accumulate by reaction 2.35 and 2.37 for the reaction of sodium bicarbonate and sodium carbonate with SO2 respectively. Evidently Verlaeten et al. (1991)[20] were unaware of Blands (1990)[16] finding that only sodium sulfate accumulates when NO was present. Additionally, since sodium pyrosulfite is stable below 150C, it too would be expected to accumulate as a product below 150C in Verlaeten et al.s (1991)[20] pathway.
Conformity to SO2 CR 3.

Verlaeten et al.s (1991)[20] pathway conforms to SO2 CR 3 above 150C since in Verlaeten et al.s (1991)[20] pathway only sodium sulfite accumulates above 150C when oxygen is not present. But below 150C, as discussed for SO2 CR 1, sodium pyrosulfite is stable and when formed will not convert to sodium sulfite. The presence or absence of oxygen plays no role in the formation of sodium pyrosulfite. Irrespective of oxygen, below 150C, sodium pyrosulfite is expected to form and accumulate by reaction of sodium bicarbonate and SO2 as described in equations 2.33. Verlaeten et al. (1991)[20] however hypothesized that NOx is removed by the pathway for temperatures greater then 150C.

119

Conformity to SO2 CR 4.

This criteria is not applicable to Verlaeten et al.s (1991)[20] pathway since NO2 formation reactions are not included in Verlaeten et al.s (1991)[20] pathway. 3.4.4.2 Conformity of Verlaeten et al.s (1991)[20] pathway to the NOx removal criteria. Table 3.15 lists the conformity of Verlaeten et al.s (1991)[20] hypothesized reaction pathway to the NOx removal criteria. Table 3-15 Conformity of Verlaeten et al.s (1991)[20] pathway to the NOx removal Criteria
NOx NOx NOx NOx NOx Criteria Type CR 1 CR 2 CR 3a CR 3b CR 3c Criteria number Pathway Verlaeten et al.s NA C C C C C-conforming, NC- non conforming, NA- not applicable NOx CR 3d

The conformance to each of the NOx removal criteria will be briefly discussed below.
Conformity to NOx CR 1.

NOx CR 1 is not applicable to Verlaeten et al.s (1991)[20] hypothesized reaction pathway because Verlaeten et al.s (1991)[20] pathway pertains only to NOx removal by sodium bicarbonate and does not include in its scope NOx removal by sodium carbonate. Verlaeten et al.s (1991)[20] pathway does not incorporate Knights (1980)[18] reaction of NO2 with sodium carbonate for the formation of sodium nitrate. Verlaeten et al. (1991)[20] provided no theory on why NO2 does not react as Knight (1980)[18] previously showed. Perhaps Verlaeten et al. (1991)[20] were unaware of Knights (1980)[18] experimental evidence.

120

Conformity to NOx CR 2.

Verlaeten et al.s (1991)[20] hypothesized reaction pathway conforms to NOx CR 2 since Verlaeten et al.s (1991)[20] experimental results served as the basis for NOx CR 2. Recall that Verlaeten et al. (1991)[20] showed that sodium carbonate (dry) when exposed to flue gas with SO2 and NO does not result in any NOx removal despite its ability to form NO2.
Conformity to NOx CR 3a

Verlaeten et al.s (1991)[20] hypothesized reaction pathway conforms to NOx CR 3a above 150C. In Verlaeten et al.s (1991)[20] hypothesized reaction pathway when SO2 is shut off, sodium pyrosulfite ceases to form as a reaction intermediate and thus NOx removal reaction equation 2.34 also ceases. The formation of sodium pyrosulfite occurs only through reaction of sodium bicarbonate with SO2. When sodium bicarbonate injection stops, sodium pyrosulfite formation would be expected to stop also.
Conformity to NOx CR 3b

Verlaeten et al.s (1991)[20] hypothesized reaction pathway conforms to NOx CR 3b since NOx removal by Verlaeten et al.s (1991)[20] pathway occurs only when sodium bicarbonate is present. Sodium bicarbonate is only available during a finite time period after injection prior to its conversion to sodium carbonate.
Conformity to NOx CR 3c.

Verlaeten et al.s hypothesized reaction pathway conforms to NOx CR 3c since as discussed previously flue gas humidity effects the equilibrium of decomposition. Increased flue gas humidity slows down decomposition extending the time period in which sodium bicarbonate remains before converting to sodium carbonate. As a result increased humidity also provides more time for the reaction of sodium bicarbonate and

121

SO2, which translates, to more sodium pyrosulfite production and more NOx removal by reaction equation 2.42.
Conformity to NOx CR 3d.

Verlaeten et al.s (1991)[20] hypothesized reaction pathway conforms to NOx CR 3d since the NOx removal reaction equation 2.34 requires oxygen as a reactant. Accordingly, without oxygen the NOx removal reaction will not occur. 3.4.4.3 Conformity of Verlaeten et al.s (1991)[20] pathway to the NO2 formation criteria. Table 3.16 lists the conformity of Verlaeten et al.s (1991)[20] hypothesized reaction pathway to the NO2 formation criteria. Since Verlaeten et al.s (1991)[20] did not hypothesize NO2 formation reactions the criteria are not applicable. Table 3-16 Conformity of Verlaeten et al.s pathway to the NO2 formation criteria. NO2 NO2 NO2 NO2 CriteriaType CR 1 CR 2 CR 3 CR 4 Criteria Number Pathway Verlaeten et al.s NA NA NA NA C-conforming, NC- non conforming , NA- not applicable 3.4.4.4 Validity and value of Verlaeten et al.s hypothesized pathway.

Verlaeten et al.s (1991)[20] pathway does not conform to all of the SO2 removal, NOx removal, and NO2 formation criteria as shown respectively in Tables 3.19, 3.20, and 3.21. Verlaeten et al.s (1991)[20] pathway does conform however with nearly all of the NOx removal criteria. There remain validity issues regarding Verlaeten et al.s (1991)[20] key reactions in the pathway. Specifically, it remains to be shown that sodium bisulfite is in fact formed by 122

the reaction of sodium bicarbonate and SO2 as Verlaeten et al.s (1991)[20] hypothesized in equation 2.33. Secondly validation of Verlaeten et al.s (1991)[20] hypothesized reaction of sodium pyrosulfite with NO and O2, as described in equation 2.34, is needed. And finally further testing of Verlaeten et al.s (1991)[20] assumption that sodium sulfite is not reactive with NO and O2 is needed to resolve the conflicting experimental findings by other researchers for this reaction.
Validity of sodium bisulfite formation by reaction of sodium bicarbonate and SO2.

There are issues surrounding the validity of Verlaeten et al.s (1991)[20] claim that sodium bisulfite forms as a reaction intermediate. Recall that this reaction intermediate is hypothesized to form by the reaction of sodium bicarbonate and SO2. There is no evidence past or present supporting the validity of equation 2.32. Verlaeten et al. (1991)[20] did not present any experimental data validating their claim that sodium bisulfite formed as a reaction intermediate. According to handbooks, sodium pyrosulfite is stable below 150C where there is no driving force to decompose the sodium pyrosulfite to sodium sulfite via reaction equation 2.35. One would therefore expect to find sodium pyrosulfite in the product formed below 150C. Recall from Section 2.2.1.3, that Howatson et al.(1980)[8] investigated the sulfation product formed by the reaction of sodium bicarbonate and SO2 at low temperatures from 90C to 150C. Howatson et al.(1980)[8] found neither sodium bisulfite nor sodium pyrosulfite from the reaction of sodium bicarbonate with SO2 at any temperature. Instead Howatson et al.(1980)[8] found the formation of sodium sulfite with small quantities of sodium sulfate. Howatson et al.s (1980)[8] analysis included crystallographic x-ray diffraction (XRD) analysis of the sulfation product. This

123

technique is able to distinguish sodium bisulfite, and sodium pyrosulfite from sodium sulfite. Howatson et al.s (1980)[8] reactions were repeated and verified by Bland (1990)[16]. Recall also that Bland (1990)[16] investigated the reaction of sodium bicarbonate with SO2 flue gas with NO. Bland (1990)[16] did not find any formation of sodium bisulfite or sodium pyrosulfite at low temperatures, instead only sodium sulfate was found to form.
Validity of NOx removal reaction by reaction of sodium pyrosulfite with NO and O2.

Recall that Verlaeten et al. (1991)[20] conducted validation experiments for the reaction of sodium pyrosulfite with NO and O2. Verlaeten et al. (1991)[20] reported that NOx removal occurred for this reaction from 129C to 227C. In particular Verlaeten et al. (1991)[20] reported the formation of sodium nitrite, sodium nitrate, and SO2 in the stoichiometric ratio described by the reaction equation 2.34. At issue with these findings is the fact that sodium pyrosulfite decomposes to sodium sulfite at temperatures above 150C according to handbooks. Exposure reactions of sodium pyrosulfite at temperatures above 150C are anticipated to involve the conversion of the sodium pyrosulfite to sodium sulfite and reaction of sodium sulfite with NO and O2. It is highly likely that Verlaeten et al.s (1991)[20] sodium pyrosulfite reactivity tests above 150C that resulted in NOx removal more than likely involved decomposition of sodium pyrosulfite to sodium sulfite and then reaction of the sodium sulfite with NO and O2 comprised of the following elementary reactions: Na2S2O5 Na2SO3 + SO2 Na2SO3 + 2NO + O2 NaNO2 + NaNO3 + SO2 Na2S2O5 + 2NO + O2 NaNO2 + NaNO3 + 2SO2 124 (3.13) (3.14) (3.15)

The above understanding of Verlaeten et al.s (1991)[20] pyrosulfite reactivity tests is consistent with the findings of other researchers. Recall that Miller (1986)[19] as described in Section 2.2.3.1 and Lai and Yesavage (1994)[21] as described in Section 2.2.3.6, found sodium sulfite to be reactive with NO and O2. Specifically, Lai and Yesavage (1994) found sodium sulfite to react with NO and O2 to result in NOx removal.
Validity of the assumption that sodium sulfite is un-reactive with NO and O2.

Question exists on the validity Verlaeten et al.s (1991)[20] hypothesized assumption that sodium sulfite is un-reactive with NO and O2. Although Verlaeten et al. (1991)[20] reported that technical grade sodium sulfite did not react with NO and O2. Their inability to obtain reactivity from technical grade sodium sulfite may not validate their assumption. Their inability to obtain reactivity from technical grade sodium sulfite may simply reflect the fact that technical grade sodium sulfite may suffer from shelf life surface oxidation contamination discussed previously in Section 2.2.5.1. Recall from Sections, 2.2.6.1.1 and 2.2.6.4 that Millers and Lai and Yesavages (1994)[21] sodium sulfite reactivity tests found sodium sulfite reactivity with NO and O2. Millers (1986)[19] and Lai and Yesavages (1994)[21] tests may have found sodium sulfite reactivity with NO and O2 because their samples may have been fresher possessing better surface purity and less surface oxidation compared to the sodium sulfite sample Verlaeten et al. (1991)[20] used. The difficulty in conducting accurate and valid reactivity tests for sodium sulfite is discussed in 2.2.6.1.

125

3.4.5

Conformity of Bortzs hypothesized reaction pathway to the criteria for SO2 removal, NO2 formation, and NOx removal.

Bortzs (1994)[45] hypothesized pathway was described in detail in Section 2.2.5.4. Recall that Bortz (1994)[45] hypothesized the reaction pathway for NOx removal during SO2 removal. 3.4.5.1 Conformity of Bortzs pathway to the SO2 removal criteria.

Table 3.17 lists the conformity of Bortzs (1994)[45] hypothesized reaction pathway to the SO2 removal criteria. Table 3-17 Conformity of Bortzs pathway to the SO2 removal criteria SO2 Criteria Type SO2 SO2 SO2 Criteria Number CR 1 CR 2 CR 3 CR 4 Pathway Bortzs C C C C C-conforming NC- non conforming NA- not applicable The conformance to each of the SO2 removal criteria will be briefly discussed below.
Conformity to SO2 CR 1.

Bortzs (1994)[45] pathway does conform to SO2 CR 1 since without NO a temperature dependent mixture of sodium sulfite and sodium sulfate are proposed to form in Bortzs (1994)[45] pathway.
Conformity to SO2 CR 2.

Bortzs (1994)[45] pathway conforms to SO2 CR 2.

Conformity to SO2 CR 3.

Bortzs (1994)[45] pathway conforms to SO2 CR 3 since without oxygen only sodium sulfite forms. In particular, without oxygen all of the reactions in Bortzs (1994)[45]

126

pathway that form sodium sulfate do not proceed without oxygen. And as discussed under SO2 CR 2 sodium sulfite was formed by the reaction of SO2 with sodium bicarbonate and sodium carbonate without oxygen in Bortzs (1994)[45] pathway.
Conformity to SO2 CR 4.

Bortzs (1994)[45] pathway conforms to SO2 CR 4. However it should be noted that according to reaction equation 2.44 in Bortzs (1994)[45] pathway a portion of SO2 removal is dependent upon NOx removal. Also it should be noted that according to reaction equation 2.41 and 2.43 the stoichiometric ratio of SO2 removal to NO removal is one to two. 3.4.5.2 Conformity of Bortzs pathway to the NOx removal criteria.

Table 3.18 lists the conformity of Bortzs (1994)[45] hypothesized reaction pathway to the NOx removal criteria. Table 3-18 Conformity of Bortzs pathway to removal criteria. NOx NOx NOx NOx NOx Criteria Type CR 1 CR 2 CR 3a CR 3b CR 3c Criteria number Pathway Bortzs C C C C C C-conforming, NC- non conforming, NA- not applicable, the NOx
NOx CR 3d NOx CR 3e

The conformance to each of the NOx removal criteria will be briefly discussed below.
Conformity to NOx CR 1.

Bortzs (1994)[45] hypothesized reaction pathway conforms to NOx CR 1. Bortzs (1994)[45] pathway acknowledges NO2 uptake by sodium carbonate as Knight previously postulated in reaction equation 2.13. This NOx removal reaction however is not considered to be the primary removal mode in Bortzs (1994)[45] pathway, instead reactions 2.41 and 2.43 are considered to be the primary removal reactions. 127

Conformity to NOx CR 2.

Bortzs (1994)[45] hypothesized reaction pathway conforms to NOx CR 2 if the reaction rate of NO2 formation by reaction equation 2.44 matches or exceeds the reaction rate of sodium nitrate formation by reaction equation 2.43. In other words no NOx removal will result for sodium carbonate if all of the sodium nitrate formed is immediately converted to NO2. If however the rate of conversion to NO2 is slower than the rate of sodium nitrate formation, then sodium nitrate accumulation will result and Bortzs (1994)[45] pathway would not conform to NOx CR 2.
Conformity to NOx CR 3a.

Bortzs (1994)[45] hypothesized reaction pathway conforms to NOx CR 3a. As discussed for SO2 CR 3, the NOx removal reactions, equation 2.41 and 2.43, additionally remove SO2. As a result whenever NOx is removed SO2 is also removed.
Conformity to NOx CR 3b.

Bortzs (1994)[45] hypothesized reaction pathway does conform to NOx CR 3b. Bortz (1994)[45] explained that NO2 was formed by a collect-and-release mechanism where sodium nitrate product first was collected and then NO2 was released. The removal of NOx and the formation of NO2 depended upon the relative reaction rates of each reaction equations 2.41 and 2.43 and 2.44 respectively.
Conformity to NOx CR 3c.

Bortzs (1994)[45] hypothesized reaction pathway conforms to NOx CR 3c. Although water vapor is not incorporated as a reactant into the reaction of sodium nitrate it is assumed to be a controlling factor in the rates of reactions for NOx removal and NO2 formation. The exact mechanism was not however provided.

128

Conformity to NOx CR 3d.

Bortzs (1994)[45] hypothesized reaction pathway conforms to NOx CR 3d since oxygen is a necessary reactant in the NOx removal reactions, equations 2.41 and 2.43. Without oxygen these reactions will not proceed and only sodium sulfite will form by reaction equations 2.45 and 2.46.
Conformity to NOx CR 3e

Bortzs (1994)[45] hypothesized reaction pathway conforms to NOx CR 3e. The NOx removal reactions, equations 2.41 and 2.43, both form sodium nitrate as the solid product. No other nitration compound is hypothesized in Bortzs (1994)[45] reaction pathway. 3.4.5.3 Conformity of Bortzs pathway to the NO2 formation criteria.

Table 3.19 lists the conformity of Bortzs (1994)[45] hypothesized reaction pathway to the NO2 formation criteria. Table 3-19 Conformity of Bortzs Pathway to Formation Criteria NO2 CriteriaType NO2 NO2 Criteria Number CR 1 CR 2 CR 3 Pathway Bortzs C C C C-conforming, NC- non conforming, NA- not applicable the NO2 NO2 CR 4 C

The conformance to each of the NO2 formation criteria will be briefly discussed below.
Conformity to NO2 CR 1.

Bortzs (1994)[45] reaction pathway conforms to NO2 CR 1 since NO2 formation depends indirectly on SO2 removal. Bortz (1994)[45] envisioned NO2 formation as a collect-andrelease mechanism of sodium nitrate accumulation. Sodium nitrate was only formed when SO2 was present according to reaction equations 2.41 and 2.43. The release

129

portion of the reaction did not however depend upon SO2 presence. NO2 formation could occur without SO2 presence if sodium nitrate was collected prior.
Conformity to NO2 CR 2.

Bortzs (1994)[45] reaction pathway conforms to NO2 CR 2. Clearly Bortz (1994)[45] was aware that NO2 formation lagged behind SO2 and NOx removal immediately following sodium bicarbonate injection since Bortzs (1994)[45] low temperature injection tests were the tests where this phenomenon was observed and the criteria established. Within Bortzs (1994)[45] reaction pathway the NO2 lag is attributed to the sodium nitrate accumulation step whereby NO2 is released only after sodium nitrate accumulation. It is somewhat unclear why incremental sodium nitrate accumulation is required instead of ongoing accumulation and subsequent release as soon as available.
Conformity to NO2 CR 3.

Bortzs (1994)[45] reaction pathway conforms to NO2 CR 3. Although Bortz (1994)[45] does not provide any clarification on the mechanism, it is assumed that water vapor retards the reaction of sodium nitrate and SO2 (equation 2.44). Bortzs (1994)[45] theory does not provide the exact reason that water vapor inhibits the reaction of sodium nitrate with SO2 is not provided. It can only be speculated that water vapor may interfere with the sodium nitrate making it impossible to in fact react with SO2. It may be speculated that since sodium nitrate is very hygroscopic water vapor may be absorbed onto the sodium nitrate surface occupying reaction sites and prohibiting direct SO2 reaction.

130

Conformity to NO2 CR 4.

Bortzs (1994)[45] reaction pathway conforms to NO2 CR 4. Oxygen was required to form NO2 since oxygen was required for sodium nitrate accumulation according to equation 2.41 and 2.43. 3.4.5.4 Validity and value of Bortzs hypothesized pathway.

Bortzs (1994)[45] hypothesized pathway has great value toward understanding the actual reaction pathway since it conforms to nearly all of the SO2 removal, NOx removal, and NO2 formation criteria as shown respectively in Tables 3.22, 3.23 and 3.24. In particular Bortzs (1994)[45] hypothesized pathway provides the novel idea that that NO2 may be formed as a result of release from a nitration compound. Although Bortzs (1994)[45] theory provides no elementary understanding of the nitration reaction, it does provide insight into the overall general reaction equation for sodium nitrate formation (equations 2.41 and 2.43). There is a need for further investigation into reaction routes for understanding sodium nitrate formation. Bortzs studies documenting the finite period of NOx removal, the associated lag time for NO2 formation, and the effect of water vapor provide insight into the role of water vapor on sodium nitrate accumulation. 3.4.6 Assessment of Lai and Yesavages experimental findings.

The valuable work of Lai and Yesavage (1994)[21] has been downplayed in the literature due to some experimental difficulties they encountered. Although Lai and Yesavage (1994)[21] did not hypothesize their own reaction pathway, they did provide validation studies of reactions hypothesized by Bland (1990) [16] and Verlaeten et al. (1991)[20].

131

Lai and Yesavages (1994)[21] experienced a number of difficulties in their experimental studies. In particular they in their 40-minute long batch style experiments with a stirred rotating propeller reactor they experienced large leakage errors resulting in large measurement errors averaging 20% error. Also, gas-sampling errors for collected samples containing both SO2 and NOx degraded while being queued for gas chromatography (GC) analysis resulting in inaccurate SO2 data measurements for samples containing both SO2 and NOx. Despite these particular problems many of their other test results were accurate and of significance. Their short 5-minute batch reactivity tests discussed in Section 2.2.3.6 did not have large errors but instead had very small leakage errors, less than 1%. Also their gas analysis of reactivity tests of sodium sulfite and sodium pyrosulfite with flue gas with NO were free of GC degradation problems since the gas stream contained only NO without SO2 and samples were analyzed immediately after sample collection without wait. Lai and Yesavages (1994)[21] reactivity results for sodium sulfite with NO and O2 provide further evidence that the sodium sulfite reaction results in NOx removal. Their test results also provide some evidence that SO2 is not evolved during this NOx removal reaction. In Lai and Yesavages (1994)[21] reactivity tests special surface preparation of the samples, namely the crushing and exposing fresh surface area along with special desiccated storage for dryness was employed. Blands reactivity tests did not include this special preparation. It is probable that both Lai and Yesavage (1994)[21] and Millers(1986)[19] found reactivity of sodium sulfite with NO and O2 because their test samples of sodium sulfite were not surface oxidized to sodium sulfate from atmospheric

132

water vapor and O2 in shelf life storage. Verlaeten et al.s theory of NOx removal must be reconsidered in light of Lai and Yesavages (1994)[21] and Millers (1986)[19] findings that sodium sulfite is reactive with NO and O2. Lai and Yesavages (1994)[21] work begins to reveal that sodium pyrosulfite in fact decomposes to sodium sulfite (reaction equation 2.35) and then reacts with NO and O2. Baseline studies of sodium pyrosulfite at temperature with and without NO are needed to further understand the role of sodium pyrosulfite decomposition. Also there is further need to determine if the reaction of sodium bicarbonate with SO2 produces sodium bisulfite or sodium pyrosulfite reaction intermediates. 3.5 Summary of conformity of all of the hypothesized reaction pathways to the

criteria for SO2 removal, NOx removal, and NO2 formation. Table 3.20 summarizes the all of the past hypothesized reaction pathways for conformity to the SO2 removal criteria. Table 3-20 Conformity of all pathways to the SO2 removal criteria.
SO2 SO2 SO2 SO2 Criteria Type CR 1 CR 2 CR 3 CR 4 Criteria Number Pathway Millers NC NC NC C Blands original C NC C C Blands final C C C C Verlaeten et al.s C NC C NA Bortzs NC NC C NC C-conforming, NC- non conforming, NA- not applicable

Table 3.21 summarizes the all of the past hypothesized reaction pathways for conformity to the NOx removal criteria.

133

Table 3-21 Conformity of all pathways to the NOx removal criteria. NOx NOx NOx NOx NOx NOx Criteria Type CR 1 CR 2 CR 3a CR 3b CR 3c CR 3d Criteria number Pathway Millers NA NA NA NA NA NA Blands original C NC C NC NC C Blands final C NC C NC NC C Verlaeten et al.s NC C C C C C Bortzs C C C C C C C-conforming, NC- non conforming, NA- not applicable

NOx CR 3e

NA C C NC C

Table 3.22 summarizes the all of the past hypothesized reaction pathways for conformity to the NO2 formation criteria. Table 3-22 Conformity of all pathways to the NO2 formation criteria NO2 CriteriaType NO2 NO2 NO2 Criteria Number CR 1 CR 2 CR 3 CR 4 Pathway Millers NC NA NC C Blands original C NC NC C Blands final NC NA NA C Verlaeten et al.s NA NA NA NA Bortzs C C C C C-conforming, NC- non conforming, NA- not applicable Recall that for a reaction pathway theory to be theoretically feasible all of the criteria must conform. None of the past researchers overall reaction pathway theories conform to all of the necessary criteria. Each of the past researchers reaction pathway theories possesses valid segments that when used together may form a new hypothesized reaction pathway for study. 3.6 Summary of key controversial reactions and end products Assessing the past hypothesized reaction pathways of past investigators identifies three main controversial reactions and associated end products. These reactions are 134

controversial because conflicting reaction results and end products have been reported for these reactions by various researchers. Further investigation and verification of these three controversial reactions is needed to dispel the controversy surrounding them and to ultimately determine their usefulness within the overall reaction pathway: 1) Controversy exists surrounding the reaction of sodium sulfite with NO and O2 (without SO2). Lai and Yesavage (1994)[21] reported NO removal by sodium sulfite. Lai and Yesavage (1994)[21] found that sodium nitrite and sodium nitrate was formed by the reaction of sodium sulfite with NO and O2 as described in Section 2.2.6.1.4. Bland (1990)[16], on the other hand, reported that sodium sulfite was un-reactive with NO and O2. Bland (1990)[16] reported that no reaction and no reaction products were formed for this reaction as described in Section 2.2.6.1.2. The controversy surrounding this reaction has been attributed to the fact that sodium sulfite is not particularly stable and is highly susceptible to surface oxidation. There is a need for highly controlled reaction studies to be conducted to verify the reactivity of sodium sulfite. Lai and Yesavages (1994)[21] work did not postulate any reaction equations but the plausible Sodium sulfite reactions with NO and O2 reaction equations are listed in equations 3.16-3.18 based on the sodium nitrite and nitrate end products found. Na2SO3 + 2NO + O2 2NaNO2 + SOx Na2SO3 + 2NO + O2 2NaNO3 + SOx Na2SO3 + 2NO + O2 NaNO2 + NaNO3 + SOx 2) Controversy exists surrounding the reaction of sodium nitrate with SO2. Kozlowski et al. (1970)[22] found that gaseous NO2 was produced by this reaction, (3.16) (3.17) (3.18)

135

as discussed in Section 2.2.6.4. But Bland (1990)[16] found no reaction by sodium nitrate with SO2. Bland did however report that sodium nitrite was reactive with SO2. Kozlowski et al.s (1970)[22] work did not postulate a reaction equation but equation 3.19 lists a plausible reaction equation based on the NO2 end product found by Kozlowski et al.s (1970)[22]. NaNO3 + SO2 NO2 + Na2SOx (3.19)

3) Controversy also exists surrounding the reaction of sodium sulfite and flue gas with SO2 and NO. Miller (1986)[19] reported that NO2 was produced by this reaction as described in Section 2.2.6.1.1. Bland (1990)[16], on the other hand, reported no reaction and no reaction products by sodium sulfite with flue gas with SO2 and NO as described in Section 2.2.6.1.2. Miller (1986)[19] did not postulate any reaction pathway equation for this reaction. NO2 formation does however result when the reactions listed in Tables 3.28 and 3.29 are combined. The controversy surrounding the reaction of sodium sulfite with SO2 and NO has been attributed to the fact that sodium sulfite is not stable and is highly susceptible to surface oxidation. There is a need for highly controlled reaction studies to be conducted to verify the reactivity of sodium sulfite. Table 3.23 summarizes all of the above controversial reactants and their end products. Table 3-23 Summary of controversial reactions and respective end products Rxn. Solid Solid End Table Reactant Gas Reactants Products Reference 3.28 Na2SO3 NO, O2 NaNO3, Lai and Yesavage[21] NaNO2, SOx --Na2SO3 SO2, NO, O2 NO2, Na2SOx Miller[19] 3.29 NaNO3 SO2 NO2, Na2SOx Kozlowski et al.[22]

List No. 1 2 3

136

These controversial reactions are essential elements for pathway development since they are critical pathway segments of the hypothesized reaction pathway for NO2 formation. 3.7 Development of a hypothesized reaction pathway

The key undisputed reactions and end products listed in Tables 3.1 through 3.6 along with the key controversial reactions and end products listed in Tables 3.28 through 3.30 equations serve as a foundation for the hypothesized reaction pathway. Figure 3.1 graphically assembles together these parsed reaction segments from the past researchers.

H2 O NaHCO3 Na2CO3 SO2

CO2 Na2SO3 SOx NO,O2 NO2 NaNO3 NO,O2 NO2 SO2 Na2SO4 SO2, O2 SOx SO2 NaNO2 NO2 Na2SOx NO CO2 SOx NO,O2 NaNO2 SO2, O2 SO2 NO2 Na2SOx NO SOx NO,O2 NaNO3 NO2 SO2, O2 Na2SO4 Na2SOx Na2SOx Na2SO4 NaNO3 NO2 O2 Na2SO4

SO2

Na2SO3 H2 O

Figure 3-1 Compilation of essential reaction pathway segments from past researchers. 3.7.1 Conformity of the foundation of the hypothesized reaction pathway.

The reaction pathway shown in Figure 3.1 was assembled from the reaction pathway segments of undisputed past research as summarized in Tables 3.1 through 3.8 and from 137

the key controversial reaction pathway segments as summarized in Tables 3.28 and 3.29. The reaction pathway shown in Figure 3.1 is referred in this development discussion as the foundation of the hypothesized reaction pathway. It is referred to as the foundation of the hypothesized reaction pathway because it incorporates the characteristic reactions essential for the actual reaction pathway. These reactions alone however do not satisfy all of the reaction criteria. A few additional reactions are needed to complete the pathway and serve as the overall reaction pathway. To determine the complete reaction pathway for NO2 formation from the foundation of the hypothesized reaction pathway requires that the foundation of the hypothesized reaction pathway be evaluated for conformity with the criteria established for SO2 removal, NOx removal and NO2 removal (listed in Tables 3.7, 3.8 and 3.9 respectively). This evaluation will identify the additional reactions needed for complete conformity. 3.7.1.1 Conformity of the foundation of the hypothesized reaction pathway to the SO2 removal criteria. Table 3.24 summarizes the conformity of the foundation of the hypothesized reaction pathway to the SO2 removal criteria. Table 3-24 Conformity of the foundation of the hypothesized reaction pathway to the SO2 removal criteria
SO2 SO2 SO2 SO2 Criteria Type CR 1 CR 2 CR 3 CR 4 Criteria Number Pathway hypothesized reaction C NC C NC pathway C-conforming, NC- non conforming, NA- not applicable

138

The SO2 removal criteria that do not conform are discussed briefly below. This evaluation of the non-conformances is used to suggest additional reaction sequences that might bring the pathway it into conformance. Conformity to SO2 CR 2. The foundation of the hypothesized reaction pathway does not conform to SO2 CR 2 since other sulfation products besides sodium sulfate accumulate according to the foundation of the hypothesized reaction pathway. Namely, in the foundation of the hypothesized reaction pathway sodium sulfite accumulates when it is not converted to sodium nitrite or sodium nitrate at a rate equivalent to its formation. Also in the hypothesized reaction pathway, Na2SOx accumulates as an end product. Na2SOx is not designated as Na2SO4. To bring the pathway into conformance with SO2 CR the sodium nitrite conversion reaction equations 3.10-3.12 must be shown to produce specifically sodium sulfate according to reaction equations 3.20-3.22. 2NaNO2 + SO2 2NO + Na2SO4 2NaNO2 + SO2 + O2 2NO2 + Na2SO4 2NaNO2 + SO2 + O2 NO + NO2 + Na2SO4 To assure no sodium sulfite accumulation it must be stipulated that the reaction rate converting sodium sulfite to sodium nitrite and nitrate must be equal to or exceed the reaction rate of sodium sulfite formation. Because the inlet SO2 to NO ratio is generally greater then 2, an additional reaction is needed to assure that all of the sodium sulfite is adequately converted without any sodium sulfite accumulation. Na2SO3 + 2NO2 + O2 2NaNO3 + SO3 (3.23) (3.20) (3.21) (3.22)

139

An inlet NO molecule may convert multiple sodium sulfite molecules since it can partake in revolving cyclic reactions after its release from sodium nitrite or sodium nitrate as either NO or NO2. Conformity to SO2 CR 4. The foundation of the hypothesized reaction pathway does not conform to SO2 CR 4. At low temperature the foundation of the hypothesized reaction pathway is expected to conform to SO2 CR 4 since sodium nitrite is primarily expected to form and NO is expected to be released by reaction with SO2. For this low temperature case, sodium sulfate accumulation occurs in this NO release reaction without NO2 formation or NOx removal. Also at high temperature the foundation of the hypothesized reaction pathway is expected to conform to SO2 CR 4 since sodium sulfate accumulation is expected to occur via sodium sulfite oxidation reaction, which also does not involve NO2 formation or NOx removal. But in the middle temperature range where sodium sulfate formation is expected to occur via the NO2 release reactions from sodium nitrate, the foundation of the hypothesized reaction pathway is not expected to conform to SO2 CR 4 without the additional reaction discussed for SO2 CR 2 equation 3.23. The inclusion of this added reaction permits cyclic chain reactions to occur. With the addition of this reaction it would be possible for an inlet molecule of NO to convert multiple molecules of sodium sulfite. 3.7.1.2 Conformity of the foundation of the hypothesized reaction pathway to the NOx removal criteria. Table 3.25 lists the conformity of the foundation of the hypothesized reaction pathway to the NOx removal criteria. 140

Table 3-25 Conformity of the foundation of the hypothesized reaction pathway to the NOx removal criteria.
NOx NOx NOx NOx NOx Criteria Type CR 1 CR 2 CR 3a CR 3b CR 3c Criteria number Pathway hypothesized reaction C NC NC NC NC pathway C-conforming, NC- non conforming, NA- not applicable NOx CR 3d NOx CR 3e

NC

NC

The NOx removal criteria that do not conform are discussed briefly below. This evaluation of the non-conformances is used to suggest additional reaction sequences that might bring the pathway it into conformance.
Conformity to NOx CR 2.

The foundation of the hypothesized reaction pathway does not conform to NOx CR 2. The foundation of the hypothesized reaction pathway reduces to the reaction of sodium carbonate with SO2 flue gas with NO when all of the reactions involving sodium bicarbonate are negated from the overall pathway while preserving all of the reactions involving sodium carbonate as shown in Figure 3.2.

141

CO2 Na2CO3 SO2 Na2SO3 SOx NO,O2 NO2 NaNO3 NO,O2 NO2 SO2 Na2SO4 SO2, O2 SOx SO2 NaNO2 NO2 Na2SOx NO SO2 Na2SOx Na2SO4 NaNO3 NO2 O2 Na2SO4

Figure 3-2 The reaction pathway of sodium carbonate with SO2 flue gas with NO reduced from the foundation of the hypothesized reaction pathway. For the sodium carbonate reaction pathway shown in Figure 3.2 NOx removal potentially occurs when sodium nitrite or sodium nitrate accumulates. In order for NOx removal not to occur the sodium nitrite and sodium nitrate formed would need to immediately convert to sodium sulfate at a rate in excess of the rate of sodium nitrite and nitrate formation.
Conformity to NOx CR 3a-e.

With the above stated rate stipulation no accumulation of sodium nitrite and sodium nitrate is expected to occur for the reactions involving sodium bicarbonate and SO2 flue gas with NO respectively. Additional reactions are needed in the foundation of the hypothesized reaction pathway to conform to NO CR 3a-e while maintaining the stipulation made for NOx CR 2. Uniquely the reaction of sodium bicarbonate with SO2 flue gas with NO shown in Figure 3.1 differs from the reaction of sodium carbonate with SO2 flue gas with NO shown in Figure 3.2 in two distinct ways: First, the pathway for sodium bicarbonate includes 142

reactions involving the reactant sodium bicarbonate prior to its conversion by decomposition to sodium carbonate, and second, the pathway for sodium bicarbonate evolves water vapor during decomposition. Water vapor is presumed to be the key agent in NOx removal. Water vapor is speculated to be the agent responsible for sodium nitrate accumulation. It is postulated to preserve and protect sodium nitrite and sodium nitrate from converting to NO2 and sodium sulfate. Water vapor is thought to bind to the surface of the sodium nitrite and sodium nitrate shielding it from further reaction with SO2. Water vapor is assumed to bind to the surface of sodium nitrite and nitrate due to hygroscopic binding forces. The water binding reactions are expressed below in equations 3.24 and 3.25. NaNO2 + 4H2O + O2 NaNO34(H2O) NaNO3 + 4H2O NaNO34(H2O) (3.24) (3.25)

It is presumed that sodium nitrite and nitrate are highly reactive with water vapor such that this reaction takes precedence to the sodium nitrite and sodium nitrate reaction with SO2. Sodium nitrate with water vapor fixed to its active sites, NaNO34(H2O), is assumed to be unable to react with SO2. The sodium nitrate fixed with water vapor to its surface will then accumulate as an end product since the water vapor prevents the sodium nitrate from reacting with SO2.
Conformity to NOx CR 3a.

With the additional reactions the foundation of the hypothesized reaction pathway conforms to NOx CR 3a since NOx removal occurs only when SO2 removal occurs since NOx removal depends on the formation of sodium sulfite and generation of water vapor.
Conformity to NOx CR 3b.

143

With the additional reactions, the foundation of the hypothesized reaction pathway conforms to NOx CR 3b since NOx removal occurs when water vapor is evolved from decomposition, which occurs during a finite time period depending on temperature immediately following sodium bicarbonate injection. As the temperature increases from the decomposition temperature, the finite time period becomes reduced since the decomposition rate increases with temperature.
Conformity to NOx CR 3c.

With the additional reactions, the foundation of the hypothesized reaction pathway conforms to NOx CR 3c since flue gas humidity affects decomposition equilibrium. Increased flue gas humidity slows down decomposition providing an increased time period when water is evolved. During this increased time period relatively more sodium nitrite and nitrate form. As a result increased flue gas humidity results in more NOx removal. Also besides water vapor from decomposition, flue gas humidity itself may result in some NOx removal by sodium nitrite and nitrate accumulation from the reaction equations 3.24 and 3.25. Flue gas humidity will result in available water vapor to be bound to any sodium nitrite and sodium nitrate. As a result the amount of sodium nitrate fixed by water vapor for accumulation increases with increasing flue gas humidity.
Conformity to NOx CR 3d.

Also the foundation of the hypothesized reaction pathway conforms to NOx CR 3d since without O2 the sodium nitrite and nitrate ceases to form.

144

Conformity to NOx CR 3e.

Sodium nitrate (with water fixed to its active sites) is the only nitration compound formed by the foundation of the hypothesized reaction pathway. 3.7.1.3 Conformity of the foundation of the hypothesized reaction pathway to the NO2 formation criteria. Table 3.26 lists the conformity of the foundation of the hypothesized reaction pathway to the NO2 formation criteria. Table 3-26 Conformity of the foundation of the hypothesized reaction pathway to the NO2 formation criteria NO2 NO2 NO2 NO2 CriteriaType CR 1 CR 2 CR 3 CR 4 Criteria Number Pathway hypothesized reaction C NC NC C pathway C-conforming, NC- non conforming, NA- not applicable The NO2 formation criteria that do not conform are discussed briefly below. This evaluation of the non-conformances is used to suggest additional reaction sequences that might bring the pathway it into conformance.
Conformity to NO2 CR 2.

The foundation of the hypothesized reaction pathway does not conform to NO2 CR 2 unless the additional reactions in are presumed true. Without the water binding reactions, NO2 would be released immediately following sodium bicarbonate injection. But with the inclusion of the water vapor binding reactions in the reaction pathway set of equations an NO2 lag period occurs following sodium bicarbonate injection. This lag period is assumed to increase with decreasing temperature since decomposition slows with decreasing temperature, extending the time period in which water vapor is evolved. The 145

selection of water vapor as the key agent to NOx removal is consistent with NO2 CR 2 since selecting water vapor rather then the presence of sodium bicarbonate provides a finite lag period in NO2 formation during the time that water vapor is evolved from decomposition.
Conformity to NO2 CR 3.

The foundation of the hypothesized reaction pathway does not conform to NO2 CR 3 since water vapor is not a reactant in any reaction leading to NO2 formation in the pathway. With the inclusion of the NOx removal reactions in the foundation of the hypothesized reaction pathway it would conform to NO2 CR 3 since water vapor would result in a reduction of available sodium nitrite and nitrate for reaction with SO2 for the formation of NO2. A portion of the sodium nitrite and sodium nitrate formed will be converted by flue gas humidity to sodium nitrate with fixed water vapor instead of being converted to NO2. 3.7.2 Assembling the overall hypothesized reaction pathway

With equations 3.20-3.25 added for conformance the foundation of the hypothesized reaction pathway transforms into the overall hypothesized reaction path as shown in Figure 3.3.

146

NO2

NaNO3 NO CO2 H2 O CO2

SO2 NaNO3 x(H2O) O2 NO,O2 or NO2 NaNO3 SO3 SO3 SO2 H2O,O2 NO2 SO2, O2 Na2SO4 H2O

Na2SO4 NO2

H2 O NaHCO3 Na2CO3 SO2

Na2SO3

Na2SO4 NaNO3 x(H2O) NO2 Na2SO4 NaNO3 x(H2O)

CO2 SO3 Na2SO4

NO,O2

NaNO2

NO Na2SO4

CO2 SO2 Na2SO3 H2 O NO,O2

SO3 NaNO2

SO2 H2O,O2 NaNO3 x(H2O) NO2 SO2, O2 Na2SO4

NO Na2SO4

SO3 CO2 SO3 Na2SO4 H2 O SO2, O2 NO,O2 or NO2 NaNO3 H2 O

SO2 NaNO3 x(H2O) NO2 Na2SO4

Figure 3-3 The overall hypothesized reaction path 3.7.2.1 Subsets of the overall hypothesized reaction pathway with respect to temperature. The overall hypothesized reaction pathway is hypothesized to proceed differently depending on temperature. The reaction subsets for sodium bicarbonate and sodium carbonate for the following temperature ranges, T< 50C, 50C < T < 120C, 120C<T<150C, 150C< T < 300C, T > 300C.

147

3.7.2.1.1

Overall hypothesized reaction pathway Subset for T< 50C.

Equations 3.26-3.30 summarize the applicable reaction equations hypothesized reaction pathway for the sodium bicarbonate reaction in SO2 flue gas with NO for the temperature range T< 50C. 2NaHCO3 + SO2 Na2SO3 + 2CO2 + H2O Na2SO3 + 2NO + O2 2NaNO2 + SO3 NaNO2 + O2 + 4H2O NaNO34(H2O) NaHCO3 + SO3 Na2SO4 + CO2 + H2O NaNO2 + SO2 Na2SO4 + NO (3.26) (3.27) (3.28) (3.29) (3.30)

In this low temperature range, T< 50C, sodium bicarbonate does not decompose and therefore remains without conversion to sodium carbonate. The sodium bicarbonate does react with SO2 to form sodium sulfite according to reaction equation 3.26. Sodium sulfite reacts with available NO and O2 to form a temperature dependent mixture of sodium nitrite and sodium nitrate; in this low temperature range it is hypothesized to form primarily sodium nitrite according to reaction equation 3.27. Water vapor evolved by the sulfation reaction 3.26 attaches itself to sodium nitrite according to reaction equation 3.28. Also SO3 evolved by the sodium nitrite formation reaction, 3.27, reacts with available sodium bicarbonate to form more sodium sulfate according to reactions 3.29. The sodium nitrite product that does not have water attached to its surface reacts with SO2 and O2 in the flue gas to form sodium sulfate and NO according to reaction 3.30. Figure 3.4 shows the schematic of the overall hypothesized reaction pathway subset for T< 50C.

148

NaHCO3

CO2 SO2 Na2SO3 H2 O NO,O2

SO3 NaNO2 H2O,O2 NaNO3 x(H2O)

NO SO2 Na2SO4

CO2 SO3 Na2SO4 H2 O

Figure 3-4 Schematic of overall hypothesized reaction pathway subset for T< 50C 3.7.2.1.2 Overall hypothesized reaction pathway, Subset for 50C<T<120C.

Equations 3.31-3.38 summarize the applicable reactions for the sodium bicarbonate reaction in SO2 flue gas with NO for the temperature range 50C < T < 120C. 2NaHCO3 Na2CO3 + H2O + CO2 2NaHCO3 + SO2 Na2SO3 + 2CO2 + H2O Na2CO3 + SO2 Na2SO3 + CO2 Na2SO3 + 2NO + O2 2NaNO2 + SO3 2NaHCO3 + SO3 Na2SO4 + 2CO2 + H2O Na2CO3 + SO3 Na2SO4 + CO2 149 (3.31) (3.32) (3.33) (3.34) (3.35) (3.36)

NaNO2 + O2 + 4H2O NaNO3x(H2O) NaNO2 + SO2 Na2SO4 + NO

(3.37) (3.38)

The temperature range 50C< T <120C is similar to the previous range described except in this range very slow decomposition occurs. Initially sodium bicarbonate slowly begins to decompose to sodium carbonate according to reaction equation 3.31. Sodium carbonate and the sodium bicarbonate react with SO2 to form sodium sulfite according to reaction equations 3.32 and 3.33. This sodium sulfite reacts with NO and O2 to form a sodium nitrite at this low temperature range according to reaction equation 3.34. SO3 is evolved by this reaction, and further reacts with available sodium bicarbonate or sodium carbonate to form sodium sulfate according to reaction equations 3.35 and 3.36. Water evolved from the sodium bicarbonate decomposition reaction 3.31, and from the SO2 removal reactions 3.32 and 3.35 binds to the sodium nitrite to form sodium nitrate with water fixed to it according to reaction equations 3.37. As water attaches to the sodium nitrate particle surface, the sodium bicarbonate decomposition reaction, equation 3.32, becomes inhibited and decomposition is terminated before all of the sodium bicarbonate is converted to sodium carbonate. Despite decomposition termination, sulfation of the non-decomposed sodium bicarbonate continues according to reaction 3.32. And this sodium sulfite reacts with NO and O2 to form sodium nitrite according to reaction equation 3.33. The sodium nitrite that forms may become bound with water vapor from flue gas humidity or from the sulfation reaction 3.32. Recall however that in this case water vapor is not generated by decomposition because decomposition has terminated. The sodium nitrite not bound with water vapor, reacts directly with SO2 and O2 to form NO and sodium sulfate according to reaction equation 3.38. This sodium sulfate product

150

adds ash layer resistance, which also contributes to a slowing down of the reaction. Figure 3.5 shows the schematic of the hypothesized reaction pathway subset for 50C< T <120C.

H2 O NaHCO3 Na2CO3 SO2

CO2 Na2SO3

SO3 CO2 SO3 Na2SO4 NO,O2 NaNO2 H2O,O2 NaNO3 x(H2O)

CO2 SO2 Na2SO3 H2 O NO,O2

SO3 NaNO2 H2O,O2 NaNO3 x(H2O)

NO SO2 Na2SO4

CO2 SO3 Na2SO4 H2 O

Figure 3-5 Schematic of overall hypothesized reaction pathway, Subset for 50C<T<120C 3.7.2.1.3 Overall hypothesized reaction pathway Subset for 120C<T<150.

Equations 3.39-3.56 summarize the applicable reactions for the sodium bicarbonate reaction in SO2 flue gas with NO for the temperature range 120C < T < 150C. 2NaHCO3 Na2CO3 + H2O + CO2 2NaHCO3 + SO2 Na2SO3 + 2CO2 + H2O Na2CO3 + SO2 Na2SO3 + CO2 151 (3.39) (3.40) (3.41)

Na2SO3 + 2NO + O2 2NaNO2 + SO3 Na2SO3 + 2NO + 2O2 2NaNO3 + SO3 NaHCO3 + SO3 Na2SO4 + CO2 + H2O Na2CO3 + SO3 Na2SO4 + CO2 NaNO2 + O2 + 4H2O NaNO3x(H2O) NaNO3 + 4H2O NaNO3x(H2O)
after decomposition ends:

(3.42) (3.43) (3.44) (3.45) (3.46) (3.47)

Na2CO3 + SO2 Na2SO3 + CO2 Na2SO3 + 2NO + O2 2NaNO2 + SO3 Na2SO3 + 2NO + 2O2 2NaNO3 + SO3 Na2CO3 + SO3 Na2SO4 + CO2 2NaNO2 + SO2 Na2SO4 + 2NO 2NaNO2 + SO2 + O2 Na2SO4 + 2NO2 2NaNO3 + SO2 Na2SO4 + 2NO2 Na2SO3 + 2NO2 + O2 2NaNO3 + SO3 Na2CO3 + NO2 2NaNO3 + NO + CO2

(3.48) (3.49) (3.50) (3.51) (3.52) (3.53) (3.54) (3.55) (3.56)

The reactions in the temperature range 120C< T <150C begin with decomposition of the sodium bicarbonate according to reaction equation 3.39. Sodium bicarbonate and sodium carbonate react with SO2 to form sodium sulfite according to reaction equations 3.40 and 3.41. Sodium sulfite reacts with NO and O2 to form a temperature dependent mixture of sodium nitrite and sodium nitrate and evolve SO3 according to reaction equations 3.42 and 3.43. The SO3 evolved reacts with available sodium bicarbonate or

152

sodium carbonate according to reaction equations 3.44 and 3.45. During decomposition when water vapor is evolved sodium nitrite and sodium nitrate react with water vapor to form fixed sodium nitrate according to reaction equations 3.46 and 3.47. Unlike the previous temperature range the layer of fixed sodium nitrate that forms in this temperature range is insufficient in inhibiting decomposition because decomposition forces within this temperature range are much more vigilant. As soon as decomposition is completed the sodium carbonate continues to react with SO2 to form sodium sulfite according to reaction equation 3.48. And sodium sulfite continues to react with NO and O2 to form a temperature dependent mixture of sodium nitrite and sodium nitrate according to reaction equations 3.49 and 3.50. The SO3 evolved reacts with available sodium carbonate according to reaction equation 3.51. When decomposition is complete no water vapor is evolved and the sodium nitrite and nitrate react with SO2 and O2 to release a temperature dependent mixture of NO and NO2 and forms sodium sulfate according to reaction equations 3.52 and 3.53 and 3.54. The NO and NO2 released by reaction equations 3.52 and 3.53 and 3.54, cycles back into the sequence by reacting with available sodium sulfite and sodium carbonate according to reaction equations 3.49, 3.50 and 3.55, 3.56. Since ash layer resistance from non-porous sodium sulfate product inhibits diffusion of reactants, it is speculated to limit the sequence. The schematic of the hypothesized reaction pathway subset for 120C< T <150C is shown in Figure 3-6.

153

NO2

NaNO3 NO CO2 H2O CO2

SO2 NaNO3 x(H2O)

Na2SO4 NO2

H2 O NaHCO3 Na2CO3 SO2

Na2SO3 NO,O2 or NO2 NaNO3 SO3 SO3 SO2 H2O,O2 NO2 SO2, O2 Na2SO4 NO Na2SO4 NaNO3 x(H2O) NO2 SO2, O2 SO3 SO2 H2O NO2 SO2, O2 Na2SO4 NaNO3 x(H2O) Na2SO4 NO Na2SO4 H2O NaNO3 x(H2O) NO2 Na2SO4 NaNO3 x(H2O)

CO2 SO3 Na2SO4

NO,O2

NaNO2

CO2 SO2 Na2SO3 H2O NO,O2

SO3 NaNO2

SO2 H2O,O2

CO2 SO3 Na2SO4 H2 O

NO,O2 or NO2

NaNO3

Figure 3-6 Schematic of overall hypothesized reaction pathway, for 120C<T 150C Figure 3.6 is the same as the overall hypothesized reaction pathway shown in Figure 3.1 except that in this temperature range sodium sulfite is not oxidized to sodium sulfate (equation 3.24). Otherwise all of the reaction equations in the overall hypothesized reaction pathway are involved in the reactions for this temperature range. 3.7.2.1.4 Overall hypothesized reaction pathway Subset for 150C<T<250C.

