University of Central Florida STARS

Electronic Theses and Dissertations, 2004-2019

2006

In-situ Ammonia Removal Of Leachate From Bioreactor Landfills

Nicole Berge University of Central Florida

Part of the Environmental Engineering Commons Find similar works at: https://stars.library.ucf.edu/etd University of Central Florida Libraries http://library.ucf.edu

This Doctoral Dissertation (Open Access) is brought to you for free and open access by STARS. It has been accepted for inclusion in Electronic Theses and Dissertations, 2004-2019 by an authorized administrator of STARS. For more information, please contact [email protected].

STARS Citation Berge, Nicole, "In-situ Ammonia Removal Of Leachate From Bioreactor Landfills" (2006). Electronic Theses and Dissertations, 2004-2019. 776. https://stars.library.ucf.edu/etd/776

IN-SITU AMMONIA REMOVAL OF LEACHATE FROM BIOREACTOR LANDFILLS

by

NICOLE D. BERGE B.S. University of South Carolina, 1999 M.S. University of South Carolina, 2001

A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the Department of Civil and Environmental Engineering in the College of Engineering and Computer Science at the University of Central Florida Orlando, Florida

Spring Term 2006

Major Professor: Debra R. Reinhart

ABSTRACT

A new and promising trend in solid waste management is to operate the landfill as a bioreactor. Bioreactor landfills are controlled systems in which moisture addition and/or air injection are used as enhancements to create a solid waste environment capable of actively degrading the biodegradable organic fraction of the waste. Although there are many advantages associated with bioreactor landfills, some challenges remain. One such challenge is the ammonia-nitrogen concentration found in the leachate. The concentrations of ammonia-nitrogen tend to increase beyond concentrations found in leachate from conventional landfills because recirculating leachate increases the rate of ammonification and results in accumulation of higher levels of ammonia-nitrogen concentrations, even after the organic fraction of the waste is stabilized. Because ammonia-nitrogen persists even after the organic fraction of the waste is stabilized, and because of its toxic nature, it is likely that ammonia-nitrogen will determine when the landfill is biologically stable and when post-closure monitoring may end. Thus an understanding of the fate of nitrogen in bioreactor landfills is critical to a successful and economic operation.

Ammonia-nitrogen is typically removed from leachate outside of the landfill.

However, additional costs are associated with ex-situ treatment of ammonia, as separate treatment units on site must be maintained or the leachate must be pumped to a publicly owned wastewater treatment facility. Therefore, the development of an in-situ nitrogen removal technique would be an attractive alternative. Several recent in-situ treatment approaches have been explored, but lacked the information necessary for field-scale

ii implementation. The objectives of this study were to develop information necessary to

implement in-situ ammonia removal at the field-scale.

Research was conducted to evaluate the kinetics of in-situ ammonia removal and

to subsequently develop guidance for field-scale implementation. An aerobic reactor and

microcosms containing digested municipal solid waste were operated and parameters

were measured to determine nitrification kinetics under conditions likely found in

bioreactor landfills. The environmental conditions evaluated include: ammonia

concentration (500 and 1000mg N/L), temperature (25o, 35o and 45oC), and oxygen concentration in the gas-phase (5, 17 and 100%).

Results suggest that in-situ nitrification is feasible and that the potential for simultaneous nitrification and denitrification in field-scale bioreactor landfills is significant due to the presence of both aerobic and anoxic areas. All rate data were fitted to the Monod equation, resulting in an equation that describes the impact of pH, oxygen concentration, ammonia concentration, and temperature on ammonia removal. In order to provide design information for a field-scale study, a simple mass balance model was constructed in FORTRAN to forecast the fate of ammonia injected into a nitrifying portion of a landfill. Based on model results, an economic analysis of the in-situ treatment method was conducted and compared to current ex-situ leachate treatment costs. In-situ nitrification is a cost effective method for removing ammonia-nitrogen when employed in older waste environments. Compared to reported on-site treatment costs, the costs associated with the in-situ ammonia removal process fall within and are on the lower end of the range found in the literature. When compared to treating the

iii leachate off-site, the costs of the in-situ ammonia removal process are always significantly lower. Validation of the laboratory results with a field-scale study is needed.

iv ACKNOWLEDGMENTS

First and foremost, I would like to begin by thanking my advisor, Dr. Debra

Reinhart, without whom this work would not have been possible. Dr. Reinhart has played many roles during my time here. She has been an extraordinary role model, mentor, friend and critic, exerting an influential impact on my future and helping to shape me both as an aspiring researcher and teacher. She has taught me how to be strong, positive, and have confidence, as she always had immense confidence in me. I have also learned from her the role teaching and research have in life and how much passion for both makes all the difference. I also thank her for providing me with numerous professional development opportunities, such as teaching and attending many conferences. Her willingness to reply to the countless daily emails I sent and for always being available when I needed to blow off steam have been invaluable. One truly could not ask for a better mentor and I feel blessed to have had the opportunity to get to know and work with her. I truly believe that I am leaving UCF not only with a better understanding of theories and as a better researcher, but also as a better person because of all I learned from her.

I would also like to thank Dr. John Dietz for all of his help during the past few years. Thank you, Dr. Dietz, for being so helpful and gracious with your time. I have learned so much from you. Dr. Dietz always dropped by my office to help with my, at times quite numerous, problems or to give advice as to how to interpret results. He taught me how to be critical of all results and how to truly understand the inherent limitations associated with them. His help and expertise were invaluable to this work and I really cannot express the immense gratitude I have for all he has done for me. And yes, Dr.

Dietz, I did finish before you retired! I just made it…barely!

v I also would like to thank my other committee members. Dr. Tim Townsend has offered invaluable guidance to this work. His comments and critiques have made this project better and I greatly appreciate all of his time and support. Thanks also go to Dr.

Andrew Randall for agreeing to serve on my committee and for always being willing to help me sort through microbiological issues. I also thank Dr. Ni-Bin Chang, who enthusiastically joined my committee late in the process and was so willing to help and positively contribute to this work. Thank you all so much.

I also acknowledge the Florida Center for Solid and Hazardous Waste and the

Environmental Research and Education foundation for funding this work. This work would not have possible without your contributions.

Thanks also go to Jackie Ceather, who helped me tremendously with some of the laboratory work. Her help and time were greatly appreciated. I would not have slept during the past five years if it weren’t for her. Thanks also go to my fellow research group colleagues, particularly those who spent so much time in the lab with me (Eyad,

Lucia and Andrea).

I also would like to extend thanks to my dear friends Sara and Bryan Stone. I couldn’t have made it without all of your support. You guys helped to keep me somewhat sane through this process and always reminded me why I was here.

Finally, I thank my family for their enduring love and support. My Mom always taught me that if I worked hard and believed in myself, I could accomplish anything. And she was right. Thanks for always being supportive. I also thank my Step-Dad for all of his support and wisdom throughout the years. And, of course, I thank my little brothers,

Weber and Mattison, whom I cherish more than anything. I am finally graduating, guys!

vi TABLE OF CONTENTS

LIST OF TABLES ...... xi LIST OF FIGURES...... xii CHAPTER 1 INTRODUCTION ...... 1 Background Information...... 1 Research Objectives and Scope of Work...... 4 Dissertation Organization ...... 6 References...... 7 CHAPTER 2 THE FATE OF NITROGEN IN BIOREACTOR LANDFILLS ...... 11 Introduction...... 11 Bioreactor Landfill Operation...... 13 Anaerobic Bioreactor Landfills...... 17 Aerobic Bioreactor Landfills ...... 19 Hybrid Bioreactor Landfills...... 21 Facultative Bioreactor Landfills ...... 24 Ammonia-Nitrogen in Leachate ...... 25 Nitrogen Transformation and Removal Processes ...... 27 Ammonification ...... 30 Ammonium Flushing ...... 31 Ammonium Sorption ...... 32 Volatilization...... 36 Nitrification...... 38 Nitrification Case Studies in Landfills...... 43 Nitrification Kinetics ...... 45 Denitrification...... 46 Heterotrophic Denitrification...... 47 Autotrophic Denitrification...... 49 Denitrification Kinetics...... 51 ANAMMOX...... 51 Dissimilatory Nitrate Reduction to Ammonium...... 52 Simultaneous Nitrogen Removal Processes...... 55 Other Nitrate Processes...... 56 Future Research Directions ...... 57 References...... 58 CHAPTER 3 THE IMPACT OF WASTE ACCLIMATION ON IN-SITU AMMONIA REMOVAL FROM BIOREACTOR LANDFILL LEACHATE ...... 70 Introduction...... 70 Materials and Methods...... 73

vii Aerobic Reactor (Waste Acclimation Process) Design and Operation ...... 73 In-Situ Nitrification Microcosm Studies...... 75 Analytical Techniques ...... 77 Results and Discussion...... 79 Aerobic Reactor (Waste Acclimation Process)...... 79 In-Situ Nitrification Microcosm Studies...... 83 Ammonia Sorption...... 84 In-Situ Nitrification Processes ...... 85 Simultaneous Nitrogen Removal Processes...... 86 Ammonia Removal Kinetics...... 90 Acclimated vs Unacclimated Processes...... 93 Field-Scale Ammonia Removal Implications...... 94 Nitrogen Mass Balances ...... 95 Conclusions...... 96 References...... 97 CHAPTER 4 THE IMPACT OF GAS-PHASE OXYGEN AND TEMPERATURE ON IN-SITU AMMONIA REMOVAL FROM BIOREACTOR LANDFILL LEACHATE ...... 101 Introduction...... 101 Experimental Materials and Methods ...... 103 Microcosm Study Operation...... 104 Analytical Techniques ...... 106 Results and Discussion...... 106 Nitrogen Removal Processes ...... 106 Ammonia-Nitrogen Removal Rates...... 113 Impact of oxygen and temperature on removal rates...... 113 Combined effects on ammonia removal ...... 122 References...... 125 CHAPTER 5 ENGINEERING SIGNIFICANCE...... 128 Introduction...... 128 Model Description...... 128 Simulation of Laboratory Experiments...... 135 Model Results ...... 138 Optimizing Treatment Depth...... 143 In-Situ Denitrification ...... 145 Field Study Plans ...... 146 Economic Analysis ...... 149 In-Situ Treatment Economics ...... 149 Cost Comparison With Other Leachate Treatment Systems ...... 153

viii References...... 156 CHAPTER 6 CONCLUSIONS AND RECOMMENDATIONS ...... 159 Conclusions...... 159 Field-Scale Implementation Recommendations...... 161 APPENDIX A MATERIALS AND METHODS...... 163 Introduction...... 164 Waste Acclimation Process ...... 164 Microcosm Studies...... 169 Failed Microcosm Designs ...... 170 Chosen Microcosm Design...... 171 Microcosm Sampling Procedure...... 172 Experiments Conducted...... 173 Impact of Waste Acclimation Experiments ...... 173 Impact of Gas-Phase Oxygen and Temperature Experiments ...... 174 Analytical Techniques ...... 177 pH...... 177 Ammonia...... 177 Anions (nitrate, nitrite, sulfate)...... 179 Alkalinity ...... 180 Gas-Phase Oxygen and Nitrogen...... 180 Gas-Phase Nitrous Oxide...... 181 Chemical Oxygen Demand (COD)...... 181 Solid-Phase Organic Nitrogen ...... 182 Volatile Solids...... 183 Moisture Content ...... 183 Gas Volume ...... 183 Ammonia Gas ...... 184 References...... 184 APPENDIX B WASTE ACCLIMATION PROCESS RESULTS...... 185 Introduction...... 186 Summary of Results...... 186 Waste Acclimation Process #1...... 186 Waste Acclimation Operation at 22oC...... 186 Waste Acclimation Operation at 35oC...... 189 Waste Acclimation Process #2...... 192 Waste Acclimation Operation at 45oC...... 193 References...... 198 APPENDIX C MODEL VALIDATION STATISTICS ...... 199 Introduction...... 200 Procedure...... 200

ix Discussion of Results...... 204 Data Sheets ...... 206 References...... 246 APPENDIX D CONFIDENCE INTERVAL DETERMINATION ...... 247 Introduction...... 248 Summary of Results...... 248 Case #1: Comparison Between Oxygen Levels at Constant Temperature Using the Monod Model Without Temperature and pH Terms ...... 249 Case #2: Comparison Between Oxygen Levels at Constant Temperature Using the Complete Monod Model (including temperature and pH terms)...... 249 Case #3: Comparison Between Temperatures at Different Oxygen Levels Using the Complete Monod Model (including temperature and pH terms)...... 250 Discussion of Results...... 250 Data Sheets ...... 252 Comparison Between Oxygen Levels at Constant Temperature Using the Monod Model Without Temperature and pH Terms...... 252 Comparison Between Oxygen Levels at Constant Temperature Using the Complete Monod Model (including temperature and pH terms) ...... 256 Comparison Between Temperatures at Different Oxygen Levels Using the Complete Monod Model (including temperature and pH terms) ...... 260 APPENDIX E MODEL CODE ...... 264 APPENDIX F PLOTS OF ALL MICROCSM DATA...... 273 Microcosms at 22oC ...... 274 Acclimated Waste ...... 274 Microcosms at 22oC ...... 290 Unacclimated Waste ...... 290 Microcosms at 35oC ...... 293 Microcosms at 45oC ...... 309 APPENDIX G ECONOMIC CALCULATIONS ...... 322 Air Flowrate Calculations ...... 323 Blower Power Requirement Calculations...... 326 Air Piping Calculations ...... 328 Leachate Injection Pump Calculations ...... 329

x LIST OF TABLES

Table 2-1. Ammonia-Nitrogen Concentrations in Both Conventional Bioreactor Landfills with Respect to Degree of Landfill Biological Stabilization (Reinhart and Townsend, 1998)...... 25 Table 3-1. Ammonia Removal Efficiencies...... 83 Table 3-2. Microcosm Gaseous Headspace Analysis...... 89 Table 4-1. Model constants for different Monod fits...... 118 Table 4-2. Dissolved oxygen concentrations (mg/L) at different temperatures and gas- phase oxygen concentrations...... 121 Table 5-1. Parameters Varied in Model Simulations...... 139 Table 5-2. Parameters Used in Model Scenarios...... 144 Table 5-3. Equipment Necessary for In-Situ Nitrogen Removal...... 152 Table 5-4. Operation Costs Associated With In-Situ Nitrogen Removal...... 152 Table 5-5. Cost Estimate of In-Situ Ammonia Removal Process...... 153 Table 5-6. Leachate Treatment Costs...... 155 Table A- 1. Typical Waste Composition...... 167 Table A- 2. List of Experiments Conducted to Evaluate the Impact of Waste Acclimation on Ammonia Removal...... 174 Table A- 3. List of Experiments Conducted to Evaluate the Impact of Gas-Phase Oxygen and Temperatures on In-Situ Nitrification...... 176 Table C- 1. Parameter Values from the Complete Multiplicative Monod Model...... 201 Table C- 2. Sigma Plot Iteration Results...... 203 Table C- 3. Parameters of the Chosen Multiplicative Monod Model...... 204 Table C- 4. Parameters of the Chosen Multiplicative Monod Model Without Oxygen Inhibition...... 205

xi LIST OF FIGURES

Figure 2-1. Dominant form of ammoniacal nitrogen in solution at 25oC at various pH levels...... 27 Figure 2-2. The potential pathways of nitrogen transformation and/or removal in bioreactor landfills...... 29 Figure 2-3. The deamination process...... 31 Figure 3-1. Waste acclimation process results: (a) Ammonia concentrations over time, (b) nitrate-nitrogen concentrations, (c) sulfate concentrations over time and (d) ammonia, nitrate, and nitrite concentrations over time from leachate from the aerobic reactor...... 81 Figure 3-2. Total ammonia-nitrogen masses recovered from microcosms over time...... 84 Figure 3-3. Ammonia-, nitrate-, and nitrite-nitrogen masses over time in all studies: (a) 200 mg N/L, (b) 500 mg N/L – Acclimated, (c) 500 mg N/L – Unacclimated, and (d) 1000 mg N/L...... 88 Figure 3-4. Ammonia removal rates for all studies: (a) 1000 mg N/L, (b) All acclimated and unacclimated data fit to Monod...... 92 Figure 4-1. Ammonia removal masses in studies at different oxygen levels at the three different temperatures: (a) 22C, (b) 35C and (c) 45C...... 110 Figure 4-2.Gaseous by-products produced in terms of the maximum mass of nitrogen converted to either nitrogen or nitrous oxide: (a) maximum mass of nitrogen initially present converted to nitrogen gas at each temperature and oxygen level. (b) maximum mass of nitrogen initially present converted to nitrous oxide at each temperature level...... 111 Figure 4-3.Ammonia removal rates and Monod equation fits for a few of the microcosm experiments: (a) Raw removal rate data and Monod fits at 22oC, (b) Raw removal rate data from tests with approximately 17% oxygen in the gas-phase at three different temperature levels, and (c) Raw removal rate data and the rate equation fit with approximately 5% oxygen in the gas-phase and at 45oC...... 115 Figure 5-1. Flow simulation module in the model...... 131 Figure 5-2. Comparison of measured and predicted ammonia concentration trends in the microcosm experiment conducted at 22oC and 5% oxygen...... 136 Figure 5-3. Comparison of measured and predicted ammonia concentration trends in the microcosm experiment conducted at 35oC and 5% oxygen...... 137 Figure 5-4. Comparison of measured and predicted ammonia concentration trends in the microcosm experiment conducted at 45oC and 17% oxygen...... 137 Figure 5-5. The impact of leachate injection rates on the required landfill depth for various gas-phase oxygen concentrations...... 140 Figure 5-6. The impact of temperature on the required landfill depth for various gas- phase oxygen concentrations...... 141 Figure 5-7. The impact of pH on the required landfill depth for various gas-phase oxygen concentrations (Note that the 7 and 7.5 trends do not differ)...... 141 Figure 5-8. The impact of hydraulic conductivity on the required landfill depth for various gas-phase oxygen concentrations...... 142 Figure 5-9. The impact of dry specific weight of waste on the required landfill depth for various gas-phase oxygen concentrations...... 142

xii Figure 5-10. Cross-section view of planned field-scale study...... 148 Figure 5-11. Plan view of planned field-scale study...... 148 Figure 5-12. Lysimeter detail (www.soilmoisture.com)...... 149 Figure A- 1. Schematic depicting the relationship between the waste acclimation process and microcosm studies...... 165 Figure A- 2. Waste Acclimation Process #1...... 165 Figure A- 3. The Second Waste Acclimation Process: (a) Reactor and (b) water bath. 166 Figure B- 1. Ammonia- and nitrate-nitrogen concentration trends over time in the waste acclimation reactor #1: 22oC...... 187 Figure B- 2. Sulfate concentration over time in the waste acclimation reactor #1: 22oC...... 188 Figure B- 3. pH, alkalinity, and ammonia-nitrogen trends over time in waste acclimation reactor #1 at 22oC...... 188 Figure B- 4. COD trend over time in waste acclimation reactor #1 at 22oC...... 189 Figure B- 5. Ammonia- and nitrate-nitrogen concentrations over time in waste acclimation reactor #1: 35oC...... 191 Figure B- 6. Nitrate and sulfate concentrations over time in waste acclimation reactor #2: 35oC...... 191 Figure B- 7. pH, alkalinity and ammonia-nitrogen trends over time in waste acclimation reactor #2: 35oC...... 192 Figure B- 8. COD trend over time in waste acclimation reactor #2: 35oC...... 192 Figure B- 9. Ammonia- and nitrate-nitrogen concentrations over time in waste ...... 196 Figure B- 10. pH, alkalinity, and ammonia-nitrogen trends over time in waste acclimation reactor #2: 45oC...... 196 Figure B- 11. Sulfate and nitrate-nitrogen trends over time in waste acclimation reactor #2: 45oC...... 197 Figure B- 12. COD trend over time in waste acclimation reactor #2: 45oC...... 197 Figure F- 1. Microcosm results from the test at the following conditions: 22oC, 500 mg N/L, 100% Oxygen and acclimated: Set 1...... 275 Figure F- 2. Microcosm results from the test at the following conditions: 22oC, 500 mg N/L, 100% Oxygen and acclimated: Set 2...... 276 Figure F- 3. Microcosm results from the test at the following conditions: 22oC, 1000 mg N/L, 100% Oxygen and acclimated: Set 1...... 277 Figure F- 4. Microcosm results from the test at the following conditions: 22oC, 200 mg N/L, 100% Oxygen and acclimated: Set 1...... 278 Figure F- 5. Microcosm results from the test at the following conditions: 22oC, 500 mg N/L, 0.7% Oxygen and acclimated: Set 1...... 279 Figure F- 6. Microcosm results from the test at the following conditions: 22oC, 500 mg N/L, 5% Oxygen and acclimated: Set 1...... 280 Figure F- 7. Microcosm results from the test at the following conditions: 22oC, 500 mg N/L, 5% Oxygen and acclimated: Set 2...... 281 Figure F- 8. Microcosm results from the test at the following conditions: 22oC, 1000 mg N/L, 5% Oxygen and acclimated: Set 1...... 282 Figure F- 9. Microcosm results from the test at the following conditions: 22oC, 500 mg N/L, 17% Oxygen and acclimated: Set 1...... 283

xiii Figure F- 10. Microcosm results from the test at the following conditions: 22oC, 1000 mg N/L, 17% Oxygen and acclimated: Set 1...... 284 Figure F- 11. Microcosm results from the test at the following conditions: 22oC, no ammonia, 100% oxygen: Biotic Control...... 285 Figure F- 12. Microcosm results from the test at the following conditions: 22oC, 500 mg N/L, 100% oxygen: Abiotic Control...... 286 Figure F- 13. Microcosm results from the test at the following conditions: 22oC, 500 mg N/L, 100% oxygen: Abiotic Control, Set 2...... 287 Figure F- 14. Microcosm results from the test at the following conditions: 22oC, 500 mg NO3-N/L, 100% oxygen: Nitrate Abiotic Control, Set 1...... 288 Figure F- 15. Microcosm results from the test at the following conditions: 22oC, 500 mg NO3-N/L, 100% oxygen: Nitrate Abiotic Control, Set 2...... 289 Figure F- 16. Microcosm results from the test at the following conditions: 22oC, 500 mg N/L, 100% oxygen: Unacclimated waste, Set 1...... 291 Figure F- 17. Microcosm results from the test at the following conditions: 22oC, 500 mg N/L, 100% oxygen: Unacclimated waste, Set 2...... 292 Figure F- 18. Microcosm results from the test at the following conditions: 35oC, 500 mg N/L, 100% oxygen: Set 1...... 294 Figure F- 19. Microcosm results from the test at the following conditions: 35oC, 500 mg N/L, 100% oxygen: Set 2...... 295 Figure F- 20. Microcosm results from the test at the following conditions: 35oC, 500 mg N/L, 100% oxygen: Set 3...... 296 Figure F- 21. Microcosm results from the test at the following conditions: 35oC, 1000 mg N/L, 100% oxygen: Set 1...... 297 Figure F- 22. Microcosm results from the test at the following conditions: 35oC, 1000 mg N/L, 100% oxygen: Set 2...... 298 Figure F- 23. Microcosm results from the test at the following conditions: 35oC, 500 mg N/L, 5% oxygen: Set 1...... 299 Figure F- 24. Microcosm results from the test at the following conditions: 35oC, 500 mg N/L, 5% oxygen: Set 2...... 300 Figure F- 25. Microcosm results from the test at the following conditions: 35oC, 1000 mg N/L, 5% oxygen: Set 1...... 301 Figure F- 26. Microcosm results from the test at the following conditions: 35oC, 1000 mg N/L, 5% oxygen: Set 2...... 302 Figure F- 27. Microcosm results from the test at the following conditions: 35oC, 500 mg N/L, 17% oxygen: Set 1...... 303 Figure F- 28. Microcosm results from the test at the following conditions: 35oC, 500 mg N/L, 17% oxygen: Set 2...... 304 Figure F- 29. Microcosm results from the test at the following conditions: 35oC, 1000 mg N/L, 17% oxygen: Set 1...... 305 Figure F- 30. Microcosm results from the test at the following conditions: 35oC, 500 mg N/L, 17% oxygen: Abiotic Control, Set 1...... 306 Figure F- 31. Microcosm results from the test at the following conditions: 35oC, no ammonia, 100% oxygen: Biotic Control, Set 1...... 307 Figure F- 32. Microcosm results from the test at the following conditions: 35oC, 500 mg NO3-N/L, 100% oxygen: Nitrate Biotic Control, Set 1...... 308

xiv Figure F- 33. Microcosm results from the test at the following conditions: 45oC, 500 mg N/L, 100% oxygen: Set 1...... 310 Figure F- 34. Microcosm results from the test at the following conditions: 45oC, 500 mg N/L, 100% oxygen: Set 2...... 311 Figure F- 35. Microcosm results from the test at the following conditions: 45oC, 500 mg N/L, 100% oxygen: Set 3...... 312 Figure F- 36. Microcosm results from the test at the following conditions: 45oC, 1000 mg N/L, 100% oxygen: Set 1...... 313 Figure F- 37. Microcosm results from the test at the following conditions: 45oC, 500 mg N/L, 5% oxygen: Set 1...... 314 Figure F- 38. Microcosm results from the test at the following conditions: 45oC, 500 mg N/L, 5% oxygen: Set 2...... 315 Figure F- 39. Microcosm results from the test at the following conditions: 45oC, 1000 mg N/L, 5% oxygen: Set 1...... 316 Figure F- 40. Microcosm results from the test at the following conditions: 45oC, 500 mg N/L, 17% oxygen: Set 1...... 317 Figure F- 41. Microcosm results from the test at the following conditions: 45oC, 500 mg N/L, 17% oxygen: Set 2...... 318 Figure F- 42. Microcosm results from the test at the following conditions: 45oC, 1000 mg N/L, 17% oxygen: Set 1...... 319 Figure F- 43. Microcosm results from the test at the following conditions: 45oC, 500 mg N/L, 100% oxygen: Abiotic Control...... 320 Figure F- 44. Microcosm results from the test at the following conditions: 45oC, no ammonia, 100% oxygen: Biotic Control...... 321

xv CHAPTER 1

INTRODUCTION

Background Information

A new and promising trend in solid waste management is to operate the landfill as a bioreactor. Bioreactor landfills are controlled systems in which moisture addition and/or air injection are used as enhancements to create a solid waste environment capable of actively degrading the biodegradable organic fraction of the waste. Several researchers have documented the benefits associated with bioreactor technology (Harper and

Pohland, 1988; Hudgens and Harper, 1998; Murphy et al., 1995; Pohland, 1995;

Reinhart, 1996; Reinhart and Townsend, 1998; Wairth and Sharma, 1998). One advantage is that increased waste degradation rates characteristic of bioreactor landfills allow for the life of a bioreactor landfill to be expanded beyond that of conventional landfills, potentially allowing for the reuse of one landfill rather than construction of many. Additionally, bioreactor landfills protect the environment and improve leachate quality. As leachate is recirculated, it is treated in-situ, significantly decreasing both the organic strength (i.e. chemical oxygen demand (COD) and biochemical oxygen demand

(BOD)) and potential impact to the environment (Pohland, 1995; Reinhart and Al-Yousfi,

1996; Reinhart and Townsend, 1998).

1 Treating the landfill as a bioreactor is an advantageous method for solid waste management. However, some challenges remain. One such challenge is the ammonia- nitrogen concentration found in the leachate. Although the organic strength of the leachate is reduced in bioreactor landfills, the concentrations of ammonia-nitrogen tend to increase beyond concentrations found in leachate from conventional landfills (Onay and

Pohland, 1998; Price et al., 2003). Recirculating leachate increases the rate of ammonification and results in accumulation of higher levels of ammonia-nitrogen concentrations, even after the organic fraction of the waste is stabilized (Burton and

Watson-Craik, 1998; Onay and Pohland, 1998). Thus the toxicity of the leachate increases, potentially inhibiting the degradation process and necessitating leachate treatment before ultimate disposal to protect receiving waters (Burton and Watson-Craik,

1998). Because ammonia-nitrogen persists even after the organic fraction of the waste is stabilized, and because of its toxic nature, it is likely that ammonia-nitrogen will determine when the landfill is biologically stable and when post-closure monitoring may end (Kjeldsen et al. 2002, Robinson and Maris, 1993). Thus an understanding of the fate of nitrogen in bioreactor landfills is critical to a successful and economic operation.

Several researchers have explored removal of ammonia-nitrogen from leachate removed from landfills (Cheung et al., 1997; Hoilijoki et al., 2000; Ilies and Mavinic,

2001; IM et al., 2001; Jokela et al., 2002; Marttined et al., 2002; Robinson and Maris,

1998; Shiskowski and Mavinic, 1998; Siegrist et al., 1998; Welander et al., 1997; XZ et al., 1999; Zouboulis et al., 2001). Removal methods often include complex sequences of physical, chemical, and/or biological processes, including methods such as chemical precipitation, nanofiltration, air-stripping, and nitrification/denitrification via various

2 reactor configurations (i.e. rotating biological filters, suspended growth reactors, and

attached growth reactors).

Nitrification/denitrification is an advantageous removal mechanism because

complete destruction of nitrogen can be achieved. Therefore it is often used and

practiced, primarily outside of the landfill (Hoilijoki et al., 2000; Martienssen and

Schops, 1997; Welander et al., 1997). However, additional costs are associated with ex-

situ treatment of ammonia, as separate treatment units on site must be maintained or the

leachate must be pumped to a publicly owned wastewater treatment facility. Therefore,

the development of an in-situ nitrogen removal technique would be an attractive

alternative. Several recent in-situ treatment approaches have been explored. These

include external nitrification and in-situ denitrification (Aljarallah and Atwater, 2000;

Burton and Watson-Craik, 1999; Ding et al., 2001; Price, 2001; Price et al., 2003; Waste

Management, 2001), incidental treatment in aerobic or semi-aerobic landfills

(Hanashima, 1999), and establishing in-situ dedicated treatment zones (Onay, 1995;

Onay and Pohland, 1998). These studies have demonstrated the efficacy of such

processes, however they are still not well understood and lack the information required

for implementation at field-scale bioreactor landfills.

Operating a successful in-situ nitrification/denitrification system at field-scale would potentially have both economic and environmental advantages. Although implementation of both nitrification and denitrification has been extensively explored and defined for wastewater, landfills are more complex biological systems, suggesting wastewater derived parameters may not be valid for similar processes in landfills. This

3 research will result in developing information necessary for field-scale implementation of

in-situ nitrification/denitrification.

Research Objectives and Scope of Work

Given the adverse environmental impacts of ammonia and the increases in concentration observed in bioreactor landfills, it is important to investigate the fate and transformation of nitrogen in a bioreactor landfill environment. In-situ nitrification/denitrification in solid waste systems is much more complex than similar processes in wastewater treatment and typical attached-growth systems. Unlike most attached-growth processes, the attachment media in landfills changes over time, shrinking in size and diminishing in organic content. Additionally, there are several parameters that

cannot be controlled or that are very difficult to control in landfills. For example, the

waste composition varies among and within landfills. Historically, parameters such as

pH, oxidation-reduction potential (ORP), moisture content, porosity, and particle size

could not be controlled. However, these parameters can now be controlled and used to aid

in creating an environment conducive to microbial degradation and biological nitrogen

removal. The goal of this research is to provide information regarding the fate of nitrogen

in bioreactor landfills, as well as to provide information required to develop guidance for

field-scale implementation of in-situ nitrification/denitrification.

The specific research objectives of this research include:

• Evaluate the fate of nitrogen in bioreactor landfills,

4 • Determine the optimal environmental conditions and kinetics of in-situ

nitrification,

• Develop an implementation strategy for in-situ nitrification and

denitrification at field-scale, and

• Complete an economic comparison between in-situ and ex-situ treatment

of ammonia.

The objectives of this research study were met by implementing laboratory studies aimed at determining and deriving the parameters and relationships required to provide guidance for field-scale implementation. Two main laboratory studies were conducted. A waste acclimation reactor was operated and was dual purpose, to provide a waste source for parallel batch microcosm studies and to demonstrate the efficacy of in-situ nitrification and denitrification. The smaller scale microcosm study experiments were conducted to evaluate the kinetics of ammonia removal under different environmental conditions likely found in bioreactor landfills and to evaluate the fate of nitrogen in bioreactor landfills. The conditions evaluated include gas-phase oxygen concentrations

(between 0.7 and 100%), temperature (22, 35, and 45oC) and ammonia concentration

(500 and 1,000 mg N/L). All experiments were conducted using older waste, as it is hypothesized that older waste environments are more optimal for in-situ nitrogen removal processes. A multiplicative Monod model that includes factors describing the impact of the environmental conditions evaluated was developed and used to develop guidance for field-scale implementation.

5 Dissertation Organization

This dissertation is organized in six chapters. Chapter 2 presents a literature review and discussion regarding the fate of nitrogen in bioreactor landfills. This work has been published in the journal Critical Reviews in Environmental Science and

Technology (Berge et al., 2005).

Chapter 3 describes results of experiments conducted evaluating in-situ ammonia removal and the impact of waste acclimation on removal kinetics. Results demonstrate that in-situ nitrification is feasible in an aerated solid waste environment. All rate data fit well to Monod kinetics and indicate mass transfer limitations may occur. Although specific rates of ammonia removal in the unacclimated waste were lower than in the acclimated waste, a relatively quick start-up of ammonia removal was observed in the unacclimated waste. This work has been published in the journal Waste Management

(Berge et al., 2006).

Chapter 4 describes experiments conducted evaluating the impact of different gas- phase oxygen concentrations and temperatures on ammonia removal. Significant ammonia removal was observed in all studies at all gas-phase oxygen concentrations and temperatures tested. An end-result of the experiments is the development of a multiplicative Monod model that includes terms describing the impact of oxygen, temperature, ammonia concentration and pH on ammonia removal (work has been submitted to the journal Environmental Science and Technology).

Chapter 5 describes the engineering significance of the experiments conducted. A mass balance model was constructed using FORTRAN and used to predict an optimal field-scale implementation plan. The economics of the in-situ nitrogen removal process

6 are presented and suggest the method is economically feasible (presented at the 2006

Intercontinental Landfill Research Symposium, Lapland, Sweden).

Chapter 6 contains the conclusions and recommendations resulting from this

research. There are several appendices included that contain more detailed information

regarding the materials and methods employed and descriptions of statistical procedures

used to evaluate the experimental results. Appendix A provides a more detailed

descritption of the materials and methods used throughout this study. Appendix B

contains the results from the waste acclimation process. Appendix C presents the procedure used for validating all parameters in the multiplicative Monod model derived based on the laboratory data. Confidence intervals were calculated on all microcosm data; details regarding this are contained in Appendix D. The model FORTRAN code is included in Appendix E and individual plots of all microcosm studies can be found in

Appendix F. Details regarding the economic comparison calculations are located in

Appendix G.

References

Aljarallah, R.S. and Atwater, J. (2000) Denitrification in Landfill Bioreactor. In Proceedings of WasteTech 2000, Orlando, Fl.

Al-Yousfi, A. B. and Pohland, F.G. (1998) Strategies for Simulation, Design, and Management of Solid Wastes Disposal Sites as Landfill Bioreactors, Practice Periodical of Hazardous, Toxic, and Radioactive Waste Management 2(1), 13- 21.

Berge, N.D., Reinhart, D.R. and Townsend, T.G. (2005) The Fate of Nitrogen in Bioreactor Landfills, Critical Reviews in Environmental Science and Technology 35 (4), 365-399.

7 Berge, N.D., Reinhart, D.R., Dietz, J. and Townsend, T. (2006) In Situ Ammonia Removal in Bioreactor Landfill Leachate, Waste Management 26 (4), 334-343.

Burton, S. A. Q. and Watson-Craik, I.A. (1998) Ammonia and Nitrogen Fluxes in Landfill Sites: Applicability to Sustainable Landfilling, Waste Management and Research 16, 41-53.

Burton, S. A. Q. and Watson-Craik, I.A. (1999) Accelerated Landfill Refuse Decomposition by Recirculation of Nitrified Leachate, In Proceedings of Sardinia 99 Seventh Waste Management and Landfill Symposium, Cagliari, Italy.

Cheung, K.C., Chu, LM and Wong, MH (1997) Ammonia Stripping as a Pretreatment for Landfill Leachate, Water Air and Soil Pollution 94 (1-2), 209-221.

Ding, A., Zhang, Z., Fu, J. and Cheng, L. (2001) Biological Control of Leachate from Municipal Landfills, Chemosphere 44, 1-8.

Hanashima, M. (1999) Pollution Control and Stabilization Process by Semi-Aerobic Landfill Type: The Fukuoka Method, In Proceedings of Sardinia 99 Seventh Waste Management and Landfill Symposium, Cagliari, Italy.

Harper, S.R. and Pohland, F.G. (1988) Landfills – Lessening Environmental Impacts, Civil Engineering 11, 66-69.

Hoilijoki, T.H, Kettunen, R.H. and Rintala, J.A. (2000) Nitrification of Anaerobically Pretreated Municipal Landfill Leachate at Low Temperature, Water Research 34 (5), 1435-1446.

Hudgens, M. and Harper, S. (1998) Operational Characteristics of Two Aerobic Landfill Systems, http://www.duesouth.net/~ecsi/page3. Accessed September 1999.

Ilies, P. and Mavinic, D.S. (2001) The Effect of Decreased Ambient Temperature on the Biological Nitrification and Denitrification of a High Ammonia Landfill Leachate, Water Resources 35(8), 2065-2072.

IM, J.H., Woo, H.J., Choi, M.W., Han, K.B., and Kim, C.W. (2001) Simultaneous Organic and Nitrogen Removal from Municipal Landfill Leachate using an Aerobic-Anaerobic System, Water Resources 35, 2403-2410.

Jokela, J.P.Y., Kettunen, R.H., Sormunen, K.M. and Rintala, J.A. (2002). Biological Nitrogen Removal from Municipal Landfill Leachate: Low-Cost Nitrification in Biofilters and Laboratory Scale in-situ Denitrification, Water Research 36(16), 4079-4087.

8 Kjeldsen, P., Barlaz, M.A., Rooker, R., Baun, A., Ledin, A. and Christensen, T.H. (2002). Present and Long-Term Composition of MSW Landfill Leachate: A Review, Critical Reviews in Environmental Science and Technology, 32(4), 297-336.