Equations 3.57-3.76 summarize the applicable reactions for the sodium bicarbonate reaction in SO2 flue gas with NO for the temperature range 150C< T < 250C.

154

During decomposition:

2NaHCO3 Na2CO3 + H2O + CO2 2NaHCO3 + SO2 Na2SO3 + 2CO2 + H2O Na2CO3 + SO2 Na2SO3 + CO2 Na2SO3 + O2 Na2SO4 Na2SO3 + 2NO + O2 2NaNO2 + SO3 Na2SO3 + 2NO + 2O2 2NaNO3 + SO3 2NaHCO3 + SO3 Na2SO4 + 2CO2 + H2O Na2CO3 + SO3 Na2SO4 + CO2 NaNO2 + O2 + 4H2O NaNO3x(H2O) NaNO3 + 4H2O NaNO3x(H2O)
after decomposition ends:

(3.57) (3.58) (3.59) (3.60) (3.61) (3.62) (3.63) (3.64) (3.65) (3.66)

Na2CO3 + SO2 Na2SO3 + CO2 Na2SO3 + O2 Na2SO4 Na2SO3 + 2NO + O2 2NaNO2 + SO3 Na2SO3 + 2NO + 2O2 2NaNO3 + SO3 Na2CO3 + SO3 Na2SO4 + CO2 2NaNO2 + SO2 Na2SO4 + 2NO NaNO2 + SO2 + O2 Na2SO4 + 2NO2 2NaNO3 + SO2 Na2SO4 + 2NO2 Na2SO3 + 2NO2 + O2 2NaNO3 + SO3 Na2CO3 + NO2 2NaNO3 + NO + CO2 155

(3.67) (3.68) (3.69) (3.70) (3.71) (3.72) (3.73) (3.74) (3.75) (3.76)

The reaction equations 3.57-3.76 for the temperature range 150C< T < 250C are very similar to the reaction equations 3.39-3.56 for the temperature range 120C< T <150C. The reaction of SO2 with sodium bicarbonate (or sodium carbonate) in this temperature range however forms a temperature dependent mixture of sodium sulfite and sodium sulfate according to reaction equations 3.58, 3.59 and 3.60 instead of simply forming sodium sulfite. Otherwise all of the reactions in the respective sequences are identical to the reactions in the temperature range 120C< T <150C. The schematic of the overall hypothesized reaction pathway subset for 150C< T < 250C is the same as the overall hypothesized reaction pathway shown in Figure 3-3. Focusing on NO2 formation only, the new hypothesized overall reaction pathway reduces to three essential steps as summarized in the simplified three-step reaction pathway described in Table 3-27. Table 3-27 The new hypothesized simplified three-step reaction pathway for NO2 formation for the overall reaction of sodium bicarbonate and SO2 flue gas with NO at 150C< T < 250C. Description of the reaction step. A sulfation step where sodium sulfite is produced by the reaction of sodium bicarbonate or sodium carbonate with SO2. 2NaHCO3 + SO2 Na2SO3 + 2CO2 + H2O Na2CO3 + SO2 Na2SO3 + CO2 A nitration step where sodium nitrate is produced by reaction of the sodium sulfite from step 1 with NO Na2SO3 + 2NO + 2O2 2NaNO3 + SO3 An NO2 release reaction step where NO2 is released from the collected sodium nitrate by reaction with SO2 in the flue gas. 2NaNO3 + SO2 Na2SO4 + 2NO2

Step Number 1

Equation Number

3.58 and 3.59

3.62

3.74

156

The schematic of the simplified three-step reaction pathway is shown in Figure 3-7.

H 2O NaHCO3 Na2CO3 SO2 Na2SO3 CO2 SO2 H2O NO,O2 or NO2

SO3 NaNO3 SO2

NO2 Na2SO4

Figure 3-7 Schematic diagram of simplified three-step reaction for NO2 formation. This reaction pathway reduces further for pathway subsets 1) without NO, 2) without O2, and 3) without SO2. Without NO the pathway reduces to reaction equations 3.77-3.80: 2NaHCO3 Na2CO3 + H2O + CO2 2NaHCO3 + SO2 Na2SO3 + 2CO2 + H2O Na2CO3 + SO2 Na2SO3 + CO2 Na2SO3 + O2 Na2SO4 (3.77) (3.78) (3.79) (3.80)

These equations do not include any reaction equations involving NO or any byproduct made by reaction with NO. These reactions are the same as the reactions established by Keener (1984), and Howatson et al. (1980) and others for the reaction of sodium bicarbonate with SO2 flue gas without NO where a temperature dependent mixture of sodium sulfite and sodium sulfate forms. Without oxygen for 150C< T < 250C the pathway reduces to reaction equations 3.81-3.83 157

2NaHCO3 Na2CO3 + H2O + CO2 2NaHCO3 + SO2 Na2SO3 + 2CO2 + H2O Na2CO3 + SO2 Na2SO3 + CO2

(3.81) (3.82) (3.83)

This set of reaction equations does not include oxygen. Note that in this set of reactions without oxygen present, the sodium bicarbonate decomposes and reacts with SO2 forming only sodium sulfite. NO2 is not formed in this pathway without O2. Without SO2 in the flue gas the reaction equations for 150C< T < 250C reduces to reaction equation 3.84: 2NaHCO3 Na2CO3 + H2O + CO2 (3.84)

All of the reaction equations that involved SO2 or any byproduct made by reaction with SO2 was eliminated. The only reaction remaining in the pathway is the decomposition of sodium bicarbonate. 3.7.2.1.5 Hypothesized reaction pathway Subset for T > 250C.

Equations 3.85-3.87 summarize the applicable reactions for the sodium bicarbonate reaction in SO2 flue gas with NO for the temperature range T > 250C. 2NaHCO3 Na2CO3 + H2O + CO2 Na2CO3 + SO2 Na2SO3 + CO2 Na2SO3 + O2 Na2SO4 Decomposition in this high temperature range is assumed to be sufficiently fast that sodium bicarbonate does not have an opportunity to react with any of the gas constituents. For this temperature range T > 250C sodium carbonate after decomposition reacts with SO2 and O2 to form sodium sulfate according to the overall reaction of the (3.85) (3.86) (3.87)

158

combined reaction equations 3.85-3.87. Sodium sulfite does not accumulate in this temperature range. The sodium sulfite that forms does not react with NO since at temperatures above 250C sodium sulfite is quickly converted to sodium sulfate according to reaction equation 3.87. The schematic of the hypothesized reaction pathway subset for T > 250C is shown in Figure 3.8.

H2 O NaHCO3 Na2CO3 SO2

CO2 Na2SO3 O2 Na2SO4

Figure 3-8 Schematic of overall hypothesized reaction pathway for T > 250C 3.8 Thermodynamic Evaluation of overall hypothesized reaction pathway.

The key undisputed reactions and end products listed in Tables 3.1 through 3.6 along with the key controversial reactions and end products listed in Tables 3.28 through 3.30 equations serve as a foundation for the hypothesized reaction pathway. Figure 3.8 graphically assembles together these parsed reaction segments from the past researchers.

159

H2 O NaHCO3 Na2CO3 SO2

CO2 Na2SO3 SOx NO,O2 NO2 NaNO3 NO,O2 NO2 SO2 Na2SO4 SO2, O2 SOx SO2 NaNO2 NO2 Na2SOx NO CO2 SOx NO,O2 NaNO2 SO2, O2 SO2 NO2 Na2SOx NO SOx NO,O2 NaNO3 NO2 SO2, O2 Na2SO4 Na2SOx Na2SOx Na2SO4 NaNO3 NO2 O2 Na2SO4

SO2

Na2SO3 H2 O

Figure 3-9 Compilation of essential reaction pathway segments from past researchers.

Equilibrium calculations of each of the proposed reactions in the new hypothesized reaction pathway to determine if the reactions are feasible under equilibrium conditions. The calculations are described and tabulated in detail in Appendix 1. Table 3.28 summarizes the Gibbs Free Energy for each reaction equation with respect to temperature. The reactions proceed to the right hand side of the equation when the Gibbs Free Energy is negative. These equilibrium calculations however do not provide any information on reaction rates.

160

Table 3-28 Summary of Gibbs Free Energy( J/mol) for each reaction equation at temperature (C).
T (C) EQ 3.57 EQ 3.58 EQ 3.59 EQ 3.61 EQ 3.62 EQ 3.72 EQ 3.73 EQ 3.74 EQ 3.75 EQ 3.76 100 13913 -44894 -58807 -98065 -243336 -225532 -288900 -143629 -179968 -25937 125 4442 -53200 -57642 -89784 -228163 -224686 -283471 -145092 -169379 -16962 150 -6335 -62171 -55836 -79638 -209035 -221835 -274927 -145529 -155944 -6641 175 -14291 -69688 -55397 -73595 -198620 -216927 -266668 -141643 -148879 555 200 -23094 -77634 -54540 -66354 -185664 -220811 -266478 -147167 -139998 8598 227 -33551 -86696 -53146 -57185 -168817 -232253 -272706 -161073 -128365 18235 250 300

-52854 -50073 -156985 -232340 -269988 -163076 -119337 27855

-51391 -36515 -132501 -240895 -271029 -175043 -102367 44247

When the Gibbs Free Energy value is negative, the equation is said to proceed to the right hand side under equilibrium conditions. Table 3.28 shows that the values of Gibbs Free Energy are negative for all of the reaction equations evaluated with the exception of particular temperature ranges for reaction equations 3.57 and 3.76. For decomposition of sodium bicarbonate, equation 3.57, the Gibbs free energy values are positive at and below 125C. These findings are consistent with the CRC Handbook of Chemistry and Physics[62] decomposition temperature of 150C for sodium bicarbonate. Sodium bicarbonate decomposition is found to occur in the literature below 150C. It is unclear if decomposition below 150C is non-equilibrium. Additionally this thermodynamic analysis in Table 3-28 shows that under equilibrium conditions sodium carbonate reacts with NO2 to form sodium nitrate, equation 3.76, only when the reaction temperature is less than 150C. This is consistent with the experimental nitration findings of Knight. Recall that Knight found that sodium nitrate formation by equation 3.76 favored low temperature. Despite numerous past investigations as described in Chapter 2 of NO2 formation by the reaction of sodium bicarbonate with SO2 flue gas, the overall reaction pathway has not yet been found by any past investigators to date. The past findings by past investigators 161

reviewed in Chapter 2 have however, revealed a number of key undisputed reactions and end products. From the past findings a set of relevant undisputed reactions and corresponding end products are compiled in this chapter for theoretical development. Knowledge of these individual related reactions are useful in determining the overall reaction sequence for the reaction of sodium bicarbonate with SO2 flue gas with NO. These known particular reactions and end products are assumed to be nested within the overall reaction pathway and serve as a framework. For example, knowledge of the reaction of sodium bicarbonate with SO2 flue gas without NO serves as part of the framework of the overall reaction pathway. Also from the set of relevant undisputed reactions and corresponding end products a list of reaction pathway criteria are established in this chapter. These reaction criteria are used to evaluate the validity of past investigators hypothesized reaction pathways for NO2 formation and NOx removal. In particular the past hypothesized reaction pathway of Miller (1986), Bland (1990), Verlaeten et al. (1991), and Bortz (1994) (described in Chapter 2 Sections 2.2.4.1-2.2.4.3 respectively) are evaluated against the set of reaction criteria established. These evaluations are useful for understanding why each of the past investigators hypothesized reaction pathways is incorrect. Also these evaluations aid in identifying areas for further investigation. The summary set of relevant undisputed reactions and relevant controversial reactions are compiled together to provide the foundation for the new hypothesized reaction pathway development. Missing pathway segments are then added to the foundation to arrive at an overall hypothesized reaction pathway consistent with all of the reaction criteria.

162

3.9

Conclusions of theoretical development. A hypothesized overall reaction pathway was theoretically developed in this chapter.

This hypothesized overall reaction pathway was developed from two main sources; the compiled list, Table 3-1, of key undisputed reactions and corresponding end products collected from past investigators and the controversial reactions, Table 3-23, as detailed in Chapter 2. The controversial reaction of sodium sulfite with NO is included in the new hypothesized reaction pathway. This reaction is considered controversial because there was disagreement on the end products produced by this reaction amongst past investigators. The reaction of sodium sulfite with NO is assumed to form sodium nitrite and sodium nitrate removing NO from the gas phase in the new overall hypothesized reaction pathway developed in this study. This reaction is assumed based on the past work of Lai and Yesavage (1994)[21] combined with the findings of Verlaeten et al. but is considered controversial because Bland (1990)[16] did not find that NO was removed or that sodium nitrite or sodium nitrate was formed by the reaction as described in Section 2.2.6.1. Additionally the controversial reaction of sodium nitrate with SO2 is used in the new hypothesized reaction pathway. This reaction is considered controversial because consistent with Bortzs (1994)[45] release theory, Kozlowski et al. (1970)[22] found that gaseous NO2 and solid sodium sulfate were produced by this reaction, but Bland (1990)[16] nonetheless reported no such formation as discussed in Section 2.2.6.4. Relevant undisputed reactions along with the above-discussed controversial reactions were used for the development of the new overall hypothesized reaction pathway. A

163

nested combination of these sets of reactions becomes the framework of the overall hypothesized reaction pathway Equilibrium thermodynamic calculations of all of the proposed reactions in the new hypothesized reaction pathway for NO2 formation were made to determine feasibility under equilibrium conditions. The thermodynamic analysis, Table 3-27, showed that under equilibrium conditions all of the reactions were feasible for the temperature ranges of interest. In particular all of the key reactions in the simplified three-step reaction pathway for NO2 formation, that were listed in Table 3.28, were shown to be thermodynamically feasible according to the analysis for equilibrium conditions. The two controversial reactions in the simplified reaction pathway are the primary focus of the experimental validation investigations to follow in this study. These controversial reactions are further experimentally investigated and verified within this study.

164

CHAPTER 4

METHODOLOGY 4.1 Introducton This chapter describes the experimental approach and methodologies used in this study. Recall that past investigations of NO2 formation found in the literature were reviewed in detail in Chapter 2. Theoretical development leading to the overall hypothesized reaction path for this study was explained in Chapter 3. 4.2 Experimental approach and scope of study Recall that the overall hypothesized reaction pathway developed in Chapter 3 and shown in Figure 3.3 conforms to all of the required reaction criteria summarized in Tables 3-2 through 3-4. Chapter 3 also showed that all of the reactions in the hypothesized reaction pathway were feasible according to thermodynamic equilibrium analysis. This study seeks to validate the hypothesized reactions in the pathway for NO2 formation as developed in the simplified three-step reaction pathway in Chapter 3, Section 3.5.2.1.5. 4.2.1 The validation status of the reactions in the overall reaction pathway. The validation status of each of the reactions in the overall reaction pathway is listed in Table 4-1.

165

Table 4-1 Reactions and validation status for hypothesized reaction pathway. Eq. 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 4.10 4.11 4.12 4.13 4.14 Reaction 2NaHCO3 Na2CO3 + H2O + CO2 2NaHCO3 + SO2 Na2SO3 + 2CO2 + H2O Na2CO3 + SO2 Na2SO3 + CO2 Na2SO3 + O2 Na2SO4 Na2SO3 + 2NO + O2 2NaNO2 + SO3 Na2SO3 + 2NO + 2O2 2NaNO3 + SO3 Na2SO3 + 2NO2 + O2 2NaNO3 + SO3 NaHCO3 + SO3 Na2SO4 + CO2 + H2O Na2CO3 + SO3 Na2SO4 + CO2 NaNO2 + O2 + 4H2O NaNO34(H2O) NaNO3 + 4H2O NaNO34(H2O) 4NaNO2 + 2SO2 + O2 2Na2SO4 + 2NO2+ 2NO 2NaNO3 + SO2 Na2SO4 + 2NO2 Na2CO3 + NO2 2NaNO3 + NO + CO2 Validation Status Undisputed Undisputed Undisputed Undisputed Controversial Controversial Controversial Undisputed Undisputed New New Undisputed Controversial Undisputed

The undisputed reaction segments, equations 4.1-4.4, 4.8, 4.9, 4.12, and 4.14, in These reactions have already been experimentally validated by past investigators as detailed in Table 3-1. Therefore these reactions are not investigated further in this study. The controversial reaction segments, equations 4.5-4.7, and 4.13, are reactions that have not yet been satisfactorily validated by past investigators. The validation of these reactions remains inconclusive since conflicting past findings exist. The reaction chemistry for sodium bicarbonate decomposition, equation 4.1, in inert nitrogen or in SO2 flue gas without NO has been previously experimentally validated by past investigators such as Erdos (1964) and Keener (1982). However in SO2 flue gas with NO it is unclear why sodium bicarbonate decomposition is inhibited.

166

Sodium bicarbonate decomposition inhibition in SO2 flue gas with NO is an area of interest in this study. Sulfation reactions of sodium bicarbonate and sodium carbonate in SO2 flue gas without NO, equations 4.2-4.4, have been experimentally validated by past investigators such as Howatson et al.(1980)[8] and Bland (1990)[16], and others. These sulfation reactions that form sodium sulfite and sodium sulfate depending upon temperature have been adequately validated and are therefore not investigated any further in this study. The reactions of sodium sulfite with NO and O2, reaction equations 4.5 and 4.6,are controversial. Recall from Chapter 2 Section 2.2.6.1 that conflicting past findings exist regarding this reaction. Both Verlaeten et al.(1990)[20] and Bland (1990)[16] found no nitrate formation by the reaction of technical grade sodium sulfite with NO and O2. Miller (1986)[19] however contended that the purity of the sodium sulfite used in these reactions was an important factor in the outcome of reactivity studies. Miller (1986)[19] pointed out that the sodium sulfite needed to be fresh and uncontaminated. Miller (1986)[19] warned that technical grade sodium sulfite might suffer from surface oxidation contamination due to shelf life storage. Lai and Yesavage (1994)[21] specially prepared sodium sulfite in the laboratory that was free of surface oxidation and showed that this specially prepared sodium sulfite reacted with NO and O2 to produce NOx removal. Lai and Yesavage (1994)[21] in particular showed that NOx removal occurred when the sodium sulfite was prepared by mechanical grinding with dry desiccator or storage, and when the sodium sulfite was formed by fresh formation from reaction of sodium bicarbonate with SO2 sodium sulfite.

167

The controversial reaction of sodium sulfite with NO and O2 is a primary focus of attention in this study. Study of specially prepared sodium sulfite with NO is of interest in this study. In particular, it is of interest to study the role of surface impurities in the reaction of sodium sulfite with NO and O2. This study seeks to determine if NO2 or sodium nitrate are formed as products of this reaction. Recall that Lai and Yesavages (1994)[21] work established that gas phase NOx removal occurred by the reaction but did not analyze the solid product. This study seeks to investigate the solid product from this reaction to verify the formation of sodium nitrite and sodium nitrate. Since the reaction of sodium sulfite with NO2 and O2, equation 4.7, has not been validated, this reaction using specially prepared sodium sulfite is experimentally investigated further in this study. The sulfation reactions of sodium bicarbonate and sodium carbonate with SO3 as described by reaction equations 4.8 and 4.9 are well established and are not investigated further in this study. The hypothesized reactions of sodium nitrite and sodium nitrate with water vapor, equations 4.10 and 4.11, have not previously been suggested by any past researchers. The finite time period of NOx removal following injection, also referred to as the NO2 lag period, noted by Bland (1990)[16] is an area for investigation in this study. NOx removal by tracking conversion of sodium bicarbonate to sodium nitrate is investigated relative to the sodium bicarbonate decomposition. In particular it is of interest to determine if NOx removal by sodium bicarbonate in SO2 flue gas with NO is linked to water vapor evolution from the decomposition of the sodium bicarbonate. The direct reactions of sodium nitrite and sodium nitrate with water vapor, equations 4.10 and 4.11, are not

168

tested in this study because they are related to NOx removal and NOx removal is not the major focus of interest of this study. Direct testing of these reactions are thought to be better suited for study after the reaction pathway for NO2 formation has been clearly established The undisputed reactions of sodium nitrite with SO2 and O2 resulting in the release of a mix of NO and NO2, equation 4.12, has already been experimentally studied by Bland (1990)[16] for the test temperatures of 115C-176C. More study is needed to understand the percent NO and NO2 that are produced by the reaction with respect to temperature however this is not investigated further in this study due to constraints of time and resources. The reaction of sodium nitrate with SO2, equation 4.13, has not been validated. Instead this reaction is controversial because Bland (1990)[16] was unable to experimentally produce NO2 from the reaction of technical grade sodium nitrate with SO2 below 175C. Kozlowski et al. (1970)[22] on the other hand showed NO2 formation and sodium sulfate formation from the reaction of molten (325C -350C) sodium nitrate with SO2. This reaction referred to as a release reaction is a primary area of interest in this study. Special preparations of sodium nitrate other than technical grade sodium nitrate are experimentally investigated to determine the role of surface impurities on reactivity for the temperature range from 110C-250C. Finally the reaction of sodium carbonate with NO2 (equation 4.14) is undisputed. Knight (1980)[18], Bland (1990) [16] and others experimentally studied this reaction extensively. No further study of this reaction is included in this study.

169

4.2.2

The simplified three-step reaction pathway for NO2 formation.

Some of the reactions in the overall reaction pathway are relevant but not directly involved in NO2 formation. These reactions that are not directly relevant to NO2 formation were eliminated from the overall reaction pathway for simplification in Chapter 3 Section 3.5.2.1.5. This simplified reaction pathway involves three essential reactions for NO2 formation. Table 4-2 reiterates the key reactions hypothesized to be responsible for NO2. This simplified reaction pathway is referred to as the three-step reaction pathway for NO2 formation described in Table 4-2. Table 4-2 The simplified three-step reaction pathway for NO2 formation from the reaction of sodium bicarbonate and SO2 flue gas with NO. Step Description of the reaction step. Number 1 A sulfation step where sodium sulfite is produced by the reaction of sodium bicarbonate or sodium carbonate with SO2. 2NaHCO3 + SO2 Na2SO3 + 2CO2 + H2O Na2CO3 + SO2 Na2SO3 + CO2 2 A nitration step where sodium nitrate is produced by reaction of the sodium sulfite from step 1 with NO Na2SO3 + 2NO + 2O2 2NaNO3 + SO3 3 An NO2 release reaction step where NO2 is released from the collected sodium nitrate by reaction with SO2 in the flue gas. 2NaNO3 + SO2 Na2SO4 + 2NO2 4.2.3 Objective and scope of study Equation Number

4.2 and 4.3

4.6

4.14

In an effort to resolve the discrepancies of the controversial reactions and fully validate the overall hypothesized reaction pathway, the controversial reactions are further investigated in this study.

170

The primary goals of this study were to determine the reaction pathway for NO2 formation for the reaction of sodium bicarbonate with SO2 flue gas with NO, and to identify plume discoloration mitigation options based on the reaction pathway. The organization of this study included preliminary tests. The laboratory reactor apparatus constructed for this study was validated to assure that it accurately simulates conditions of sorbent injection in a bag-house in a field application. Peak NO2 was measured from the reactor at a full range of temperatures for the reaction of sodium bicarbonate with SO2 flue gas with NO to compare with the values reported by Bland from a past study discussed in Chapter 2 Section 2.2.4.1 Figure 2-11. After the equipment was validated, this study included preliminary material characterization measurements. Sodium bicarbonate decomposition inhibition in SO2 flue gas with NO was studied. Conversion tracking experiments were used to study why decomposition was inhibited for the reaction of sodium bicarbonate in SO2 flue gas with NO. Conversion tracking experiments were also used to study why sodium nitrate accumulated only during a short finite time period for reactions of sodium bicarbonate in SO2 flue gas with NO. Also conversion tracking was used to study the effect of flue gas water vapor on sodium nitrate accumulation for sodium bicarbonate reaction with SO2 flue gas with NO. Surface area changes, and microstructure changes of the sodium bicarbonate during reaction with SO2 flue gas with NO were evaluated relative to the surface area and microstructure changes for the reaction of sodium bicarbonate with inert nitrogen and with SO2 flue gas without NO. Also the sulfation products from sodium bicarbonate reaction in SO2 flue gas without NO with respect to temperature were identified. Also decomposition of sodium bisulfite and sodium pyrosulfite was studied

171

including material determination of the conversion products. Recall from Chapter 2, Section 2.2.6.2 that Verlaeten et al.s (1990)[20] reported that sodium pyrosulfite reacted with NO and O2 to form sodium nitrite and sodium nitrate. It was of particular interest in this study to investigate the thermal stability of sodium pyrosulfite at the reaction temperatures of those past tests. Experimental studies using the reactor apparatus were conducted to validate two key reactions in the hypothesized reaction pathway for NO2. First, the reaction of sodium sulfite with NO as described by reaction equation 4.6 in the hypothesized reaction pathway was experimentally validated. Recall that it was one of the main reactions in the simplified three-step reaction pathway described in Section 4.1.2. The reaction tests were designed to determine if the reaction resulted in NOx removal or more specifically in sodium nitrate formation. Product material analysis was conducted to identify the product formation. These tests were also intended to determine if NO2 was formed by this reaction. Evidence to refute past hypotheses that stated that sodium sulfite catalytically converted NO to NO2 was sought. Also as a side study, the reaction of sodium sulfite with NO2, reaction equation 4.7, was experimentally investigated for NOx removal. Second, the reaction of sodium nitrate with SO2 as described by reaction equation 4.13 was validated. This study was designed to determine if NO2 was evolved by the reaction. Additionally this investigation was intended to identify the formation of sodium sulfate products. X-ray diffraction analysis was used to identify the sulfation products including the specific polymorphic form of the sulfation products formed.

172

4.3

Apparatus

Two experimental test systems were used for this study- a thermal gravimetric analyzer and an isothermal fixed bed reactor. Both systems are described in detail below. 4.3.1 Thermal gravimetric analyzer. A thermal gravimetric analyzer (TGA) was initially selected for the reactor for all experiments. The TGA was selected because it was envisioned that this reactor type would be capable of monitoring reaction conversions. The Dupont TGA selected for use consisted of a cylindrical quartz tube with an external programmable furnace. Inside of the quartz tube a cantilever style scale was mounted with a sample pan fastened to its end. The quartz tube included a gas inlet and outlet drain allowing gas to pass over the sample pan. Electronically, the scale and thermocouple were able to continuously monitor weight and temperature changes at the pan. The TGA reactor set up was found to be useful to monitor sodium bicarbonate and sodium pyrosulfite decomposition under inert gas conditions. Weight losses were accurately monitored with respect to temperature for these decomposition reactions with inert gas. See Chapter 5, Section 5.3.5 for the results from these investigations. The TGA set up was not found suitable for monitoring reactions of sodium bicarbonate with SO2 flue gas with and without NO. The pan style reactor was unable to accurately simulate the gas solid contact reactions of the bag-house. In particular, in the TGA reactor flue gas passed over the TGA sample pan instead of being forced through and around the sorbent as would be the case in the bag-house at the filter where the dustcake was positioned. Preliminary TGA trials indicated very low conversions. The TGA set up proved to be gas phase limited. Additionally the TGA was unable to distinguish 173

conversion of sodium bicarbonate to sodium nitrate from conversion to sodium carbonate since both compounds possessed similar molecular weights, 84 grams/mole versus 84.98 grams per mole respectively. As a result the TGA reactor was not used as the primary reactor for study. 4.3.2 Isothermal fixed bed reactor.

To simulate the bag-house reactions an isothermal fixed bed reactor was constructed for use. This type of reactor was selected because it provided an isothermal reaction zone and because it simulated the gas solid contact of a bag-house filter. The fixed bed configuration provided superior gas solid mixing compared to the boat style TGA where the reactant gas passed only over the top surface of the sample pile. The fixed bed reactor shown in Figure 4-2 was designed and built with the following components: a flue gas metering station; a gas pre-heater; a powder injection module; a main isothermal reaction zone; a quick quench transfer zone for sample removal; a gas analyzer station for SO2 and NOx monitoring; a data acquisition system.

174

300C

300 C

300C

300 C

Powder Injection Module

N2

Data Acquisition Computer

RS232 Water cooling Heater and Insulation Joined Connection

Heater Controllers

Gas Preheater Zone Chilled N2

Mesh Screen Bed Support Valve Thermocouple Electrical wire

Heated SO2 Expansion Chamber Rotometers

MFC MFC

Isothermal Zone

Drain to hood Bridge Powder Removal

SO2 NOx NO

CO2

O2

N2

NO

SO2

Isothermal Fixed Bed

N2

Flue Gas Mixing Station

Quick Quench Bed

Gas Analyzer Station To Gas Analyzers

Figure 4-1 Apparatus

175

4.3.2.1

The flue gas mixing station

A flue gas mixing station was included in the reactor to meter specified concentrations of N2, CO2, O2, SO2, and NO flows. This mixing station permitted different flue gas mixtures to be studied. The N2, CO2, O2, SO2, and NO were individually supplied from high purity compressed gas tanks. The gasses were dispensed individually using two stage regulators on each compressed tank. A supply pressure of 20 pounds per square (psi) inch was used on the N2, CO2, and O2 tanks and transferred to metering devices through inch diameter flexible polyethylene tubing. The pure SO2 and NO were metered with a supply pressure of 25 psi using special corrosion resistant stainless steel and non-reactive silicone inch tubing. These lines used to transfer the SO2 and NO were specially cleaned with alcohol, purged and dried with dry nitrogen to avoid the formation of a sticky substance on the inner tube walls. Additionally a heated (70C) expansion chamber was installed prior to the SO2 metering device to assure gas phase rather then liquid phase SO2 transfer into the SO2 metering device and to dampen out pressure surges from regulator breathing. Separate volumetric metering devices for N2, CO2, O2, SO2, NO, NO2, and H2O were installed to provide flexible tailoring of desired flue gas mixtures for each experiment. Precision glass tube needle valve rotometers from Porter Instrument Inc. were employed to meter the N2, CO2, and O2 flows. The three glass tube rotometers were calibrated by a bubble meter. The calibration charts for the rotometer settings versus the actual flow rates delivered are shown respectively in Figures 4.2 through 4.4.

176

Calibration for Nitrogen Rotometer

2500 Actual N2 Flow Rate (cc/min)

2000

1500 Regulator @ 20 psi Porter B-125-40-Sap Room temperature Bubble meter w ith stop-w atch Y = 148.5 X - 181.0

1000

500

0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Rotom eter setting

Figure 4-2 Nitrogen rotometer calibration curve Y=148.5*X 181 Figure 4-3 shows that the N2 rotometer was linear between 100 and 2100 cc/min.
Calibration for Carbon Dioxide Rotometer
250 Actual CO2 Flow Rate (cc/min)

200

150 Regulator @ 20 psi Porter B-125-10-SS Room temperature Bubble meter w ith stop-w atch Y = 30.60 X - 87.37

100

50

0 0 1 2 3 4 5 6 7 8 9 10 Rotom eter setting

Figure 4-3 Carbon dioxide rotometer calibration curve. Y=30.6*X - 87.4 Figure 4-4 shows that the CO2 rotometer was found to be linear from 65 to 230 cc/min. 177

Calibration for Oxygen Rotometer

80 Actual Oxygen Flow Rate (cc/min) 70 60 50 40 30 20 10 0 0 2 4 6 8 10 12 14 16 Rotom eter setting Regulator @ 20 psi Porter B-125-6-G Room temperature Bubble meter w ith stop-w atch Y = 6.93 x - 25.13

Figure 4-4 Oxygen rotometer calibration curve. Y= 6.93*X - 25.13 Figure 4-5 shows that the oxygen rotometer was found to be linear from 12 to 75 cc/min. Precision electronic mass flow controllers by Porter Instruments were employed to meter the very low SO2 and NO flows. Both of the mass flow controllers were factory calibrated as shown respectively in Figure 4-5 and 4-6.

178

SO2 Mass Flow Controller (MFC) Calibration

10 9 Actual SO2 Flow Rate (cc/min) 8 7 6 5 4 3 2 1 0 0 10 20 30 40 50 60 70 80 90 100 MFC Setting (% full scale)

Porter Factory calibrated MFC Y=0.01*X

Figure 4-5 SO2 Mass flow controller calibration curve Y= 0.01*X

NO Mass Flow Controller (MFC) Calibration
5 4.5 Actual NO Flow Rate (cc/min) 4 3.5 3 2.5 2 1.5 1 0.5 0 0 10 20 30 40 50 60 70 80 90 100 MFC Setting (% full scale)

Porter Factory calibrated MFC Y=0.05*X

Figure 4-6 NO mass flow controller calibration curve Y= 0.05*X

179

Figure 4-6 shows that the SO2 mass flow controller delivered 10cc/min set at full scale, and Figure 4-7 shows that the NO mass flow controller delivered 5cc/min full scale. Directly combining metered NO with O2 using a tee joint and 3 length of tube for mixing made NO2. Selectively combining NO with O2 at room temperature without the presence of any third body molecules was experimentally shown to convert all NO to NO2. Table 4.3 lists the percent concentrations of the flue gas mixtures used in this study. SO2 flue gas with NO and with humidity was made by bubbling a premixed gas mixture of SO2 flue gas with NO through a water bath with precise heat control. In particular 5% H2O vapor was achieved by sparging the flue gas mixture through a heated water bath. The water vapor concentration was measured by a mass balance method where the bath volume change was measured per volumetric gas flow. Table 4-3 Flue gas mixtures used in experiments Mixture ID Inert flow Flue gas without SO2 and without NO SO2 flue gas without NO SO2 flue gas with NO SO2 flue gas with NO2 SO2 flue gas with NO without O2 Inert Flow with humidity SO2 flue gas with NO with humidity CO2 (%) 0 14 14 14 14 14 0 14 O2 (%) 0 3 3 3 3 0 0 3 N2 (%) 100 83 83 83 83 86 95 78 SO2 (%) 0 0 0.3 0.3 0.3 0.3 0 0.186 NO (%) 0 0 0 0.1 0 0.1 0 0.25 NO2 (%) 0 0 0 0 0.1 0 0 0 H2O (%) 0 0 0 0 0 0 5 5

180

To simulate the laminar bag-house flow conditions an air-to-cloth ratio of approximately 1386 cc/min was used. Table 4-4 lists the flow rates, which were respectively metered for each gas in the flue gas mixtures studied. Table 4-4 Flow (cc/ min) 1386 1386 Flue gas mixture flow rates used in experiments NO2 CO2 O2 N2 SO2 NO (cc/ (cc/ (cc/ (cc/ (cc/ (cc/ min) min) min) min) min) min) 0 0 1386 0 0 0 197 44 1145 0 0 0

Mixture ID Inert N2 Flue gas without SO2 and without NO SO2 flue gas without NO SO2 flue gas with NO SO2 flue gas with NO2 SO2 flue gas with NO without O2 N2 with NO N2 with NO2 Inert N2 with humidity SO2 flue gas with NO with humidity

H2O (cc/ min) 0 0

1386 1386 1386 1386

197 197 197 197

44 44 44 0

1140 1139 1139 1183

4.1 4.1 4.1 4.1

0 1.3 0 1.3

0 0 1.3 0

0 0 0 0

1386 1386 1386 1386

0 0 0 194

0 0 0 42

1385 1385 1317 1075

0 0 0 2.6

1.3 0 0 3.5

0 1.3 0 0

0 0 69 69

4.3.2.2

The gas pre-heater.

The gas pre-heater was designed to pre-heat the incoming flue gas mixture. This stage was required to maintain stable temperatures in the reaction zone. It was constructed out of 0.75 diameter stainless steel tubing and extended horizontally 6 inches in length into the main reactor. The pre-heat stage was heated with a wrapped ceramic woven heat tape around the outside of the stainless steel tubing. The heated stage was insulated with a 181

wrap of a 2-inch thick fiber spun ceramic insulation blanket and a outer wrap of ceramic cloth resembling an ace bandage. The heat tapes were plugged into a receptacle with a power supply controlled by a heater controller driven by an externally mounted K type thermocouple. Table 4-5 provides the settings used in the pre-heater to obtain the desired test temperatures. Table 4-5 Pre-heater control settings for gas test temperatures Gas Temperature Pre-heater Control Setting (C) 111 2.0 150 2.5 175 3.0 200 3.5 250 4.0 300 4.5

4.3.2.3

Powder injection module.

A sorbent powder injection module was installed in the system to provide a controlled vestibule for sorbent queuing. The injection module was designed to maintain sorbent powder at room temperature and to prohibit airflow into the main reaction chamber during tests. The sealed style injection module also prohibited the toxic flue gases from escaping into the laboratory during test. The injection module was constructed out of stainless steel tubing, fittings and valves. It consisted of a short vertically oriented diameter tube chamber with two diameter manually operated swivel ball valves. The ball valves were mounted at opposite ends of the load lock chamber. The sorbent powder was poured into the load lock by opening the top ball valve while the bottom ball valve was kept shut. Once the powder was loaded into the lock, the top of the lock was closed with a cap and the holding chamber was purged and back filled with nitrogen through the 182

cap. The powder was injected into the main isothermal fixed bed reactor by opening the bottom hatch ball valve and allowing the powder to drop by gravity onto the mesh screen below. Additionally the injection module was repeatedly tapped with a wrench during loading to assure that all powder from the holding chamber was dislodged. Because the injection module butted up against the top of the main heated reactor zone, the load lock chamber of the injection module needed to be cooled. A copper tube wrap that circulated chilled water was used to cool the injection module. Tests were conducted to determine the time required for the bed zone to reach steady state temperature of 150C after 0.3 grams of sodium bicarbonate was injected into the fixed bed from room temperature. The bed decreased to 143C in 1 second after injection and returned to150C steady state after 10 seconds. 4.3.2.4 The main isothermal reaction zone.

The main isothermal reaction zone of the reactor was designed to pass the flue gas mixture through the bed of injected powder at isothermal conditions. It was constructed out of four vertically mounted pieces of stainless steel tubing (2.5, 6, 2 and 4). The top of the 2.5 tube was connected to the injection module discussed in Section 4.3.2.3 and the bottom of the 2.5 tube was connected to a diameter tee joint fitting positioned sideways. The bottom of this tee joint the 6 long tube was vertically connected. And to the bottom of the 6 long tube another diameter tee joint fitting was attached. To the bottom of this tee joint the 2 long tube was vertically connected. And to the bottom of the 2 long tube a union connector connected it to a tee joint. Between the bottom of the 2 long tube and the union connector a durable stainless steel mesh screen (325x 2300 mesh) was stretched and wedged across horizontally with the 183

fitting feral at the bottom of the joint. The mesh screen was designed to serve as the supported bed for the injected powder; it was intended to simulate a fabric bag in a baghouse. At the bottom of the tee a thermocouple was fed from the outside to the inside through an end cap. The thermal couple was fed through to probe the temperature at the bottom side of the mesh screen inside the reactor. The side leg of this bottom most tee joint fitting directed the gas out of the reactor and to the gas analyzers. Additionally an inlet port with a valve was attached to provide nitrogen carrier gas for transferring the powder to the quick quench removal zone. During all experiments the drain valve to the analyzers was left opened and the nitrogen carrier gas valve was kept closed. Only after reaction experiments were completed were these valve positions changed to allow a high velocity pulse of nitrogen carrier gas through the bed to entrain and transfer the reacted powder from the mesh screen in the isothermal reaction zone to the quick quench sample removal zone. Two ceramic woven heat tapes heated the vertical tube assembly. Insulated with a 4 thick fiber spun ceramic insulation blanket wrapped tightly with ceramic weave fabric in a bandage fashion. The heat tapes were powered separately by two thermal controllers driven by shielded thermal couples mounted to the outer diameter of the reactor tubes. These thermal couples were attached to the upper and middle tee joints. The tee joints were fabricated out of thick wall stainless steel. In an effort to avoid electrical and thermal signal instability and to avoid heating tape shorting the heating tapes were not wrapped directly on the thermal couple tips. To obtain the desired isothermal temperatures the heaters were calibrated by a thermal couple that was temporarily inserted into the reactor and slid along the vertical

184

reactor column. The temperature was probed from the injection point down to the supported bed. The calibration curves for the two heaters were experimentally obtained and are shown in Figure 4.7 and 4.8.

Calibration of Reaction Zone Heater Controller # 1

250

Actual Temperature Inside Reactor (C)

200

150

100

50

0 0 50 100 150 200 250 300 350 400 450

Y= 0.524114*X - 21.3177

Controller Setting (C)

Figure 4-7 Calibration curve of heater controller 1 Y=0.524114- 21.3177

185

Calibration of Reaction Zone Heater Controller #2

250

Actual Temperature Inside Reactor (C)

200

150

100

50

0 0 50 100 150 200 250 300 350 400 450 Controller Setting (C)

Y= 0.595918*X - 18.5727

Figure 4-8 Calibration curve of heater controller 2 Y= 0.595918-18.5727 The most commonly used settings in the experiments are listed in Table 4-6. Table 4-6 Heater control settings for test gas temperatures Gas Temperature Heater #1 Heater #2 Control Setting Control Setting (C) 111 252 217 150 327 283 175 375 325 200 422 367 250 518 451 300 613 535 4.3.2.5 The quick-quench transfer zone for sample removal. The quick-quench zone was designed to quickly cool and purge the reacted sample and to provide easy withdrawal of the reacted powder after test. The quickquench zone was attached to the main isothermal reactor via a horizontal diameter bridge tube with a central swivel shut off ball valve. The bridge was attached between the main isothermal reactor and the quick-quench zone. 186

The quick-quench zone was constructed out of stainless steel tubing and vertically mounted. To the bottom of the bridge tee joint a 2 long tube was vertically connected. And to the bottom of the 2 long tube a union connector connected it to another tee joint. Between the bottom of this 2 long tube and the union connector another durable stainless steel mesh screen (325x 2300 mesh) was stretched and wedged horizontally at the fitting feral position at the bottom of the joint. The mesh screen was designed to catch the reacted powder transferred from the main reactor. The bottom of the union was connected to another tee joint with a thermocouple feed through in the end cap. The thermal couple was fed through to probe the temperature at the bottom side of the quick quench zone. The gas was directed out the side leg of this tee joint to the laboratory hood for neutralization and discharge. At the top of quick-quench zone an inlet port was connected for chilled nitrogen gas to flow. The nitrogen gas used for quenching was dry and chilled by copper tubing submerged in a liquid nitrogen cooled methanol bath maintained at -37C. The entire quick-quench assembly including part of the bridge was wrapped with copper tubing with circulating chilled water to assure that heat transfer from the main reaction zone did not cause the quick-quench zone to heat. The water-cooling coils were arranged such that the quick-quench zone could be opened easily to remove the reacted powder for analysis. Sample powder was entrained and transferred from the main isothermal reaction zone via a high pressure, high velocity pulse of carrier nitrogen gas into the quick-quench zone. The transfer was made by opening the bridge ball valve, and opening the quickquench drain valve allowing the gas pulse to travel into the quick-quench zone out the drain. The mesh screen in the quick-quench zone collected the sample transfer. Once the

187

transfer was made the bridge ball valve was quickly closed and the chilled nitrogen was turned on and allowed to drain out the bottom of the quick-quench zone, cooling the collected powder in the process. After the powder was sufficiently cooled from the reaction temperature and thawed to room temperature to avoid water condensation upon opening to the air. The quick-quench chamber was unlocked and the powder was removed for product analysis. The quenching procedure was primarily employed for reaction tracking experiments where the reaction was monitored with respect to time. The quick-quench zone provided a means of removing the sample without having to wait hours for the isothermal reaction zone to cool before opening up to the air. The quickquench zone avoided extraneous reactions which could occur from removing the hot sorbent in air by providing cooling under inert nitrogen gas conditions. 4.3.2.6 The gas analyzer station for SO2 and NOx monitoring.

The gas analyzer station was designed to analyze gas flow for NOx and SO2 content. The gas analyzer station was capable of analyzing gas from either the reactor inlet or outlet. The gas analyzer station consisted of two chemiluminescent NOx analyzers serially connected to a SO2 analyzer. One of the NOx analyzers was used to monitor NO and the other NOx analyzer was used to monitor total NOx. NO2 was calculated by subtraction, [NOx] - [NO] = [NO2] The chemiluminescent NOx analyzers used in this project were very old (vintage 1965) and required considerable refurbishment prior to use. In particular, the numerous gas tubing connections within each unit were leak tested, repaired and tightened wherever appropriate. Internally the ozone producing devices were rebuilt including the replacement of the oxygen inlet pressure regulators. Internally, the NOx converter 188

canisters were also rebuilt replacing heaters, thermocouples, controllers, power supplies and insulation to assure a steady converter temperature of 650C. Internally, all electronic switching solenoids for gas flows were replaced with manual on-off ball valves. The vacuum pumps were also rebuilt and refurbished to eliminate signal noise to obtain steady pumping speed and 25 inches of Hg pressure. Shims were installed in the pumps to fix the effects of wear. The high gain and low suppression used on the NOx analyzer photo multiplier tubes (PMT) were set to HI and LO respectively to compensate for tube life wear. Also for each PMT the dark current was compensated (without ozone flowing) after resetting the PMT gain and suppression levels. The NOx analyzers did not possess any automatic calibration components. After gas analysis the gas was directed to the laboratory hood for neutralization and discharge. 4.3.2.7 The data acquisition computer system.

NOx, the NO, and the SO2 gas analysis data from the analyzers was automatically recorded by a computer RS 232 interface. Strawberry Tree hardware and software were used to acquire and digitally store analog signals of the NOx, NO and SO2. The NO and NOx signals were taken as voltage readings and signal processed with a multiplication factor of 1000. A 1000 ppm NOx analyzer signal produced 1 volt and Strawberry Tree recorded this 1 volt signal as a value of 1000. Because the NOx analyzers did not possess any automatic calibration components, the NO and NOx data collected from each test was manually recalculated in a spreadsheet after each test to account for the calibration factors obtained through manual calibration.

189

4.3.3

Precision and accuracy The gas flow rotometers were calibrated against bubble meter readings for each

experiment. The heated zones were calibrated to internally mounted thermal couples for each experiment. The thermal couples were calibrated to ice and boiling water for each experiment. The NOx signal scatter was measured as + 15ppm for a calibration gas. NOx and SO2 analyzers were calibrated with premixed span gas for each experiment. The NOx analyzers were checked to assure that CO concentration did not produce any phantom signals. No phantom signals were found when special CO calibration gas was directed through the NOx analyzers. Table 4-7 Gas flow meter precision and accuracy Repeatability Cv Flow Rate Accuracy ( + %) (cc/min) ( + %) 6.3x10-3 1.1x10-3 --1.6x10-4 3.6x10-5 197.2 44.3 1140.6 4.1 1.3 Total 1386 5 5 5 1 1 0.25 0.25 0.25 -----

Gas

Drift ( + %) 0.5 0.5 0.5 -----

CO2 O2 N2 SO2 NO

4.4 4.4.1

Analytical methods Elemental analysis

A LECO 932 CHNS Elemental Determinator was used for elemental analysis of hydrogen, nitrogen and sulfur. This analytical device determined the elemental analysis of the product by standardized gas analysis method. In particular, a small, 2 mg, sample, was automatically combusted at 1000C in an oxygen rich flow sample. Non-dispersive infrared analysis of the gas was automatically conducted internal to the analyzer

190

illuminating the gas with an IR source, from a heated NiCr wire (850C), through a window in the duct and detecting any absorbency from H2O for hydrogen analysis, and SO2 for sulfur analysis. A thermal conductivity, whetstone bridge, device within the LECO 932 analyzer served to measure nitrogen gas concentrations for nitrogen analysis (Note thermal NOx was not formed in the combustion column. The LECO CHNS 932 was computerized and automated requiring the operator to input only the sample weight into the computer, and load the encapsulated sample into the instrument. The instrument was calibrated with a single point during each session by running several duplicates of a NIST certified standard of sulfamethazone. Complete calibration curves of the LECO CHNS analyzer were experimentally obtained by measuring the hydrogen, nitrogen, and sulfur content of other NIST certified standards while maintaining the sulfamethazone calibration point. Sample size was varied from 2 mg to obtain denominations of the certified values. Table 4-8 lists the NIST standards used. And the calibration curves obtained by this process are shown in Figure 4.9. Table 4-8 analyzer Name of Standard 3,5 dinitro sulfamethazone cystine soil #9 soil #162 soil #8 CaCO3 NIST standards used to calibrate elemental Carbon (wt%) 9.91 -29.99 2.866 1.040 13.61 12 Hydrogen (wt%) 0.475 5.07 5.03 0.776 0.103 1.89 -Nitrogen (wt%) 3.3025 20.13 11.66 0.150 -0.729 -Sulfur (wt%) -11.52 26.68 -----

191

Calibration Curve for Hydrogen

1.8 1.6 Actual Hydrogen (wt%) 1.4 1.2 1 0.8 0.6 0.4 0.2 0 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 Measured Hydrogen (w t%)

Instrument calibrated at H= 5.07

Y=.999*X - 0.04

Calibration Curve for Nitrogen

25

Instrument calibrated at N=20.13

20 Actual Nitrogen (wt%)

15

10

Y=0.964*X - 0.001

0 0 5 10 15 20 25 Measured NItrogen (w t%)

Calibration Curve for Sulfur

30.000 25.000 Actual Sulfur (wt%) 20.000 15.000 10.000 5.000 0.000 0.000

Instrument calibrated at S=11.52

for S<2: Y=0.043 + 5.262*X -2.458*X^2 for 2<S<12: Y=1.923 + 1.029*X -0.017*X^2

5.000

10.000

15.000

20.000

25.000

Measured Sulfur (w t%)

Figure 4-9 Calibration curves for elemental analysis. 192

The calibration curves in Figure 4.9 show that the hydrogen and nitrogen channels were very linear and required no adjustment after the instrument was calibrated to the single point. The sulfur calibration curve however shows that the sulfur channel required post data collection adjustment according to the curve fitting equations obtained. For sulfur values measuring less then 2.00% the following quadratic adjustment was applied to the LECO measured values: Y=0.043 + 5.262*X -2.458*X2 where Y is the actual value X is the measured value and for sulfur values between 2.00 and 12.00 the following quadratic adjustment was applied to the LECO measured values: Y=1.923 + 1.029*X -0.017*X2 (4.16) (4.15)

However samples that were analyzed for carbon, hydrogen and nitrogen only (not sulfur) were weighed and encapsulated in tin capsules. Samples that were analyzed for carbon, hydrogen and nitrogen and sulfur were weighed and encapsulated in platinum capsules with a small amount of V2O5 catalyst added into the capsules. The V2O5 was added to aid in combustion of samples with sodium sulfate content. 4.4.2 X-ray diffraction analysis

Reaction product samples collected from the experiments were analyzed by x-ray diffraction to identify the crystallographic compounds contained in the product. A small sample of powder collected from each reaction, 5mg, was evenly distributed onto 1 square section in the center of a well-greased glass microscope slide. The sticky silicone grease was used to adhere the powder to the slide. The slide was mounted into the 193

sample holder of a Rigaku D-2000 Powder X-ray diffractometer. The diffractometer used a copper source with a 2dsin() equal to 1.54. The unit was set with 40KV and 40MA. The angle of rotation for the analysis was from 15 degrees to 65 degrees, stepping in 0.05-degree increments at a scan speed of 1degree/minute. Data processing of the scan was conducted after test on a separate computer using Rigaku search-match software supported with the International Centre for Diffraction Data (ICDD) Powder Diffraction File (PDF) database. The standard powder diffraction files from the Joint Committee on Powder Diffraction Standards (JCPDS) are contained in the ICDD PDF database. The software had a search and match engine to interpret the x-ray diffraction scans. It took a peak file, from the reaction sample, and matched it with materials from its database that possessed good fits. 4.4.3 BET surface area measurements

Samples of the reaction product were used for BET surface area measurements. A special BET sample holder was employed to allow BET measurements with a small amount of sample powder, 0.05grams. The BET sample size was limited by the fact that the overall reaction sample was small, 0.3 grams, and other testing was needed with the remainder. BET samples were not out-gassed above room temperature in the BET sample holder prior to analysis to assure that the product was not changed from its quick quench form. The BET sample was however mounted into the BET sample holder and purged with dry inert gas for 1 hour. All reaction samples were carefully stored under desiccated conditions during BET queue. Error analysis was conducted by measuring the surface area of the reaction products collected from five repeat reaction experiments.