Martienssen, M. and Schops, R. (1997) Biological Treatment of Leachate From Solid Waste Landfill Sites—Alterations in the Bacterial Community During the Denitrification Process, Water Research 31 (5), 1164-1170.

Marttinen, S.K., Kettunen, R.H., Sormunen, K.M., Soimasuo, R.M. and Rintala, J.A. (2002) Screening of Physical-Chemical Methods for Removal of Organic Material, Nitrogen and Toxicity from Low Strength Landfill Leachates, Chemosphere 46, 851-858.

Murphy, R.J., Jones, D.E. and Stessel, R.I. (1995) Relationship of Microbial Mass and Activity in Biodegradation of Solid Waste, Waste Management and Research 13, 485-497.

Onay, T.T. (1995) In-situ Attenuation of Nitrogeneous Compounds in Controlled Landfills, MS Thesis, Dept. of Civil and Environmental Engineering, University of Pittsburgh, Pittsburgh, PA.

Onay, T.T. and Pohland, F.G. (1998) In-situ Nitrogen Management in Controlled Bioreactor Landfills, Water Resources 32 (5), 1383-1392.

Pohland, F.G. (1995) Landfill Bioreactors: Historical Perspective, Fundamental Principles and New Horizons in Design and Operations, In EPA Seminar Publication: Landfill Bioreactor Design and Operation.

Price, G. A., Barlaz, M. A., and Hater, G. R. (2003). Nitrogen Management in Bioreactor Landfills, Waste Management, 23(7), 675-688.

Price, G.A. (2001) Long-Term Nitrogen Management in Bioreactor Landfills, MS Thesis, North Carolina State University, Raleigh, NC.

Reinhart, D.R. (1996) Full-Scale Experiences with Leachate Recirculating Landfills: Case Studies, Waste Management and Research 14, 347-365.

Reinhart, D.R. and Al-Yousfi, A.B. (1996) The Impact of Leachate Recirculation on Municipal Solid Waste Landfill Operating Characteristics, Waste Management and Research 14, 337-346.

Reinhart, D.R. and Townsend, T.G. (1998) Landfill Bioreactor Design and Operation, Boca Raton, New York, USA: CRC Press LLC.

9 Robinson, H.D. and Maris, P.J. (1993) The Treatment of Leachates From Domestic Waste In Landfill Sites, Journal of Water Pollution Control Federation 57(1), 30.

Shiskowski, D. M., and Mavinic, D. S. (1998). Biological Treatment of a High Ammonia Leachate: Influence of External Carbon During Initial Startup, Water Research, 32(8), 2533-2541.

Siegrist, H., Reithaar, S., Koch, G. and Lais, P. (1998) Nitrogen Loss in a Nitrifying Rotating Contractor Treating Ammonium-Rich Wastewater Without Organic Carbon, Water Science and Technology 38 (8-9), 241-248.

Stensel, H.D., Loehr, R.C., and Lawrence, A.W. (1973) Biological Denitrification of Suspended-Growth Denitrification, Journal of Water Pollution Control Federation 45(2), 249-261.

Warith, M. and Sharma, R. (1998) Technical Review of Methods to Enhance Biological Degradation in Sanitary Landfills, Water Quality Research Journal of Canada 33(3), 417-437.

Waste Management (2001) The Bioreactor Landfill: The Next Generation of Landfill Management, A White Paper, http://www.wastemanagement.com, Accessed Sept. 15, 2001.

Welander, U., Henrysson, T. and Welander, T. (1997) Nitrification of Landfill Leachate Using Suspended-Carrier Biofilm Technology, Water Resources 32 (5), 2351- 2355.

XZ, L., QL, Z., and XD, H. (1999) Ammonium Removal From Landfill Leachate by Chemical Precipitation, Waste Management 19 (6), 409-415.

Youcai, Z., Hua, L., Jun, W. and Guowei, G. (2002) Treatment of Leachate by Aged- Refuse-Based Biofilter, Journal of Environmental Engineering 128(7), 662-668.

Zouboulis, A.I., Loukidou, M.X., and Christodoulou, K. (2001) Enzymatic Treatment of Sanitary Landfill Leachate, Chemosphere 44, 1103-1108.

10 CHAPTER 2

THE FATE OF NITROGEN IN BIOREACTOR LANDFILLS

NOTE: This paper has been previously published as: Berge, N.D., Reinhart, D.R. and Townsend, T.G. (2005) The Fate of Nitrogen in Bioreactor Landfills, Critical Reviews in Environmental Science and Technology 35 (4), 365-399 (DOI: 10.1080/10643380590945003). The journal Critical Reviews in Environmental Science and Technology can be found at www.tandf.co.uk/journals/titles/10643389.asp and this article can be accessed on the web at the following address: http://journalsonline.tandf.co.uk/(2na1lirimfayc145spcv1dn5)/app/home/contribution.asp ?referrer=parent&backto=issue,2,3;journal,4,35;linkingpublicationresults,1:106794,1

Introduction

A new and promising trend in solid waste management is to operate the landfill as a bioreactor. Bioreactor landfills are controlled systems in which moisture addition (often leachate recirculation) and/or air injection are used to create a solid waste environment capable of actively degrading the readily biodegradable organic fraction of the waste.

Several researchers have documented the benefits associated with bioreactor technology

(Murphy et al., 1995; Pohland, 1995; Reinhart, 1996; Reinhart and Townsend, 1998;

Warith and Sharma, 1998). One advantage is that increased waste degradation rates characteristic of bioreactor landfills permit the life of a bioreactor landfill to be expanded beyond that of conventional landfills through recovery of valuable airspace. As leachate is recirculated, it is treated in-situ, decreasing its organic strength and thus potential

impact to the environment. In-situ treatment potentially reduces the length of the post-

11 closure care period and associated costs (Pohland, 1995; Reinhart and Al-Yousfi, 1996;

Reinhart and Townsend, 1998). Additionally, bioreactor landfills stimulate gas production; the majority of the methane is produced earlier in the life of the landfill allowing for more efficient capture and subsequent use.

Although the organic strength of the leachate is significantly reduced in bioreactor landfills, ammonia-nitrogen remains an issue. The ammonia-nitrogen concentrations found in leachate from bioreactor landfills are greater than those found in leachate from conventional landfills (Barlaz et al., 2002; Onay and Pohland, 1998). Ammonia-nitrogen tends to accumulate in both systems because there is no degradation pathway for ammonia-nitrogen in anaerobic systems. However, in bioreactor landfills, moisture addition and/or recirculating leachate increases the rate of ammonification, resulting in accumulation of higher levels of ammonia-nitrogen, even after the organic fraction of the waste is degraded (Barlaz et al., 2002; Burton and Watson-Craik, 1998; Onay and

Pohland, 1998; Price et al., 2003). The increased ammonia-nitrogen concentrations intensifies the toxicity of the leachate to aquatic species (Ward et al., 2002), potentially inhibiting the degradation process and necessitating leachate treatment before ultimate disposal to protect receiving waters (Burton and Watson-Craik, 1998). It has been suggested that ammonia-nitrogen is one of the most significant long-term pollution problem in landfills (Barlaz et al., 2002) and it is likely that the presence of ammonia- nitrogen will determine when the landfill is biologically stable and when post-closure monitoring may end (Price et al., 2003). Thus an understanding of the fate of nitrogen in bioreactor landfills and possible mechanisms for ammonia-nitrogen removal is critical to both a successful and economic operation.

12 As more and more landfills transition operation to bioreactors, more attention

must be paid to how operating the landfill in such a manner may affect the fate of

nitrogen. The in-situ physical, chemical and biological processes in bioreactor landfills

differ from those typically observed when operating a landfill conventionally, potentially

resulting in different nitrogen transformation and removal processes. The fate of nitrogen

in bioreactor landfills is not well understood. Because of the adverse impact ammonia-

nitrogen has on the environment, an understanding of nitrogen transformation processes

in bioreactor landfills is necessary to ensure adverse environmental impacts and/or

treatment costs are minimized by expanding the current use of landfills to include in-situ

leachate treatment.

This paper discusses the nitrogen transformation and removal processes that may

occur in bioreactor landfills. Little research has been conducted evaluating the fate of

nitrogen in bioreactor landfills, or in conventional landfills for that matter. However, it is

suspected that processes that typically occur in wastewater treatment and in soils will also

occur in bioreactor landfills, but in a much less controlled fashion, as the inherent

variability and heterogeneities in bioreactor landfills do not allow for them to be operated

with the high level of control possible in wastewater treatment processes. Using

wastewater and soil literature, as well as landfill related literature, nitrogen removal and

transformation processes that may occur in bioreactor landfills are discussed and

evaluated in this review.

Bioreactor Landfill Operation

Traditionally, landfills have been thought of as storage and containment systems, functioning primarily to entomb the waste. Recently, however, the focus of solid waste

13 management has changed to regarding the landfill as a complex biological system

capable of managing solid waste in a more proactive manner, acting to degrade the

readily biodegradable material (Pohland, 1975; Reinhart et al., 2002; Reinhart and

Townsend, 1998). Because bioreactor landfill environments are different from

conventional landfills, there is potential for a greater number of nitrogen transformation

and removal processes to occur and for them to occur to a greater extent than in

conventional landfills. System design of bioreactor landfills provides the flexibility in the

location and duration of liquid and air injection, allowing for adjustment of pH,

oxidation-reduction potential (ORP), and moisture content to create an environment

conducive to microbial degradation and biological nitrogen removal. System design is

rigid with respect to parameters such as waste composition and age (i.e. organic carbon

content); waste components cannot be controlled and vary from landfill to landfill, while

waste age varies from location to location within a landfill. Thus, in a landfill, the active

control of in-situ reactions and nitrogen removal/transformation is generally restricted by

the location and volume of injected liquid and air.

Liquid addition to landfills has many advantages associated with it. Leachate

recirculation involves the collection and redistribution of leachate through the landfill.

Moisture addition and movement are important factors affecting waste biodegradation resulting in an increase in the moisture content of the waste and distribution of nutrients throughout the landfill, respectively. Optimal levels of moisture content have been found to be between 40 and 70%, on a wet weight basis (Barlaz et al., 1990). Much research has been conducted evaluating the benefits associated with increasing the moisture content of solid waste and can be found elsewhere (Reinhart and Townsend, 1998). At times,

14 insufficient leachate is available and it is necessary to supplement with other liquids such

as groundwater, stormwater, wastewater, or surface water.

Achieving uniform liquid distribution is difficult. Waste heterogeneities and

differences in compaction within landfills create distribution challenges. Injected liquid

will flow around areas with lower hydraulic conductivities and channel through the waste

following preferential flow pathways formed by areas of higher hydraulic conductivites;

the areas of higher hydraulic conductivity may be due to waste heterogeneity or differences in compaction ratios (McCreanor and Reinhart, 2000). The non-uniform distribution that occurs results in portions of the landfill (on both a micro- and macro level) having various moisture contents and thus different waste degradation rates; therefore, several microbial consortiums will be present, potentially in close proximity to one another, allowing for different types of microbial degradation and thus nitrogen removal/transformations to occur simultaneously. Differential settlement may also occur as a result of the changes in waste degradation with respect to location. There are different methods that can be used to reinject leachate or add liquid to landfills, including horizontal trenches and vertical injection pipes. These recirculation methods have been reviewed elsewhere (Reinhart, 1996; Reinhart and Al-Yousfi, 1996). Reintroduction rates, for horizontal trenches, vary from 0.15 to 0.30 gpm/ft trench, while vertical injection rates in wells are generally from 0.5 to 2.5 gpm (Reinhart and Townsend, 1998).

Air addition has also been used as an enhancement and has been shown to enhance degradation processes in landfills at both the field and laboratory scale

(American Technologies, 1997; Florida Center for Solid and Hazardous Waste, 2004;

Leikam et al., 1999; Merz and Stone, 1970; Murphy et al., 1995; Read et al., 2001;

15 Stessel and Murphy, 1992). Adding air uniformly throughout the waste is also a

challenge. Not only do waste heterogeneities and compaction affect the air distribution,

the presence of moisture does as well. Air will take the path of least resistance, thus there

will likely be areas of an aerobic landfill in which air does not reach, resulting in anoxic

or anaerobic pockets within the waste mass.

Generally, bioreactor landfills undergo the same degradation processes as

conventional landfills, just at a faster rate and to a greater extent because of the

optimization of in-situ conditions. However, degradation pathways may vary depending on the operation of the bioreactor landfill. Compared with conventional landfills, bioreactor landfills have shown a more rapid and complete waste conversion and stabilization process (Harper and Pohland, 1988; Pohland, 1995; Reinhart, 1996; Warith and Sharma, 1998). Increased waste degradation rates characteristic of bioreactor landfills

may allow for the life of a bioreactor landfill site to be expanded beyond that of a

conventional landfill, potentially allowing for the reuse of one site rather than

construction of many. Because waste degradation rates increase in bioreactor landfills,

airspace may be created by settlement and filled prior to closure. Moisture injection

increases the rate of initial settlement due to additional unit weight, and, over time,

increases the extent of waste degradation, all resulting in the recovery of a significant

volume of airspace. For example, Reinhart and Al-Yousfi (1996) reported that for one

landfill 13-15% settlement occurred over a four-year period when recirculating leachate;

a dry control cell at the same site settled only 8-12%. Bioreactor landfills also provide a

means to store and /or treat leachate. As leachate is recirculated, it is treated in-situ via

naturally occurring processes such as adsorption, ion exchange, and mechanical filtration

16 (Pohland et al., 1992), significantly decreasing both the organic strength (i.e. chemical

oxygen demand (COD) and biochemical oxygen demand (BOD) by almost 50%), and

heavy metal content, thus reducing impact to the environment were the leachate to reach

the groundwater or surface water (Pohland, 1995; Reinhart and Al-Yousfi, 1996). Not

only can leachate be treated within bioreactor landfills, but it may also be stored by

adsorption by the waste, rather than stored external to the landfill. Perhaps the biggest

advantage of bioreactor landfills is the reduction of landfill biological stabilization time

(Reinhart and Al-Yousfi, 1996). This reduction in time has been repeatedly proven

through the reduction of COD half-lives in landfills utilizing leachate recirculation

(Reinhart and Townsend, 1998); COD half-lives in leachate from conventional landfills

have been calculated to be around 10 years, whereas for bioreactor landfills, the COD

half-life of the leachate is closer to 230 to 380 days (Reinhart and Al-Yousfi, 1996;

Reinhart and Townsend, 1998).

Four types of bioreactor landfills have been explored, each with different operating schemes to obtain optimal results: anaerobic, aerobic, facultative, and hybrid systems. Each bioreactor type is a patented process (Ham and Viste, 1999; Hater and

Green, 2001; Hater and Green, 2002; Hudgins et al., 2000).

Anaerobic Bioreactor Landfills

Anaerobic bioreactor landfills are those in which moisture addition is practiced.

Sources of liquid addition may include groundwater, stormwater, infiltrating rainfall, or leachate. Moisture content adjustment results in enhanced methane production, which has been repeatedly demonstrated in several laboratory, pilot and field scale studies (Doedens

17 and Cord-Landwehr, 1989; Florida Center for Solid and Hazardous Waste, 2004; Otieno,

1989; Pohland, 1975; Reinhart, 1996; Tittlebaum, 1982; Townsend et al., 1996). Because

waste degradation is enhanced in anaerobic bioreactors and organic material is returned

to the landfill via leachate recirculation (Reinhart and Al-Yousfi, 1996), methane is

produced at a much faster rate. The total volume of gas produced also increases, as

organics in the leachate are recycled and then biodegraded within the landfill. The

majority of gas production may be confined to a few years, earlier in the life of the

landfill, than traditionally occurs in conventional landfills, allowing for more efficient

capture and subsequent use (Reinhart and Al-Yousfi, 1996). Gas production time frames

are highly dependent on the moisture content of the waste. Modeling of gas production

from bioreactor landfills requires different parameters than used for conventional landfills

(Barlaz et al., 2002; Faour, 2003). As the parameters are fitted for wet landfills, the time

for 99% of the methane to be produced may decrease by almost fourteen fold (Faour,

2003). Although the vast majority of the gas will be produced relatively early after

closing the landfill (within 20 years), limited methane production may continue over long

periods of time due to wetting of previously unreached dry areas.

Anaerobic bioreactor landfills are more effective at degrading the solid waste than

conventional landfills. However, when compared to other types of bioreactor landfills,

anaerobic systems tend to have lower temperatures and slower degradation rates (Merz and Stone, 1970; Stessel and Murphy, 1992). A disadvantage to operating the landfill as an anaerobic bioreactor is the accumulation of ammonia-nitrogen. In anaerobic bioreactor landfills, the ammonia-nitrogen present in the leachate is continually returned to the

landfill where there is no degradation pathway for ammonia in anaerobic environments.

18 An advantage of operating the bioreactor anaerobically when compared to other bioreactor landfill types is that air is not added; therefore the operational costs are less than what would be incurred aerobically and methane can be captured and reused.

Aerobic Bioreactor Landfills

Adding air to landfills has been shown to enhance degradation processes in landfills, as aerobic processes tend to degrade organic compounds typically found in municipal solid waste (MSW) in shorter time periods than anaerobic degradation processes (American Technologies, 1999; Leikam et al., 1999; Murphy et al., 1995; Read et al., 2001; Stessel and Murphy, 1992). Reported advantages of operating the landfill aerobically rather than anaerobically include increased settlement, decreased metal mobility, reduced ex-situ leachate treatment required, lower leachate management and methane control costs, and reduced environmental liability (Environmental Control

Systems, 1999; Read et al., 2001). Both laboratory and field-scale studies have been conducted showing the effectiveness of the aerobic bioreactor landfill system (Leikam et al., 1999; Merz and Stone, 1970; Smith et al., 2000).

Many of the nitrogen transformation/removal process are favored by aerobic processes, including nitrification and ammonia air stripping or volatilization. Air stripping and volatilization may be favored in aerobic bioreactor landfills because of higher pH levels and temperatures that are inherent in an aerobic environment. The additional gas flow associated with air injection may also induce greater masses of ammonia-nitrogen removal.

19 During aerobic degradation of MSW, biodegradable materials are converted

mostly to carbon dioxide and water. Little, if any, methane is produced which may be

viewed as either an advantage or disadvantage, depending on whether methane collection

and use as an energy source is desired or required. Methane is a potent greenhouse gas,

thus if it cannot be efficiently controlled and collected in anaerobic landfills, its

production can be a local environmental concern. Further, the solid waste environment

during aerobic degradation has a fairly neutral pH (Hanashima, 1999; Merz and Stone,

1970; Read et al., 2001; Stessel and Murphy, 1992), which decreases metal mobility.

Volatile organic acid production is decreased in aerobic bioreactors because the

anaerobic fermentation processes are limited. However, volatile acid and methane

production may still occur in anaerobic pockets within the landfill.

The aerobic process generates a considerable amount of heat, leading to elevated

in-situ temperatures as high as 66oC (American Technologies, 1997; Merz and Stone,

1970; Stessel and Murphy, 1992). The elevated temperatures increase evaporation, which results in a significant loss of leachate. As a consequence, there is less leachate to manage

(Environmental Control Systems, 1999). The high temperatures may limit certain biological nitrogen transformation processes from occurring although no data regarding temperature effects are available. Additionally, the combination of the high temperatures and presence of air may create a fire potential. However, by minimizing methane production and ensuring proper moisture contents, fire potential is lessened.

Odors often associated with anaerobic systems, such as hydrogen sulfide and volatile acids, are reduced in aerobic bioreactor landfills. Aerobic processes do have some odor associated with them, however, it is an earthy smell. Some odorous

20 compounds emitted by aerobic composting include methanethiol, which has a pungent sulfide odor (Miller, 1992).

Hybrid Bioreactor Landfills

Another, less studied type of bioreactor landfill that shows promise is the hybrid bioreactor. This type of bioreactor landfill is still in the early stages of development.

Hybrid bioreactor landfills involve the combination of both aerobic and anaerobic conditions. Two types of these aerobic/anaerobic systems have been explored: short term cycling of air injection into the landfill and sequencing of aerobic and anaerobic conditions.

Cycling of air injection into the landfill is defined as a pattern of alternating in- situ aerobic and anaerobic conditions that is repeated throughout the life-cycle of the landfill, while sequencing of air-injection into the landfill involves an initial aerobic phase, followed by a final anaerobic phase. Because there are many advantages associated with both aerobic and anaerobic degradation processes, researchers see combining the processes as a way to maximize the potential of a bioreactor landfill.

There are some components in both the waste and leachate that are recalcitrant in anaerobic conditions, but degradable in aerobic environments, such as lignins and aromatic compounds. Utilizing one of these hybrid techniques may allow for the leachate and/or waste to be treated more completely (Berge, 2001; Pichler and Kogner-Knabner,

2000; Reinhart et al., 2002). Operating a bioreactor landfill as a hybrid system may serve to combine several nitrogen transformation and removal processes, such as nitrification

21 and denitrification, potentially resulting in complete in-situ removal of nitrogen from

landfills.

A few laboratory studies have been completed evaluating the effect of cyclic air-

injection on the performance of bioreactor landfills (Berge, 2001; Pichler and Kogner-

Knabner, 2000; Ziehmann and Meier, 1999). Each cyclic air-injection system evaluated achieved a more biologically stable leachate with respect to COD in a shorter period of time than that experienced by purely aerobic systems. Ziehmann and Meier (1999)

conducted both laboratory and pilot scale studies evaluating this technique. Three

bioreactor systems were operated for 180 days. Anaerobic and aerobic conditions were

cycled based on the methane concentration measured; once the methane concentration

reached 2.5%, by volume, air was added. Results from the laboratory study showed that

the leachate from the reactor in which aerobic and anaerobic conditions were alternated

had lower concentrations of total organic carbon and COD than those from either the

anaerobic or aerobic reactors. However, when operating the pilot-scale study, there was

little difference between the cyclic and continuously aerobic reactors, suggesting that the

advantages of the cyclic system seen in the small-scale studies may not be realized at

field scale. Each study was conducted over short time periods, so additional long-term

studies are needed to evaluate this process further.

A few studies have also evaluated the effect of a sequencing air-injection system

(Boni et al., 1996; Stegmann and Spendlin, 1989). In this system, waste is placed in lifts.

The first lift is aerated for a period of time; when the second lift is placed, aeration of the

first layer stops and aeration of the second layer commences. Leachate is continuously

recirculated. This process continues until the landfill is filled (Hater and Green, 2001). It

22 is hypothesized that this system acts to speed typical anaerobic degradation processes,

specifically the onset of methanogenesis. By initially aerobically degrading the waste, the temperature of the waste is increased and the extent of the acidogenic phase is reduced, therefore allowing for the early onset of methanogenesis. Fletcher et al. (1989) conducted a study that demonstrated the effect of increasing temperatures on methane production. Air was briefly added to an older landfill using vertical injection wells to promote aerobic activity. As a result of the air addition, local temperatures increased by

17oC. Methane production was stimulated as a result of the increase in temperature.

Stegmann and Spendling (1989) conducted lysimeter tests evaluating sequencing of air addition. In their studies, waste was loosely placed in thin layers (from 0.4 to 2 m) with no cover to allow natural air diffusion into the waste; leachate was also recirculated.

In the lysimeters with the 0.4-m lifts, a new loosely placed lift of waste was applied every

6 weeks. Another lysimeter was operated with waste placed in 2-m lifts; after two years, another 2-m layer of loose waste was placed. The addition of waste lifts prevented air intrusion into the lower layers of waste, resulting in the lower layers becoming anaerobic.

The investigators found that the waste placed in thinner layers resulted in the production of methane earlier. Because of the initial aerobic degradation of the readily biodegradable organics, it was hypothesized that the organic acid production was reduced and did not reach concentrations inhibitory to methane production. Methane production rates were not measured.

23 Facultative Bioreactor Landfills

Facultative bioreactor landfills are operated with the intent of actively degrading

the waste mass and, at the same time, controlling high ammonia-nitrogen concentrations

typically found in the leachate from bioreactor landfills. In facultative systems, leachate

is removed from the bioreactor landfill and nitrified in an external treatment system prior

to recirculation (Hater and Green, 2002). Thus, the ammonia-nitrogen concentrations of

the treated leachate are low to non-existent, while the nitrate levels are high. As the

nitrate-rich leachate is recirculated and passes through the landfill, denitrification occurs

since several microorganisms, including facultative microorganisms, use the nitrate for

respiration. Although this type of bioreactor has not been evaluated in many studies, there

is laboratory evidence suggesting that implementation of such a system is plausible

(Price, 2001; Price et al., 2003; US Environmental Protection Agency and Waste

Management, 2003). Price et al. (2003) conducted a laboratory scale study

demonstrating the ability of this process to denitrify nitrified leachate as it passed through

the waste. The Outer Loop Landfill in Louisville, Kentucky is in the process of using this

approach for controlling nitrogen discharges (US Environmental Protection Agency and

Waste Management, 2003). A disadvantage of this technique is that external treatment of

leachate for ammonia-nitrogen removal must occur, which adds an extra step to the

bioreactor landfill process and can be both difficult and costly because of high levels of

ammonia-nitrogen in the leachate. Additionally, while denitrification of the leachate is occurring, methane production may be halted until the nitrate is consumed. It has been shown that methane production quickly resumes after nitrate is depleted (Price et al.,

2003).

24 Ammonia-Nitrogen in Leachate

The ammonia-nitrogen in leachate is derived from the nitrogen content of the waste; the concentration is dependent on the rate of solubilization and/or leaching from the waste. The nitrogen content of MSW is less than 1%, on a wet-weight basis

(Tchobanoglous et al., 1993), and is composed primarily of the proteins contained in yard wastes, food wastes, and biosolids (Burton and Watson-Craik, 1998). As the proteins are hydrolyzed and fermented by microorganisms, ammonia-nitrogen is produced. This process is termed ammonification, yielding concentrations ranging from less than detection levels to over 5,000 mg/L (Ehrig, 1989; Grosh, 1996; Qasim and Chiang, 1994;

Reinhart and Townsend, 1998).

Leachate composition is quite variable, depending highly on waste composition, moisture content of the waste, and age of the landfill. Table 2-1 provides ammonia- nitrogen concentration ranges for both conventional and bioreactor landfills as a function of waste age as summarized by Reinhart and Townsend (Reinhart and Townsend, 1998).

Table 2-1. Ammonia-Nitrogen Concentrations in Both Conventional Bioreactor Landfills with Respect to Degree of Landfill Biological Stabilization (Reinhart and Townsend, 1998). Concentration (mg/L as N) Stabilization Phase Conventional Bioreactor Landfills Landfills Transition 120 - 125 76 - 125 Acid Formation 2 - 1030 0 - 1800 Methane Fermentation 6 - 430 32 - 1850 Final Maturation 6 - 430 420 - 580

25 Removal of ammonia-nitrogen from leachate to low levels is necessary because of

its aquatic toxicity and oxygen demand in receiving waters. Several researchers have

conducted tests to measure the toxicity of leachate, concluding that ammonia-nitrogen

significantly contributed to the toxic nature of the leachate (Barlaz et al., 2002; Kjeldsen

et al., 2002; Ward et al., 2002). In landfill leachate, the vast majority of the ammonia-

+ nitrogen species will be in the form of the ammonium ion (NH4 ) because pH levels are generally less than 8.0 (Read et al., 2001; Reinhart et al., 2002; US Environmental

Protection Agency and Waste Management, 2003). Figure 2-1 provides the distribution of ammonia and ammonium as a function of pH. Dissolved unionized ammonia

(predominant at pH levels above 10) is more toxic to anaerobic degradation processes

than ammonium ions but should not be present in significant concentrations in a landfill.

Ammonia-nitrogen concentrations greater than 500 mg-N/L are inhibitory to the

degradation process (Lay et al., 1997). Ammonium concentrations between 50 and 200

mg/L have been shown to be beneficial to anaerobic degradation processes in wastewater

treatment, while ammonium concentrations between 200 and 1,000 mg/L have been

shown to have no adverse effect. Concentrations ranging from 1,500 to 5,500 mg/L have

been shown to have inhibitory effects at higher pH levels and concentrations above 5,800

mg/L have been shown to be toxic to some microorganisms (Lay et al., 1997). However,

the effect of ammonium concentrations on landfilled waste degradation has not been

reported.

26 100 s ie

c 80 e p s

l ammonium

a 60 c

a ammonia i n 40 mmo a

l 20 a t o t f 0

% o 468101214 solution pH

Figure 2-1. Dominant form of ammoniacal nitrogen in solution at 25oC at various pH levels.

Nitrogen Transformation and Removal Processes

Currently, ammonia-nitrogen is treated in leachate ex-situ to the landfill (Cheung et al., 1997; Hoilijoki et al., 2000; Ilies and Mavinic, 2001; Marttinen et al., 2002;

Shiskowski and Mavinic, 1998; Welander et al., 1997). Ammonia-nitrogen removal methods often include complex sequences of physical, chemical, and/or biological processes, including chemical precipitation, nanofiltration, air-stripping, and biological nitrification/denitrification via various reactor configurations (i.e. rotating biological filters, suspended and attached growth reactors). However, operating the landfill as a bioreactor provides opportunities for in-situ nitrogen transformation and removal processes. Little research has been conducted evaluating the fate of nitrogen in bioreactor landfills; however, understanding the possible nitrogen transformations is important when considering potential leachate management options. When adding air to landfills,

27 biological processes such as nitrification traditionally found and expected only in landfill cover soils as a result of air diffusion may now occur within the waste mass.

Additionally, recirculating nitrified leachate allows for denitrification processes to occur in anoxic areas found in both anaerobic and aerobic bioreactor landfills. Figure 2-2 illustrates the potential nitrogen transformation and/or removal pathways that may occur in bioreactor landfills.

The heterogeneous nature of solid waste complicates the nitrogen cycle in bioreactor landfills. Because the waste is heterogeneous, portions of the landfill may contain different amounts of nutrients, be at different temperatures, have different moisture levels, and may be at different ORPs. Environmental conditions greatly affect the transformation and removal of nitrogen. Thus, within one landfill cell, there may be many nitrogen transformation processes occurring simultaneously or sequentially.

Processes commonly found in wastewater treatment processes and in soils, such as ammonification, sorption, volatilization, nitrification, denitrification, anaerobic ammonium oxidation (ANAMMOX) and nitrate reduction, may all occur in bioreactor landfills. This section discusses how the nitrogen transformation and removal processes found in wastewater and/or soils may also occur in bioreactor landfills based on the current knowledge associated with each process.

28 Organic Nitrogen N2 COD - N2 SO4 HS or H2S Reduction (depending on pH) COD SO4

Sulfide Ammonification Oxidation O Heterotrophic Autotrophic 2 Denitrification Denitrification

NO3 Reduction

O2 NO Nitrification 3 NH4-N NO2

Volatilization ANAMMOX Partial Heterotrophic Denitrification

COD Free N2 NH3

N2 N2O NO

Figure 2-2. The potential pathways of nitrogen transformation and/or removal in bioreactor landfills.

29 Ammonification

Proteins present in the waste are the major source of ammonia-nitrogen. This conversion of organic nitrogen to ammonia-nitrogen by heterotrophic bacteria is termed

ammonification. Ammonification is a two-step process consisting of the enzymatic

hydrolysis of proteins by aerobic and anaerobic microorganisms releasing amino acids

and the subsequent deamination or fermentation (depending on aerobic vs. anaerobic

conditions) of the acids to carbon dioxide, ammonia-nitrogen, and volatile fatty acids

(Burton and Watson-Craik, 1998). During deamination, amine groups are liberated to

form ammonia or ammonium, depending on the pH, and alkalinity is slightly elevated

(Burton and Watson-Craik, 1998). The deamination process is illustrated in Figure 2-3.

Once ammonification occurs, the ammonia-nitrogen is dissolved in the leachate and is

ready to be transformed and/or removed via volatilization, sorption, or biological

processes when in an aerobic environment. The pH also increases during ammonification.

Any free ammonia that is present is highly reactive and has been found to combine with

organic matter (i.e. carboxyls, quinine hydroxls), making them more biodegradable

(Palmisano and Barlaz, 1996). Thus, in landfills, any ammonia that is produced within

the landfill may redissolve and react with organic matter before exiting the landfill.

Little research has been conducted evaluating the rate of ammonification in

landfills. However, ammonia-nitrogen release from wastes has been evaluated in both

solid waste digestion and composting studies (de Laclos et al., 1997; Schwab et al.,

1994). Ammonification rates were not quantified, although the generation trends appear

to follow first-order reaction kinetics. Additionally, ammonification occurs during the

organic hydrolysis phase of landfill stabilization, which is also often represented by first-

30 order kinetics (Al-Yousfi, 1992; Halvadakis, 1983). In compost, ammonification has been found to be optimal between 40 and 50oC (Ross and Harris, 1982).

H

Amino

H C COOH Acid

NH 2

H

H C COOH

Deaminating Bacteria + NH (ex. Citrobacter) 4

NH - NH3 2 H+

Figure 2-3. The deamination process.

Ammonium Flushing

The mass of ammonia-nitrogen that can be leached from the waste is controlled by the volume of water passed through the landfill, the nitrogen content of the waste, and the ammonia-nitrogen concentration in the bulk liquid. Reducing ammonia-nitrogen concentrations by washout and dilution to acceptable levels within a landfill requires the addition of large volumes of water. The Institute of Waste Management Sustainable

Landfill Working Group (1999) reported that at a solid waste moisture content of 30%

31 (wet weight basis) and an initial liquid-phase ammonia-nitrogen concentration of 5,833

mg/L as N, a flushing volume of approximately 2.4 m3/tonne of waste was necessary to reduce the nitrogen concentration to 2 mg N/L. It was also noted that other studies had been conducted suggesting that flushing volumes between 5 and 7.5 m3/tonne of waste were needed to adequately reduce nitrogen concentrations in the landfill. No time frames for this reduction to occur were given. The effectiveness of flushing will be dependent on hydraulic conductivity of the waste, as it will be harder to introduce liquid in areas of lower permeability. As the hydraulic conductivity decreases, the time required for leaching to occur increases, as does the ammonification process.

Purcell et al. (1999) conducted a laboratory-scale study evaluating the flushing of ammonia-nitrogen from landfills. It was found that as flushing rates increased from 435 mm/yr to 2195 mm/yr, the release of ammonium-nitrogen from the waste and thus overall removal from the reactors increased. The main mechanisms of ammonia-nitrogen removal were found to be washout and dilution from the incoming water.

Flushing results in the removal of ammonia-nitrogen from landfills by adding large volumes of water, which must be treated externally. When operating the landfill as a bioreactor, leachate is recycled, and hence ammonia-nitrogen is continually reintroduced to the landfill while additional ammonia is solubilized into the leachate.

Ammonium Sorption

Sorption of ammonia-nitrogen to waste may be significant in bioreactor landfills

because of the high ammonium concentrations present. Ammonium is known to sorb

onto various inorganic and organic compounds (Laima, 1994). The amount of ammonium

32 sorbed on some organics has been reported to exceed the mass found in the bulk liquid

(Nielson, 1996). Sorption of ammonium to the waste will allow for temporary storage of

ammonium prior to it being used in other processes, such as nitrification and

volatilization, and may also result in the slow dissolution of ammonium over time

(Heavey, 2003).

Sorption is dependent on pH, temperature, ammonium concentration and ionic

strength of the bulk liquid. For ammonia to sorb to waste particles, it must be in the form

+ of ammonium (NH4 ). At pH levels expected in a landfill, the dominant form of the ammonia species is the ammonium ion (Metcalf and Eddy, 1991), as depicted in Figure

2-1. As ionic strength of the bulk liquid increases, sorption of ammonium tends to decrease (Heavey, 2003; Nielson, 1996) due to ion exchange effects. The sorbed ammonium is released and exchanged with other ions present in the bulk liquid, especially those with higher selectivity or concentration. A common procedure used to extract sorbed ammonium from solid particles involves the addition of a sodium or potassium sulfate solution. The sodium or potassium ions exchange with the ammonium, allowing for the ammonium to desorb from the waste. The conductivity of landfill leachate is generally high (approximately 7,000 umhos/cm) (Kjeldsen et al., 2002) and thus may influence ammonium sorption. The effect of the ionic strength in leachate on the sorption of ammonium needs to be evaluated.

In marine sediments, sorption of ammonium has been shown to follow a

Langmuir-type isotherm (Laima, 1994). However, Nielson (1996) conducted a study of ammonium sorption to activated sludge and found the data did not fit well to that isotherm type. Little work has been conducted evaluating ammonium desorption, which

33 is important to predict the amount of ammonium that will be available for treatment over time (Nielson, 1996). Nielson (1996) found that a portion of sorbed ammonium on activated sludge flocs was oxidized and used during the nitrification processes, however, a significant portion of the sorbed ammonium remained unoxidized, even when the ammonium in the bulk water was almost completely removed via nitrification.

Ammonium desorption kinetics may be dependent on ammonium removal in the bulk liquid; as the ammonium concentration in the bulk liquid decreases, potentially due to flushing or other removal processes, ammonium is likely to be desorbed from the waste to regain equilibrium (Heavey, 2003). Ionic strength affects were not evaluated in this study, however, they were noted to potentially impact the sorption and desorption properties of ammonium.

Ammonium sorption in soils has also been measured. Kwok and Loh (2003) conducted a laboratory-scale study evaluating the cation exchange capacity of different soil types in Singapore. In each isotherm study, ammonium sorption followed a

Freundlich isotherm; sorption increased with increasing exchange capacity. Van

Raaphorst and Malchaert (1996) conducted ammonium sorption studies on various sea sediments and found the sorption data to also follow a Freundlich isotherm. Additionally,

Van Raaphorst and Malchaert (1996) conducted a study in which they measured the mass of ammonium that could be extracted from a sediment using a potassium chloride solution over a 40-hour time period. They found that exchange of ammonium was initially rapid (during first 10 hours), but declined significantly after 10 hours. During the

40-hour test, not all of the ammonium was recovered, suggesting that some of it was tightly bound to the soil particles.

34 Studies evaluating the sorptive capacity of fresh waste have not been conducted,

however, the sorptive capacity of peat and soil has been studied. Heavey (2003) found

+ that peat (which may simulate well oxidized waste) could sorb 18 to 27 mg NH4 /g dry peat. It seems probable that more sorption occurs in older solid waste than in younger waste because older waste has a smaller particle size and thus a larger surface area yielding more available reactive sites for sorption. Additionally, older waste contains more recalcitrant organic particles (predominantly humic and fulvic acids) to which ammonium may sorb. Further, as waste ages, there may be changes in the surface charges of the waste, resulting in higher levels of sorption. The presence of complex organics has been shown to influence ammonium sorption; the ammonium ions may fix irreversibly to these molecules. He et al. (1998) found that approximately 15% of the radiolabeled ammonium they added to soil samples was associated with the humic fractions, however, the ammonium was recoverable using a series of several different types of extraction media. Reinhart (1989) conducted long-term desorption isotherms with various organic pollutants and MSW and found that as time increased, the mass of sorbed compound increased. It was suggested that sorption may be time-dependent; the compound may sorb deeper into the solid-phase over time, ultimately allowing for sorption of more mass over time. Similar phenomenon of irreversible sorption in soils has been observed (Brusseau and Rao, 1989; Morrissey and Grismer, 1999; Van Der Zee and Van Riemsdijk, 1986) and will likely occur with ammonium. More research on the sorption and desorption of ammonia-nitrogen on MSW is necessary.

35 Volatilization

In conventional landfills, ammonia makes up approximately 0.1 to 1.0% (dry volume basis) of landfill gas exiting the landfill (Tchobanoglous et al., 1993). Ammonia is not a greenhouse gas, so its impact on the environment is not as harmful as methane, however, there are some adverse health effects that may result from exposure to the gas.

Ammonia has a pungent odor and is a respiratory tract irritant. Also, ammonia gas can dissolve in the moisture on skin and form ammonium hydroxide, a corrosive chemical that can cause skin irritation (Matheson TriGas, 2002).

Volatilization only occurs when free ammonia is present. At pH levels above 10.5 to 11.5, the majority of the ammonia-nitrogen present in solution is in the form of free ammonia gas (NH3), as depicted in Figure 2-1. The free ammonia concentration at a particular pH level may be computed via equation 2.1.

+ pH [NH4 - N] *10 []NH - N = (2.1) 3 K a +10 pH Kw

+ where NH3-N is the free ammonia concentration, mass/volume, NH4 -N is the ammonium concentration, mass/volume, Ka is the acid dissociation constant, and Kw is the water ionization fraction (10-14).

As temperature increases, more of the ammonia is converted to free ammonia gas because of the temperature dependence of the acid dissociation constant. At a pH level of

7, under standard conditions (i.e. temperature is 25oC and pressure is 1atm), 0.56% of ammonia present is in the form of free ammonia. When the temperature increases to

36 60oC, a temperature commonly found in aerobic landfills, the percentage of free ammonia present at pH 7 increases to 4.90%. Ammonia volatilization has been measured in numerous compost studies. Results have shown that as temperature increases, the dominant ammonia removal mechanism becomes volatilization. Sanchez-Monedero et al.

(2001) found that at temperatures above 40oC, the only ammonia removal mechanism observed in compost was volatilization. Tiquia and Tam (2000) also found that at temperatures above 40oC and at pH levels of 7 and above, the majority of nitrogen removed from compost is via volatilization.

Airflow also plays an important role in ammonia-nitrogen volatilization. As air is introduced, it begins to agitate the leachate, creating a removal pathway for dissolved free ammonia to volatilize and leave the landfill. Airflow also dilutes the concentration of gas-phase ammonia-nitrogen above the leachate, increasing the driving force for dissolved ammonia-nitrogen to partition to the gaseous phase (Henry et al., 1999;

Thomas, 1982).

Ritzkowski and Stegmann (2003) conducted a laboratory-scale study in which the mass of ammonia-nitrogen volatilized from the waste mass was measured. All gas emissions from a simulated aerobic bioreactor landfill exited through an acid scrubber to capture any ammonia-nitrogen that may have been volatilized. It was found that at a pH of 7.4 and a temperature of 35oC, 50% of the ammonia-nitrogen initially present in the leachate was volatilized. The air flowrate was not reported.

37 Nitrification

Nitrification has been successfully used in wastewater treatment processes as a means to convert ammonium-nitrogen to nitrite and nitrate for decades and the mechanisms in which it is conducted and operated have been thoroughly studied

(Abeliovich, 1985; Gupta and Sharma, 1996; Schmidt et al., 2003; Van Loosdrecht and

Jetten, 1998). The purpose of this section is not to thoroughly review the nitrification process, but rather to discuss how nitrification may occur in bioreactor landfills. More detailed information about nitrification can be found elsewhere (Grady et al., 1999;

Schmidt et al., 2003).

Nitrification is a two-step aerobic process in which ammonia-nitrogen/ammonium is microbially oxidized to nitrite and nitrate via obligate aerobe, autotrophic, chemolithotrophic microorganisms. Because nitrification is an aerobic process, it is almost non-existent in conventional landfills and in bioreactor landfills in which air is not added. In those systems, nitrification is restricted to upper portions of the landfill or the cover where air may infiltrate (Burton and Watson-Craik, 1998). In landfills in which air is purposely added, nitrification can be a significant nitrogen removal pathway.

During the first step of nitrification, Nitrosomonas bacteria oxidize ammonia- nitrogen to nitrite, according to the following reaction (Rittman and McCarty, 2001):

+ - + NH4 + 1.5 O2 Æ NO2 + 2 H + H2O (2.2)

38 The second step of the nitrification process is the oxidation of nitrite to nitrate by

Nitrobacter bacteria (or the more recently implicated Nitrospira) according to the following reaction (Rittman and McCarty, 2001):

- - NO2 + 0.50 O2 Æ NO3 (2.3)

Nitrifiers must fix and reduce inorganic carbon to use as their carbon source

(Rittman and McCarty, 2001), resulting in low cell yields and thus small maximum specific growth rates. Additionally, nitrification results in the consumption of alkalinity as nitrous acid is formed. The first step of nitrification is often the limiting step, as the

Nitrosomonas bacteria grow more slowly than Nitrobacter or Nitrospira (Grady et al.,

1999). Some heterotrophic microorganisms are able to nitrify, however, their specific nitrifying rates are considered generally three to four orders of magnitude lower than that of the autotrophs (Gupta, 1997; Schmidt et al., 2003). Thus, heterotrophic nitrification is generally considered to be a minor pathway. Some of the heterotrophic nitrifiers are able to denitrify (reduce nitrate) aerobically as well.

Nitrification has also been documented to naturally occur in soils (Bengtsson et al., 2003; Burger and Jackson, 2003; Vestgarden and Kjonaas, 2003). Nitrification processes in soil generally result from the addition of nitrogen fertilizers and the diffusion of oxygen.

Nitrification may occur in bioreactor landfills in which air is added. Although the metabolic processes associated with nitrification may be essentially the same in landfills and wastewater treatment processes, the operation, control, and potential extent of such processes is not the same. Nitrification in landfill environments is complicated by oxygen

39 and temperature limitations, heterotrophic bacteria competition, and potentially pH inhibition. Oxygen is a required element for nitrification. Adding air to a landfill would be dual-purpose: to nitrify, removing the ammonia-nitrogen, and to enhance the degradation of solid waste. However, maintaining and controlling sufficient oxygen levels within the landfill, especially considering the heterogeneous nature of solid waste and the high temperatures characteristic of aerobic landfills, may be difficult and may result in oxygen limitations (dissolved oxygen concentration declines with temperature increases) and thus reduced nitrification rates. Additionally, oxygen may become limiting to nitrifiers in areas within the landfill containing large amounts of organic carbon (newly placed waste) due to competition with heterotrophs. Under oxygen-limiting conditions, autotrophic ammonia-oxidizing bacteria may produce nitric and nitrous oxides, which would be a distinct disadvantage of this technique as they are potent greenhouse gases

(Burton and Watson-Craik, 1998). Heterotrophic nitrifiers are also capable of producing nitrous oxide.

Cheng et al. (2004) measured the production of both nitric and nitrous oxides in

Chinese agricultural soils in which high levels of fertilizer were added. Different types of soils were tested to determine which conditions resulted in higher gas production. Both nitric and nitrous oxide production from nitrification was observed. Production could be correlated with the pH of the system; soils that were more basic (pH > 8) resulted in the highest concentrations of nitrous oxide, while the more acidic soils produced the least.

Khalil et al. (2004) also conducted a study evaluating the production of nitrous oxide in soils, paying particular attention to the influence of oxygen on nitrous oxide production.

They found that as oxygen decreased, the mass of nitrous oxide from nitrification

40 increased. In landfills, there may be areas in which oxygen concentrations are limiting, thus, nitrous oxide production via nitrification may result. However, long residence times are expected, so the nitrous oxide may be converted to nitrogen gas before exiting the landfill.

When air is added to landfills, in-situ temperatures generally increase, often as high as 55o to 66oC (American Technologies, 1997; Merz and Stone, 1970), which is a temperature range potentially inhibitory to nitrification (Lubkowitz-Bailey and Steidel,

1999; Murthy et al., 2000; Willers et al., 1998). Willers et al. (1998) reported that pure

Nitrosomonas cultures have a thermal death point between 54o and 58oC. In landfills, there may be pockets of lower temperatures, allowing for the nitrifiers to be protected.

Additionally, nitrifiers that may be present within biofilms on waste particles may be temporarily protected from high temperatures. At these high temperature levels, volatilization may become the predominant ammonia-nitrogen removal mechanism.

Sanchez-Monedero et al. (2001) completed studies evaluating the dynamics of nitrogen transformations during organic waste composting. They reported that nitrification did not occur when temperatures rose above 40oC. Several studies evaluating nitrification in thermophilic wastewater processes have been conducted (Juteau et al., 2004; Lubkowitz-

Bailey and Steidel, 1999; Murthy et al., 2000). Juteau et al. (2004) found that nitrification did not occur under thermophilic conditions. However, Lubkowitz-Baily and

Steidel (1999) and Willers et al. (1998) found that nitrification was achievable at temperatures as high as 44oC in wastewater and 50oC in veal-calf slurry, respectively, although the rate of nitrification was decreased significantly at both temperature levels. In higher temperature environments, other types of bacteria may be responsible for

41 conversion of ammonium to nitrite (Mevel and Prieur, 1998). Methanotrophs have been shown to oxidize ammonium to nitrite under thermophilic conditions (53oC), however, nitrification by the methanotrophs was highly dependent on oxygen and methane concentrations; at methane concentrations above 84uM, nitrification was inhibited

(Mevel and Prieur, 1998). In hydrothermal vents, thermophilic heterotrophic nitrifiers have been isolated and found to convert ammonium to nitrite at temperatures as high as

65oC, thus conversion of ammonium to nitrite at high temperatures is possible (Mevel and Prieur, 1998). Heterotrophic nitrifiers generally have lower ammonium conversion rates than autotrophic nitrifiers, but in environments in which autotrophic processes are inhibited, heterotrophic processes may occur and be the dominant nitrogen conversion process.

It is suspected that in-situ nitrification may be optimized when operated in landfill cells containing older waste, because, as in composting, as the age of the waste increases, the temperature of the system decreases due to reduced biological activity

(Tchobanoglous et al., 1993; Vesilind et al., 2002). Additionally, since older waste contains fewer biodegradable organics, less competition with heterotrophs for oxygen will occur. Sanchez-Monedero et al. (2001) also reported that nitrification did not occur in compost processes until the majority of the organic matter was degraded, something also seen in wastewater treatment processes (Carrera et al., 2004; Lee et al., 2002).

Additionally, in older waste, more recalcitrant organics, such as humic acids, are present.

In leachate collected during the methanogenic stage of degradation, almost 60% of the dissolved organics present were in the form of high molecular weight compounds (i.e. humic and fulvic acids) (Kjeldsen et al., 2002). Humic acid has been shown to inhibit

42 nitrification, resulting in the build-up of nitrite concentrations. Bazin et al. (1991) conducted a study in which humic acid was added to columns containing glass beads and pure cultures of nitrifying microorganisms. At input levels of 100 micrograms/cm3, the humic acid additions had no adverse effect on nitrification rates and aided in buffering the pH of the system. However, when humic acid was added at rates above that level, nitrification was inhibited. The mechanism of inhibition was not stated. It is suspected that in landfills humic acids may affect nitrification, although more work needs to be conducted evaluating the extent of such effects.

pH may also be a complication during nitrification processes in landfills. The pH of leachate in aerobic landfills is generally near neutral, or slightly above (Read et al.,

2001; Stessel and Murphy, 1992). The alkalinity of leachate is generally in the range of

1,000 to 10,000 mg/L as calcium carbonate (Tchobanoglous et al., 1993). Because nitrification destroys alkalinity, there may not be sufficient alkalinity present to buffer pH changes that would result from nitrification of high ammonia-nitrogen leachates. It is possible that alkalinity may need to be added to the landfill to buffer the leachate.

Nitrification Case Studies in Landfills

Several researchers have evaluated the potential use of in-situ, or partially in-situ, nitrification processes in landfills. Youcai et al. (2002) conducted a study in which a biofilter consisting of old waste (8 to 10 years old) was used to treat leachate. Aerobic portions existed at the top and bottom of the system (air was not supplied, rather was drawn in from the atmosphere via convection), while the middle of the system was anaerobic. It is important to note these conditions (aerobic and anaerobic) were never

43 shown experimentally nor was the ORP measured. A removal of 99.5% of the ammonia- nitrogen in leachate was observed. Elevated concentrations of nitrate and nitrite were measured, indicating the ammonia-nitrogen was converted biologically. Additionally, 20-

30% of total nitrogen in the leachate was removed suggesting in-situ nitrification and denitrification occurred sequentially in the landfill.

Incidental treatment of nitrogen in aerobic or semi-aerobic landfills has also been observed. Hanashima (1999) described lysimeters operated under aerobic and semi- aerobic conditions over a three to 20-year period. Aerobic test cells were continuously supplied with air via a feed pipe to the bottom of the cell. The semi-aerobic cell was constructed with a large drainage pipe in contact with the atmosphere to provide aeration to the bottom of the cell while maintaining the upper portion of the landfill under anaerobic conditions. Leachate was recycled to both cells. Comparison with the performance of conventional anaerobic cells suggested that nitrogen removal under both semi-aerobic and aerobic conditions was significantly greater than under anaerobic conditions.

The most efficient method evaluated to date is complete in-situ removal of nitrogen using dedicated zones. Onay and Pohland (1998) completed an in-situ nitrification/denitrification laboratory study in which a three-component system was used to facilitate the process. A laboratory study was conducted to evaluate a conceptual idea of an anoxic denitrification zone located near the surface of the reactor, an anaerobic zone to simulate methanogenic conditions in the middle, and at the bottom, an aerobic nitrification zone. When utilizing leachate recirculation among the zones, approximately

95% nitrogen removal was achieved. Onay and Pohland (1998) completed another study

44 during which the reactors were connected in series, but with no leachate recycle, just a single pass. Nitrogen removal was observed with this set of experiments as well, however only 30 to 52% removal of nitrogen in the leachate was achieved. Onay and

Pohland (1998) suggested application of this type of system in the field by having different portions of the landfill serving as treatment zones; the upper portion of the landfill would be anoxic, the middle anaerobic, and the bottom aerobic (air naturally added via convection through leachate collection pipes).

Nitrification Kinetics

Traditional nitrification kinetics in wastewater systems are derived from the net growth rates of both Nitrosomonas and Nitrobacter, with the growth rate of

Nitrosomonas considered as the rate-limiting step and thus the most critical from a design perspective. Monod kinetics are often used, as they describe first-order substrate-limiting growth at low ammonia-nitrogen concentrations and zero-order at higher concentrations

(Grady et al., 1999; Rittman and McCarty, 2001). Because ammonia oxidation is the rate- limiting step, it is often used as the overall rate of nitrification. Several environmental factors influence the rate and must be accounted for in the rate expression, including pH, dissolved oxygen (DO) concentrations, and temperature. These factors are included in the rate expression of ammonia oxidation in a multiplicative Monod manner (de Renzo,

1978). The Monod relationship can also be modified to account for substrate inhibition, which could be relevant at high ammonia-nitrogen concentrations.

The nitrification process in solid waste environments may be better approximated by fixed-film theory rather than suspended, as the waste may act as an attachment surface

45 for the microorganisms (Palmisano and Barlaz, 1996; Senior, 1995). In fact, a bioreactor landfill may contain both suspended and fixed-film populations, but it seems likely that in most cases the greater portion of the biomass will be associated with biofilms. This means that diffusion of electron acceptors and donors and other mass transfer limitations become significant. In landfills, mass transfer of ammonium and/or oxygen may be a bigger factor than in wastewater treatment because of the large particle sizes of the waste and because the liquid to solid ratio is much smaller than in typical wastewater treatment processes. Mass transfer limitations would likely become apparent in the value of the half-saturation constant in the Monod model (Pohland, 1992). The half-saturation constants in wastewater for nitrification are generally 1 to 2mg/L N; a much larger value may indicate mass transfer limitations. In addition, the presence of biofilms increases the possibility of multiple micro-environments (e.g. even an aerobic region may contain biofilms with anoxic depths and thus possibly simultaneous denitrification). Thus, it is unlikely the kinetics of in-situ nitrification will fit well to strict Monod or biofilm kinetic models, rather an expression including both types of consortia may be appropriate.

Denitrification

Denitrification has been applied in many wastewater treatment processes. The intent of this section is not to review the denitrification process, rather to discuss how denitrification may occur in bioreactor landfills. Information regarding denitrification processes may be found elsewhere (Biesterfeld et al., 2003; Ekama and Wentzel, 1999;

Etchebehere et al., 2001; Foglar and Briski, 2003; Grady et al., 1999; Mateju et al., 1992;

Schmidt et al., 2003). In-situ denitrification is also complicated in solid waste systems,

46 although it may be easier to implement than nitrification. Denitrifiers are more robust than nitrifiers, however they require a sufficient organic carbon source for high nitrate removal rates. Because of the carbon needs, denitrification may occur most efficiently in young waste, rather than in older, partially oxidized waste. Price et al. (2003) evaluated the potential need for an external carbon source in the laboratory and noted that a fresh layer of refuse contained sufficient carbon to stimulate significant nitrate consumption. If a sufficient organic carbon source is not readily available, partial denitrification may occur which may lead to the production of harmful intermediates (N2O and NO), which are potent greenhouse gases (Cheng et al., 2004; Khalil et al., 2004).

Typically, in-situ denitrification occurs in anoxic bioreactor landfills. However, because of the potential for anoxic pockets to be present in aerobic systems, denitrification may also occur in portions of aerobic bioreactor landfills that air does not reach.

Heterotrophic Denitrification

Denitrification is an anoxic process that reduces nitrate to nitrite, nitric oxide, nitrous oxide, and finally nitrogen gas, as shown in reactions 2.4 – 2.7 (Rittman and

McCarty, 2001):

- - + - NO3 + 2 e + 2 H Æ NO2 + H2O (2.4)

- - + NO2 + e + 2 H Æ NO + H2O (2.5)

- + 2 NO + 2 e + 2 H Æ N2O + H2O (2.6)

47 - + N2O + 2 e + 2 H Æ N2(g) + H2O (2.7)

Typically, denitrifying bacteria are heterotrophic, facultative aerobes, which use nitrate as an electron acceptor when oxygen is absent or limiting. A potential advantage of denitrification is the simultaneous carbon and nitrate destruction without requiring oxygen input (Grady et al., 1999). Denitrification also recovers half of the alkalinity consumed during nitrification. It is important to note that processes in which nitrate is used as a terminal electron acceptor are energetically favored over acetogenic, sulfate reduction, and methanogenic processes. Thus in landfills in anaerobic/anoxic environments in which nitrate reduction occurs, inhibition of such processes may occur.

Researchers have evaluated in-situ, or partially in-situ, denitrification at both laboratory- and field-scale. Burton and Watson-Craik (1999) operated a landfill test cell designed to denitrify externally nitrified leachate. Nitrate returned to the landfill cell was efficiently consumed under the anoxic/anaerobic landfill conditions, confirmed using labeled isotopic nitrate. Both Waste Management (2003) and Aljarallah and Atwater

(2000) have completed similar studies at field and laboratory-scale, respectively.

Aljarallah and Atwater (2000) noted that denitrification was feasible in a bioreactor landfill, however methane production and waste degradation were hindered. A carbon balance was conducted on the leachate and solid waste in their study and found that as the nitrate concentration increased, less carbon was released in either the liquid or gas form, suggesting that waste degradation was inhibited by high nitrate concentrations (i.e. 800 mg/L-N). Additionally, it was noted that poor leachate quality was produced (high organic strength). Jokela et al. (2002) conducted a similar laboratory study demonstrating

48 that in-situ denitrification is possible and can result in the elimination of nitrogen.

Ammonia was detected in the effluent from the solid waste column, which was attributed to either release from the waste or high leachate COD to nitrate ratio, which may promote the reduction of nitrate to ammonia-nitrogen (see Figure 2-1). It was also concluded that at an oxidized nitrogen loading rate below 3.8 g N/total solids-day, methanogenesis was not inhibited. High leachate COD concentrations inhibited nitrification in the ex-situ process, presumably due to competition for available oxygen.

Price et al. (2003) also conducted studies evaluating the ability of older waste

(with low organic carbon) to denitrify nitrified leachate. It was shown that the landfill does have the capacity to denitrify, as significant nitrate consumption was observed, and that fresh waste contained enough organic carbon to support denitrification, while older waste required the addition of an external carbon source. Additionally, it was observed that methanogenic activity was inhibited during denitrification, but quickly resumed following nitrate removal.

Autotrophic Denitrification

Nitrate removal in wastewaters containing high sulfur concentrations or reduced sulfur sources, such as hydrogen sulfide, may occur via autotrophic denitrification.

2- Thiobacillus denitrificans use an inorganic sulfur source (i.e. H2S, S, SO3 ) rather than an organic carbon source when reducing nitrate to nitrogen gas (Onay and Pohland, 2001) according to reaction 2.8.

- - + 2- 2 NO3 + 1.25 HS + 0.75 H Æ N2 + 1.25 SO4 + H2O (2.8)

49

This nitrate removal mechanism produces sulfate. At low carbon to nitrogen ratios this removal mechanism is favored over heterotrophic Denitrification (Koenig and Lui, 1996).

Autotrophic denitrification may occur in landfills, especially in older landfills or older portions of landfills where the carbon to nitrogen ratio may be low. The increased sulfate concentrations may have an adverse effect on methane production rates by limiting the amount of organic carbon available to the methanogens due to competition with sulfidogens.

While operating their reactors, Onay and Pohland (2001) observed the presence of autotrophic denitrification. To confirm their findings, a spike of nitrate was added and gas samples from the headspace of the reactor were measured for nitrogen and hydrogen sulfide. It was found that 13 days after the nitrate spike, the hydrogen sulfide present in the gas-phase disappeared. After the nitrate source was exhausted, the sulfate was converted back to hydrogen sulfide. Onay and Pohland (2001) concluded that autotrophic denitrification accounted for between 15% and 55% of the nitrate conversion to nitrogen gas, with the variation being attributed to the mass of organics present in the system.

Additionally, it was stated that autotrophic denitrification is advantageous, as it converts nitrate to nitrogen gas in the absence of an organic carbon source and can utilize inorganic sulfur compounds. High sulfate concentrations (increased to approximately 350 mg/L sulfate) were produced, however, the impact of sulfate on methanogenesis was not quantified.

50 Denitrification Kinetics

Traditionally, Monod kinetics are used to describe denitrification in wastewater systems. The nitrate removal rate is dependant on several factors that must be accounted for in the rate expression. Because an organic carbon source is desirable for rapid denitrification, the amount present in the system affects the rate, as does the biodegradability of the carbon source. Additionally, pH and dissolved oxygen (DO) levels affect the denitrification rate and can be accounted for in a Monod expression in a multiplicative manner.

As in nitrification, the denitrification process in solid waste may be better approximated by fixed-film theory rather than suspended, as the waste may act as an attachment surface for the microorganisms (Palmisano and Barlaz, 1996; Senior, 1995).

Mass transfer effects may also be severe in denitrification processes and may be reflected in higher half-saturation values when fitting the data to the Monod model (Pohland,

1992). It is unlikely the kinetics of in-situ denitrification will fit well to either strict

Monod or biofilm kinetic models, rather an expression combining both types of consortia may be appropriate.

ANAMMOX

Biological oxidation of ammonia-nitrogen may also occur under anaerobic conditions and is termed the ANAMMOX process (ANaerobic AMMonium OXidation).

Bacteria capable of ANAMMOX use ammonium as the electron donor and nitrite as the electron acceptor, as shown in reaction 2.9 (Jetten et al., 1998; Jetten et al., 2001):

51 + - + + NH4 + 1.26 NO2 + 0.085 CO2 + 0.02 H Æ N2 + 0.017 H

- + 0.24 NO3 + 1.95 H2O (2.9)

There has been little research concerning ANAMMOX in solid waste environments, however, studies conducted in wastewater have shown that ANAMMOX readily occurs

(Hao et al., 2002; Jetten et al., 1998; Jetten et al., 2001; Schmidt et al., 2003).

Researchers have determined that the microorganisms most often responsible for the

ANAMMOX process are from the Planctomycetales group (Jetten et al., 1998; Ye and

Thomas, 2001). This process is generally favorable in environments in which retention time is long, operation is stable, nitrite is present and electron donors that would cause nitrite reduction via denitrification are absent. Because of the potential for anaerobic regions located within an aerobic landfill, this biological ammonia-nitrogen removal mechanism may incidentally occur simultaneously with nitrification. However, the growth rates of the ANAMMOX bacteria are extremely slow, thus ammonia-nitrogen removal is slow as well. It is questionable whether or not the ANAMMOX microorganisms will be able to compete with denitrifiers for nitrate and nitrite within landfills (Burton and Watson-Craik, 1998). Removal rates have been shown to be less than half that of aerobic nitrification (Ye and Thomas, 2001).

Dissimilatory Nitrate Reduction to Ammonium

Dissimilatory nitrate reduction to ammonium (DNRA) in anaerobic or anoxic environments may also occur in landfills according to reaction 2.10.

52 - + + NO3 + 2 H + 4 H2 Æ NH4 + 3 H2O (2.10)

As shown, ammonium is produced as a result of nitrate reduction. This pathway is generally favored when the microbes are electron acceptor (nitrate) limited in high organic carbon environments (Price, 2001; Tiedje, 1988) and has been shown to occur readily in anaerobic digestion and anoxic sediments where the redox potential is low

(Tiedje, 1988). DNRA is favored over denitrification in anaerobic and anoxic environments in environments with a high COD to nitrate ratio because in an electron acceptor limiting environment it is more advantageous for the microorganisms to metabolize nitrate to ammonium and gain eight electrons per mole of nitrate than denitrify and only gain five electrons per mole of nitrate (Tiedje, 1988). In electron acceptor rich environments (higher COD to nitrate ratios), denitrification is usually the favored nitrate reduction process because the greatest need by the microorganisms is to gain energy. The microbes responsible for the DNRA process differ from denitrifiers in that they are generally fermentive (obligate anaerobes, facultative anaerobes, and aerobes), using nitrate as electron sink, rather than being respiratory and using nitrate as a terminal electron acceptor (Cole, 1990; Tiedje, 1988).

DNRA depends highly on redox conditions and the amount of labile carbon available (Bonin, 1995; Fazzolari et al., 1998; Yin et al., 2002). Yin et al. (2002) conducted experiments in Chinese and Australian paddy soils and found that the partitioning of nitrate that was reduced to ammonium and to that being denitrified was greatly dependent on the amount of labile carbon present, which was demonstrated by an increase in ammonium production with increasing carbon. Buresh and Patrick (1981) conducted an experiment on estuarine sediment and found that approximately 15% of the

53 nitrate was converted to ammonium at a redox potential of 0 mV. When decreasing the redox potential to -200 mV, approximately 35 to 42% of the nitrate was reduced to ammonium, while an increase in redox potential (300 mV) resulted in a significant decline of ammonium production, supporting the theory that DNRA is optimal in low redox environments. When nitrate is added to systems, a general increase in the redox potential occurs. If nitrate is added to environments with a sufficiently low redox potential, DNRA may be favored. However, if the nitrate addition results in an increase in redox above 0 mV, denitrification of the nitrate is more likely. Oxygen also impacts

DNRA, however, it is less sensitive to changes in oxygen than Denitrification (Fazzolari et al., 1998).

In anaerobic or anoxic areas within the bioreactor landfill in which low nitrate concentrations are present in areas containing young waste (high degradable organic carbon) and low redox potentials, DNRA may be favored over Denitrification (Tiedje,

1988). The dissimilatory nitrate reduction pathway is not desired because it results in an increase in ammonium concentration. However, this removal mechanism may be limited because of competition from the denitrifiers for nitrate. The nitrate reducing bacteria require a tenfold greater population than denitrifiers to reduce 50% of the nitrate (Price et al., 2003). Bonin (1995) reported a ratio of 1.8:1.0 denitrifiers to DRNA microbes are generally present in an environment. In landfills, there is generally adequate denitrifying populations naturally present to out-compete any DNRA capable microorganisms. Price et al. (2003) conducted laboratory studies in solid waste evaluating the denitrification capacity of the waste and found that there was no noticeable increase in ammonium due to DNRA. However, the redox potential of the laboratory reactors was not measured.

54 Because there had been several additions of nitrate to each reactor, it is possible the redox potential was high enough to inhibit DNRA activity.

Simultaneous Nitrogen Removal Processes

Simultaneous nitrification and denitrification has been observed in wastewater processes, particularly in trickling filter and other biofilm processes. Because the potential for anoxic pockets in aerobic landfills is high, simultaneous nitrification and denitrification may occur in aerobic bioreactor landfills. Pochana and Keller (1999) conducted experiments evaluating the factors that may affect simultaneous processes in activated sludge flocs. They determined that the most influential parameters are DO, particle size, and carbon source. Of particular interest is that as the floc size increases, the potential for anoxic zones around the particles increases due to oxygen flux limitations.

Solid waste particles are large compared to activated sludge flocs, thus the probability of oxygen flux limitations is high, supporting the likelihood of simultaneous processes.

Because landfills are heterogeneous and may support several different micro- environments simultaneously (i.e. aerobic, anaerobic, and anoxic), several combinations of nitrogen transformation processes mentioned may be present. In aerobic bioreactor landfills, it is possible that partial nitrification (only resulting in the production of nitrite) followed by either ANAMMOX or denitrification will occur naturally because of the heterogeneous nature of the in-situ environment. There will be portions of the landfill that are aerated well, some only partially aerated, and others not aerated at all. As leachate flows from one section of the landfill to another, it is possible that it will come into contact with aerobic, anoxic and anaerobic regions, leading to multiple nitrogen

55 transformation processes. For example, leachate ammonium may be converted only to nitrite before the leachate flows to an anaerobic pocket. In that anaerobic pocket, the nitrite may then be converted to nitrogen gas. The hydraulic conductivity of the landfill will be a factor, as the time during which the leachate remains in each type of environment will ultimately determine the extent of the reactions that may occur. The ability to predict which nitrogen transformations will occur allows for more strategic design and operation of bioreactor landfills.

Other Nitrate Processes

Nitrate may also have an abiotic fate in landfills. Leachate generally does not have high nitrate concentrations, however, nitrate may be present if nitrification has occurred in-situ, or if the leachate is nitrified externally and then reinjected to the landfill.

Nitrate sorption has been shown to occur in soils, although not to the extent that ammonium sorption has been observed. Kwok and Loh (2003) measured nitrate sorption in six different soils. Sorption was detected, but in small amounts (average for all soils was 0.004 mol/kg). Kowalenko and Yu (1995) also evaluated the sorption of nitrate on soils and found up to 34% of additional nitrate was removed from soils when performing an extraction using potassium chloride. Sorption of nitrate by waste is probable. Because of the large variability of waste types, there is bound to be particles with negative charges that would allow for anion exchange.

Another fate of nitrate is the abiotic transformation via iron. Davidson et al.

(2003) proposed a method of abiotic nitrate removal called the “ferrous wheel hypothesis” in which reduced iron (Fe(II)) abiotically converts nitrate to nitrite in

56 anaerobic environments; nitrite then reacts with the dissolved organic matter to produce dissolved organic nitrogen. No evidence of this occurring in compost or solid waste has been reported, however, because leachate typically contains large amounts of iron(II) (3 –

5500mg/L) (Kjeldsen et al., 2002) the possibility exists. Iron(II) has also been shown to reduce nitrate in basic solutions to ammonia-nitrogen. Fanning (2000) reported that a pH of 8 was optimal for the reduction, however, the reduction proceeded at lower pH levels, just at slower rates. Additionally, it was suggested that the reaction may be influenced or catalyzed by the presence of silver and copper. Silver is not generally found in leachates, but copper can be found at levels ranging from 0.005 to 10 mg/L (Kjeldsen et al., 2002).

Studies in acid forest soil have observed the disappearance of nitrate via an abiotic mechanism. Dail et al. (2001) conducted a radiolabeled study attempting to determine the fate of nitrate. They found that nitrate was incorporated into an insoluble organic nitrogen form in both live and sterile soils, suggesting abiotic fate attenuation.

Additionally, in their study, there was more attenuation of nitrate in soils with larger amounts of organic carbon, suggesting the abiotic conversion is related to the soil carbon content. Because of the large organic carbon content found in landfills, this nitrate transformation mechanism could easily occur.

Future Research Directions

An understanding of the fate of nitrogen and possible mechanisms for ammonia- nitrogen removal in bioreactor landfills may significantly increase the capability of bioreactor landfills to more completely treat leachate in-situ. Bioreactor landfills are currently one of the most advantageous methods available for solid waste management,

57 but still have significant undeveloped potential with respect to in-situ leachate and waste treatment. An understanding of the fate of nitrogen, and thus the ways in which nitrogen can be removed/treated allows for this undeveloped potential to be better developed.

Additionally, understanding the fate of nitrogen may aid in developing methods to remediate old landfills (Ritzkowski and Stegmann, 2003).

Little research has been conducted evaluating the potential processes of nitrogen transformation and removal in bioreactor landfills and is needed before an in-depth understanding of the processes can be achieved and used to optimize the operation of bioreactor landfills. Both laboratory- and full-scale studies should be completed to evaluate the hypothesized, but untested, nitrogen transformation processes. To date, no controlled full-scale studies purposely evaluating in-situ nitrification as a nitrogen transformation process have been conducted. Additionally, laboratory-scale studies need to be conducted to gain a better understanding of the rates and kinetics of the nitrogen transformation processes, as well as to develop design requirements for an in-situ nitrogen removal system to facilitate full-scale testing.

References

Abeliovich, A. (1985). Nitrification of ammonia in wastewater: Field observations and laboratory studies, Water Research 19(9), 1097-1099.

Al-Yousfi, A. B. (1992). Modeling of Leachate and Gas Production and Composition at Sanitary Landfills, Dissertation, University of Pittsburgh.

Aljarallah, R. S., and Atwater, J. (2000). Denitrification in Landfill Bioreactors. Waste Tech 2000, Orlando, Fl.

58 American Technologies, I. (1997). Final Report: Development of an Aerobic Landfill Bioreactor Operational Plan for the Columbia County Baker Place Road Landfill, Grovetown, Georgia. Southeastern Technologies, Inc., Grovetown, Georgia.

American Technologies, I. (1999). Final Report: Evaluation of Waste Placement Options for Optimum Aerobic Landfill System Performance. Southeastern Technologies.

Barlaz, M. A., Ham, R. K., and Schafer, D. M. (1990). Methane Production from Municipal Refuse: A Review of Enhancement Techniques and Microbial Dynamics, Critical Reviews in Environmental Control, 19(6), 557.

Barlaz, M. A., Rooker, A. P., Kjeldsen, P., Gabr, M. A., and Borden, R. C. (2002). Critical Evaluation of Factors Required To Terminate the Postclosure Monitoring Period at Solid Waste Landfills, Environmental Science and Technology, 36(16), 3457-3464.

Bazin, M. J., Rutili, A., Gaines, A., and Lynch, J. M. (1991). Humic Acid Relieves pH- Inhibition of Nitrification in Continuous-Flow Columns, FEMS Microbiology Reviews, 85, 9-14.

Bengtsson, G., Bengtson, P., and Mansson, K. F. (2003). Gross nitrogen mineralization-, immobilization-, and nitrification rates as a function of soil C/N ratio and microbial activity, Soil Biology and Biochemistry, 35(1), 143-154.

Berge, N. D. (2001). Laboratory Study of Aerobic and Anaerobic Landfill Bioreactors and the Influence of Air-Injection Patterns on Aerobic Bioreactor Performance, M.S., University of Sotuh Carolina, Columbia.

Biesterfeld, S., Farmer, G., Figueroa, L., Parker, D., and Russell, P. (2003). Quantification of denitrification potential in carbonaceous trickling filters, Water Research, 37(16), 4011-4017.

Boni, M. R., Delle, S., and Lombardi, E. R. (1996). Aerobic-Anaerobic Operation of a Lab-Scale MSW Sanitary Landfill. Twelfth International Conference on Solid Waste Technology and Management, Philadelphia, PA.

Bonin, P. (1995). Anaerobic Nitrate Recuction to Ammonium in Two Strains Isolated From Coastal Marine Sediment: A Dissimilatory Pathway, FEMS Microbiology Ecology, 19, 27-38.

Brusseau, M. L., and Rao, P. S. C. (1989). The influence of sorbate-organic matter interactions on sorption nonequilibrium, Chemosphere, 18(9-10), 1691-1706.

Buresh, R. J., and Patrick, W. H. (1981). Nitrate Redction to Ammonium and Organic Nitrogen in an Estuarine Environment, Soil Biology and Biochemistry, 13, 279- 283.

59

Burger, M., and Jackson, L. E. (2003). Microbial immobilization of ammonium and nitrate in relation to ammonification and nitrification rates in organic and conventional cropping systems, Soil Biology and Biochemistry, 35(1), 29-36.

Burton, S. A. Q., and Watson-Craik, I. A. (1998). Ammonia and Nitrogen Fluxes in Landfill Sites: Applicability to Sustainable Landfilling, Waste Management and Research, 16, 41-53.