194

This analysis determined a standard deviation of 0.1 m2/gram. As a result a confidence interval, of 2 sigma, 95% confidence of + 0.2 was assigned to all data. 4.4.4 Scanning electron microscope microstructure analysis

Microstructure analysis of particle product was conducted with a Scanning Electron Microscope (SEM). Powder product for analysis was specially prepared for the microscope. The powder product sparsely sprinkled onto double-sided conductive copper tape that was mounted on a 1 diameter metallic microscope-mounting slug. This mount was then sputtered for 12 minutes with gold to obtain a gold thin (500 to 700 angstroms thick) metallic coating. The gold sputtered onto the sample possessed a 4angstrom intrinsic grain structure, which for this analysis was non-visible. After sputtering, the SEM analysis was conducted in a Cambridge 90B SEM was used with a 25KV secondary electron beam. Powder particle surface analysis was conducted on a representative particle from the powder. Snapshots were taken at 20,000X magnification. 4.4.5 Time-of-flight static secondary ion mass spectrometry

This analysis technique was used to determine the product formed by the gas solid reaction of NO and O2 gas with sodium sulfite solid. Chapter 6 Section 6.3.3.1 discusses the findings and results of these experiments. This technique was selected because provided material analysis of the product formed on the outer surface of the sorbent particle. Other techniques such as X-ray diffraction, Electron Spectroscopy for Chemical Analysis (ESCA), and infrared (IR) were initially tried but were abandoned because they were unable to identify the outer surface product; instead they identified material beneath the outer surface that was not of interest.

195

Time-of-Flight static Secondary Ion Mass Spectrometry (TOF-SIMS) analysis was however found to provide identification of the outer surface product formation. The analysis was conducted at Physical Electronics Laboratory at the TOF-SIMS equipment manufacturers US headquarters in Eden Prairie, Minnesota. A PHI 7200 TOF-SIMS instrument was used for the analysis. The TOF-SIMS analysis technique bombarded the surface under analysis with short (1 ns) pulses of energetic primary ions (less than 1013 ions/cm2). The ions were generated with an 8keV Cs+ from a Tungsten heated frit ion beam source. These primary ion pulses cause were secondary ions (material) from the sample surface to desorb. The primary ion pulses cause collision cascades that knock off material from the outermost mono-layer of the sample. A stacked pair of multi channel plates and a narrow post acceleration gap (10kV) is used to produce an extraction potential between the sample surface and the mass analyzer. All secondary ions are accelerated to the same energy and therefore travel through the analyzer with different velocities, depending on their mass/charge ratio. Precise measurement of time of flight is detected and counted using a 253 stop time-to-digital converter with 156 ps time resolution for accurate determination of the particle mass. Measuring the arrival times of all the different species at the detector and calculating the corresponding mass collect the full mass spectrum. The achievable mass resolution of the analyzer is typically m/m > 7500 at m/z = 29 and m/m > 10,000 at m/z = 119. The set up of the positive ion scans included a 11.250 ns Time, 1.9661x105 pulses, 1.0894x105 integral, 1.5013x104maximum and for the negative ion scans the set up included an 3.750 ns Time, 6.1440 x105 pulses, 1.0263 x106 integral, 6.4632x104maximum.

196

4.4.6

X-ray photoelectron spectroscopy /electron spectroscopy for chemical analysis X-ray photoelectron Spectroscopy (XPS) also known as Electron Spectroscopy

for Chemical Analysis (ESCA) was used to analyze reaction product formed by the reaction of sodium sulfite with NO and O2. The results of this analysis are described in Chapter 6. The XPS analysis was conducted at an outside laboratory called Physical Electronics Laboratory. This technique is a widely used method for investigating the chemical composition of solid surfaces. It is a non-destructive method that provides elemental identification, chemical species identification, accurate quantification and spatial distribution information. The analysis is conducted under ultrahigh vacuum. A mono energetic (Al K 1486.6 eV) soft X-ray beam of less than 10 microns is irradiated on to the sample with known energy. The incoming x-rays penetrate into the surface of the sample. They interact with the atoms in the surface by the photoelectric effect, causing electrons to be emitted. Since the mean free path of the electrons is very small, the electrons that are detected originate only from the top few atomic layers. Photoelectrons from within 30 Angstroms of the surface are emitted as a result of the Xray absorption. The kinetic energy of the emitted photoelectrons is measured with an electron spectrometer. The spectrum obtained is a plot of the number of emitted electrons per energy interval versus their kinetic energy. Binding energy is determined from the following equation: BE = h-KE where: BE is the binding energy of the atomic orbital from which it originates (4.17)

197

h is the energy of the photon from the x-ray KE is the measured kinetic energy The binding energy associated with each detected peak is used to establish its elemental identity and chemical state. The binding energy may be regarded as an ionization energy of the atom for the particular shell involved. 4.5 4.5.1 4.5.1.1 Test Procedures Procedures for preliminary tests. Test procedure to investigate the effect of temperature on NO2 formation from the reaction of sodium bicarbonate and SO2 flue gas with NO. An experimental test procedure was devised to measure NO2 formation from the reaction of sodium bicarbonate with SO2 flue gas with NO at various temperatures. NO2 formation was evaluated from the reaction at 111C, 150C, 175C, 200C, 250C and 300C in the reactor apparatus. The SO2 flue gas with NO was mixed and flowed according to the percentages and volumetric flow rates listed in Table 4-3 and 4-4 respectively. The reactor preheat controller and heater controllers were set according to the set points listed in Tables 4-5 and 4-6 respectively for the specific reaction temperatures. Batch samples of sodium bicarbonate (0.35 grams) were injected into the isothermal fixed bed at the desired temperature and held for one hour. The powder mass mean diameter was approximately 30 microns. The gas analyzers monitored exit gas NO and NOx from the reaction through the data acquisition system. NO2 formation was measured by subtraction. Immediately after powder injection a pressure drop prohibited accurate data collection. After approximately one minute the system pressure stabilized 198

and data collection resumed. For each reaction NO2 formation was measured for 1hour. The peak NO2 formation, which occurred after injection, was used for analysis in the results section. No material analysis was conducted on the products removed from the reactor after test. The NO2 formation results from the six investigated temperatures are presented in Chapter 5, Section 5.2. For validation purposes, the data obtained in this study have been compared against the NO2 formation data previously reported by Bland. Appendix 4 lists the specific step-by-step procedure used for these tests. 4.5.1.2 4.5.1.2.1 Test procedures for material characterization studies. Conversion tracking test procedures for reactions related to sodium bicarbonate decomposition inhibition. Conversion of sodium bicarbonate to sodium carbonate was studied at 111C to investigate the cause of sodium bicarbonate decomposition inhibition in SO2 flue gas with NO. It was envisioned that an understanding of decomposition inhibition might lead to some further understanding of the overall pathway for NO2 formation. Table 4-9 lists the reactions investigated for understanding decomposition inhibition, including the flue gas mixture, the sorbent type, and the reaction temperatures studied.

199

Table 4-9 List of conversion tracking reactions at 111C. Flue Gas Mixture N2 SO2 flue gas without NO SO2 flue gas with NO SO2 flue gas with NO2 N2 with NO and O2 N2 with NO2 N2 with NO and O2 N2 with NO2 Sorbent Type NaHCO3 NaHCO3 NaHCO3 NaHCO3 NaHCO3 NaHCO3 Na2CO3 Na2CO3 Temp 111C X X X X X X X X

In particular the set of experiments listed in Table 4-9 evaluate the effect of NOx on sodium bicarbonate decomposition. Note that two reactions with sodium carbonate are included in the 111C study. These reactions are included as a frame of reference since Knight previously documented the reaction of NO2 with sodium carbonate. The experiments were conducted in the reactor apparatus described in Section 4.3.2. All of the reactions in this section were conducted at 111C. The pre-heater and heater controller were set according to the set points designated in Tables 4-5 and 4-6 for 111C. For each of the reactions listed in Table 4-9 the specified gas mixture was made according to the designated percentages and flows listed in Table 4-3 and 4-4. The tests involved reacting 0.3 grams of sodium bicarbonate (or sodium carbonate) with the flue gas mixture by dropping the powder into the fixed bed. The samples were kept in the reactor with flow respectively for an increment of 50, 120, 200, 400, or 600 minutes. Gas analysis was not conducted during the timed reaction increment. After the respective increment was complete the reaction sample was promptly transferred to the quick quench zone for quick purging with dry nitrogen and quick cooling by the quench system described in Section 4.2.2.5. After thawing, the reaction products were removed from the 200

apparatus at room temperature and stored in a dry nitrogen desiccator. Collecting reaction products for each reaction at the prescribed time increments made it possible to track the progress of conversion for each reaction. Reaction products collected from the reaction sets at incremental tracking times were elementally analyzed for hydrogen, nitrogen, and sulfur content by elemental analysis described in Section 4.4.1. X-ray diffraction analysis was used to determine the crystallographic compounds formed by the reactions. And conversion to these compounds was then calculated from the elemental analysis data as described in Appendix 2. The results of this experimental study are presented in Chapter 5, 5.3.1. 4.5.1.2.2 Conversion tracking test procedures for the study of the finite time period of sodium nitrate accumulation. Conversion tracking experiments were also employed to study the finite time period of sodium nitrate accumulation. Conversion tracking was thought to reveal new information as to why sodium nitrate is accumulated only during a finite time period. Additionally this study was thought to uncover the effect of temperature on sodium nitrate accumulation. These conversion tracking experiments were conducted exactly the same as the conversion tracking experiments described above in Section 4.5.1.2.1 for the study of decomposition inhibition except that the reactions studied in this section used reactor temperatures of 150C and 250C. The pre-heater and heater controllers were adjusted accordingly to achieve the different reactor temperatures as described in Section 4.3.2.2 Table 4-5. For the reactions at these temperatures, reaction time increments of 3, 10, 15, 30, 45, 60, and 300 minutes were selected for measuring conversion. Table 4-10 lists the reactions selected for study of sodium nitrate accumulation. 201

Table 4-10 Reactions conducted to study conversion. Flue Gas Mixture N2 SO2 flue gas without NO SO2 flue gas with NO SO2 flue gas with NO2 N2 with NO and O2 N2 with NO2 Sorbent Type NaHCO3 NaHCO3 NaHCO3 NaHCO3 NaHCO3 NaHCO3 Temp 150C X X X X X X Temp 250C X X X

The results of these conversion experiments are given in Chapter 5, Section 5.3.2.
Preparations of sodium nitrate for reactivity tests.

Four preparations of sodium nitrate were tested for reactivity with SO2 as described in 4.5.6.2. The method used for each sodium nitrate preparation is described in detail below:
Technical grade sodium nitrate.

An ultra pure laboratory grade technical grade material (99.9999 NaNO3) was shipped direct from the manufacturer in a vacuum-sealed container. The vacuum seal was broken immediately prior to the time when the sodium nitrate was injected into the reactor for the reactivity test.
Sodium nitrate prepared fresh by reaction of sodium carbonate with NO2.

Sodium nitrate was prepared by reacting injected sodium carbonate with NO2 at 111C. This method was based on the findings of Knight (1977)[18] as described in Chapter 2, Section 2.2.3. Sodium carbonate, 0.225 grams, was injected into the reactor at 111C with 1000 ppm NO2, with a balance of dry nitrogen. The NO2 flow was held for 1 hour and then shut off. The sodium nitrate product was maintained in dry nitrogen up until the reactivity test with SO2 and O2 was conducted. 202

Sodium nitrate prepared fresh by reaction of sodium sulfite with NO and O2. Sodium nitrate was prepared fresh for reactivity tests by reaction of sodium sulfite with NO and O2 where the sodium sulfite was made fresh by reaction of sodium bicarbonate with SO2. Material analysis in Chapter 6 showed that in addition to sodium nitrate this reaction resulted in some sodium nitrite formation. This nitration preparation is referred to as a two step because the preparation involves the following two steps: 1) Sodium bicarbonate (3 grams) was injected into the reactor at the desired temperature with 7200 ppm SO2 balance dry nitrogen flow (total flow of 1386 cc/min). Flow of SO2 was maintained for 15 minutes and then shut off, and purged completely from the reactor with nitrogen pulses. 2) NO and O2 flow was then turned on to allow 3600 ppm NO and 3% O2 flow into the reactor for 15 minutes and then shut off, and purged completely from the reactor with high flows of dry nitrogen pulses. Both SO2 and O2 were then turned on together to allow 3000 ppm SO2 and 3% O2 flows into the reactor while outlet NO and NOx (NOx - NO = NO2) gas analysis was recorded.
Method of Producing Reaction Product for Material Analysis

The different preparations of sodium nitrate were tested for their reactivity with SO2 and with SO2 and O2. For the reactivity tests where NO2 was found to form, a method was devised to collect sufficient quantity of sulfation product for X-ray diffraction analysis. For the preparation of sodium nitrate formed by reaction of sodium carbonate with NO2, the following steps were used: 1) Sodium carbonate, 0.225 grams, was injected into the reactor at 111C with 1000 ppm NO2, with a balance of dry nitrogen.

203

2) NO2 flow was held for 3 hours and then shut off. 3) Both SO2 and O2 (3000 ppm SO2 and 3% O2) were then turned on together and held for 2 hours and then shut off. These steps were followed to increase the amount of reaction product formed in an effort to aid in material analysis. X-ray diffraction analysis was used to obtain product analysis for this product. The material analysis results of the product are given in Chapter 7. For the study of sulfation product formation Table 4-11 lists the selected reactions conducted. Table 4-11 Reactions conducted to study sulfation product formation. Temp Temp Temp Sorbent Flue Gas 111C 150C 250C Type Mixture SO2 flue gas NaHCO3 X X X without NO SO2 flue gas NaHCO3 X X X with NO SO2 flue gas X X NaHCO3 without NO2 X SO2 and O2 NaNO3 freshly made [Na2CO3 + NO2] SO2 and O2 NaNO3 freshly made X [Na2SO3 + NO, O2]

Sulfation product was obtained from reaction of sodium bicarbonate with SO2 flue gas with and without NO, and with and without NO2, and from the reaction of sodium nitrate with SO2 and O2. Two preparations of sodium nitrate were evaluated. To obtain sulfation product crystallographic compound identification, x-ray diffraction (XRD) analysis was used. The analysis method used is described in Section 4.4.2. The XRD 204

analysis required substantial sulfation conversion, since small amounts of conversion were undetectable. As a result the samples analyzed for sulfation product identification possessed long reaction time increments of 300 minutes. The sulfation product identification results are presented in Chapter 5, Section 5.3.4.
The test procedure to check if the product formed by the three step reaction is reactive with NO and O2.

This preparation is referred to as a four step preparation and is made by adding the following additional step to the previously described three step preparation: NO and O2 flow were then turned on to allow 3600 ppm NO and 3% O2 flow into the reactor for 15 minutes and then shut off, and purged completely from the reactor with high flows of dry nitrogen pulses. 4.5.1.2.3 Procedure to study surface area changes for the reaction of sodium bicarbonate with SO2 flue gas with NO at 150 and 250C. The test procedure described in Section 4.5.1.2.2 was also used to study surface area changes, sulfation product formation and conversion at 150 and 250C. However the pre-heater and heater controller set points were adjusted to the levels listed Table 4-5 and 4-6 for 150C and 250C. For the surface area studies, surface area changes were tracked for the sodium bicarbonate reaction in SO2 flue gas with NO and in SO2 flue gas without NO. Table 412 lists these selected reactions studied.

205

Table 4-12 changes.

Reactions conducted to study surface area Sorbent Type NaHCO3 NaHCO3 NaHCO3 Temp 150C X X X Temp 250C X X X

Flue Gas Mixture N2 SO2 flue gas without NO SO2 flue gas with NO

To track surface area with respect to time, surface area measurements were conducted for reactions with reaction hold time increments of 0.8, 3, 10, 60, and 300 minutes. The sample product from each experiment was analyzed for surface area by the method described in Section 4.4.3. The results for these surface area investigations are presented in Chapter 5, Section 5.3.3 4.5.2 4.5.2.1 Procedures for reaction validation tests. Test procedure to validate the reaction of sodium sulfite with NO and O2.

The reaction of sodium sulfite with NO and O2 was studied using the reactor apparatus described in 3.2.2. Table 4-13 lists the selected reactions investigated. Table 4-13 Selected Reactions for sodium sulfite with NO and O2 study. Sorbent Temp Type Flue Gas Mixture 175C Na2CO3 blank N2 with NO(50ppm) and O2 X freshly made Na2SO3 N2 with NO(50ppm) X freshly made Na2SO3 N2 with NO(50ppm) and O2 X technical grade Na2SO3 N2 with NO(50ppm) X technical grade Na2SO3 N2 with NO(50ppm) and O2 X

206

The reactivity tests with NO and O2 involved two primary steps: the special preparation of solid sodium sulfite, and the reactivity test with gas analysis. The following general procedure was used for these tests: 1) For experiments testing the reactivity of freshly made sodium sulfite, the sodium sulfite was freshly produced by injecting sodium bicarbonate (5.6 grams) into the reactor at 111C with 7200 ppm SO2 balance dry nitrogen flow (total flow of 1386 cc/min). The SO2 was flowed through the reactor for 1 hour and then shut off, and purged completely from the reactor with nitrogen pulses. 2) With nitrogen flowing (total flow of 1386 cc/min), the temperature in the reactor was adjusted to the desired test temperature of 175C. 3) The NO, NOx, and SO2 analyzers were calibrated and prepared for outlet monitoring. 4) NO flow was then turned on to allow 50 ppm NO flow into the reactor while outlet NO, NOx (NOx - NO = NO2), and SO2 analysis was recorded. 5) O2 flow was then turned on to allow 3% oxygen flow into the reactor while outlet NO, NOx (NOx - NO = NO2), and SO2 analysis was recorded. A blank system test was conducted to check and assure that the test procedure, the apparatus, and the sodium carbonate substrate did not produce NOx artifacts. For the blank system test step 1 was skipped. Instead, sodium carbonate was injected during step 2 after the reactor reached 175C. Steps 3 through 5 were followed without change.

207

4.5.2.1.1

Special preparations of sodium sulfite for reactivity tests.

As described in Chapter 2, Section 2.2.2.2, Howatson et al. (1980) showed that sodium sulfite could be produced by reacting sodium bicarbonate with SO2 flue gas at 111C. Fresh sodium sulfite was produced by this method for the reactivity studies discussed in Chapter 6. Sodium bicarbonate was reacted with 7000 ppm SO2, balance dry nitrogen gas at a total flow rate of 1386 cc/min. While preparing sodium sulfite, oxygen was not included in the gas mixture as an extra precaution to avoid the possibility of sodium sulfate formation. The freshly made sodium sulfite formed by this method was kept inside the reactor without exposing it to the atmosphere. It was then tested for reactivity with NO and O2 immediately. Between being freshly formed and being tested with NO and O2 the reactor was purged for 10 minutes with dry nitrogen gas to assure that all of the SO2 was flushed from the system. During the nitrogen purge the reactor temperature was adjusted to the desired reaction temperature being evaluated (111C, and 175C) Once thoroughly purged of SO2, the 50 ppm NO was inlet into the reactor with the freshly formed sodium sulfite. Reaction observations were made by comparing the reactor inlet NO concentration with the reactor exit NO concentration. After steady state NO concentrations were obtained the oxygen valve was then opened and the reaction of the freshly made sodium sulfite with NO and O2 was observed.
Freshly made sodium sulfite reactivity tests

To test the reactivity of freshly made sodium sulfite with NO and O2 steps 1 through 5 were followed. Freshly made sodium sulfite was made in step 1 of the sequence by reacting injected sodium bicarbonate with SO2 flue gas without NO at 111C. This method of making sodium sulfite was based on the findings of Howatson et al.s (1980) 208

[8]

as described in Chapter 2, Section 2.2.2.2. Dry nitrogen with SO2 was used in step 1

along with a 111C temperature to assure sodium sulfite formation without sodium sulfate. Note that after the sodium sulfite was freshly formed it was kept in the reactor with dry nitrogen flowing and then without exposure to the air it was reacted with NO and O2 in steps 4 and 5. The freshly made sodium sulfite was not removed from the system or exposed to atmospheric conditions prior to reactivity test. The reactivity tests were conducted at 175C because this was the temperature where sodium bicarbonate with SO2 flue gas with NO was reported to produce the greatest amount of NO2 (see Chapter 5 Section 5.2 Figure 5-1).
Technical grade sodium sulfite reactivity tests.

The tests conducted with technical grade sodium sulfite used the General Procedure described above except that step 1 was skipped and instead technical grade sodium sulfite was injected into the reactor during step 2 when the bed was at 175C with dry nitrogen flow. Steps 3 through 5 were followed without change. 4.5.2.1.2 Method of producing reaction product for material analysis.

A special procedure was used to obtain reaction product for material analysis. Since material analysis was easier with more product the procedure was modified to increase the product formation. All five steps of the general method described above were followed except that step 1 was held for 12 hours instead of 1 hour in an effort to fully convert the sodium bicarbonate to sodium sulfite. And additionally the freshly formed sodium sulfite was reacted with a higher, 3600 ppm, NO concentration in step 4 and held for 20 minutes. The material analysis results of the product formed are described in Chapter 6 Section 6.3.3.1. 209

4.5.2.2

Test procedure to validate the reaction of sodium nitrate with SO2 and O2 . The reaction of sodium nitrate with SO2 and O2 was also studied using the reactor

apparatus described in Section 4.3.2. Different preparations of sodium nitrate were evaluated: technical grade sodium nitrate, sodium nitrate freshly prepared by reacting sodium carbonate with NO2, and sodium nitrate freshly prepared by reacting sodium sulfite with NO and O2. Oxygen was also studied for its effect in the reactions. When reactivity was found, the effect of temperature on NO2 formation was evaluated Table 414 lists the selected reactions investigated. Table 4-14 Selected reactions for reactivity study of sodium nitrate with SO2 and O2. No. Solid reactant Gas Temperature (C) 1 blank SO2, O2 175 2 technical grade NaNO3 SO2 175 3 technical grade NaNO3 SO2, O2 175 4 freshly made NaNO3 175 SO2 [Na2CO3+NO2] 5 freshly made NaNO3 111, 150, 175, SO2, O2 200, 250 [Na2CO3+NO2] 6 freshly made NaNO3 175 SO2 [Na2SO3+NO,O2] 7 freshly made NaNO3 111, 150, 175, SO2, O2 200, 250 [Na2SO3+NO,O2] SO2, O2 8 Product of 175 {[(Na2SO3+NO,O2)+SO2,O2] +NO, O2} The reactivity testing involved the following steps: 1) For experiments testing the reactivity of preparations of freshly made sodium nitrate, the sodium nitrate was either inserted or freshly produced in the reactor

210

where the reactivity tests were conducted. Any gases used for sodium nitrate preparation were purged out of the reactor completely with dry nitrogen pulses. 2) With nitrogen flowing (total flow of 1386 cc/min), the temperature in the reactor was then adjusted to the desired reactivity test temperature (i.e. 111C, 150C, 175C, 200C, 250C). 3) The NO and NOx analyzers were then calibrated and prepared for outlet monitoring. Both SO2 and O2 were then turned on together to allow 3000 ppm SO2 and 3% O2 flows into the reactor while outlet NO and NOx (NOx - NO = NO2) gas analysis was recorded. Oxygen was not added for experiments 2 and 6 of the list given in Table 4-14. 4.5.3 4.5.3.1 Procedures for mitigation study tests. Test procedure to investigate the effect of oxygen percent on NO2 formation from the reaction of sodium bicarbonate and SO2 flue gas with NO. The test procedure used to investigate the effect of oxygen percent in the flue gas on NO2 formation from the reaction of sodium bicarbonate with SO2 flue gas with NO was similar to the procedure described in Section 4.5.1.1 to investigate the effect of temperature on NO2 formation except that the reactor temperature was not varied from experiment to experiment. Instead the temperature was maintained at 175C. Also the sodium bicarbonate sorbent used in these experiments was of a coarser size then the powder used in the validation tests. The powder mass mean diameter was approximately 175 microns. Samples of 0.35 grams were injected into the reactor for each test. Five

211

different oxygen percents were studied, 0%, 0.5%, 3%, 5%, and 20%. Table 4-15 provides the adjusted flows for the flue gas mixture components. Table 4-15 Flow Rate (cc/min) 1386 Flue gas mixes with various oxygen percents. CO2 O2 N2 SO2 NO NO2 (cc/min) (cc/min) (cc/min) (cc/min) (cc/min) (cc/min) 197.2 0 1183.4 4.1 1.3 0

Mixture ID

SO2 flue gas (0% O2) with NO SO2 flue gas (0.5% O2) with NO SO2 flue gas (3% O2) with NO SO2 flue gas (5% O2) with NO SO2 flue gas (20% O2) with NO

1386

197.2

7.0

1176.4

4.1

1.3

1386

197.2

44.3

1139.1

4.1

1.3

1386

197.2

69.3

1114.1

4.1

1.3

1386

197.2

277.2

906.2

4.1

1.3

Since Bland had previously reported NO2 formation at 0% oxygen in flue gas, the test conducted in this study at 0% oxygen also served as further validation of the test set up and the procedures. These tests were also designed to provide parameter guidance on flue gas re-circulation for NO2 mitigation. The test results for the oxygen investigations are presented in Chapter 8.

212

CHAPTER 5

PRELIMINARY TESTS 5.1 Introduction Several topics related to the reaction pathway for NO2 formation for the reaction of sodium bicarbonate with SO2 flue gas with NO were included in this preliminary study chapter. These studies provide useful preliminary information. Experiments were conducted to validate and assure that the fixed bed reactor constructed for the laboratory experimentation in this study operated properly. The apparatus validation studies aimed to show that equipment set-up simulated the bag-house injection reactions. To assure that the data collected was of a good quality, measurements of NO2 formation versus temperature were collected from this laboratory set-up and compared to reputable past results found in the literature. Additionally in this chapter, preliminary material characterization was conducted. To clarify why the literature reported that sodium bicarbonate decomposition was inhibited for reactions in SO2 flue gas with NO at low temperatures, sorbent conversion tracking was used to study decomposition inhibition for sodium bicarbonate reaction in SO2 flue gas with NO. Also since no explanation or pathway particulars have been provided to explain why past researchers found that sodium nitrate accumulation occured only during a finite time period immediately following injection of sodium bicarbonate into SO2 flue gas with NO, conversion tracking was also used in this chapter to study this finite period of sodium nitrate accumulation for these reactions. Surface area and microstructure changes of sodium bicarbonate were studied for its reactions in SO2 flue 213

gas with NO. As further verification of the experimental set up, sulfation product collected from the reaction of sodium bicarbonate with SO2 flue gas without NO was analyzed for sulfation compound identification and compared to reputable past findings in the literature. In this section, sulfation compounds and associated polymorphic forms were reviewed. Finally in this chapter, decomposition conversion of sodium pyrosulfite and sodium bisulfite in inert gas was studied. All of the preliminary tests were intended to provide useful information related to the reaction pathway for NO2 formation. 5.2 Test for proper operation of apparatus.- Measurements of NO2 versus

temperature for the reaction of sodium bicarbonate with SO2 flue gas with NO. In this section the reaction of sodium bicarbonate and SO2 flue gas with NO with respect to temperature was studied for system validation. This study was used to provide a quality assurance check of the reactor apparatus, and the experimental procedure including gas analysis. These validation tests were intended to verify that the fixed bed reactor and the associated procedures were able to produce data that was comparable to data obtained by past investigators. In particular NO2 formation was measured for the injection of sodium bicarbonate into SO2 flue gas with NO at various temperatures. The following temperatures were evaluated: 111C, 150C, 175C, 200C, 250C, and 300C. These NO2 measurements were compared with the NO2 measurements previously reported by Bland (1990)[16]. The experimental test procedure for these experiments is described in Section 4.4. Figure 5.1 compares the NO2 formation with respect to temperature measurements obtained in this study to the NO2 formation measurements previously

214

reported by Bland (1990)[16]. The data collected in this study are labeled as Stein (2001) while the data collected previously by Bland (1990)[16] is respectively labeled.

300

Stein (2001)
250

Peak NO2 Formation (ppm)

Bland (1990)

200

150

100

50

0 0 50 100 150 200 250 300 350 400

Temperature (C)

Figure 5-1 Validation test results of peak NO2 formation versus temperature. The data obtained in this study are shown with the solid line. The data obtained by Bland (1990) are dashed.

Figure 5.1 shows that the data collected in this study (Stein, 2001) agrees with the data previously collected by Bland (1990)[16]. Although the data sets are not exactly alike, they both follow the same general trends. There are slight differences in the data sets because the reactor apparatuses used were different; Bland (1990)[16] for example, used a fluidized bed reactor while in this study a fixed bed reactor was used. In both data sets the

215

maximum NO2 formation is close to 175C. And in both data sets minimum NO2 formation occurs at low temperature increasing up to the maximum. Also in both data sets the NO2 formation decreases with increasing temperature above the maximum. From low temperature up to 175C (the NO2 max), NO2 formation is speculated to be controlled by the sodium sulfate ash layer.

NaHCO3/Na2CO3 SO2 Na2SO3 Na2NO3 Na2SO4

NO

SO2

Figure 5-2 Snapshot of ash layers from reaction sequence.

Figure 5.2 shows a generalized schematic of the ash layer formation during reaction. For SO2 removal, SO2 is required to diffuse through the sodium sulfate, nitrate and sulfite ash layers to get to the sodium bicarbonate or the decomposed sodium carbonate core for reaction, 2NaHCO3 + SO2 Na2SO3 + 2CO2 + H2O Na2CO3 + SO2 Na2SO3 + CO2 Similarly for NOx removal, NO is required to diffuse through the sodium sulfate and sodium nitrate ash layer to get to the sodium sulfite ash layer for reaction, Na2SO3 + 2NO + 2O2 2NaNO3 + SO3 (5.3) (5.1) (5.2)

For NO2 formation, SO2 is required to diffuse through the sodium sulfate ash layer to get to the sodium nitrate ash layer for reaction, 216

2NaNO3 + SO2 Na2SO4 + 2NO2

(5.4)

For this temperature range, diffusion through the sodium sulfate ash layer increases with temperature. For the temperature range of 175C and greater NO2 formation is believed to be limited by the sodium sulfite formation at this temperature range. Recall that it is well established that sodium sulfite formation decreases with temperature due to the increase in the sodium sulfite oxidation reaction rate, Na2SO3 + O2 Na2SO4 5.3 Material characterization tests. (5.5)

In these preliminary material characterization studies conversion of sodium bicarbonate was studied. X-ray diffraction analysis of sulfation product from the reaction of sodium bicarbonate with SO2 flue gas without NO was studied. Conversion tracking analysis was conducted to investigate decomposition inhibition of sodium bicarbonate in SO2 flue gas with NO. Conversion tracking analysis was also conducted to study the finite NOx removal period during the reaction of sodium bicarbonate with SO2 flue gas with NO. Surface area and microstructure changes were measured during these conversion tests to determine their effect on NOx removal. Also the effect of water vapor concentration on NOx removal was studied. 5.3.1 Sodium bicarbonate decomposition inhibition in SO2 flue gas with NO. Bland (1990)[16] previously reported that sodium bicarbonate decomposition was inhibited at low temperatures when reacted with SO2 flue gas with NO compared to the reaction in inert gas. In this chapter and section this decomposition inhibition is studied

217

further. Conversion tracking is used to better understand the cause of decomposition inhibition. Table 5.1 provides a complete list of the conversion tracking reactions experimentally studied in this section including the flue gas mixtures, the sorbent type, the reaction temperature, the associated figure number, and whether or not decomposition was found to be inhibited for the respective reaction. All of these experiments were conducted at 111C. Table 5-1 Conversion Tracking Reactions. Flue Gas Mixture N2 SO2 flue gas without NO SO2 flue gas with NO N2 with NO and O2 N2 with NO2 SO2 flue gas with NO2 N2 with NO2 N2 with NO and O2 Sorbent Type NaHCO3 NaHCO3 NaHCO3 NaHCO3 NaHCO3 NaHCO3 Na2CO3 Na2CO3 Figure Number Figure 5.3 Figure 5.4 Figure 5.5 Figure 5.6 Figure 5.7 Figure 5.8 Figure 5.9 Figure 5.10 Decomposition Inhibition no no yes yes yes yes Not applicable Not applicable

Note from Table 5.1 that decomposition inhibition occurs for all of the reactions that contain some form of NOx in the flue gas. For each of the reactions evaluated, conversion was tracked with respect to time by measuring the products formed at incremental time periods. Chapter 4, Section 4.4.1.2.1 provides a detailed description of the experimental method used for these tracking experiments. The conversion products measured were residual sodium bicarbonate, sodium carbonate, sodium sulfite, sodium sulfate, and sodium nitrate. Elemental analysis was used to quantitatively measure conversion. And X-ray diffraction analysis was used to qualitatively identify the particular SO2 removal product and the NOx removal product formed. Appendix 2 218

provides a description of the calculation method used for the reaction conversions graphed. The scale of the y-axis for the conversion graphs for residual sodium bicarbonate, sodium carbonate, and sulfation is 0 to 100% in Figures 5.3-5.10. But because only a small amount of sodium nitrate conversion was found for nearly all of the reactions investigated, the scale of the y-axis conversion tracking graphs was modified to 12%. Since using a 0 to 100% scale a 1% conversion to sodium nitrate was indistinguishable from a 3% conversion to sodium nitrate, the scale for sodium nitrate was reduced to a 12% max to allow more accurate inspection and comparison of these conversion results. First, baseline conversions for reaction of sodium bicarbonate in inert nitrogen gas at 111C are shown in Figure 5.3a-d respectively for residual sodium bicarbonate, sodium nitrate, sodium carbonate, and sodium sulfite.

219

(a
100 90 Sodium Bicarbonate in Product % 80 70 60 50 40 30 20 10 0

Conversion of NaHCO3 in Inert Nitrogen Gas at 111C: Sodium Bicarbonate in the Product versus Reaction Time

50

100

150

200

250

300

350

400

450

Tim e (m inutes)

(b
Sodium Nitrate in Product % 12 10 8 6 4 2 0 0

Conversion of NaHCO3 in Inert Nitrogen Gas at 111C: Sodium Nitrate in the Product versus Reaction Time

50

100

150

200

250

300

350

400

450

Tim e (m inutes)

(c
Sodium Carbonate in Product %

Conversion of NaHCO3 in Inert Nitrogen Gas at 111C: Sodium Carbonate in the Product versus Reaction Time
100 90 80 70 60 50 40 30 20 10 0 0 50 100 150 200 250 300 350 400 450 Tim e (m inutes)

(d
100 Sodium Sulfite in Product % 90 80 70 60 50 40 30 20 10 0 0

Conversion of NaHCO3 in Inert Nitrogen Gas at 111C: Sodium Sulfite in the Product versus Reaction Time

50

100

150

200

250

300

350

400

450

Tim e (m inutes)

Figure 5-3 Conversion of sodium bicarbonate in inert nitrogen gas at 111C. Gas composition was 100% nitrogen. (a) residual sodium bicarbonate, (b) sodium nitrate, (c) sodium carbonate, (d) sodium sulfite. Note full scale on the sodium nitrate versus time graph is 12 %.

220

As expected Figures 5-3a and c show sodium bicarbonate decomposition to sodium carbonate. With time the amount of residual sodium bicarbonate decreases converting to sodium carbonate. The amount of residual sodium bicarbonate drops to 10% after 125 minutes. Likewise, the amount of sodium carbonate conversion increases to nearly 90% within 125 minutes. Because NOx was not in the gas mixture, no NOx removal or nitration product is found to form in Figure 5.3b. And because SO2 was not in the gas, no SO2 removal or sulfation product was found to form in 5-3d. Next, a cursory look at conversion of sodium bicarbonate in SO2 flue gas without NO is shown in Figure 5-4a-d respectively for residual sodium bicarbonate, sodium nitrate, sodium carbonate, and sodium sulfite.

221

(a
100 90 Sodium Bicarbonate in Product % 80 70 60 50 40 30 20 10 0

Conversion of NaHCO3 in SO2 Flue Gas without NO at 111C: Sodium Bicarbonate in the Product versus Reaction Time

50

100

150

200

250

300

350

400

450

Tim e (m inutes)

(b
Sodium Nitrate in Product % 12 10 8 6 4 2 0 0

Conversion of NaHCO3 in SO2 Flue Gas without NO at 111C: Sodium Nitrate in the Product versus Reaction Time

50

100

150

200

250

300

350

400

450

Tim e (m inutes)

(c
100 Sodium Carbonate in Product % 90 80 70 60 50 40 30 20 10 0

Conversion of NaHCO3 in SO2 Flue Gas without NO at 111C: Sodium Carbonate in the Product versus Reaction Time

50

100

150

200

250

300

350

400

450

Tim e (m inutes)

(d
100 Sodium Sulfite in Product % 90 80 70 60 50 40 30 20 10 0

Conversion of NaHCO3 in SO2 Flue Gas without NO at 111C: Sodium Sulfite in the Product versus Reaction Time

50

100

150

200

250

300

350

400

450

Tim e (m inutes)

Figure 5-4 Conversion of sodium bicarbonate in SO2 flue gas without NO at 111C. Gas composition was 14% CO2, 3% O2, 83% N2, 0.3% SO2. (a) residual sodium bicarbonate, (b) sodium nitrate, (c) sodium carbonate, (d) sodium sulfite. Note full scale on the sodium nitrate versus time graph is 12 %.

222

Because this data set was intended to be cursory in nature, no data points were collected below 200 minutes. The conversion results show that the sodium bicarbonate decomposed by 200 minutes, without any gross decomposition inhibition. X-ray diffraction analysis described in Chapter 6, Section 6.4.5, Table 6-5 determined that the sulfation product formed for this reaction was sodium sulfite. This sulfation product determination confirmed the past sulfation product findings of both Howatson et al.(1980)[8]and Bland (1990)[16]. Next, the conversion results for sodium bicarbonate with SO2 flue gas with NO is shown in Figure 5.5 including conversion of sodium bicarbonate to sodium carbonate, and sodium nitrate and sodium sulfate.

223

(a)
100 90 Sod ium Bicar bonate in 80 70 Prod uct % 60 50 40 30 20 10 0 0

Conve rsion of Na HCO 3 in S O 2 Flue Ga s w ith NO a t 111C: S odium Bica rbona te in the P roduct ve rsus Re a ction Tim e

50

100

150

200

250

300

350

400

450

T im e (m inute s )

(b)
So dium Nitr ate in Prod uct %

Conve rsion of Na HCO 3 in S O 2 Flue Ga s w ith NO a t 111C: S odium Nitra te in the Product ve rsus Re a ction Tim e

(b
12 10 8 6 4 2 0 0 50 100 150 200 250 300 350 400 450

Tim e (m inute s )

(c) (c
So dium Car bonate in Pr oduct

Conve rsion of Na HCO 3 in SO 2 Flue Ga s w ith NO a t 111C: Sodium Ca rbona te in the P roduct ve rsus Re a ction Tim e

100 90 80 70 60

50 40 30 20 10 0 0 50 100 150 200 250 300 350 400 450

Tim e (m inu te s )

(d (d)
Sodium Su lfate in Prod uct %

Conve rsion of Na HCO 3 in SO 2 Flue Ga s w ith NO a t 111C: S odium Sulfa te in the Product ve rsus Re a ction Tim e

100 90 80 70 60 50 40 30 20 10 0 0 50 100 150 200 250 300 350 400 450

Tim e (m inu te s )

Figure 5-5 Conversion of sodium bicarbonate in SO2 flue gas with NO at 111C. Gas composition was 14% CO2, 3% O2, 83% N2, 0.3% SO2 and 0.1% NO. (a) residual sodium bicarbonate, (b) sodium nitrate, (c) sodium carbonate, (d) sodium sulfate. Note full scale on the sodium nitrate versus time graph is 12 %.

224

Figure 5-5a and c show that sodium bicarbonate decomposition is significantly inhibited in this reaction since only a very small amount of sodium bicarbonate converts to sodium carbonate, 10% over 450 minutes. Recall from Figure 5-3 that under inert nitrogen conditions 90% of the sodium bicarbonate decomposed to sodium carbonate during the first 125 minutes. These results clearly confirm Blands (1990)[16] previous finding that decomposition of sodium bicarbonate is inhibited when sodium bicarbonate is reacted with SO2 flue gas with NO. Compared to the reaction of sodium bicarbonate in SO2 flue gas without NO shown in Figure 5-4d, a much smaller amount of sulfation product is formed for the reaction in SO2 flue gas with NO shown in Figure 5-5d. The reduction in sulfation product formation is thought to be due to the ash layer resistance from the sodium sulfate product compared to the ash layer resistance of sodium sulfite formed when SO2 flue gas is without NO. Ash layer diffusion resistance through sodium sulfate is greater than the ash layer diffusion through sodium sulfite. Figure 5-5b shows that there is a small amount of conversion to sodium nitrate for this reaction. The sodium nitrate formed is speculated to be responsible for sodium bicarbonate decomposition inhibition. Specifically, the hygroscopic forces from the sodium nitrate formation surrounding the particle are speculated to oppose the forces of decomposition and cause decomposition inhibition. The hygroscopic forces from the nitrate strongly bind water vapor to the particle surface preventing evolution of water vapor from decomposition. Relative to the findings of Chapter 6, where sodium nitrate was shown to react with SO2 and O2 resulting in its conversion to sodium sulfate and NO2, in this reaction,

225

sodium nitrate is found to accumulate and are not shown to be destroyed by reaction with SO2 and O2. The hygroscopic forces of the sodium nitrate are hypothesized in this case to prevent reaction of sodium nitrate with SO2 and O2. Next, conversion for the reaction of sodium bicarbonate in NO and O2 (balance inert nitrogen) was studied and is shown in Figure 5-6 including conversion of sodium bicarbonate to sodium carbonate, and sodium nitrate and sodium sulfite/sulfate.

226

(a
100 90 Sodium Bicarbonate in Product % 80 70 60 50 40 30 20 10 0

Conversion of NaHCO3 in Nitrogen Gas with NO,O2 at 111C: Sodium Bicarbonate in the Product versus Reaction Time

50

100

150

200

250

300

350

400

450

Tim e (m inutes)

(b
Sodium Nitrate in Product % 12 10 8 6 4 2 0 0

Conversion of NaHCO3 in Nitrogen Gas with NO,O2 at 111C: Sodium Nitrate in the Product versus Reaction Time

50

100

150

200

250

300

350

400

450

Tim e (m inutes)

(c
100 Sodium Carbonate in Product % 90 80 70 60 50 40 30 20 10 0

Conversion of NaHCO3 in Nitrogen Gas with NO,O2 at 111C: Sodium Carbonate in the Product versus Reaction Time

50

100

150

200

250

300

350

400

450

Tim e (m inutes)

(d
100 Sodium Sulfite in Product % 90 80 70 60 50 40 30 20 10 0

Conversion of NaHCO3 in Nitrogen Gas with NO,O2 at 111C: Sodium Sulfite in the Product versus Reaction Time

50

100

150

200

250

300

350

400

450

Tim e (m inutes)

Figure 5-6 Conversion of sodium bicarbonate in nitrogen gas with NO, O2 at 111C. Gas composition 96.9% nitrogen, 0.1% NO, 3% O2. (a) residual sodium bicarbonate, (b) sodium nitrate, (c) sodium carbonate, (d) sodium sulfite/sulfate. Note full scale on the sodium nitrate versus time graph is 12 %.

227

Note this reaction was conducted in gas with NO and O2 but without SO2. The conversion results in Figure 5-6 show that decomposition is inhibited similar to the inhibition found in the previous reaction shown in Figure 5-5. A small percentage (less then 19%) of the product decomposes to sodium carbonate while 80% of the product remains un-decomposed as sodium bicarbonate after 450 minutes. The fact that decomposition was found to proceed without inhibition for the reaction in SO2 flue gas without NO as shown in Figure 5-4 indicates that decomposition inhibition is not related to sulfation alone. Decomposition inhibition is believed to be associated with sodium nitrate formation, since decomposition was found to be inhibited for the reaction of sodium bicarbonate with NO and O2 gas alone without SO2, as well as, for the reaction in SO2 flue gas with NO. Next, conversion data was also collected for the reaction of sodium bicarbonate with NO2 and O2, as shown in Figure 5-7 including conversion of sodium bicarbonate to sodium carbonate, and sodium nitrate and sodium sulfite/sodium sulfate.

228

(a
100 90 Sodium Bicarbonate in Product % 80 70 60 50 40 30 20 10 0 0

Conversion of NaHCO3 in Nitrogen Gas with NO2 at 111C: Sodium Bicarbonate in the Product versus Reaction Time

50

100

150

200

250

300

350

400

450

Tim e (m inutes)

(b
Sodium Nitrate in Product % 12 10 8 6 4 2 0 0

Conversion of NaHCO3 in Nitrogen Gas with NO2 at 111C: Sodium Nitrate in the Product versus Reaction Time

50

100

150

200

250

300

350

400

450

Tim e (m inutes)

(c
100 Sodium Carbonate in Product % 90 80 70 60 50 40 30 20 10 0 0

Conversion of NaHCO3 in Nitrogen Gas with NO2 at 111C: Sodium Carbonate in the Product versus Reaction Time

50

100

150

200

250

300

350

400

450

Tim e (m inutes)

(d
100 Sodium Sulfite in Product % 90 80 70 60 50 40 30 20 10 0 0

Conversion of NaHCO3 in Nitrogen Gas with NO2 at 111C: Sodium Sulfite in the Product versus Reaction Time

50

100

150

200

250

300

350

400

450

Tim e (m inutes)

Figure 5-7 Conversion of sodium bicarbonate in nitrogen gas with NO2 at 111C. Gas composition was 97% nitrogen, 0.1% NO2, 2.9% O2. (a) residual sodium bicarbonate, (b) sodium nitrate, (c) sodium carbonate, (d) sodium sulfite/sodium sulfate. Note full scale on the sodium nitrate versus time graph is 12 %.

229

Figures 5-7a and c show that decomposition is completely inhibited in this reaction. Decomposition is found to be more inhibited in this reaction, than in the reaction of sodium bicarbonate with NO and O2, less sodium bicarbonate is found to convert to sodium carbonate. For this reaction, the larger amount of nitration product is speculated to provide stronger hygroscopic opposition forces against decomposition. Figure 5-7b shows that over 12% of the product was converted to sodium nitrate compared to the less than 1% converted by the reaction in NO and O2. The large amount of nitration was expected since Knight (1977)[18] previously established that NO2 was more reactive than NO and O2. Decomposition is more inhibited for this reaction where a greater amount of nitrate product formation is found to form as shown in Figure 5-7b compared to 5-6b. The net action of the hygroscopic forces is speculated to be greater from the greater amount of sodium nitrate conversion. Next, Figure 5-8 shows the conversion results for the reaction of sodium bicarbonate with SO2 flue gas with NO2 including conversion of sodium bicarbonate to sodium carbonate, and sodium nitrate and sodium sulfate.

230

(a
100 90 Sodium Bicarbonate in Product % 80 70 60 50 40 30 20 10 0 0

Conversion of NaHCO3 in SO2 Flue Gas with NO2 at 111C: Sodium Bicarbonate in the Product versus Reaction Time

50

100

150

200

250

300

350

400

450

Tim e (m inutes)

(b
Sodium Nitrate in Product % 12 10 8 6 4 2 0 0

Conversion of NaHCO3 in SO2 Flue Gas with NO2 at 111C: Sodium Nitrate in the Product versus Reaction Time

50

100

150

200

250

300

350

400

450

Tim e (m inutes)

(c
100 Sodium Carbonate in Product % 90 80 70 60 50 40 30 20 10 0 0

Conversion of NaHCO3 in SO2 Flue Gas with NO2 at 111C: Sodium Carbonate in the Product versus Reaction Time

50

100

150

200

250

300

350

400

450

Tim e (m inutes)

(d
100 Sodium Sulfate in Product % 90 80 70 60 50 40 30 20 10 0 0

Conversion of NaHCO3 in SO2 Flue Gas with NO2 at 111C: Sodium Sulfate in the Product versus Reaction Time

50

100

150

200

250

300

350

400

450

Tim e (m inutes)

Figure 5-8 Conversion of sodium bicarbonate in SO2 flue gas with NO2 at 111C. Gas composition was 14% CO2, 3% O2, 83% N2, 0.3% SO2 and 0.1% NO2. (a) residual sodium bicarbonate, (b) sodium nitrate, (c) sodium carbonate, (d) sodium sulfate. Note full scale on the sodium nitrate versus time graph is 12 %.

231

Figure 5-8 shows that decomposition is fully inhibited for this reaction. Because a relatively large amount of sodium nitrate conversion was previously found for the reaction of sodium bicarbonate and NO2 (without SO2) shown in Figure 5-7, there was some anticipation that similarly a large amount of sodium nitrate conversion might accumulate for this reaction with SO2 flue gas with NO2. Conversion to sodium nitrate shown in Figure 5-7b is not, however, large. The ash layer resistance from sodium sulfate formation is thought to be the reason why considerably less sodium nitrate conversion is found for this reaction in SO2 flue gas with NO2 compared to the amount found for the reaction in NO2 (without SO2) in Figure 5-7b. The sodium nitrate conversion for this reaction in SO2 flue gas with NO2 is greater than the amount found to form for the reaction of sodium bicarbonate with SO2 flue gas with NO as shown in Figure 5-5b. Sodium sulfate polymorphic Form I was found to be the sulfation conversion product by this reaction. The formation of this polymorphic form is an indicator that the reaction of sodium nitrate with SO2 and O2 has occurred. Next, the conversion obtained for the reaction of sodium carbonate with NO2 and O2 at 111C was also studied and the results are shown in Figure 5-9 including conversion of sodium carbonate to sodium nitrate.

232

(a
100 90 Sodium Bicarbonate in Product % 80 70 60 50 40 30 20 10 0 0

Conversion of Na 2CO3 in Nitrogen Gas with NO2 at 111C: Sodium Bicarbonate in the Product versus Reaction Time

50

100

150

200

250

300

350

400

450

Tim e (m inutes)

(b
Sodium Nitrate in Product % 12 10 8 6 4 2 0 0

Conversion of Na 2CO3 in Nitrogen Gas with NO2 at 111C: Sodium Nitrate in the Product versus Reaction Time

50

100

150

200

250

300

350

400

450

Tim e (m inutes)

(c
100 Sodium Carbonate in Product % 90 80 70 60 50 40 30 20 10 0 0

Conversion of Na 2CO3 in Nitrogen Gas with NO2 at 111C: Sodium Carbonate in the Product versus Reaction Time

50

100

150

200

250

300

350

400

450

Tim e (m inutes)

(d
100 Sodium Sulfite in Product % 90 80 70 60 50 40 30 20 10 0 0

Conversion of Na 2CO3 in Nitrogen Gas with NO2 at 111C: Sodium Sulfite in the Product versus Reaction Time

50

100

150

200

250

300

350

400

450

Tim e (m inutes)

Figure 5-9 Conversion of sodium carbonate in nitrogen gas with NO2 at 111C. Gas composition was 97% nitrogen, 0.1% NO2, 2.9% O2. (a) residual sodium bicarbonate, (b) sodium nitrate, (c) sodium carbonate, (d) sodium sulfite/sodium sulfate. Note full scale on the sodium nitrate versus time graph is 12 %.