Burton, S. A. Q., and Watson-Craik, I. A. (1999). Accelerated Landfill Refuse Decomposition by Recirculation of Nitrified Leachate. Sardinia 99 Seventh Waste Management and Landfill Symposium, Cagliari, Italy.

Carrera, J., Vicent, T., and Lafuente, J. (2004). Effect of influent COD/N ratio on biological nitrogen removal (BNR) from high-strength ammonium industrial wastewater, Process Biochemistry, 39(12), 2035-2041.

Cheng, W., Tsuruta, H., Chen, G., and Yagi, K. (2004). N2O and NO Production in Various Chinese Agricultural Soils by Nitrification, Soil Biology and Biochemistry, 36, 953-963.

Cheung, K. C., Chu, L. M., and Wong, M. (1997). Ammonia Stripping as a Pretreatment for Landfill Leachate, Water Air and Soil Pollution, 94(1-2), 209-221.

Cole, J. A. (1990). Physiology, Biochemistry, and Genetics of Nitrate Dissimilation to Ammonia. Denitrification in Soil and Sediment, N. P. Revsbeck and J. Sorensen, eds., Plenum Press, New York, NY, 57-76.

Dail, D. B., Davidson, E. A., and Chorover, J. (2001). Rapid Abiotic Transformation of Nitrate in an Acid Forest Soil, Biogeochemistry, 54, 131-146.

Davidson, E. A., Chorover, J., and Dail, D. B. (2003). A Mechanism of Abiotic Immobilization of Nitrate in Forest Ecosystems: The Ferrous Wheel Hypothesis, Global Change Biology, 9, 228-236. de Laclos, H. F., Desbois, S., and Saint-Joly, C. (1997). Anaerobic Digestion of Municipal Solid Organic Waste: Valorga Full-Scale Plant in Tilburg, The Netherlands, Water Science and Technology, 36(6-7), 457-462. de Renzo, D. J. (1978). Pollution Technology Review No. 44: Nitrogen Control and Phosphorous Removal in Sewage Treatment, Noyes Data Corporation, Park Ridge, New Jersey.

Doedens, H., and Cord-Landwehr, K. (1989). Leachate Recirculation. In Sanitary Landfilling: Process, Technology, and Environmental Impact, R. C. T.H. Christensen, and R. Stegman, ed., Academic Press Ltd., London, England.

60 Ehrig, H. J. (1989). Leachate Quality. Sanitary Landfilling: Process, Technology, and Environmental Impact, R. C. T.H. Christensen, and R. Stegman, ed., Academic Press, New York.

Ekama, G. A., and Wentzel, M. C. (1999). Denitrification kinetics in biological N and P removal activated sludge systems treating municipal wastewaters, Water Science and Technology, 39(6), 69-77.

Environmental Control Systems. (1999). "Method and System for Treating Bio- Degradable Waste Material Through Aerobic Degradation." Environmental Control Systems, Inc. (ECS), USA.

Etchebehere, C., Errazquin, I., Barrandeguy, E., Dabert, P., Moletta, R., and Muxi, L. (2001). Evaluation of the denitrifying microbiota of anoxic reactors, FEMS Microbiology Ecology, 35(3), 259-265.

Fanning, J. C. (2000). The Chemical Reduction of Nitrate in Aqueous Solution, Coordination Chemistry Reviews, 199, 159-179.

Faour, A. (2003). First-order Kinetic Gas Generation Model Parameters for Wet Landfills, M.S. Thesis, University of Central Florida., Orlando, FL.

Fazzolari, E., Nicolardot, B., and Germon, J. C. (1998). Simultaneous Effects of Increasing Levels of Glucose and Oxygen Partial Pressures on Denitrification and Dissimilatory Nitrate Reduction to Ammonium in Repacked Soil Cores, European Journal of Soil Biology, 34(1), 47-52.

Fletcher, P. (1989). Landfill Gas Enhancement Technology -- Laboratory Studies and Field Research. Energy from Biomass and Wastes, XIII.

Florida Center for Solid and Hazardous Waste. (2004). "Bioreactor Landfill Demonstration Project."

Foglar, L., and Briski, F. (2003). Wastewater denitrification process--the influence of methanol and kinetic analysis, Process Biochemistry, 39(1), 95-103.

Grady, C. P., Daigger, G. T., and Lim, H. C. (1999). Biological Wastewater Treatment, Marcel Dekker, Inc, New York, NY.

Grosh, C. (1996). Analysis of Florida MSW Leachates, MS, University of Central Florida, Orlando.

Gupta, A. B. (1997). Thiosphaera pantotropha: A Sulphur Bacterium Capable of Simultaneous Heterotrophic Nitrification and Aerobic Denitrification, Enzyme and Microbial Technology, 21, 589-595.

61 Gupta, S. K., and Sharma, R. (1996). Biological oxidation of high strength nitrogenous wastewater, Water Research, 30(3), 593-600.

Halvadakis, C. P. (1983). Methanogenesis in Solid-Waste Landfill Bioreactors, PhD. Dissertation, Stanford University.

Ham, R.K. and Viste, D.R. (1999). Landfill Bioreactor, Montgomery Watson, US Patent #: 5,984,580.

Hanashima, M. (1999). Pollution Control and Stabilization Process by Semi-Aerobic Landfill Type: The Fukuoka Method. Sardinia 99 Seventh Waste Management and Landfill Symposium, Cagliari, Italy.

Hao, X., Heijnen, J. J., and Van Loosdrecht, M. (2002). Model-Based Evaluation of Temperature and Inflow Variations on a Partial Nitrification-ANAMMOX Biofilm Process, Water Research, 36, 4839-4849.

Harper, S. R., and Pohland, F. G. (1988). Landfills - Lessening Environmental Impacts, Civil Engineering, 11, 66-69.

Hater, G.R. and Green, R. (2001). Sequential Aerobic/Anaerobic Solid Waste Landfill Operation, Waste Management, Inc., US Patent #: 6,283,676.

Hater, G.R. and Green, R. (2002). Facultative Landfill Bioreactor, Waste Management, Inc., US Patent #: 6,398,958.

He, X. T., Stevenson, F. J., Mulvaney, R. L., and Kelley, K. R. (1988). Incorporation of Newly Immobilized 15N into Stable Organic Forms in Soil, Soil Biology and Biochemistry, 20(1), 75-81.

Heavey, M. (2003). Low-Cost Treatment of Landfill Leachate Using Peat, Waste Management, 23, 447-454.

Henry, C., Sullivan, D., Rynk, R., Dorsey, K., and Cogger, C. (1999). Managing Nitrogen from Biosolids, Report to the Northwest Biosolids Management Association.

Hoilijoki, T. H., Kettunen, R. H., and Rintala, J. A. (2000). Nitrification of Anaerobically Pretreated Municipal Landfill Leachate at Low Temperature, Water Research, 35(5), 1435-1446.

Hudgins, M., Bessette, B.J., March, J., and McComb, S. (2000) Aerobic Landfill Bioreactor, American Technologies, Inc., US Patent #: 6,024,513.

Ilies, P., and Mavinic, D. S. (2001). The Effect of Decreased Ambient Temperature on the Biological Nitrification and Denitrification of a High Ammonia Landfill Leachate, Water Research, 35(8), 2065-2072.

62

Institute of Waste Management Sustainable Landfill Working Group. (1999). The Role and Operation of the Flushing Bioreactor.

Jetten, M. S. M., Strous, M., van de Pas-Schoonen, K. T., Schalk, J., van Dongen, U. G. J. M., van de Graaf, A. A., Logemann, S., Muyzer, G., van Loosdrecht, M. C. M., and Kuenen, J. G. (1998). The anaerobic oxidation of ammonium, FEMS Microbiology Reviews, 22(5), 421-437.

Jetten, M. S. M., Wagner, M., Fuerst, J., van Loosdrecht, M., Kuenen, G., and Strous, M. (2001). Microbiology and application of the anaerobic ammonium oxidation ('anammox') process, Current Opinion in Biotechnology, 12(3), 283-288.

Jokela, J. P. Y., Kettunen, R. H., Sormunen, K. M., and Rintala, J. A. (2002). Biological Nitrogen Removal from Municipal Landfill Leachate: Low-Cost Nitrification in Biofilters and Laboratory Scale in-situ Denitrification, Water Research, 36(16), 4079-4087.

Juteau, P., Tremblay, D., Ould-Moulaye, C., Bisaillon, J., and Beaudet, R. (2004). Swine Waste Treatment by Self-Heating Aerobic Thermophilic Bioreactors, Water Research, 38(3), 539-546.

Khalil, K., Mary, B., and Renault, P. (2004). Nitrous Oxide Production by Nitrification and Denitrification in Soil Aggregates as Affected by O2 Concentration, Soil Biology and Biochemistry, 36, 687-699.

Kjeldsen, P., Barlaz, M.A., Rooker, R., Baun, A., Ledin, A. and Christensen, T.H. (2002). Present and Long-Term Composition of MSW Landfill Leachate: A Review, Critical Reviews in Environmental Science and Technology, 32(4), 297-336.

Koenig, and Lui. (1996). Autotrophic Denitrification of Landfill Leachate Using Elemental Sulphur, Water Science and Technology, 34(5-5).

Kowalenko, C. G., and Yu, S. (1995). Assessment of Nitrate Adsorption in Soils by Extraction, Equilibration and Column-Leaching Methods, Canadian Journal of Soil Science, 49-57.

Kwok, C. K., and Loh, K. C. (2003). Effects of Singapore Soil Type on Bioavailability of Nutrients in Soil Bioremediation, Advances in Environmental Research, 7, 889- 900.

Laima, M. C. J. (1994). Is KCl a Reliable Extractant of 15NH4+ Added to Coastal Marine Sediments? Biogeochemistry, 27, 83-95.

63 Lay, J. J., Li, Y. Y., Noike, T., Endo, J., and Ishimoto, S. (1997). Analysis of environmental factors affecting methane production from high-solids organic waste, Water Science and Technology, 36(6-7), 493-500.

Lee, H. S., Park, S. J., and Yoon, T. I. (2002). Wastewater treatment in a hybrid biological reactor using powdered minerals: effects of organic loading rates on COD removal and nitrification, Process Biochemistry, 38(1), 81-88.

Leikam, K., Heyer, K. U., and Stegmann, R. (1999). Aerobic in situ Stabilization of Completed Landfills and Old Sites, Waste Management and Research, 17, 555- 562.

Lubkowitz-Bailey, E., and Steidel, R. C. (1999). Investigation of High Temperature Nitrification. 1999 ASCE Environmental Engineering Conference, Norfolk, Va.

Marttinen, S. K., Kettunen, R. H., Sormunen, K. M., Soimasuo, R. M., and Rintala, J. A. (2002). Screening of Physical-Chemical Methods for Removal of Organic Material, Nitrogen and Toxicity from Low Strength Landfill Leachates, Chemosphere, 46, 851-858.

Mateju, V., Cizinska, S., Krejci, J., and Janoch, T. (1992). Biological water denitrification--A review, Enzyme and Microbial Technology, 14(3), 170-183.

Matheson TriGas. (2002). Material Safety Data Sheet for Ammonia/Air Gas Mixtures. Parsippany, NY.

McCreanor, P. T., and Reinhart, D. R. (2000). Mathematical Modeling of Leachate Routing in a Leachate Recirculating Landfill, Water Research, 34(4), 1285-1295.

Merz, R. C., and Stone, R. (1970). Special Studies of a Sanitary Landfill. NTIS PB 196 148, National Technical Information Service, Washington DC, USA.

Metcalf and Eddy, I. (1991). Wastewater Engineering: Treatment, Disposal, and Reuse, McGraw Hill, New York, NY.

Mevel, G., and Prieur, D. (1998). Thermophilic Heterotrophic Nitrifiers Isolated from Mid-Atlantic Ridge Deep-Sea Hydrothermal Vents, Canadian Journal of Microbiology, 44(8), 723-733.

Miller, F. C. (1992). Minimizing Odor Generation. Science and Engineering of Composting: Design, Environmental, Microbiological and Utilization Aspects, H. A. J. Hoitink and H. M. Keener, eds., Renaissance Publications, Worthington, OH.

64 Morrissey, F. A., and Grismer, M. E. (1999). Kinetics of volatile organic compound sorption/desorption on clay minerals, Journal of Contaminant Hydrology, 36(3- 4), 291-312.

Murphy, R. J., Jones, D. E., and Stessel, R. I. (1995). Relationship of Microbial Mass and Activity in Biodegradation of Solid Waste, Waste Management and Research, 13, 485-497.

Murthy, S. N., Novak, J. T., Holbrook, D., and Surovik, F. (2000). Mesophilic Aeration of Autothermal Thermophilic Aerobically Digested Biosolids to Improve Plant Operations, Water Environment Research, 72(4), 476-483.

Nielson, P. H. (1996). Adsorption of Ammonium to Activated Sludge, Water Research, 30(3), 762-764.

Onay, T. T., and Pohland, F. G. (1998). In-situ Nitrogen Management in Controlled Bioreactor Landfills, Water Research, 32(5), 1383-1392.

Onay, T. T., and Pohland, F. G. (2001). Nitrogen and Sulfate Attenuation in Simulated Landfill Bioreactors, Water Science and Technology, 44(2-3), 367-372.

Otieno, F. O. (1989). Leachate Recirculation in Landfills as a Management Technique. Sardinia 1989, Second International Landfill Symposium, Calgari, Italy.

Palmisano, A. C., and Barlaz, M. A. (1996). Microbiology of Solid Waste, CRC Press, Inc., Boca Raton, Fl.

Pichler, M., and Kogner-Knabner, I. (2000). Chemolytic Analysis of Organic Matter during Aerobic and Anaerobic Treatment of Municipal Solid Waste, Journal of Environmental Quality, 29, 1337-1344.

Pochana, K., and Keller, J. (1999). Study of Factors Affecting Simultaneous Nitrification and Denitrification, Water Science and Technology, 39(6), 61-68.

Pohland, F. G. (1975). Sanitary Landfill Stabilization with Leachate Recycle and Residual Treatment. EPA-600/2-75-043.

Pohland, F. G. (1992). Anaerobic Treatment: Fundamental Concepts, Applications, and New Horizons. Design of Anaerobic Processes for the Treatment of Industrial and Municipal Wastes, J. F. Malina and F. G. Pohland, eds., Technomic, Lancaster, PA.

Pohland, F. G. (1995). Landfill Bioreactors: Historical Perspective, Fundamental Principles and New Horizons in Design and Operations. EPA Seminar: Landfill Bioreactor Design and Operation.

65 Pohland, F. G., Cross, W. H., Gould, J. P., and Reinhart, D. R. (1992). The Behavior and Assimilation of Organic Priority Pollutants Codisposed with Municipal Refuse. Cr-812158.

Price, G. A. (2001). Long-Term Nitrogen Management in Bioreactor Landfills, MS Thesis, North Carolina State University, Raleigh, NC.

Price, G. A., Barlaz, M. A., and Hater, G. R. (2003). Nitrogen Management in Bioreactor Landfills, Waste Management, 23(7), 675-688.

Purcell, B. E., Butler, A. P., Sollars, C. J., and Buss, S. E. (1999). Leachate Ammonia Flushing From Landfill Simulators, J. CIWFM, 13.

Qasim, S. R., and Chiang, W. (1994). Sanitary Landfill Leachate, Technomic Publishing Co., Inc., Lancaster, Lancaster.

Read, A. D., Hudgins, M., and Philips, P. (2001). Perpetual Landfilling Through Aeration of the Waste Mass; Lessons from Test Cells in Georgia (USA), Waste Management, 21, 617-629.

Reinhart, D. R. (1989). Fate of Selected Organic Pollutants During Landfill Codisposal with Municipal Refuse, PhD Dissertation, Georgia Institute of Technology, Atlanta.

Reinhart, D. R. (1996). Full-Scale Experiences with Leachate Recirculating Landfills: Case Studies, Waste Management and Research, 14, 347-365.

Reinhart, D. R., and Al-Yousfi, A. B. (1996). The Impact of Leachate Recirculation on Municipal Solid Waste Landfill Operating Characteristics, Waste Management and Research, 14, 337-346.

Reinhart, D. R., McCreanor, P. T., and Townsend, T. G. (2002). The Bioreactor: Its Status and Future, Waste Management and Research, 20(2), 172-186.

Reinhart, D. R., and Townsend, T. G. (1998). Landfill Bioreactor Design and Operation, CRC Press LLC, Boca Raton.

Rittman, B. E., and McCarty, P. L. (2001). Environmental Biotechnology: Principles and Applications, McGraw Hill, New York, NY.

Ritzkowski, M., and Stegmann, R. (2003). Emission Behaviour of Aerated Landfills: Results of Laboratory Scale Investigations. Sardinia 2003 Ninth Waste Management and Landfill Symposium, Cagliari, Italy.

Ross, R. C., and Harris, P. J. (1982). Some Factors Involved in Phase II of Mushroom Compost Preparation, Science Horticulture, 17, 223-229.

66

Sanchez-Monedero, M. A., Roig, A., Paredes, C., and Bernal, M. P. (2001). Nitrogen Transformation during Organic Waste Composting by the Rutgers System and its Effects on pH, EC and Maturity of the Composting Mixtures, Bioresource Technology, 78, 301-308.

Schmidt, I., Sliekers, O., Schmid, M., Bock, E., Fuerst, J., Kuenen, J., Jetten, M., and Strous, M. (2003). New Concepts of Microbial Treatment Processes for the Nitrogen Removal in Wastewater, FEMS Microbiology Reviews, 27, 481-492.

Schwab, B. S., Ritchie, C. J., Kain, D. J., Dobrin, G. C., King, L. W., and Palmisano, A. C. (1994). Characterization of Compost from a Pilot Plant-Scale Composter Utilizing Simulated Solid Waste, Waste Management and Research, 12, 289-303.

Senior, E. (1995). Microbiology of Landfill Sites, CRC Press, Inc., Boca Raton, Fl.

Shiskowski, D. M., and Mavinic, D. S. (1998). Biological Treatment of a High Ammonia Leachate: Influence of External Carbon During Initial Startup, Water Research, 32(8), 2533-2541.

Smith, M. C., Gattie, D. K., Boothe, D. D., and Das, K. C. (2000). Enhancing Aerobic Bioreduction under Controlled Conditions in a Municipal Solid Waste Landfill through The Use of Air Injection and Water Recirculation, Advances in Environmental Research, 3(4), 459-471.

Stegmann, R., and Spendlin, H. H. (1989). Enhancement of Degradation: German Experiences. Sanitary Landfilling: Process, Technology and Environmental Impact, T. H. Christensen, R. Cossu, and R. Stegmann, eds., Academic Press, London.

Stessel, R., and Murphy, R. J. (1992). A Lysimeter Study of the Aerobic Landfill Concept, Waste Management and Research, 10, 485-503.

Tchobanoglous, F., Theisen, H., and Vigil, S. A. (1993). Integrated Solid Waste Management: Engineering Principles and Management Issues, McGraw-Hill, New York, NY.

Thomas, R. G. (1982). Volatilization from Soil. Handbook of Chemical Property Estimation Methods: Environmental Behavior of Organic Compounds, W. J. Lyman, W. F. Reehal, and D. H. Rosenblatt, eds., McGraw-Hill.

Tiedje, J. M. (1988). Ecology of Denitrification and Dissimilatory Nitrate Reduction to Ammonium. Biology of Anaerobic Microorganisms, J. B. A. Zehnder, ed., John Wiley and Sons, New York, NY.

67 Tiquia, S. M., and Tam, N. F. Y. (2000). Fate of Nitrogen During Composting of Chicken Litter, Environmental Pollution, 100, 535-541.

Tittlebaum, M. E. (1982). Organic Carbon Content Stabilization Through Landfill Leachate Recirculation, Journal of Water Pollution Control Federation, 54.

Townsend, T. G., Miller, W., Lee, H., and Earle, J. (1996). Acceleration of Landfill Stabilization Using Leachate Recycle, Journal of Environmental Engineering, 122(4), 263-268.

US Environmental Protection Agency and Waste Management. (2003). Landfills as Bioreactors: Research at the Outer Loop Landfill, Louisville, Kentucky. First Interim Report. US Environmental Protection Agency and Waste Management Inc.

Van Der Zee, S. E. A. T. M., and Van Riemsdijk, W. H. (1986). Sorption kinetics and transport of phosphate in sandy soil, Geoderma, 38(1-4), 293-309.

Van Loosdrecht, M. C. M., and Jetten, M. S. M. (1998). Microbiological conversions in nitrogen removal, Water Science and Technology, 38(1), 1-7.

Van Raaphorst, W., and Malschaert, J. F. P. (1996). Ammonium Adsorption in Superficial North Sea Sediments, Continental Shelf Research, 16(11), 1415-1435.

Vesilind, P. A., Worrell, W. A., and Reinhart, D. R. (2002). Solid Waste Engineering, Brooks/Cole, Pacific Grove, Ca.

Vestgarden, L. S., and Kjonaas, O. J. (2003). Potential nitrogen transformations in mineral soils of two coniferous forests exposed to different N inputs, Forest Ecology and Management, 174(1-3), 191-202.

Ward, M. L., Bitton, G., Townsend, T. G., and Booth, M. (2002). Determining Toxicity of Leachates from Florida Municipal Solid Waste Landfills Using a Battery-of- Tests Approach, Environmental Toxicology, 17, 258-266.

Warith, M., and Sharma, R. (1998). Technical Review of Methods to Enhance Biological Degradation in Sanitary Landfills, Water Quality Research Journal of Canada, 33(3), 417-437.

Welander, U., Henrysson, T., and Welander, T. (1997). Nitrification of Landfill Leachate Using Suspended-Carrier Biofilm Technology, Water Research, 32(5), 2351- 2355.

Willers, H. C., Derikx, P. L. J., ten Have, P. J. W., and Vijin, T. K. (1998). Nitrification Limitation in Animal Slurries at High Temperatures, Bioresource Technology, 64(1), 47-54.

68

Ye, R. W., and Thomas, S. M. (2001). Microbial Nitrogen Cycles: Physiology, genomics and applications, Current Opinion in Microbiology, 4, 307-312.

Yin, S. X., Chen, D., Chen, L. M., and Edis, R. (2002). Dissimilatory nitrate reduction to ammonium and responsible microorganisms in two Chinese and Australian paddy soils, Soil Biology and Biochemistry, 34(8), 1131-1137.

Youcai, Z., Hua, L., Jun, W., and Guowei, G. (2002). Treatment of Leachate by Aged- Refuse-Based Biofilter, Journal of Environmental Engineering, 128(7), 662-668.

Ziehmann, G., and Meier, J. (1999). Alternating Aerobic/Anaerobic Treatments of MSW: Large Scaled Experiments. Seventh International Waste Management and Landfill Symposium, Calgari, Italy.

69 CHAPTER 3

THE IMPACT OF WASTE ACCLIMATION ON IN-SITU AMMONIA REMOVAL FROM BIOREACTOR LANDFILL LEACHATE

NOTE: This paper has been previously published as: Berge, N.D., Reinhart, D.R., Dietz, J. and Townsend, T. (2006) In Situ Ammonia Removal in Bioreactor Landfill Leachate, Waste Management 26 (4), 334-343 (DOI: 10.1016/j.wasman.2005.11.003). The journal Waste Management can be found at http://www.elsevier.com/wps/find/journaldescription.cws_home/404/description?navope nmenu=-2 and this article can be accessed on the web at the following address: http://dx.doi.org/10.1016/j.wasman.2005.11.003.

Introduction

Bioreactor landfills are a new trend in waste management in which in-situ conditions are controlled by adding moisture (often leachate recirculation) and/or air to create a solid waste environment capable of actively degrading the readily biodegradable organic fraction of the waste. Several researchers have documented the many benefits associated with bioreactor technology (Pohland, 1995; Reinhart, 1996; Reinhart and

Townsend, 1998). Despite the numerous advantages associated with operating the landfill as a bioreactor, challenges remain. The persistence of NH3-N in the leachate is one such challenge. Although the organic strength of the leachate is significantly reduced in bioreactor landfills, the NH3-N accumulates because there is no degradation pathway for ammonia in anaerobic systems. Moisture addition and/or recirculating leachate increases

70 the rate of ammonification and results in accumulation of higher levels of NH3-N concentrations than those found in leachate from conventional landfills, even after the organic fraction of the waste is stabilized (Burton and Watson-Craik, 1998; Ehrig, 1989;

Onay and Pohland, 1998; Price et al., 2003). The increased ammonia concentrations intensify the toxicity of the leachate, necessitating leachate treatment before ultimate disposal to protect receiving waters (Burton and Watson-Craik, 1998). It has been suggested by researchers that NH3-N is a significant long-term pollution problem in landfills and it is likely that its presence will influence when post-closure monitoring may end (Kjeldsen et al., 2002; Price et al., 2003).

Removal of NH3-N from leachate is currently practiced ex-situ. However, ex-situ treatment of high strength leachate can be difficult and costly. Thus, the development of an in-situ removal technique would be an attractive alternative, potentially yielding both economic and environmental advantages. Using biological nitrogen removal processes, such as nitrification and denitrification, would be advantageous, as complete nitrogen removal may be achieved. In the past, oxygen has not been added to landfills, thus nitrification has not typically occurred in-situ; however, air addition has more recently been practiced at a number of landfills worldwide, resulting in enhancement of solid waste degradation (ECS, 1999; Merz and Stone, 1970; Reinhart et al., 2002; Read et al.,

2001; Ritzkowski and Stegmann, 2003; Stessel and Murphy, 1992).

Recent laboratory studies have shown the efficacy of in-situ biological nitrogen removal processes in solid waste environments (Jokela et al., 2002; Onay and Pohland,

1998; Youcai et al., 2002). Youcai et al. (2002) conducted a study in which leachate passed through a biofilter consisting of old waste (8 to 10 years old) with both anaerobic

71 and aerobic sections. A removal of 99.5% of the ammonia in leachate was observed, coupled with elevated concentrations of nitrate and nitrite, indicating the ammonia was converted biologically. Onay and Pohland (1998) also completed an in-situ nitrification/denitrification laboratory study in which high removals of nitrogen in the leachate were observed and attributed to the nitrification and denitrification processes.

Additionally, Hanashima (1999) observed incidental removal of nitrogen in aerobic or semi-aerobic landfills.

These studies evaluated in-situ, or partially in-situ, nitrogen removal and demonstrated the potential efficacy of such processes in landfills, however, they lack the depth necessary to implement the processes at field-scale. The purpose of this study is to collect information to aid in the implementation of nitrification at field-scale, allowing for a more informed approach to designing and operating bioreactor landfills.

Specifically, ammonia removal rates in a solid waste environment were observed and measured, yielding valuable insight into how nitrification may be implemented in the field. Additionally, the rate data were fit to the Monod equation producing an equation describing ammonia removal. Because adding air to a landfill can be costly, it is anticipated that air addition may occur more as a polishing step in the operation of a landfill during which air is added to designated, older portions of the landfill to nitrify ammonia-rich leachate and degrade some organics not removed under anaerobic conditions. Laboratory-scale studies evaluating the rates of ammonia removal in old wastes were conducted and ammonia removal kinetics were evaluated at different ammonia concentrations.

72 Materials and Methods

A waste acclimation process was operated and was dual purpose, to provide an acclimated waste source for parallel batch microcosm studies and to demonstrate the efficacy of in-situ nitrification and denitrification. In this study, acclimated waste is defined as waste that has been exposed to a nitrifying microbial population and is capable of removing ammonia concentrations as high as 1000 mg N/L. The microcosm studies were smaller scale experiments conducted to evaluate the kinetics of ammonia removal using waste from the waste acclimation process.

Aerobic Reactor (Waste Acclimation Process) Design and Operation

A 133-L reactor was designed to allow for leachate draining and recirculation, air addition, and gas sampling. To prevent clogging of the leachate drain, a layer of gravel was placed at the bottom of the reactor. Digested municipal solid waste (MSW) was obtained from an aerobic MSW compost facility located in Sumter County, Florida,

USA. All MSW received by the facility first goes through a materials recovery facility in which materials that have recycling value are removed (i.e. plastic, paper, metals, glass).

The process is not 100% efficient, thus a large amount of plastic, paper, and glass remain in the waste. Once through the sorting process, the waste is combined with biosolids from a wastewater treatment facility and placed in an aerated vessel. The mixture remains in the composting vessel for at least 72-hours. When the compost is removed, it is placed in a holding area. The waste used in this study has remained in that holding area for at least one year.

73 The aerobic reactor was initially seeded with 14 kg of digested MSW (i.e. compost). Six hundred grams of approximately 4-cm long wood chips were added to the reactor to promote air distribution throughout the matrix. Initially, 6 L of deionized (DI) water and 2 liters of mixed liquor from a local nitrifying wastewater plant were added to the system to initiate leachate production and aid in developing a nitrifying microbial population. All liquid was immediately drained and 3.95 liters of the leachate were recirculated. Moisture-saturated air was added to both the bottom and middle of the compost matrix at a total rate of 2.77 L/min. The air was saturated with moisture prior to introduction to the reactor to replenish any water lost due to evaporation and was added continuously throughout the duration of the study.

After two days, leachate was removed and then 2 L recirculated; this process was repeated every two to three days. Samples were routinely removed and analyzed for pH, chemical oxygen demand (COD), alkalinity, sulfate, nitrate, nitrite, and NH3-N to evaluate whether nitrification or other nitrogen removal processes were occurring. If needed, tap water was also added to bring the recirculated volume to two liters. Gas- phase samples were extracted from three sample ports located on the side of the reactor and used to measure the in-situ oxygen concentrations to evaluate the efficiency of the aeration system. Digested MSW samples were periodically removed from the reactor and used in parallel microcosm studies. Additional digested MSW was added to the reactor as needed to maintain the same waste mass. Periodically, samples of the digested MSW from the aerobic reactor were collected and solid-phase organic nitrogen, moisture content and volatile solids content of the waste were measured.

74 The aerobic reactor was operated for 717 days and was periodically spiked with small concentrations of NH3-N to ensure the compost was populated with nitrifying bacteria.

In-Situ Nitrification Microcosm Studies

Microcosm experiments were conducted in 3-L foil gas sampling bags (SKC, Inc.,

Pennsylvania) modified to permit addition of solid components to the bag. A gas sampling port was located at the top of each bag. Each microcosm was loaded with 200 g of acclimated, digested MSW taken from the aerobic reactor along with 20 g of wood chips. The waste source was chosen to simulate older waste from a landfill, where aeration and nitrification processes would most probably be employed. Fifteen mL of ammonium bicarbonate solution (concentration varied depending on target ammonia level) was added to each system resulting in moisture levels at field capacity

(approximately 63% on a wet-weight basis). After each bag was loaded with digested

MSW and ammonia, the bags were purged with helium and then filled with pure oxygen.

Air was not used because the high nitrogen concentration would limit the potential to complete mass balances on nitrogen in the systems. Also, use of pure oxygen ensured adequate oxygen. Each bag was sealed to allow for all nitrogen products to be controlled, captured, and measured. Both liquid- and gas-phase samples were taken from the bags over time and percent recoveries of nitrogen species were calculated. The microcosms were inverted once each day to simulate leachate recirculation.

Ammonia-nitrogen removal rates were measured for two series of experiments; the first series was conducted on acclimated waste at constant temperature (22oC), with

75 an initial oxygen concentration of 100% in the headspace, and with varying ammonia concentrations (200, 500 and 1000 mg N/L). One duplicate study at the 500 mg N/L concentration was conducted to demonstrate the reproducibility of the data. The second series of experiments was conducted to evaluate the effect of waste acclimation on ammonia removal rates. The experiments were conducted using unacclimated waste at

22oC, with an initial oxygen concentration of 100% oxygen in the headspace and with an ammonia concentration of 500 mg N/L. An abiotic control was conducted under the same environmental conditions and at an ammonia concentration of 500 mg N/L. The controls contained waste that had been autoclaved at 120oC for 2 hours. Because biological activity was halted, this set of controls functioned to evaluate whether changes in ammonia concentration could be attributed to physical or chemical processes. A biotic control with no ammonia added was also conducted under the same environmental conditions with no ammonia added to evaluate whether ammonia was produced via ammonification during the studies.

Each microcosm set consisted of several different bags operated in a batch mode and each bag was destructively sampled over time. Each bag was destroyed during each sample time because there was not sufficient leachate volume for repetitive samples and because the potential for sorption precluded the analysis of leachate alone. After disassembling each microcosm, a mild extraction procedure was conducted to desorb any

NH3-N or nitrate from the waste.

In all bags, both liquid- and gas-phase parameters were measured, including NH3-

N (leachate and sorbed masses), nitrate (leachate and sorbed masses), nitrite, sulfate, pH, nitrogen and nitrous oxide gases, and gas-phase oxygen. Nitrogen and nitrous oxide were

76 measured because there may be micro-anoxic areas in which denitrification occurred.

Nitrous oxide was not measured in all studies because of initial procedural issues.

Analytical Techniques

COD, biochemical oxygen demand (BOD), pH, and alkalinity were measured using methods found in Standard Methods (1995). Ammonia-nitrogen in the leachate was measured using an ion-specific electrode (Fisher Scientific, Inc.) via the known-addition method. The form of NH3-N present in solution is dependent on the solution pH. In all studies conducted, the pH was below 8.0, therefore the dominant form of ammonia is ammonium. All anions, including nitrite, nitrate, and sulfate, were measured using a DX-

120 ion chromatograph (Dionex, Inc.) equipped with an AS-14 column and using a bicarbonate/carbonate eluent. Prior to anion analysis, all samples were centrifuged and filtered using a 0.45-micron nitrocellulose filter.

In the microcosm studies, a mild extraction procedure was used to measure any

NH3-N and nitrate that was sorbed to the waste matrix. After disassembling the microcosms, 210 mL of DI water were added to each system and the microcosms were subsequently drained. The drained leachate was analyzed for pH, nitrite, nitrate, and sulfate. Three hundred mL of a 0.5-M sodium sulfate solution were added to the remaining waste to desorb NH3-N and nitrate. The system was shaken at 245 rpm for 1.5- hours, subsequently drained, and ammonia and nitrate concentrations were measured.

Preliminary work was completed to ensure the volume and concentration of sodium sulfate and shaking time were sufficient. Ammonia was measured using an ion specific electrode. Nitrate in the extract was measured using a nitrate-selective electrode (Fisher

77 Scientific, Inc.). The electrode was used in place of the ion chromatograph for measuring nitrate in the extract because of sulfate interference. A known-addition method ensured no interferences from the sample matrix influenced the measured concentration.

Oxygen, nitrogen and methane in the headspace were analyzed using a gas chromatograph (GC) (Shimadzu, Inc) equipped with a TCD detector and two packed 13X molecular sieve columns (Alltech Associates, Inc.) in series (to achieve adequate separation of the oxygen and nitrogen peaks). The injector, detector, and oven temperatures were 25oC. Nitrous oxide was measured with a GC (Varian, Inc.) equipped with an electron capture detector (ECD) in conjunction with an AT-Q 30-m capillary column (Alltech Associates, Inc); the injector, detector, and oven temperatures were

110oC, 110oC, and 60oC, respectively.

In each microcosm, gas volume in the headspace was measured using a water displacement technique. Each bag was submerged in water and the volume of water displaced measured. The gas volume was equal to the displaced water volume corrected by the volume of the waste and liquid. The pressure applied to the bags was minimal and thus neglected.

Solid-phase organic nitrogen was measured using a modified macro-kjeldahl method (Standard Methods, 1995). Modifications to the Standard Method include the use of a larger volume of digestion reagent (100 mL) and a longer acid hydrolysis step (6 hrs) during which ground solid samples were shaken in 200 mL of solution prior to digestion

(100 mL of digestion solution and 100 mL DI). Moisture content was measured by drying solids in an oven at 105oC for 24 hours. Volatile solids were measured by heating 1 g ground waste samples at 550oC for 2 hours.

78 Results and Discussion

Aerobic Reactor (Waste Acclimation Process)

On days 30, 74, 139, and 230, spikes of NH3-N were added to the reactor to determine whether nitrification was occurring in-situ, as shown in Figure 3-1(a). After each spike of ammonia, an initial decline of ammonia followed by an increase in both nitrate and sulfate (see Figures 3-1(b) and (c)) was observed. However, the concentration of nitrate and nitrite observed was never as high as would have been stoichiometrically expected if only nitrification had occurred, suggesting that both nitrification and denitrification were occurring within the system. Mass balances could not be conducted on this reactor because of inefficient gas capture. Although air was continuously added to the system, a sulfide odor was detected in the leachate, suggesting the presence of anaerobic pockets in the reactor.

Because of the presence of anoxic pockets coupled with the lower than expected appearance of nitrate and nitrite, a spike of nitrate was added to the leachate recirculation stream on day 26 to confirm whether or not denitrification was occurring in-situ. Figure

3-1(b) depicts the spiking and subsequent disappearance of nitrate from the leachate stream. The nitrate decrease was coupled with an increase in sulfate concentration (see

Figure 3-1(c)), suggesting a portion of nitrate removal may be attributed to autotrophic denitrification. Autotrophic denitrification follows reaction 3.1 and is favored in environments with a low biodegradable environment in the presence of inorganic sulfur compounds, such as hydrogen sulfide (Koenig and Lui, 1996).

- - + 2- 1/5 NO3 + 1/8 HS + 3/40 H Æ 1/10 N2 + 1/8 SO4 + 1/10 H2O (3.1)

79 1200 N) - 1000 on (mg/L 800

600 en Concentrati 400 trog

nia-Ni 200 Ammo 0 0 100 200 300 400 500 Time (Days) (a) Ammonia-nitrogen concentration in the leachate from the aerobic reactor.

250

200 (mg/L-N) n

150 ncentratio

100

50 Nitare-Nitrogen Co

0 0 100 200 300 400 500 Days (b) Nitrate-nitrogen concentrations in the leachate from the aerobic reactor. Arrows represent NH3-N spikes.