233

As expected, based on the previous studies of Knight (1977)[18], a large amount of conversion to sodium nitrate occurs by this reaction. More conversion to sodium nitrate occurs for this reaction compared to any of the previous reactions studied (Figures 5-5b through Figure 5-8b). Note that the conversion to sodium nitrate for this reaction is greater than the conversion to sodium nitrate found for the reaction of sodium bicarbonate and NO2. For the case of the reaction of sodium bicarbonate and NO2 at 111C recall that decomposition was inhibited and therefore does not produce added surface area of sodium carbonate for reaction. Recall that Knight (1977)[18] previously reported that at low temperature, sodium carbonate was highly reactive with NO2. Finally Figure 5-10 shows the conversion results for the reaction of sodium carbonate with NO and O2 at 111C including conversion of sodium carbonate to sodium nitrate.

234

(a
100 90 Sodium Bicarbonate in Product % 80 70 60 50 40 30 20 10 0 0

Conversion of Na 2CO3 in Nitrogen Gas with NO,O2 at 111C: Sodium Bicarbonate in the Product versus Reaction Time

50

100

150

200

250

300

350

400

450

Tim e (m inutes)

(b
Sodium Nitrate in Product % 12 10 8 6 4 2 0 0

Conversion of Na 2CO3 in Nitrogen Gas with NO,O2 at 111C: Sodium Nitrate in the Product versus Reaction Time

50

100

150

200

250

300

350

400

450

Time (m inutes)

(c
100 Sodium Carbonate in Product % 90 80 70 60 50 40 30 20 10 0

Conversion of Na 2CO3 in Nitrogen Gas with NO,O2 at 111C: Sodium Carbonate in the Product versus Reaction Time

50

100

150

200

250

300

350

400

450

Tim e (m inutes)

(d
100 Sodium Sulfite in Product % 90 80 70 60 50 40 30 20 10 0

Conversion of Na 2CO3 in Nitrogen Gas with NO,O2 at 111C: Sodium Sulfite in the Product versus Reaction Time

50

100

150

200

250

300

350

400

450

Time (m inutes)

Figure 5-10 Conversion of sodium carbonate in nitrogen gas with NO and O2 at 111C. Gas composition 96.9% nitrogen, 0.1% NO, 3% O2. (a) residual sodium bicarbonate, (b) sodium nitrate, (c) sodium carbonate, (d) sodium sulfite/sodium sulfate. Note full scale on the sodium nitrate versus time graph is 12 %.

235

A very small amount of conversion to sodium nitrate was found for this reaction and is shown in Figure 5-10b. As expected, based on the previous findings of Knight much less conversion to sodium nitrate was expected for this reaction in NO and O2 than for the reaction with NO2 shown in Figure 5-9. The amount of sodium nitrate formed by the reaction of sodium carbonate in NO and O2 is comparable to the amount of sodium nitrate formed for by reaction of sodium bicarbonate in NO and O2. Both sodium carbonate and sodium bicarbonate are relatively non-reactive with NO and O2, however it is important to note that in both cases a small amount of conversion to sodium nitrate is found. 5.3.2 Sodium nitrate accumulation by sodium bicarbonate reaction with SO2 flue gas with NO. Recall that Bland (1990)[16] (see EPRI GS 6850 page 8-8) reported that a finite time period of NOx removal existed immediately following sodium bicarbonate injection into SO2 flue gas with NO for the temperature range between 135C and 150C. Also recall that Blands (1990)[16] studies revealed that NO2 formation began after this finite NOx removal period ended. Bland (1990)[16] noted that the time period when NOx was removed was the time period when NO2 formation lagged. In this chapter Blands reported finite period of NOx removal and NO2 lag for the reaction of sodium bicarbonate with SO2 flue gas with NO was investigated further. First experiments were conducted to verify the existence of the finite period of NOx removal. Experiments were conducted to understand why NOx removal is limited to this finite time period. Sodium nitrate accumulation was closely examined to understand why it accumulated during this time period rather than converting to sodium sulfate by 236

reaction with SO2 and O2 as discussed further in Chapter 6. The selected reactions studied in this section are listed in Table 5.2. All of the reaction experiments weere conducted at 150C. Table 5-2 Reactions to study finite NOx removal period Sorbent Type NaHCO3 NaHCO3 NaHCO3 NaHCO3 NaHCO3 Na2CO3 Na2CO3 Na2CO3 Figure No. Figure 5-11 Figure 5-12 Figure 5-13 Figure 5-14 Figure 5-15 Figure 5-16 Figure 5-17 Figure 5-18

Flue Gas Mixture N2 SO2 flue gas with NO N2, NO and O2 N2 with NO2 SO2 flue gas with NO2 N2 with NO2 SO2 flue gas with NO SO2 flue gas with NO2

Tracking of product conversion from sodium bicarbonate to sodium carbonate, sodium nitrate, sodium sulfate or sodium sulfite was conducted during each of the reactions listed in Table 5.2. For NOx removal, sodium nitrate was tracked. The experimental method used is described in Chapter 4, Section 4.4.1.2.2. The conversion calculation algorithm from the elemental analysis of carbon, hydrogen, nitrogen and sulfur is described in Appendix 2. The sulfation product determination of sodium sulfate for each experiment was determined by X-Ray diffraction analysis as reported in Chapter 6, Section 6.4.5 Table 6.5. The term sodium nitrate accumulation is used interchangeably with the term NOx removal in the following discussion. Note that all

237

of the reactions in Table 5.2 were conducted at the temperature of 150C; a temperature above 111C the temperature used for the reactions in the previous section. It has been established that decomposition and sulfation rates are greater at 150C compared to at 111C. As a baseline, the conversion results for the reaction of sodium bicarbonate with inert gas at 150 C are shown in Figure 5-11a-d respectively for residual sodium bicarbonate, sodium nitrate, sodium carbonate, and sodium sulfite/sodium sulfate.

238

(a
Sodium Bicarbonate in Product %

Conversion of NaHCO3 in Inert Nitrogen Gas at 150C: Sodium Bicarbonate in the Product versus Reaction Time
100 90 80 70 60 50 40 30 20 10 0 0 5 10 15 20 25 30 35 40 45 50 55 60 65 Tim e (m inutes)

(b
Sodium Nitrate in Product % 12 10 8 6 4 2 0

Conversion of NaHCO3 in Inert Nitrogen Gas at 150C: Sodium Nitrate in the Product versus Reaction Time

10

15

20

25

30

35

40

45

50

55

60

65

Tim e (minutes)

(c
Sodium Carbonate in Product %

Conversion of NaHCO3 in Inert Nitrogen Gas at 150C: Sodium Carbonate in the Product versus Reaction Time
100 90 80 70 60 50 40 30 20 10 0 0 5 10 15 20 25 30 35 40 45 50 55 60 65 Tim e (m inutes)

(d
Sodium Sulfate in Product %

Conversion of NaHCO3 in Inert Nitrogen Gas at 150C: Sodium Sulfate in the Product versus Reaction Time
100 90 80 70 60 50 40 30 20 10 0 0 5 10 15 20 25 30 35 40 45 50 55 60 65 Tim e (m inutes)

Figure 5-11 Conversion of sodium bicarbonate in inert nitrogen at 150C. Gas composition was 100% nitrogen. (a) residual sodium bicarbonate, (b) sodium nitrate, (c) sodium carbonate (d) sodium sulfite/sodium sulfate. Note full scale on the sodium nitrate graph is 12 %.

239

The baseline in Figure 5-11 shows that sodium bicarbonate decomposes to sodium carbonate, 2NaHCO3 Na2CO3 + H2O + CO2 The period of time that the sodium bicarbonate decomposes to sodium carbonate is referred to as the finite period of decomposition. Figure 5-11 shows that the finite period of decomposition is 15 minutes from start to finish. The conversion results for the reaction of sodium bicarbonate with SO2 flue gas with NO at 150C are shown in Figure 5-12a-d respectively for residual sodium bicarbonate, sodium nitrate, sodium carbonate, and sodium sulfate. (5.6)

240

(a
100 90 Sodium Bicarbonate in Product % 80 70 60 50 40 30 20 10 0 0

Conversion of NaHCO3 in SO2 Flue Gas with NO at 150C Sodium Bicarbonate in the Product versus Reaction Time

10

15

20

25

30

35

40

45

50

55

60

65

Tim e (m inutes)

(b
Sodium Nitrate in Product % 12 10 8 6 4 2 0 0

Conversion of NaHCO3 in SO2 Flue Gas with NO at 150C Sodium Nitrate in the Product versus Reaction Time

10

15

20

25

30

35

40

45

50

55

60

65

Tim e (m inutes)

(c
100 Sodium Carbonate in Product % 90 80 70 60 50 40 30 20 10 0 0

Conversion of NaHCO3 in SO2 Flue Gas with NO at 150C Sodium Carbonate in the Product versus Reaction Time

10

15

20

25

30

35

40

45

50

55

60

65

Tim e (m inutes)

(d
100 Sodium Sulfate in Product % 90 80 70 60 50 40 30 20 10 0 0

Conversion of NaHCO3 in SO2 Flue Gas with NO at 150C Sodium Sulfate in the Product versus Reaction Time

10

15

20

25

30

35

40

45

50

55

60

65

Tim e (m inutes)

Figure 5-12 Conversion of sodium bicarbonate in SO2 flue gas with NO at 150C. Gas composition was 14% CO2, 3% O2, 83% N2, 0.3% SO2 and 0.1% NO. (a) residual sodium bicarbonate, (b) sodium nitrate, (c) sodium carbonate, (d) sodium sulfate. Note full scale on the sodium nitrate versus time graph is 12 %.

241

The following key observations are drawn from the conversion results shown in Figure 512: Figure 5-12 shows that sodium nitrate predominantly accumulates during the first 15 minutes after injection. Very little sodium nitrate accumulation occurs after 15 minutes, distinguishing the finite time period of NOx removal where sodium nitrate is accumulated. This time period of accumulation is referred to as the finite time period of sodium nitrate accumulation. Figure 5-12 shows that the finite time period of sodium nitrate accumulation coincides with the finite time period of decomposition of the sodium bicarbonate in this reaction. Sodium nitrate accumulation and sodium bicarbonate decomposition both occur during the first 15 minutes. Figure 5-12 shows that decomposition is slightly inhibited for this reaction in SO2 flue gas with NO. Decomposition of sodium bicarbonate settles to a steady state conversion level of 75% after 15 minutes compared to the steady state conversion level of over 90% after 15 minutes obtained in inert nitrogen at 150C shown in Figure 5-11. Sodium sulfate accumulation is found to be ongoing and does not correspond to any finite time period. The amount of sodium sulfate accumulation is not significantly lowered after 15 minutes but is ongoing. Sodium sulfate was the only sulfation product found to form in this product according to the X-ray diffraction analysis. Sodium sulfite was not found in the product. Comparing the sulfation conversion in Figure 5-12d with the sulfation conversion for the same reaction but at 111C as shown in Figure 5-5d, it is apparent that a higher

242

sulfation rate occurs at 150C (40% conversion to sodium sulfate in 45 minutes) compared to at 111C (10% conversion to sodium sulfate in 400 minutes). Up to this point the finite time period of sodium nitrate accumulation has not been extensively studied. It remains unclear why sodium nitrate accumulates only during this finite time period immediately following injection. It is unclear why sodium nitrate accumulation does not occur whenever sodium sulfate is accumulated. Since sodium nitrate accumulation is found to coincide with sodium bicarbonate decomposition in Figure 5-12; it is suspected that water vapor evolved from decomposition plays a key roll in sodium nitrate accumulation. It is postulated that the strong hygroscopic forces binding water vapor (available from decomposition) to sodium nitrite or nitrate, shields SO2 and O2 from reacting with sodium nitrate. Water vapor from sodium bicarbonate decomposition is therefore defined as the agent responsible for the finite time period of sodium nitrate formation. This postulation is consistent with the findings presented in Chapter 6 where only freshly formed sodium nitrate was found to react with SO2 and O2. In Chapter 6 it is found that sodium nitrate that has been exposed to atmospheric humidity (not freshly formed) was un-reactive with SO2 and O2. Further investigation is needed however to provide more evidence that water vapor interacts with sodium nitrate to prevent reaction of the sodium nitrate with SO2 and O2. The reaction pathway for this reaction is postulated to proceed with the following reaction equations: Sodium bicarbonate reacts with SO2 to form sodium sulfite, 2NaHCO3 + SO2 Na2SO3 + 2CO2 + H2O Sodium bicarbonate decomposes to sodium carbonate: (5.7)

243

2NaHCO3 Na2CO3 + H2O + CO2 Sodium sulfite reacts with NO and O2 to form sodium nitrite or nitrate, Na2SO3 + 2NO + O2 2NaNO2 + SO3 Na2SO3 + 2NO + 2O2 2NaNO3 + SO3

(5.8)

(5.9) (5.10)

Water vapor from decomposition is speculated to bind to available sodium nitrite or sodium nitrate as postulated here, NaNO2 + H2O + O2 NaNO3(H2O) NaNO3 + H2O NaNO3(H2O) (5.11) (5.12)

Sodium nitrite and nitrate that is free of surface bound water vapor reacts with SO2 for conversion to sodium sulfate and NOx. For example sodium nitrite or nitrate formed after decomposition ends reacts with SO2 resulting in conversion to sodium sulfate, 2NaNO2 + SO2 Na2SO4 + 2NO 2NaNO3 + SO2 Na2SO4 + 2NO2 (5.13) (5.14)

The NO2 formed by reaction equation 5.14 may react with the sodium carbonate formed by decomposition resulting in additional sodium nitrate formation. This reaction has been well established by Knight, Na2CO3 + 3NO2 2NaNO3 + CO2+ NO (5.15)

The sodium sulfite formed by equation 5.8 is converted to sodium nitrite or nitrate by equation 5.9 and 5.10. And this sodium nitrate formed is either, bound with water vapor through equation 5.12 and shielded from reaction with SO2 and O2, or destroyed by reaction with SO2 and O2 according to equation 5.14. Figure 5-12b shows that no additional sodium nitrate is accumulated after decomposition ends. Without water vapor evolution from decomposition, the sodium 244

nitrate that forms is not bound with water vapor and protected from the destructive reactions with SO2 and O2. As a result any sodium nitrate that forms after decomposition ends is destroyed or converted by reaction with SO2 and O2 to sodium sulfate and NO2. It may be contemplated that perhaps counter diffusion rather than the effect of water vapor is responsible for the finite time period when NO2 is not formed. It may be considered that counter diffusion from decomposition of sodium bicarbonate inhibits SO2 from diffusing towards and reacting with available sodium nitrate to produce NO2 according to equation 5-14. If counter diffusion obstructed incoming SO2 from diffusing in it would also obstruct NO from diffusing into the particle for sodium nitrate formation during decomposition. But there is evidence that counter diffusion does not stop SO2 diffusion into the particle since sulfation compounds are shown to form continually throughout decomposition when there is counter diffusion. But Figure 5-12b shows that sodium nitrate formation is not stopped during decomposition but is on the contrary the only time period when it forms. Figure 5-12 shows that the sodium nitrate accumulated during decomposition remains accumulated after decomposition ends. This sodium nitrate that has formed during decomposition is not destroyed or converted after decomposition ends. It is speculated that the water vapor that during decomposition is bound to the accumulated sodium nitrate by hygroscopic forces remains in tact after decomposition ends since the hygroscopic forces remain after decomposition. These hygroscopic forces that bind the water to the sodium nitrate and shield it from converting to sodium sulfate by reaction with SO2 and O2, are speculated to be strong and permanent.

245

Next, the conversion results for the reaction of sodium bicarbonate with NO, O2 at 150C are shown in Figure 5-13a-d respectively for residual sodium bicarbonate, sodium nitrate, sodium carbonate, and sodium sulfate.

246

(a
100 90 Sodium Bicarbonate in Product % 80 70 60 50 40 30 20 10 0

Conversion of NaHCO3 in Nitrogen Gas with NO,O2 at 150o C: Sodium Bicarbonate in the Product versus Reaction Time

10

15

20

25

30

35

40

45

50

55

60

65

Tim e (m inutes)

(b
Sodium Nitrate in Product % 12 10 8 6 4 2 0

Conversion of NaHCO3 in Nitrogen Gas with NO,O2 at 150o C: Sodium Nitrate in the Product versus Reaction Time

10

15

20

25

30

35

40

45

50

55

60

65

Time (m inutes)

(c
100 Sodium Carbonate in Product % 90 80 70 60 50 40 30 20 10 0

Conversion of NaHCO3 in Nitrogen Gas with NO,O2 at 150o C: Sodium Carbonate in the Product versus Reaction Time

10

15

20

25

30

35

40

45

50

55

60

65

Time (m inutes)

(d
100 Sodium Sulfite in Product % 90 80 70 60 50 40 30 20 10 0

Conversion of NaHCO3 in Nitrogen Gas with NO,O2 at 150o C: Sodium Sulfite in the Product versus Reaction Time

10

15

20

25

30

35

40

45

50

55

60

65

Time (m inutes)

Figure 5-13 Conversion of sodium bicarbonate in nitrogen with NO and O2 at 150C. Gas composition was 96.9% nitrogen, 0.1% NO, 3% O2. (a) Residual sodium bicarbonate, (b) sodium nitrate, (c) sodium carbonate, (d) sodium sulfite/sodium sulfate. Note full scale on the sodium nitrate versus time graph is 12 %.

247

The following key observations were made from Figure 5-13: Figure 5-13 shows that less sodium nitrate accumulation occurs for this reaction compared to the amount found for the reaction in SO2 flue gas with NO, Figure 5-12b. As a result, this reaction is not speculated to be the only reaction responsible for sodium nitrate accumulation for the overall reaction in SO2 flue gas with NO. Instead NOx removal and sodium nitrate accumulation for this reaction in NO and O2 is speculated to occur by reaction of sodium sulfite with NO and O2. Figure 5-13 shows that the sodium nitrate accumulation in nitrogen with NO and O2 is ongoing; it is not limited to a finite time period compared to the finite time period of sodium nitrate accumulation found for reaction in SO2 flue gas with NO. Sodium nitrate accumulation in 5-13 occurs during and after sodium bicarbonate decomposition. Figure 5-13 shows that decomposition is not inhibited. The small amount of sodium nitrate accumulated does not inhibit decomposition at this temperature. At this temperature a larger amount of sodium nitrate accumulation is needed to inhibit decomposition. The following reaction sequence is postulated to occur for this reaction: Sodium bicarbonate initially begins to decompose, 2NaHCO3 Na2CO3 + H2O + CO2 Sodium bicarbonate and sodium carbonate react with NO: NaHCO3 + NO + O2 NaNO3(H2O) + CO2 Na2CO3 + NO + 2O2 2NaNO3 + CO2 Na2CO3 + NO + 2O2 + H2O 2NaNO3(H2O) + CO2 (5.17) (5.18) (5.19) (5.16)

248

Conversion results for the reaction of sodium bicarbonate with NO2, O2 and nitrogen at 150C are shown in Figure 5-14a-d respectively for residual sodium bicarbonate, sodium nitrate, sodium carbonate, and sodium sulfite/sodium sulfate.

249

(a
100 90 Sodium Bicarbonate in Product % 80 70 60 50 40 30 20 10 0 0

Conversion of NaHCO3 in Nitrogen Gas with NO2 at 150o C: Sodium Bicarbonate in the Product versus Reaction Time

10

20

30

40

50

60

70

80

90

100

110

120

Tim e (m inutes)

(b
Sodium Nitrate in Product % 12 10 8 6 4 2 0 0

Conversion of NaHCO3 in Nitrogen Gas with NO2 at 150o C: Sodium Nitrate in the Product versus Reaction Time

10

20

30

40

50

60

70

80

90

100

110

120

Tim e (m inutes)

(c
100 Sodium Carbonate in Product % 90 80 70 60 50 40 30 20 10 0 0

Conversion of NaHCO3 in Nitrogen Gas with NO2 at 150o C: Sodium Carbonate in the Product versus Reaction Time

10

20

30

40

50

60

70

80

90

100

110

120

Tim e (m inutes)

(d
100 Sodium Sulfite in Product % 90 80 70 60 50 40 30 20 10 0 0

Conversion of NaHCO3 in Nitrogen Gas with NO2 at 150o C: Sodium Sulfite in the Product versus Reaction Time

10

20

30

40

50

60

70

80

90

100

110

120

Tim e (m inutes)

Figure 5-14 Conversion of sodium bicarbonate in nitrogen with NO2 at 150C for 120 minutes. Gas composition was 97% nitrogen, 0.1% NO2, 2.9% O2. (a) residual sodium bicarbonate, (b) sodium nitrate, (c) sodium carbonate, (d) sodium sulfite/sodium sulfate. Note full scale on the sodium nitrate versus time graph is 12 %.

250

The following key observations are made from Figure 5-14: Figure 5-14shows that a large amount of sodium nitrate accumulation occurs compared to the reaction with NO and O2, Figure 5-13, and the reaction with SO2 flue gas with NO, Figure 5-12. Figure 5-14shows that the sodium nitrate that is accumulated is not limited to the finite time period of sodium bicarbonate decomposition. Sodium nitrate accumulation in Figure 5-14b is ongoing after decomposition. Unlike the case where SO2 was present in the flue gas, as shown in Figure 5-12, sodium nitrate accumulation for this reaction without SO2, occurs during and after (beyond 65 minutes) sodium bicarbonate decomposition. Figure 5-14 shows that decomposition is inhibited. Compared to the baseline reaction of sodium bicarbonate with inert nitrogen shown in Figure 5-11, the finite time period of sodium bicarbonate decomposition is significantly longer. The finite time period of decomposition for this reaction is approximately 65 minutes compared to the 15 minutes measured under inert conditions, Figure 5-11. Decomposition inhibition appears to be associated with the large amount of sodium nitrate that forms in this reaction. This sodium nitrate with its hygroscopic nature is thought to bind water vapor to the particle surface and counteract the forces of decomposition. The following reaction sequence is postulated to occur for this reaction: Sodium bicarbonate initially begins to decompose, 2NaHCO3 Na2CO3 + H2O + CO2 Sodium bicarbonate and sodium carbonate react with NO2: NaHCO3 + NO2 + O2 NaNO3(H2O) + CO2 (5.21) (5.20)

251

Na2CO3 + 2NO2 + O2 + H2O 2NaNO3(H2O) + CO2 Na2CO3 + 2NO2 + O2 2NaNO3 + CO2

(5.22) (5.23)

These results also provide further evidence that SO2 plays a significant role in the finite period of sodium nitrate formation. The finite time period of sodium nitrate formation was not found to form when SO2 was not present in the flue gas, and was found to form when SO2 was in the flue gas. The much greater amount of sodium nitrate accumulation for this reaction compared to the reaction in NO and O2 as shown in Figure 5-13b, confirms Knights (1977)[18] past finding that NO2 is more reactive with sodium carbonate compared to with NO and O2. For this reaction in NO2 gas at 150C, sodium bicarbonate decomposition is not nearly as inhibited as was found for the same reaction at 111C as shown in Figure 5-7. But recall that at 150C in NO and O2 gas shown in Figure 5-13 where a smaller amount of sodium nitrate is accumulated, sodium bicarbonate decomposition is not inhibited. These results provide evidence that sodium bicarbonate decomposition inhibition is influenced directly by the amount of sodium nitrate accumulated. The conversion results for the reaction of sodium bicarbonate in SO2 flue gas with NO2 at 150C are shown in Figure 5-15a-d respectively for residual sodium bicarbonate, sodium nitrate, sodium carbonate, and sodium sulfate.

252

(a
100 90 Sodium Bicarbonate in Product % 80 70 60 50 40 30 20 10 0 0

Conversion of NaHCO3 in SO2 Flue Gas with NO2 at 150C: Sodium Bicarbonate in the Product versus Reaction Time

10

15

20

25

30

35

40

45

50

55

60

65

Tim e (m inutes)

(b
Sodium Nitrate in Product % 12 10 8 6 4 2 0 0

Conversion of NaHCO3 in SO2 Flue Gas with NO2 at 150C: Sodium Nitrate in the Product versus Reaction Time

10

15

20

25

30

35

40

45

50

55

60

65

Tim e (m inutes)

(c
100 Sodium Carbonate in Product % 90 80 70 60 50 40 30 20 10 0 0

Conversion of NaHCO3 in SO2 Flue Gas with NO2 at 150C: SodiumCarbonate in the Product versus Reaction Time

10

15

20

25

30

35

40

45

50

55

60

65

Tim e (m inutes)

(d
Sodium Sulfate in Product %

Conversion of NaHCO3 in SO2 Flue Gas with NO2 at 150C: Sodium Sulfate in the Product versus Reaction Time
100 90 80 70 60 50 40 30 20 10 0 0 5 10 15 20 25 30 35 40 45 50 55 60 65 Tim e (m inutes)

Figure 5-15 Conversion of sodium bicarbonate in SO2 flue gas with NO2 at 150C. Gas composition was 14% CO2, 2.9% O2, 83% N2, 0.3% SO2 and 0.1% NO2. (a) residual sodium bicarbonate, (b) sodium nitrate, (c) sodium carbonate, (d) sodium sulfate. Note full scale on the sodium nitrate versus time graph is 12 %.

253

The conversion results for the reaction in SO2 flue gas with NO2 in Figure 5-15 closely resemble the results obtained for the reaction in SO2 flue gas with NO shown in Figure 512. Sodium nitrate is predominantly accumulated during the first 15 minutes after injection. After 15 minutes very little sodium nitrate accumulation occurs relative to the other conversions, distinguishing a finite time period of NOx removal where sodium nitrate is accumulated. The finite time period of sodium nitrate accumulation coincides with the finite time period of decomposition for sodium bicarbonate. Sodium nitrate accumulation and sodium bicarbonate decomposition both occur during the first 15 minutes. Figure 5-15 shows that decomposition is slightly inhibited for this reaction in SO2 flue gas with NO. Instead of 100% decomposition of sodium bicarbonate, Figure 5-15 shows decomposition settles to a steady state conversion level of 60% after 15 minutes compared to the steady state conversion level of over 90% after 15 minutes obtained in inert nitrogen at 150C shown in Figure 5-11. The sodium sulfate accumulation is found to be ongoing and does not correspond to any finite time period. The amount of sodium sulfate accumulation is not significantly lowered after 15 minutes but is ongoing. Sodium sulfate is the only sulfation product found to form in the product. Sodium sulfite is not found in the product. Comparing this sulfation conversion at 150C in Figure 5-15d with the sulfation conversion at 111C as shown in Figure 5-5d, it is apparent that more sulfation product is formed at 150C, where 40% conversion to sodium sulfate occurs in 45 minutes, compared to at 111C, where 10% conversion to sodium sulfate occurs in 400 minutes. 254

The reaction equations in the pathway for this reaction are speculated to be as follows: Sulfation with sodium bicarbonate, 2NaHCO3 + SO2 Na2SO3 + 2CO2 + H2O Decomposition begins: 2NaHCO3 Na2CO3 + H2O + CO2 (5.25) (5.24)

Sodium nitrate forms by direct reaction of NO2 and sodium bicarbonate and carbonate, 2NaHCO3 + 2NO2 + O2 2NaNO3(H2O) + 2CO2 Na2CO3 + 2NO2 + O2 + H2O 2NaNO3(H2O) + CO2 Na2CO3 + 2NO2 + O2 2NaNO3 + CO2 And sodium nitrate forms by reaction of sodium sulfite with NO2, Na2SO3 + 2NO2+ O2 2NaNO3 + SO3 (5.29) (5.26) (5.27) (5.28)

Hygroscopic forces cause available water vapor, from equation 5.24 and 5.25 to bind with the sodium nitrate: NaNO3 + H2O NaNO3(H2O) The sodium nitrate free of surface bound water vapor reacts with SO2 converting to sodium sulfate and NOx, 2NaNO3 + SO2 Na2SO4 + 2NO2 The NO2 formed by reaction equation 5.31 may react with the sodium carbonate according to reaction equation 5.23. Note that any sodium sulfite that forms by equation 5.24 is converted to sodium sulfate by reaction with NO2 in equation 5.26 and subsequent reaction with SO2 by (5.31) (5.30)

255

equation 5.30 and 5.31. And any sodium nitrite or nitrate that forms is destroyed by SO2 if it is not bound with water. For the reaction in SO2 flue gas with NO2 larger sodium nitrate accumulation would have been anticipated compared to in SO2 flue gas with NO since more accumulation is found for NO2 without SO2 than for NO without SO2. But the amount of sodium nitrate that accumulates is nearly equivalent to the amount of sodium nitrate that accumulates for the reaction in SO2 flue gas with NO shown in Figure 5-12b. It is speculated that the ash layer diffusion resistance of the sodium sulfate that forms affects sodium nitrate accumulation. The reaction of sodium carbonate with NO2 (without SO2) is shown respectively for residual sodium bicarbonate, sodium nitrate, sodium carbonate, and sodium sulfate in Figure 5-16.

256

(a
Sodium Bica rbona te in Product % 100 90 80 70 60 50 40 30 20 10 0 0

Conversion of Na2CO3 in NO2 at 150C: Sodium Bicarbonate in the Product versus Reaction Time

10

20

30

40

50

60

70

80

90

100

110

120

Tim e (m inute s)

(b
50 S odium Nitra te in Product % 40

Conversion of Na2CO3 in NO2 at 150C: Sodium Nitrate in the Product versus Reaction Time

30

20

10

0 0 60 Tim e (m inute s) 120 180

(c
Sodium Ca rbona te in P roduct %

Conversion of Na2CO3 with NO2 at 150C: Sodium Carbonate in the Product versus Reaction Time
100 90 80 70 60 50 40 30 20 10 0 0 10 20 30 Tim e (m inute s) 40 50 60

Conversion of Na2CO3 with NO2 at 150C: Sodium Sulfate in the Product versus Reaction Time

(d
Sodium Sulfa te in Product %

100 90 80 70 60 50 40 30 20 10 0 0 60 Tim e (m inute s) 120 180

Figure 5-16 Conversion of sodium carbonate with NO2 at 150C. Gas composition was 97% nitrogen, 0.1% NO2, 2.9% O2. (a) residual sodium bicarbonate, (b) sodium nitrate, (c) sodium carbonate, (d) sodium sulfate. 257

Figure 5-16 shows that sodium nitrate is accumulated for this reaction. There is however no finite time period of sodium nitrate accumulation; instead sodium nitrate accumulation is ongoing. The amount of sodium nitrate accumulation for this reaction is much smaller than the amount found for the same reaction at 111C as shown in Figure 5-9. This is consistent with Blands (1990)[16] previous findings that showed an inverse relation of sodium nitrate accumulation with respect to temperature. Also the amount of sodium nitrate accumulation in Figure 5-16 is smaller than the amount found for the reaction of sodium bicarbonate in NO2 gas as shown in Figure 5-14. The added surface area that forms by decomposition is thought to enhance sodium nitrate accumulation. More activated and more porous sodium carbonate forms through decomposition compared to the crystalline sodium carbonate. Conversion results for the reaction of sodium carbonate with SO2 flue gas with NO, and for the reaction of sodium carbonate with SO2 flue gas with NO2 at 150C are shown respectively in Figure 5-17a-d and 5-18a-d for residual sodium bicarbonate, sodium nitrate, sodium carbonate, and sodium sulfate.

258

(a
Sodium Bica rbona te in P roduct % 100 90 80 70 60 50 40 30 20 10 0 0

Conversion of Na2CO3 in SO2 Flue Gas with NO at 150C: Sodium Bicarbonate in the Product versus Reaction Time

10

20

30 Tim e (m inute s)

40

50

60

(b
S odium Nitra te in P roduct %

Conversion of Na2CO3 in SO2 Flue Gas with NO at 150C: Sodium Nitrate in the Product versus Reaction Time
100 90 80 70 60 50 40 30 20 10 0 0 Tim e (m inute s) 60

Conversion of Na2CO3 in SO2 Flue Gas with NO at 150C: Sodium Carbonate in the Product versus Reaction Time
S odium Ca rbona te in Product %

(c

100 90 80 70 60 50 40 30 20 10 0 0 10 20 30 Tim e (m inute s) 40 50 60

Conversion of Na2CO3 in SO2 Flue Gas with NO at 150C: Sodium Sulfate in the Product versus Reaction Time
100

(d

Sodium Sulfa te in P roduct %

90 80 70 60 50 40 30 20 10 0 0 Tim e (m inute s) 60

Figure 5-17 Conversion of sodium carbonate in SO2 flue gas with NO at 150C. Gas composition was 14% CO2, 3% O2, 83% N2, 0.3% SO2 and 0.1% NO. (a) residual sodium bicarbonate, (b) sodium nitrate, (c) sodium carbonate, (d) sodium sulfate.

259

(a
Sodium Bica rbona te in Product % 100 90 80 70 60 50 40 30 20 10 0 0

Conversion of Na2CO3 in SO2 Flue Gas with NO2 at 150C: Sodium Bicarbonate in the Product versus Reaction Time

10

20

30 Tim e (m inute s)

40

50

60

(b
100 S odium Nitra te in Product % 90 80 70 60 50 40 30 20 10 0 0

Conversion of Na2CO3 in SO2 Flue Gas with NO2 at 150C: Sodium Nitrate in the Product versus Reaction Time

60 Tim e (m inute s)

(c
Sodium Ca rbona te in P roduct %

Conversion of Na2CO3 in SO2 Flue Gas with NO2 at 150C: Sodium Carbonate in the Product versus Reaction Time
100 90 80 70 60 50 40 30 20 10 0 0 10 20 30 Tim e (m inute s) 40 50 60

Conversion of Na2CO3 in SO2 Flue Gas with NO2 at 150C: Sodium Sulfate in the Product versus Reaction Time

(d
Sodium Sulfa te in Product %

100 90 80 70 60 50 40 30 20 10 0 0 Tim e (m inute s) 60

Figure 5-18 Conversion of sodium carbonate in SO2 flue gas with NO2 at 150C. Gas composition was 14% CO2, 2.9% O2, 83% N2, 0.3% SO2 and 0.1% NO2. (a) residual sodium bicarbonate, (b) sodium nitrate, (c) sodium carbonate, (d) sodium sulfate. 260

The conversion results in Figure 5-17b and 5-18b for the reaction of sodium carbonate with SO2 flue gas with NO and NO2 show that sodium nitrate was not accumulated in either of these reactions. No sodium nitrate accumulation was found for these reactions of sodium carbonate with SO2 flue gas with NO or NO2. Recall that some sodium nitrate accumulated in the reaction of sodium carbonate with NO2 (without SO2 present) as shown in Figure 5-16. These results illustrate that when SO2 is present it destroys the sodium nitrate that is not bound with water vapor. These results illustrate that when no source of water vapor was present in the system, the sodium nitrate that formed was destroyed by SO2 and O2. In these reactions of sodium carbonate and SO2 flue gas with NO and NO2 there is no source of water vapor to bind to any of the sodium nitrate that forms and protect it from the destructive reactions of SO2. Water vapor is not in this case evolved since there is no sodium bicarbonate present to decompose. As a result, any sodium nitrate that forms, was left unshielded and was destroyed by SO2 and O2. This finding was supported by the X-ray diffraction analysis since the X-ray diffraction analysis showed that the sulfation product formed was sodium sulfate Form I. Form I was the polymorphic form formed by reaction of sodium nitrate with SO2. Figure 5-17 shows that the amount of sodium sulfate accumulated by the reaction of sodium carbonate and SO2 flue gas with NO is considerably less then the amount accumulated by the reaction of sodium bicarbonate with SO2 flue gas with NO as shown in Figure 5-14. Less sodium sulfate accumulation is thought to be due to the fact that no

261

fresh surface area or surface activation occurs during the reaction with sodium carbonate as it does with sodium bicarbonate. 5.3.2.1 The effect of flue gas water vapor on sodium nitrate accumulation for sodium bicarbonate reaction with SO2 flue gas with NO. The effect of water vapor on the finite NOx removal period was studied. Laboratory experiments were conducted where water vapor was added to the flue gas in the form of humidity. Instead of using a standard flue gas mix of SO2 flue gas with NO, a humidified mix of SO2 flue gas with NO was used. The flue gas was mixed with 5% water vapor by bubbling the gas through a fritted glass sparge in a heated water bath. A more detailed description of the experimental methodology is described in Chapter 4 Section 4.4.1.2.2. Table 5-3 lists the humidified reactions studied in this section. Table 5-3 Reactions to study finite NOx removal period. Flue Gas Mixture N2 with humidity N2 with humidity SO2 flue gas with NO and humidity SO2 flue gas with NO and humidity Sorbent Type NaHCO3 NaHCO3 NaHCO3 NaHCO3 Temp 150C X X X Temp 250C X Figure No. Figure 5-19 Figure 5-20 Figure 5-21 Figure 5-22

First, Figure 5-19 shows the baseline reaction results for sodium bicarbonate in inert nitrogen with 5% humidity at 150C.

262

(a
100 90 Sodium Bicarbonate in Product % 80 70 60 50 40 30 20 10 0

Conversion of NaHCO3 in Inert Nitrogen Gas at 150C: Sodium Bicarbonate in the Product versus Reaction Time

25

50

75

100

125

150

175

200

225

250

275

300

Tim e (m inutes)

(b
Sodium Nitrate in Product %

Conversion of NaHCO3 in Inert Nitrogen Gas at 150C: Sodium Nitrate in the Product versus Reaction Time
100 90 80 70 60 50 40 30 20 10 0 0 25 50 75 100 125 150 175 200 225 250 275 300 Tim e (m inutes)

(c
Sodium Carbonate in Product %

Conversion of NaHCO3 in Inert Nitrogen Gas at 150C: Sodium Carbonate in the Product versus Reaction Time
100 90 80 70 60 50 40 30 20 10 0 0 25 50 75 100 125 150 175 200 225 250 275 300 Tim e (m inutes)

Sodium Sulfate in Product %

(d

Conversion of NaHCO3 in Inert Nitrogen Gas at 150C: Sodium Sulfate in the Product versus Reaction Time
100 90 80 70 60 50 40 30 20 10 0 0 25 50 75 100 125 150 175 200 225 250 275 300 Tim e (m inutes)

Figure 5-19 Conversion of sodium bicarbonate in humidified inert nitrogen at 150C. Gas composition was 95% nitrogen, 5% H2O. (a) residual sodium bicarbonate, (b) sodium nitrate, (c ) sodium carbonate, (d) sodium sulfate.

263

The decomposition reaction in inert nitrogen with water vapor shown in Figure 5-19 closely resembles the decomposition reaction without added water vapor at 150C shown in Figure 5-11. Both decomposition reactions possess similar time periods of decomposition. Evidently decomposition equilibrium is not significantly affected at this temperature by this change in humidity. The reaction temperature of 150C is far enough away from the temperature where decomposition is considered to be noticeable (111C). Figure 5-20 shows the baseline conversion results for the reaction of sodium bicarbonate in inert nitrogen with 5% water vapor at 250C.

264

(a
Sodium Bicarbonate in Product %

Conversion of NaHCO3 in Inert Nitrogen Gas at 250C: Sodium Bicarbonate in the Product versus Reaction Time
100 90 80 70 60 50 40 30 20 10 0 0 50 100 150 Time (m inute s) 200 250 300

(b
Sodium Nitrate in Product %

Conversion of NaHCO3 in Inert Nitrogen Gas at 250C: Sodium Nitrate in the Product versus Reaction Time
100 90 80 70 60 50 40 30 20 10 0 0 50 100 150 Time (m inute s) 200 250 300

(c
Sodium Carbonate in Product %

Conversion of NaHCO3 in Inert Nitrogen Gas at 250C: Sodium Carbonate in the Product versus Reaction Time
100 90 80 70 60 50 40 30 20 10 0 0 50 100 150 Time (m inute s) 200 250 300

(d
Sodium Sulfate in Product %

Conversion of NaHCO3 in Inert Nitrogen Gas at 250C: Sodium Sulfate in the Product versus Reaction Time
100 90 80 70 60 50 40 30 20 10 0 0 50 100 150 Time (m inute s) 200 250 300

Figure 5-20 Conversion of sodium bicarbonate in humidified inert nitrogen with 5% H2O at 250C. Gas composition was 95% nitrogen, 5% H2O. (a) residual sodium bicarbonate, (b) sodium nitrate, (c ) sodium carbonate, (d) sodium sulfate.

265

Up to this point, no experimental reactions have been conducted at 250C. As expected at the higher temperature, Figure 5-20 shows that the decomposition time period is significantly shorter compared to at 150C as shown in Figure 5-19. For this sodium bicarbonate decomposition in inert nitrogen with water vapor at 250C, the decomposition time period occurs within 3 minutes compared to the 15 minutes observed at 150C, in Figure 5-19. Figure 5-20 shows that 97 % of the sodium bicarbonate is converted to sodium carbonate in the first 3 minutes. This shows a more complete conversion of sodium bicarbonate to sodium carbonate compared to at 150C, speculated to be due to the higher temperature. At 150C Figure 5-19 shows that only 88% of the sodium bicarbonate is converted to sodium carbonate in the first 10 minutes. Next, the conversion results for the reaction of sodium bicarbonate and SO2 flue gas with NO with 5% water vapor at 150C are shown in Figure 5-21.

266

(a
100 90 Sodium Bicarbonate in Product % 80 70 60 50 40 30 20 10 0

Conversion of NaHCO3 in non standard SO2 Flue Gas with NO at 150o C: Sodium Bicarbonate Product versus Reaction Time

25

50

75

100

125

150

175

200

225

250

275

300

Tim e (m inutes)

(b
Sodium Nitrate in Product % 12 10 8 6 4 2 0 0

Conversion of NaHCO3 in non standard SO2 Flue Gas w/ NO at 150o C: Sodium Nitrate Product versus Reaction Time

25

50

75

100

125

150

175

200

225

250

275

300

Tim e (m inutes)

(c
100 Sodium Carbonate in Product % 90 80 70 60 50 40 30 20 10 0

Conversion of NaHCO3 in non standard SO2 Flue Gas w/ NO at 150o C: Sodium Carbonate Product versus Reaction Time

25

50

75

100

125

150

175

200

225

250

275

300

Tim e (m inutes)

(d
100 Sodium Sulfate in Product % 90 80 70 60 50 40 30 20 10 0

Conversion of NaHCO3 in non standard SO2 Flue Gas w/ NO at 150o C: Sodium Sulfate Product versus Reaction Time

25

50

75

100

125

150

175

200

225

250

275

300

Tim e (m inutes)

Figure 5-21 Conversion of sodium bicarbonate in humidified SO2 flue gas with NO at 150C. Gas composition was 14% CO2, 3% O2, 5% H2O, 78% N2, 0.19% SO2 and 0.25% NO. (a) residual sodium bicarbonate, (b) sodium nitrate, (c) sodium carbonate, (d) sodium sulfate. Note full scale on the sodium nitrate versus time graph is 12 %. The following observations were made from the conversion results in Figure 5-21: 267

Figure 5-21 shows that decomposition is inhibited. The 45-minute finite time period of decomposition is significantly longer than the 15-minute finite time period of decomposition measured for the baseline reaction in Figure 5-19. The 45-minute time period of decomposition is larger that the 15-minute finite time period of decomposition measured for the same reaction without humidity, as shown in Figure 5-12. Figure 5-21 shows that sodium nitrate accumulation occurs during the time period of sodium bicarbonate decomposition, during the first 45 minutes, and after decomposition ends, after 45 minutes. The accumulation after decomposition ends however is much less then the accumulation that occurs during decomposition. Recall from Figure 5-12 that no sodium nitrate accumulation occurred after decomposition ended for the flue gas without humidity. The small steady sodium nitrate accumulation that occurs after decomposition ends is thought to be due to available water vapor from the humidified flue gas in these experiments. The flue gas humidity is thought to bind to sodium nitrate formation by hygroscopic forces and then shield the sodium nitrate from destructive reaction with SO2 and O2. Figure 5-21 shows that sodium nitrate accumulation does not occur during surface area production. Recall from Figure 5-23 that surface area is primarily produced during the first 10 minutes. Recall that Figure 5-21 shows that for the reaction in SO2 flue gas with NO the total time period of decomposition is slowed to 45 minutes speculated to be due to sodium nitrate and sodium sulfate formation. Figure 5-21 shows that a small amount of sodium nitrate is initially accumulated but that sodium nitrate accumulation is suspended during the first ten minutes when the surface area is produced. Figure 5-21 shows that the bulk of the sodium nitrate accumulation occurs after the first 10 minutes

268

when the surface area has been produced. It is speculated that the vigorous physical changes of molecular restructuring from decomposition inhibit sodium nitrate accumulation. The conversion results obtained in this study for the reaction of sodium bicarbonate with humidified SO2 flue gas with NO with 5% H2O at 250C are shown in Figure 5-22.

269

(a
100 90 Sodium Bicarbonate in Product % 80 70 60 50 40 30 20 10 0 0

Conversion of NaHCO3 in SO2 Flue Gas with NO at 250C: Sodium Bicarbonate in the Product versus Reaction Time

25

50

75

100

125

150

175

200

225

250

275

300

Tim e (m inutes)

(b
Sodium Nitrate in Product % 12 10 8 6 4 2 0 0

Conversion of NaHCO3 in SO2 Flue Gas with NO at 250C: Sodium Nitrate in the Product versus Reaction Time

25

50

75

100

125

150

175

200

225

250

275

300

Tim e (m inutes)

(c
100 Sodium Carbonate in Product % 90 80 70 60 50 40 30 20 10 0 0

Conversion of NaHCO3 in SO2 Flue Gas with NO at 250C: Sodium Carbonate in the Product versus Reaction Time

25

50

75

100

125

150

175

200

225

250

275

300

Tim e (m inutes)

(d
100 Sodium Sulfate in Product % 90 80 70 60 50 40 30 20 10 0 0

Conversion of NaHCO3 in SO2 Flue Gas with NO at 250C: Sodium Sulfate in the Product versus Reaction Time

25

50

75

100

125

150

175

200

225

250

275

300

Tim e (m inutes)

Figure 5-22 Conversion of sodium bicarbonate in SO2 flue gas with NO at 250C. Gas composition was 14% CO2, 3% O2, 5% H2O, 78% N2, 0.19% SO2 and 0.25% NO. (a) residual sodium bicarbonate, (b) sodium nitrate, (c ) sodium carbonate, (d) sodium sulfate. Note full scale on the sodium nitrate versus time graph is 12 %.

270

For this reaction at 250C Figure 5-22 shows that there is a very small finite time period of decomposition. Sodium nitrate accumulation corresponds directly to this finite time period of decomposition. Sodium bicarbonate decomposes within the first 10 minutes, and sodium nitrate accumulation occurs during this first 10 minutes. Unlike the reaction at 150C shown in Figure 5-21, at this higher temperature with the humidity in the flue gas, sodium nitrate is not found to accumulate after decomposition ends. Sodium nitrate is found to decline after decomposition ends. This loss of accumulated sodium nitrate is speculated to be due to reaction of the accumulated sodium nitrate with SO2. It is speculated that at this temperature because the hygroscopic forces are not strong the sodium nitrate may not be well shielded from SO2 reaction. It is speculated that a higher water vapor concentration in the flue gas mixture is needed at this temperature to preserve the accumulated sodium nitrate. According to Figure 5-22, sulfation increases linearly beyond 60 minutes and is evidently not adversely affected by the dramatic surface area reduction noted in Figure 525. 5.3.3 Sodium bicarbonate surface area and microstructure changes that occur during reaction with SO2 flue gas with NO. Surface area measurements were made on product collected from the reaction of sodium bicarbonate with humidified SO2 flue gas with NO. Surface area measurements were made after incremental times of 0.8, 3.3, 10, 60, and 300 minutes at 150C and 250C. In addition, baseline surface area measurements were made on product from the reaction of sodium bicarbonate with inert humidified gas at the same time increments. A description of the experimental test procedure used is described in Chapter 4 Section 4.6. 271

Also a description of the humidified gas mixture used for these experiments is described in Chapter 4 Section 4.2.2.1 Table 4.3 and 4.4. The surface area analysis method used is described in Chapter 4, Section 4.3.3. Microstructure analysis of product from the reactions for times of 10 minutes and 300 minutes were made using the method described in Chapter 4, Section 4.3.4. Figure 5.23 shows the surface area results obtained for the reaction of sodium bicarbonate with inert nitrogen at 150C compared to the reaction of sodium bicarbonate with SO2 flue gas with NO at 150C.
Surface Area Development of Sodium Bicarbonate Reaction Products at 150o C
10 BET Surface Area (m 2/gram) 9 8 7 6 5 4 3 2 1 0 0 50 100 150 200 250 300 Tim e (minutes) Inert nitrogen gasw ith 5% w ater vapor SO2 flue gas w ith NO w ith 5% w aer vapor

Figure 5-23 BET surface area measurements versus time for the reaction products of sodium bicarbonate in inert nitrogen. Humidity of gas at 5% water vapor, reaction at 150C. Figure 5.23 shows that surface area is primarily formed during the first ten minutes of both of the reactions. Surface area is not significantly formed after the first 10 minutes in either reaction. Figure 5.23 shows that greater surface area forms for the baseline reaction in inert nitrogen with humidity compared to the reaction in SO2 flue gas with NO 272

with humidity. One possible reason for this is that the formation of sodium nitrate and the ongoing sodium sulfate formation both restrict surface area production from decomposition. The corresponding conversion tracking results can be found in Chapter 8 Figures 5-19 and 5-21 respectively. Figure 5-24 compares the surface microstructure from Scanning Electron Microscope (SEM) photos of the product formed by the reaction in SO2 flue gas with NO after 10 minutes of reaction compared with after 300 minutes of reaction at 150C.

Figure 5-24 Photo Micrographs of product particle surface from the reaction for sodium bicarbonate with SO2 flue gas with NO at 150C (a) after 10 minutes, (b) after 300 minutes. SO2 flue gas with NO at 150C. Figure 5-24(a) shows the high surface area popcorn like microstructure formed after 10 minutes from decomposition of the sodium bicarbonate. This microstructure closely resembles the microstructure found to form after 10 minutes in inert nitrogen. No discernible features in the microstructure formed in inert nitrogen were observed compared to that formed in the SO2 flue gas with NO at either 10 or 300 minutes even though surface area reductions were measured. The slight reductions in surface area that 273

are shown in Figure 5-23 after 300 minutes for both reactions respectively may however be due to small amounts of sintering. Figure 5-25 compares the surface area measurements for the reaction of sodium bicarbonate with humidified SO2 flue gas with NO with the surface area measurements for the reaction in humidified inert nitrogen at 250C.
Surface Area Development of Sodium Bicarbonate Reaction Products at 250o C
10 BET Surface Area (m 2/gram) 9 8 7 6 5 4 3 2 1 0 0 50 100 150 200 250 300 Tim e (m inutes) Inert nitrogen gas w ith 5% w ater vapor SO2 flue gas w ith NO w ith 5% w ater vapor

Figure 5-25 BET surface area measurements versus time for the reaction products of sodium bicarbonate in inert nitrogen and in SO2 flue gas with NO at 250C. Figure 5-25 shows that high surface area is formed initially within the first 10 minutes for both reactions. For the reaction in inert nitrogen, Figure 5-25 shows that considerably more sintering occurred for the reaction at 250C compared to at 150C as shown in Figure 5-24. In particular, for the reaction at 250C in SO2 flue gas with NO vapor, a dramatic drop in surface area is shown to occur after 10 minutes and remains out to 60 minutes. The product conversion graphs for these reactions are shown in Figures 5-20 and Figure 5-22 respectively.