80

(b) nitra Figure 3-1.Wasteacclim (d) Ammonia-,nitrate-andn (c) Sulfateconcentratio nitrate, andnitriteconcentrations ov t

e-n Sulfate Concentration (mg/L) Ammonia-Nitrogen Concentrations (mg/L-N) 10 20 40 60 80 100 120 i 00 trogen con 200 400 600 800 0 0 0 0 0 0 0 0 5 0 0 0 c entra ns overtim 520 ation processresu 1 0 0 ti itr ons, (c)sulfateco from repres ite -nitrogen conce theaerobicreacto e 54 ent N in er ti 200 0 Time (D the leach m 81 H lts: (a)Ammoniaconcentrations overtime, e Day 3 fromleachatefro -N spikes. ay ncentration s s 5 ) 60 ate from n tra 30 0 tions ov r .

N Nitrate-N A

m i tri theaerobicreactor.Arrows m t s e o overtim - 580 n Nit i er tim a-N i r tro og m i 400 g t en r theaerobicreactor. e oge n

e n f e a r om thele 60 nd (d)ammonia, 0 0 100 200 300 400 500 5 0 Nitrate- and Nitrite-Nitrogen Concentration (mg/L-N) 0 achate

Onay and Pohland (2001) also suspected autotrophic denitrification was occurring in their laboratory study and led to observed sulfate production.

Based on the stoichiometry of reaction 3.1, during the nitrate spike, more nitrate disappeared (approximately 70%) than can be accounted for by the increase in sulfate concentration. Thus, it is possible nitrate removal may also be attributed to heterotrophic denitrification, resulting in an undetectable conversion of nitrate and/or nitrite to nitrogen gas which could not be measured. If all the nitrate unaccounted for is assumed to have been converted to nitrogen gas, almost 70% of the nitrate was denitrified via heterotrophic denitrification. Thirty percent of the nitrate was converted to sulfate via autotrophic denitrification based on the sulfate concentrations measured. Both the COD and BOD in the reactor leachate during the nitrate spike were low (2,000 mg/L and 1 mg/L, respectively). The COD concentration was stable, suggesting the COD remaining is recalcitrant and that an environment favorable to the autotrophic denitrifiers is present.

On days 500, 515, 518, and 538, the system was spiked with larger concentrations of NH3-N (1000 mg N/L) to acclimate the system to higher ammonia concentrations. As shown in Figure 3-1, all spikes were followed with an increase in nitrate and, at times, nitrite, indicating nitrification was proceeding. The sulfate concentrations continued to increase as the reactor was spiked. On average, during each spike, the sulfate concentrations measured suggest conversion of 10 to 15% of the nitrate via autotrophic denitrification. Both the COD and BOD remained fairly constant throughout each spike.

Solid-phase parameters were measured on the waste over time. The parameters remained fairly constant. The organic nitrogen, moisture content, and volatile solids averaged 14 mg N/g dry waste, 63% (wet weight) and 56% (by weight), respectively.

82 In-Situ Nitrification Microcosm Studies

In all microcosm studies, ammonia readily disappeared (Table 3-1). The lower removal observed in the 1000 mg N/L is most probably due to pH inhibition. Figure 3-2 presents the NH3-N masses over time in all studies. During each study, nitrate, nitrite, nitrogen gas, and nitrous oxide were detected, all indicating that simultaneous processes were occurring. Note that an initial concentration of 100% oxygen was used in all studies. The oxygen concentration in the headspace of each bag declined over time, but never reached levels low enough to limit ammonia removal. The lowest oxygen concentration measured was 21% (in the study with the greatest mass of NH3-N). In the majority of the studies, the oxygen concentration remained greater than 60%. High levels of oxygen may be slightly inhibitory to the nitrification process (Park and Noguera,

2004), thus the ammonia removal rates measured may be slightly lower than what would occur in a landfill environment in which the oxygen concentration is more realistic (less than 21%).

Table 3-1. Ammonia Removal Efficiencies. Initial Ammonia % Removal Concentration (mg N/L) 500 (acclimated) 99 500 (acclimated) 99 500 (unacclimated) 99 500 (unacclimated) 97 200 98 1000 95

83

500 mg N/L (a) - Unacclimated 120 500 mg N/L (b) - Unacclimated 500 mg N/L (a) - Acclimated 500 mg N/L (b) - Acclimated 100 200 mg N/L - Acclimated 1000 mg N/L - Acclimated Abiotic Control (500 mg/L-N) 80 ss (mg N)

en Ma 60

40 nia-Nitrog

Ammo 20

0 0 5 10 15 20 Time (Days) Figure 3-2. Total ammonia-nitrogen masses recovered from microcosms over time.

Ammonia Sorption

A high amount of ammonia sorption was observed during the microcosm studies.

Most ammonia sorption was reversible, as it was recovered through mild extractions while conducting the abiotic studies (see Figure 3-2). However, in the microcosms sampled on day zero for all studies, some ammonia was unrecoverable. This initial sorption amounted to 10 to 20% of the total ammonia added. It is hypothesized that this initial unrecovered mass was immediately and tightly fixed to complex molecules present in the waste matrix, such as humic substances as has been observed in soil systems

(Nommik and Vahtras, 1982). It is possible that a portion of the initially fixed ammonia desorbs at some point in time during the studies. However, the exact mass that may

84 desorb with time is unknown. Studies conducted in soil systems have shown that less than 15% of initially fixed ammonia that is not easily exchanged with an extraction media is biologically available (Nommik and Vahtras, 1982).

In-Situ Nitrification Processes

The ammonia removal in all studies can be attributed in part to nitrification. In almost all studies, the decline in ammonia concentration (see Figure 3-2) was coupled with an increase in nitrite/nitrate concentrations (Figure 3-3). In some studies, a stoichiometric increase in nitrate concentrations was not observed because denitrification was also occurring. However, the increases in nitrite/nitrate were never as high as were stoichiometrically predicted, suggesting other processes were occurring. The pH in all the systems declined over time from an initial value around 7.0 to 6.0 (depending on study).

The drop in pH also suggests nitrification had occurred, since alkalinity is destroyed during nitrification.

Abiotic controls were also operated in which the waste was autoclaved for two hours. Figure 3-2 shows the mass of ammonia in an abiotic control operated at an ammonia concentration of 500 mg N/L and the ammonia masses in all active studies. The ammonia concentration remained relatively constant in the abiotic control, while it declined rapidly in the active studies, suggesting ammonia removal can be attributed primarily to biotic processes. Ammonia assimilation is expected to be minimal, thus the primary biotic process responsible for ammonia removal should be nitrification.

A biotic control was also operated in which no ammonia was added to the microcosms to evaluate whether ammonia was being produced during the experiments

85 via ammonification. No ammonia was produced, nor was there any additional nitrate produced during the biotic control experiment, suggesting that ammonification did not occur during these studies and the ammonia removal measured is not offset by ammonia production. Because it is expected that the ammonia removal processes will occur in older portions of the landfill, most of the nitrogen from the waste will already have been solubilized and ammonification processes should not be the rate limiting step during ammonia removal. However, if ammonia removal processes were to be applied in younger waste in which significant amounts of ammonia have not yet been released to the leachate, ammonification may be the rate limiting step in the ammonia removal process.

Simultaneous Nitrogen Removal Processes

Masses of both nitrogen and nitrous oxide gases were measured. Cumulative masses of each gas measured are shown in Table 3-2. The data suggest denitrification had occurred along with nitrification. The occurrence of denitrification was not surprising because the presence of micro-anoxic areas within the digested MSW was expected and this phenomenon was observed during the waste acclimation process. Quantification of denitrification processes was complicated by the potential diffusion of atmospheric nitrogen gas into the bags. During each experiment, a bag containing only oxygen was monitored to determine whether diffusion of nitrogen was occurring. A small amount of nitrogen diffusion did periodically occur, however, because the mass of nitrogen produced in each bag was small, diffusion of atmospheric nitrogen may impact the nitrogen mass balances (Table 3-2).

86 25

N) 20 g m ( Ammonia-Nitrogen ies Nitrate-Nitrogen

ec 15 Nitrite-Nitrogen Sp en

og 10 Nitr f o

ss 5 Ma

0 0123456 Time (Days) (a) 200 mg N/L Study

60 Ammonia-Nitrogen Nitrate-Nitrogen 50 Nitrite-Nitrogen g N) m ( s 40 e i c e p S 30 en og r

Nit 20 ss of a

M 10

0 01234567

Time (Days) (b) 500 mg N/L (a) – Acclimated

87 70

60

N) Ammonia-Nitrogen 50 Nitrate-Nitrogen Nitrite-Nitrogen

40

30

20 of Nitrogen Species (mg s s a

M 10

0 0246810121416

Time (days) (c) 500 mg N/L - Unacclimated

120

Ammonia-Nitrogen 100 Nitrate-Nitrogen Nitrite-Nitrogen

s (mg N) 80 ecie

Sp 60 ogen

40 of Nitr s

Mas 20

0 0 5 10 15 20 25 30 Time (Days) (d) 1000 mg N/L Figure 3-3. Ammonia-, nitrate-, and nitrite-nitrogen masses over time in all studies: (a) 200 mg N/L, (b) 500 mg N/L – Acclimated, (c) 500 mg N/L – Unacclimated, and (d) 1000 mg N/L.

88 Table 3-2. Microcosm Gaseous Headspace Analysis. Initial Ammonia Concentration Maximum Nitrogen Gas Maximum Nitrous Oxide (mg N/L) Produced (% of total N Measured (% of total N initially present) initially present) 500 (acclimated) 10.8 12.2 500 (acclimated) 13.2 5.2a 500 (unacclimated) 13.6 NMb 500 (unacclimated) 26.5 8.5 200 13.0 9.4 1000 4.9 15.6 Nitrogen Diffusion Studiesc 6.1 0 a during this set, not all data points were collected because of experimental difficulties. b NM means not measured c the average value is presented, not the maximum.

Nitrous oxide concentrations in the gas-phase were the result of either partial nitrification or denitrification (Mummey et al., 1994; Venterea and Rolston, 2000).

Nitrous oxide production as a result of partial denitrification would occur because of the presence of high oxygen concentrations in the gas-phase, which act to inhibit the production of the nitrous oxide reductase enzyme that is necessary to convert nitrous oxide to nitrogen gas. Otte et al. (1996) showed the sensitivity of the nitrous oxide enzyme to oxygen concentrations. Low carbon to nitrogen ratios can also lead to nitrous oxide production (Hong et al., 1993). Nitrous oxide can also be a byproduct of nitrification and generally results when there is a low partial pressure of oxygen (Khalil et al., 2004; Mummey et al., 1994). In these studies, the oxygen concentration was high

(100%) and the carbon to nitrogen ratio was estimated to be fairly small, suggesting the nitrous oxide production was a result of partial denitrification. Without conducting a study with labeled nitrogen species, the exact cause of the nitrous oxide production cannot be determined. The mass of nitrogen added that was converted to nitrous oxide

89 was significant at times, reaching levels as high as 16% of the total nitrogen added.

Nitrous oxide production is a concern because it is a potent greenhouse gas. Price et al.

(2003) also observed nitrous oxide production while conducting studies evaluating the denitrification potential of waste. However, they hypothesized that the nitrous oxide may be converted to nitrogen gas before it exits the landfill because of the high residence times expected.

Ammonia Removal Kinetics

Because the mass of ammonia that disappeared cannot be accounted for when adding the masses of all byproducts, the kinetic rates calculated are reported as ammonia removal kinetics, not nitrification kinetics. However, it is believed the ammonia removal kinetics would be very similar to nitrification kinetics. All NH3-N removal rates were calculated using a central difference method of analyses (equation 3.2):

Ct−1 − Ct+1 RR = (3.2) tt+1 − tt−1

where, RR is the rate of ammonia change at time t (mg N/day), C is the total NH3-N mass

(mg N), and t is the time (days). The total NH3-N mass was found by adding the masses of both sorbed and liquid-phase ammonia.

In all studies, the ammonia disappearance rates decreased with increasing time and decreasing concentration. During the 1000 mg N/L study, the disappearance rates appeared to decrease with time and ammonia mass more significantly than in other

90 studies (see Figure 3-4(a)). This decrease is probably due to inhibition by low pH levels.

The pH in this study had decreased to 5.94, a level that is reportably inhibitory to nitrification activity (Grady et al., 1999; Metcalf and Eddy, 1991), particularly in short- term batch studies. pH inhibition was not observed at any other initial ammonia concentrations. It is possible that alkalinity will need to be added to landfills with high ammonia concentrations to prevent pH inhibition from occurring. The lowest oxygen concentration measured during the 1000 mg N/L test was 21%, sufficient enough to maintain ammonia removal.

During the studies, the ammonia removal rate appeared to reach a maximum level at high ammonia concentrations, suggesting the data follow Monod kinetics, as was expected because the primary removal mechanism is biological. The data were normalized by the dry mass of waste present in each microcosm test and plotted against the total ammonia concentration, as shown in Figure 3-4(b). Because some of the ammonia was sorbed to the waste matrix, the total concentration of ammonia was calculated by dividing the total ammonia mass (i.e. ammonia mass measured in the liquid plus the mass sorbed) by the total volume of liquid initially present in the microcosm.

This concentration of ammonia was used because these are the concentrations that would be measured in the recirculated leachate stream entering the landfill. The ammonia rates were normalized by the dry mass of waste, allowing for the rate data found in this study to be easily applied to other landfill scenarios since the mass and moisture content of the waste landfilled is generally known. The density of the waste (dry basis) was 0.23 g/cm3.

Using the density, the rates can be correlated to waste volume, allowing for the application of this relationship to field-scale.

91 0.20 6.6 y) da * 0.18 6.5 0.16 Rate Data pH 0.14 6.4

0.12 g N/g dry waste 6.3 m

( 0.10 pH 6.2 Rate 0.08 val o 0.06 pH Inhibited Rates 6.1

Rem 0.04 6.0 0.02 monia

Am 0.00 5.9 0 100 200 300 400 500 600

Total Ammonia Concentration (mg/L-N) (a) 1000 mg N/L

Rate Data -- Unacclimated 0.25 Rate Data - Acclimated

day) Monod Fit - Acclimated Monod Fit - Unacclimated ste* a 0.20 w y

0.15

Acclimated Rate (mg N/g dry waste*day) = (0.196*C )/(C + 59.6) N N te (mg N/g dr a

R 0.10 l a v o m e R

0.05 a i Unacclimated Rate (mg N/g dry waste*day) = (0.117*C )/(C + 147) N N

Ammon 0.00 0 100 200 300 400 500 600 700 Total Ammonia Concentration (mg/L-N) (b) All acclimated and unacclimated data fit to the Monod equation (minus inhibited data from the 1000mg N/L study)

Figure 3-4. Ammonia removal rates for all studies: (a) 1000 mg N/L, (b) All acclimated and unacclimated data fit to Monod.

92 The normalized rate data were fit to the Monod equation using SigmaPlot (SSPS,

Inc.). The Monod equation follows equation 3.3:

kC R = N (3.3) KS + CN

where, R is the ammonia removal rate (mg N/g dry waste-day), Ks is the half-saturation constant (mg N/L), CN is the total NH3-N concentration (mg N/L), and k is specific rate of removal of ammonia (mg N/day-g dry waste).

Figure 3-4(b) includes all the rate data from all the acclimated microcosm studies, except for the data in which pH inhibition occurred in the 1000 mg N/L study. The

Monod fit to the data was good, with a correlation coefficient of 0.79. The half- saturation constant resulting from the fit was 59.6 mg N/L. This half-saturation value is much higher than typically found in wastewater treatment processes (Grady et al., 1999), indicating that the ammonia removal rate is dependent on concentration over a larger concentration range than in wastewater. Additionally, because the first-order portion of the curve is relatively large (thus a large half-saturation constant), it is possible that the rate is being limited by mass transfer of either oxygen or ammonia (Pohland, 1992).

Acclimated vs Unacclimated Processes

Studies were conducted to evaluate the effect of the waste acclimation process on the ammonia removal rate. The rates of ammonia removal for both acclimated and unacclimated studies conducted at 500 mg N/L are shown on Figure 3-4(b). During the

93 unacclimated studies, the ammonia removal rate appeared to reach a maximum level at high ammonia concentrations, similar to the acclimated studies, although the rate reached was lower than that in the acclimated studies suggesting that the acclimated waste had a greater capacity to remove ammonia. The unacclimated data fit well (r2 = 0.91) to the

Monod equation (see Figure 3-4(b)); a half-saturation constant of 147 mg N/L was determined. The half-saturation constant for the unacclimated set is higher than for the acclimated set because the slope of the first-order portion of the unacclimated curve is

40% lower than the acclimated, indicating the maximum rate of ammonia removal occurs at higher concentrations.

Field-Scale Ammonia Removal Implications

Using the data obtained, hypotheses may be made regarding a field-scale implementation strategy. First, as was shown during the 1000 mg N/L study, in systems containing high NH3-N concentrations and low buffering capacity, the pH may drop below the level in which nitrification may proceed efficiently. In those instances, a buffer may need to be added. If denitrification also occurs, the pH drop will not be as substantial.

The effect of both oxygen concentration and temperature on NH3-N removal is unknown. In this study, a very high oxygen concentration was used. No landfill will have

100% oxygen. Studies evaluating the effect of different oxygen concentrations are underway and results will be applied to the Monod model in a multiplicative manner. The same is true for temperature effects.

94 By completing a mass balance around the ammonia removal zone and using the dry density of the waste, the leachate recirculation rate, and the Monod kinetic expression obtained, the required landfill volume and times for a given removal efficiency of ammonia can be calculated. However, it is important to note that the information gained using the kinetic data from this study may not be universally valid. Using these kinetics assumes that field conditions are similar to the laboratory conditions, and that may not be the case because of mass transfer differences in the field, as the waste will have different particle sizes and both air and liquid will be flowing through the landfill. Additionally, the environmental conditions present in the field may be different than those tested in these studies (22oC and 100%). Thus, it is suggested estimation of treatment times and volumes be integrated with the effect of oxygen and temperature.

Nitrogen Mass Balances

Mass balances were conducted on each system. Although all byproducts of nitrification and denitrification were measured, not all of the ammonia that disappeared was recovered in the byproducts measured. For example, over a period of 15 days, the recovery of nitrogen species (nitrate, nitrite, ammonia, nitrogen gas and nitrous oxide) decreased from 90 to 30% of nitrogen. It is possible that other processes were occurring and resulted in the attenuation of either ammonia or nitrate. For example, assimilation of ammonia and nitrate could not be measured and may have accounted for a small portion of the unaccounted ammonia disappearance; however, it is not likely to be the primary cause for the decrease in recovery of nitrogen over time. Additionally, sorption of some

95 of the nitrogen species could have contributed to the low recovery of nitrogen over time.

Alternatively, abiotic conversion of nitrate may have occurred.

Another possibility for the unaccounted-for nitrogen stems from experimental difficulties. It is possible that some of the gaseous endproducts were leaked to the atmosphere and thus not measured. Leakage of the gases is not likely, as atmospheric pressure conditions were maintained in all bags. Also, the masses of nitrogen added were small; it is possible that at times the mass of nitrogen byproducts produced were not detectable, resulting in a smaller mass of nitrogen being measured.

Maintaining a gas-tight bag in which there is no diffusion of atmospheric gases is challenging, particularly when biological nitrogen production expected is small. It is possible that some of the nitrogen gas measured in the bags may have resulted from atmospheric nitrogen gas diffusion into the bags. Many researchers have conducted studies in which a complete nitrogen mass balance was not possible (Micks et al., 2004;

Sjoberg and Persson, 1998).

Conclusions

Ammonia removal via nitrification and denitrification is feasible in bioreactor landfills, readily occurring in decomposed solid waste environments. Results suggest that nitrification and denitrification may occur simultaneously in one aerobic landfill cell

(even under low biodegradable C:N conditions), rather than requiring two separate cells containing two different in-situ environments (i.e. anoxic and aerobic), which is significant when developing field-scale guidance for implementation of such processes.

Also, demonstrating denitrification can occur in older portions of the landfill (via both

96 heterotrophic and autotrophic denitrification) provides valuable insight as to where nitrification and denitrification can occur in landfills.

Based on the ammonia removal rates measured in these studies, unacclimated solid waste environments do not appear to be an issue, as ammonia will still be removed at high rates. Acclimation also does not appear to take long periods of time. When examining the data from the aerobic reactor, the time for each ammonia spike to be removed is approximately 30 days, a short period when considering the life span of a landfill. Before using the removal rate expressions obtained to estimate times and areas of waste necessary to remove NH3-N from landfill leachate, more laboratory studies are needed to determine how other environmental conditions (i.e. temperature and gas-phase oxygen concentrations) affect ammonia removal. Additionally, field-scale studies are also needed to evaluate scale-up of the laboratory data.

References

Berntson, G.M. and Aber, J.D. (2000). Fast Nitrate Immobilization in N Saturated Temperate Forest Soils, Soil Biology and Biochemistry 32, 151-156.

Burton, S. A. Q. and Watson-Craik, I.A. (1998). Ammonia and Nitrogen Fluxes in Landfill Sites: Applicability to Sustainable Landfilling, Waste Management and Research 16, 41-53.

Davidson, E.A., Hart, S.C., and Firestone, M.K. (1992). Internal Cycling of Nitrate in Soils of a Mature Coniferous Forest, Ecology 73(4), 1148-1156.

Davidson, E.A., Chorover, J. and Dail, R. (2003). A Mechanism of Abiotic Immobilization of Nitrate in Forest Ecosystems: the Ferrous Wheel Hypothesis, Global Change Biology 9, 228-236.

Ehrig, H. J. (1989). Leachate Quality. In Sanitary Landfilling: Process, Technology, and Environmental Impact, T.H. Christensen, R. Cossu, and R. Stegman,Eds.; Academic Press, New York.

97

Environmental Control Systems, Inc. (ECS) (1999). Method and System for Treating Bio-Degradable Waste Material Through Aerobic Degradation, US Patent 5,888,022, 1999.

Grady, C.P., Daigger, G.T. and Lim, H.C. (1999). Biological Wastewater Treatment, Second Edition, Marcel Dekker, Inc. New York, NY.

Hanashima, M. (1999). Pollution Control and Stabilization Process by Semi-Aerobic Landfill Type: The Fukuoka Method, In Proceedings of Sardinia 99 Seventh Waste Management and Landfill Symposium, Cagliari, Italy.

Hong, Z., Hanaki, K., and Matsuo, T. (1993). Greenhouse-Gas – N2O Production During Denitrification in Wastewater Treatment, Water Science and Technology 28, 203- 207.

Jokela, J.P.Y., Kettunen, R.H., Sormunen, K.M. and Rintala, J.A. (2002). Biological Nitrogen Removal from Municipal Landfill Leachate: Low-Cost Nitrification in Biofilters and Laboratory Scale in-situ Denitrification, Water Research 36(16), 4079-4087.

Khalil, K., Mary, B., and Renault, P. (2004). Nitrous Oxide Production by Nitrification and Denitrification in Soil Aggregates as Affected by O2 Concentration, Soil Biology and Biochemistry, 36 687-699.

Kjeldsen, P., Barlaz, M.A., Rooker, R., Baun, A., Ledin, A. and Christensen, T.H. (2002). Present and Long-Term Composition of MSW Landfill Leachate: A Review, Critical Reviews in Environmental Science and Technology, 32(4), 297-336.

Koenig and Lui (1996). Autotrophic Denitrification of Landfill Leachate Using Elemental Sulphur, Water Science and Technology, 34 (5-6).

Kowalenko, C.G. and Yu, S. (1995). Assessment of Nitrate Adsorption in Soils by Extraction, Equilibration and Column-Leaching Methods, Canadian Journal of Soil Science, 49-57.

Leikam, K., Heyer, K.U. and Stegmann, R. (1999). Aerobic in situ Stabilization of Completed Landfills and Old Sites, Waste Management and Research 17, 555- 562.

Merz, R.C. and Stone, R. (1970). Special Studies of a Sanitary Landfill, NTIS PB 196 148. Washington DC, USA.: National Technical Information Service.

Metcalf and Eddy, Inc. (1991). Wastewater Engineering: Treatment, Disposal, and Reuse, McGraw Hill, New York, NY.

98 Micks, P., Aber, J.D., Boone, R.D., and Davidson, E.A. (2004). Short-Term Soil Respiration and Nitrogen Immobilization Response to Nitrogen Application in Control and Nitrogen-Enriched Temperate Forests, Forest Ecology and Management in press.

Mummey, D.L., Smith, J.L., and Bolton, H. (1994). Nitrous Oxide Flux from a Shrub- Steppe Ecosystem: Sources and Regulations, Soil Biology and Biochemistry 26(2), 279-286.

Nommik, H. and Vahtras, K. (1982). Retention and Fixation of Ammonium and Ammonia in Soils, In Nitrogen in Agricultural Soils, Edited by F.J. Stevenson, Madison, WI.

Onay, T.T. and Pohland, F.G. (1998). In-Situ Nitrogen Management in Controlled Bioreactor Landfills, Water Research 32 (5), 1383-1392.

Onay, T.T. and Pohland, F.G. (2001). Nitrogen and Sulfate Attenuation in Simulated Landfill Bioreactors, Water Science and Technology 44(2-3), 367-372.

Otte, S., Grobbenn, N.G., Robertson, L.A., Jetten, M.S.M., and Kuenen, J.G. (1996). Nitrous Oxide Production by Alcaligenes faecalis Under Transient and Dynamic Aerobic and Anaerobic Conditions, Applied Environmental Microbiology 62, 2421-2426.

Park, H.D. and Noguera, D.R. (2004). Evaluating The Effect Of Dissolved Oxygen On Ammonia-Oxidizing Bacterial Communities In Activated Sludge, Water Research 38, 3275-3286.

Pohland, F.G. (1992). Anaerobic Treatment: Fundamental Concepts, Applications, and New Horizons, In Design of Anaerobic Processes for the Treatment of Industrial and Municipal Wastes, edited by Malina and Pohland.

Pohland, F.G. (1995). Landfill Bioreactors: Historical Perspective, Fundamental Principles and New Horizons in Design and Operations, In EPA Seminar Publication: Landfill Bioreactor Design and Operation.

Price, G.A., Barlaz, M.A. and Hater, G.R. (2003). Nitrogen Management in Bioreactor Landfills, Waste Management 23(7), 675-688.

Read, A.D., Hudgins, M. and Philips, P. (2001). Perpetual Landfilling Through Aeration of the Waste Mass; Lessons from Test Cells in Georgia (USA), Waste Management 21, 617-629.

Reinhart, D.R. (1996). Full-Scale Experiences with Leachate Recirculating Landfills: Case Studies, Waste Management and Research 14, 347-365.

99 Reinhart, D.R. and Townsend, T.G. (1998). Landfill Bioreactor Design and Operation, Boca Raton, New York, USA: CRC Press LLC.

Reinhart, D.R., McCreanor, P.T., and Townsend, T., (2002). The Bioreactor: Its Status and Future, Waste Management and Research 20(2), 172-186.

Ritzkowski, M. and Stegmann, R. (2003). Emission Behaviour of Aerated Landfills: Results of Laboratory Scale Investigations, In Proceedings of Sardinia 2003 Ninth Waste Management and Landfill Symposium, Cagliari, Italy.

Sjoberg, R.M. and Persson, T. (1998). Turnover of Carbon and Nitrogen in Coniferous 15 Forest Soils of Different N-Status and Under Different NH4-N Application Rate, Environmental Pollution 102SI, 385-393.

Standard Methods for the Examination of Water and Wastewater (1995) 19th Ed., American Public Health Association/American Water Works Association/Water Environment Federation. Washington, DC, USA.

Stark, J.M. and Hart, S.C. (1997). High Rates of Nitrification and Nitrate Turnover in Undisturbed Coniferous Forests, Letters to Nature 385, 61-64.

Stessel, R. and Murphy, R. (1992). A Lysimeter Study of the Aerobic Landfill Concept, Waste Management and Research 10, 485-503.

Venterea, R. and Roltston, D. (2000). Mechanisms and Kinetics of Nitric and Nitrous Oxide Production During Nitrification in Agricultural Soil, Global Change Biology 6, 303-316.

Youcai, Z., Hua, L., Jun, W. and Guowei, G. (2002). Treatment of Leachate by Aged- Refuse-Based Biofilter, Journal of Environmental Engineering 128(7), 662-668.

100 CHAPTER 4

THE IMPACT OF GAS-PHASE OXYGEN AND TEMPERATURE ON IN-SITU AMMONIA REMOVAL FROM BIOREACTOR LANDFILL LEACHATE

NOTE: Portions of this paper have been submitted for publication as: Berge, N.D., Reinhart, D.R., Dietz, J. and Townsend, T. (submitted, 2006) The Impact of Gas-Phase Oxygen and Temperature on In Situ Ammonia Removal in Bioreactor Landfill Leachate, Environmental Science and Technology. The journal Environmental Science and Technology can be found at http://pubs.acs.org/journals/esthag/.

Introduction

As more and more landfills transition to operating as bioreactors, challenges encountered during and/or as a result of their operation must be addressed. Removal of ammonia-nitrogen is one such challenge. Ammonia-nitrogen concentrations in leachate from bioreactor landfills are generally higher than those from conventional landfills

(Berge et al. 2005; Burton and Watson-Craik, 1998). Leachate ammonia-nitrogen is a significant long-term pollution problem that may greatly influence when post-closure care of a landfill may end or be reduced (Kjeldsen et al., 2002), thus creating considerable economic impacts if left untreated. Developing an effective method to treat and remove ammonia is of great importance to advancing bioreactor technology.

Currently, ammonia-nitrogen is removed from leachate outside of the landfill.

However, ex-situ treatment can be both difficult and costly. Although bioreactor landfills

101 result in higher leachate ammonia-nitrogen concentrations, they may provide a waste environment that can be more conducive to in-situ nitrogen removal mechanisms (Berge et al., 2005). Researchers have successfully demonstrated the efficacy of in-situ nitrogen removal processes, including both nitrification (Hanashima, 1999; Jokela et al., 2002;

Onay and Pohland, 1998; Youcai et al., 1998) and denitrification (Burton and Watson-

Craik, 1998; Price et al., 2003). However, these studies did not provide the information necessary to understand what may influence in-situ nitrification processes in solid waste environments. In each study, the impact of changes in the environment on ammonia removal was not evaluated, nor was there an effort to define ammonia removal rates.

To fairly evaluate such processes for field-scale implementation, information regarding nitrification kinetics under different environmental conditions found in bioreactor landfills is necessary. Berge et al. (2006) conducted laboratory-scale experiments evaluating the kinetics of ammonia-nitrogen removal in digested municipal solid waste (simulating old waste) at 22oC and with 100% oxygen in the gas-phase.

Although this work provided the framework for conducting such experiments, studies evaluating the impact of different gas-phase oxygen concentrations and temperatures, two factors greatly influencing nitrification, are necessary before being able to hypothesize what may happen when implementing at the field-scale.

The purpose of this study is to determine the impact of various gas-phase oxygen concentrations and temperatures on in-situ ammonia removal processes in bioreactor landfills. When air is introduced into landfills, the gas-phase oxygen concentrations will vary from location to location within the landfill because of the heterogeneous nature of solid waste, potentially resulting in oxygen limited areas and thus reduced nitrification

102 rates. Temperatures found in landfills (up to 70oC) also vary spatially within the waste and may increase to levels inhibitory to nitrification. Additionally, high temperatures may adversely impact the amount of dissolved oxygen (D.O.) available for nitrification to proceed. Laboratory-scale experiments evaluating ammonia removal rates in a digested solid waste environment at different average gas-phase oxygen concentrations (between

0.7 and 100%) and temperatures (22, 35 and 45oC) were conducted and evaluated. All experiments were conducted in older waste environments, following Berge et al. (2006), as it is hypothesized to be a more suitable environment for in-situ nitrification processes; air injection costs would be minimized because any oxygen demand associated with organics present in younger waste would be lower. A multiplicative Monod model was developed with terms describing the impact of ammonia concentration, gas-phase oxygen concentration, and temperature on ammonia removal.

Experimental Materials and Methods

A reactor was operated to provide an acclimated waste source for parallel batch microcosm studies. Details regarding the design and operation of the acclimation reactor can be found elsewhere (Berge et al., 2006). Acclimated waste is defined as waste that has been exposed to a nitrifying microbial population and the temperature being tested in the parallel microcosm studies, and is thus capable of removing ammonia concentrations as high as 1000 mg N/L at each temperature evaluated. The waste acclimation process was placed in a constant temperature room and a water bath to maintain constant temperatures within the waste mass. Smaller batch microcosm studies were operated with various gas-phase oxygen concentrations and under different temperatures to

103 evaluate the kinetics of ammonia removal. Waste from the waste acclimation reactor was used as the waste source in each microcosm experiment.

Microcosm Study Operation

Microcosm experiments were conducted in foil gas sampling bags (SKC, Inc.,

Pennsylvania) modified to permit addition of solid components to the bag. Bag size was dependent on the type of experiment being conducted. Tests with 100% oxygen were conducted in 3-L bags, while tests with lower oxygen concentrations were conducted in

10-L bags. Gas-phase oxygen concentrations were controlled as an indirect measure of dissolved oxygen concentrations in the liquid phase. A gas sampling port was located at the top of each bag. Details regarding the loading of the microcosms can be found elsewhere (Berge et. al., 2006). After each bag was loaded with digested MSW and ammonia, the bags were purged with helium and then filled with gas containing the appropriate oxygen concentration. Studies were conduced at average gas-phase oxygen concentrations between 0.7 and 100%. Gas mixtures consisted of oxygen and helium and were added using a 1.5-L gas-tight syringe. Air was not used because the high nitrogen concentration would limit the potential to complete mass balances on nitrogen. Each bag was sealed to allow for all nitrogen products to be controlled, captured, and measured.

All bags were placed in a constant temperature incubator. Temperatures of 22, 35 and

45oC were evaluated. Each microcosm set consisted of several different bags operated in a batch mode and each bag was destructively sampled over time. Both liquid- and gas- phase samples were taken from the bags over time and percent recoveries of nitrogen species were calculated.

104 Three series of experiments were conducted; the first series was conducted on acclimated waste at 22oC, with varying initial ammonia concentrations (500 and 1000 mg

N/L) and varying target gas-phase oxygen concentrations (0.7, 5, 17 and 100%). Over time, the oxygen concentrations declined slightly due to the demand imposed by nitrifying and heterotrophic microorganisms. To maintain a near constant average concentration in the gas-phase, the gas composition of the bags was measured every 12- hours and, if needed, a small volume of 100% oxygen was added to increase the oxygen concentration to the target level. Ammonia removal rates were quantified for each experiment. The second and third series of experiments were conducted at 35 and 45oC, respectively, and at the same ammonia and gas-phase oxygen concentrations. An exception is the 0.7% oxygen level; this level was only tested at 22oC. In each of the series of experiments, one duplicate was conducted at each temperature and oxygen level evaluated.

Abiotic controls were conducted at each temperature with 100% oxygen in the gas-phase and at an ammonia concentration of 500 mg N/L. The control contained waste that had been autoclaved at 120oC for two hours to halt biological activity and functioned to evaluate whether changes in ammonia concentration could be attributed to physical or chemical processes. Biotic controls with no ammonia added were also conducted at each temperature level to evaluate ammonia production via ammonification.

Each microcosm set consisted of several different bags operated in a batch mode and each bag was destructively sampled over time (Berge et al. 2006). Both liquid- and gas-phase parameters were measured in all bags, including ammonia-nitrogen (leachate and sorbed masses), nitrate (leachate and sorbed masses), nitrite, sulfate, pH, nitrogen

105 and nitrous oxide gases, and gas-phase oxygen. As the temperature in the bags increased, the potential for liberation of free ammonia was possible. Thus, samples were taken to determine whether gas-phase ammonia was present in the head-space of each bag.

Analytical Techniques

Chemical oxygen demand (COD), pH, alkalinity, anions (i.e. nitrite, nitrate, sulfate), gaseous compounds (oxygen, nitrous oxide) were measured following methods outlined by Berge et al. (2006). Gaseous ammonia was measured by extracting the gas from each bag and passing it through an indicating boric acid solution. In the 100% oxygen studies, the volume in the headspace of each bag was measured using a water displacement technique. In the lower oxygen studies, the headspace volume was measured by extracting the gas with a 1.5-L gas-tight syringe.

Results and Discussion

Nitrogen Removal Processes

Significant ammonia removal was observed in all studies at all gas-phase oxygen concentrations and temperatures tested, with the exception of the controls (Figure 4-1).

Within the studies conducted at each temperature, the slowest ammonia removal occurred in the tests with the lowest oxygen concentrations (0.7% at 22oC and 5% at both 35 and

45oC). During all studies, masses of nitrite, nitrate, nitrous oxide and nitrogen gas were detected, suggesting that a portion of the ammonia removal can be attributed to the nitrification process and that simultaneous nitrogen removal processes were occurring

106 (i.e., denitrification). Other researchers have observed simultaneous processes occurring in similar types of experiments (e.g., Holman and Wareham, 2005). The pH in all studies decreased over time from an initial average level of approximately 7.7 to a final pH of approximately 6.0, also suggesting that the ammonia removed was primarily due to nitrification.

Results from abiotic controls operated at 500 mg N/L at each temperature level are also presented in Figure 4-1. As shown, the mass of ammonia remained fairly constant in the control experiments, suggesting the majority of ammonia removal is due to biotic processes. In the control conducted at 35oC, the last point showed a slight increase in ammonia mass. Ammonia assimilation is expected to be minor, thus the primary biotic process responsible for ammonia removal is nitrification. A biotic control was also operated in which no ammonia or nitrate was produced, suggesting that ammonification and dissimilatory nitrate reduction to ammonia did not occur and that removal is not offset by ammonia production.

The masses of nitrite and nitrate measured were never as high as would be stoichiometrically expected, confirming other processes were contributing to nitrogen removal. The nitrate and nitrite productions at 22 and 35oC were similar. At 45oC, there was little nitrite production. Researchers evaluating the impact of temperature on nitrification in wastewater treatment units have observed temperature-related trends with respect to nitrite and nitrate production (Bougard et al., 2006); noting that nitrite concentrations generally increase with temperature because the growth rate of ammonia- oxidizers is higher than nitrite-oxidizers at temperatures above 15oC. In the experiments conducted in this study, this phenomenon was not observed. Operating the waste

107 acclimation process may have dampened any temperature impact on nitrite production.

The highest nitrite concentration measured was approximately 200 mg N/L at 35oC.

Although nitrite has been reported to be inhibitory to nitrification in wastewater at levels around 20 mg N/L (Charley et al., 1980), the high concentration did not appear to be inhibitory in these experiments.