274

Figure 5-26 compares the microstructure of the particle product after 10 minutes and 300 minutes for this reaction in SO2 flue gas with NO with water vapor at 250C.

Figure 5-26 Photo Micrographs of product particle surface from the reaction for sodium bicarbonate with SO2 flue gas with NO at 250C (a) after 10 minutes, (b) after 60 minutes. The microstructure after 10 minutes of reaction as shown in Figure 5-26(a) is not as popcorn-like as was formed at 150C after 10 minutes, as shown in Figure 5-26(a). The nodular structure shows distinguishable coalescence of the micelles. Figure 5-26(b) also shows a significant change in microstructure for the reaction after 60 minutes. Instead of a popcorn-like microstructure Figure 5-26(b) shows a structure composed of large densely packed re-crystallized grains. This structure is consistent with the drop in surface area shown in Figure 5-25. Figure 5-27 shows the phase diagram[80] for sodium nitrate with sodium sulfate.

275

Figure 5-27 Phase diagram[80] of sodium nitrate and sodium sulfate. Temperatures are given in C. The phase diagram shows a eutectic melt point at 301C, lower than either of the melt points of sodium sulfate or sodium nitrate. It is speculated that the re-crystallization is caused by this eutectic melt point depression from the sodium nitrate and sodium sulfate product mixture. 5.3.4 Sulfation product analysis for sodium bicarbonate reaction in SO2 flue gas without NO. Preliminary studies were undertaken to identify and establish the polymorphic phase formation of the sulfation reaction products formed by the reaction of sodium bicarbonate with SO2 flue gas without NO at 111C, 150 and 250C. Howatson et al.(1980) [8]

276

identified the polymorphic phase of the sodium sulfate product formed by the reaction of sodium bicarbonate and SO2 flue gas without NO. Sodium sulfate possesses a complex temperature-sensitive material structure exceeding that of any single material system excluding silica, according to Kracek (1930)[70]. Kracek (1930)[70] identified five distinct polymorphic states of sodium sulfate with respect to thermal analysis. Table 5-4 lists the common name, the crystallographic structure, the range of stability, the density and the X-Ray diffraction file of the five polymorphic phases of sodium sulfate. The phases are numbered with Roman numerals, I through V, beginning with the stable high temperature phase according to petrography convention. Table 5-4 Polymorphic phases of sodium sulfate Polymorphic Common Name Phase Form V Thenardite Form IV Form III --inverted Thenardite, altered Thenardite, metastable Thenardite ---high temperature phase Crystallographic Structure Orthorhombic --Orthorhombic Max Stable Temp. Range 25 to 195C 195 to 210C 25 to 244C Density 2.6640 --2.6970 XRD file 371465 --241132

Form II Form I

--Hexagonal

Not stable 244 to moltenC

-----

--1-990

Thermal studies of sodium sulfate have shown that at room temperature Form V and Form III are stable and may be found. Form V more commonly known, as thenardite is the polymorphic form of anhydrous sodium sulfate at ordinary room temperature. It is the form of sodium sulfate that is crystallized from solution. Form III more commonly

277

known as metastable thenardite is also found at ordinary room temperature but is formed only when the sodium sulfate has had a temperature excursion above 200C. The presence of water vapor has been found to accelerate phase transformations especially from Form III to Form V. Na2SO4 Form II has no region of stable existence at low pressures. The polymorphic phases of the sulfation product collected from conversion experiments of the reaction of sodium bicarbonate with SO2 flue gas without NO were studied. Polymorphic phase identification is accomplished through X-ray diffraction (XRD) analysis as described in Chapter 4 Section 4.3.2. Table 5-5 shows the X-ray diffraction results of the products formed by the listed reactions. Table 5-5 Summary of sulfation product analysis by X-ray diffraction Solid reactant NaHCO3 NaHCO3 NaHCO3 NaHCO3 Gas SO2 flue gas without NO Exp.411 SO2 flue gas without NO Exp 75 SO2 flue gas without NO Exp 419 SO2 flue gas without NO Exp 91 Temp. (C) 111 150 175 250 XRD of Products Na2SO3 Na2SO3, Na2SO3 Na2SO4-III Na2SO4-III

The X-Ray diffraction product analysis from the reaction of sodium bicarbonate with SO2 flue gas without NO at 150C after 720 minutes is shown in Figure 5-28.

278

(a
100 90 80 Intensity % 70 60 50 40 30 20 10 0

X-ray diffraction Analysis of Product from Reaction of Sodium Bicarbonate with SO2 Flue Gas without NO at 150C

15

20

25

30

35

40

45

50

55

60

65

2Theta (degrees)

(b
100 90 80 70 Intensity % 60 50 40 30 20 10 0 15

Joint Committee on Powder Diffraction (JCPDS) and International Center of Diffraction Data (ICDD) Standard for Sodium Sulfite, File #37-1488

20

25

30

35

40

45

50

55

60

65

2Theta (degrees)

(c
100 90 80 70 Intensity % 60 50 40 30 20 10 0 15

Joint Committee on Powder Diffraction (JCPDS) and International Center of Diffraction Data (ICDD) Standard for Sodium Bicarbonate, File #15-700

20

25

30

35

40

45

50

55

60

65

2Theta (degrees)

Figure 5-28 X-ray diffraction analysis of product from reaction of sodium bicarbonate with SO2 flue gas without NO at 150C. The scan in (a) is matched with the standard files for sodium sulfite in (b), and sodium bicarbonate in (c). Identical product was found for the reaction in SO2 flue gas without NO at 111C.

279

Figure 5-28(a) shows that the sulfation product analysis for this case matches the x-ray diffraction standard reference file for sodium sulfite (PDF 37-1488) Figure 7.4(b). Additionally the X-ray analysis shows that residual sodium bicarbonate (PDF 15-700) Figure 7.4(c). Similarly the same sulfation products were found to form by x-ray diffraction for the reaction at 111C (sodium bicarbonate with SO2 flue gas without NO after 600 minutes). Although not shown the x-ray diffraction analysis similarly found that sodium sulfite was formed for this reaction without NO at 111C. It was anticipated that sodium sulfite would be found for these reactions since Howatson et al.(1980) [8] had previously reported that sodium sulfite formed for this reaction at both 111C and 150C. Figure 5-29 shows the x-ray diffraction scan of the product collected from the reaction of sodium bicarbonate with SO2 flue gas without NO at 250C after 585 minutes.

280

(a
100 90 80 Intensity % 70 60 50 40 30 20 10 0

X-ray diffraction Analysis of Product from Reaction of Sodium Bicarbonate with SO2 Flue Gas without NO at 250C

15

20

25

30

35

40

45

50

55

60

65

2Theta (degrees)

(b
100 90 80 70 Intensity % 60 50 40 30 20 10 0 15

Joint Committee on Powder Diffraction (JCPDS) and International Center of Diffraction Data (ICDD) Standard for Sodium Sulfate III, File #24-1132

20

25

30

35

40

45

50

55

60

65

2Theta (degrees)

Joint Committee on Powder Diffraction (JCPDS) and International Center of Diffraction Data (ICDD) Standard for Sodium Bicarbonate, File #15-700

(c
Intensity %

100 90 80 70 60 50 40 30 20 10 0 15 20 25 30 35 40 45 50 55 60 65 2Theta (degrees)

Figure 5-29 X-ray diffraction analysis of product from reaction of sodium bicarbonate with SO2 flue gas without NO at 250C. The scan in (a) is matched with the standard files for sodium sulfate Form III in (b), and sodium bicarbonate in (c). 281

The analysis shows that the sulfation product diffraction scan, Figure 5-29 (a) matches the standard reference file for sodium sulfate polymorphic Form III (PDF 24-1132). Note that for this reaction at 250C no sodium sulfite was found to form as it was for the reaction at 111C and at 150C. These results correspond to Howatson et al.s (1980) [8] previous findings where sodium sulfate Form III was found after reaction at 250C. Although Kracek and Gibson (1930) [72] reported that sodium sulfate Form I would form under equilibrium conditions at 250C (above 241C), it is no surprise that polymorphic Form III (PDF 24-1132) is found since the phase studies of Kracek and Gibson (1930) [72] indicate that upon cooling sodium sulfate Form I the phase inverts to Form III unless otherwise stabilized. It should also be noted that if the Form III phase were exposed to high humidity from the atmosphere or another source the phase would invert to Form V. 5.3.5 Sodium bisulfite and sodium pyrosulfite decomposition products.

Sodium bisulfite, NaHSO3, and sodium pyrosulfite, Na2S2O5 (commonly called sodium meta-bisulfite), were studied here to determine their decomposition temperatures and associated conversion compounds. Sodium bisulfite and sodium pyrosulfite were materials studied previously by Verlaeten et al.(1991)[20]. Verlaeten et al.(1991)[20] acknowledged that sodium bisulfite readily decomposed to sodium pyrosulfite. Verlaeten et al. (1991)[20] studied the reactivity of sodium pyrosulfite with NO and O2 and speculated that it resulted in NOx removal according to the following reaction equations: 2NaHSO3 Na2S2O5 + H2O Na2S2O5 + 2NO + O2 NaNO2 + NaNO3 + 2SO2 (5.32) (5.33)

Within Verlaeten et al.(1991)[20] experimental studies, the decomposition of sodium pyrosulfite was not evaluated closely. As a result the decomposition of the sodium 282

pyrosulfite was evaluated in this study to establish its conversion product and temperature. TGA analyses of technical grade sodium bisulfite and sodium pyrosulfite were conducted. Analysis showed that the technical grade sodium bisulfite was not in fact sodium bisulfite but was converted to sodium meta-bisulfite. The manufacturer of the sodium bisulfite confirmed that the technical grade sodium bisulfite was actually shipped as sodium meta-bisulfite due to a natural conversion that occurs shortly after its manufacture. The TGA analysis of the weight percent versus temperature of sodium pyrosulfite is shown in Figure 5-30.
Thermal Gravametric Analysis of Sodium Pyrosulfite

110 100 90 80 70 60 50 40 30 20 10 0

Weight (%)

100

200 300 Temperature (C)

400

Figure 5-30 pyrosulfite

Thermal gravimetric analysis of sodium

The thermal gravimetric analysis in Figure 5-30 shows that sodium pyrosulfite starts at its original weight percent of 100% and begins to decompose near 140C. The weight percent drops and stabilizes at 66%. The 66 weight % is representative of the weight percent loss for conversion of sodium pyrosulfite to sodium sulfite where the molecular weight of sodium pyrosulfite is 190.10 and the molecular weight of sodium sulfite is 283

126.04. (126.04/190.1x100 = 66.3%) . The converted product obtained from this TGA decomposition was analyzed by wet chemistry and X-ray diffraction analysis and identified by both methods to be sodium sulfite. These findings provide evidence that clarify Verlaeten et al.s said claim that sodium pyrosulfite reacts directly with NO and O2 for NOx removal at or above 140C. Based on these TGA results sodium pyrosulfite is unstable near or above 140C. It is plausible that at or above 140C sodium pyrosulfite decomposes to sodium sulfite and the sodium sulfite actually engages in reaction with NO and O2 according to the following two postulated reaction equations: Na2S2O5 Na2SO3 + SO2 (above 140C) (5.34) (5.35)

Na2SO3 + 2NO + O2 NaNO2 + NaNO3 + SO2 5.4 Conclusions

This chapter provided the results of the preliminary studies conducted. First in this chapter, tests were conducted to check proper operation of the reactor apparatus constructed for laboratory study. These tests were intended to verify that the apparatus simulated reactions in a bag-house injection application. These tests showed that the reactor apparatus and NOx measuring equipment were operating properly. To verify the reactor apparatus, NO2 measurements with respect to temperature were collected from the reactor apparatus for the reaction of sodium bicarbonate with SO2 flue gas with NO. These NO2 measurements were compared to NO2 measurements previously reported by Bland (1990) [16] and were found to be in good agreement. The preliminary studies in this chapter included material characterization tests. Decomposition inhibition of sodium bicarbonate in SO2 flue gas with NO was studied by 284

tracking conversion. These conversion-tracking studies verified that sodium bicarbonate decomposition was indeed inhibited at 111C in SO2 flue gas with NO, as Bland had previously reported. Additionally, conversion tracking of various related reactions revealed that decomposition of sodium bicarbonate was inhibited for other flue gas mixtures containing NOx. Gas mixtures containing NO or NO2 without SO2 were found to result in decomposition inhibition despite the fact that these reactions resulted in very little sodium nitrate product formation. These experiments revealed the great influence that sodium nitrate accumulation, albeit small, has on decomposition inhibition. The amount of sodium nitrate accumulated in these tracking studies was found to impact the extent of decomposition inhibition where greater amounts of sodium nitrate accumulation resulted in greater sodium bicarbonate decomposition inhibition. It is speculated that decomposition inhibition results from effects of hygroscopic forces from sodium nitrate product accumulation. In particular the hygroscopic forces generated from the sodium nitrate product layer on the exterior of the particle are envisioned to inhibit decomposition by opposing water vapor evolution and instead binding it to the sodium nitrate. The preliminary studies in this chapter included material characterization studies of the finite time period of sodium nitrate accumulation. Conversion tracking was used to study this finite period of sodium nitrate accumulation. Blands previous finding that sodium nitrate accumulation is finite in nature for the reaction of sodium bicarbonate and SO2 flue gas with NO at 150C was experimentally verified. Conversion tracking experiments revealed that a finite time period of sodium nitrate accumulation existed and that it corresponded directly to the finite time period of sodium bicarbonate

285

decomposition for reactions in SO2 flue gas with NO. Sodium nitrate accumulation was found to cease when decomposition of sodium bicarbonate ended for the reaction in SO2 flue gas with NO. Theoretically it was speculated that sodium nitrate accumulated during the finite time period of sodium bicarbonate decomposition as a result of the water vapor evolved from decomposition. In particular, it was speculated that available water vapor evolved from decomposition became bound to sodium nitrate by hygroscopic forces. The water vapor bound to the sodium nitrate was speculated to shield and protect the sodium nitrate from destructive reaction with SO2 and O2, which would otherwise result in reaction and formation of NO2. It was speculated that when sodium bicarbonate decomposition ends sodium nitrate accumulation stops because no water vapor is available to bind and protect the sodium nitrate from reaction with SO2 and O2. Instead it is speculated that after sodium bicarbonate decomposition ends, sodium nitrate is converted to NO2 and sodium sulfate by reaction with SO2 and O2. For the reaction of sodium bicarbonate with either NO or NO2 without SO2 it was found that sodium nitrate accumulates continually in a non-finite manner. For this reaction case, sodium nitrate is speculated to accumulate continually whether there is a source of water vapor or not, because there is no SO2 in the flue gas to react with the sodium nitrate to destroy or convert it to NO2 and sodium sulfate. These findings importantly highlight the importance of both water vapor availability and the presence of SO2 in the flue gas in producing the finite period of sodium nitrate accumulation. Conversion tracking experiments in humidified flue gas revealed that some sodium nitrate accumulation occurs outside of the finite period of decomposition, since

286

the flue gas humidity provides a source of water vapor. It is speculated that the flue gas humidity provides some water vapor to bind and shield to sodium nitrate. Also Bortzs past humidity test results were found to be consistent with this theory. Bortzs past humidity tests showed that the amount of sodium nitrate accumulation increased in reactions where the flue gas humidity was increased. With greater flue gas humidity more water vapor becomes available to bind to sodium nitrate and shield or protect it from destructive reactions with SO2 and O2. The preliminary studies in this chapter included material characterization studies of surface area and microstructure changes of sodium bicarbonate during reaction with SO2 flue gas with NO as compared with reactions conducted in inert nitrogen and in SO2 flue gas with NO. This analysis confirmed that surface area increased during decomposition of the sodium bicarbonate for all reaction gas mixtures studied. At 150C the surface area formation for the reaction in SO2 flue gas with NO was found to be less than the surface area formation found for the reaction in inert nitrogen or in SO2 flue gas without NO. The reduced surface area production for the reaction in SO2 flue gas with NO was speculated to be due to sodium sulfate product formation. At 250C for the first 10 minutes high surface area formed for the reaction in SO2 flue gas with NO comparable to the surface area formation found for the reaction in inert nitrogen and in SO2 flue gas without NO. For the reaction in SO2 flue gas with NO, beyond 10 minutes at 250C, the surface area was found to dramatically drop. The microstructure analysis of product collected after 10 minutes at 250 C in SO2 flue gas with NO correspondingly showed structural re-crystallization. In this high temperature case the surface area and

287

microstructure changes are speculated to be due sintering brought on by the low melt eutectic that exists between sodium sulfate and sodium nitrate. The preliminary studies in this chapter included material characterization studies of sulfation product analysis for sodium bicarbonate reaction with SO2 flue gas without NO. These tests verified Howatson et al.s previous findings that at low temperatures sodium sulfite forms and at high temperatures sodium sulfate polymorphic Form III forms, and at temperatures between a mixture of sodium sulfite and sodium sulfate polymorphic Form III forms. The preliminary studies in this chapter included material characterization studies of sodium pyrosulfite and sodium bisulfite. These studies revealed that sodium bisulfite, at room temperature is unstable and converts to sodium pyrosulfite. Thermal gravimetric analysis along with x-ray diffraction analysis revealed that at or above 140C sodium pyrosulfite was unstable and decomposed to sodium sulfite. These findings clarified Verlaeten et al.s (1991)[20] previous claim that above 140C sodium pyrosulfite reacted with NO and O2 to produce sodium nitrite and sodium nitrate. These findings refute Verlaeten et al.s (1991)[20] claim that above 140C sodium pyrosulfite reacts with NO and O2 to form sodium nitrite and sodium nitrate. These findings instead provide evidence that above 140C the sodium pyrosulfite converts by decomposition to sodium sulfite and that sodium sulfite reacts with NO and O2 to form sodium nitrite and sodium nitrate 5.5 Follow on tests Further work is needed on the effect of water vapor on the reaction of sodium nitrate and SO2 and O2. accumulation and. Fresh sodium nitrate formed by reaction of 288

sodium bicarbonate or sodium carbonate with NO2 or formed by sodium sulfite with NO and O2 should be tested for its reactivity with SO2 and O2 to form NO2 and sodium sulfate. This same reaction should be repeated with a step where humidity is added after the sodium nitrate formation step to definitively show that absorbed humidity inhibits the SO2 and O2 reaction with sodium nitrate. Further studies are needed to evaluate the effect of increasing the water vapor percent on the sodium nitrate accumulation rate after decomposition ends. This study showed that no sodium nitrate accumulation occurred after decomposition for flue gas without water vapor, while some sodium nitrate accumulation after the end of decomposition was found when a flue gas mix with 5% water vapor was added. Follow on studies are needed to investigate sodium nitrate accumulation after the end of decomposition for flue gas with 10% and 15 % water vapor concentrations to better establish the trend. The preliminary studies that showed that sodium nitrate formation was linked to the inhibition of sodium bicarbonate decomposition suggests that follow on testing is needed to determine if likewise sodium nitrate product formation for the reaction of sodium pyrosulfite with flue gas containing NOx inhibits decomposition of the sodium pyrosulfite.

289

CHAPTER 6 VALIDATION OF THE REACTION OF SODIUM SULFITE WITH NO AND O2. 6.1 Introduction The reaction of sodium sulfite with NO and O2 is studied in this chapter. Recall from Chapter 3 Section 3.7.2.1.4 that this reaction is in the overall hypothesized reaction pathway for the reaction of sodium bicarbonate with SO2 flue gas with NO. Recall also that in particular a temperature dependant mixture of sodium nitrite and sodium nitrate is hypothesized to form by the reaction of sodium sulfite with NO and O2 according to the following reaction equations: Na2SO3 + 2NO + O2 2NaNO2 + SO3 Na2SO3 + 2NO + 2O2 2NaNO3 + SO3 Additionally the reaction of sodium sulfite with NO and O2 is a key reaction in the simplified three-step reaction pathway for NO2 formation reiterated in Table 6-1. Table 6-1 The simplified three-step reaction pathway for NO2 formation from the reaction of sodium bicarbonate and SO2 flue gas with NO. Step Number 1 Description of the reaction step. A sulfation step where sodium sulfite is produced by the reaction of sodium bicarbonate or sodium carbonate with SO2. 2NaHCO3 + SO2 Na2SO3 + 2CO2 + H2O Na2CO3 + SO2 Na2SO3 + CO2 A nitration step where sodium nitrate is produced by reaction of the sodium sulfite from step 1 with NO Na2SO3 + 2NO + 2O2 2NaNO3 + SO3 An NO2 release reaction step where NO2 is released from the collected sodium nitrate by reaction with SO2 in the flue gas. 2NaNO3 + SO2 Na2SO4 + 2NO2 (6.1) (6.2)

290

In Table 6-1 the second step of the three-step reaction path is shaded to highlight the reaction studied in this chapter. This chapter seeks to validate and provide evidence that sodium sulfite reacts with NO and O2 to form sodium nitrate as hypothesized. 6.1.1 Examination of past studies of the reaction of sodium sulfite with NO and O2 . In the literature there is little agreement on the outcome of the reaction of sodium sulfite and NO and O2. This reaction remains controversial in its validation status. Past studies of this reaction are closely examined in this section to reveal reasons why the past findings in the literature are not in agreement on the outcome of this reaction. Bland (1990)[16] and others theoretically speculated that the reaction of sodium sulfite with NO and O2 resulted in NO2 formation. In particular it was speculated that in this reaction sodium sulfite catalyzed NO to NO2. Bland (1990)[16] conducted screening tests of technical grade sodium sulfite with NO and O2 in a packed bed from 121C to 211C and found no NO2 formation and no NOx removal by the reaction. Verlaeten et al.(1991)[20] similarly tested the reaction of sodium sulfite with NO and O2 in a fluidized bed and also found no NO2 formation or NOx removal. To date, NO2 formation by this reaction has not been demonstrated. Miller (1986)[19] noted and recognized that sodium sulfite was easily contaminated under atmospheric conditions. Specifically Mellor (1947)[61] published in a handbook that sodium sulfite was rapidly oxidized at 15 C in moist air to sodium sulfate. Miller (1986)[19] did not test the reaction of sodium sulfite with NO and O2, but did test the reaction of sodium sulfite with SO2 flue gas with NO. For the reaction of sodium sulfite with SO2 flue gas with NO Miller (1986)[19] reported that a very small but 291

measurable amount of NO2 was formed. Miller (1986)[19] thought that the sodium sulfite used in these experiments was contaminated with surface oxidation and recommended that sodium sulfite reactant in future tests be used that was free of surface oxidation. No special method to prepare sodium sulfite free of surface oxidation contamination was suggested by Miller (1986)[19]. A number of past researchers reported NOx removal in the form of sodium nitrite and sodium nitrate for the reaction of sodium sulfite with NO and O2. Lai and Yesavage (1994)[21] conducted reactivity studies of sodium sulfite with NO and O2 and reported NOx removal in the form of sodium nitrite and sodium nitrate for these reactions. NO2 was not found to form in these tests by Lai and Yesavage (1994)[21]. Close examination of Lai and Yesavages (1994)[21] test methodology revealed that the sodium sulfite used in these tests was not plain technical grade sodium sulfite purchased and shipped from an outside manufacturer but was a special preparation of sodium sulfite. In particular technical grade sodium sulfite that was specially crushed with a mortar and pestle and carefully stored in a dry desiccator prior to testing was found to be reactive with NO and O2. Also sodium sulfite prepared by freshly reacting sodium bicarbonate with SO2 at low temperature in the test vessel where it was to be reacted with NO and O2 was found to be reactive with NO and O2. In both cases the sodium sulfite was specially prepared in a manner in which surface oxidation from moist air exposure was minimized. In the first case, mechanical crushing provided a method where fresh sodium sulfite from the particulate core was exposed and kept free of oxides by storing in a dry dessicator. In the second case the sodium sulfite tested was formed fresh in the test vessel free of air and moisture.

292

A number of past researchers including, Ackermann and Powell (1967)[53], Martin et al. (1981)[54], Ito et al. (1984)[55], Takeuchi et al. (1977)[56], Weisweiler and Blumhofer (1984)[57], Sada et al. (1988)[58], Littlejohn et al. (1986)[59] and Burnell et al. (1977)[76] reported NOx removal for the reaction of aqueous sodium sulfite with flue gas with NO and O2. Ito et al. (1984)[55] in particular noted that NOx was removed by an aqueous sodium sulfite solution when the pH was maintained above 5.5. It is relevant to note also that Verlaeten et al.(1991)[20] found NOx removal in the form of sodium nitrite and nitrate by the reaction of technical grade sodium bisulfite with NO and O2 and by reaction of technical grade sodium pyrosulfite with NO and O2. For both of these reactions Verlaeten et al.(1991)[20] reported that NO2 was not produced but that these reactions resulted in NOx removal in the form of sodium nitrite and sodium nitrate. Verlaeten et al.(1991)[20] postulated the following reaction equations: 2NaHSO3 Na2S2O5 + H2O Na2S2O5 + 2NO + O2 NaNO2 + NaNO3 + 2SO2 Lai and Yesavage (1994)[21] confirmed that sodium nitrite and sodium nitrate were formed by the reaction of sodium pyrosulfite with NO and O2 as claimed by Verlaeten et al.(1991)[20]. Because it was established in the Preliminary Tests in Chapter 5 Section 5.3.5 that sodium bisulfite and sodium pyrosulfite thermally decomposed to sodium sulfite at temperatures greater than 140 C, the reactivity tests of sodium bisulfite and sodium pyrosulfite above 140C by Verlaeten et al.(1991)[20] are interpreted in this study to show that sodium sulfite formed fresh by sodium pyrosulfite or sodium bisulfite thermal decomposition is reactive with NO and O2. (6.3) (6.4)

293

The discrepant findings between all of these past investigations show the importance that preparation of sodium sulfite has on the outcome of the tests. Miller (1986)[19] previously discussed the tendency of technical grade sodium sulfite to be poisoned by surface oxidation from exposure to moist air. Sodium sulfite prepared without exposure to moist air was found by these past researchers to be reactive with NO and O2. And sodium sulfite that was not prepared specially to avoid moist air exposure was repeatedly found by past researchers to be un-reactive with NO and O2. As suggested by Mellor it is maintained here that surface oxidation of the sodium sulfite occurs rapidly in moist air at room temperature. It is speculated that this surface oxidation of sodium sulfite to sodium sulfate prohibits the sodium sulfite in the core beneath from reaction with NO and O2. For Millers (1986)[19] fluidized bed tests of sodium sulfite with SO2 flue gas with NO it is speculated that the fluidized bed provided some means of mechanical abrasion to initially expose fresh sodium sulfite and react with NO and O2 and form sodium nitrite and sodium nitrate as reported by other past researchers, but because SO2 was in this case in the flue gas the sodium nitrate is speculated to react with the SO2 according to Bortzs (1994)[45] release theory where NO2 was released by the reaction of sodium nitrate with SO2, 2NaNO3 + SO2 Na2SO4 + 2NO2 + 2SO2 (6.5)

It is speculated that the sodium sulfate that forms by this reaction is more tightly bound to the particle then the initial surface layer of sodium sulfate oxidation that was formed at room temperature from atmospheric exposure to moist air. It is speculated that the formation of this more tightly bound sodium sulfate layer is not abraded by the

294

fluidization and results in an end to the reaction of sodium sulfite with NO and O2 due to ash layer resistance to diffusion 6.2 Experimental approach The reaction of sodium sulfite with NO and O2 was experimentally studied. In particular, different sodium sulfite preparations were evaluated for their reactivity with NO and O2. These tests were intended to validate that the reaction of sodium sulfite with NO and O2 results in sodium nitrate. These tests were also intended to show the importance of surface preparation on sodium sulfite reactivity with NO and O2. Technical grade sodium sulfite, and specially prepared sodium sulfite made by freshly reacting sodium bicarbonate with SO2 at low temperature in the test vessel where it was to be tested for reactivity with NO and O2 were both studied. Additionally reactivity tests of these two grades of sulfite were evaluated respectively for their reactivity with NO without oxygen. Table 6.2 lists the reactions conducted.

295

Table 6-2 Investigated reactions. Test No. 1 2 3 4 5 Sorbent Type Na2CO3 blank technical grade Na2SO3 high purity technical grade Na2SO3 high purity freshly made Na2SO3 freshly made Na2SO3 Flue Gas Mixture N2 with NO(50ppm) and O2 N2 with NO(50ppm) N2 with NO(50ppm) and O2 N2 with NO(50ppm) N2 with NO(50ppm) and O2 Reactor Temp Type 175C Fixed bed X Fixed bed X Fixed bed X Fixed bed X Fixed bed X

The experimental procedure used for these experiments is described in detail in Chapter 4 Section 4.4.2.1. Test number 2 and 4 are experiments designed to indicate whether reaction occurs without oxygen. Test numbers 1 through 5 were conducted in a fixed bed reactor with NO, NO2 and SO2 monitors at the inlet and outlet. NOx removal was measured by subtraction of inlet and outlet NO and NO2. The reaction products from these tests were analyzed using elemental analysis, X-ray diffraction, infrared (IR), Time of Flight static Secondary Ion Mass Spectrometry (TOF-SIMS), and Electron Spectroscopy for Chemical Analysis (ESCA). 6.3 Experimental results of the reaction tests.

Table 6.3 lists respectively the general results for the experiments conducted in the fixed bed reactor.

296

Table 6-3 Experimental results of sodium sulfite reactions Test No. 1 2 3 4 5 Type of Sodium Sulfite Na2CO3 blank Technical Grade Na2SO3 Technical Grade Na2SO3 Freshly Made Na2SO3 Freshly Made Na2SO3 Gas Mixture NO and O2 NO (without O2) NO and O2 NO (without O2) NO and O2 Observed Results No NOx removed, No NO2 formed No NOx removed, No NO2 formed No NOx removed, No NO2 formed No NOx removed, No NO2 formed NOx was removed, No NO2 formed

6.3.1

Experimental result of blank test The blank experiment resulted in no measurable NOx removal and no NO2

formation. A small short pressure induced swing disrupting NOx analysis was experienced when the oxygen gas valve was turned-on (in step 5 of the procedure described in Chapter 4 Section 4.4.2.1). This pressure swing resulted in a small short dip in the NOx gas analysis as shown in Figure 6.1.
NOx vs Time
100 NOx (ppm) 75 50 25 0 0 1 2 3 4 5 6 7

Blank Run: Na 2CO3 + NO + O2 @ 111o C exp 410

O2 on

Tim e (m in)

Figure 6-1 NOx analysis for blank run.

297

The dip in NO and NOx shown in Figure 6.1 was induced by a pressure swing when the oxygen valve was opened. Data from the first 30 seconds in subsequent reaction experiments were not considered relevant because of this observed pressure induced swing in the blank. Because no NOx removal was found for this reaction no product analysis was conducted. 6.3.2 Experimental results of technical grade sodium sulfite reaction with NO and O2. Technical grade sodium sulfite was tested for its reactivity with NO without O2 and then tested for its reactivity with NO with O2. The tests showed that the technical grade sodium sulfite was not reactive with either NO without oxygen or with NO with oxygen. Neither NOx removal nor NO2 formation was detected for either of the reaction tests with NO without oxygen or NO with oxygen. Because there was no reaction detected, no product analysis was conducted. It was speculated that this technical grade sodium sulfite did not react with NO and O2 because its surface was contaminated with sodium sulfate oxidation from exposure to moist air during its storage and handling under atmospheric conditions. 6.3.3 Experimental results of freshly made sodium sulfite reaction with NO and O2 . A special preparation of freshly made sodium sulfite was tested for its reactivity with NO without O2 and with NO with O2. The freshly made sodium sulfite was produced in the reactor vessel where it was subsequently tested with NO and O2 as described in more detail in Chapter 4 Section 4.4.2.1.1. A methodology was used where

298

the sodium sulfite was made by reacting sodium bicarbonate with SO2 flue gas at low temperature in the reactor vessel and maintained in this sealed vessel under inert nitrogen conditions prior to test. This special preparation that was not exposed to atmospheric conditions was then tested for reactivity with NO and O2 in this same test vessel without breaking the vessel open to the atmosphere. This special preparation was intended to provide sodium sulfite that was free of surface oxidation from atmospheric exposure to moist air. The tests of this special preparation of freshly made sodium sulfite were found to be un-reactive with NO without O2. Neither NO2 formation nor NOx removal was found for this reaction with this special sodium sulfite preparation with NO without oxygen. However with oxygen present, this special preparation of freshly made sodium sulfite was found to react with the NO and O2. This reaction was found to result in NOx removal as shown in Figure 6.2.
NOx vs Time
100 NOx (ppm)

Reaction: freshly-made Na2SO3 +NO+O2 @ 175oC exp 412

75 50 25 0 0 1 2 3 4 5

O2 on

Time (min)

Figure 6-2 NOx versus time for the reaction of freshly made sodium sulfite with NO and O2 at 175C. Figure 6.2 shows that for this reaction of the special preparation of freshly made sodium sulfite with NO and O2 the starting NO concentration was reduced to zero for over 4 299

minutes after the oxygen was added to the NO in the system. This NOx removal was clearly measurable and sustained relative to the disturbance in NOx experienced in the blank test when the O2 valve was opened. No NO2 formation was found to form for this reaction. A significant temperature rise was recorded for this reaction. The temperature rose from 175C to 284C within the first 2 minutes of the reaction start. This reaction is therefore considered exothermic consistent with the thermodynamic analysis conducted in Chapter 3 Section 3.5.3. The SO2 analyzer for this reaction did not detect SO2 formation. Within the conversion of the sodium sulfite to sodium nitrate it is speculated that SO3 is evolved by the reaction since SO2 was not detected. More follow on work is needed to verify this speculation. 6.3.3.1 Analysis of reaction product formed by specially prepared freshly made sodium sulfite with NO and O2. Various analysis methods were used to determine the composition of the solid product formed by the NOx removal reaction of freshly made sodium sulfite and NO and O2. The following list of analysis methods were implemented: elemental analysis, X-ray diffraction (XRD), infrared (IR), Time of Flight static Secondary Ion Mass Spectrometry (TOF-SIMS) and Electron Spectroscopy for Chemical Analysis (ESCA). Figure 6-3 shows a simplified diagram of the hypothetically reacted particle.

300

Diagram of Reacted Particle

Legend
Na2CO3 Core Freshly-made Na2SO3 Unknown Reaction Product

Figure 6-3 Diagram of reacted particle from experiment Figure 6-3 shows that the reacted particle includes an un-reacted core of sodium carbonate and a layer around this core of the specially prepared freshly formed sodium sulfite from the reaction of sodium bicarbonate with SO2, and an outer layer of unknown reaction product from the reaction of the specially prepared freshly formed sodium sulfite with NO and O2. Table 6.3 lists the depth of penetration of the various material analysis methods used.

301

Table 6-4 Comparison of sampling depth for each analysis method. Method" Sampling Depth Elemental Not applicable, bulk XRD 1000 A IR 50 A XPS ESCA 1-50A TOF-SIMS 1-10 A

X-ray diffraction and infrared analysis were not effective in identifying the product since the depth of penetration of these analysis methods was too deep. Elemental analysis was not effective in determining the precise compound formed in the outer surface layer of the particle but was useful in detecting nitrogen uptake to indicate that a compound containing nitrogen was formed. XRD and IR were unable to analyze surface layers. The product was identified through SIMS and ESCA analysis since these methods were capable of surface layer analysis. 6.3.3.1.1 Elemental Analysis

Bulk elemental analysis using the LECO elemental analyzer, described in more detail in Chapter 4 Section 4.3.1 was conducted on product collected from reaction with NO and O2. This analysis revealed the total nitrogen content in the particles. Baseline analysis was made on freshly made sodium sulfite particles to compare with the particles additionally reacted with NO and O2. This analysis was not intended to identify the crystallographic or molecular form of the product compound. But this analysis was instead used to indicate whether or not nitrogen was contained in the compound formed. Further more detailed surface analysis was conducted on samples found to contain nitrogen.

302

The elemental analysis of the reaction product from the specially prepared freshly formed sodium sulfite with NO and O2 found the product contained 0.2% nitrogen content outside of the baseline. This analysis definitively shows that the product of the reaction did in fact contain nitrogen. Although the nitrogen content found was small it was measurable. Material analysis was then used to identify the molecular compound formed containing nitrogen. The X-ray diffraction and infrared analysis both had a depth of penetration for analysis that exceeded the product layer thickness. As a result, these methods were unable to detect the surface product but instead were able to identify un-reacted sodium sulfite and the sodium bicarbonate beneath. 6.3.3.1.2 Time of flight static secondary ion mass spectrometry analysis.

Because XRD and IR were found to be unsuccessful in identifying the product layer formed on the outer most surface of the collected particles, a Time of Flight static Secondary Ion Mass Spectrometry (TOF-SIMS) analysis method was used. This method was used because of its known ability to identify surface mono-layer information. In Chapter 4 Section 4.3.5 the TOF-SIMS method is described in detail. This SIMS analysis showed that the product formed on the particle surface was a mixture of sodium nitrite and sodium nitrate. Table 6.5 lists the essential positive and negative ions for the reference standards of sodium nitrite and sodium nitrate according to Briggs et al.s (1989)[75] from the SIMS Handbook.

303

Table 6-5 Positive and negative ions in the reference standard for sodium nitrite and sodium nitrate Positive Ions [Na]+ [Na2O]+ [Na2OH] + Negative Ions [NO2][NO3] [NaNO2.NO3] [Na(NO3)2] Mass Charge 23 62 63 Mass Charge 46 62 131 147

The positive ion SIMS scan from 0 to 26 m/z for the analysis of the product is shown in Figure 6.4(a). Figure 6.4(b) provides the sodium nitrate reference standard SIMS scan for that range.

304

a
14000 12000 10000 Counts 8000 6000 4000 2000 0 0

SIMS Scan of Product: Na 2SO3 + NO + O2 Positive Ion Scan 0-26 m/z, normalized (x 1/5.3)

[Na]+

10

15

20

25

Mass Charge (m/z)

b
14000 12000 10000 Counts 8000 6000 4000 2000 0 0

SIMS Reference Scan of Sodium Nitrate Positive Ion Scan 0-26 [Na]+

10

15

20

25

Mass Charge (m /z)

Figure 6-4 SIMS Positive ion scan of high intensity peak counts, 0-14000, versus mass charge, 0-26 m/z. a) for unknown reaction product from NaSO3 + NO + O2 b) sodium nitrate reference standard. Note from Figure 6-4 that this SIMS positive scan of the product possesses the large Na+ peak at 23 that matches the sodium nitrate reference standard. The positive ion SIMS scan from 26 to 66 m/z are shown in Figure 6.5(a). Figure 6.5(b) provides the sodium nitrate reference standard SIMS scan for that range.

305

a
300 250 200 Counts 150 100 50 0

SIMS Scan of Product: Na 2SO3 + NO + O2 Positive Ion Scan 26-66 m/z, norm alized (x 5.3) [Cx Hy ]+ [Na2OH]+ [Na2O]
+

26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 56 58 60 62 64 66 Mass Charge (m /z)

b
300 250 200 Counts 150 100 50 0

SIMS Reference Scan of Sodium Nitrate Positive Ion Scan 26-66 [Cx Hy ]+ [Na2OH]+ [Na2O]+

26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 56 58 60 62 64 66 Mass Charge (m /z)

Figure 6-5 SIMS positive ion scan of high intensity peak counts, 0-300, versus mass charge, 26-66 m/z. a) for unknown reaction product from NaSO3 + NO + O2 b) sodium nitrate reference standard. Figure 6.5 shows that the SIMS scan of the product also contains the [Na2O]+ and the [Na2OH]+ peaks respectively at 62 and 63 m/z that matches the sodium nitrate reference standard. Hydrocarbon ions are also present in both the product scan and the standard reference scan attributable to handling and mounting. The negative ion SIMS scan from

306

0 to 100 m/z is shown in Figure 6.6(a). Figure 6.6(b) provides the sodium nitrate reference standard SIMS scan for that range.
a
24000 20000 16000 Counts 12000 8000 4000 0 0 10 20 30 40 50 60 70 80 90 100 Mass Charge (m /z)

SIMS Scan of Product: Na 2SO3 + NO + O2 Negative Ion Scan 0-100 m/z, normalized (x1/5.3)

[H][O][CH]-

[OH]-

[NO2][C2H][Cl]-

[NO3]-

b
24000 20000 16000 Counts 12000 8000 4000 0 0 10

SIMS Reference Scan of Sodium Nitrate Negative Ion Scan 0-100 m/z

[O]-

[OH][H][CH][C2H][Cl]-

[NO2]-

[NO3]-

20

30

40

50

60

70

80

90

100

Mass Charge (m/z)

Figure 6-6 SIMS negative ion scans of high intensity peaks counts, 0-24000, versus mass charge, 0-100 m/z. a) for unknown reaction product from NaSO3 + NO + O2 b) sodium nitrate reference standard. Figure 6.6 shows that both the product scan and the sodium nitrate reference scan possess [NO2]-, and [NO3]- peaks at 46 and 62 respectively. Other contaminants are present in

307

both samples, such as chlorine, oxygen, and hydroxide from handling. The negative ion SIMS scan from 100 to 200 m/z is shown in Figure 6.7(a). Figure 6.7(b) provides the sodium nitrate reference standard SIMS scan for that range.

a
1000 900 800 700 Counts 600 500 400 300 200 100 0 100 110

SIMS Scan of Product: Na 2SO3 + NO + O2 Negative Ion Scan 100-200 m/z, normalized (x 1/5.3)

[Na(NO3)2]-

[NaNO2.NO3]-

120

130

140

150

160

170

180

190

200

Mass Charge (m /z)

b
1000 900 800 700 Counts 600 500 400 300 200 100 0 100 110

SIMS Reference Scan of Sodium Nitrate Negative Ion Scan 100-200 m/z

[Na(NO3)2][NaNO2.NO3]-

120

130

140

150

160

170

180

190

200

Mass Charge (m /z)

Figure 6-7 SIMS negative ion scans of high intensity peaks counts, 0-1000, versus mass charge, 100-200 m/z. a) for unknown reaction product from NaSO3 + NO + O2 b) sodium nitrate reference standard.

308

Figure 6.7 shows that both the product and the sodium nitrate reference standard possess [NaNO2.NO3]-, and [Na(NO3)2]- peaks at 131 and 147 respectively. The unknown outer reaction product layer shown in Figure 6.3 has been determined by this SIMS analysis. All of the major peak locations in the SIMS scans of the product layer shown in Figures 6-4 through 6-7 match the peak locations of the sodium nitrate reference SIMS scan. The relative intensity of each of these peaks however is not proportional to the intensity of each of the reference peaks in the SIMS scan. Some of the peak intensities in the sodium nitrate reference are larger then those found in the product SIMS scan. This finding is an indication that sodium nitrite may be together with the sodium nitrate in the product. ESCA/XPS analysis is used to further determine more definitively the presence of sodium nitrite and sodium nitrate together. 6.3.3.1.3 Electron Spectroscopy for chemical analysis

X-ray photoelectron Spectroscopy (XPS) also known as Electron Spectroscopy for Chemical Analysis (ESCA) was used to analyze the reaction product formed by the reaction of sodium sulfite with NO and O2. This method was selected because it is capable of analyzing surface layers as indicated from Table 6-2; and because this method provides detailed compound specificity. This method is able to differentiate and identify the presence of similar compounds such as sodium nitrite and sodium nitrate that have different oxidation states. This method is described in further detail in Chapter 4 Section 4.3.6. Figure 6.8 shows the ESCA scan of the unknown outer product layer.

309

Figure 6-8 ESCA scan of product from the reaction of sodium sulfite with NO and O2.

310

In the ESCA scan in Figure 6.8 nitrogen in multiple chemical states is detected in small quantities on the surface of the powder. The binding energies are consistent with sodium nitrite and sodium nitrate along with organic nitrogen from handling. 6.4 Conclusions The reaction of sodium sulfite with NO and O2 was hypothesized to form a temperature dependent mixture of sodium nitrate and sodium nitrate in the overall reaction pathway for the reaction of sodium bicarbonate with SO2 flue gas with NO. In the pathway, the formation of sodium nitrite and sodium nitrate were hypothesized as separate elemental reactions, Na2SO3 + 2NO + O2 2NaNO2 + SO3 Na2SO3 + 2NO + 2O2 2NaNO3 + SO3 2Na2SO3 + 4NO + 3O2 2NaNO3 + 2NaNO2 + 2SO3 (6.6) (6.7) (6.8)

Reaction equation 6.7, the reaction that forms sodium nitrate, is an important reaction with respect to the overall reaction pathway. This reaction is a key reaction in the threestep reaction pathway for NO2 formation where NO2 was hypothesized to be released from the sodium nitrate by a subsequent reaction with SO2. In the literature there is little agreement on the outcome of the reaction of sodium sulfite and NO and O2. Some past researchers maintained that NO2 was produced by the reaction. But to date NO2 formation by this reaction has not been demonstrated by any past researcher or through experiments run in this study. Some past researchers have claimed that sodium sulfite is un-reactive with NO and O2 and produces nothing. Miller recognized that it was very difficult to conduct accurate reactivity tests using sodium sulfite. Miller noted that most laboratory grade sodium sulfite samples possess surface 311

oxidation from shelf life exposure and handling under atmospheric conditions in moist air. Recall that it has been well established that sodium sulfite readily oxidizes in moist air. Close examination of a number of past laboratory tests of sodium sulfite reveals that sodium sulfite prepared free of surface oxidation and maintained under inert conditions without moist air exposure was reactive with NO and O2. In particular fresh un-oxidized sodium sulfite formed by mechanically crushing laboratory grade sodium sulfite and storing in a dry desiccator was found by past investigations to react with NO and O2. And fresh un-oxidized sodium sulfite formed by reacting sodium bicarbonate with SO2 at low temperature in the test vessel where it was subsequently reacted with NO and O2 was found to be reactive with NO and O2. Also because it was established in the preliminary tests in Chapter 5 Section 5.3.5 that sodium bisulfite and sodium pyrosulfite thermally decomposed to sodium sulfite at temperatures greater than 140C, the reactivity tests of sodium bisulfite and sodium pyrosulfite above 140C by Verlaeten et al.(1991)[20] may be interpreted to show that sodium sulfite formed fresh by sodium pyrosulfite or sodium bisulfite thermal decomposition is reactive with NO and O2. In this chapter sodium nitrite and sodium nitrate were found to form by the reaction of sodium sulfite with NO and O2 as hypothesized. In this study a special sodium sulfite preparation that was free of surface oxidation from atmospheric exposure was tested for its reactivity with NO and O2 and compared to the reactivity of technical grade sodium sulfite that was exposed to atmospheric conditions. These tests showed that sodium sulfite prepared fresh and free of surface oxidation from moist air exposure was reactive with NO and O2. These tests also showed that sodium sulfite that was not kept

312

away from moist air exposure, such as technical grade sodium sulfite, was un-reactive with NO and O2. In this study sodium sulfite was specially prepared similar to the preparation that Lai and Yesavage (1994)[21] used in their reactivity tests, but in this study the preparation and reactivity tests were conducted in a fixed bed reactor rather than in a continuously stirred reactor (CSR) as Lai and Yesavage (1994)[21] used. The fixed bed reactor used in this study did not have leaks to the atmosphere, as was the case of the CSR reactor used by Lai and Yesavage (1994)[21]. But like Lai and Yesavage (1994)[21], sodium sulfite was formed by reacting sodium bicarbonate with SO2 at low temperature in the test vessel where it was subsequently reacted with NO and O2. The test of this specially prepared sodium sulfite with NO and O2 was found to produce NOx removal. These tests found no NO2 formation by this reaction, refuting past claims. Elemental analysis showed that the product contained nitrogen. After extensive analysis the particular reaction product was identified. These material tests showed that the reaction product resided on the outermost edge of the particles in a very thin layer. Due to this location, special material analysis, Time of Flight static Secondary Ion Mass Spectrometry (TOF-SIMS) analysis and by Electron Spectroscopy for Chemical Analysis (ESCA) were used. It was found that sodium nitrite and sodium nitrate product had formed by the reaction. SO2 analysis during the reactivity tests indicated that no SO2 was generated by the reaction. It is speculated as hypothesized that SO3 is instead generated and reacts with sodium carbonate in the core of the particle to form sodium sulfate. Also from the reaction tests the reaction of sodium sulfite with NO and O2 was found to be exothermic,

313

consistent with the findings of the thermodynamic analysis in Chapter 3 Section 3.5.3. A significant temperature rise was recorded from this reaction of specially prepared sodium sulfite with NO and O2. These tests also showed that oxygen was required for the reaction to proceed with NO. 6.5 Areas for further studies Follow on tests are needed to definitively show that freshly, thermally decomposed sodium pyrosulfite subsequently reacted with NO and O2 produces sodium nitrite and sodium nitrate. Follow on work is needed to verify that sulfur trioxide gas is produced by the reaction of sodium sulfite and NO and O2. The use of a special SO3 analyzer was not available for this study but should be pursued in future studies. In addition it is suggested that pure freshly formed sodium sulfite, without a core of sodium bicarbonate or sodium carbonate, be used in future studies to test for SO3 formation. A core of sodium bicarbonate or sodium carbonate may bias SO3 detection since any SO3 produced by the sodium sulfite reaction may react with sodium bicarbonate or sodium carbonate in the core and result in sodium sulfate formation. It is recommended that instead of using partially converted sodium sulfite made from a sodium bicarbonate conversion with SO2 as diagramed in Figure 6.3, it is preferable to start with sodium bisulfite or sodium pyrosulfite and heat and hold above the decomposition temperature to convert to fresh sodium sulfite. Converting sodium bisulfite or sodium pyrosulfite thermally has the advantage of providing fresh sodium sulfite without a core of sodium bicarbonate or sodium carbonate. This conversion technique has the advantage that complete conversion could be achieved in a relatively short period of time compared the sodium 314

bicarbonate conversion technique. The sodium bisulfite or sodium pyrosulfite conversion method is thought to be a better method for accurate SO3 detection. Follow on work is also needed to study the polymorphic form of the sodium sulfate product formed by reaction of SO3 and sodium bicarbonate or sodium carbonate. Follow on work is also needed to determine the effect of temperature on the resulting percent mix of sodium nitrite and sodium nitrate formed for the reaction of sodium sulfite with NO and O2. A study analogous to the study that Howatson et al. (1980) conducted on the reaction of sodium bicarbonate with SO2 and O2 is needed for the reaction of sodium sulfite and NO and O2. In particular this information would be useful for modeling and mitigation. Additionally follow on work is needed to study the product formation from the reaction of sodium sulfite with NO2. Follow on tests are needed to show that the specially prepared sodium sulfite made in the reactor by the reaction of sodium bicarbonate and SO2 at low temperature becomes poisoned by exposure to moist air. It should be shown that freshly formed sodium sulfite exposed to moist air becomes poisoned and is un-reactive with NO and O2.

315

CHAPTER 7

VALIDATION OF THE REACTION OF SODIUM NITRITE WITH SO2 7.1 Introduction The reaction of sodium nitrate with SO2 was studied in this chapter. This chapter sought to validate and provide evidence that NO2 is formed by the reaction of sodium nitrate with SO2. Recall from Chapter 3 that in the overall hypothesized reaction pathway of sodium bicarbonate with SO2 flue gas with NO that this reaction of sodium nitrate with SO2 is hypothesized to form NO2. 2NaNO3 + SO2 Na2SO4 + 2NO2 This reaction was also hypothesized to be a key reaction in the simplified three-step reaction pathway developed in Chapter 4, Section 4.2.2 for NO2 formation for the reaction of sodium bicarbonate with SO2 flue gas with NO. Table 7-1 reiterates the simplified three-step reaction pathway for NO2 formation hypothesized. Table 7-1 The simplified three-step reaction pathway for NO2 formation from the reaction of sodium bicarbonate and SO2 flue gas with NO. Step Description of the reaction step. Number 1 A sulfation step where sodium sulfite is produced by the reaction of sodium bicarbonate or sodium carbonate with SO2. 2NaHCO3 + SO2 Na2SO3 + 2CO2 + H2O Na2CO3 + SO2 Na2SO3 + CO2 2 A nitration step where sodium nitrate is produced by reaction of the sodium sulfite from step 1 with NO Na2SO3 + 2NO + 2O2 2NaNO3 + SO3 3 An NO2 release reaction step where NO2 is released from the collected sodium nitrate by reaction with SO2 in the flue gas. 2NaNO3 + SO2 Na2SO4 + 2NO2 (7.1)

316

Note that the reaction of sodium nitrate with SO2 as shaded in Table 7-1 is the third step of the simplified three-step reaction pathway for NO2 formation.

7.1.1

Examination of past studies of the reaction of sodium nitrate with SO2. There is little agreement in the past literature on the outcome of the reaction of

sodium nitrate with SO2. Bortz (1994)[45] speculated that NO2 was released by the reaction of sodium nitrate and SO2 according to reaction equation 7.1. Kozlowski et al. (1970) [5] noted that as early as 1892, Hodgkinson and Young found reaction between SO2 and solid nitrates. Kozlowski et al. (1970) [5] notes that Hodgkinson and Young reported that this reaction resulted in complete conversion of the nitrate to sulfate along with an evolution of nitrous fumes. Kozlowski et al. (1970) [5] in the laboratory reported reaction between molten alkali nitrates at 325C with SO2. Kozlowski et al. (1970) [5] like Hodgkinson and Young (1892) found sodium sulfate and gaseous NO2 formation by this reaction. Likewise, Quanquin et al (1967)[77] in U.S. Patent 3,348,914 stated that aqueous alkali metal nitrate solutions mixed with sulfuric acid resulted in conversion of the metal nitrates to sulfates, and the release of nitric oxide. Bland (1990) [16] on the other hand, screened in the laboratory, technical grade sodium nitrate with SO2 and O2 in a fixed bed reactor and reported that there was no NO2 formation and no NOx removal. Sodium nitrate is hygroscopic in nature and as a result attracts water vapor to its surface. The following reaction equation was hypothesized in Chapter 4 Section 4.1.1, Table 4-1 to reflect the hygroscopic attraction of water vapor to the surface of sodium nitrate,

332

NaNO3 + 4H2O NaNO34(H2O)

(7.2)

In the manufacture of sodium nitrate, its hygroscopic nature is well known and special surface additives are commonly added to sodium nitrate powders to keep them clumpfree. Water vapor bound to the surface of sodium nitrate is postulated in this study to affect its reactivity with SO2. It is thought that surface bound water vapor acts to shield sodium nitrate from reaction with SO2. The tests of Kozlowski et al. (1970) [5], which showed NO2 formation for the reaction of sodium nitrate, and SO2, were conducted at relatively high temperature where hygroscopic forces are speculated to be weak. Additionally in Quanquin et al.s (1967)[77] tests the sodium nitrate tested was in an aqueous state where hygroscopic forces did not apply since in the aqueous solution nitrate ions were free and mobile for reaction with SO2. 7.2 Experimental approach The reaction of sodium nitrate with SO2 was experimentally studied. In particular, different grades of sodium nitrate were evaluated for reactivity with NO and O2. These tests were intended to validate that sodium nitrate reacts with SO2 to produce NO2 and sodium sulfate. Special attention was given to the preparation of the sodium nitrate used in these tests. Because sodium nitrate was known to be hygroscopic in nature, having a high propensity to absorb water vapor from the atmosphere it is exposed to, special precautions were taken in these studies to avoid water vapor from binding to the sodium nitrate prior to test. Three preparations of sodium nitrate were tested in this chapter for their respective reactivity with SO2 and O2 . Ultra high purity technical grade sodium nitrate was one grade of sodium nitrate tested. Also tests were conduct with 333

sodium nitrate made by freshly reacting sodium carbonate with NO2 at low temperature in the same reactor vessel where it was tested with SO2 and maintained under inert dry conditions prior to test. Similarly tests were conducted with sodium nitrate freshly made by the two-step process described in steps one and two of the simplified three-step reaction pathway listed in Table 7-1. This freshly made sodium nitrate was made by first reacting sodium bicarbonate with SO2 to form sodium sulfite at low temperature, and then reacting this sodium sulfite with NO and O2 to form a mixture of sodium nitrite and sodium nitrate as discussed in Chapter 6. This freshly formed sodium nitrate was made in the same test vessel where it was tested for its reactivity with SO2, and maintained under inert dry conditions prior to test without exposure to the atmosphere. Table 7-2 lists all of the described reactions investigated in this chapter. Table 7-2 Selected reactions for reactivity study of sodium nitrate with SO2 and O2. No. Solid reactant Gas Temperature (C) 1 Blank SO2, O2 175 2 technical grade NaNO3 SO2 175 3 technical grade NaNO3 SO2, O2 175 4 freshly made NaNO3 175 SO2 [Na2CO3+NO2] 5 freshly made NaNO3 111, 150, 175, SO2, O2 200, 250 [Na2CO3+NO2] 6 freshly made NaNO3 175 SO2 [Na2SO3+NO,O2] 7 freshly made NaNO3 111, 150, 175, SO2, O2 200, 250 [Na2SO3+NO,O2] SO2, O2 8 Product of 175 {[(Na2SO3+NO,O2)+SO2,O2] +NO, O2} All of the reactions listed in Table 7-2 were evaluated at 175C except for reactions 5 and 7 in Table 7-2, which were evaluated additionally at 111C, 150C, 334

175C, 200C, and 250C. The experimental procedure used for these experiments is described in detail in Chapter 4 Section 4.4.2.2. Test number 2, 4 and 6 are experiments designed to indicate whether reaction occurs without oxygen. All of the tests were conducted in the fixed bed reactor with NO, NO2 and SO2 monitors at the inlet and outlet. NOx removal was measured by subtraction of inlet and outlet NO and NO2. The reaction products from these tests were analyzed using X-ray diffraction. 7.3 Experimental results of reaction tests. Table 7-3 lists respectively the general results for the experiments conducted in the fixed bed reactor. Table 7-3 Reaction results for reactivity study of sodium nitrate with SO2 and O2 at 175C. No. Solid reactant Gas NO or NO2 formation 1 blank SO2, O2 No 2 technical grade NaNO3 SO2 No 3 technical grade NaNO3 SO2, O2 No 4 freshly made NaNO3 No SO2 [Na2CO3+NO2] 5 freshly made NaNO3 Yes SO2, O2 NO2 formed [Na2CO3+NO2] 6 freshly made NaNO3 Yes SO2 NO formed [Na2SO3+NO,O2] 7 freshly made NaNO3 Yes SO2, O2 NO and NO2 [Na2SO3+NO,O2] formed SO2, O2 8 Product of No {[(Na2SO3+NO,O2)+SO2,O2] +NO, O2}

335

7.3.1

Experimental results of blank test. The blank experiment resulted in no measurable NOx removal and no NO2

formation. A small short pressure induced swing disrupting NOx analysis was experienced when the oxygen gas valve was turned-on (in step 6 of the procedure). This pressure swing resulted in a small short dip in the NOx gas analyzer signal. 7.3.2 Experimental results of SO2 and O2 reaction with technical grade sodium nitrate. Because past researchers such as Bland tested the reactivity of technical grade sodium nitrate with SO2 and O2 it was reinvestigated. The technical grade sodium nitrate tested was an ultra pure laboratory grade material (99.9999 NaNO3) shipped from the manufacturer in a vacuum-sealed container. Despite the high purity and lack of extensive handling after manufacture, this preparation of sodium nitrate no reaction was found to occur when reacted with SO2 and with SO2 and O2. It was suspected that surface contamination on the technical grade sodium nitrate from atmospheric exposure to humidity may have obstructed the sodium nitrate from reacting with SO2 or with SO2 and O2. 7.3.3 Experimental results of SO2 and O2 reaction with specially prepared sodium nitrate prepared fresh by reaction of sodium carbonate with NO2. To avoid surface contamination, sodium nitrate was made fresh in the reactor where the reactivity test was conducted. From the work of Knight (1977)[18] as described in Chapter 2, Section 2.2.2.1, sodium nitrate was formed by the reaction of sodium carbonate with NO2 according to the following equation,

336

Na2CO3 + NO2 2NaNO3 + NO + CO2

(7.3)

As a result, sodium nitrate was prepared fresh for reactivity testing by reacting sodium carbonate and NO2. The detailed methodology of this preparation is described in Chapter 4 Section 4.9.2. This special preparation of sodium nitrate was tested for its reactivity with SO2 and with SO2 and O2. No reactivity was found for this specially prepared sodium nitrate with SO2 alone. This specially prepared sodium nitrate was however found to be reactive with SO2 and O2. NO2 and sodium sulfate were found to form by this reaction. 7.3.3.1 The effect of temperature on NO2 formation from the reaction of the specially prepared sodium nitrate and SO2 and O2. The effect of temperature on the reaction of this specially prepared sodium nitrate with SO2 and O2 was studied. The resulting amount of NO2 produced versus temperature for this reaction is shown in Figure 7.1. For comparison, the amount of NO2 versus temperature formed by the reaction of sodium bicarbonate and SO2 flue gas with NO as reported by Bland (1990) [16] is shown in Figure 7.1.

337

300

250

Peak NO2 Formation (ppm)

200

Reaction of
150

NaNO3 + SO2,O2 Collected data

100

Reaction of
50

NaHCO3 + SO2

0 0 50 100 150 200 250 300 350 400

Temperature (C)

Figure 7-1 NO2 formation versus temperature for the reaction of NaNO3 with SO2 and O2. The dotted curve is Blands reported findings for NO2 formation by the reaction of NaHCO3 with SO2 flue gas with NO. Data sets are fit with splines. Gas composition was 3% O2, 96.7% N2, 0.3% SO2. Figure 7.1 shows that the amount of NO2 formation versus temperature from the reaction of SO2 and O2 with this special preparation of sodium nitrate corresponds to the general trend of NO2 formation from the reaction of sodium bicarbonate in SO2 flue gas with NO. Both reactions produce a maximum NO2 formation between 175C and 200C, and both show that NO2 formation is low at temperatures below 150C and above 275C.

338

7.3.4

Experimental results of SO2 and O2 reaction with specially prepared sodium nitrate prepared fresh by reaction of sodium sulfite with NO and O2 .

It was established in Chapter 6 that the reaction of sodium sulfite with NO and O2 resulted in the formation of fresh sodium nitrate and sodium nitrite. This special preparation of sodium nitrate is referred to in this study as the product of the two-step reaction, where the following are the two steps involved in its formation: A) The reaction of sodium bicarbonate with SO2 and O2 at low temperature to produce sodium sulfite. 2NaHCO3 + SO2 Na2SO3 + 2CO2 + H2O B) The reaction of the step A sodium sulfite product with NO and O2. Na2SO3 + 2NO + 2O2 2NaNO3 + SO3 Na2SO3 + 2NO + O2 2NaNO2 + SO3 (7.5) (7.6) (7.4)

The detailed methodology of this special preparation is described in Chapter 4 Section 4.4.1.2.2.1. NO was found to form in addition to NO2 by the reaction of the specially prepared three-step nitrate with SO2 and O2. The NO versus temperature for this reaction is shown in Figure 7.2.

339

NO Formation by 3 Step Sequence versus Temperature

180 160 140 120 NO (ppm) 100 80 60 40 20 0 0 50 100 150 Tem perature (C) 200 250 300

Figure 7-2 NO formation versus temperature for the threestep reaction sequence. The three step reaction sequence is comprised of: (1) sodium bicarbonate reaction with SO2, (2) the product reaction with NO and O2, (3) the product reaction with SO2 and O2. Figure 7.3 shows that NO formation increases with temperature above 150C. The NO found to form from these reactions is speculated to be a result of the reaction of SO2 and O2 with sodium nitrite from the two-step preparation process. The following are postulated reaction equations for the reaction of the specially prepared two-step nitrate with SO2 and O2: 2NaNO2 + SO2 + O2 Na2SO4 + 2NO2 2NaNO2 + SO2 Na2SO4 + 2NO 2NaNO3 + SO2 Na2SO4 +2NO2 (7.7) (7.8) (7.9)

The formation of NO by the equation 7.8 is consistent with Blands (1990) [16] previous findings that NO is released when sodium nitrite is reacted with SO2.

340

The product formed in these reactions is referred to as the product of the three-step reaction where equations 7.7, 7.8, and 7.9 are the third-step of the reaction sequence in the overall reaction pathway for NO2 formation. The nitration product (sodium nitrite and nitrate) from this two-step process was also tested for its reactivity with SO without oxygen. Without the oxygen present in the gas the reaction was found to produce only NO. The solid product formed by this reaction without oxygen was not analyzed for sulfation product identification. 7.3.4.1 The effect of temperature on NO2 formation by the SO2 reaction with specially prepared sodium nitrate prepared fresh by reaction of sodium sulfite with NO and O2. The nitration product from the two-step (sodium nitrite and nitrate) process was tested for its reactivity with SO2 and O2. It was found that NO2 and sodium sulfate was formed. The amount of NO2 produced by this three-step reaction path is graphed versus temperature in Figure 7-3.

341

300

250

Peak NO2 Formation (ppm)

Legend
200

Three Step Reaction:

1)

150

NaHCO3 + SO2 + NO, O2 + SO2,O2 scale factor 0.62 R i f

2)

100
3)

50

0 0 50 100 150 200 250 300 350 400

Temperature (C)

Figure 7-3 NO2 formation from the three step reaction sequence versus temperature. NO2 formation from the three-step reaction sequence (solid line) compared with the reaction of sodium bicarbonate with SO2 flue gas with NO (dotted line). The three-step reaction sequence is comprised of: (1) sodium bicarbonate reaction with SO2, (2) the reaction product (sodium sulfite) reacted with NO and O2, (3) the reaction product (sodium nitrite and nitrate) reacted with SO2 and O2. Both data sets are fit with splines. The trend in NO2 formation by the three-step reaction sequence with respect to temperature in Figure 7.2 closely matches the previous trend reported for the other specially prepared sodium nitrate. Also the trend in NO2 formation corresponds to the trend in NO2 formation found by Bland for the reaction of sodium bicarbonate and SO2 flue gas with NO as shown in Figure 7.1 and 7.2.

342

7.3.5

Experimental results of SO2 and O2 reaction with specially prepared sodium nitrate prepared fresh by the reaction of the three-step product with NO and O2. In an effort to verify that sodium sulfate was produced by the three step reaction

sequence, the product of the three-step reaction sequence described above was reacted with NO and O2 and checked for sodium nitrate formation. The detailed methodology of this four-step preparation is described in Chapter 4 Section 4.9.2. This four-step preparation of possible sodium nitrate was tested for its reactivity with SO2 and O2 to determine if NO or NO2 was formed. Neither NO nor NO2 was not found to form, confirming that sodium nitrate was not formed by the four step preparation. And confirming that sodium sulfite was not produced by the reaction of sodium nitrite or sodium nitrate with SO2 and O2. These results confirmed that the three-step reaction process of NO2 formation (equation 7.5- 7.8) was not cyclic. 7.4 X-ray diffraction analysis of sulfate reaction products from specially

prepared sodium nitrate reaction with SO2 and O2. Identification of the polymorphic form of sodium sulfate that forms by the reaction of sodium bicarbonate with SO2 flue gas with NO has not been thoroughly investigated in the past literature. Recall that Howatson et al.(1980) [8] identified the polymorphic phase of the sodium sulfate product formed by the reaction of sodium bicarbonate and SO2 flue gas without NO. Howatson et al.(1980)[8] did not however identify the polymorphic phase form of the sodium sulfate formed by the reaction of sodium bicarbonate and SO2 flue gas with NO. No past researchers have reported on the polymorphic phases of sodium sulfate formed by the reaction of sodium bicarbonate and 343

SO2 flue gas with NO. Recall however that Bortz postulated that sodium sulfate along with NO2 was formed by the reaction of sodium nitrate and SO2, equation 7.1. Recall that preliminary X-ray diffraction analysis was conducted in Chapter 5 Section 5.3.4 Table 5.4 identifying the polymorphic phases of the reaction products formed by the reaction of sodium bicarbonate with SO2 flue gas without NO at 111C, 150C, 175C and 250C. X-ray diffraction analysis was used in this study to determine the polymorphic phase of the reaction products formed by the following reactions of interest: 1) The reaction of different preparations of sodium nitrate with SO2 and O2 at 150C, 175 and 200C. 2) The reaction of sodium bicarbonate with SO2 flue gas with NO at 111C, 150C and 250C. 3) The reaction of sodium bicarbonate with SO2 flue gas with NO2 at 111C. Polymorphic phase identification was accomplished through X-ray diffraction (XRD) analysis. The X-ray diffraction method used is described in Chapter 4 Section 4.3.2. Recall that sodium sulfate possesses a complex temperature-sensitive material structure exceeding that of any single material system excluding silica, according to Kracek (1930)[70]. And recall that Kracek (1930)[70] identified five distinct polymorphic states of sodium sulfate with respect to thermal analysis X-ray diffraction material analysis was conducted on reaction products to determine the sulfation product formed for the reactions of interest. This analysis was intended to provide evidence that the reaction of sodium nitrate with SO2 and O2, 344

equation 7.1, was the reaction segment responsible for NO2 formation in the overall reaction pathway of sodium bicarbonate with SO2 flue gas with NO. Table 7.4 shows the X-ray diffraction results of the products formed by the reactant and gas mixtures listed including those reported in the preliminary studies for the reaction of sodium bicarbonate with SO2 flue gas without NO.

345

Table 7-4 diffraction. Solid reactant freshly made NaNO3 [Na2CO3 + NO2] freshly made NaNO3 [Na2CO3 + NO2] NaHCO3

Summary of product analysis by X-ray Gas SO2, O2 Exp 366 SO2, O2 Exp 397 SO2 flue gas without NO Exp.411 SO2 flue gas without NO Exp 75 SO2 flue gas without NO Exp 419 SO2 flue gas without NO Exp 91 SO2, O2 Exp 420 SO2 flue gas with NO Exp 180 SO2 flue gas with NO Exp 102 SO2 flue gas with NO Exp 107 SO2 flue gas with NO2 Exp 192 Temp. (C) 150 200 111 XRD of Products Na2SO4-I Na2SO4-I Na2SO3

NaHCO3

150

Na2SO3,

NaHCO3

175

Na2SO3 Na2SO4-III Na2SO4-III

NaHCO3

250

freshly made NaNO3 [Na2SO3+NO,O2] three step NaHCO3

175

Na2SO4-I

111

Na2SO4-I

NaHCO3

150

Na2SO4-I

NaHCO3

250

Na2SO4-III

NaHCO3

111

Na2SO4-I

The method of product collection for the reaction of the freshly prepared sodium nitrate with SO2 and O2 is described in Chapter 4 Section 4.4.2.2.3. Product from reaction of sodium nitrate preparations and SO2 and O2 were analyzed by X-ray diffraction. Two separate sodium nitrate preparations were evaluated. Product from the reaction of SO2

346

and O2 with sodium nitrate prepared by reaction of sodium carbonate with NO2 was evaluated. And product from the reaction of SO2 and O2 with sodium nitrate was evaluated. The sodium nitrate was prepared by reaction of sodium sulfite with NO and O2, where the sodium sulfite was made fresh by reaction of sodium bicarbonate with SO2 at low temperature (also referred to as product from the three step reaction). The reaction products from these two reactions were found both to be sodium sulfate Form I. This finding was unexpected. The formation of Form I polymorphic phase by the reaction of sodium nitrate and SO2 and O2 is of a unique nature, since this polymorphic Form is expected only under equilibrium conditions above 244C. Neither of the two reactions of sodium nitrate with SO2 and O2 were however conducted at temperatures above 244C, and all x-ray diffraction testing was done at room temperature under atmospheric conditions where sodium sulfate Form I would convert to Form III unless stabilized in some way. The non-equilibrium, stabilized presence of Form I sodium sulfate by the reaction of sodium nitrate and SO2 and O2 is a unique marker for this reaction. Sodium sulfate Form I was not found to form at any temperature by the reaction of sodium bicarbonate with SO2 flue gas without NO. At temperature equal to or below 150C, the reaction of sodium bicarbonate with SO2 flue gas without NO, sodium sulfite was found to form as anticipated based on the previous reported product identification by Howatson et al. At temperature greater than 150C but below 250C, the reaction product of sodium bicarbonate with SO2 flue gas without NO was found to be a mixture of sodium sulfite and sodium sulfate Form III as anticipated based on the previous reported product identification by Howatson et al. And at a temperature greater than or equal to 250C, the reaction product of sodium bicarbonate with SO2 flue gas without NO was 347

found as expected according to Howatson et al.s previous findings to be sodium sulfate Form III. Although the reaction was conducted above 244C where sodium sulfate Form I would be expected to form, under equilibrium conditions after cooling Form I would be expected to transform to Form III if no stabilization was available. For the reaction of sodium bicarbonate with SO2 flue gas with NO, as Bland previously reported, sodium sulfate was found to form regardless of temperature. Importantly, below 250C, sodium sulfate Form I was found to form by this reaction. Finding sodium sulfate Form I is some evidence that the reaction of sodium nitrate with SO2 and O2 is involved since Form I is a marker of this reaction. Form I is found to form for the reaction of sodium bicarbonate with SO2 flue gas with NO at the temperatures where for the reaction of sodium bicarbonate with SO2 flue gas without NO some sodium sulfite forms. Above 250 C, without NO, no sodium sulfite forms. Therefore with NO above 250 C, the hypothesized three-step pathway would be obstructed by the absence of sodium sulfite formation, since sodium nitrate is speculated to form by reaction of sodium sulfite with NO and O2. Figure 7.4 shows the x-ray diffraction scan of the product collected from the reaction of sodium bicarbonate with SO2 flue gas with NO at 111C after 665 minutes.

348

(a
100 90 80 Intensity % 70 60 50 40 30 20 10 0 15

X-ray diffraction Analysis of Product from Reaction of Sodium Bicarbonate with SO2 Flue Gas with NO at 111C

20

25

30

35

40

45

50

55

60

65

2Theta (degrees)

(b
100 90 80 Intensity % 70 60 50 40 30 20 10 0 15

Joint Committee on Powder Diffraction (JCPDS) and International Center of Diffraction Data (ICDD) Standard for Sodium Sulfate I, File #1-990

20

25

30

35

40

45

50

55

60

65

2Theta (degrees)

(c
100 90 80 Intensity % 70 60 50 40 30 20 10 0 15

Joint Committee on Powder Diffraction (JCPDS) and International Center of Diffraction Data (ICDD) Standard for Sodium Nitrate, File #36-1474

20

25

30

35

40

45

50

55

60

65

2Theta (degrees)

Figure 7-4 X-ray diffraction analysis of product from reaction of sodium bicarbonate with SO2 flue gas with NO at 111C. The scan in (a) is matched with the standard files for sodium sulfate Form I in (b), and sodium nitrate in (c). 349

The analysis in Figure 7.4 shows that the sulfation product diffraction scan matches the standard reference file for sodium sulfate Form I (PDF 1-990). In addition the analysis reveals that the product contains some sodium nitrate (PDF 36-1474). Although not shown, analysis was also conducted on the product from the reaction of sodium bicarbonate with SO2 flue gas with NO at 150C. Like the reaction at 111C, sodium sulfate Form I (PDF 1-990) and sodium nitrate (PDF 36-1474) were found in the product. Finding sodium sulfate Form I (PDF 1-990) formation for these reactions at 111C and 150C is unexpected since Kracek (1930)[70] reported that under equilibrium conditions it is not possible for sodium sulfate Form I to form at temperatures below 241C. Additionally as stated above Kracek and Gibson (1930) [72] showed that under normal atmospheric conditions at room temperature with some humidity, sodium sulfate Form I rapidly inverts to Form III. The analysis of the product from the reaction of sodium bicarbonate with SO2 flue gas with NO at 250C after 585 minutes was found in this study to match the reference scan for sodium sulfate Form III (PDF 24-1132). This phase is the same phase found to form by the reaction of sodium bicarbonate with SO2 flue gas without NO at 250C (Figure 7.5). The sodium sulfate formed by the reaction of sodium bicarbonate or sodium carbonate and SO2 flue gas with or without NO at 250C was found to be different from the stabilized sodium sulfate Form I formed by the reaction of sodium nitrate with SO2. The X-ray diffraction analysis of the sulfation products analyzed in these preliminary tests is summarized in Table 7.5.

350

Table 7-5 Summary of sulfation products determined by X-ray diffraction Flue Gas Mixture NaHCO3+ SO2 flue gas without NO NaNO3+SO2+O2 Temp 111C Temp 150C Temp 250C Na2SO4-III PDF 24-1132 NT Na2SO4-III PDF 24-1132 NT

Na2SO3 Na2SO3 PDF 37-1488 PDF 37-1488

Na2SO4-I Na2SO4-I PDF 1-990 PDF 1-990 NaHCO3+ SO2 Na2SO4-I Na2SO4-I flue gas with NO PDF 1-990 PDF 1-990 NaHCO3+ SO2 NT Na2SO4 -I flue gas with NO2 PDF 1-990 PDF= Powder Diffraction File, NT= Not Tested

The stabilized formation of sodium sulfate Form I at 111C and 150C in SO2 flue gas with NO indicates the occurrence of a special non-equilibrium reaction process different from the sulfate Form III formation mechanism that occurs at 250C in SO2 flue gas with or without NO. The formation of sodium sulfate Form III at 250C in SO2 flue gas with or without NO is thought to be due to oxidation via the two elementary reactions described by Keener and Khang (1993)[52]: Na2CO3 + SO2 Na2SO3 + CO2 Na2SO3 + O2 Na2SO4 Na2CO3 + SO2 + O2 Na2SO4 + CO2 (7.10) (7.11) (7.12)

For the reaction of sodium bicarbonate in SO2 flue gas with NO, the formation of sodium sulfate Form I was unexpected since under equilibrium conditions this Form arises only when the sodium sulfate product has been exposed to temperature greater than 244. Also Form I is not stable upon cooling to room temperature but would be expected to invert to a lower temperature form unless stabilized. Finding Form I from these reactions indicated that the sodium sulfate had formed due to special conditions which 351

include stabilization since it does not invert. The stabilized sodium sulfate Form I was speculated to form as a result of the reaction of sodium nitrate and SO2 and O2. Note that for the reactions in SO2 flue gas with NO it was not found to form at 250C since at this temperature the sodium sulfate is hypothesized to form due to oxidation of sodium sulfite not by the three step reaction pathway. At 250C, based on the work of Howatson et al. sodium sulfite did not form and as a result sodium nitrate did not form by the three-step reaction pathway. It is important to recall from Chapter 2 Figure 2.6 and from Figures 7.5 that sodium sulfite was produced by the reaction of sodium bicarbonate and SO2 for temperatures between 111C and 150C, Na2CO3 + SO2 Na2SO3 + CO2 (7.13)

Sodium sulfite reacted with NO (or NO2) to produce sodium nitrite and sodium nitrate: Na2SO3 + 2NO + O2 2NaNO2 + SO3 Na2SO3 + 2NO + 2O2 2NaNO3 + SO3 Na2SO3 + 2NO2 + O2 2NaNO3 + SO3 (7.14) (7.15) (7.16)

The sodium nitrite and nitrate reacted with SO2 to form this stabilized sodium sulfate Form I, 2NaNO2 + SO2 Na2SO4 + 2NO 2NaNO2 + SO2 + O2 Na2SO4 + 2NO2 2NaNO3 + SO2 Na2SO4 + 2NO2 7.5 Conclusions The reaction of sodium nitrate with SO2 was studied. It was found that NO2 and sodium sulfate were formed by this reaction when O2 was included in the flue gas. These (7.17) (7.18) (7.19)

352

findings validate the third step of the hypothesized three-step reaction pathway described by equation 7.1. Various preparations of sodium nitrate were tested for reactivity with SO2 and O2 revealing the importance of keeping the sodium nitrate surface fresh and free from hygroscopically bound water vapor. Bound water vapor is postulated to shield the sodium nitrate from reaction with SO2 and O2 according to the following reaction equation, NaNO3 + 4H2O NaNO34(H2O) (7.22)

Shelf life contamination from atmospheric humidity was implicated as the reason that non-freshly prepared sodium nitrate samples failed to react with SO2 and O2. Manufacturer purchased technical grade sodium nitrate although possessing high bulk purity was found to be non-reactive with SO2 and O2. Surface contamination from shelf life humidity exposure was again implicated as the cause of this lack of reactivity. Sodium nitrate is known to be hygroscopic, and therefore moisture would readily bind to its surface if available. Experiments showed that with a fresh preparation of sodium nitrate, that was unexposed to atmospheric humidity, reaction with SO2 and O2 proceeded. Two methods of fresh preparation were evaluated and both resulted in reaction with SO2 and O2. Sodium nitrate freshly prepared by reacting sodium carbonate with NO2 according to Knights (1977)[18] previous findings was shown to be reactive with SO2 and O2. And sodium nitrate, freshly prepared, by the first two steps of the simplified three-step reaction pathway described in Table 7-1, was found to be reactive with SO2 and O2. Stepwise, this sodium nitrate was formed fresh by reacting sodium bicarbonate with SO2

353

at low temperature to form sodium sulfite and then reacting this sodium sulfite with NO and O2 as investigated in detail in Chapter 6. Although the speculated reaction equation of sodium nitrate with SO2 (equation 7.1) does not parenthetically include oxygen, the experiments conducted in this chapter showed that oxygen was required for NO2 formation. Without oxygen, NO2 was not found to form for the reaction of SO2 with the sodium nitrate formed by the reaction of sodium carbonate and NO2. Without oxygen, NO2 was also not found to form for the reaction of SO2 with the sodium nitrate formed by the first two steps of the three-step reaction pathway. But NO was found to form by the reaction of SO2 without oxygen with the sodium nitrate formed by the first two steps of the three step reaction pathway. Finding NO formation by this reaction further confirms that the reaction of sodium sulfite with NO and O2 forms a mixture of sodium nitrite and sodium nitrate. Recall that Bland in screening found that sodium nitrite reacted with SO2 without O2 produced NO evolution. NO2 formation with respect to temperature was evaluated respectively for the reaction of both forms of freshly prepared sodium nitrate with SO2 and O2 and compared to the NO2 formation found from the reaction of sodium bicarbonate with SO2 flue gas with NO. The NO2 formation with respect to temperature for both freshly prepared sodium nitrates with SO2 and O2 closely corresponded to the NO2 formation found from the reaction of sodium bicarbonate with SO2 flue gas with NO. For both of these reactions maximum NO2 formation was found to occur at 175C.

354

Also X-ray diffraction analysis of the product found the formation of the same polymorphic form of sodium sulfate, Form I for both reactions at all reaction temperatures. 7.6 Areas for further studies More work is needed to determine the polymorphic form of sodium sulfate formed by the reaction of sodium nitrite with SO2 and O2 and the reaction of sodium nitrate with SO2 and O2 at other temperatures then those tested in this study to verify the generalized trend. Additionally more work is needed to study the polymorphic form of sodium sulfate formed from the reaction of sodium nitrite/nitrate, formed by the threestep reaction sequence, with SO2 and O2 with respect to temperature. This information would be useful to identify if a unique form of sodium sulfate arises by this reaction that may be identified as a marker of the reaction. Further investigation of the reaction of sodium nitrate formed by the three-step reaction sequence with SO2 and O2 is needed to determine the effect of temperature on the resultant NO to NO2 ratio. More work is needed to measure the strength of hygroscopic forces associated with sodium nitrate and water vapor with respect to temperature. Exposure tests of freshly made sodium nitrate/nitrite, formed by the three-step reaction sequence, with water vapor are needed to verify that it results in subsequent deactivation with SO2 and O2. More surface analysis work is needed to study the changes that water vapor makes to the surface of sodium nitrate.

355

CHAPTER 8 MITIGATION OF PLUME DISCOLORATION FOR SODIUM BICARBONATE INJECTION IN SO2 FLUE GAS WITH NO 8.1 Introduction With the reaction pathway knowledge obtained from this study efforts were made to conceive of possible mitigation techniques to avoid plume discoloration. In this chapter the validated reaction pathway investigated in this study will be looked at for possible actions that may be implemented in the field to avoid NO2 plume discoloration during sodium bicarbonate injection. Table 8-1 lists the reactions hypothesized in the overall reaction pathway for NO2.

356

Table 8-1 Reactions hypothesized in the overall reaction pathway for NO2.

Eq. 8.1 8.2 8.3 8.4 8.5 8.6 8.7 8.8 8.9 8.10 8.11 8.12 8.13 8.14 8.15 8.16

Reaction 2NaHCO3 Na2CO3 + H2O + CO2 2NaHCO3 + SO2 Na2SO3 + 2CO2 + H2O Na2CO3 + SO2 Na2SO3 + CO2 Na2SO3 + O2 Na2SO4 Na2SO3 + 2NO + O2 2NaNO2 + SO3 Na2SO3 + 2NO + 2O2 2NaNO3 + SO3 Na2SO3 + 2NO2 + O2 2NaNO3 + SO3 NaHCO3 + SO3 Na2SO4 + CO2 + H2O Na2CO3 + SO3 Na2SO4 + CO2 NaNO2 + O2 + 4H2O NaNO34(H2O) NaNO3 + 4H2O NaNO34(H2O) 2NaNO2 + SO2 Na2SO4 + 2NO 2NaNO2 + SO2 Na2SO4 + 2NO 2NaNO2 + SO2 + O2 Na2SO4 + 2NO2 2NaNO3 + SO2 Na2SO4 + 2NO2 Na2CO3 + NO2 2NaNO3 + NO + CO2

357

The reactions listed in Table 8-1 are schematically assembled and shown in Figure 3-3. It was recognized that whatever mitigation method was selected, it was required to curtail NO2 formation in addition to removing SO2. Each of the following approaches to mitigation of NO2 were investigated: 1) Adjusting the flue gas oxygen concentration 2) Adjusting the duct temperature 3) Injecting flue gas additives 4) Removing NO from the flue gas prior to sodium bicarbonate injection 5) Injecting sodium carbonate after flue gas desulfurization with sodium bicarbonate injection to capture NO2. 8.2 8.2.1 Plume discoloration mitigation options. Plume discoloration mitigation by limiting oxygen concentration in the flue gas. Experimental data revealed that oxygen was required for NO2 formation. In terms of the reaction pathway and its individual reactions as listed in Table 8-1, two essential reactions required oxygen. First oxygen was required in the nitration reaction where sodium sulfite was reacted with flue gas NO, reaction equations 8.5 and 8.6. Without oxygen present, sodium sulfite, from sulfation, was found not to react with NO in the flue gas. Additionally in the absence of oxygen, sodium nitrate was found to be un-reactive with SO2, and therefore would not convert to NO2 and sodium sulfate by reaction with SO2 according to equation 8.15. This mitigation option was experimentally studied further and is described below.

358

The effect of oxygen concentration on NO2 formation was studied for the reaction of sodium bicarbonate and SO2 flue gas with NO. NO2 formation by this reaction was measured for oxygen concentrations ranging from 0% to 20%. A description of the experimental test procedure used is described in Chapter 4 Section 4.5.3.1. These tests were used to establish the relationship between oxygen concentration and NO2 formation. Knowledge of this relationship was intended for use for NO2 mitigation particularly in evaluating flue gas re-circulation. Recall from Chapter 2 that Bland (1990)[16] (Section 2.3.4.1) previously reported no NO2 formation by the reaction of sodium bicarbonate and SO2 flue gas with NO when the oxygen concentration was zero. Figure 8-1 shows the results obtained for the NO2 measurements at oxygen concentrations of 0%, 0.5%, 3%, 5% and 20% percent. Note that twenty percent oxygen in flue gas was the upper limit of the oxygen concentration used in this study. This upper limit was selected because no more then twenty percent oxygen is found in air, and the flue gas oxygen percent reflects the amount of excess air after combustion based on the fuel to air ratio used for combustion.
NO2 Formation versus Oxygen Percent for the Reaction of NaHCO3 and SO2 Flue Gas with NO

600 Peak NO2 (ppm) 500 400 300 200 100

0 0 2 4 6 8 10 12 14 16 18 20 Oxygen in Flue Gas (%)

359

Figure 8-1 NO2 Formation versus O2 percent in flue gas. Note from Figure 8-1 that as Bland (1990)[16] had reported previously, this study found that the NO2 formation was zero for the reaction of sodium bicarbonate with SO2 flue gas with NO with zero percent oxygen. Additionally, Figure 8-1 shows that NO2 formation increases dramatically when the oxygen concentration is increased from 0% and 5%. And the test results shown in Figure 8-1 also show that the NO2 formation remains constant for oxygen concentrations above 5%. From these results it is realized that if the oxygen concentration is reduced below 5% then NO2 formation is reduced. These results indicate that oxygen reduction methods for flue gas may be used to mitigate NO2 formation. For example flue gas re-circulation methods may be implemented to reduce the oxygen concentration from 3% to 1% to obtain over a 50% reduction in NO2 formation. 8.2.2 Plume discoloration mitigation by adjusting the injection flue gas temperature. Mitigation of NO2 was also studied by controlling reaction temperature during sodium bicarbonate injection. In particular from a reaction pathway standpoint, the temperature may be tailored to minimize formation of sodium sulfite from reaction equations 8.7 and 8.3. Recall that the sodium sulfite is a key constituent in the NO2 formation pathway. Recall that sodium sulfite is one of the primary constituents in the simplified three-step reaction pathway. Howatson et al.s (1980) investigations previously revealed according to Figure 2-6 that sodium sulfite formation was minimum at low temperature below 95C and at high temperatures above 250C. This is consistent with the NO2 formation measurements found in Chapter 5 Figure 5-1 with respect to 360

temperature. Recall that in Chapter 5 NO2 formation was measured from the reaction of sodium bicarbonate and SO2 flue gas with NO and found to be minimum at low temperatures below 110C and at high temperatures above 250C. At high temperatures above 250C, the sodium sulfite that forms by reaction equation 8.2 and 8.3 is preferentially oxidized to sodium sulfate according to reaction equation 8.5 as opposed to reacting with NO and O2. Secondly from a reaction pathway perspective, temperature may be tailored to minimize sodium nitrite and nitrate formation by the reaction of sodium sulfite and NO and O2, according to equations 8.5 and 8.6. Recall that sodium nitrate is a primary constituent in the simplified three-step reaction pathway for NO2 formation. In particular, duct temperatures may be set where the outcome of the reaction of sodium sulfite with NO and O2 produces little sodium nitrite and sodium nitrate. Further follow on testing is however needed to determine the temperature relationship for these reactions. Based on thermodynamic analysis however, in Chapter 3 Section 3.8, for sodium nitrate and nitrite formation, no temperature in the operating range appears to limit the equilibrium tendency for formation. In conclusion, tailoring the duct reaction temperatures to below 100C or to above 300C is an option that may be used to minimize NO2 formation. Injecting sodium bicarbonate above 300C as opposed to below 100C would be preferred since Bland reported that SO2 removal was considerably better at higher temperature as opposed to at lower temperature. However, injection with high temperature may incur losses from the required heating of the flue gas and may involve extra construction and maintenance costs in building a higher temperature bag-house to handle the higher temperature. 361

8.2.3

Plume discoloration mitigation by injecting flue gas additives. Adding flue gas additives to the duct during sodium bicarbonate injection may be

used to minimize NO2 formation. In particular from a reaction pathway perspective, flue gas additives may be added in an effort to convert gas phase NO2 to NO. Also additives may be added to impede one or more of the key reactions in the pathway. For example, additives might be injected to promote oxidation of sodium sulfite to sodium sulfate to minimize available sodium sulfite in the reaction pathway. Or additives may be used to bind to sodium sulfite and shield it from reaction with NO and O2. Likewise additives may be used to bind to sodium nitrate to shield it from reaction with SO2 and O2. Hooper (1989)[79] patented a urea additive for brown plume control. Injection of water vapor additive has also been suggested by Bortz to curtail NO2 formation. Additions of water vapor would beneficially aid in flue gas NOx removal by fixing water vapor to sodium nitrate and shielding the accumulated sodium nitrate from destructive conversion reaction with SO2. Bortzs study of added flue gas humidity confirmed that adding water vapor to the flue gas reduces NO2 formation and increases NOx removal (sodium nitrate formation). It should be noted that adding water by humidification or water spray may have negative system impacts. For example, corrosion issues with duct and stack materials may arise with water vapor additions. Additionally it is speculated that water vapor additions to flue gas may cause undesirable plume opacity issues. Although NOx would be removed, and the flue gas would be cleaner by water vapor additions, the white plume coloration known to be associated with water vapor slippage out a stack is believed to be unacceptable to communities living in the vicinity of facilities and may raise a public nuisance issues with respect to Health and Safety Codes.

362

8.2.4

Plume discoloration mitigation by removing NO from the flue gas prior to sodium bicarbonate injection. Another mitigation method considered was removing NO from the flue gas prior

to sodium bicarbonate injection. From a reaction pathway perspective sodium nitrate would not form by equation 8.6 in the reaction pathway if NO were scrubbed from the reaction flue gas. NO2 would not form if NO were removed since NO2 formation by the reaction pathway requires sodium nitrate formation. Recall that in the reaction pathway NO2 is formed by the three step reaction pathway where NO and O2 is first reacted with sodium sulfite, equation 8.6 to form sodium nitrate and subsequently by the reaction of this sodium nitrate with SO2 and O2, equation 8.14. It should be noted that physically removing NO from flue gas containing SO2 is a challenge. Very few practical methods to remove NO from flue gas with SO2 have been invented to date. Combustion flame temperature control, low NOx burners, selective catalytic reduction system, and selective non-catalytic reduction systems with ammonia injection are methods but all require extensive hardware and are costly. A simple, efficient, and economical process for NO removal during sodium bicarbonate injection has been invented. The method includes two stages for removal of NO and SO2 from flue gas. The first stage removes all NO by injection of fresh unoxidized sodium sulfite. The second stage removes SO2 by sodium bicarbonate injection. Special parameters are required in the second stage to form sodium sulfite product. And the freshly formed, un-oxidized sodium sulfite product from this stage may be used in the first stage for injection. This method in its entirety is provided in Appendix C.

363

According to past development aqueous solutions of sodium sulfite may also be used to remove NO from gas streams. Cost and operating complexity may make this wet scrubbing technique an unattractive option. In addition if wet scrubbing were used to remove the NO from the flue gas, then it may be simpler to additionally scrub out SO2 with the NO in this aqueous system and not install a second stage. 8.2.5 Plume discoloration mitigation by injecting sodium carbonate after the sodium bicarbonate flue gas desulfurization injection step.

Another possible option to mitigate NO2 formation is to inject sodium carbonate after the sodium bicarbonate flue gas desulfurization (FGD) step. For this mitigation option the fact that NO2 is produced by sodium bicarbonate FGD becomes an advantage. Knight (1977)[18] and others have established that NO2 may be removed by reaction with sodium carbonate. Removal of NO2 by sodium carbonate is especially fitting as a second stage to sodium bicarbonate FGD since no gas stream reheating would be required since NO2 removal is optimized at low temperature. Also this second stage injection approach may by feasible since sodium carbonate is a readily available material and is a material that is easy to handle. More work and investigation into this mitigation method is needed to determine if the residual SO2 after the sodium bicarbonate flue gas desulfurization (FGD) step would adversely interfere in NO2 removal or if it were of a low enough concentration to allow necessary NO2 removal to proceed.

364

8.2.6

Conclusions In this chapter several mitigation methods for NO2 were discussed. Experimental

investigations were conducted to investigate mitigation of NO2 by reducing flue gas oxygen concentration during sodium bicarbonate injection. These laboratory experiments showed that NO2 formation during sodium bicarbonate injection was reduced by reduction of flue gas oxygen concentration below 5%. NO2 formation was found to proportionally decrease when the oxygen concentration was decreased below 5% with complete NO2 mitigation when the flue gas oxygen concentration was zero. Duct temperature changes were also explored as plume discoloration mitigation. NO2 formation with respect to temperature data for the reaction of sodium bicarbonate in SO2 flue gas with NO was analyzed to determine if practical duct injection temperature controls were possible to provide NO2 mitigation. The data showed that NO2 formation was minimum when flue gas temperatures were reduced below 110C and above 250C. The injection of flue gas additives was also considered as a method of plume discoloration mitigation. Additions of water vapor were considered. Theoretically, it was believed that injection of water vapor would provide NO2 mitigation. In particular it was speculated that NO2 formation would be reduced by water vapor additions by providing additional sodium nitrate shielding from reaction with SO2 and O2. It was noted that Bortz had previously suggested a process step of water vapor injection to aid in NOx collection and NO2 mitigation. Although it was determined to be an effective mitigation method it was concluded that water vapor injection presented negative side effects making it undesirable. In particular it was noted that water vapor additions posed significant increased risk of public nuisance violation because water vapor additions

365

themselves caused a white visible plume. Also it was noted that with water injection increased hardware corrosion issues existed. Also it was noted that Hooper previously patented a method of ammonia and urea additives. Ammonia and urea additives however were noted to be costly and have disadvantages associated with slip. Also mitigation methods were suggested where NO was removed from the flue gas prior to sodium bicarbonate injection. A mitigation method was devised and patented based on the reaction pathway findings found in this study. This patented mitigation method involved a two step process where NO was removed from the flue gas prior to sodium bicarbonate injection by dry injection of sodium sulfite. Appendix C provides a copy of this complete U.S. Patent. Post injection of sodium carbonate, after the sodium bicarbonate injection step was considered as a mitigation method. It was theorized that adding an extra sodium carbonate injection step might be an effective means to subsequently capture the NO2 formed in the sodium bicarbonate step.

366

CHAPTER 9 CONCLUSIONS 9.1 9.1.1 Chapter Summaries Summary of Chapter 1 Sodium bicarbonate sorbent has been under development since the early nineteen sixties as a sorbent option for scrubbing SO2 from coal combustion flue gas. This sorbent injection technique potentially offers a low cost option to utilities for compliance with Phase II SO2 Acid Rain regulations within the Clean Air Act. This technology however has not yet been fully introduced into the market because of unresolved plume discoloration problems discovered during sorbent injection demonstration tests in the late1980s. To allow the technology of dry sodium bicarbonate injection to fully enter the marketplace for Acid Rain scrubbing applications, it is necessary that the plume discoloration problems first be resolved. The reaction pathway of NO2 formation, the gaseous constituent responsible for this plume discoloration during sodium bicarbonate injection, was investigated in this study. In addition this study considered and devised mitigation methods to avoid plume discoloration. 9.1.2 Summary of Chapter 2

Chapter 2 reviewed the pertinent literature. Chapter 2 included a review of literature related to plume discoloration. Because plume discoloration depended on a combination of a number of complex parameters including flue gas concentration, temperature, stack

367

diameter, and atmospheric meteorology it was not discovered as a problem until very late in the research and development cycle of this technology. The literature review revealed that NO2 was the primary agent responsible for this plume discoloration. The literature review revealed important information related to the overall reaction chemistry. The literature review provided a foundation of reaction information related to: 1) The decomposition of sodium bicarbonate. 2) The reaction of sodium bicarbonate with SO2 flue gas without NO 3) The reaction of sodium carbonate with NO without SO2 4) The reaction of sodium bicarbonate with NO and NO2 5) The reaction of sodium bicarbonate with SO2 flue gas with NOx. In particular it was found in the literature that: 1) Sodium bicarbonate decomposition was inhibited from 121C to 162C for decomposition in SO2 flue gas with NO. 2) Sodium bicarbonate reacted with SO2 flue gas without NOx to form a temperature dependent mixture of sodium sulfite and sodium sulfate. 3) Only sodium sulfate, sodium nitrate, and NO2 were formed by the reaction of sodium bicarbonate with SO2 flue gas with NO. No sodium sulfite formed by the reaction of sodium bicarbonate with SO2 flue gas with NOx. Additionally the NO2 formation depended directly on active SO2 removal. When sodium bicarbonate injection was stopped, NO2 formation ceased. And when SO2 flow was stopped, NO2 formation also ceased. The amount of NO2 formation depended upon the SO2 concentration. NO2 formation increased when the ratio of inlet SO2 to NO concentration increased.

368

4)

The amount of total sulfation formed for the reaction of sodium bicarbonate with SO2 flue gas with NO was significantly smaller than the amount formed for the reaction of sodium bicarbonate with SO2 flue gas without NO.

5)

For the reaction of sodium bicarbonate with SO2 flue gas with NO, sodium nitrate was formed only during a short finite time period immediately after sodium bicarbonate injection. The amount of sodium nitrate formed depended on temperature. Maximum sodium nitrate was formed at low temperature. Sodium nitrate formation decreased with increasing temperature.

6)

For the reaction of sodium bicarbonate with SO2 flue gas with NO, NO2 formation depended upon temperature. NO2 formation was small at low temperature, T<150C and also small at high temperatures above 300C. SO2 formation was greater at 175C. NO2 did not form during the time period immediately after injection when NOx was removed but NO2 was formed after the finite period of NOx removal ended.

7)

The reaction of sodium bicarbonate with SO2 flue gas with NO without oxygen produced only sodium sulfite; no sodium nitrate or sodium sulfate was formed.

8)

For the reaction of sodium carbonate with SO2 flue gas with NO, sodium sulfate was formed but sodium nitrate was not formed.

9)

For the reaction of sodium carbonate and NO2, sodium nitrate was formed.