Figure 4-2 presents the maximum percentage of nitrogen initially present that was converted to nitrogen and nitrous oxide gases during all studies conducted at initial ammonia concentrations of 500 mg N/L. The most complete denitrification occurred in the 0.7% oxygen study at 22oC, as evidenced by the lack of nitrous oxide and the highest amount of nitrogen gas (87% of nitrogen was converted to nitrogen gas). Lower oxygen levels tended to create a more optimal environment for denitrification. For the most part, the nitrogen gas produced decreased with increasing oxygen concentration, which was expected and indicates that denitrification decreases with increasing oxygen concentrations. Although operating at lower oxygen levels slows the ammonia removal process, the most complete nitrogen removal was observed and would require adding less air to a landfill, alleviating some economic burden imposed by its addition. Being able to achieve complete nitrogen removal in an aerated system is very advantageous when considering field-scale operation as complete nitrogen removal can occur in one aerated cell.

108 80

70

N) 60 g m 50 ss ( a 40 a M i 30 n

20

Ammo 10

0 0 5 10 15 20 25 Time (Days) 500 mg N/L - 100% 500 mg N/L - 0.7% 500 mg N/L - 5% 500 mg N/L - 17% 500 mg N/L - Abiotic Control Biotic Control

(a) 22C

120 ) 100 g N m

( 80 s s a 60 M a 40 oni m

m 20 A

0 02468 Time (Days) 500 mg N/L - 100% 500 mg N/L - 5% 500 mg N/L - 17% 500 mg N/L - Abiotic Control Biotic Control

(b) 35C

109 100 )

N 80 g m 60 ss ( a 40 a M i n

o 20 m m

A 0 02468 Time (Days) 500 mg N/L - 17% 500 mg N/L - 100% 500 mg N/L - Abiotic Control 500 mg N/L - 5% Biotic Control

(c) 45C

Figure 4-1. Ammonia removal masses in studies at different oxygen levels at the three different temperatures: (a) 22C, (b) 35C and (c) 45C.

87% 45 s 40 22 C Ga n

e 35 35 C og

r 45 C t i 30 N s 25 Note: The test at d a e 0.7% oxygen was c 20 only conducted at odu

r 15 o

P 22 C

N 10 % .

x 5 a M 0 100% 20% 5% 1%

(a) maximum mass of nitrogen initially present converted to nitrogen gas at each temperature and oxygen level

110 20 e

id 18 22 C x

O 16 s

u 35 C o

r 14 t i 45 C N 12 as

10 ced

u Note: The test at d 8 o r 0.7% oxygen was P 6

N only conducted at

% 4 o

x. 22 C

Ma 2

0 100% 20% 5% 1%

(b) maximum mass of nitrogen initially present converted to nitrous oxide at each temperature level

Figure 4-2.Gaseous by-products produced in terms of the maximum mass of nitrogen converted to either nitrogen or nitrous oxide: (a) maximum mass of nitrogen initially present converted to nitrogen gas at each temperature and oxygen level. (b) maximum mass of nitrogen initially present converted to nitrous oxide at each temperature level.

The greatest nitrous oxide production occurred at 35oC (13 – 18% of the maximum mass of nitrogen initially present was converted to nitrous oxide), the temperature in which the fastest ammonia removal rates were measured, while the lowest was at 45oC, a temperature that was slightly inhibitory to the microbial processes. Nitrous oxide production can be a result of either partial denitrification or partial nitrification

(Mummey et al., 1994; Venterea and Rolston, 2000). Nitrous oxide production during denitrification may occur because of the presence of oxygen, while production during nitrification generally occurs when low oxygen is present (Khalil et al., 2004). Without using a radio-labeled tracer, it is impossible to determine the exact source of the N2O,

111 however, inferences can be made. Besides the study conducted at 22oC and 0.7% oxygen, the lowest nitrous oxide was produced in the 20% oxygen studies (2% of the maximum mass of nitrogen initially present that was converted to N2O). At both 100 and

5% oxygen concentrations, higher N2O was produced suggesting that it may be a result of either nitrification or denitrification, depending on the concentration of oxygen present.

The production of N2O is not a desirable outcome as it is a potent greenhouse gas. Price et al. (2003) also measured N2O in some laboratory-scale experiments while evaluating the denitrification capacity of waste, but hypothesized that nitrous oxide emissions may not be an issue at field-scale because of the high residence times generally found in landfills during which N2O may be converted to N2.

Nitrogen mass balances were conducted on all studies. In each study, there was some initial irreversible sorption occurring in each bag (approximately 5 to 10% of the nitrogen added) resulting from sorption to complex molecules present in the waste matrix, such as humic substances, as has been observed in soil systems (Nommik and

Vahtras, 1982). This initial mass was unaccounted for when summing all nitrogen species measured (i.e., nitrate, nitrite, ammonia) and was assumed to remain constant throughout the study. Results from the abiotic controls confirm this conclusion. Over time, the percentage of nitrogen in the form of ammonia decreased and increases in by-products were measured. The mass balances do not always resolve to 100%. On average, the percent recovery of nitrogen declined with time from approximately 96% to an average level of 66% recovery of nitrogen after an average of 6 days. This reduction may be due to several reasons, including assimilation, abiotic removal of nitrate, or uncaptured nitrogen gas. Other researchers have conducted studies in which a complete nitrogen

112 balance was not possible (Micks et al., 2004; Sjoberg and Persson, 1998). Using radiolabeled nitrogen would aid in determining the exact fate of the nitrogen.

Ammonia-Nitrogen Removal Rates

Because all of the ammonia that disappeared could not be accounted for, the kinetic rates are reported as ammonia-nitrogen removal rates, not nitrification rates.

However, it is believed that ammonia removal rates are very similar to what the nitrification rates would be. All ammonia-nitrogen removal rates were calculated using a central difference method of analyses (Berge et al. 2006).

Impact of oxygen and temperature on removal rates

All rate data were normalized by the dry mass of waste present in each microcosm test and plotted against the total ammonia concentration, as shown in Figure 4-3. Because some of the ammonia was sorbed to the waste matrix, the total concentration of ammonia was calculated by dividing the total ammonia mass (i.e. ammonia mass measured in the liquid plus the recoverable sorbed mass) by the total volume of liquid initially present in the microcosm. This ammonia concentration was used because these are the concentrations that would be measured in the recirculated leachate stream entering the landfill. All ammonia removal rates were normalized by the dry mass of waste, allowing for the rate data found in this study to be easily applied to other landfill scenarios since the mass and moisture content of the waste landfilled are generally known.

113 ) 0.6 22C - 100% O2 y 22C - 5% O2 da

- Monod Fit - 100% O2

ste Monod Fit - 5% O2 0.5 dry wa g

/ 0.4

(mg N 0.3 e t

val Ra 0.2 o m e R

0.1 monia m

A 0.0 0 200 400 600 800 1000 1200 Total Ammonia Concentration (mg N/L) (a) Raw removal rate data and Monod fits at 22oC

35C - 17% O2 ) 1.2 y 22C - 17% O2 a 45C - 17% O2 Monod Fit - 35C ste-d

a 1.0 Monod Fit - 22C

w Monod Fit - 45C

0.8

(mg N/g dry 0.6 te a

0.4 moval R e

0.2

Ammonia R 0.0 0 200 400 600 800 1000 1200 Total Ammonia Concentration (mg N/L)

(b) Raw removal rate data from tests with approximately 17% oxygen in the gas-phase at three different temperature levels

114 0.6 y) a Removal Rate Fit

ste-d 0.5 Raw Rate Data (5% O2 at 45C) y wa

0.4 N/g dr g m 0.3 te ( a

0.2 Removal R 0.1

Ammonia 0.0 0 200 400 600 800 1000 1200 Total Ammonia Concentration (mg N/L)

(c) Raw removal rate data and the rate equation fit with approximately 5% oxygen in the gas-phase and at 45oC

Figure 4-3.Ammonia removal rates and Monod equation fits for a few of the microcosm experiments: (a) Raw removal rate data and Monod fits at 22oC, (b) Raw removal rate data from tests with approximately 17% oxygen in the gas-phase at three different temperature levels, and (c) Raw removal rate data and the rate equation fit with approximately 5% oxygen in the gas-phase and at 45oC.

In almost every study, the removal rates appeared to reach a maximum level at higher concentrations, suggesting Monod-like removal had occurred, as was expected because Monod kinetics generally describe biological reactions. However, the rates from the studies conducted at 45oC and 5% oxygen deviated from the Monod model (Figure 4-

3).

The normalized rate data at each temperature were first separately fit to a multiplicative Monod equation to take into account the impact of both ammonia and gas-

115 phase oxygen concentrations. The rate data at 45oC and 5% oxygen were excluded from this analysis because they approach first-order removal, not Monod-like removal. The oxygen concentrations used were averages calculated for each individual bag. Thus, in the 5% oxygen study, the individual bag oxygen concentrations ranged from 4.0 to 5.0%.

At higher oxygen levels (100%), inhibition of ammonia removal occurred, thus an

Andrew’s inhibition term was chosen to describe the impact of oxygen. Different inhibition terms were evaluated, however, the best fit was found with the Andrew’s inhibition term. Inhibition by high ammonia concentrations was not observed. All data were fit to this model using SigmaPlot (SSPS, Inc.). The multiplicative Monod equation used, keeping temperature and pH constant, follows equation 4.1:

⎛ ⎞ ⎜ ⎟ ⎛ kC ⎞⎜ %O ⎟ ⎜ N ⎟ 2 R = ⎜ 2 ⎟ (4.1) ⎜ K + C ⎟ ⎝ S N ⎠⎜ %O2 ⎟ ⎜ KO + %O2 + ⎟ ⎝ K I ⎠

where, R is the ammonia removal rate (mg N/g dry waste-day), Ks is the half- saturation constant (mg N/L), CN is the total NH3-N concentration (mg N/L), k is the specific rate of removal of ammonia (mg N/g dry waste-day), KO is the oxygen half- saturation constant (%O2), %O2 is the gas-phase concentration of oxygen (%), and KI is a term to represent the inhibition observed at high oxygen concentrations (%O2).

The data from all tests (500 and 1000 mg N/L) at each oxygen level were combined. Table 4-1 presents the constants and r2 values associated with each fit at each temperature. The p-values are also shown in Table 4-1. Although the model parameters

116 are not always statistically significant (p-value < 0.05), the parameters are mechanistically significant and a required element of each fit. The r2 values are a result of the scatter associated with each fit; data scatter is a result of the inherent heterogeneities associated with solid waste and biological systems, as well as the fact that the pH was not the same in each bag. Interestingly, at 35oC, there was no statistically significant difference (confidence interval of 75%) among all the different oxygen levels, suggesting oxygen concentrations had little impact on nitrification rates at this temperature. At 22 and 45oC, however, the 100 and 17% oxygen studies were found to be statistically different (75% confidence level).

The half-saturation values of each fit are relatively high when compared to those generally found in wastewater treatment (approximately 1 mg N/L; Grady et al., 1999) suggesting that the ammonia removal rate is dependent on concentration over a larger range in landfills than in wastewater and that the rate may be limited by the mass transfer of oxygen and/or ammonia (Pohland, 1992).

117 Table 4-1. Model constants for different Monod fits. 2 Monod Model k Ks Ko2 KI t n z r o Fit at 22 C: ⎛⎞0.972 96.1 5.69 17.3 - - - 0.78 (0.0202) (0.0140) (0.1679) (0.1443) ⎛⎞kC ⎜⎟%O R = ⎜⎟N ⎜⎟2 K+C⎜ %O2 ⎟ ⎝⎠SNKO+% +2 ⎜⎟O2 2 ⎝⎠KI o Fit at 35 C: ⎛⎞1.04 212 3.35 390 - - - 0.69 (0.0002) (0.0092) (0.1129) (0.4257) ⎛⎞kC ⎜⎟%O R = ⎜⎟N ⎜⎟2 K+C⎜ %O2 ⎟ ⎝⎠SNKO+% +2 ⎜⎟O2 2 ⎝⎠KI o Fit at 45 C: ⎛⎞1.78 315 9.4 29.6 - - - 0.63 ⎛⎞kC ⎜⎟%O (0.0711) (0.0702) (0.1991) (0.1155) R = ⎜⎟N ⎜⎟2 K+C⎜ %O2 ⎟ ⎝⎠SNKO+% +2 ⎜⎟O2 2 K ⎝⎠I Fit with all rate data: 1.17 167 2.84 206 33.8 7.66 6.13 0.75 ⎛⎞ n (<0.0001) (<0.0001) (0.0089) (0.031) (<0.0001) (<0.0001) (<0.0001) ⎜⎟⎛T ⎞ ⎛kC ⎞%O ⎛T ⎜1− ⎟⎞1 N ⎜2 ⎟⎝t ⎠⎛⎞ R= ⎜⎟ 2 ⎜e⎟⎜⎟z− pH KC+⎜ %O ⎟⎜t ⎟1+10() ⎝SN⎠KO++% 2 ⎝⎠⎝⎠ ⎜O2 2 ⎟ ⎝KI ⎠ Fit excluding 100% oxygen rate data: 0.899 159 2.46 - 34.7 4.56 5.40 0.74

n (<0.0001) (0.0031) (0.0056) (<0.0001) (0.0009) (<0.0001) ⎛⎞T ⎛⎞⎛⎜⎟1− ⎞ ⎛⎞kCN %1O2 T⎝⎠t ⎛⎞ Re= ⎜⎟⎜⎟⎜⎟⎜⎟z− pH K+C⎜⎟K+%O⎜t ⎟⎝⎠1+10() ⎝⎠SN⎝⎠O2 2 ⎝⎠

* p-value for each parameter is shown in parentheses * - dash indicates the parameter was not used in that fit

118 Additionally, as the temperature increases, the Ks value also increases, suggesting mass transfer limitations increase with temperature. The highest Ko value was found at

45oC, as was expected, because as temperature increases, the D.O. concentration decreases, potentially creating an oxygen limitation. At higher temperatures, the sensitivity to gas-phase oxygen increases. At both 22 and 45oC, the studies conducted at

100% oxygen resulted in inhibition of removal. At 35oC, however, no inhibition

o occurred, which explains the significantly higher KI determined for the 35 C rate fit.

There was significant scatter in the rate data at the 0.7% oxygen study (22oC) because maintaining a constant oxygen concentration of 0.7% was difficult and greatly influenced the ability to fit the data to the one-dimensional Monod model. During this test, the gas-phase oxygen concentrations ranged from 0.5 to 1.0%, thus each rate point was not measured at the exact same oxygen concentration, resulting in significant variability in measured rates. In this oxygen range, the nitrification rate is sensitive to oxygen, thus small fluctuations in oxygen concentration have a greater impact on ammonia removal rates than at higher oxygen levels.

At the 5% oxygen study conducted at 45oC, the rates appeared to follow a first- order model (Figure 4-3), indicating that mass transfer issues dominated the removal at this oxygen and temperature level. The sensitivity to oxygen increases with temperature, so small changes in gas-phase concentrations may create large rate changes. Because the rate is first-order, it suggests that the liquid-film diffusional resistence is high

(Harremoes, 1978). When Beccari et al. (1992) evaluated the effects of D.O. on nitrification in suspended biomass, they reported that internal diffusion resistance should not be neglected when measuring overall kinetics, especially when the D.O. in the liquid

119 was less than 2 mg/L. The removal rate at 5% oxygen and 45oC can be predicted with equation 4.2.

1.1 RC= 0.00027 N (4.2)

At each temperature, the slowest removal occurred at the lowest oxygen concentration evaluated. The theoretical D.O. concentration in the bulk liquid at each gas-phase oxygen concentration and temperature tested is included in Table 4-2; the estimates assume clean water and neglect conductivity and total solids impacts. The D.O. concentrations in the biofilm drive ammonia removal in these experiments and will be significantly less than that in the bulk liquid. Stewart (1998) reported that the D.O. in a denitrifying biofilm is approximately 40% less than what is found in the surrounding bulk liquid. Mass flux of ammonia in the biofilm may also be a limitation. The ammonia concentration has been reported to be 50% less than that in the bulk-liquid (Stewart,

1998). The D.O. level in the bulk-liquid at 0.7% oxygen and 22oC is approximately 0.29 mg/L, within a range that has been reported to limit nitrification rates (Stenstrom and

Poduska, 1980). At 5% oxygen and 45oC, the D.O. in the bulk-liquid is higher, 1.50 mg/L, but is still at a level in which nitrification processes are inhibited. At this higher temperature, it is possible both mesophiles and thermophiles were present and resulted in a greater uptake of the available oxygen, thus out competing the nitrifiers.

At 22 and 45oC, the experiments conducted at 100% oxygen show some degree of inhibition when compared to those studies conducted under lower oxygen levels. Other researchers have observed similar phenomena when conducting studies at high oxygen levels (Charley et al., 1980; Jones and Paskins, 1982; Park and Noguera, 2004). As

120 reported in Table 4-2, the D.O. concentration with 100% oxygen in the gas-phase is approximately 42 mg/L, a D.O. value that is almost four times greater than saturation values at 21% gas-phase oxygen. It is possible that had the waste been acclimated to

100% oxygen prior to running the experiments, the process would not have been inhibited. Charley et al. (1980) found that immediately after adding pure oxygen to a wastewater process the nitrification rate decreased and remained low for five days, after which the rate increased. Jones and Paskins (1982) found similar results. In the 100% oxygen study at 35oC, however, no inhibition was observed.

Table 4-2. Dissolved oxygen concentrations (mg/L) at different temperatures and gas- phase oxygen concentrations. Gas-Phase Oxygen 22oC 35 oC 45 oC Concentration (%) 0.7 0.29 0.24 0.21 5 2.10 1.73 1.50 20 8.42 6.92 6.01 100 42.08 34.59 30.1

When comparing the impact of temperature alone, ammonia removal was fastest at 35oC and slowest at 22oC. Surprisingly, there was substantial, although slower, removal at 45oC. In wastewater, nitrification has been found to be limited at temperatures above 40 – 45oC (Willers et al., 1998). Willers et al. (1998) found that nitrification rates dropped sharply in digested pig slurry when temperatures were above 40oC, and they dropped sharply when temperatures were above 45oC in veal-calf slurry. However, El-

Zanfaly (1981) measured nitrying microbial activity in soil up to 50oC. In landfills, it is possible that the microorganisms will be protected from the high temperatures by the

121 waste particles. At higher temperatures, concern arises with the loss of free ammonia gas is likely. Theoretically, at the pH, ammonia concentrations and temperatures tested, a maximum of 15% of the ammonia could have been in found the form of gaseous ammonia. However, samples taken from the microcosms showed no ammonia gas.

Combined effects on ammonia removal

All rate data were pooled and a single multiplicative Monod model was developed. Because the pH did not remain constant in the experiments, a pH term traditionally used to describe pH impacts on nitrification was added to the model.

Additionally, a term to account for the impact of changing temperatures was also added.

The temperature term was constructed so that both the increase and decrease in rates due to changing temperature could be modeled. The complete multiplicative Monod model is presented in equation 4.3.

⎛⎞ n ⎜⎟⎛⎞T ⎛⎞kC %OT⎛⎞⎜⎟1− 1 N ⎜⎟ 2 ⎝⎠ t ⎛ ⎞ (4.3) R= ⎜⎟ 2 ⎜⎟e⎜zp− H ⎟ KC+ ⎜⎟%O ⎜⎟t 11+ 0() ⎝⎠ SN KO ++% 2 ⎝⎠ ⎝ ⎠ ⎜⎟O2 2 ⎝⎠KI

where, T is temperature (oC), pH is the average system pH, and t, n and z are fitting parameters.

Sigma Plot was used to fit all data to this equation. Table 4-1 presents the values of the parameters found. A fairly high correlation coefficient was determined (r2=0.75), indicating a good fit. All model parameters were found to be statistically significant (p <

122 0.05). A graph of the predicted rates using the Monod model vs the measured removal rates is shown in Figure 4-4a.

Because it is highly unlikely that 100% oxygen will ever be added to landfills, a multiplicative Monod model based only on the lower oxygen concentrations was also developed. The only difference in this model structure is the absence of the oxygen inhibition term. Using Sigma Plot, the data (excluding the data from the 100% oxygen studies) were fitted to the Monod model; parameter results are presented in Table 4-1. All parameters in this model were found to be significant (p < 0.05). The predicted vs observed rates for this model are shown in Figure 4-4b. The results at 22 and 45oC were not found to be statistically different (75% confidence) from one another, while significant difference was found when comparing rates at 35 and 45oC (75% confidence) and when comparing 22 and 35oC (80% confidence), indicating that temperature is the most significant factor in the nitrification process and that operating at different temperature levels does in fact change the ammonia removal rates. The significance of temperature, however, may be influenced by the impact it has on the oxygen levels in the liquid-phase.

Caution must be taken when applying either of these multiplicative equations to other systems. Only three different temperatures were evaluated. Based on the model, the temperature in which the maximum removal rate occurs is 34oC, however, it may actually be between 35 and 45oC. Also, the pH values were not held constant in this study, thus the pH effect in this model may be based on compounding impacts of other influential parameters (i.e. oxygen and ammonia concentration). Differences in mass transfer are also a potential limitation of this model when applying it to other systems. It is also

123 important to note that toxicity of ammonia at levels as high as 1,000 mg N/L was not observed.

0.8 g / 0.7 R2 = 0.64

g N

y) 0.6 m

a ( d

s 0.5 e t

a 0.4 aste- 0.3 d R y w e t r

c 0.2 d i d

e 0.1 r

P 0 0 0.2 0.4 0.6 0.8 Measured Rates (mg N/g dry waste-day)

(a) Monod model including oxygen inhibition

0.8 y 0.7 R2 = 0.69 g dr / 0.6

g N 0.5 m ( ay)

d s 0.4

e e- t

a 0.3 ast d R w 0.2 e t 0.1 ic

d

e 0 r P 0 0.2 0.4 0.6 0.8 Measured Rates (mg N/g dry waste-day)

(b) Monod model excluding data at 100% oxygen

Figure 4-4. Predicted versus observed rates for the multiplicative Monod fits: (a) Monod model including oxygen inhibition and (b) Monod model excluding data at 100% oxygen.

124 Based on the relationships derived in this study, guidance can be given with respect to implementation of in-situ nitrification processes the field-scale. Temperature is an important parameter in in-situ nitrification processes because temperatures in landfills may be high. Higher temperatures have an adverse impact on the available D.O.; thus, at higher temperatures, more air needs to be added to maintain higher gas-phase oxygen concentrations if high ammonia removal rates are desired. Additionally, nitrification is sensitive to pH. At a pH of 6.5, the rate is 35% lower than when the pH is 7.5. Below a pH of 6.5, the rates decrease by approximately 5% for each 0.1 drop in pH. When nitrifying leachate with 1,000 mg N/L, the pH will likely drop, thus injection rates should be reduced or a pH buffer added to ensure nitrification is not inhibited.

References

Beccari, M., Di Pinto, A.C., Ramadori, R., and Tomei, M.C. (1992). Effects of Dissolved Oxygen and Diffusion Resistances on Nitrification Kinetics, Water Research 26(8), 1099-1104.

Berge, N.D., Reinhart, D.R., Dietz, J., and Townsend, T. (2006). In-Situ Ammonia Removal in Bioreactor Landfill Leachate, Waste Management 26(4), 334-343.

Berge, N.D., Reinhart, D.R., and Townsend, T. (2005). The Fate of Nitrogen in Bioreactor Landfills, Critical Reviews in Environmental Science and Technology 35, 365-399.

Bougard, D., Bernet, N., Cheneby, D., and Delgenes, J.P. (2006). Nitrification of a High- Strength Wastewater in an Inverse Turbulent Bed Reactor: Effect of Temperature on Nitrite Accumulation, Process Biochemistry 41, 106-113.

Burton, S. A. Q. and Watson-Craik, I.A. (1998). Ammonia and Nitrogen Fluxes in Landfill Sites: Applicability to Sustainable Landfilling, Waste Management and Research 16, 41-53.

125 Charley, R.C., Hooper, D.G. and McLee, A.G. (1980). Nitrification Kinetics in Activated Sludge at Various Temperatures and Dissolved Oxygen Concentrations, Water Research 14, 1387-1396.

El-Zanfaly, H. (1981). Temperature as an Environmental Factor: Its Effect on Nitrification, Environment International, 5(3), 145-148.

Grady, C.P., Daigger, G.T. and Lim, H.C. (1999). Biological Wastewater Treatment, Second Edition, Marcel Dekker, Inc. New York, NY.

Hanashima, M. (1999). Pollution Control and Stabilization Process by Semi-Aerobic Landfill Type: The Fukuoka Method, In Proceedings of Sardinia 99 Seventh Waste Management and Landfill Symposium, Cagliari, Italy.

Harremoes, P. (1978). Biofilm Kinetics, in Water Pollution Microbiology, Volume 2, edited by R. Mitchell, John Wiley and Sons, 71-109.

Holman, J.B. and Wareham, D.G. (2005). COD, Ammonia and Dissolved Oxygen Time Profiles in the Simultaneous Nitrification/Denitrification Process, Biochemical Engineering Journal 22, 125-133.

Jokela, J.P.Y., Kettunen, R.H., Sormunen, K.M. and Rintala, J.A. (2002). Biological Nitrogen Removal from Municipal Landfill Leachate: Low-Cost Nitrification in Biofilters and Laboratory Scale in-situ Denitrification, Water Research 36(16), 4079-4087.

Jones, G.L. and Paskins, A.R. (1982). Influence of High Partial Pressure of Carbon Dioxide and/or Oxygen, Journal of Chem. Tech. Biotechnol. 32, 213-223.

Khalil, K., Mary, B., and Renault, P. (2004). Nitrous Oxide Production by Nitrification and Denitrification in Soil Aggregates as Affected by O2 Concentration, Soil Biology and Biochemistry, 36 687-699.

Kjeldsen, P., Barlaz, M.A., Rooker, R., Baun, A., Ledin, A. and Christensen, T.H. (2002). Present and Long-Term Composition of MSW Landfill Leachate: A Review, Critical Reviews in Environmental Science and Technology, 32(4), 297-336.

Micks, P., Aber, J.D., Boone, R.D., and Davidson, E.A. (2004). Short-Term Soil Respiration and Nitrogen Immobilization Response to Nitrogen Application in Control and Nitrogen-Enriched Temperate Forests, Forest Ecology and Management in press.

Mummey, D.L., Smith, J.L., and Bolton, H. (1994). Nitrous Oxide Flux from a Shrub- Steppe Ecosystem: Sources and Regulations, Soil Biology and Biochemistry 26(2), 279-286.

126 Nommik, H. and Vahtras, K. (1982). Retention and Fixation of Ammonium and Ammonia in Soils, In Nitrogen in Agricultural Soils, Edited by F.J. Stevenson, Madison, WI.

Onay, T.T. and Pohland, F.G. (1998). In-Situ Nitrogen Management in Controlled Bioreactor Landfills, Water Research 32 (5), 1383-1392.

Park, H.D. and Noguera, D.R. (2004). Evaluating The Effect Of Dissolved Oxygen On Ammonia-Oxidizing Bacterial Communities In Activated Sludge, Water Research 38, 3275-3286.

Pohland, F.G. (1992). Anaerobic Treatment: Fundamental Concepts, Applications, and New Horizons, In Design of Anaerobic Processes for the Treatment of Industrial and Municipal Wastes, edited by Malina and Pohland.

Price, G.A., Barlaz, M.A. and Hater, G.R. (2003). Nitrogen Management in Bioreactor Landfills, Waste Management 23(7), 675-688.

Stewart, P.S. (1998). A Review of Experimental Measurement of Effective Diffusive Permeabilities and Effective Diffusion Coefficients in Biofilms, Biotechnology and Bioengineering 59(3), 261-272.

Stenstrom, M.K., and Poduska, R.A. (1980). The Effect of Dissolved Oxygen Concentration on Nitrification, Water Research 14, 643-649.

Sjoberg, R.M. and Persson, T. (1998). Turnover of Carbon and Nitrogen in Coniferous 15 Forest Soils of Different N-Status and Under Different NH4-N Application Rate, Environmental Pollution 102SI, 385-393.

Venterea, R. and Roltston, D. (2000). Mechanisms and Kinetics of Nitric and Nitrous Oxide Production During Nitrification in Agricultural Soil, Global Change Biology 6, 303-316.

Willers, H.C., Derikx, P.L.J., ten Have, P.J.W. and Vijin, T.K. (1998). Nitrification Limitation in Animal Slurries at High Temperatures, Bioresource Technology 64(1), 47-54.

Youcai, Z., Hua, L., Jun, W. and Guowei, G. (2002). Treatment of Leachate by Aged- Refuse-Based Biofilter, Journal of Environmental Engineering 128(7), 662-668.

Zhu, S. and Chen, S. (2002). The Impact of Temperature on Nitrification Rate in Fixed Film Biofilters, Aquacultural Engineering 26, 221-237.

127 CHAPTER 5

ENGINEERING SIGNIFICANCE

Introduction

Determining what the results of the in-situ nitrification experiments may mean with respect to field-scale implementation is an important aspect of this study and the first step in developing appropriate field-scale guidance. With field-scale guidance, larger scale studies can be more knowledgeably designed and operated to validate the laboratory results obtained from this study. Such a study is planned for Spring 2006 at the New

River Regional Landfill.

In order to provide design information for a field-scale study, a simple mass balance model was constructed in FORTRAN to forecast the fate of ammonia injected into a nitrifying portion of a landfill. Based on model results, an economic analysis of the in-situ treatment method was conducted and compared to current ex-situ leachate treatment costs.

Model Description

A mass balance approach was used to model nitrification occurring within the landfill using the nitrification rate equation determined from the experiments conducted

128 in this study (Chapter 4, Table 4-1). The model was constructed to allow the user to determine which ammonia removal rate equation (with and without oxygen inhibition) is used in the rate determination. The model code can be found in Appendix E.

The purpose of the model was to determine design parameters that would provide information necessary to implement a field-scale plan including: volume of ammonia-rich leachate that can be treated, how long it would take to treat it, and what depth of the landfill would be required to achieve treatment. To accomplish this, many simplifying assumptions were made, including:

1. The landfill is modeled as multiple CSTRs in series.

2. Flow through the landfill is uniform. Liquid follows a wetting front at the speed

of the saturated hydraulic conductivity of the waste. In a landfill, this will not

necessarily be the case. For the most part, the landfill will not be saturated and

thus the liquid will move at a slower rate. Using the saturated hydraulic

conductivity will result in a more conservative design. The injection rate of

leachate was chosen to ensure that saturated conditions would not exist (i.e., less

than saturated hydraulic conductivity).

3. Simultaneous air and liquid flow through the waste mass are possible. The

amount of liquid present in each CSTR is never at saturation, which allows for

concurrent air and liquid flow. As stated previously, the injection rate of leachate

was chosen to ensure that saturated conditions would not exist.

129 4. Uniform flow is also assumed, however, because the waste is heterogeneous,

there will be preferential flow pathways allowing some liquid to move at a faster

rate. To account for the possibility of short circuiting, relatively high hydraulic

conductivities were used (10-2 and 10-3 cm/sec).

5. Initially, each CSTR contains an excess amount of leachate (i.e., it is above field

capacity (field capacity is the maximum amount of water that can be held against

the pull of gravity), but not saturated). As leachate is injected into the landfill,

flow moves at the rate of the saturated hydraulic conductivity. Once injection

stops, flow percolates through the landfill at a slower rate. Flow continues to

leave each CSTR until the volume of liquid remaining is equivalent to the volume

associated with field capacity. However, unsaturated flow equations were not

used to simplify the model. A flowchart describing the liquid flow processes used

in the model is presented in Figure 5-1.

6. The model assumes flow is injected uniformly over the entire impacted area.

Thus, the first CSTR in this model is located just below the point of leachate

injection.

7. The model assumes that the radius of influence of the leachate being injected is

equal to or less than the radius of influence of the injected air. According to Jain

(2005a), the minimum radius of influence of injected leachate is 8 m. Powell

130 (2005) found that the radius of influence of air at a flowrate of 1 m3/min was

approximately 17.4 m.

8. Leachate injection rates were chosen based on rates provided from field-scale data

(Jain, 2005a). Injection was assumed to occur at a single vertical well. Injection

rates both above and below the rate reported by Jain (2005a) were used to

describe the impact of leachate injection on ammonia removal processes.

Figure 5-1. Flow simulation module in the model.

131 9. The only reaction involving nitrogen in this model is nitrification. In reality, there

may be several simultaneous processes occurring. Ammonification is assumed

negligible. Note that it is assumed that this process will be occurring in older

areas of a landfill in which most of the nitrogen present as proteins have already

hydrolyzed and fermented. The experiments conducted in old waste confirmed

this phenomenon. Volatilization of ammonia is also neglected. In the laboratory-

scale experiments, no volatilization was observed. In a landfill, it is possible for

some gaseous ammonia to be present if the pH, ammonia concentration and

temperature are high enough. Denitrification of some of the nitrate produced will

occur, as suggested in the lab experiments, however, it is not included in the

model, nor is nitrate production simulated.

10. It is assumed conditions in the field will be similar to those in the laboratory. No

scale-up factors were used in the model and mass transfer of ammonia was

assumed to be the same. In the field, mass transfer will most probably increase

because of leachate and gas movement not present in batch microcosm studies. A

safety factor of 5 was applied to all model results to create a more conservative

estimate.

11. No aerobic degradation of the waste is included in the model. Because the

simulations are occurring in older waste environments, it is assumed the majority

of the organic carbon has been degraded and little additional waste degradation is

occurring.

132 12. The oxygen concentration within the landfill is the same everywhere. In reality,

oxygen concentration will vary throughout the landfill, ranging from 21 (near

injection wells) to 0%. This model was not constructed to simulate oxygen

consumption nor was it constructed to handle spatially differing oxygen

concentrations. To evaluate what may happen at different average gas-phase

oxygen concentrations, the model was run for several scenarios, each with a

different oxygen concentration. Additionally, it was assumed that the oxygen

concentrations chosen are possible to maintain within a field-scale landfill and

were chosen based on available field data (Hudgins, 2006; Powell, 2005).

13. Similar to the case with oxygen, temperature within the landfill is assumed to be

constant. Several iterations were conducted evaluating the impact of different in-

situ temperatures.

14. The pH was also assumed to be constant within the landfill, which may not be the

case. Several simulations were conducted in which different pH values were used

to determine how pH impacts nitrification. It is also important to note that

although the rate equation predicts some nitrification will occur at low pH values

(i.e., 5), in fact nitrification would be severely inhibited. Thus when using the

model, care must be taken when analyzing results in low pH areas.

Model input parameters include:

133 • All constants required in the multiplicative Monod model

• Average gas-phase oxygen concentration in landfill

• Average temperature of the waste

• Average pH present in the landfill

• Time to stop leachate injection

• Saturated hydraulic conductivity

• Dry specific weight of the waste

• Concentration of ammonia in the landfill prior to leachate injection

• Concentration of ammonia in the leachate being injected to the landfill

• Leachate injection rate

• Radius of influence for each injection well

• Field capacity of the waste

Equations 5.1 – 5.5 depict how ammonia concentration changes in each CSTR over time. First, a mass balance on the ammonia concentration around each CSTR is conducted, as shown in Equation 5.1. Equations 5.2 and 5.3 present the rate equations both with removal rates inhibited by high oxygen values and those that are not, respectively. Because there is not expected to be 100% oxygen in the gas-phase in landfills, the equation with oxygen inhibition (Equation 5.2) is not needed. Note that the rate equation is multiplied by the specific weight of the waste and volume of the landfill.

To determine the liquid volume in each CSTR (Vlq), a volume balance is conducted

(equation 5.4) during each time step. The ammonia concentration for the next time step is given in equation 5.5.

134

VVin tt−1 out CCNN− − R dCN ∆∆tt = (5.1) dt Vlq

⎛⎞7.66 t ⎛⎞T ⎛⎞1.17C ⎜⎟%1OT⎛⎞⎜⎟1− ⎛⎞ Re = N ⎜⎟ 2 ⎜⎟⎝⎠ 33.8 P V inhib ⎜⎟t 2 ⎜⎟⎜⎟()6.13− pH ()DLF (5.2) 167 + CN ⎜⎟%O2 33.8 ⎝⎠11+ 0 ⎝⎠⎜⎟2.84 ++%O ⎝⎠ ⎝⎠2 206

4.56 t ⎛⎞T ⎛⎞⎜⎟1− ⎛⎞0.90CON ⎛⎞% 2 T⎝⎠34.7 ⎛⎞1 Re= ⎜⎟ t ⎜⎟ ⎜⎟ ⎜⎟ 5.40− pH ()PDLV F (5.3) 156 ++CO2.46 % ⎜⎟34.7 () ⎝⎠N ⎝⎠2 ⎝⎠⎝⎠11+ 0

t−1 (5.4) VVlq =−in Vout +Vfc +Vlq

tt+1 CCNN=+ d CN (5.5)

where, CN is the ammonia concentration (mg N/L), t is time (min), Vin is the volume of liquid entering each CSTR (L), Vout is the volume exiting the CSTR (L), Vfc is the volume of liquid associated with the field capacity of the waste (L), R is the ammonia removal rate (mg N/day), PD is the dry specific weight of waste (g/L), VLF is the volume of the CSTR (L), dCN is the change in ammonia concentration (mg N/L) (calculated by rearranging equation 5.1), and %O2 is the gas-phase concentration of oxygen (%).

Simulation of Laboratory Experiments

Once constructed, the model was used to simulate the conditions of the microcosm experiments to validate that it was correctly predicting what occurred in the

135 laboratory. The microcosms were batch experiments, thus the flow module was set to zero in these simulations. The microcosm specific weight, initial ammonia concentrations and temperatures were used as initial parameters. An average system pH of 6.8 was used in each simulation. The gas-phase oxygen concentrations used were averages calculated from each test. Simulations were conducted using the ammonia removal rate equation both with and without inhibition due to high oxygen. Figures 5-2 to 5-4 present examples of the results from three different simulations. As shown, both of the predicted ammonia concentration trends are very close to those actually measured in the laboratory. The small difference between the two predicted trends is most probably because the oxygen levels were well below the inhibitory levels (100% oxygen). Results from these simulations confirm that the model is accurately predicting what happened in the microcosms.