The literature review in Chapter 2 also provided descriptions of previously hypothesized reaction pathways for NO2 formation and NOx removal:

369

1) Miller (1986) hypothesized that NO2 was formed by the reaction of oxygen with a reaction intermediate-sodium nitrogen sulfonate, Na2SO3(NO)2. The sodium nitrogen sulfonate was hypothesized to form by reaction of sodium sulfite and NO. 2) Bland (1990) hypothesized that NO2 was formed by a heterogeneous catalytic reaction of a sodium sulfur-oxide intermediate and NO. The sodium sulfur-oxide compound was hypothesized to form as a reaction intermediate from the reaction of sodium bicarbonate and SO2. 3) Verlaeten et al. (1991) hypothesized that NOx was removed by reaction of flue gas NO with sodium bisulfite. The sodium bisulfite was hypothesized to form as a reaction intermediate from the reaction of the sodium bicarbonate with SO2. 4) Bortz (1994) hypothesized that NO2 was formed by the reaction of sodium nitrate and SO2. The sodium nitrate was hypothesized to form as a reaction product of sodium bicarbonate and SO2 flue gas with NO. Also the literature review in Chapter 2 provided a description of past findings on: 1) The reaction of sodium sulfite with NO and O2. 2) The reaction of sodium pyrosulfite and sodium bisulfite with NO and O2. 3) The reaction of sodium nitrite with SO2 and O2 and of the reaction of and sodium nitrate with SO2 and O2. For each of these reactions, there was no clear agreement in the literature on the reaction outcomes. It was reported by some past investigators that no reaction product formed by the reaction of sodium sulfite and NO and O2. While other past investigators reported that

370

NOx was removed by the reaction. One researcher reported that NO2 formed by this reaction. . It was reported by some past investigators that the reaction of sodium pyrosulfite and sodium bisulfite with NO and O2 resulted in NOx removal in the form of sodium nitrite and sodium nitrate by this reaction; while other past investigators reported that sodium pyrosulfite and sodium bisulfite was un-reactive with NO and O2. It was reported by one past investigator that the reaction of sodium nitrate and SO2 and O2 produced NO and NO2; while another past investigator reported that no reaction products were produced by this reaction. All of the findings discussed in the literature review in Chapter 2 were used as a basis or foundation for theoretical development of the reaction pathway hypothesized for investigation in this study. 9.1.3 Summary of Chapter 3

A reaction pathway for NO2 formation and NOx removal was theoretically developed and hypothesized for this study in Chapter 3. A list of past undisputed relevant reactions and associated end-products were compiled in Chapter 3 (Table 3.6). From the list of past undisputed relevant reactions and associated end products, criteria for SO2 removal, NOx removal, and NO2 formation were established in Chapter 3 (Tables 3-7,3-8 and 3-9 respectively). These criteria were then used to assess the past hypothesized reaction pathways from Miller (1986)[19], Bland (1990)[16], Verlaeten et al. (1991)[20], and Bortz (1994)[45]. All of the undisputed reaction segments within these past hypothesized reaction pathways found to conform with the criteria were then grouped together and combined to assemble a foundation for a new reaction pathway hypothesis (Tables 3-3 371

Table 3-6 and Figure 3-1). Additionally the conforming controversial reaction segments, from the past investigators were considered for further experimental investigation in this study. In particular: 1) The controversial reaction of sodium sulfite with NO was selected for further study in this investigation to determine if either NO2 or sodium nitrate was produced by this reaction. 2) The controversial reaction of sodium nitrate with SO2 was selected for further experimental study in this investigation to determine if NO2 was formed by this reaction. Chapter 3 also included a thermodynamic feasibility study of both the undisputed and controversial reactions in the overall hypothesized reaction pathway. The thermodynamic calculation method used was provided in Appendix A. These thermodynamic calculations showed that all of the individual reactions in the overall hypothesized reaction pathway were thermodynamically feasible. 9.1.4 Summary of Chapter 4

Chapter 4 included a description of the methodology used for the experimental investigations in this study. The experimental approach, the laboratory apparatus, the analytical methods, and the test procedures used in the experimental section of this study were all included in Chapter 4. The overall reaction pathway developed in Chapter 3 involving the fifteen individual reactions was simplified to a simple three-step reaction pathway for NO2 formation in Chapter 4. Table 9-1 summarizes this simplified three-step reaction pathway for NO2 formation developed in Chapter 4. 372

Table 9-1 The simplified three-step reaction pathway for NO2 formation from the reaction of sodium bicarbonate and SO2 flue gas with NO. Step Description of the reaction step. Number 1 A sulfation step where sodium sulfite is produced by the reaction of sodium bicarbonate or sodium carbonate with SO2. 2NaHCO3 + SO2 Na2SO3 + 2CO2 + H2O (9.1) Na2CO3 + SO2 Na2SO3 + CO2 2 A nitration step where sodium nitrate is produced by reaction of the sodium sulfite from step 1 with NO Na2SO3 + 2NO + 2O2 2NaNO3 + SO3 (9.2) 3 An NO2 release reaction step where NO2 is released from the collected sodium nitrate from step 2 by reaction with SO2 in the flue gas. 2NaNO3 + SO2 Na2SO4 + 2NO2 (9.3)

The experimental approach used in this study was provided in Chapter 4. It was described that the experimental studies specifically focused on validation of the nonvalidated reactions in simplified three-step reaction pathway. In particular this study focused on both the validation of step number 2 in the simplified three-step reaction pathway, the nitration step and on the validation of step number 3 in the simplified threestep reaction pathway, the NO2 release step. The reactor apparatus used in the experimental portion of this study was described in Chapter 4. Both the thermal gravimetric analyzer and the isothermal fixed bed reactor apparatus were described. In particular a detailed description of the fixed bed reactor was given including description of: 1) The special precision flue gas mixing station for metering desired flue gas mixes of CO2, O2, N2, water vapor and SO2 and NO, 2) The special gas pre-heater to guarantee isothermal conditions.

373

3) The special powder injection module with inert gas back fill to minimize atmospheric gas interference, 4) The special isothermal reaction zone with precision temperature monitoring and with charge transfer capability for reaction product analysis, 5) The special quick quench transfer zone for quick sample chilling under inert gas conditions and configured with a wide mouth hatch opening for reaction product removal, 6) The special gas analyzer station with continuous inlet and outlet SO2 and NOx monitors, 7) The computer data acquisition system with data management for inlet and outlet SO2 and NOx measurements, thermal couple outputs, and SO2 and NO mass flow controller outputs. Chapter 4 also included a description of the analytical testing methods used in the experimental section of this study. In particular detailed descriptions of the elemental analysis, X-ray diffraction analysis, BET surface area measurements, scanning electron micro-structural analysis, Time-of-Flight static Secondary Ion Mass Spectrometry, and X-ray photoelectron Spectroscopy (XPS) /Electron Spectroscopy for Chemical Analysis (ESCA) were provided. Chapter 4 also described the experimental test procedures used in the laboratory reaction studies including the procedures used in the preliminary studies, the reaction validation studies, and the mitigation option studies.

374

9.1.5

Summary of Chapter 5 The results of the preliminary studies were provided in Chapter 5. In particular the

results of the initial reactor apparatus check runs were presented in Chapter 5. For these tests NO2 measurements with respect to temperature were collected from the reactor apparatus for the reaction of sodium bicarbonate with SO2 flue gas with NO. NO2 measurements with respect to temperature from these check runs were found to match the NO2 measurements obtained and reported previously in a laboratory setting by Bland (1990) [16]. The fixed bed apparatus was found from these tests to satisfactorily simulate the bag-house injection reactions. These tests assured that the reactor apparatus and NOx measuring equipment operated properly. Also, preliminary studies in Chapter 5 included material characterization work. These preliminary studies included the study of decomposition inhibition of sodium bicarbonate in SO2 flue gas with NO by conversion tracking. Appendix B provided the particular conversion tracking calculations used. These conversion-tracking studies showed as Bland had previously reported that sodium bicarbonate decomposition was indeed inhibited at 111C in SO2 flue gas with NO. Additionally, the conversion tracking results from sodium bicarbonate reactions in other gas mixtures revealed that decomposition of sodium bicarbonate was inhibited for flue gas mixtures that contained NOx. Interestingly, inhibition of sodium bicarbonate decomposition was found to occur for gas mixtures containing NO without SO2 and for NO2 without SO2. These experiments revealed correlation between sodium nitrate accumulation, of whatever amount, and decomposition inhibition. The amount of sodium nitrate accumulated was found to impact the extent of decomposition inhibition. Greater amounts of sodium

375

nitrate accumulation were found to result in greater sodium bicarbonate decomposition inhibition. It was speculated that decomposition inhibition was a result of the hygroscopic forces from the sodium nitrate product. These hygroscopic forces were speculated to oppose water vapor evolution forces of sodium bicarbonate decomposition. Sodium nitrate accumulation from the reaction of sodium bicarbonate and SO2 flue gas with NO was studied in further detail in the preliminary studies of Chapter 5. Conversion tracking was used to study this sodium nitrate accumulation. The finite sodium nitrate accumulation period previously reported by Bland was experimentally verified in Chapter 5 to occur for the reaction of sodium bicarbonate and SO2 flue gas with NO in this study. These conversion tracking experiments confirmed that that sodium nitrate indeed accumulated during a finite time period immediately following sodium bicarbonate injection. These conversion tracking results additionally showed that this finite time period of sodium nitrate accumulation corresponded directly to the finite time period of sodium bicarbonate decomposition. Sodium nitrate accumulation was shown to cease when decomposition ended. It was speculated that water vapor evolution from sodium bicarbonate decomposition played a key role in sodium nitrate accumulation. In particular, it was postulated that water vapor was bound to sodium nitrate product by its own hygroscopic forces. It was speculated that the bound water vapor on the sodium nitrate product prevented SO2 from reacting with the sodium nitrate product. It was speculated that the sodium nitrate product would otherwise react with SO2 and result in NO2 formation. Under this speculated theory, NO2 would be formed whenever water vapor ceased to be

376

available such as after decomposition was ended and water vapor had stopped being evolved. This speculated theory was consistent with the findings found for the reaction of sodium bicarbonate with flue gas with NOx not containing SO2. For this case without SO2 the tracking data showed, as anticipated by the theory, that sodium nitrate accumulation was not limited to a finite time period when sodium bicarbonate decomposed. Instead sodium nitrate accumulation was found to accumulate both during and after sodium bicarbonate decomposition as the theory expects. Without SO2 in the flue gas sodium nitrate accumulated after sodium bicarbonate decomposition because there was no destruction or conversion of the sodium nitrate to NO2 by SO2. These findings for this case importantly highlight the fact that sodium nitrate accumulation does not depend solely on the presence of water vapor but also depends on the presence of SO2 in the flue gas. Conversion tracking experiments in humidified flue gas were found to be consistent with this speculated theory. The conversion tracking experiments in humidified SO2 flue gas with NO resulted in some sodium nitrate accumulation after decomposition ended as the theory would anticipate. Chapter 5 also included studies of surface area and microstructure changes of sodium bicarbonate reacted with SO2 flue gas with NO. Surface area and microstructure changes measured for sodium bicarbonate reacted in inert nitrogen were compared to surface area and microstructure changes measured for sodium bicarbonate reacted in SO2 flue gas with and without NO. This analysis confirmed that surface area increased during decomposition of the sodium bicarbonate regardless of the reaction gas mixture studied.

377

The surface area formation for the reaction in SO2 flue gas with NO at 150C was found to be less than the surface area formation found for the reaction in inert nitrogen or in SO2 flue gas without NO at 150C. The reduced surface area production for the reaction in SO2 flue gas with NO was speculated to be due to sodium sulfate ash layer product formation. At 250C for the first 10 minutes of reaction of sodium bicarbonate in SO2 flue gas with NO the surface area formation was found to be comparable to the surface area formation for reaction in inert nitrogen and for the surface area formation found in SO2 flue gas without NO. The surface area for the reaction in SO2 flue gas with NO was however found to dramatically drop after 10 minutes at 250C. The microstructure analysis of this product collected after the first 10 minutes at 250C in SO2 flue gas with NO correspondingly showed structural re-crystallization not found in the microstructures formed in inert nitrogen and in SO2 flue gas without NO. These significant surface area and microstructure differences were thought to be due to eutectic melting between sodium sulfate and sodium nitrate product at this temperature. Chapter 5 included the results of the sulfation product analysis from reaction products from sodium bicarbonate the reaction with SO2 flue gas without NO. These sulfation product analysis results confirmed Howatson et al.s (1980)[8] previous findings that at lower temperatures sodium sulfite was formed, and at the higher temperatures sodium sulfate polymorphic Form III was formed, and that between these two temperature regimes a mixture of sodium sulfite and sodium sulfate polymorphic Form III was formed.

378

Chapter 5 also included thermal decomposition investigations of sodium pyrosulfite and sodium bisulfite. It was first established that at or above room temperature, sodium bisulfite was unstable and was fully converted to sodium pyrosulfite. Thermal gravimetric analysis revealed that at or above 140C sodium pyrosulfite was unstable and decomposed. Product analysis revealed that the sodium pyrosulfite decomposition resulted in conversion to sodium sulfite. These findings importantly put into question Verlaeten et al.s (1991)[20] assertion that above 140C sodium pyrosulfite reacted directly with NO and O2 and produced sodium nitrite and sodium nitrate. With this new knowledge that at or above 140C sodium pyrosulfite decomposes to sodium sulfite, it is plausible that sodium pyrosulfite reacted with NO and O2 at or above 140C involves sodium pyrosulfite decomposition to sodium sulfite and subsequent direct reaction of this sodium sulfite with the NO and O2 to form sodium nitrite and sodium nitrate. 9.1.6 Summary of Chapter 6 In Chapter 6 the reaction of sodium sulfite and NO was experimentally investigated. The reaction of sodium sulfite with NO was hypothesized to form a temperature dependent mixture of sodium nitrite and sodium nitrate within the hypothesized overall reaction pathway for sodium bicarbonate with SO2 flue gas with NO according to the following separate elemental reactions, Na2SO3 + 2NO + O2 2NaNO2 + SO3 Na2SO3 + 2NO + 2O2 2NaNO3 + SO3 2Na2SO3 + 4NO + 3O2 2NaNO3 + 2NaNO2 + 2SO3 (9.4) (9.5) (9.6)

379

Equation 9.5, where sodium nitrate was formed was of particular interest in this study since this reaction was a key hypothesized reaction in the simplified three-step reaction pathway (Table 9-1) for NO2 formation. This reaction was selected for further investigation in this study because there was little agreement in the literature on the reaction outcome. Some past researchers maintained that NO2 was produced by this reaction. But to date, no NO2 formation by this reaction has been demonstrated or documented. Other past researchers claimed that nothing was produced by this reaction because it was believed that sodium sulfite was un-reactive with NO. Importantly in Chapter 6 it was noted that it has been well established in the literature that sodium sulfite readily oxidizes to sodium sulfate in moist air. Miller (1986)[19] noted the difficulty of conducting reactivity tests with sodium sulfite since the sodium sulfite samples tested possessed surface oxidation from shelf life exposure and handling under atmospheric conditions with moist air. All of the tests by past investigators that used technical grade sodium sulfite, in fact, resulted in no reaction with NO and O2. As Miller suggested, it was speculated that no reaction was obtain for these tests because the sodium sulfite used in these experiments was contaminated or surface oxidized from moist air exposure during shelf life or handling. Chapter 6 included close examination of the past laboratory tests of sodium sulfite with NO and O2 by past investigators. It was noted from this examination that sodium sulfite was reactive with NO and O2 when it was prepared special. In particular it was noted that past investigators found sodium sulfite reactivity with NO when the sodium sulfite was prepared in a way that kept it free from exposure to moist air. Sodium sulfite formed by mechanically crushing technical grade sodium sulfite and storing it in a dry

380

desiccator was found by one past investigator to react with NO and O2 resulting in NOx removal. In this case, the sodium sulfite tested was freed of surface oxidation by mechanically rubbing during crushing and maintained oxide free by storing in dry conditions. Also, it was noted that sodium sulfite that was prepared fresh by reacting sodium bicarbonate with SO2 at low temperature within the test vessel where it was later reacted with NO and O2 was found to be reactive with NO and O2 and result in NOx removal. In this case the sodium sulfite was free of surface oxidation since it was prepared in the test vessel where it was later tested and was not exposed to moisture or air and was maintained under inert nitrogen flow prior to test. Chapter 6 included the results of the laboratory experiments conducted to test the reaction of sodium sulfite with NO. The experimental laboratory tests of this reaction found certain preparations of sodium sulfite reactive with NO and O2. Specifically, sodium sulfite that was prepared by reacting sodium bicarbonate with SO2 at low temperature in the fixed bed reactor where it was subsequently reacted with NO and O2 was found to result in NOx removal. In these laboratory tests the special preparation of sodium sulfite was kept under inert conditions in the fixed bed prior to the NO and O2 reactivity test. These findings validated the previous findings of Lai and Yesavage (1994)[21] except that in this study reactivity was achieved in a fixed bed reactor without any air leaks as opposed to Lai and Yesavages (1994)[21] continuously stirred reactor (CSR) which reportedly had significant air leakage problems. A significant temperature rise was recorded to occur during this reaction of specially prepared sodium sulfite with NO and O2. These exothermic findings are consistent with the thermodynamic calculations made in Chapter 3 for this reaction.

381

The laboratory tests experimentally found that oxygen was required to be present with NO for the reaction of this specially prepared sodium sulfite to result in NOx removal. Elemental analysis of the product found from these laboratory experiments of the reaction of the specially prepared sodium sulfite with NO and O2 showed that the product contained nitrogen. Subsequent material tests showed that the reaction product resided in a very thin layer on the outermost edge of the reacted particles. The particular reaction product formed by this reaction was identified by both Time of Flight static Secondary Ion Mass Spectrometry (TOF-SIMS) analysis and by Electron Spectroscopy for Chemical Analysis (ESCA) as a mixture of sodium nitrite and sodium nitrate. SO2 was not found to be evolved by this reaction of specially prepared sodium sulfite and NO and O2. It was therefore speculated that SO3 was instead evolved and that it subsequently reacted with available sodium carbonate and formed sodium sulfate. It was experimentally found from the laboratory testing that technical grade sodium sulfite was un-reactive with NO and O2 verifying the previous findings of past investigators. It was speculated that technical grade sodium sulfite was un-reactive in this case due to surface oxidation contamination that had occurred from shelf life and handling exposure of this sodium sulfite grade to atmospheric air and moisture. 9.1.7 Summary of Chapter 7 The reaction of sodium nitrate with SO2 was studied in Chapter 7. In the overall reaction pathway for the reaction of sodium bicarbonate with SO2 flue gas with NO the reaction of sodium nitrate with SO2 was hypothesized to form NO2 and sodium sulfate according to the following reaction, 382

2NaNO3 + SO2 Na2SO4 + 2NO2

(9.7)

This reaction was hypothesized in Chapter 4 to be the third step of the simplified threestep reaction pathway for NO2 formation listed in Table 9-1. Chapter 7 reiterates that there was little agreement in the literature on the outcome of the reaction of sodium nitrate with SO2. Some past researchers reported that sodium nitrate was un-reactive with SO2; while other past researchers reported that NOx was released and sodium sulfate formed by this reaction. In Chapter 7 provides the experimental results of reaction of various preparations of sodium nitrate with SO2 and O2. Through these tests the importance of keeping the sodium nitrate surface fresh and free from hygroscopically bound water vapor was revealed. In Chapter 7 it was experimentally found through laboratory testing that certain preparations of sodium nitrate were reactive with SO2 when O2 was included in the flue gas. Manufacturer purchased technical grade sodium nitrate possessing ultra high bulk purity was found to be non-reactive with SO2 and O2. Shelf life contamination from atmospheric humidity was implicated as the reason that this sodium nitrate preparation failed to react with SO2 and O2. In Chapter 7 it is noted that sodium nitrate is hygroscopic. Moisture readily binds to the sodium nitrate surface. It was speculated that water vapor binds according to the following reaction equation, NaNO3 + 4H2O NaNO34(H2O) (9.8)

It was speculated, that sodium nitrate bound with water vapor was shielded from reaction with SO2 and O2.

383

Therefore in Chapter 7 preparations of sodium nitrate that were kept unexposed to humidity were experimentally tested for reactivity with SO2 and O2. It was found that these sodium nitrate preparations that were kept unexposed to humidity reacted with SO2 and O2 to produce NO2 and sodium sulfate. Sodium nitrate freshly prepared by Knights (1977)[18] method of reacting sodium carbonate with NO2 was found to be reactive with SO2 and O2 resulting in NO2 and sodium sulfate formation. And sodium nitrate, freshly prepared, by the first two steps of the simplified three-step reaction pathway described in Table 9-1 (reacting sodium bicarbonate with SO2 at low temperature to form sodium sulfite and then reacting this sodium sulfite with NO and O2 to form sodium nitrate) was also found to be reactive with SO2 and O2 resulting in NO2 and sodium sulfate formation. Although oxygen was not included in the speculated reaction equation of sodium nitrate with SO2 (equation 9.7), the laboratory experiments in Chapter 7 showed that oxygen was absolutely required for this reaction where NO2 and sodium sulfate were formed. Without oxygen, the laboratory experiments found that NO2 was not formed by the reaction of SO2 and either of the moisture free sodium nitrate preparations. The reaction of SO2 without oxygen with the moisture free sample formed by the first two steps of the simplified three-step reaction pathway was however found to form NO, further validating that the first two steps of the simplified three-step reaction pathway produced a mixture of sodium nitrite and sodium nitrate. Bland had previously established that sodium nitrite reacted with SO2 without O2 to form NO. In Chapter 7 NO2 formation with respect to temperature was evaluated respectively for the reaction of both forms of freshly prepared sodium nitrate with SO2 and O2 and compared to the NO2 formation found from the reaction of sodium bicarbonate with SO2

384

flue gas with NO. The NO2 formed with respect to temperature for the reaction of both preparations of sodium nitrate with SO2 and O2 closely corresponded to the NO2 formed from the reaction of sodium bicarbonate with SO2 flue gas with NO. For both of these preparations maximum NO2 formation was found to occur at 175C. Sodium sulfate, Form I was identified by x-ray diffraction analysis to form at all reaction temperatures for both preparations of specially prepared sodium nitrate with SO2 and O2. 9.1.8 Summary of Chapter 8 In Chapter 8 knowledge of the reaction pathway for NO2 formation was used to define methods to mitigate plume discoloration. Experimental investigations were conducted in Chapter 8 to investigate mitigation of NO2 by reducing flue gas oxygen concentration during sodium bicarbonate injection. These laboratory experiments showed that NO2 formation during sodium bicarbonate injection was reduced by reduction of flue gas oxygen concentration below 5%. NO2 formation was found to proportionally decrease when the oxygen concentration was decreased below 5% with complete NO2 mitigation when the flue gas oxygen concentration was zero. Duct temperature changes were also explored as plume discoloration mitigation in Chapter 8. NO2 formation with respect to temperature data for the reaction of sodium bicarbonate in SO2 flue gas with NO was analyzed to determine if practical duct injection temperature controls were possible to provide NO2 mitigation. The data showed that NO2 formation was minimum when flue gas temperatures were reduced below 110C and above 250C.

385

Injection of flue gas additives was also considered in Chapter 8 as a method of plume discoloration mitigation. In particular, water vapor additions were considered. Theoretically, water vapor additions are speculated to provide NO2 mitigation by shielding sodium nitrate from reaction with SO2 and O2. It was noted that Bortz had previously suggested employing a process step of water vapor injection to aid in NOx collection and NO2 mitigation. Water vapor injection however has negative side effects that make its use undesirable. Water vapor additions pose a threat of public nuisance violation because these additions themselves cause a white visible plume. Also it was noted that with water injection increased hardware corrosion issues might arise. Chapter 8 noted that Hooper previously patented a method of ammonia and urea additives. Ammonia and urea additives however were noted to be costly and have disadvantages associated with slip. Mitigation methods in which NO was removed from the flue gas prior to sodium bicarbonate injection were also discussed in Chapter 8. A mitigation method was devised in this study in which NO was removed by sodium sulfite injection prior to sodium bicarbonate injection. In this patented process special parameters were specified for the sodium bicarbonate injection step such that sodium sulfite product is formed. This freshly made sodium sulfite may then be transferred back to the first stage for injection for NO removal. Appendix C provides a complete copy of this U.S. Patent. Another NO2 mitigation option for sodium bicarbonate flue gas desulfurization discussed in Chapter 8 was the addition of a scrubbing step for NO2 after sodium bicarbonate flue gas desulfurization. In particular it was suggested that sodium carbonate might be injected after sodium bicarbonate flue gas desulfurization to scrub NO2 out.

386

9.2

Overall Conclusions In conclusion this study hypothesized a reaction pathway for NO2 formation for the

reaction of sodium bicarbonate with SO2 flue gas with NO, thermodynamically showed that the proposed reaction pathway was feasible, and then experimentally validated each of the three primary core reactions in the reaction pathway. This study revealed that the reaction pathway was not obvious to past researchers because experimental contamination had interfered in many of the past reactivity tests. After recognizing that contamination issues played an important role, special experimental methods were implemented to avoid contamination and provide accurate results. The reaction pathway for NO2 formation was found to stem from the formation of sodium sulfite. In this study freshly made sodium sulfite was shown to be reactive with NO and O2 to produce sodium nitrite and sodium nitrate. The sodium nitrate in turn was shown to be reactive with SO2 and O2 and directly responsible for the NO2 formation and sodium sulfate formation. This study has provided a clear understanding of the reaction pathway of NO2 formation, which occurs during flue gas desulfurization with dry sodium bicarbonate injection. With knowledge of this simplified three-step reaction pathway for NO2 formation, practical mitigation methods for plume discoloration have been suggested for implementation to allow the dry sodium bicarbonate injection method to enter the marketplace as a viable control technology. In particular, as an option for utilities in complying with Clean Air Act Phase II Acid Rain regulations the patented method suggested in Appendix C may be implemented to dually scrub SO2 and NOx from flue gas while avoiding plume discoloration.

387

BIBLIOGRAPHY [1] Quarles, J. and Lewis, W. H., Jr., The New Clean Air Act, A Guide to the Clean Air Program as Amended in 1990, Morgan, Lewis & Bockius, Washington D.C., 1990. [2] Keeth, R. J., Ireland, P. A., Moser, R. E., Economic Evaluation of Twenty-Four FGD Systems, Proceedings: Tenth Symposium on Flue Gas Desulfurization, EPRI CS5167, Volume 1, Electric Power Research Institute, Palo Alto, California, May 1987. [3] Dyni, J. R., Colorado Nahcolite Deposits: Geology and Outlook for Development, Transactions of the Society of Mining Engineers of AIME, Vol. 270, pp1853-1859, Annual Meeting, Las Vegas, Nevada, Feb. 1980. [4] Mocek, K. and Wald, J., A New Czechoslavak Process to Activate Soda Ash for Dry Scrubbing of Flue Gases, Proceedings of the fifth Annual International Pittsburgh Coal Conference, No.100, Pittsburgh, Pennsylvania, Sept., 1988. [5] Maulbetsch, J. and Offen, G., Sodium Sorbent Removal of NOx, EPRI Journal, Vol. 12, No. 7, Electric Power Research Institute, Palo Alto, California, Oct./Nov. 1987. [6] Howatson, J., Smith, J. W., Outka, D. A., Dewald, H .D., Nahcolite Properties Affecting Stack Gas Pollutant Absorption, Proceedings of the 5th National Conference on Energy and the Environment, American Institute of Chemical Engineers, Dayton, OH, 1977. [7] Carson, J.R., Removal of Sulfur Dioxide and Nitric Oxide from a flue Gas Stream by Two Sodium Alkalis of various sizes, Masters thesis, The University of Tennessee, 1980. [8] Howatson, J, Dewald, H. D., Outka, D. A.,Diller, D. J., and Cain, M. B., Reaction of Nahcolite with Sulfur Dioxide, JAPCA Vol. 30, pp.1229-1230, 1980. [9] Keener, T. C., Thermal Decomposition of Sodium Bicarbonate and its Effect on the Reaction of Sodium Bicarbonate and Sulfur Dioxide in a Simulated Flue Gas, Ph.D. dissertation, The University of Tennessee, 1982. [10] Keener, T. C. and Davis, W. T., Study of the Reaction of SO2 with NaHCO3 and Na2CO3, JAPCA, Vol. 34, No. 6, pp651-654, 1984. [11] Muzio, L. J., and Arand, J.K., Bench Scale Study of the Dry Removal of SO2 with Nahcolite and Trona, EPRI CS-1744, Electric Power Research Institute, Palo Alto, California, March 1984. [12] NaTec, History of Dry Sodium Bicarbonate Injection Technology, Technical Bulletin, Houston, Texas, 1990.

388

[13] Muzio, L. J., Sonnichsen, T. W., Green, L. P. Brines, H. G., Hooper, R. G, "22 MW Coal Fired Demonstration of Dry SO2 Scrubbing with Sodium Sorbent Compounds," Proceedings: Second Conference on Fabric Filter Technology for Coal Fired Power Plants, Denver, Colorado, EPRI CS-3257, Electric Power Research Institute, Palo Alto, California, pp. 9-1 to 9-32, Nov. 1983. [14] Fuchs, M. R., Glass, J. R., Galloway, C. S., Blythe,G. M., Muzio, L .J., Hooper, R. G., Full Scale Demonstration of Desulfurization by Dry Sodium Injection, Final Report, EPRI GS-6860, Electric Power Research Institute, Palo Alto, California, July 1990. [15] Bland, V. V. and Martin, C. E., Full-Scale demonstration of Additives for NO2 Reduction With Dry Sodium Desulfurization, EPRI GS-6852, Electric Power Research Institute, Palo Alto, California, June 1990. [16] Bland,V.V., Evaluation of Dry Sodium Sorbent Utilization in Combustion Gas SOx/ NOx Reduction, EPRI GS-6850, Electric Power Research Institute, Palo Alto, California, May 1990. [17] U.S. Environmental Protection Agency, Compilation of Air Pollution Emission Factors, 4th ed. Document No. AP-42, EPA, Research Triangle Park, North Carolina, Sept. 1985. [18] Knight, J.H., The Use of Nahcolite for Removal of Sulfur Dioxide and Nitrogen Oxides from Flue Gas, The Superior Oil Company Report, Oil Shale Department, 1977. [19] Miller, S. and Chang, R., EPRI internal report, Electric Power Research Institute, Palo Alto, California, May 1986. [20] Verlaeten, J. DeSoete, G. G., Ninane, L., and Blondel, J. M., Combined SO2/NOx Abatement By Sodium Bicarbonate Dry Injection, EPA/EPRI SO2 Control Symposium, Washington D. C., Dec. 1991. [21] Lai, Y. and Yesavage, V. F., Simultaneous Removal of SO2 and NOx By Selected Sodium Based Dry Sorbents, Final Technical Report to Public Service Co., Colorado School of Mines, 1994. [22] Kozlowski, T. R., Bartholomew, R. F., and Garfinkel, H. M., The Reaction of SO2 with Molten Nitrates and Nitrites, J. Inorg. Nucl. Chem, Vol. 32, pp. 401-409, 1970. [23] Stern, A. C., editor, Air Pollution 3rd Edition, Vol. V, Academic Press, New York, 1977. [24] White, W. H, Patterson, D.E., On the Relative Contributions of NO2 and Particles to the Color of Smoke, Atmospheric Environment, Vol. 15, No.10\11. pp. 20972104,1981.

389

[25] Charlson, R. J. and Ahlquist, N.C. Brown Haze: NO2 or Aerosol?, Atmospheric Environment , Vol. 3, p653-656, 1969. [26] Horvath H., On the Brown Color of Atmospheric Haze, Atmospheric Environment, Vol. 5, pp. 333-334, 1971. [27] Husar, R. B. and White, W. H., On the Color of the Los Angeles Smog, Atmospheric Environment, Vol.10, p199-204, 1976. [28] Brewster, D., Observations on the Lines of the Solar Spectrum, and on Those Produced by the Earths Atmosphere, and by the Action of Nitrous Acid Gas, Trans. R. Soc. Edinburgh, Vol. 12, pp. 519-530, 1834. [29] Davidson, J. A., Cantrell, C. A., McDaniel, A. H., Shetter, R. E., Mandronch, S., Calvert, J. G., Visible-ultraviolet Absorption Cross Sections for NO2 as a Function of Temperature, J. Geophys. Res. Vol. 93 p. 7105, 1988. [30] Bass, A.M., Ledford, A. E. Jr., and Laufer, A.H., Extinction Coefficients of NO2 and N2O4, J. Res. Natl.Bur. Stand., Sect. A, 80, pp. 143-166, 1976. [31] Schneider, W., Moortgat, G. K., Tyndall, G. S., and Burrows, J.P., Absorption Cross Sections of NO2 and Visible Region (200-700) at 298K, J. Photochem. Photobiol., Vol. 40, pp. 195-217, 1987. [32] Hsu, D. H., Monts, D. L., and Zare, R. N., Spectral Atlas of Nitrogen Dioxide, 5530 to 6480 Angstroms, Academic Press, Orlando, Fl., 1978. [33] Lindau, L., NO2 Effect on Flue Gas Opacity, J.Air Waste Manage.Assoc., p. 1098, 1991. [34] Hardison, L.C., Techniques for Controlling the Oxides of Nitrogen, JAPCA, Vol. 20, No. 6, pp. 377-382, 1970. [35] Melo, O. T., Stevens, R .D., The Occurrence and Nature of Brown Plumes in Ontario, Atmospheric Environment, Vol. 15, No. 12, pp. 2521-2529, 1981. [36] Kircher, O and Hougen, O. A., Effect of Water Vapor on the Catalytic Oxidation of Nitric Oxide, A.I.Ch.E. Journal, Vol. 3, pp. 331-334, 1957. [37] Duecker, W.W., and West, J.R., editors, The Manufacture of Sulfuric Acid, American Chemical Society, Robert E. Krieger Publishing Co., Huntington, NY, pp 103133, 1959. [38] Liu, H. and Chaffee, R., Evaluation of Fabric Filter as Chemical Contactor for Control of Sulfur Dioxide from Flue Gas, Air Preheater Company, National Air Pollution Control Administration Contract No. PH22-68-51, p. 82, Dec. 1969. 390

[39] Felsvang, K., Morsing, P.,and Veltman,P., Acid Rain Prevention Through New SOx/NOx Dry Scrubbing Process, Symposium on Flue Gas Desulfurization, New Orleans, Louisiana, Nov. 1983. [40] Markussen, J. M. and Drummond C. J, Investigation of Spray Drying and Dry Injection Processes: Enhanced NOx Removal Using Lime-Sodium Hydroxide Sorbent, The Coal Utilization and Environmental Contractors Review Meeting, Pittsburgh Energy Technology Center (PETC), July 1985. [41] Farber, P.S., Huang, H. S., and Allen, J. W., NOx and SOx Removal in a High Sulfur Spray Dryer System, Processing and Utilization of High Sulfur Coals II, edited by Chugh, V. P., Caudle, R .D., Elsevier Science Publishing Company, Inc. pp. 481-492, 1987. [42] Oxley, J.H., Quick Response Inquiries-Telephone Interviews, Battelle Columbus Laboratories, 1986. [43] Lowel, P. S., et al., Testing of Sodium Sorbent Injection for Simultaneous SO2, NOx, and Particulate Removal in a Fabric Filter, Proceedings: Third Conference on Fabric Filter Technology for Coal Fired Power Plants, Scottsdale, Arizona,Electric Power Research Institute, Palo Alto, California, Nov., 1985. [44] Hooper, R.G. and Bland, V. V., Combined SO2 and NOx Removal From Flue Gas by Injection of Sodium Sorbents, Proceedings: Third Conference on Fabric Filter Technology for Coal Fired Power Plants, Scottsdale, Arizona,Electric Power Research Institute, Palo Alto, California, Nov., 1985. [45] Bortz, S.J., Evaluation of NOx Removal and NO2 Emissions in Dry Sodium Desulfuization Processes, EPRI TR-103667, Electric Power Research Institute, Palo Alto, California, March, 1994 [46] Barrall, E. M. and Rogers, L. B., Differential Thermal Analysis of the Decomposition of Sodium Bicarbonate and its Simple Double Salts, J. Inorg. Nucl.Chem., Vol. 28, pp. 41-51, 1966. [47] Subramanian, K. S., Radha Krishnan, T. P. and Sundram, A. K., Thermal Decomposition Kinetics of Sodium Bicarbonate by Differential Thermal Analysis, J. Thermal Analysis, Vol. 4, pp. 89-93, 1972. [48] Templeton, C.C., Pressure-Temperature Relationship for Decomposition of Sodium Bicarbonate from 200 to 600F, Journal of Chemical and Engineering Data, Vol. 23, No. 1, pp. 7-11, 1978. [49] Keener, T. C., Frazier, G. C. and Davis, W. T., Thermal Decomposition of Sodium Bicarbonate, Chem. Engng. Commun. Vol. 33, pp. 93-105, 1985. 391

[50] Hu, W., Smith, J. M., Dogu, T. and Dogu, G., Kinetics of Sodium Bicarbonate Decomposition, A.I.Ch.E. Journal, Vol. 32, pp. 1229-1230, 1986. [51] Stern, F.R., Bench Scale Study of Sulfur and Nitrogen Oxides Adsorption by Nahcolite and Trona, Master's thesis, University of North Dakota, p.31, 1978. [52] Keener, T. C. and Khang, S. J., Kinetics of the Sodium Bicarbonate-Sulfur Dioxide Reaction., Chemical Engineering Science, Vol. 48, No. 16, pp. 2859-2865, 1993. [53] Ackermann, M. W., and Powel, R. E., Air Oxidation of Hydroxylamine-Nsulfonate, Inorg.Chem., Vol.6, p.1718-1720, 1967. [54] Martin, L. R., Damschen, D. E., Judeikis, H. S., The Reactions of Nitrogen Oxides with SO2 in Aqueous Aerosols, Atmospheric Environment, Vol.15, p.191, 1981. [55] Ito, Yukio, Masuda, Etsuro, and Yasumoto, Yoshiro, Production of Trisodium Imidodisulfate from Nitrogen Oxides, Sulfur Dioxide, and Sodium Hydroxide, Ind. Eng. Chem. Process Des. Dev., Vol. 23, pp. 382-383, 1984. [56] Takeuchi, H., Ando, M., and Kizawa, N., Absorbtion of Nitrogen Oxides in Aqueous Sodium Sulfite and Bisulfite Solutions, Ind. Eng. Chem., Process Des. Dev., Vol. 16, No. 3, pp. 303-308, 1977. [57] Weisweiler, W., and Blumhofer, R., Absorption of NOx in Aqueous Solutions of Na2SO3/NaHSO3 and Simultaneous Absorption of NOx and SO2 in NaOH, Ger. Chem. Eng., Vol. 7, p 241-247, 1984. [58] Sada, E., Kumanzawa, H., and Yoshikawa, Y., Simultaneous Removal of NO and SO2 by Absorption into Aqueous Mixed Solutions, AIChE Journal, Vol. 34, No. 7, pp. 1215-1220, July 1988. [59] Littlejohn, D., Hu, K. Y., and Chang, G., Kinetics of the Reaction of Nitric Oxide with Sulfite and Bisulfite Ions in Aqueous Solution, Inorg. Chem. 25, pp. 3131-3135, 1986. [60] Takeuchi, H., and Yamanaka, Y., Simultaneous Absorption of SO2 and NO2 in Aqueous Solutions of NaOH and Na2SO3, Ind. Eng. Chem., Process Des. Dev., Vol. 17, No. 4, p. 389, 1978. [61] Mellor, J. W., A Comprehensive Treatise on Inorganic and Theoretical Chemistry, Longmans, Green and Company, New York, p. 263, 1947. [62] Lide, D.R., CRC Handbook of Chemistry and Physics, 75th Edition, CRC Press, Boca Raton, p 4-99, 1994.

392

[63] Code of Federal Regulations, Multipurpose Additives, Title 21, Volume 3, Sec.172.824, U.S. Government Printing Office, Washington DC, 1997. [64] Stopperka, K., Wolf, F., Suss, G., Investigation of the Reaction between Nitrogen Dioxide and Sulfur Dioxide in the Gaseous Phase, Zeitschrift Fur anorganische und allgemeine Chemie, Vol. 359, pp. 14-29, 1968. [65] Muzio,L.J. and, Arand, J.K., Bench Scale Study of the Dry Removal of SO2 With Nahcolite and Trona, Final Report, EPRI CS-1744 (1981). [70] Kracek, F. C., The Polymorphisim of Sodium Sulfate:I Thermal Analysis, The Journal of Physical Chemistry, Vol. 33, pp1281-1303, 1930. [71] Kracek, F. C. and Gibson, R.E., The Polymorphisim of Sodium Sulfate: II. The Densities for Anhydruous Sodium Sulfate at 25 Degrees Celcius, The Journal of Physical Chemistry, Vol. 33, pp.1304-1308, 1930. [72] Kracek, F. C. and Gibson, R.E., The Polymorphisim of Sodium Sulfate: III. Dilatometer Investigations, The Journal of Physical Chemistry, Vol. 34, pp188-206, 1930. [73] Kracek, F. C. and Ksanda, C.J., The Polymorphisim of Sodium Sulfate: IV X-Ray Analysis , The Journal of Physical Chemistry, Vol. 34, pp1741-1744, 1930. [74] Mehrotra, B.N., Structural Transformations and Phase Stabilization in Na2SO4Na2CO3 System,, Phase Transitions, Vol. 16/17, pp.431-436, 1989 [75] Briggs, D., Brown, A., Vickerman, J.C., Handbook of Static Secondary Ion Mass Spectrometry (SIMS), John Wiley & Sons, New York, 1989 pp. 116-117. [76] Burnell, L. K., Hyde, J.D., Kolak, N.P., Gibbs, R.E., Control of Nitrogen Oxides From Stationary Source Effluents, United States Patent 4,140,651, Feb. 20, 1979. [77] Quanquin, B. and Trimbach, H., Process For Recovering Oxides Of Nitrogen, United States Patent, No. 3,348,914, Oct. 24, 1967. [78] Barin, Ihsan, Thermochemical Data of Pure Substances, Volume I and II, Third Edition,VCH, New York, 1990. [79] Hooper, Richard G, Method for ESP Brown Plume Control, U.S. Patent #4,844,915, July, 4, 1989. [80] Cook, L. P. and McMurdie, H.F., Phase Diagrams for Ceramics, Volume VII. Sangster, J. and Pelton, A.D.; The American Ceramic Society, Columbus, Ohio, 1989, Figure 7050.

393

APPENDIX A A.1 Introduction

The following are the spread-sheets for the calculation of Gibbs free energy of each reaction with respect to temperature. Each reaction is considered a closed system. A.2 Gibbs free energy equation

Gibbs free energy, Gr, is calculated by the following equation: Gr = R * T * Ln (K r ) where: Gr R T Kr is Gibbs free energy of the reaction is the universal gas constant is the temperature is the equilibrium constant for the reaction (A. 1)

The overall Gibbs free energy of the reaction is equal to the summation of all of the reactants and product Gibbs free energy ratio with their respective stoichiometric coefficients: log K r = ! i * log K f (i ) where: Kr i Kf( i ) is the equilibrium constant for the reaction is the stoichiometric coefficient of substance i is the equilibrium constant for each reactant and product (A. 2)

substances

394

The stoichiometric coefficients of the products are by definition positive while the stoichiometric ratio of the reactants are by definition considered negative. Each of the reactant and product log Kf( i ) values with respect to temperature have been previously tabulated in a thermodynamic table most commonly known as the JANF tables. More recently the original set of data of the JANF tables have been expanded to include more substances and at more temperatures by Barin (1990)[78]. For temperatures that were not specifically tabulated in the Barin (1990)[78] tables linear interpolation was used. It is assumed that the reaction proceeds to the right hand side if the Gibbs free energy is found to be negative and to the left hand side if it is positive. An example of a calculation for the following reaction: C + H2O CO + H2 at 1000K is as follows: log K r = ! i *log K f (i ) log Kr =-1* log KC + -1* log KH2O + 1*log KCO +1* log KH2 at 1000 K the following are found in the Barin(1990)[78] tables: log KC =0 log KH2O = 10.062 log KCO =10.461 log KH2 = 0 therefore the summation is equal to: log Kr =-1*0 + -1* 10.062 + 1*10.461 +1* 0 log Kr =0.399
Kr =100.399 =2.506

(A. 3)

(A. 4) (A. 5)

(A. 6) (A. 7) (A. 8) (A. 9)

(A. 10) (A. 11) (A. 12)

395

Gr = R * T * Ln( K r ) = 8.31441 * 1000 * Ln(2.506) = -7638

(A. 13)

Because the Gibbs free energy is negative the reaction is expected to proceed to the right hand side at 1000 K forming CO and H2 for this example. A.3 Reactions thermodynamically evaluated.

Thermodynamic calculations are evaluated for the following reactions: 2NaHCO3 Na2CO3 + H2O + CO2 Na2CO3 + SO2 + O2 Na2SO4 + CO2 2NaHCO3 + SO2 Na2SO3 + 2CO2 + H2O Na2CO3 + SO2 Na2SO3 + CO2 Na2SO3 + 2NO + O2 2NaNO2 + SO3 Na2SO3 + 2NO + 2O2 2NaNO3 + SO3 2NaNO2 + SO2 Na2SO4 + 2NO 2NaNO2 + SO2 + O2 Na2SO4 + 2NO2 2NaNO3 + SO2 Na2SO4 + 2NO2 Na2CO3 + NO2 2NaNO3 + NO + CO2 Na2SO3 + 2NO2 + O2 2NaNO3 + SO3 A.4 (A. 14) (A. 15) (A. 16) (A. 17) (A. 18) (A. 19) (A. 20) (A. 21) (A. 22) (A. 23) (A. 24)

Equilibrium calculations from 27 to 627C for each of the reactions.

Spread sheet for equation A.14. The equilibrium calculations for reaction equation A. 14 are listed below

396

Table APPENDIX A-1 Equilibrium calculations for reaction equation A. 14

2 T (C) 27 50 100 125 127 150 175 185 200 227 T(K) 300 323 373 398 400 423 448 458 473 500 2 2 2 2 2 2 2 2 2 2 2 NaHCO3 "-->" 1 NaHCO3 148.389 138.861 120.403 110.818 106.963 101.234 95.006 92.515 88.779 82.053 1 1 1 1 1 1 1 1 1 1 1 Na2CO3 1 Na2CO3 182.385 171.053 149.096 137.696 133.115 126.296 118.883 115.918 111.471 103.465 1 1 1 1 1 1 1 1 1 1 1 H2O H2O 39.792 37.366 32.665 30.224 29.245 27.784 26.195 25.560 24.607 22.891 1 1 1 1 1 1 1 1 1 1 1 1 CO2 CO2 68.666 64.726 57.096 53.133 51.535 49.171 46.601 45.573 44.031 41.255 logK Ktotal Gibbs J/m -5.935 0.000 34087 -4.577 0.000 28303 -1.948 0.011 13913 -0.583 0.261 4442 -0.031 0.931 237 0.782 6.057 -6335 1.666 46.375 -14291 2.020 104.684 -17711 2.550 355.042 -23094 3.505 3198.895 -33551

Reaction A.14 switches its equilibrium direction between 127C and 150C. The analysis shows that above 150C reaction A.14 proceeds to the right resulting in the thermal decomposition of sodium bicarbonate.

397

Spread sheet for equation A.15. Table APPENDIX A-2 Equilibrium calculations for reaction equation A. 15
1 Na2CO3 T (C) 27 50 100 125 127 150 175 185 200 227 241 250 275 284 300 306 327 350 400 427 450 450 500 527 550 627 T(K) 300 323 373 398 400 423 448 458 473 500 514 523 548 557 573 579 600 623 673 700 723 723 773 800 823 900 1 Na2CO3 1 182.385 1 171.053 1 149.096 1 137.696 1 133.115 1 126.296 1 118.883 1 115.918 1 111.471 1 103.465 1 100.702 1 98.926 1 93.993 1 92.217 1 89.060 1 87.876 1 83.732 1 80.501 1 73.478 1 69.686 1 67.013 1 67.013 1 61.938 1 59.198 1 57.327 1 51.062 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 SO2 SO2 52.255 49.275 43.492 40.490 39.299 37.489 35.522 34.735 33.555 31.430 30.690 30.214 28.892 28.416 27.570 27.253 26.143 25.269 23.370 22.345 21.687 21.687 20.255 19.482 18.952 17.178 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 O2 "-->" 1 O2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Na2SO4 1 Na2SO4 220.973 207.060 179.638 165.483 160.483 151.327 141.375 137.394 132.616 124.015 120.041 117.929 112.064 109.953 106.199 104.791 99.864 95.906 87.302 82.656 79.693 79.693 73.251 69.772 67.457 59.705 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 CO2 CO2 68.666 64.726 57.096 53.133 51.535 49.171 46.601 45.573 44.031 41.255 40.295 39.678 37.965 37.348 36.251 35.840 34.400 33.273 30.824 29.501 28.656 28.656 26.818 25.825 25.167 22.965 logK 55 51.46 44.15 40.43 39.6 36.71 33.57 32.31 31.62 30.38 28.94 28.47 27.14 26.67 25.82 25.5 24.39 23.41 21.28 20.13 19.65 19.65 17.87 16.92 16.34 14.43 Ktotal 1.0E+55 2.9E+51 1.4E+44 2.7E+40 4.0E+39 5.2E+36 3.7E+33 2.1E+32 4.2E+31 2.4E+30 8.8E+28 2.9E+28 1.4E+27 4.6E+26 6.6E+25 3.2E+25 2.4E+24 2.6E+23 1.9E+21 1.3E+20 4.5E+19 4.5E+19 7.5E+17 8.3E+16 2.2E+16 2.7E+14 Gibbs J/mol -315881 -318203 -315247 -308056 -303282 -297308 -287928 -283334 -286344 -290759 -284820 -285034 -284767 -284360 -283232 -282675 -280151 -279196 -274140 -269713 -271969 -271969 -264527 -259096 -257532 -248631

Under equilibrium conditions A.15 proceeds to the right at all temperatures calculated. Spread sheet for equation A.16. Table APPENDIX A-3 Equilibrium calculations for reaction equation A. 16
2 NaHCO3 1 T (C) T(K) 27 300 50 323 100 373 125 398 127 400 150 423 175 448 185 458 200 473 227 500 2 2 2 2 2 2 2 2 2 2 2 NaHCO3 148.389 138.861 120.403 110.818 106.963 101.234 95.006 92.515 88.779 82.053 1 1 1 1 1 1 1 1 1 1 1 SO2 SO2 52.255 49.275 43.492 40.490 39.299 37.489 35.522 34.735 33.555 31.430 "--->" 1 Na2SO3 2 1 1 1 1 1 1 1 1 1 1 1 Na2SO3 176.166 165.128 143.727 132.618 128.175 121.509 114.263 111.365 107.017 99.192 2 2 2 2 2 2 2 2 2 2 2 CO2 CO2 68.666 64.726 57.096 53.133 51.535 49.171 46.601 45.573 44.031 41.255 1 1 1 1 1 1 1 1 1 1 1 1 H2O H2O 39.792 37.366 32.665 30.224 29.245 27.784 26.195 25.560 24.607 22.891 logKtotal 4.26 4.95 6.29 6.98 7.27 7.68 8.13 8.30 8.57 9.06 Ktotal 1.8E+04 8.9E+04 1.9E+06 9.6E+06 1.8E+07 4.8E+07 1.3E+08 2.0E+08 3.7E+08 1.1E+09 Gibbs J/mol -24450 -30602 -44894 -53200 -55634 -62171 -69688 -72815 -77634 -86696

Under equilibrium conditions A.16 sulfation proceeds to the right at all temperatures calculated.

398

Spread sheet for equation A.14. Table APPENDIX A-4 Equilibrium calculations for reaction equation A. 17
1 T (C) 27 50 100 125 127 150 175 185 200 227 241 250 275 284 300 306 327 350 400 427 450 450 500 527 550 627 T(K) 300 323 373 398 400 423 448 458 473 500 514 523 548 557 573 579 600 623 673 700 723 723 773 800 823 900 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Na2CO3 1 Na2CO3 182.385 171.053 149.096 137.696 133.115 126.296 118.883 115.918 111.471 103.465 100.702 98.926 93.993 92.217 89.060 87.876 83.732 80.501 73.478 69.686 67.013 67.013 61.938 59.198 57.327 51.062 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 SO2 SO2 52.255 49.275 43.492 40.490 39.299 37.489 35.522 34.735 33.555 31.430 30.690 30.214 28.892 28.416 27.570 27.253 26.143 25.269 23.370 22.345 21.687 21.687 20.255 19.482 18.952 17.178 "-->" 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Na2SO3 Na2SO3 176.166 165.128 143.727 132.618 128.175 121.509 114.263 111.365 107.017 99.192 96.483 94.741 89.903 88.161 85.064 83.903 79.839 76.657 69.740 66.005 63.618 63.618 58.428 55.626 53.754 47.486 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 CO2 CO2 68.666 64.726 57.096 53.133 51.535 49.171 46.601 45.573 44.031 41.255 40.295 39.678 37.965 37.348 36.251 35.840 34.400 33.273 30.824 29.501 28.656 28.656 26.818 25.825 25.167 22.965 logK 10.19 9.53 8.24 7.57 7.30 6.89 6.46 6.28 6.02 5.55 5.39 5.28 4.98 4.87 4.68 4.61 4.36 4.16 3.72 3.47 3.57 3.57 3.05 2.77 2.64 2.21 Ktotal Gibbs J/mol 1.6E+10 -58537 3.4E+09 -58906 1.7E+08 -58807 3.7E+07 -57642 2.0E+07 -55872 7.9E+06 -55836 2.9E+06 -55397 1.9E+06 -55104 1.1E+06 -54540 3.6E+05 -53146 2.4E+05 -52997 1.9E+05 -52854 9.6E+04 -52265 7.5E+04 -51983 4.8E+04 -51391 4.1E+04 -51139 2.3E+04 -50128 1.4E+04 -49611 5.2E+03 -47866 3.0E+03 -46569 3.7E+03 -49468 3.7E+03 -49468 1.1E+03 -45174 5.9E+02 -42440 4.4E+02 -41631 1.6E+02 -38096

Under equilibrium conditions A.17 sulfation proceeds to the right at all temperatures calculated.