400 Temperature = 22C L) / Raw Data Oxygen = 5% 350 g N Predicted Trend m (

300 n

o Predicted Trend w / Inhib. i

t 250 a r t

n 200 e

nc 150 o

C 100 a i n

o 50 m

m 0 A 0123456 Time (Days)

Figure 5-2. Comparison of measured and predicted ammonia concentration trends in the microcosm experiment conducted at 22oC and 5% oxygen.

136 500 Raw Data Temperature = 35C

) Oxygen = 5% L

/ 450 Predicted Trend

g N 400 Predicted Trend w / Inhib. 350 n (m o i 300 t a

tr 250 n e

c 200 n o 150 C a i 100 n o

m 50 m

A 0 0123456 Time (Days)

Figure 5-3. Comparison of measured and predicted ammonia concentration trends in the microcosm experiment conducted at 35oC and 5% oxygen.

500 Raw Data Temperature = 45C Oxygen = 17%

L) 450 Predicted Trend /

g N 400 Predicted Trend w / Inhib.

m 350 n ( o i 300 t a r

t 250 n e 200 nc o 150 C

a 100 ni 50 mmo

A 0 0 0.5 1 1.5 2 2.5 3 3.5 4 Time (Days)

Figure 5-4. Comparison of measured and predicted ammonia concentration trends in the microcosm experiment conducted at 45oC and 17% oxygen.

137 Model Results

Several model simulations were run evaluating how different operational parameters impact ammonia removal. Table 5-1 presents the conditions that were varied.

From an operations perspective, the leachate injection rate and % oxygen in the gas-phase are parameters that can be controlled to achieve desirable in-situ nitrification rates.

It is important to note that the specific weight used in the model is a dry, compacted specific weight because the removal rates were normalized in terms of mass of dry waste. Dry specific weight can be converted to wet specific weight using the moisture content of the waste. A typical wet, compacted specific weight of fresh municipal solid waste (MSW) is 600 g/L or 1000 lb/yd3 (typical dry specific weight is

480 g/L or 800 lb/yd3). Fresh MSW has large particle sizes; older waste has smaller particles sizes and thus a higher specific weight, closer to that of approximately 700 g/L

(1200 lb/yd3, wet specific weight). Assuming a typical moisture content of waste (20%, by weight), a dry specific weight of 570 g/L is equivalent to a dry specific weight similar to older waste (960 lb/yd3, dry specific weight). The dry specific weight of the waste used in the microcosm experiments was 230 g/L (480 lb/yd3, wet specific weight). Simulations will be conducted using both 480 and 570 g/L to simulate the impact of different particle sizes.

138 Table 5-1. Parameters Varied in Model Simulations

Leachate Dry %O2 Temp. Hydraulic pH Ammonia Injection Specific (C) Conductivity Concentration Rate (L/day) weight (cm/sec) (mg N/L) (g/L) 4000 600 0 22 .01 6 1000 3000 500 1 35 .001 6.5 1500 400 2 45 .0001 7 300 5 10 21

Initially, simulations were conducted to evaluate how long injection of ammonia- rich leachate could continue without it reaching the bottom of the landfill and thus the environment. It was found that while nitrification was occurring, pumping of the leachate could continue for long periods because the mass rate of removal was similar to the mass rate of injection.

Important parameters associated with the treatment process determined from these simulations are the ammonia removal time and depth of waste required to remove the ammonia. The time it took for ammonia to be removed to less than 1 mg N/L was less than five days. Because this is a short time with respect to the pumping time, the depth of waste required to remove the ammonia was found to be the controlling parameter. The depth of landfill required is defined as the depth of waste necessary for the final ammonia concentration to be less than 1 mg N/L. Varying the conditions as provided in Table 5-1, the depth for each environment was determined and compared. Figures 5-5 to 5-9 present results from the simulations.

Based on the results, the higher the gas-phase oxygen concentration, the lower the impacted depth, as was expected because nitrification rates increase with available oxygen. Generally, impacted waste depth increases with increasing leachate injection rate

139 (Figure 5-5), hydraulic conductivity (Figure 5-8), and decreasing pH (Figure 5-7) and dry specific weight of the waste (Figure 5-9). The impact of temperature was found to be variable; the impacted depth decreases from 22 to 35oC, but then increases between 35 and 45oC because of the nitrification rate changes caused by temperature. Also, as the specific weight increases (Figure 5-9), minor changes in required depth result.

4000 L/day 0.5 pH = 7 3000 L/day Temperature = 35C 0.45 1500 L/day Dry Density = 500 g/L 0.4 Hyd. Cond. = 0.001 cm/sec ill f 0.35 d ) n m a 0.3 L h (

d 0.25 e pt r e i

D 0.2 u q

e 0.15 R 0.1 0.05 0 1251021 % Ox yge n

Figure 5-5. The impact of leachate injection rates on the required landfill depth for various gas-phase oxygen concentrations.

140 22C 0.7 pH = 7 35C Injection = 4000 L/day 0.6 45C Dry Density = 500 g/L

l Hyd. Cond. = 0.001 cm/sec l 0.5 dfi ) n 0.4 La d th (m e p r i e 0.3 D qu e 0.2 R

0.1

0 1 2 5 10 21 % Oxygen

Figure 5-6. The impact of temperature on the required landfill depth for various gas- phase oxygen concentrations.

6 0.6 Injection = 4000 L/day 7 Hyd. Cond. = 0.001cm/sec 0.5 7.5 Temperature = 35C Dry Density = 500 g/L

l l 0.4 dfi ) n La d th (m 0.3 e p r i e D

qu 0.2 e R 0.1

0 1 2 5 10 21 % Oxygen

Figure 5-7. The impact of pH on the required landfill depth for various gas-phase oxygen concentrations (Note that the 7 and 7.5 trends do not differ).

141 pH = 7 1.2 Temperature = 35C Dry Density = 500 g/L 1 Injection = 4000 L/day th

p 0.01 cm/sec e

D 0.8

l 0.001 cm/sec l dfi ) 0.0001 cm/sec n a

(m 0.6 L d e r i

u 0.4 q e R 0.2

0 1 2 5 10 21 % Oxygen

Figure 5-8. The impact of hydraulic conductivity on the required landfill depth for various gas-phase oxygen concentrations.

600 g/L pH = 7 0.8 500 g/L Temperature = 35C Hyd. Cond. = 0.001 cm/sec 0.7 400 g/L

Injection = 4000 L/day

h 300 g/L t 0.6 p e D ll

i 0.5 f d ) n a

(m 0.4 L d e 0.3 ir u q e 0.2 R

0.1

0 12345 % Oxygen

Figure 5-9. The impact of dry specific weight of waste on the required landfill depth for various gas-phase oxygen concentrations.

142 Optimizing Treatment Depth

At all oxygen concentrations, ammonia was removed in a fairly short depth of waste. In field-scale landfills, the oxygen concentrations will be variable (from 21 to 0%).

Scenarios evaluating the worst-case environmental conditions and those with more typical conditions expected were conducted. Table 5-2 presents the conditions evaluated and the resulting required depth of waste.

The worst-case scenario estimate of the depth of waste required to attain a final ammonia concentration less than 1 mg N/L can be used to ensure a conservative design.

A depth of 1.4 meters was determined under the worst-case conditions. Because the model assumes the environment in the field is exactly the same as that in the laboratory, this depth estimate may be small because the laboratory experiments were operated under optimal conditions. In the field, ammonia and oxygen mass transfer issues are not known, thus their impact on nitrification rates is also unknown. To account for scale-up issues

(such as those associated with mass transfer), a factor of safety of five was applied, yielding a required treatment depth of approximately 7.2 meters. Note that this is the depth below the point of leachate injection.

Treatment depth is greatly influenced by the hydraulic conductivity of the waste.

In the worst-case scenario, a 10-2 cm/sec saturated hydraulic conductivity was used. This is a very high value and it is unlikely that flow in the landfill will be that fast. Typical hydraulic conductivities range from 10-4 to 10-7 cm/sec (McCreanor and Reinhart, 2000).

As shown in Table 5-2, the more typical conditions (with saturated hydraulic conductivities of 10-3 and 10-4) result in lower depth requirements (2.2 and 1.7 meters, respectively). Additionally, it is important to note that the specific weight used in the

143 worst-case scenario is not likely for older landfills (480 g/L). In the more typical scenarios, a specific weight of 570 g/L was used.

Table 5-2. Parameters Used in Model Scenarios. Scenario Worst-Case More Likely Most Likely Conditions Conditions Conditions pH 6.0 6.5 6.5 Temperature (oC) 22 35 35 Average % oxygen 1 1 1 Hydraulic Conductivity (cm/sec) 10-2 10-3 10-4 Dry specific weight (g/L) 475 570 570 Leachate injection rate (L/m2-day) 20 20 20 Predicted required depth (m) 1.43 0.43 0.34 Depth with a factor of safety (m) 7.2 2.2 1.7

To minimize treatment depth, the following can be done:

1. Add more oxygen to keep the gas-phase oxygen concentration high. At

10% oxygen, for example, the required treatment depth (assuming worst-

case conditions and a safety factor of five is applied) decreases to

approximately 3.5 meters.

2. Apply the nitrification process in areas of the landfill in which the pH is

above 6.5 and the temperature is above 22oC and below 45oC.

3. Lower leachate injection rates could also be practiced. If a rate of 15

L/m2-day is injected (assuming worst-case conditions), the required

144 treatment depth remains similar (7.2 meters), however, at a rate of 8 L/m2-

day, the depth required decreases to 5.4 meters.

It is suggested that in-situ nitrification occur later in the life of a bioreactor landfill, serving more as a polishing step. At this time the organic fraction of the waste has been degraded, to minimize the oxygen demand and high temperatures imparted by heterotrophic microbial action. It is important to note that even in older waste environments there will still be some oxygen demand imparted by the heterotrophs. In the simulations conducted in this work, it was assumed that the only process using oxygen was nitrification. Thus, when scaling up to field-scale, more oxygen may be required to satisfy both the nitrifiers and heterotrophs. Because adding air to a landfill is expensive, addition to smaller, dedicated nitrification zones within the landfill is suggested. Thus, in- situ nitrification processes (air injection) would begin after leachate recirculation has been practiced and the amount of biodegradable organic material in the leachate and waste is stable with respect to time.

In-Situ Denitrification

Although in-situ denitrification was not extensively explored in the laboratory experiments, simultaneous ammonia and nitrate removal was observed. In the field, this same phenomenon is expected to occur, perhaps to a greater extent than that observed in the laboratory experiments because of the potential for a larger number of micro-anoxic regions within the field-scale landfill. When conducting laboratory experiments with low gas-phase oxygen concentrations (0.7%), the most complete nitrogen removal was

145 observed, indicating that denitrification will occur in areas containing low oxygen concentrations. Thus, in-situ ammonia removal is expected to also result in removal of the majority of the nitrate. Other researchers have studied denitrification of nitrate-rich leachate in landfills and found that denitrification occurs quickly and readily (Aljarallah and Atwater, 2000; Burton and Watson-Craik, 1999; Price et al., 2003). If the nitrate is not completely removed from the leachate, the nitrate-rich leachate could be injected in newer areas of the landfill to create an anoxic area where it will be removed quickly.

Field Study Plans

The objectives of the field-scale study include:

• Demonstrate the potential for field-scale nitrification processes,

• Measure ammonia removal rates,

• Validate laboratory study results, and

• Evaluate scale-up issues.

To meet the objectives, a small field-scale study is planned in which air will be added and parameters in both the liquid- and gas-phases will be measured. The liquid- phase parameters to be measured include: pH, alkalinity, ammonia, nitrate, sulfate, nitrite, and chemical oxygen demand. Both oxygen and nitrous oxide will be measured in the gas-phase. While operating the field study, it is also suggested that tests be conducted to determine the specific oxygen uptake rate of the waste environment. Tests evaluating

146 the mass transfer efficiency of oxygen are also important to conduct during the field study.

The study will be implemented in an older portion of the landfill in which the majority of biological activity has already occurred. Two air injection wells and one leachate injection well will be placed in the study area with eight sampling points, as shown in Figures 5-10 and 5-11. The leachate injection well will consist of a 5-cm diameter PVC pipe installed to a depth of 6 meters, with the bottom 3 meters screened.

The air injection wells will also consist of 5-cm PVC and will be installed to depths of approximately 12 meters, with the bottom 6-m screened.

Each sampling location will be equipped to monitor parameters in both the leachate and gas, as well as temperature. At each sampling point, a lysimeter (Soil

Moisture, Inc., see Figure 5-12) coupled with a thermocouple and a tube to allow for gas measurements will be installed in a 5-cm hole. The bottom of the hole will be filled with a soil slurry (where the porous tip, thermocouple and gas measurement tube will be located) and backfilled with bentonite. The depths and locations of each sample location are shown in Figures 5-10 and 5-11.

Once the field site is instrumented, leachate (with an ammonia concentration of at least 500 mg N/L) and air will be injected. Samples will be taken from each location weekly. The results from this study will be the first step in evaluating the laboratory- based implementation strategies developed for in-situ ammonia removal. Specifically, attention will be paid to ammonia removal time frames and the relation to oxygen concentration in the gas-phase, as well as changes in in-situ leachate parameters, such as nitrate, nitrite, pH, and alkalinity.

147 Lysimeter Sampling Tubes Air Injection Leachate and NH4 Injection

3 m 1.5 m 1.5 m 1.5 m 3m

3 m

4.6 m 4.6 m 6 m

6 m = Lysimeter

= Well Screen 11 m 12 m

Figure 5-10. Cross-section view of planned field-scale study.

1 m

3 m 1.5 m

2.4 m

3 m

= Leachate Injection

= Air injection = Sampling wells

Figure 5-11. Plan view of planned field-scale study.

148

Figure 5-12. Lysimeter detail (www.soilmoisture.com).

Economic Analysis

An economic analysis of the proposed in-situ nitrification process was conducted and compared to treatment costs at landfills discharging their leachate to publicly owned treatment works.

In-Situ Treatment Economics

It was assumed that any landfill choosing to implement an in-situ nitrogen removal process was already operating as a bioreactor, thus costs associated with construction of the leachate injection piping were not taken into account. Additionally, it was assumed that air injection would occur in the leachate injection wells, so installation of new wells was not required. However, the costs of blowers and piping from the blower to the injection well were included in the analysis, as it was assumed the landfill had only

149 been operating anaerobically. Costs were calculated based on leachate flow to one well and then normalized to the volume of leachate treated.

The necessary air flowrate required to completely remove the mass of ammonia added was calculated based on the stoichiometry of nitrification (4.5 g O2/g N) and the mass of nitrogen being injected into the landfill assuming the leachate contained 1,000 mg N/L. Because the transfer of gas-phase oxygen to dissolved oxygen is unknown, a safety factor of 100 was applied, corresponding to a mass transfer efficiency of approximately 1%. Mass transfer of oxygen in the field is unknown. Field-scale studies need to be conducted to determine the actual oxygen transfer efficiency. The resulting total air flowrate for each leachate injection rate was calculated to be 4.2, 3.1, and 1.6 m3/min, respectively, for leachate injection rates of 20, 15 and 8 L/m2-day. More detailed calculations can be found in Appendix G. Based on work by Jain et al. (2005b) and assuming a screen length of 6 m (air will be injected into wells 12-m deep), the flowrates cannot be applied in one well, rather air injection must occur in two wells. It is suggested that two of three wells in an injection cluster be used to inject air. Using two wells and taking into account both the headlosses in the piping and the relationship derived by Jain et al. (2005b), the required discharge pressure of the blower at each flowrate was calculated as 39, 24 and 14 kN/m2 (for each respective injection rate of 20, 15 and 8

L/m2-day). Appendix G contains more detailed information regarding these calculations.

Table 5-3 lists the equipment necessary for in-situ nitrogen removal. Table 5-4 presents the operational costs for the proposed treatment system, including details regarding their calculation. All costs and expenditures (over a design life of 10 years) were converted to a present day value using an interest rate of 8%. The total volume of

150 leachate treated over the design life was calculated based on the flowrates provided in the model simulations (20, 15 and 8 L/m2-day) to find the cost per gallon (English units are used because this is the convention for reporting treatment costs) of leachate. Results from this analysis are shown in Table 5-5.

It is important to note that these costs include ammonia removal only. In the laboratory experiments, complete nitrogen removal was achieved at times, so any additional treatment for other nitrogen species would be reduced. If any nitrate remains in the leachate, it can be easily treated by reinjecting the leachate in a biological active anoxic portion of the landfill at minimal cost; the landfill will have already been recirculating leachate so no capital costs would be incurred. Costs associated with removing other parameters (i.e., nitrate, biochemical oxygen demand, total suspended solids) were not calculated. Additionally, it is important to note that the economics of this process are dependent on the mass transfer of oxygen. In these simulations, a1% transfer efficiency was assumed. The economics of this process are sensitive to the amount of air added. Thus, differences in transfer efficiency in the field may significantly alter the total cost of the process. It is possible that the transfer efficiency in the field will be different, resulting in an increase in the cost of the process.

151 Table 5-3. Equipment Necessary for In-Situ Nitrogen Removal. Equipment Capital C ost Assumption Leachate Injection System ___ Already acquired. No (i.e., pumps, piping, wells) additional cost Air Blower $13,000, for two pumps a Must purchase. Blower life is 10 yearsb Air Injection Piping $220 Must purchase. Assume 91 meters of 5-cm diameter piping.c Life of piping is assumed to be 10 years a Costs based on verbal quote provided by Global Technology and Engineering (2006) b Design life provided by pump manufacturer. More detailed calculations can be found in Appendix G. c Cost provided by Lowes for schedule 40, 5-cm diameter pvc pipe, $6.04 per 3 meters. Additional cost was added to account for valves and fittings. More detailed calculations can be found in Appendix G.

Table 5-4. Operation Costs Associated With In-Situ Nitrogen Removal. Description Assumptions Cost ($/yr) Air Blower Electricity costs $0.10/kWha, 0.45 – 3 kW 360 – Electricity (depending on leachate injection rate) and blower 2,600 runs for 365 days/yearb Leachate Injection Electricity costs $0.10/kWha, 0.02 – 0.04 kW 10 – 30 Pump Electricity (depending on leachate injection rate) is required to pump the max flowrate for 365 days/yearc Maintenance and Maintenance costs are assumed to be 5% of the 10,730 Landfill Personnel pump cost per year. Landfill personnel will need to devote 2 hr/day to this operation (Hourly wage is $15/hr). a Electricity cost estimate from Department of Energy for 2005 (www.eia.doe.gov/fuelelectric.html) b kW estimated based on equation provided by Reynolds and Richards (1996). The backpressures used were calculated for each flowrate and are described in text. More details regarding the calculations can be found in Appendix G. c kW estimated on information provided in Hwang and Hita (1987) and Meadows (2006). A recommended discharge pressure of 34 kN/m2 was used (Townsend, 2006). More details regarding the calculations can be found in Appendix G.

152 Table 5-5. Cost Estimate of In-Situ Ammonia Removal Process. Leachate Volume Total Process Cost $/gal of Leachate Injection Rate Leachate in Present Value ($)a Treateda,b (L/m2-day) Treated (gal)b 20 75,000 3.86 x 106 0.019 15 68,800 2.89 x 106 0.024 8 64,700 1.45 x 106 0.045 a Based on capital and operating costs b English units are used because this is the convention for reporting treatment costs

Cost Comparison With Other Leachate Treatment Systems

Table 5-6 presents cost information from different types of leachate treatment systems for several different landfills. The costs represent the construction and operation of different leachate treatment methods. All costs were converted to equivalent locations and times using indices provided by Engineering News-Record (ENR). First, the costs were normalized to the same location (to Atlanta, GA) using the closest city ENR indices

(ENR, 2006b) to try to create a common basis for comparison. In some cases, a city where the treatment occurred was unknown and the value was thus not corrected. The spatial corrected values were then converted to present day costs using temporal construction cost indices provided by Engineering News Record (ENR, 2006a).

However, it is important to note that the costs provided from each source did not include detailed information, such as individual construction costs. The ENR construction cost indices were applied to the total leachate treatment costs. As shown, there is significant variability in the costs (ranges from $0.004 – $0.40/gal of treated leachate). Some of the reported costs include transportation of leachate to publicly owned treatment works, while some are purely on-site ex-situ treatment costs. Treatment costs also vary based on the volume of leachate treated. Few studies reported the actual treated leachate volume,

153 creating another level of variability in the cost information. Additionally, the reported costs reflect removal of additional leachate constituents, such as BOD, COD and total suspended solids.

Because of the inherent differences associated with each cost, it is difficult to make a clear-cut comparison between them and the cost of the proposed in-situ nitrogen removal process. However, the treatment cost associated with the in-situ method is small and at the lower end of range of reported leachate treatment costs. Based on the cost information in Table 5-6, on-site leachate treatment ranges from $0.004 – 0.18/gal, while off-site leachate treatment ranges from $0.06 – 0.40/gal. Compared to the on-site treatment costs, the costs associated with the in-situ ammonia removal process fall within and are on the lower end of the range found in the literature. When compared to treating the leachate off-site, the costs of the in-situ ammonia removal process are always significantly lower.

Assuming that the final ammonia concentration in the treated leachate is less than

1 mg N/L, no further treatment of ammonia would be required before discharging.

According to the laboratory experiments, denitrification also occurred, potentially resulting in complete nitrogen removal, reducing the requirement for any nitrogen treatment. Also, when adding air, any remaining BOD should be decreased significantly.

Thus, little treatment following an in-situ nitrogen removal process would be required, potentially allowing direct discharge of the leachate, making this an attractive economic alternative. Leachate transportation costs to a publicly owned treatment works would also be avoided, as would any risk associated with the transportation. Risk associated with accidental leakage from the landfill would also be reduced.

154 Table 5-6. Leachate Treatment Costs. Source Description of Leachate Treatment Method Leachate Treatment Cost ($/gal)a,b General Paper on Bioreactor Landfills Costs to haul and treat leachate based on information provided 0.06 – 0.14 (Levin, 1999) by Florida Solid Waste Manager Polk County (Laux, 2006) Cost to haul and treat leachate 0.11 Yolo County, CA (Yazdani et al., 2002) Cites a general disposal cost 0.017 Countryside Landfill, Illinois (RMT, 2006) Cost includes hauling and treating of waste 0.07 Darebin Parklands (Darebin Parklands, 2006) Cost to pump to sewer and treat leachate 0.42 Fauquier County Landfill (Leachate Recirc., 2006) Cost to transport and treat leachate 0.18 Essex-Windsor Regional Landfill (Cuthill and Pepper, Cost to transport and treat leachate (POTW is 15 miles away) 0.06 2001) Case Study in Georgia (Darragh, 1997) Cost of off –site treatment 0.12 Buncombe County Landfill, NC (Reinhart, 2006) Cost to transport and treat leachate 0.40 Waste Management Site (Hater, 2006) Costs are based on several ex-situ, on-site treatment methods, 0.004 – 0.017 including: ex-situ nitrification tank, SBR, and wetland units Waste Management (Waste Management, 2000) Costs are based on several different treatment schemes 0.04 – 0.18 Conestoga Landfill (Pall Corp., 2006) Costs based on on-site treatment using different biological and 0.03 – 0.05 chemical methods Orchard Hill Landfill, Michigan (Dunson, 1997) Cost of on-site treatment cost, no transportation added 0.05 a Costs were corrected to present value using the construction cost indexes provided in ENR. All costs were also corrected to the same location (Atlanta, GA) using the ENR city construction cost index. b English units are used because this is the convention for reporting treatment costs.

155 References

Aljarallah, R. S., and Atwater, J. (2000). Denitrification in Landfill Bioreactors. Waste Tech 2000, Orlando, Fl.

Burton, S. A. Q. and Watson-Craik, I.A. (1998). Ammonia and Nitrogen Fluxes in Landfill Sites: Applicability to Sustainable Landfilling, Waste Management and Research 16, 41-53.

Cuthill, J. and Pepper, T. (2001), Leachate Recirculation at the Essex-Windsor Regional Landfill: A Case Study of a Leachate Bio-Reactor Pilot Project, in Proceedings from the 6th Annual SWANA Landfill Symposium, June 18-20, 2001, San Diego, CA, 167-183.

Casey, Lee. (2006). Personal Communications.

Darebin Parklands (2006). Darebin Parklands Water Treatment, Accessed February 2006, http://www.dcmc.org.au/parklands/leachate.php.

Department of Energy (2006). Energy Information Administration, Accessed February 2006, www.eia.doe.gov/fuelelectric.html.

Darragh, T.M. (1997). Comparison of Leachate Recirculation and Bioreactor Technology, proceedings of the Solid Waste Association of North America 2nd Annual Landfill Symposium, Sacramento, CA, August 4-6, 1997.

Dunson, C. (1997). Laundering Leachate, Waste Age, May 1, 1997, http://wasteage.com/issue_19970501/.

Engineering News-Record (2006a). Construction Cost Index History (1915-2006), http://enr.construction.com/features/conEco/subs/0604-bldIndexHist.asp.

Engineering News-Record (2006b). 20 City Construction Cost Index History (1913- 2006), http://web.lexis nexis.com/universe/document?_m=4c42c0f74d36d815bf6112b18ac94e4c&_docn um=17&wchp=dGLbVtb-zSkVA&_md5=8f1065620388b073a270178bad8c9cc9.

Florida Center for Solid and Hazardous Waste (FCSHW) (1998). New River Regional Landfill Bioreactor Demonstration Project.

Global Technology Engineering (2006). Personal Communications.

Hater, G. (2006). Personal Communications.

Hudgins, M. (2006). Personal Communications.

156 Hwang, N. and Hita, C. (1987). Fundamentals of Hydraulic Engineering Systems, Prentice-Hall, Inc., New Jersey.

Jain, P. (2005a). Moisture addition at bioreactor landfills using vertical wells: Mathematical modeling and field application, PhD dissertation, Univ. of Florida.

Jain, P., Powell, J., Townsend, T. and Reinhart, D. (2005b). Air Permeability of Waste in a Municipal Solid Waste Landfill, Journal of Environmental Engineering 131(11), 1565-1573.

Leachate Recirculation System Pays for Itself. (2006). Accessed February 2006, http://www.netafim-usa-wastewater.com/Wastewater/Leachate-article-reprint.pdf.

Laux, S. (2006). Financial Performance Comparison of Traditional to Bioreactor Landfills, presentation at the SWANA E-Session, November 2, 2006.

Levin, S. (1999). Personal Communications.

McCreanor, P and Reinhart, D. (2000). Mathematical modeling of leachate routing in a leachate recirculating landfill, Water Research 34(4), 1285-1295.

Meadows, M.E. (2006). Personal Communications.

Pall Corporation (2006). On-Site Leachate Treatment at BFI's Conestoga Landfill Provides: Superior Effluent Quality and Predictable Operating Costs, Accessed February 2006, www.pall.com/articles_water_8146.asp.

Powell, J. (2005). Trace Gas Quality, Temperature Control and Extent of Influence from Air Addition at a Bioreactor Landfill, ME Thesis, University of Florida.

Price, G.A., Barlaz, M.A. and Hater, G.R. (2003). Nitrogen Management in Bioreactor Landfills, Waste Management 23(7), 675-688.

Reinhart, D.R. (2006). Personal Communications.

Reynolds, T. and Richards, P. (1996). Unit Operations and Processes in Environmental Engineering, 2nd Edition, Boston, MA: PWS Publishing Company.

RMT, Inc. (2006). Solid Waste Management Project Experience, Accessed February 2006, www.rmtinc.com/public/docs/swm_project.pdf.

Tchobanoglous, G., Theisen, H. and Vigil, S. (1993). Integrated Solid Waste Management: Engineering Principals and Management Issues, McGraw-Hill, Boston, MA.

Townsend, T. (2006). Personal Communications.

157 Waste Management (2000). The Bioreactor Landfill: A White Paper, Accessed February 2006, www.wm.com/WM/environmental/Bioreactor/WMWhitePaper.pdf.

Yazdani, R., Kieffer, J., and Akau, H. (2002). Full Scale Landfill Bioreactor Project at the Yolo County Central Landfill: Final Report, California Integrated Waste Management Report (www.yolocounty.org/recycle/bioreactor.htm).

158 CHAPTER 6

CONCLUSIONS AND RECOMMENDATIONS

Conclusions

Results demonstrate that in-situ nitrification and denitrification are feasible in decomposed aerated solid waste environments at various gas-phase oxygen concentrations. Additionally, results indicate the potential for simultaneous nitrification and denitrification (even under low biodegradable C:N conditions) in field-scale bioreactor landfills is significant due to the presence of both aerobic and anoxic areas.

Unacclimated solid waste environments do not appear to be an issue, as ammonia was removed. Rate data fit well to the Monod equation (r2 = 0.79 and 0.91, for both acclimated and unacclimated conditions, respectively) with specific rates of removal of

0.196 and 0.117 mg N/day-g dry waste and half-saturation constants of 59.6 mg N/L and

147 mg N/L for acclimated and unacclimated wastes, respectively. Although specific rates of ammonia removal in the unacclimated waste are lower than in the acclimated waste, a relatively quick start-up of ammonia removal was observed in the unacclimated waste. Acclimation also does not appear to take long periods of time.

Results also indicate that ammonia removal can readily occur at various gas-phase oxygen levels and over a range of temperatures (22, 35 and 45oC). Slowest rates occurred with lower gas-phase oxygen concentrations. All rate data, except at 45oC and 5%

159 oxygen, fit well (r2 = 0.75) to a multiplicative Monod equation with terms describing the impact of oxygen, pH, temperature and ammonia concentration. The ammonia half- saturation values are relatively high when compared to those generally found in wastewater treatment, suggesting that the rate may be limited by the mass transfer of either oxygen or ammonia. Additionally, as the temperature increases, the ammonia half- saturation value also increases, suggesting mass transfer limitations increase with temperature. The highest oxygen half-saturation value was found at 45oC, as was expected, because as temperature increases, the dissolved oxygen concentration in the liquid-phase decreases, potentially creating an oxygen limitation. At higher temperatures, the sensitivity to gas-phase oxygen increases. A multiplicative Monod model was developed that contains terms describing the impact of gas-phase oxygen, temperature, pH and ammonia concentration on ammonia removal. Other important observations include:

• Ammonia toxicity at levels as high as 1,000 mg N/L did not occur during any

of the experiments,

• Nitrous oxide is formed, but it is unclear whether the gas will exit the landfill

before being converted entirely to nitrogen gas,

• At 100% oxygen, the ammonia removal rates are inhibited either due to the

high oxygen partial pressure or because of a lack of acclimation to the high

oxygen levels,

160

• Temperature is a significant factor in in-situ removal; removal can occur at

45oC, although the rate is slower than that at 35oC,

• Ammonia removal does occur at low gas-phase oxygen concentrations (0.7%

oxygen), but it slower than at higher levels, and

• pH is a significant factor in the nitrification process.

The laboratory results demonstrate denitrification can occur in older portions of the landfill, providing valuable insight as to where nitrification and denitrification can occur in landfills. Integrating the multiplicative Monod equation with a mass balance model, it was found that in-situ ammonia removal could occur while pumping leachate in the landfill, allowing for removal to occur quickly. However, validation of the laboratory results with a field-scale study is needed. Mass transfer in a field-scale system will be much different than that in the laboratory. Tests need to be conducted to evaluate scale- up issues before routine in-situ treatment can occur.

Field-Scale Implementation Recommendations

It is suggested that in-situ nitrification occur later in the life of a bioreactor landfill, serving more as a polishing step. At this time the organic fraction of the waste has been degraded to minimize the oxygen demand and high temperatures imparted by heterotrophic microorganisms. Because adding air to a landfill is expensive, addition to

161 smaller, dedicated nitrification zones within the landfill is suggested. Thus, in-situ nitrification processes should occur after the organic fraction in the leachate and waste has stabilized with respect to time.

Based on the laboratory experiments, in-situ nitrification should occur in older portions of the landfill in which the pH is near neutral, the temperature is 45oC or lower and there is gas-phase oxygen present. Low concentrations of oxygen still resulted in ammonia removal. The higher the oxygen concentration, the faster the removal.

However, adding air can be expensive, thus adding less air may be more economically attractive. When injecting leachate with high ammonia concentrations (> 1,000 mg N/L), it may be necessary to inject a buffer solution or inject leachate at a slower rate.

Monitoring of the pH in the leachate exiting the landfill is important to gage whether or not treatment within that area can continue effectively.

The laboratory results suggest that while nitrifying, a portion of the produced nitrate will be converted to nitrogen gas. If complete nitrogen removal does not occur

(i.e., nitrate is not completely removed) in the aerated area, the nitrate rich leachate could be injected in other biologically active anoxic areas. Other researchers have found in-situ denitrification to occur readily, thus this is not expected to be problematic. Based on model results, one area of the landfill, if properly buffered (either naturally or artificially), could nitrify effectively during continuous leachate injection for over a year.

Validation of laboratory results is necessary to confirm the model results. When implementing this process at the field-scale, mass transfer issues may significantly alter the nitrification kinetics.

162 APPENDIX A

MATERIALS AND METHODS

163 Introduction

Two types of experiments were conducted: a waste acclimation reactor and microcosm studies. The waste acclimation reactor was operated to provide a waste source containing nitrifying and/or denitrifying microorganisms for parallel microcosm studies, as depicted in Figure A-1. Acclimated waste is defined as waste that has been exposed to a nitrifying microbial population and the temperature being tested, thus is capable of removing ammonia concentrations as high as 1000 mg N/L at each temperature. Additionally, the waste acclimation process was used to demonstrate the potential for nitrification and denitrification processes to occur prior to implementing the microcosm studies. The microcosm studies were operated with the purpose of quantifying the kinetics of the nitrification and/or denitrification process under different environmental conditions.

Waste Acclimation Process

Two waste acclimation reactors were operated. The first process was operated for

984 days and consisted of a 133-L reactor designed to allow for leachate draining and recirculation, air addition, and gas sampling (see Figure A-2). The second reactor was designed similarly, however, was approximately 110 L and was placed in a water bath to more effectively control temperature of the waste (see Figure A-3); it was operated for

330 days. To prevent clogging of the leachate drain, a layer of gravel was placed at the bottom of each reactor.

164 Ammonium MLSS Off Gases Bicarbonate NH3

Microcosms

Composted

Waste Acclimated ~50 lbs Waste Air Injection

Leachate COD, pH, NO2, NO3, NH3, SO4, alkalinity

Figure A- 1. Schematic depicting the relationship between the waste acclimation process and microcosm studies.

Figure A- 2. Waste Acclimation Process #1.

165 Water Bath

Reactor with

Compost

(a) Reactor (b) Water bath schematic

Figure A- 3. The Second Waste Acclimation Process: (a) Reactor and (b) water bath.

Digested municipal solid waste (MSW) was obtained from an aerobic MSW compost facility located in Sumter County, Florida, USA and was used as the waste source. This particular waste source was chosen because older waste environments were desired. Old waste is defined as waste that is highly degraded; in which the readily biodegradable materials have been degraded, with mostly recalcitrant organics remaining.

The composition of the waste accepted by the facility is shown in Table A-1 (Florida

Department of Environmental Protection, 2000). All waste received by the facility first goes through a materials recovery facility in which materials that have recycling value are removed (i.e. plastic, paper, metals, glass). The process is not 100% efficient, thus a large amount of plastic, paper, and glass remain in the waste. Once through the sorting process, the waste is combined with biosolids from a wastewater treatment facility and placed in an aerated vessel. The mixture remains in the composting vessel for at least 72-

166 hours. When the compost is removed, it is placed in a holding area. The waste used in this study has remained in that holding area for at least one year.

Table A- 1. Typical Waste Composition. Waste Constituent % Composition of waste, by weight Paper 22 Plastics 8 Textiles 2 Food Waste 5 Yard Trash 9 Glass 3 Metals 15 Construction and Demolition 25 Debris Misc. 11

The aerobic reactor was initially seeded with 14 kg of digested MSW (i.e. compost). Six hundred grams of approximately 4-cm long wood chips were added to the reactor to promote air distribution throughout the matrix. Initially, 6 L of deionized (DI) water and 2 liters of mixed liquor from a local nitrifying wastewater plant were added to the system to initiate leachate production and aid in developing a nitrifying microbial population. All liquid was immediately drained and 3.95 liters of the leachate were recirculated. Moisture-saturated air was added to both the bottom and middle of the compost matrix at a total rate of 2.77 L/min in the first reactor and 6.44 L/min in the second reactor. The air was saturated prior to introduction to each reactor to replenish any water lost due to evaporation and was added continuously throughout the duration of the study.

167 After two days, leachate was removed and then 2 L recirculated; this process was repeated every two to three days. Samples were routinely removed and analyzed for pH, chemical oxygen demand (COD), alkalinity, sulfate, nitrate, nitrite, and NH3-N to evaluate whether nitrification or other nitrogen removal processes were occurring. If needed, tap water was also added to bring the recirculated volume to two liters. Gas- phase samples were extracted from three sample ports located on the side of the reactor and used to measure the in-situ oxygen concentrations to evaluate the efficiency of the aeration system. Digested MSW samples were periodically removed from the reactor and used in parallel microcosm studies. Additional digested MSW was added to the reactor as needed to maintain the same waste mass. Periodically, samples of the digested MSW from the aerobic reactor were collected and solid-phase organic nitrogen, moisture content and % volatile solids of the waste were measured.

Temperature in the reactors was controlled to maintain temperatures of 22, 35 and

45oC. The first reactor was first operated at room temperature, which was 22oC; thus no temperature control was necessary to maintain it. After all the microcosm studies at 22oC were conducted, the first reactor temperature was increased by placing it in a temperature controlled room operated to maintain a constant temperature of 35oC. While operating at an elevated temperature, 4 L of leachate and/or tap water were recirculated (instead of the

2 L at room temperature) due to high evaporative loss of moisture.

The second reactor was heated ultimately to 45oC. Because this reactor was exposed to temperatures close to the reported death point of nitrifiers/denitrifiers, the reactor was placed in both a temperature controlled room and a water bath. The room temperature was maintained at 40oC. The reactor was spiked with ammonium bicarbonate

168 at 40oC. Once it was apparent the waste was acclimated, the water bath temperature was slowly ramped up to 45oC using a heating cable (Fisher Scientific, Inc) and temperature controller. Once 45oC was attained, the temperature within the waste remained constant and the system was spiked with ammonium bicarbonate to ensure the waste contained an active nitrifying and/or denitrifying microbial population. While operating at an elevated temperature, 4 L of leachate and/or tap water were recirculated (in place of the 2 L recirculated at room temperature) due to high evaporative loss of moisture.