399

Spread sheet for equation A.18. Table APPENDIX A-5Equilibrium reaction equation A. 18
1 Na2SO3 2 T (C) 27 50 100 125 127 150 175 185 200 227 241 250 275 284 300 306 327 350 400 427 450 450 500 527 550 627 T(K) 300 323 373 398 400 423 448 458 473 500 514 523 548 557 573 579 600 623 673 700 723 723 773 800 823 900 NO 1 O2 "--> 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 NaNO2 NaNO2 50.435 46.836 39.863 36.243 34.786 32.622 30.270 29.329 27.918 25.378 24.438 23.834 22.156 21.552 20.700 20.380 19.261 18.295 16.194 15.060 14.345 14.345 12.790 11.950 11.401 9.563 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Na2SO3 2 NO 1 O2 1 176.166 2 -15.074 1 0 1 165.128 2 -14.170 1 0 1 143.727 2 -12.419 1 0 1 132.618 2 -11.510 1 0 1 128.175 2 -11.143 1 0 1 121.509 2 -10.600 1 0 1 114.263 2 -10.010 1 0 1 111.365 2 -9.774 1 0 1 107.017 2 -9.420 1 0 1 99.192 2 -8.783 1 0 1 96.483 2 -8.563 1 0 1 94.741 2 -8.421 1 0 1 89.903 2 -8.028 1 0 1 88.161 2 -7.886 1 0 1 85.064 2 -7.635 1 0 1 83.903 2 -7.540 1 0 1 79.839 2 -7.210 1 0 1 76.657 2 -6.951 1 0 1 69.740 2 -6.389 1 0 1 66.005 2 -6.086 1 0 1 63.618 2 -5.892 1 0 1 63.618 2 -5.892 1 0 1 58.428 2 -5.471 1 0 1 55.626 2 -5.243 1 0 1 53.754 2 -5.092 1 0 1 47.486 2 -4.587 1 0

calculations
SO3 SO3 64.573 60.601 52.896 48.897 47.304 44.898 42.283 41.237 39.668 36.844 35.862 35.231 33.478 32.847 31.725 31.304 29.831 28.675 26.161 24.803 23.933 23.933 22.042 21.021 20.327 18.003 logK 19.43 17.48 13.73 11.78 10.99 9.83 8.58 8.08 7.33 5.97 5.38 5.00 3.94 3.56 3.33 3.24 2.93 2.51 1.59 1.09 0.79 0.79 0.13 -0.22 -0.44 -1.18 0.00

for

Ktotal Gibbs J/mol 2.7E+19 -111565 3.0E+17 -108118 5.4E+13 -98065 6.1E+11 -89784 9.7E+10 -84137 6.8E+09 -79638 3.8E+08 -73595 1.2E+08 -70843 2.1E+07 -66354 9.4E+05 -57185 2.4E+05 -52958 1.0E+05 -50073 8.8E+03 -41371 3.7E+03 -37990 2.1E+03 -36515 1.7E+03 -35925 8.6E+02 -33702 3.2E+02 -29936 3.9E+01 -20459 1.2E+01 -14607 6.2E+00 -10920 6.2E+00 -10920 1.4E+00 -1989 6.0E-01 3354 3.6E-01 6944 6.6E-02 20383 1.0E+00 0

Under equilibrium conditions reaction A.18 proceeds to the right at temperatures below 500C.

400

Spread sheet for equation A.19. Table APPENDIX A-6 Equilibrium calculations for reaction equation A. 19
1 T (C) 27 50 100 125 127 150 175 185 200 227 241 250 275 284 300 306 327 350 400 427 450 450 500 527 550 627 T(K) 300 323 373 398 400 423 448 458 473 500 514 523 548 557 573 579 600 623 673 700 723 723 773 800 823 900 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Na2SO3 2 Na2SO3 176.166 165.128 143.727 132.618 128.175 121.509 114.263 111.365 107.017 99.192 96.483 94.741 89.903 88.161 85.064 83.903 79.839 76.657 69.740 66.005 63.618 63.618 58.428 55.626 53.754 47.486 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 NO NO -15.074 -14.170 -12.419 -11.510 -11.143 -10.600 -10.010 -9.774 -9.420 -8.783 -8.563 -8.421 -8.028 -7.886 -7.635 -7.540 -7.210 -6.951 -6.389 -6.086 -5.892 -5.892 -5.471 -5.243 -5.092 -4.587 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 O2 "--> 2 O2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 NaNO3 1 NaNO3 63.793 59.107 50.035 45.323 43.420 40.611 37.559 36.338 34.506 31.209 29.970 29.173 26.960 26.281 25.075 24.622 23.230 21.979 19.259 17.790 16.865 16.865 14.854 13.768 13.059 10.684 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 SO3 SO3 64.573 60.601 52.896 48.897 47.304 44.898 42.283 41.237 39.668 36.844 35.862 35.231 33.478 32.847 31.725 31.304 29.831 28.675 26.161 24.803 23.933 23.933 22.042 21.021 20.327 18.003 logK 46.14 42.03 34.08 29.94 28.26 25.81 23.16 22.10 20.50 17.64 16.44 15.68 13.55 13.02 12.08 11.73 10.87 9.88 7.72 6.55 5.83 5.83 4.26 3.42 2.87 1.06 Ktotal Gibbs J/mol 1.4E+46 -265006 1.1E+42 -259884 1.2E+34 -243336 8.8E+29 -228163 1.8E+28 -216373 6.5E+25 -209035 1.4E+23 -198620 1.2E+22 -193743 3.2E+20 -185664 4.3E+17 -168817 2.8E+16 -161820 4.8E+15 -156985 3.6E+13 -142168 1.0E+13 -138851 1.2E+12 -132501 5.3E+11 -129972 7.4E+10 -124884 7.5E+09 -117815 5.2E+07 -99428 3.5E+06 -87778 6.8E+05 -80688 6.8E+05 -80688 1.8E+04 -63086 2.6E+03 -52334 7.5E+02 -45293 1.1E+01 -18247

Under equilibrium conditions reaction A-19 proceeds to the right at all temperatures calculated.

401

Spread sheet for equation A.20. Table APPENDIX A-7 Equilibrium calculations for reaction equation A. 20
2 T (C) T(K) 27 300 50 323 100 373 125 398 127 400 150 423 175 448 185 458 200 473 227 500 241 514 250 523 275 548 284 557 300 573 306 579 327 600 350 623 400 673 427 700 450 723 450 723 500 773 527 800 550 823 627 900 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 NaNO2 1 NaNO2 50.435 46.836 39.863 36.243 34.786 32.622 30.270 29.329 27.918 25.378 24.438 23.834 22.156 21.552 20.700 20.380 19.261 18.295 16.194 15.060 14.345 14.345 12.790 11.950 11.401 9.563 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 SO2 "--->" 1 Na2SO4 2 SO2 52.255 49.275 43.492 40.490 39.299 37.489 35.522 34.735 33.555 31.430 30.690 30.214 28.892 28.416 27.570 27.253 26.143 25.269 23.370 22.345 21.687 21.687 20.255 19.482 18.952 17.178 NO

1 Na2SO4 2 NO logK Ktotal Gibbs J/mol -216526 1 220.973 2 -15.074 37.70 5.0E+37 -221216 1 207.060 2 -14.170 35.77 5.9E+35 1 179.638 2 -12.419 31.58 3.8E+31 -225532 -224686 1 165.483 2 -11.510 29.49 3.1E+29 -224574 1 160.483 2 -11.143 29.33 2.1E+29 -221835 1 151.327 2 -10.600 27.39 2.5E+27 -216927 1 141.375 2 -10.010 25.29 2.0E+25 -214400 1 137.394 2 -9.774 24.45 2.8E+24 -220811 1 132.616 2 -9.420 24.38 2.4E+24 -232253 1 124.015 2 -8.783 24.26 1.8E+24 1 120.041 2 -8.563 23.35 2.2E+23 -229763 -232340 1 117.929 2 -8.421 23.20 1.6E+23 -239239 1 112.064 2 -8.028 22.80 6.4E+22 1 109.953 2 -7.886 22.66 4.6E+22 -241629 -240895 1 106.199 2 -7.635 21.96 9.1E+21 -240509 1 104.791 2 -7.540 21.70 5.0E+21 1 99.864 2 -7.210 20.78 6.0E+20 -238684 -240262 1 95.906 2 -6.951 20.14 1.4E+20 -241764 1 87.302 2 -6.389 18.76 5.8E+18 -241477 1 82.656 2 -6.086 18.02 1.0E+18 -242678 1 79.693 2 -5.892 17.53 3.4E+17 -242678 1 79.693 2 -5.892 17.53 3.4E+17 -243811 1 73.251 2 -5.471 16.48 3.0E+16 -243581 1 69.772 2 -5.243 15.90 8.0E+15 -244507 1 67.457 2 -5.092 15.52 3.3E+15 -245134 1 59.705 2 -4.587 14.23 1.7E+14

Under equilibrium conditions reaction A.20 proceeds to the right at all temperatures calculated.

402

Spread sheet for equation A.21. Table APPENDIX A-8 Equilibrium calculations for reaction equation A. 21
2 NaNO2 1 T (C) 27 50 100 125 127 150 175 185 200 227 241 250 275 284 300 306 327 350 400 427 450 450 500 527 550 627 T(K) 300 323 373 398 400 423 448 458 473 500 514 523 548 557 573 579 600 623 673 700 723 723 773 800 823 900 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 NaNO2 50.435 46.836 39.863 36.243 34.786 32.622 30.270 29.329 27.918 25.378 24.438 23.834 22.156 21.552 20.700 20.380 19.261 18.295 16.194 15.060 14.345 14.345 12.790 11.950 11.401 9.563 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 SO2 SO2 52.255 49.275 43.492 40.490 39.299 37.489 35.522 34.735 33.555 31.430 30.690 30.214 28.892 28.416 27.570 27.253 26.143 25.269 23.370 22.345 21.687 21.687 20.255 19.482 18.952 17.178 1 O2 "--->" 1 Na2SO4 2 1 O2 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Na2SO4 220.973 207.060 179.638 165.483 160.483 151.327 141.375 137.394 132.616 124.015 120.041 117.929 112.064 109.953 106.199 104.791 99.864 95.906 87.302 82.656 79.693 79.693 73.251 69.772 67.457 59.705 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 NO2 NO2 -8.945 -8.617 -7.982 -7.652 -7.517 -7.322 -7.110 -7.026 -6.899 -6.670 -6.592 -6.541 -6.401 -6.351 -6.261 -6.228 -6.110 -6.018 -5.819 -5.712 -5.643 -5.643 -5.494 -5.414 -5.361 -5.182 logK 49.96 46.88 40.46 37.20 36.58 33.95 31.09 29.95 29.43 28.49 27.29 26.96 26.06 25.73 24.71 24.32 22.98 22.01 19.90 18.77 18.03 18.03 16.43 15.56 14.98 13.04 Ktotal Gibbs J/mol 9.1E+49 -286928 7.6E+46 -289896 2.9E+40 -288900 1.6E+37 -283471 3.8E+36 -280109 8.9E+33 -274927 1.2E+31 -266668 8.9E+29 -262599 2.7E+29 -266478 3.1E+28 -272706 2.0E+27 -268557 9.2E+26 -269988 1.1E+26 -273373 5.4E+25 -274379 5.1E+24 -271029 2.1E+24 -269611 9.5E+22 -263955 1.0E+22 -262519 8.0E+19 -256453 5.8E+18 -251501 1.1E+18 -249561 1.1E+18 -249561 2.7E+16 -243105 3.6E+15 -238343 9.6E+14 -236045 1.1E+13 -224630

Under equilibrium conditions reaction A.21 proceeds to the right at all temperatures calculated.

403

Spread sheet for equation A.22. Table APPENDIX A-9 Equilibrium calculations for reaction equation A. 22
2 NaNO3 T (C) 27 50 100 125 127 150 175 185 200 227 241 250 275 284 300 306 327 350 400 427 450 450 500 527 550 627 T(K) 300 323 373 398 400 423 448 458 473 500 514 523 548 557 573 579 600 623 673 700 723 723 773 800 823 900 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 NaNO3 63.793 59.107 50.035 45.323 43.420 40.611 37.559 36.338 34.506 31.209 29.970 29.173 26.960 26.281 25.075 24.622 23.230 21.979 19.259 17.790 16.865 16.865 14.854 13.768 13.059 10.684 1 SO2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 SO2 52.255 49.275 43.492 40.490 39.299 37.489 35.522 34.735 33.555 31.430 30.690 30.214 28.892 28.416 27.570 27.253 26.143 25.269 23.370 22.345 21.687 21.687 20.255 19.482 18.952 17.178 "--->" 1 Na2SO4 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Na2SO4 220.973 207.060 179.638 165.483 160.483 151.327 141.375 137.394 132.616 124.015 120.041 117.929 112.064 109.953 106.199 104.791 99.864 95.906 87.302 82.656 79.693 79.693 73.251 69.772 67.457 59.705 2 NO2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 NO2 -8.945 -8.617 -7.982 -7.652 -7.517 -7.322 -7.110 -7.026 -6.899 -6.670 -6.592 -6.541 -6.401 -6.351 -6.261 -6.228 -6.110 -6.018 -5.819 -5.712 -5.643 -5.643 -5.494 -5.414 -5.361 -5.182 logK 23.24 22.34 20.11 19.04 19.31 17.97 16.51 15.93 16.25 16.83 16.23 16.29 16.45 16.27 15.96 15.84 15.04 14.64 13.78 13.31 12.99 12.99 12.30 11.93 11.67 10.80 Ktotal 1.7E+23 2.2E+22 1.3E+20 1.1E+19 2.0E+19 9.3E+17 3.3E+16 8.6E+15 1.8E+16 6.7E+16 1.7E+16 1.9E+16 2.8E+16 1.9E+16 9.1E+15 6.9E+15 1.1E+15 4.4E+14 6.0E+13 2.0E+13 9.8E+12 9.8E+12 2.0E+12 8.4E+11 4.6E+11 6.2E+10 Gibbs J/mol -133488 -138130 -143629 -145092 -147873 -145529 -141643 -139698 -147167 -161073 -159695 -163076 -172575 -173518 -175043 -175565 -172773 -174639 -177484 -178330 -179793 -179793 -182009 -182655 -183808 -186000

Under equilibrium conditions reaction A.22 proceeds to the right at all temperatures calculated.

404

Spread sheet for equation A.23. Table APPENDIX A-10 Equilibrium calculations for reaction equation A. 23
1 Na2CO3 1 NO2 "--->" 2 NaNO3 1 NO 1 CO2

T (C) T(K) 1 Na2CO3 27 300 1 182.385 50 323 1 171.053 100 373 1 149.096 125 398 1 137.696 127 400 1 133.115 150 423 1 126.296 175 448 1 118.883 185 458 1 115.918 200 473 1 111.471 227 500 1 103.465 241 514 1 100.702 250 523 1 98.926 275 548 1 93.993 284 557 1 92.217 300 573 1 89.060 306 579 1 87.876 327 600 1 83.732 350 623 1 80.501 400 673 1 73.478 427 700 1 69.686 450 723 1 67.013 450 723 1 67.013 500 773 1 61.938 527 800 1 59.198 550 823 1 57.327 627 900 1 51.062

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

NO2 -8.945 -8.617 -7.982 -7.652 -7.517 -7.322 -7.110 -7.026 -6.899 -6.670 -6.592 -6.541 -6.401 -6.351 -6.261 -6.228 -6.110 -6.018 -5.819 -5.712 -5.643 -5.643 -5.494 -5.414 -5.361 -5.182

2 NaNO3 1 NO 1 CO2 2 63.793 1 -15.074 1 68.666 2 59.107 1 -14.170 1 64.726 2 50.035 1 -12.419 1 57.096 2 45.323 1 -11.510 1 53.133 2 43.420 1 -11.143 1 51.535 2 40.611 1 -10.600 1 49.171 2 37.559 1 -10.010 1 46.601 2 36.338 1 -9.774 1 45.573 2 34.506 1 -9.420 1 44.031 2 31.209 1 -8.783 1 41.255 2 29.970 1 -8.563 1 40.295 2 29.173 1 -8.421 1 39.678 2 26.960 1 -8.028 1 37.965 2 26.281 1 -7.886 1 37.348 2 25.075 1 -7.635 1 36.251 2 24.622 1 -7.540 1 35.840 2 23.230 1 -7.210 1 34.400 2 21.979 1 -6.951 1 33.273 2 19.259 1 -6.389 1 30.824 2 17.790 1 -6.086 1 29.501 2 16.865 1 -5.892 1 28.656 2 16.865 1 -5.892 1 28.656 2 14.854 1 -5.471 1 26.818 2 13.768 1 -5.243 1 25.825 2 13.059 1 -5.092 1 25.167 2 10.684 1 -4.587 1 22.965

logK 7.74 6.33 3.63 2.23 1.63 0.82 -0.06 -0.42 -0.95 -1.91 -2.44 -2.78 -3.74 -3.84 -4.03 -4.11 -3.97 -4.20 -4.71 -4.98 -4.88 -4.88 -5.39 -5.67 -5.77 -6.13

Ktotal Gibbs J/mol 5.5E+07 -44442 2.2E+06 -39168 4.3E+03 -25937 1.7E+02 -16962 4.3E+01 -12513 6.6E+00 -6641 8.6E-01 555 3.8E-01 3671 1.1E-01 8598 1.2E-02 18235 3.6E-03 23999 1.7E-03 27855 1.8E-04 39188 1.4E-04 40977 9.3E-05 44247 7.9E-05 45504 1.1E-04 45625 6.3E-05 50137 2.0E-05 60648 1.0E-05 66725 1.3E-05 67495 1.3E-05 67495 4.1E-06 79752 2.2E-06 86779 1.7E-06 90970 7.3E-07 105690

Under equilibrium conditions reaction A.23 proceeds to the right up to 150C. Above 150C this reaction proceeds to the left.

405

Spread sheet for equation A.24. Table APPENDIX A-11 Equilibrium calculations for reaction equation A. 24
1 Na2SO3 2 NO2 1 O2 "--> 2 NaNO3 1 SO3

T (C) T(K) 27 300 50 323 100 373 125 398 127 400 150 423 175 448 185 458 200 473 227 500 241 514 250 523 275 548 284 557 300 573 306 579 327 600 350 623 400 673 427 700 450 723 450 723 500 773 527 800 550 823 627 900

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Na2SO3 176.166 165.128 143.727 132.618 128.175 121.509 114.263 111.365 107.017 99.192 96.483 94.741 89.903 88.161 85.064 83.903 79.839 76.657 69.740 66.005 63.618 63.618 58.428 55.626 53.754 47.486

2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2

NO2 -8.945 -8.617 -7.982 -7.652 -7.517 -7.322 -7.110 -7.026 -6.899 -6.670 -6.592 -6.541 -6.401 -6.351 -6.261 -6.228 -6.110 -6.018 -5.819 -5.712 -5.643 -5.643 -5.494 -5.414 -5.361 -5.182

1 O2 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0

2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2

NaNO3 63.793 59.107 50.035 45.323 43.420 40.611 37.559 36.338 34.506 31.209 29.970 29.173 26.960 26.281 25.075 24.622 23.230 21.979 19.259 17.790 16.865 16.865 14.854 13.768 13.059 10.684

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

SO3 64.573 60.601 52.896 48.897 47.304 44.898 42.283 41.237 39.668 36.844 35.862 35.231 33.478 32.847 31.725 31.304 29.831 28.675 26.161 24.803 23.933 23.933 22.042 21.021 20.327 18.003

logK 33.88 30.92 25.20 22.23 21.00 19.26 17.36 16.60 15.46 13.41 12.50 11.92 10.30 9.95 9.33 9.10 8.67 8.01 6.58 5.80 5.33 5.33 4.31 3.76 3.41 2.25

Ktotal Gibbs J/mol 7.6E+33 -194603 8.3E+30 -191204 1.6E+25 -179968 1.7E+22 -169379 1.0E+21 -160838 1.8E+19 -155944 2.3E+17 -148879 4.0E+16 -145545 2.9E+15 -139998 2.6E+13 -128365 3.2E+12 -123026 8.3E+11 -119337 2.0E+10 -108035 8.9E+09 -106101 2.1E+09 -102367 1.3E+09 -100869 4.7E+08 -99613 1.0E+08 -95559 3.8E+06 -84739 6.3E+05 -77754 2.1E+05 -73805 2.1E+05 -73805 2.0E+04 -63792 5.7E+03 -57572 2.6E+03 -53755 1.8E+02 -38751

Under equilibrium conditions reaction A.24 proceeds to the right at all temperatures calculated. A.5 Summary of all equilibrium calculations from 27 to 627 C.

Summary of equilibrium calculations for the new hypothesized reactions: The Table A.1 is a summary of the Gibbs Free energy for the reactions with respect to temperature

406

Table APPENDIX A-12 Summary spread sheet

T (C) EQ A. 14 EQ A. 15 EQ A. 16 EQ A. 17 EQ A. 18 EQ A. 19 EQ A. 20 EQ A. 21 EQ A. 22 EQ A. 23 EQ A. 24 27 34087 -315881 -24450 -58537 -111565 -265006 -216526 -286928 -133488 -44442 -194603 50 28303 -318203 -30602 -58906 -108118 -259884 -221216 -289896 -138130 -39168 -191204 100 13913 -315247 -44894 -58807 -98065 -243336 -225532 -288900 -143629 -25937 -179968 125 4442 -308056 -53200 -57642 -89784 -228163 -224686 -283471 -145092 -16962 -169379 127 237 -303282 -55634 -55872 -84137 -216373 -224574 -280109 -147873 -12513 -160838 150 -6335 -297308 -62171 -55836 -79638 -209035 -221835 -274927 -145529 -6641 -155944 175 -14291 -287928 -69688 -55397 -73595 -198620 -216927 -266668 -141643 555 -148879 185 -17711 -283334 -72815 -55104 -70843 -193743 -214400 -262599 -139698 3671 -145545 200 -23094 -286344 -77634 -54540 -66354 -185664 -220811 -266478 -147167 8598 -139998 227 -33551 -290759 -86696 -53146 -57185 -168817 -232253 -272706 -161073 18235 -128365 241 -284820 -52997 -52958 -161820 -229763 -268557 -159695 23999 -123026 250 -285034 -52854 -50073 -156985 -232340 -269988 -163076 27855 -119337 275 -284767 -52265 -41371 -142168 -239239 -273373 -172575 39188 -108035 284 -284360 -51983 -37990 -138851 -241629 -274379 -173518 40977 -106101 300 -283232 -51391 -36515 -132501 -240895 -271029 -175043 44247 -102367 306 -282675 -51139 -35925 -129972 -240509 -269611 -175565 45504 -100869 327 -280151 -50128 -33702 -124884 -238684 -263955 -172773 45625 -99613 350 -279196 -49611 -29936 -117815 -240262 -262519 -174639 50137 -95559 400 -274140 -47866 -20459 -99428 -241764 -256453 -177484 60648 -84739 427 -269713 -46569 -14607 -87778 -241477 -251501 -178330 66725 -77754 450 -271969 -49468 -10920 -80688 -242678 -249561 -179793 67495 -73805 450 -271969 -49468 -10920 -80688 -242678 -249561 -179793 67495 -73805 500 -264527 -45174 -1989 -63086 -243811 -243105 -182009 79752 -63792 527 -259096 -42440 3354 -52334 -243581 -238343 -182655 86779 -57572 550 -257532 -41631 6944 -45293 -244507 -236045 -183808 90970 -53755 627 -248631 -38096 20383 -18247 -245134 -224630 -186000 105690 -38751

407

Figure A.1 graphs Gibbs free energy versus temperature for each reaction equation

G ibbs Fre e Ener gy

150000 100000 50000 T(K) 0 Gibbs F ree Energ 0 -50000 -100000 -150000 -200000 -250000 -300000 -350000 Temper atur e 100 200 300 400 500 600 700 EQ A .14 EQ A. 15 EQ A. 16 EQ A.17 EQ A. 18 EQ A. 19 EQ A. 20 EQ A. 21 EQ A. 22 EQ A. 23 EQ A. 24

Figure APPENDIX A-1 Summary of Gibbs free energy versus temperature for all of the reaction equations.

408

Table APPENDIX A-13 Summary of reaction directions Eq No. 14 15 16 17 18 Equation 2NaHCO3 Na2CO3 + H2O + CO2 Na2CO3 + SO2 + O2 Na2SO4 + CO2 2NaHCO3 + SO2 Na2SO3 + 2CO2 + H2O Na2CO3 + SO2 Na2SO3 + CO2 Na2SO3 + 2NO + O2 2NaNO2 + SO3 Na2SO3 + 2NO + 2O2 2NaNO3 + SO3 2NaNO2 + SO2 Na2SO4 + 2NO 2NaNO2 + SO2 + O2 Na2SO4 + 2NO2 2NaNO3 + SO2 Na2SO4 + 2NO2 Na2CO3 + NO2 2NaNO3 + NO + CO2 Na2SO3 + 2NO2 + O2 2NaNO3 + SO3 Comments proceeds to the right at temperature greater than 150C. proceeds to the right at all temperatures calculated proceeds to the right at all temperatures calculated proceeds to the right at all temperatures calculated proceeds to the right for temperatures less than 500C calculated proceeds to the right at all temperatures calculated proceeds to the right at all temperatures calculated proceeds to the right at all temperatures calculated proceeds to the right at all temperatures calculated proceeds to the right for temperatures up to 150C calculated proceeds to the right at all temperatures calculated

19 20 21 22 23

24

409

APPENDIX B CALCULATIONS OF CONVERSION FOR SODIUM BICARBONATE AND SODIUM CARBONATE TRACKING REACTION EXPERIMENTS B.1 Equations for molar percentage of each compound in the product.

For the complex system of reactions involved in the sodium bicarbonate pathway with flue gas conversion definitions were not readily definable because interacting reactions were simultaneously occurring. Conversion was not simply conversion from one compound to another; but instead it involved overlapping conversions. For example as sodium bicarbonate dehydration conversion to sodium carbonate was occurring, SO2 reaction with the freshly formed sodium carbonate was converting the sodium carbonate to sodium sulfite or sulfate. Similarly, sodium sulfite was converting simultaneously to sodium nitrate by reaction with NO and water vapor released by dehydration. Because of the numerous overlapping reactions a different approach to studying conversion was used. To study conversion the composition of the product of the reaction was tracked with time. The molar percentage of each products formed is defined for the flue gas reaction. The possible products found to form for flue gas reactions with sodium bicarbonate are (residual) sodium bicarbonate, sodium carbonate, sodium nitrate, and sodium sulfite and sodium sulfate. The molar percent residual sodium bicarbonate, X NaHCO3 is defined as the molar percent of Na in the form of sodium bicarbonate in the product. By this definition the numerical expression is calculated by the following equation:

410

X NaHCO3 =

mNaHCO3 mNaHCO3 + 2mNa2 CO3 + 2mNa2 SO3/ 4 + m NaNO3

* 100

(B.1)

where: mNaHCO3 is the moles of sodium bicarbonate per 100 grams of product (mol/100g), mNa2 CO3 mNa2 SO3/ 4 is the moles of sodium carbonate per 100 grams of product (mol/100g), is the moles of sodium sulfite or sulfate per 100 grams of product (mol/100g), mNaNO3 is the moles of sodium nitrate per 100 grams of product (mol/100g).

Note for the reaction experiments conducted either sodium sulfite or sodium sulfate was formed (not both). The molar percent sodium carbonate, X Na2 CO3 is defined as the as the molar percent of Na in the form of sodium carbonate in the product: X Na2 CO3 = 2mNa2 CO3 m NaHCO3 + 2m Na2 CO3 + 2m Na2 SO3/ 4 + mNaNO3 * 100 (B.2)

The molar percent sodium sulfite, X Na2 SO3 is defined as the as the molar percent of Na in the form of sodium sulfite or sulfate in the product: X Na2 SO3/ 4 = 2mNa2 SO3/ 4 mNaHCO3 + 2mNa2CO3 + 2mNa2 SO3/ 4 + m NaNO3 * 100 (B.3)

And the molar percent sodium nitrate, X NaNO3 is defined as the as the molar percent of Na in the form of sodium nitrate in the product: X NaNO3 = mNaNO3 mNaHCO3 + 2mNa2 CO3 + 2mNa2 SO3/ 4 + mNaNO3 * 100 (B.4)

411

B.2

Equations for elemental weight percent of hydrogen, nitrogen and sulfur.

In Chapter 5 conversion of sodium bicarbonate by reaction with flue gas was studied with respect to time. Reactions were conducted for incremental time periods, 0.83, 3.3,10,60,and 300 minutes and product from each reaction was collected after quick quench and analyzed for elemental hydrogen, nitrogen, and sulfur analysis. A mathematical method was devised to calculate the molar percent of residual sodium bicarbonate, sodium carbonate, sodium sulfite or sodium sulfate (depending on temperature), and sodium nitrate for each of the products collected. A LECO 932 Elemental Determinator was employed for elemental hydrogen, H, nitrogen, N, and sulfur, S, analysis. Also for products with high weight percents of sulfur, a LECO Total Sulfur Analyzer was used for sulfur analysis. The elemental measurements were obtained on a weight percent basis. Mathematically, weight percent hydrogen, wt%(H), the weight percent nitrogen, wt%(N), and the weight percent sulfur, wt%(S), is expressed by definition respectively: wt%(H) = MW *100 H * fNaHCO 3 (MW * fNaHCO + MW * fNa2CO3 + MW * fNa2SO4 + MW * fNaNO ) NaHCO Na2CO Na2SO NaNO 3 3 3 /4 3 3 (B.5)

wt%(N) =

MW *100 N * fNaNO 3 (MW * fNaHCO + MW * fNa2CO3 + MW * fNa2SO4 + MW * fNaNO ) NaHCO Na2CO Na2SO NaNO 3 3 3 /4 3 3

(B.6)

412

wt%(S) =

MW *100 S * fNa2SO 3/ 4 (MW * fNaHCO + MW * fNa2CO3 + MW * fNaNO ) NaHCO Na2CO Na2SO4 * fNa2SO4 + MW NaNO 3 3 3 3 3

(B.7)

where: MWH is the molecular weight of hydrogen (grams/mole) MWN is the molecular weight of nitrogen (grams/mole) MWS is the molecular weight of sulfur (grams/mole) MWNaHCO3 is the molecular weight of sodium bicarbonate (grams/mole) MWNa2 CO3 is the molecular weight of sodium carbonate (grams/mole)
MW Na2 SO3 is the molecular weight of sodium sulfite (grams/mole) MWNa2 SO4 is the molecular weight of sodium sulfate (grams/mole)

MWNaNO3 f NaHCO3 f Na2 CO3

f Na2 SO3 f Na2 SO4

is the molecular weight of sodium nitrate (grams/mole) is the weight fraction of the product that is sodium bicarbonate is the weight fraction of the product that is sodium carbonate is the weight fraction of the product that is sodium sulfite is the weight fraction of the product that is sodium sulfate is the weight fraction of the product that is sodium nitrate

f NaNO3

Note that for the particular test temperatures used in the experiments either sodium sulfite or sodium sulfate, not both were formed. It is also known that the product is composed only of residual sodium bicarbonate, sodium carbonate, sodium sulfite or sulfate and sodium nitrate. The sum of the weight fractions of all of the compounds in the product make up the whole and is therefore equal to 1. 413

f NaHCO + f Na2CO3 +* f Na2SO/ 4 + f NaNO =1 3 3

(B.8)

B.3

Determining the weight fractions of the respective compounds in the product.

The four weight fractions f NaHCO3 , f Na2CO3 , fNa2SO4 , f NaNO3 , of the compounds in the product are unknown. They were found by solving the set of linear algebraic equations A3.5, A3.6, A3.7, and A3.8. Matrix algebra combined with a spreadsheet was used to manage the large number of computations. Figure A3.1 provides more computational detail. For equations A3.5, A3.6, A3.7, and A3.8 the molecular weights of the elements and compounds are known from handbooks as listed in Table B-1.

414

Table APPENDIX B-1 compounds.

Molecular weights of relevant Molecular Weight 1.01 14.00 32.06 84.00 105.99 126.04 142.04 84.98

Compound or Element Name hydrogen nitrogen sulfur sodium bicarbonate sodium carbonate sodium sulfite sodium sulfate sodium nitrate

415

Figure APPENDIX B-1 Details of numerical computations for weight fractions. The reaction equations A3.5, A3.6, A3.7, and A3.8, are rearranged into linear matrix equation AX=B where A is the four by four matrix of the constant values in parenthesis, and X represents the four unknown weight fractions, and B is the 1 by 4 matrix of constant values from the right hand side of the equations:
$ $ ' wt%(H) $ ' wt%(H) ' wt%(H) $ ' wt%(H) f NaHCO * MWNa2CO3 " + f Na2SO4 * % * MWNa2SO4 " + f NaNO3 * % * MWNaNO3 " = 0 *% * MWNaHCO MWH " + f Na2CO3 * % 3 3 # # & 100 # & 100 & 100 # & 100 $ ' wt%(N) $ $ ' wt%(N) $ ' wt%(N) ' wt%(N) f NaHCO *% * MWNaHCO * MWNa2CO3 " + f Na2SO4 * % * MWNa2SO4 " + f NaNO *% * MWNaNO3 MWN " = 0 + f Na2CO3 * % 3 3 " 3 # # & 100 # & 100 # & 100 & 100 ' wt%(S) $ + f Na2CO3 f NaHCO *% * MWNaHCO 3 3 " & 100 # f NaHCO + f Na2CO3 + f Na2SO4 + f NaNO3 = 1 3 ' wt%(S ) $ *% * MWNa2CO3 " + f Na2SO4 & 100 # ' wt%(S) $ *% * MWNa2SO4 MWS " + f NaNO3 & 100 # ' wt%(S) $ *% * MWNaNO3 " = 0 & 100 #

From this matrix arrangement Cramers rule can be used in a spreadsheet computer program to solve the following numerical solutions: f NaHCO3 = det( A1 ) det( A2 ) det( A3 ) det( A4 ) , f Na2 SO4 = , f Na2CO3 = , f NaNO3 = det( A) det( A) det( A) det( A)

where Aj is the matrix obtained by replacing the entries in the jth column of A by the entries in the matrix B which are zero.

416

B.4

Calculating the moles of each compound in the product.

From the weight fractions, f NaHCO3 , f Na2CO3 , fNa2SO4 , f NaNO3 , the molar fractions are obtained for a 100gram sample. For the 100 gram sample, the weight of each compound in the product is calculated by multiplying the respective weight fractions times 100 grams. The moles are then found by multiplying by each of these respective weights by their respective molecular weights:
mNaHCO3 = 100 * f NaHCO3 MWNaHCO3

(B.9)

mNa2 CO3 =

100 * f Na2 CO3 MWNa2 CO3 100 * f Na2 SO4 MW Na2 SO4 100 * f NaNO3 MWNaNO3

(B.10)

m Na2 SO4 =

(B.11)

mNaNO3 =

(B.12)

417

APPENDIX C UNITED STATES PATENT OF NITROGEN DIOXIDE MITIGATION PROCESS

United States Patent Keener , et al.

6,214,308

April 10, 2001

Process for removal of nitric oxides and sulfur oxides from flue gases

Abstract

A two stage method is described for the removal of NO.sub.x and SO.sub.x from flue gas. The first stage removes all NO.sub.x by sodium sulfite sorbent injection. The second stage removes SO.sub.x by sodium bicarbonate injection. The sodium sulfite product formed in the second stage is transferred to the first stage for injection.

Inventors Keener; Timothy C. (Cincinnati, OH); Khang; Soon-Jai (Cincinnati, OH); Stein; : Assignee
The University of Cincinnati (Cincinnati, OH) Antoinette Weil (Menlo Park, CA)

: Appl.
208353

418

No.: Filed:
December 9, 1998

Current U.S. Class: Intern'l Class: Field of Search:

423/239.1; 423/244.07; 423/244.08

B01D 053/50; B01D 053/56 423/235,239.1,243.08,244.07,244.08

References Cited

U.S. Patent Documents

3305307 3977844 4140651 4197278 4425313 4645652 4681045 4839147 5059406 5165902

Feb., 1967 Aug., 1976 Feb., 1979 Apr., 1980 Jan., 1984 Feb., 1987 Jul., 1987 Jun., 1989 Oct., 1991 Nov., 1992

Spormann et al. Van Slyke Burnell et al. Gehri et al. Cooper Kimura Dvirka et al. Lindbauer et al. Sheth et al. Bortz et al.

423/243. 48/202. 252/188. 423/242. 423/235. 423/235. 110/345. 423/235. 423/244. 423/235.

419

Other References

Chu et al. "Removal of SO.sub.2 and NO.sub.x from Stack Gas . . . " JAPCA vol. 39 No. 2 pp. 175-179, Feb. 1989.

Primary Examiner: Griffin; Steven P. Assistant Examiner: Vandy; Timothy C Attorney, Agent or Firm: Flehr Hohbach Test Albrittion & Herbert LLP

Claims

What is claimed is:

1. The process for removing nitric oxides and sulfur oxides from flue gases comprising the steps of injecting sodium sulfite particles and water vapor into the flue gas stream which reacts with the nitric oxides in the flue gases to form sodium nitrate, removing the sodium nitrate and unreacted sodium sulfite from the flue gases, and thereafter injecting sodium bicarbonate particles into the flue gas stream to react with the sulfur dioxide in the gas stream and form sodium sulfite particles collecting the sodium sulfite particles and injecting the collected sodium 420

sulfite particles into the flue gas stream to react with said nitric oxides.

2. The process of claim 1 in which the temperature of the flue gases which react with the sodium sulfite particles and water vapor is maintained between 100.degree. C. and 150.degree. C.

3. The process of claim 2 in which the temperature is about 110.degree. C.

4. The process of claim 1, 2 or 3 in which the temperature of the flue gases which react with the sodium bicarbonate particles is between 100.degree. C. and 150.degree. C.

5. The process of claim 1, 2 or 3 in which the particle size of the sodium sulfite particles is less than 120 micrometers in diameter.

6. The process of claim 5 in which the particle size of the sodium bicarbonate particles is less than 120 micrometers in diameter.

7. The process of claim 1, 2 or 3 in which the stoichiometric ratio of sodium sulfite with respect to nitric oxide is greater than one.

8. The process of claim 1, 2 or 3 in which the stoichiometric ratio of sodium bicarbonate with respect to sulfur oxide is greater than one. 421

Description

BRIEF DESCRIPTION OF THE INVENTION

A process is described for the removal of nitric oxides and sulfur oxides from flue gases. More particularly, a two-stage process is described for the removal of nitric oxides and sulfur oxides from flue gases.

BACKGROUND OF THE INVENTION

To conform with government regulations, primarily in the Acid Rain section, Title IV, of the Clean Air Act, energy providers that burn sulfur containing fuels are increasingly responsible for assuring that exit flue gas emissions meet low levels of SO.sub.x and NO.sub.x. With these stipulated regulations, new methods of flue gas scrubbing for SO.sub.x and NO.sub.x are increasingly important for energy providers.

SO.sub.x in combustion flue gases are created due to sulfur oxidation by the combustion processes where the sulfur originates in the coal or oil fossil fuel. Likewise NO.sub.x is formed 422

partially by the oxidation of nitrogen in the fuel during combustion but also by the oxidation of nitrogen in the air mixture at high temperature and dwell time during combustion.

Many processes have been proposed and used for the removal of SO.sub.x from the flue gases including dry injection, slurry injection (spray dryer), and wet scrubbing. The injection of dry alkaline sorbents such as lime and sodium based salts has been recognized as an available technology for control of SO.sub.2. Flue gas desulfurization by the injection of dry sodium-based reagents has been investigated in the laboratory since the 1960s. Pilot plant and full scale testing has successfully been demonstrated. These processes are of interest because they require low capital investment. The use of these materials has gained popularity in regions such as the western United States where the materials are geologically available relative to other desulfurization reagents.

In the early 1990s it was discovered that sodium sorbents used for desulfurization in SO.sub.2 flue gas which contained NO resulted in an unwanted brown discoloration of the exit plume. This unwanted side effect jeopardized the feasibility of the process. A number of processes have been proposed to eliminate the formation of the brown plume including the use of ammonia and urea injection. These processes produce undesirable side effects of their own, namely ammonia, urea and greenhouse gases.

Many processes have been proposed to reduce NO.sub.x emissions from flue gas, namely low 423

NO.sub.x burners, combustion flame temperature reduction, flue gas recirculation, selective catalytic reduction, selective non-catalytic reduction with and without ammonia injection. These processes, however, require intensive capital investment.

A handful of processes have also been proposed to remove the combination of both SO.sub.x and NO.sub.x. Sodium based sorbents have been suggested for NO.sub.x mitigation as additives with calcium based reagents, such as in the Niro process where sodium hydroxide or sodium sulfite solutions are used as additives in a wet slurry process. It has been reported that NO.sub.2 and NO decomposed to N.sub.2 by a wet sulfite solution. In a gas liquid system, absorption of SO.sub.2 to form sulfites or bisulfites participate in reactions with NO.sub.x.

Prior art processes using sodium bicarbonate sorbent for combined SO.sub.x and NO.sub.x removal provide relatively low removal rates of both SO.sub.x and NO.sub.x, and low sorbent utilization.

OBJECTS AND SUMMARY OF THE INVENTION

It is a general object of the present invention to provide a method and apparatus for removal of nitric oxides and sulfur oxides from flue gases which avoids the shortcomings of the prior art.

It is a further object of the present invention to provide a simple, efficient and economical 424

process for the removal of nitric oxides and sulfur oxides from flue gases.

It is a further object of the present invention to provide a process which removes nitric oxides from flue cases followed by the removal of sulfur oxides.

There is provided a process in which nitric oxides are removed from flue gases in a first stage by treating the flue gases with unoxidized sodium sulfite under conditions which minimize the reaction of sodium nitrite or nitrate with the sulfur oxides in the flue gases to minimize the formation of sodium sulfate and then removing sulfur oxide in a second stage by reacting the sulfur oxides with sodium bicarbonate.

BRIEF DESCRIPTION OF THE DRAWINGS

Additional objects and advantages of the present invention will become apparent to those skilled in the art by reference to the following detailed description thereof and the drawing which schematically shows the steps of the process.

DETAILED DESCRIPTION OF THE INVENTION

425

In our effort to provide a simple process for the efficient and economical removal of nitric oxides and sulfur oxides from flue gases, we considered the reaction of nitric oxide with sodium sulfite and sodium sulfate, and the reaction of sulfur oxide with sodium bicarbonate in the presence of nitric oxides.

First we considered the reaction path for the formation of NO.sub.2 during desulfurization of flue gas containing SO.sub.2 and NO. Prior to this invention the pathway for NO.sub.2 formation was not well understood.

1) We found that sodium bicarbonate first reacted with SO.sub.2 in the SO.sub.2 flue gases containing NO to initially form a temperature dependent mixture of sodium sulfite and sodium sulfate.

2) For temperatures where sodium sulfite was formed (below 250.degree. C.) we found a second reaction step. The sodium sulfite proceeded to react with NO and O.sub.2 or NO.sub.2 to form a temperature dependent mixture of sodium nitrite and sodium nitrate.

3) The third reaction step we found was that sodium nitrite and sodium nitrate reacted with SO.sub.2 and O.sub.2 to form sodium sulfate and a temperature dependent mixture of NO and NO.sub.2.

426

4) The accumulation of sodium nitrite or nitrate was found to occur only when water vapor was available to react with sodium nitrite and fix itself on the surface of sodium nitrate. The removal of SO.sub.2 and NO.sub.x was limited by the formation of a sodium sulfate ash layer in the final step in NO.sub.2 formation. From this we realized that since SO.sub.2 and NO.sub.x removal is significantly limited by the formation of sodium sulfate, and that the formation of sodium sulfate should be avoided. In particular, our novel process avoids sodium sulfate formation during SO.sub.2 removal. This is achieved in two ways: first, sodium sulfate formation is avoided by taking out all NO.sub.x from the gas stream prior to the reaction of SO.sub.2 with sodium bicarbonate. Without NO.sub.x in the gas stream, sodium sulfite is formed by the reaction of sodium bicarbonate with SO.sub.2. Since there is no NO.sub.x present, the formation of sodium nitrite and nitrate are prevented, which prevents the further reaction of the sodium nitrite and nitrate with SO.sub.2 which would form sodium sulfate and NO or NO.sub.2. Secondly, sodium sulfate formation is avoided by maintaining the temperature of the reaction of sodium bicarbonate with SO.sub.2 flue gas below 150.degree. C. to assure no oxidation of sodium sulfite by O.sub.2.

From the above observations we arrived at the two stage process of the present invention. The first stage is used to remove all NO.sub.x from the gas stream and the second stage is used for SO.sub.2 removal. The reactions and the reaction temperatures described and defined for the removal of SO.sub.2 by sodium bicarbonate have been well established. What makes our process unique is the staging technique which removes NO.sub.x from the flue gas prior to its contact 427

and reaction with sodium bicarbonate during the SO.sub.2 removal.

NO.sub.x removal in the first stage is achieved by reacting fresh unoxidized sodium sulfite powder with NO.sub.x in the flue gas. Sodium sulfate formation is avoided in the first stage to provide high NO.sub.x removal and high sorbent utilization. Two conditions are applied to the first stage to avoid sodium sulfate formation. First the temperature in the first stage is maintained below 150.degree. C., preferably around 110.degree. C., to minimize the reaction of sodium nitrite or nitrate with the SO.sub.2 in the flue gas. Based on reaction tests of sodium nitrate with SO.sub.2 it was found that lower temperatures result in less sodium sulfate formation. Secondly, water vapor or an additive such as sodium hydroxide, which provides water vapor, may be added to promote oxidation of sodium nitrite to sodium nitrate and to prevent the sodium nitrate from reacting with the SO.sub.2.

FIG. 1 is a schematic diagram of apparatus suitable for carrying out the process of the present invention. The flue gas from combustion is shown applied to the conduit 11 which directs the flue gas into the first stage apparatus. In the present example, a cyclone collector 12, although other particulate control devices would be acceptable. The flue gas contains SO.sub.2, NO.sub.x, CO.sub.2, O.sub.2 and N.sub.2. Sodium sulfite, NO.sub.2 SO.sub.3 is also continuously injected into the flue gas stream by a suitable injector 13. In the preferred embodiment, water vapor or sodium hydroxide, NaOH, is continuously injected into the flue gas stream by a suitable injector 14. For existing systems with fabric filters the first stage collector 12 is placed before the fabric 428

filter. The flue gas temperature at the first stage 12 should be less than 150.degree. C., preferably about 110.degree. C. This may require the addition of a heat extractor (not shown) prior to this first stage 12. The temperature is adjusted below 150.degree. C. to obtain more NO.sub.x removal per unit of sorbent injection. Depending upon the flue gas volumetric flow rate a stoichiometric ratio greater than one is needed for sodium sulfite sorbent injection into the first stage. The sorbent preferably has a particle size less than 120 micrometers in diameter; the particle size for best NO.sub.x removal. The sodium sulfite sorbent must be of a fresh grade and must not have any surface oxidation of sodium sulfate from storage life or moist air exposure. The sodium sulfite sorbent injected into stage one may be recycled through stage one as long as it does not have any surface oxidation, sodium sulfate, formed from storage or any other conditions. The sodium sulfite sorbent injected into stage one may be the product formed at stage two, as will be presently described, as long as no sodium sulfate has formed in the stage two reaction or from storage or handling during transfer from stage two to stage one. The cyclone collector 12, which collects the stage one particulates, could be replaced by a fabric filter baghouse, an ESP, or any other collector. The collector choice is based on the gas solid contact needed to provide adequate NO.sub.x, removal and to avoid NO.sub.x flow to the second stage. The unreacted sodium sulfite, Na.sub.2 SO.sub.3 and the sodium nitrite, NaNO.sub.3 formed in the reaction of the nitric oxides with the sodium sulfite, are collected as illustrated at 17.

In the process it is essential that substantially all flue gas NO.sub.x is removed in the first stage such that the second stage can remove high amounts of SO.sub.2. If all of the NO.sub.x is not 429

removed in the first stage the second stage will not obtain high SO.sub.2 removals since sodium sulfate product will form in the second stage and cause significant ash layer resistance and inhibit high rates of SO.sub.2 removal. Also, if substantially all of the NO.sub.x is not removed in the first stage it is likely that NO.sub.2 will form at the exit of the second stage. To obtain full removal of flue gas NO.sub.x by the first stage, plenty of sodium sulfite must be injected into the first stage. The sodium sulfite should be of a fresh nature, whereby there is no oxide, sodium sulfate, scale formed on its outer surface prohibiting it from reacting with flue gas NO.sub.x. To assure that the sodium sulfite is free of the inhibiting sodium sulfate scale, the sodium sulfite should be processed fresh without shelf life or exposure to moist air. Sodium sulfite must be stored in sealed dry air free containers.

The remaining flue gases, SO.sub.2, CO.sub.2, O.sub.2, and N.sub.2, from stage one flow into the baghouse 18. Sodium bicarbonate, NaHCO.sub.3 powder is continuously injected into the remaining fuel gas stream by a suitable injector 19.

Alternate sorbent injection may be used to remove NO.sub.x in the first stage such as sodium bisulfite, sodium pyrosulfite, sodium sulfite heptahydrate or sodium bicarbonate or nahcolite (NaHCO.sub.3), or sodium sesquicarbonate or trona (Na.sub.2 CO.sub.3. NaHCO.sub.3.2H.sub.2 O). For these alternate sorbents their time of flight and total dwell time allowed in the flue gas duct must be limited to their respective dehydration or water loss periods to assure that NO.sub.x is being removed by these materials and that NO.sub.2 is not being 430

formed. Additionally, water vapor or other water containing materials may be injected with the sorbent to enhance NO.sub.x removal and assure no NO.sub.2 formation. The second stage temperature is between 100.degree. C. and 150.degree. C. so that the injected sodium bicarbonate sorbent dehydrates simultaneously with desulfurization and such that maximum SO.sub.2 removal is obtained. The sodium bicarbonate sorbent is preferably less than 120 micrometer in diameter for superior SO.sub.2 removal. A stoichiometric ratio greater than one is used for this sorbent in the SO.sub.2 flue gas. The sorbent is injected at a distance from the particulate collection device 18 to increase the reaction time and maximize SO.sub.2 removal. The sodium bicarbonate reacts with the SO.sub.2 to form sodium sulfite, as shown at 21, which can be reinjected into the first stage as described above. The clean gases CO.sub.2, O.sub.2 and N.sub.2 flow through conduit 22 into the atmosphere.

There has been provided a simple, efficient and economical process which removes nitric oxides from flue gases followed by removal of sulfur oxides to exhaust clean gases into the atmosphere from the combustion of sulfur containing fuels.

****

431