Once all the microcosm studies at 45oC were conducted, the waste temperature was slowly increased to determine the temperature in which nitrification ceased. The temperature was increased by increasing the water bath temperature. Once nitrification ceased, the temperature was decreased to 45oC. The reactor was then spiked with ammonium bicarbonate to see whether nitrification was recoverable. This small study was conducted to determine whether high temperatures were fatal to the in-situ nitrification process.

Microcosm Studies

Microcosm studies were conducted at various oxygen, temperature and ammonia concentrations. The waste source for each study is acclimated waste from the acclimation reactors. Great care was taken in determining appropriate microcosm containers and an adequate operation plan in which all objectives could be met. The most important parameter used to determine the optimal container type was oxygen/nitrogen gas diffusion/intrusion. The following discussion describes the different microcosm assemblies tested.

169 Failed Microcosm Designs

Initially, 1-L plastic high-density polyethylene containers were tested. In these systems, air was continuously added through the bottom of the plastic container. The air passed through the compost and exited the top of the reactor, passing through a solution of indicating boric acid to capture any ammonia that may have been stripped. Several problems existed with this reactor configuration. First, adding air equally to a set of reactors was problematic with the available equipment. Also, when adding air, and thus large amounts of atmospheric nitrogen, nitrogen mass balances were not able to be conducted. A few tests were conducted in which air was not continuously added, rather added intermittently via a syringe. During these tests, it became apparent the plastic containers were not an adequate container type as both oxygen and nitrogen diffusion and intrusion occurred. Diffusion occurred through the plastic walls. Because the walls of the plastic containers were thin, problems were encountered when maintaining a gas-tight seal around necessary fittings (i.e. gas sampling/extraction ports). In an attempt to aid in minimizing oxygen/nitrogen intrusion, the containers were placed underwater. However, water leaked into each container and was thus unsuccessful.

Next, 3-L tedlar gas sampling bags (SKC, Inc.) were used. Because these bags were not designed for solids introduction, the bags were modified to permit compost addition. The bottom of each bag was cut open and lined with a layer of duct tape. To seal the bag, a device called a “clip-n-seal” (http://clipnseal.com/) was used to seal the bag. A layer of silicone sealant and then another layer of duct tape was used to seal the ends of the bag, following the clip-n-seal placement. In these bags, air was not continuously added nor used. To allow for a nitrogen mass balance to be conducted,

170 100% oxygen was used as the oxygen source. Problems with oxygen and nitrogen diffusion through the bag walls were observed. In an attempt to minimize diffusion, a series of experiments were conducted in which the bags were sealed and placed under water. Because of the high buoyancy of the bags, two bricks were placed on top of each bag to ensure they remained underwater. Unfortunately, due to the pressure exerted on each bag by both the water and bricks, gas volume was lost. Thus, these bags were deemed unacceptable.

Chosen Microcosm Design

Finally, 3- and 10-L foil gas sampling bags (SKC, Inc., Pennsylvania) were obtained. Tests confirmed that oxygen/nitrogen diffusion was non-existent or minimal.

Thus, foil gas-sampling bags were chosen as the microcosm container. Each bag was modified to permit addition of solid components to the bag. All bags were sealed using duct tape, sealant, and a clip-n-seal sealer (as described previously). A gas sampling port was located at the top of each bag. Each microcosm was loaded with 200 g of acclimated, digested MSW taken from the aerobic reactor along with 20 g of wood chips. The waste source was chosen to simulate waste from older portions of the landfill, where nitrification processes would most probably be employed at field-scale landfills. A volume of 15 mL of ammonium bicarbonate solution (concentration varied depending on target ammonia level) was added to each system resulting in moisture levels at field capacity (approximately 63% on a wet-weight basis).

After each bag was loaded with digested MSW and ammonia, the bags were purged with helium and then filled with the appropriate gas mixture. Target gas-phase

171 oxygen concentrations of 100, 17, 5 and 0.7% were used. The 3-L bags were used when

100% oxygen conditions were tested and the 10-L bags were used when testing the lower oxygen concentrations. The 10-L bags were chosen for the lower oxygen tests because the larger volume aided in maintaining a more constant gas-phase oxygen concentration.

In each 3-L bag experiment, 1200 mL of 100% oxygen was added to the gas-phase via a

1.5-L gas-tight syringe. In the 10-L bags, a volume of 5200 mL of a mixture of oxygen and helium was added. Again, air was not used because the high nitrogen concentration would limit the potential to complete mass balances on nitrogen in the systems.

Microcosm studies were also conducted at three different temperatures: 22, 35 and 45oC. The tests conducted at 22oC were left at room temperature, while the tests conducted at the higher temperatures were placed into incubators to ensure all tests were conducted at constant temperatures.

Microcosm Sampling Procedure

Obtaining an adequate volume of a liquid sample from the microcosms for repeated analysis over time was not possible. Also, the potential for sorption of constituents onto the waste matrix was high, therefore analysis of leachate alone was not adequate. Thus, each study consisted of multiple microcosms and at each sample time, one microcosm was disassembled. After disassembling the microcosms, 210 mL of DI water were added to each system and the microcosms were subsequently drained. The drained leachate was analyzed for pH, nitrite, nitrate, and sulfate. After removal of the initial small sample, a mild extraction procedure was performed to remove any ammonia and/or nitrate sorbed onto the waste matrix. Three hundred mL of a 0.5-M sodium sulfate

172 solution were added to each system. The bags were then shaken at approximately 245 rpm for 1.5 hours. They were then drained and ammonia and nitrate concentrations were measured using ion-specific electrodes.

Experiments Conducted

Microcosm experiments were conducted to evaluate the impact of waste acclimation on ammonia removal and to evaluate different gas-phase oxygen and temperatures on ammonia removal.

Impact of Waste Acclimation Experiments

Table A-2 presents all the studies conducted to evaluate the impact of waste acclimation on ammonia removal. Three Ammonia-nitrogen removal rates were measured for two series of experiments; the first series was conducted on acclimated waste at constant temperature (22oC), with an initial oxygen concentration of 100% in the headspace, and with varying ammonia concentrations (200, 500 and 1000 mg N/L). One duplicate study at the 500 mg N/L concentration was conducted to demonstrate the reproducibility of the data. The second series of experiments was conducted to evaluate the effect of waste acclimation on ammonia removal rates. The experiments were conducted using unacclimated waste at 22oC, with an initial oxygen concentration of

100% oxygen in the headspace and with an ammonia concentration of 500 mg N/L for comparison.

173 An abiotic control was conducted under the same environmental conditions and at an ammonia concentration of 500 mg N/L. The controls contained waste that had been autoclaved at 120oC for 2 hours and an ammonia solution. Because biological activity was halted, this set of controls functioned to evaluate whether changes in ammonia concentration could be attributed to physical or chemical processes. A biotic control with no ammonia added was also conducted under the same environmental conditions with no ammonia added to evaluate whether ammonia was produced via ammonification during the studies.

Table A- 2. List of Experiments Conducted to Evaluate the Impact of Waste Acclimation on Ammonia Removal. Experiment Temperature Oxygen Ammonia Conc. Description # Conc. (%) (mg N/L) 1 22 100 200 Acclimated 2 22 100 500 Acclimated 3 22 100 500 Acclimated 4 22 100 1000 Acclimated 5 22 100 500 Unacclimated 6 22 100 500 Unacclimated 7 22 100 0 Biotic 8 22 100 500 Abiotic Control 9 22 100 500 Abiotic Control 10 22 100 0 500 mg/L NO3 Control

Impact of Gas-Phase Oxygen and Temperature Experiments

Table A-3 presents all the studies conducted to evaluate the impact of gas-phase oxygen and temperature on ammonia removal. Three series of experiments were conducted; the first series was conducted on acclimated waste at 22oC, with varying ammonia loadings (500 and 1000 mg N/L) and varying average gas-phase oxygen

174 concentrations (0.7, 5, 17 and 100%). Over time, the oxygen concentrations declined slightly due to the demand imposed by nitrifying and heterotrophic microorganisms. To maintain a constant average concentration in the gas-phase, the gas composition of the bags was measured every 12-hours and, if needed, a small volume of 100% oxygen was added to increase the oxygen concentration to the target level. Ammonia removal rates were quantified for each experiment. The second and third series of experiments were conducted at 35 and 45oC, respectively, and at the same ammonia and gas-phase oxygen concentrations. An exception is the 0.7% oxygen level; this level was only tested at 22oC.

In each of the series of experiments, one duplicate was conducted at each temperature and oxygen level evaluated.

Abiotic controls were conducted at each temperature with 100% oxygen in the gas-phase and at an ammonia concentration of 500 mg N/L. The control contained waste that had been autoclaved at 120oC for 2 hours to halt biological activity and functioned to evaluate whether changes in ammonia concentration could be attributed to physical or chemical processes. Biotic controls with no ammonia added were also conducted at each temperature level to evaluate whether ammonia was produced via ammonification during the studies.

175 Table A- 3. List of Experiments Conducted to Evaluate the Impact of Gas-Phase Oxygen and Temperatures on In-Situ Nitrification. Experiment # Temperature Oxygen Ammonia Conc. Control Type Conc. (%) (mg N/L) 1 22 100 200 2 22 100 500 3 22 100 500 4 22 100 1000 5 22 5 500 6 22 5 500 7 22 5 1000 8 22 20 500 9 22 20 1000 10 22 20 1000 11 22 1 500 12 22 100 0 Biotic 13 22 100 500 Abiotic 14 22 100 500 Abiotic 15 22 100 0 500 mg/L NO3 16 22 100 0 500 mg/L NO3 17 35 100 500 Abiotic 18 35 100 0 Botic 19 35 100 0 500 mg/L NO3 20 35 100 500 21 35 100 500 22 35 100 500 23 35 100 1000 24 35 100 1000 25 35 5 500 26 35 5 500 27 35 5 1000 28 35 5 1000 29 35 20 500 30 35 20 500 31 35 20 1000 32 45 100 500 33 45 100 500 34 45 100 500 35 45 100 500 36 45 100 1000 37 45 5 500 38 45 5 500 39 45 5 1000 40 45 20 500 41 45 20 500 42 45 20 1000 43 45 100 500 Abiotic 44 45 100 0 Biotic

176 Analytical Techniques

pH

pH was measured using a Research AR-25 pH/ISE/mV meter (Fisher Scientific,

Inc) and combination junction pH gel filled electrode (Fisher Scientific, Inc) with an automatic temperature control probe. This particular electrode was chosen for its accuracy, quick response time and its specific ability to measure pH in samples that contain a large number of solids. The junction in this probe is larger than most probes so as to not clog as easily as other pH electrodes. Because this electrode is sensitive to small changes in temperature, the automatic temperature control probe was placed in all samples prior to measurement. The pH electrode was calibrated twice a week using ready-made pH buffer solutions of 4, 7 and 10 (Fisher Scientific, Inc.).

Ammonia

Ammonia was measured using a Thermo Electron Ammonia Gas-Sensing electrode (Thermo Orion, Inc.) and a Research AR-25 pH/ISE/mV meter (Fisher

Scientific, Inc). This electrode measures the concentration of free ammonia present in the sample. To measure ammonia, 100 mL of sample was placed into a 150-mL glass beaker

(tall beakers were used to minimize the exposed surface area) with a stir bar. The electrode was lowered half-way into the sample at a 45o angle while slow mixing of the sample commenced. Then, 2 mL of ionic strength adjuster (40% sodium hydroxide solution, Fisher Scientific, Inc.) was added to increase the pH of the sample to above 12

(to ensure the majority of the ammonia species present is in the form of free ammonia

177 gas). The beaker and electrode assembly was then covered tightly with parafilm. A reading was taken after the mV reading on the meter stabilized. The probe was rinsed between samples.

The ammonia probe was calibrated prior to every use. Standards were made using a 1,000 mg N/L solution (ammonium chloride). Standard concentrations were determined for each specific sampling event based on anticipated sample concentrations. During every calibration event, at least three standards were used, most often four. Quality control tests were conducted and showed that the probe calibration was only valid for 2-3 hours.

Every sample was measured using two methods: direct read and known addition.

The direct read method is straightforward. The probe is calibrated, and then using the calibration curve, the sample concentration is determined. Because there is high potential for interferences when using ion-specific electrodes (ISE), a known addition method was also used. In the known addition method, a known volume of a known ammonia concentration is added to the sample. Generally, 2 mL of 1,000 mg N/L were added to samples with concentrations less than 200 mg N/L. A volume of 4 mL was added to samples with higher concentration. All reported concentrations are those measured using the known addition method. The difference in the mV readings is then used to determine the sample concentration using equation A.1:

CV C = std std S ∆mV (A.1) Slope ()VVSS+ td10 −VS

178 where, CS is the concentration of the sample, Cstd is the concentration of the standard added, Vstd is the volume of standard added, VS is the sample volume, DmV is the change in mV readings, and slope is the slope of the line.

Anions (nitrate, nitrite, sulfate)

Nitrite, nitrate and sulfate concentrations were measured using a DX-120 ion chromatograph (IC, Dionex, Inc) equipped with an AS-14 column and a 0.002-M sodium bicarbonate/0.008-M sodium carbonate eluent. Prior to anion analysis, all samples were centrifuged and filtered using a 0.45-micron nitrocellulose filter. A flowrate of 1.05 mL/min was used and was found to achieve the most optimal separation of all peaks.

Each sample was run in duplicate and with one spike.

Nitrate was also measured in samples taken from microcosms after the extraction procedure was conducted. These samples were measured using a nitrate-specific electrode. The IC was not used because of the high sulfate concentrations (a result of the

0.5-M sodium sulfate solution). One hundred mL of sample was placed in a 100-mL glass beaker. Two mL of ionic strength adjuster (Fisher Scientific, Inc.) and 2 mL of nitrate suppressor solution were added to each beaker to minimize the number of interferences present in the samples. As with ammonia, both direct read and known addition methods were used. All samples were spiked with 4 mL of a 1,000 mg NO3-N/L standard.

179 Alkalinity

Alkalinity was measured in leachate samples from the waste acclimation process only. Alkalinity was not measured in the microcosm studies because there was insufficient leachate volume. A sample of 50 mL was added to a 150-mL beaker and was then titrated with a 0.2 N sulfuric acid solution until an end-point of 4.5 was attained.

Gas-Phase Oxygen and Nitrogen

Gas-phase oxygen and nitrogen were measured using a Shimadzu 14A gas chromatograph (GC) equipped with a thermal conductivity detector and two packed 13X molecular sieve columns (Alltech Associates, Inc.) in series. Helium was used as the carrier gas and was set at a flowrate of 30 mL/min. The two columns were placed in series to achieve adequate separation of the oxygen and nitrogen peaks. Separation of oxygen and nitrogen peaks was particularly difficult in these samples because the oxygen concentration was high (100%), while the nitrogen concentration was significantly lower

(at times <1%). Inadequate peak separation was achieved with one column. Other procedures (such as cooling the column with liquid nitrogen) were attempted to achieve adequate separation; however, the best results were obtained when placing two columns in series. The injector, detector, and oven temperatures were all 25oC and the sample volume was 0.5 mL. All samples were injected using a gas-tight syringe. Gas standards used to calibrate the GC were mixed using appropriate volumes of 100% oxygen, helium and 15% (balance was helium) nitrogen. All standards were mixed just prior to injection.

The GC was calibrated prior to every use.

180 Gas-Phase Nitrous Oxide

Nitrous oxide was measured with a GC (Varian, Inc.) equipped with an electron capture detector (ECD) in conjunction with an AT-Q 30-m capillary column (Alltech

Associates, Inc); the injector, detector, and oven temperatures were set at 110oC, 110oC, and 60oC, respectively. Both nitrogen and a methane/argon gas mixture were supplied to the GC. A gas-tight syringe was used to inject samples (0.4 mL). The GC was calibrated prior to every use using standards consisting of 100% helium and 1% nitrous oxide

(balance was helium).

Chemical Oxygen Demand (COD)

COD was measured using a slightly modified version of the standard method

(Standard Methods, 1999). Two mL of a potassium dichromate digestion solution, 4 mL of a sulfuric acid reagent, and 2 mL of sample were added to each vial. The vials were then capped, mixed and heated at 150oC for a period of 2 hours. A blank in which the sample volume was replaced with 2 mL of distilled water was also heated. The vials were subsequently cooled. To measure the COD concentration, the cooled vials were titrated with a 0.013-M ferrous ammonium sulfate (FAS) solution (4.9-g FAS and 20-mL sulfuric acid). The FAS molarity was measured before every titration by standardizing it against 2 mL of digestion reagent.

181 Solid-Phase Organic Nitrogen

Solid-phase organic nitrogen was measured using a modified macro-kjeldahl method (Standard Methods, 1995). Modifications to the Standard Method include the use of a larger volume of digestion reagent (100 mL) and an acid hydrolysis step (6 hrs) during which ground solid samples were shaken in 200 mL of solution prior to digestion

(100 mL of digestion solution and 100 mL DI). Wheat germ served as the solid-phase organic nitrogen standard. Initial tests were conducted verifying this test procedure.

Ninety -101% of the nitrogen theoretically present in the wheat germ was recovered.

One gram of ground waste was first placed in a 200-mL glass beaker. Then 100 mL of the digestion solution (potassium hydroxide and cupric sulfate solution) and 100 mL of distilled water were added to each beaker. The mixture was shaken at approximately 60 rpm for 6 hours, after which the samples were flushed into digestion flasks. Boiling chips were added and the mixture was digested at medium heat until white smoke was observed. The samples were digested for an additional 30 min. Next, 300 mL of distilled water and 50 mL of the sodium hydroxide and sodium thiosulfate solution

(see Standard Methods) were added to each flask. The samples were heated and the condensate was collected in 50 mL of an indicating boric acid solution. A 0.2 N sulfuric acid solution was used to titrate and measure the mass of nitrogen in each sample. Each sample was measured in duplicate and spiked with wheat germ (used as the solid-phase organic nitrogen standard) to ensure proper measurement occurred.

182 Volatile Solids

To measure the volatile solids content, waste was taken from the waste acclimation process and was dried in the oven at 105oC. At least 20 g of waste were then ground using a Wiley Cutting Mill (Fisher Scientific, Inc.). Once ground, the waste was mixed and 1 g was placed in a ceramic dish. The waste was heated at 550oC for 2 hours.

Moisture Content

Moisture content samples were taken on waste from the waste acclimation process at the beginning of every microcosm study. Waste samples were dried at 105oC for at least 24 hours. Weights before and after drying were used to determine the mass of moisture present in the waste.

Gas Volume

In the 100% oxygen studies, the volume in the headspace was measured using a water displacement technique. Each bag was submerged in water and the volume of water displaced measured. The gas volume was equal to the displaced water volume corrected by the volume of the waste and liquid. The pressure applied to the bags was minimal and thus neglected. In the lower oxygen studies, the headspace volume was measured by extracting the gas with a 1.5-L gas-tight syringe (Hamilton, Inc.).

183 Ammonia Gas

Ammonia was measured in the gas-phase of every microcosm. The gas was extracted from the bags using a 1.5-L gas-tight syringe and passed through 50 mL of an indicating boric acid solution (20-g boric acid, 10-mL mixed indicator in 1-L) and 50 mL of distilled water. The gas-tight syringe was connected to a diffuser stone assembly to allow for fine gas bubbles to pass through the boric acid, optimizing the transfer efficiency of ammonia gas to the liquid-phase. If a green color was observed, ammonia gas was present in the samples. The boric acid was titrated with a 0.02-N sulfuric acid solution to determine the mass of nitrogen present in the gas.

References

Florida Department of Environmental Protection (2000) Solid Waste Management in Florida, 1998,http://www.dep.state.fl.us/waste/categories/recycling/pages/00.htm.

Standard Methods for the Examination of Water and Wastewater (1995) 19th Ed., American Public Health Association/American Water Works Association/Water Environment Federation. Washington, DC, USA.

184 APPENDIX B

WASTE ACCLIMATION PROCESS RESULTS

185 Introduction

Two waste treatment reactors were operated to create an acclimated waste source for the parallel microcosm studies. Separate acclimation reactors were operated to allow for simultaneous operation of microcosm studies at different temperatures. In addition to developing an acclimated source, the reactors were operated to demonstrate the efficacy of in-situ nitrification processes. System design did not allow for process operation to be as tightly controlled as the parallel microcosm studies. In each acclimation reactor, it was not possible to capture all emissions, preventing mass balances to be conducted. As a result, measurements and observations from these processes serve to provide qualitative operational guidance to implementing such processes in solid waste environments.

Summary of Results

Waste Acclimation Process #1

Waste acclimation process #1 was operated for 984 days. During the first 753 days, the reactor was operated at 22oC. From day 754 to 984 the reactor temperature remained at 35oC.

Waste Acclimation Operation at 22oC

The majority of the results from this portion of the waste acclimation process were described elsewhere (Chapter 3, p.75-78). In addition to what was presented, spikes of ammonium bicarbonate were added to the reactor on days 518, 538, 585, 609

186 and 715 (approximately 1000 mg N/L) to acclimate the waste to high ammonia concentrations. As shown in Figure B-1, all spikes were followed with an increase in nitrate and, at times, nitrite, indicating nitrification was proceeding. The sulfate concentrations continued to increase after each ammonium bicarbonate spike as a result of autotrophic denitrification (Figure B-2). The pH and alkalinity trends are presented in

Figure B-3. Each trend declines with ammonia concentration, also indicating nitrification was occurring. The COD concentrations are presented in Figure B-4.

1400 700

1200 Ammonia-Nitrogen 600 Nitrate-Nitrogen 1000 500

800 400

600 300

400 200

200 100

0 0 0 75 150 225 300 375 450 525 600 675 750 Days fro m Start-up

Figure B- 1. Ammonia- and nitrate-nitrogen concentration trends over time in the waste acclimation reactor #1: 22oC.

187

1500 1400 1300 1200 ) L / 1100 g

m 1000 (

n 900 io

at 800 r t 700 cen

n 600

o 500 e C 400 at lf

u 300 S 200 100 0 0 75 150 225 300 375 450 525 600 675 750

Days from Start-up

Figure B- 2. Sulfate concentration over time in the waste acclimation reactor #1: 22oC.

pH 9 1800 NH3-N 8.5 Alkalinity 1600 d 8 n

1400 a ) ) L 7.5 / 3 1200 L N g/ 7 CO

pH 1000 Ca n (m

6.5 g e g 800 m o 6 ( y tr t i i

600 N

n - i

5.5 l a a k 400 ni 5 o Al

m

4.5 200 m

A 4 0 0 75 150 225 300 375 450 525 600 675 750 Days from Start-up

Figure B- 3. pH, alkalinity, and ammonia-nitrogen trends over time in waste acclimation reactor #1 at 22oC.

188 25000 22500 20000 17500 )

L 15000 / g

m 12500

D ( 10000

CO 7500 5000 2500 0 0 100 200 300 400 500 600 700 Days from Start-up

Figure B- 4. COD trend over time in waste acclimation reactor #1 at 22oC.

Waste Acclimation Operation at 35oC

While at 35oC, the reactor was periodically spiked (days 755, 824, 836, 852, 919, and 938) with ammonium bicarbonate (see Figure B-5). After each spike, nitrate and, at times nitrite, were produced. As observed at 22oC, the masses produced were never as high as were stoichiometrically expected had nitrification been the only process occurring, suggesting either simultaneous or sequential processes were occurring.

However, after each spike, higher masses of nitrate were accumulated than that observed at 22oC, suggesting that less denitrification was occurring at 35oC. Because each increase in nitrate concentration was followed by a decline in concentration, it is apparent that denitrification was still occurring, just at a slower rate than observed at 22oC.

Throughout this period, little nitrite was produced. In wastewater processes, researchers evaluating the impact of temperature on nitrification have observed temperature related trends with respect to nitrite and nitrate production (Bougard et al., 2006); nitrite

189 concentrations generally increase with temperature because the growth rate of ammonia- oxidizers is higher than nitrite-oxidizers at temperatures above 15oC (Bougard et al.,

2006). However, this phenomenon was not observed.

The sulfate concentration trend differed from that observed at 22oC. At 22oC, the sulfate concentration dramatically increased with each increase in nitrate. At 35oC, however, the sulfate concentration trend was less impacted by increases in nitrate; the trend still mirrored that of the nitrate concentration trend (Figure B-6).

The pH trend is much more pronounced than the trends at 22oC (see Figures B-3 and B-7). At 35oC, the pH decreased more substantially after each ammonium bicarbonate spike. The lowest pH measured in the leachate was approximately 6.2. This low pH may have contributed to the decline in denitrification activity observed. Glass and

Silverstein (1998) found that denitrification of wastewater with a nitrate concentration of

1350 mg N/L was inhibited pH levels below 7.0. The COD remained stable at 35oC

(Figure B-8) and was slightly lower than that observed at 22oC. The average solid-phase organic nitrogen content was 14 mg N/g dry waste, similar to that at 22oC. The organic nitrogen remained stable throughout the experiment.

190 1400 Ammonia-Nitrogen 1000 Nitrate-Nitrogen 1200 875 750 1000

625 800 500 600 375

400 250

200 125

0 0 750 800 850 900 950 1000 Days from Start-up

Figure B- 5. Ammonia- and nitrate-nitrogen concentrations over time in waste acclimation reactor #1: 35oC.

1500

1250

) SO4

L / g NO3 m 1000 ( n

o i t a r

t 750 en

c n o 500 e C at

f l u

S 250

0 750 800 850 900 950 1000 Days from Start-up

Figure B- 6. Nitrate and sulfate concentrations over time in waste acclimation reactor #2: 35oC.

191 pH 9 1800 NH3-N 8.5 Alkalinity 1600 d

8 n

1400 a ) ) L 7.5 / 3

1200 L N g/ 7 CO pH 1000 (m Ca n 6.5 g

800 m oge 6 ( y tr t i i

600 N n - 5.5 i l

a a 5 400 k

Al moni

4.5 200 m A 4 0 750 800 850 900 950 1000 Days from Start-up

Figure B- 7. pH, alkalinity and ammonia-nitrogen trends over time in waste acclimation reactor #2: 35oC.

5000 4500 4000 3500 )

L 3000 / g

m 2500 (

D 2000

CO 1500 1000 500 0 750 825 900 975 Days from Start-up

Figure B- 8. COD trend over time in waste acclimation reactor #2: 35oC.

Waste Acclimation Process #2

Waste acclimation process #2 was operated for 330 days. Initially, the reactor was operated at 40oC. Once acclimated, the reactor temperature was increased and maintained at 45oC.

192 Waste Acclimation Operation at 45oC

As was done in the waste acclimation reactor #1, the second reactor was periodically spiked with ammonium bicarbonate on days 8, 50, 82, 98, 167, 186, 205,

211, 226, 251, 271 and 307 (see Figure B-9). The reactor remained at 40oC for the first

100 days. After that, the temperature was increased to 45oC. While at the higher temperature, differences in acclimation time were experienced. During the first couple of days of some spikes, acclimation of the waste to the ammonia concentrations appeared to take longer. This phenomenon was observed by comparing the ammonia removal slopes after taking into account the mass of ammonia removed via biological activity and that removed via dilution. The average solid-phase organic nitrogen was 15 mg N/g dry waste and remained stable throughout the experiment.

Although there was no nitrite detected in the leachate at this temperature, increases in nitrate concentration were observed, providing evidence that ammonia removal was in part due to nitrification. High nitrate concentrations (at times > 3,000 mg

N/L) were measured. The high nitrate concentrations did not appear to have an inhibitory impact on nitrification. The decrease in nitrate concentration between day 240 and 250

(Figure B-9) was due to dilution by the addition of water and buffering agents to the reactor.

The pH and alkalinity trends also indicate that nitrification was occurring (Figure

B-11), as both trends declined with ammonia concentration. At this temperature, ammonia concentrations, and pH levels, it is possible that some ammonia was lost as gaseous ammonia. The theoretical maximum amount of free ammonia gas that may have

193 been present is 15%. The COD trends in this reactor (Figure B-12) remained stable, similar to the trends observed at other temperatures (Figures B-4, B-8 and B-12).

In this reactor, the nitrate concentrations accumulated over time and at times were close to what was stoichiometrically expected, suggesting that the amount of denitrification that had occurred was lower than that observed at the lower temperatures.

It is possible that denitrification was inhibited by the high nitrate concentrations measured (always > 1,000 mg N/L and at times > 3,000 mg N/L) or potentially by the increase in temperature. However, there is no evidence that this is true; Glass and

Silverstein (1998) conducted a study evaluating denitrification capacity in wastewater with nitrate concentrations as high as 2,700 mg N/L and observed no inhibition. The high temperature did not completely inhibit nitrification, thus it is unlikely to completely inhibit denitrification. Denitrification may have been inhibited by the lower pH levels observed at this temperature. However, there seems to be no trend associated with the nitrate concentrations. At times in which the pH is above 7.5, there is no evidence of denitrification; the accumulation of nitrate is fairly constant with time. Glass and

Silverstein (1998) also observed that nitrite accumulation increased with pH. The lack of nitrite production in the leachate from this reactor may be tied to the lower pH values measured.

The sulfate concentration trend over time is presented in Figure B-11. The sulfate trend does mirror the nitrate concentration (as was observed during the operation at 22 and 35oC), however, each increase in sulfate concentration following an increase in nitrate was not as great as that observed at 22oC (see Figure B-2), most probably resulting from the lower amount of denitrification that had occurred (more nitrate accumulated).

194 During the last 130 days, the pH in the compost from this reactor was lower than usual (Figure B-10). Despite attempts to increase the buffering capacity, the pH would not remain stable at an elevated value. The exact cause of this is unknown. Other researchers found that while conditions transition between mesophilic and thermophilic temperature ranges, more acid is produced because of degradation of complex organic compounds at the higher temperatures (Tuomela et al., 2000). However, this generally occurs during the earlier stages of a compost process in which larger masses of degradable carbon are present (Sundberg et al., 2004), not during late stages in the compost process which is the case in this instance. At thermophilic temperatures, lignin is degraded and could potentially result in acid accumulation and thus lower pH levels. The optimal temperature for lignin degradation in compost has been found to be between 40 and 50oC (Tuomela et al., 2000). The microorganisms responsible for the degradation are fungi and actinomycetes. It is important to note that only at this temperature fungus was observed on the wood chips, potentially indicating degradation of lignocellulose materials. If this did occur, it is interesting to note that any acids produced did not appear to be used as a carbon source for denitrification, suggesting that denitrification was in fact inhibited.

195 2500 5000 NH4 2000 NO3 4000

1500 3000

1000 2000

500 1000

0 0 0 50 100 150 200 250 300 350 Days from Start-up Figure B- 9. Ammonia- and nitrate-nitrogen concentrations over time in waste

9 pH 2000

8.5 NH3-N 1800

Alkalinity d 8 acclimation reactor #2. 1600 n a ) ) L

7.5 1400 / 3

L N g/

7 1200 CO

pH (m Ca

6.5 1000 n g

e m ( og 6 800 y tr i t i N n - 5.5 600 i l a a ni 5 400 k Al

4.5 200 mmo A 4 0

0 50 100 150 200 250 300 350 Days from Start-up

Figure B- 10. pH, alkalinity, and ammonia-nitrogen trends over time in waste acclimation reactor #2: 45oC.

196

1400 SO4 4998 NO3 1200 4284 L) g/ L)

/

m 1000 3570 N ( n e

g n g o m i o t 800 2856 e a r r t t i on (

i n t N e 600 2142 a r t ate- n onc tr e

i C

400 1428 N e t

onc a f C l

u 200 714 S 0 0 0 50 100 150 200 250 300 350 Days from S tart-up

Figure B- 11. Sulfate and nitrate-nitrogen trends over time in waste acclimation reactor #2: 45oC.

2500

2000 ) L

/ 1500 g m

D ( 1000 CO

500

0 0 50 100 150 200 250 300 350 Days from Start-up

Figure B- 12. COD trend over time in waste acclimation reactor #2: 45oC.

197 References

Bougard, D., Bernet, N., Cheneby, D., and Delgenes, J.P. (2006) Nitrification of a High- Strength Wastewater in an Inverse Turbulent Bed Reactor: Effect of Temperature on Nitrite Accumulation, Process Biochemistry 41, 106-113.

Glass, C. and Silverstein, J. (1998). Denitrification Kinetics of High Nitrate Concentration Water: pH effect on inhibition and nitrite accumulation, Water Research 32(3), 831-839.

Tuomela, M., Vikman, M., Hatakka, A. and Itavaara, M. (2000). Biodegradation of Lignin in a Compost Environment: A Review, Bioresource Technology 72(2), 169-183.

Sundberg, C, Smars, S. and Jonsson, H. (2004). Low pH as an Inhibiting Factor in the Transition from Mesophilic to Thermophilic Phase in Composting, Bioresource Technology 95(2), 145-150.

198 APPENDIX C

MODEL VALIDATION STATISTICS

199 Introduction

A multiplicative Monod model was developed to describe the impact of ammonia concentration, gas-phase oxygen concentrations, pH and temperature in in-situ nitrification in solid waste environments. To determine the statistical validity of the model parameters (to determine whether they are a required element of the model), a backward elimination procedure involving fitting all rate data to several variations of the multiplicative Monod model was conducted (Petruccelli et al., 1999). The procedure used to determine the validity of each term is described below.

Procedure

1. Add all potential terms to the model, including inhibition of both ammonia and

oxygen, and terms evaluating the impact of temperature and pH (Chapter 4, Table 4-

1):

⎛⎞⎛⎞ n ⎜⎟⎜⎟⎛⎞T kC %1O ⎛⎞T ⎜⎟1− ⎜⎟N ⎜⎟2 ⎝⎠t ⎛⎞ Re= 22⎜⎟⎜⎟zp− H ⎜⎟CO⎜⎟% ⎜⎟t 11+ 0() KC++N %OK++2 ⎝⎠⎝⎠ ⎜⎟sN 2 O 2 ⎜⎟K ⎜⎟K ⎝⎠IN, ⎝⎠I

Some Definitions:

⎛⎞kC ⎜⎟N = Classic Monod K + C ⎝⎠sN

C 2 N = Ammonia inhibition term KIN,

200

⎛⎞ ⎜⎟%O ⎜⎟2 2 = Oxygen term with an Andrew’s inhibition term ⎜⎟%O %OK++2 ⎜⎟ 2 O 2 ⎝⎠KI

n ⎛⎞T ⎛⎞T ⎜⎟1− ⎜⎟e⎝⎠t ⎜⎟ = Temperature term ⎝⎠t

⎛⎞1 = pH term ⎜⎟()zp− H ⎝⎠11+ 0

2. Rate data from all the experiments conducted were pooled. Sigma Plot (SSPS, Inc.)

was used to fit all data to the model shown in step 1. Both parameter estimates and

analysis of variance (ANOVA) test results were evaluated to determine the validity of

each parameter. Table C-1 presents the results.

Table C- 1. Parameter Values from the Complete Multiplicative Monod Model. Parameter Parameter p-value Value k 1.167 < 0.0001 Ks 167 < 0.0001 KI,N 3324719 0.5175 KO2 2.84 0.0091 KI 206 0.0317 t 33.81 < 0.0001 n 7.66 < 0.0001 z 6.13 < 0.0001

The r2 value of the fit was 0.75 and the F-value was 45.18. Based on the

statistics performed, it is evident that the ammonia inhibition term (KI,N) is not

statistically required (p-value > 0.05 and a very high number), suggesting that

201 ammonia inhibition had not occurred and will not occur at ammonia concentrations

present in landfills.

3. Next, several iterations using Sigma Plot were performed. During the first set of

iterations, one parameter was removed from the model, one at a time. For example,

all rate data were fit to the multiplicative model minus the k term. During the next

iteration, the k term was replaced and the ammonia inhibition term removed. This

procedure was continued until all the parameters were removed once. The F-values of

each iteration were compared to determine the relative validity of each parameter.

4. Next, iterations were run in which two parameters were removed from the model. For

example, the ammonia inhibition term was removed and subsequently, one at a time,

the other parameters were removed and then replaced. The F-values of each iteration

were compared to determine the relative validity of each set of parameters.

5. The next set of iterations involved the removal of three parameters at a time. The

ammonia inhibition term and the half-oxygen saturation constant were always

removed. Subsequently, one other parameter was removed and then replaced. The F-

values of each iteration were compared to determine the relative validity of each set

of parameters.

202 6. The procedure of removing parameters was continued until all parameters were

removed (except the classic Monod model). Table C-2 presents the results from this

entire procedure.

Table C- 2. Sigma Plot Iteration Results. 2 Case # k KI,N Ko2 KI n Temp. pH F r Factor Factor 1 X X X X X X X 40.95 0.73 2 X X X X X X 52.3 0.75 3 X X X X X X 53.2 0.75 4 X X X X X X 38.2 0.68 5 X X X X X X 48.2 0.73 6 X X X X X X 34.35 0.66 7 X X X X X 34.1 0.61 8 X X X X X X 41.3 0.7 9 X X X X X 63.3 0.75 10 X X X X X 46.2 0.68 11 X X X X X 58.4 0.73 12 X X X X X 41.6 0.66 13 X X X X 43 0.61 14 X X X X X 49.7 0.7 15 X X X X 55.4 0.67 16 X X X X 58.32 0.68 17 X X X X 34 0.56 18 X X X 33.3 0.48 19 X X X X 45.2 0.63 20 X X X 74.5 0.67 21 X 94.5 0.46 22 X X 60.3 0.63 23 X X X 45.6 0.56 24 X X 50.4 0.48 25 X X 53.8 0.5 26 X 88.1 0.44 27 No Parameters – Only Classic Monod 85.7 0.43 Note: X indicates the value was used in the fit, while a blank cell indicates the values that were removed from each fit.

203 Discussion of Results

The highest F-values were found when fitting all data to the classic Monod model with a term to account for the impact of pH. It was not surprising that the classic Monod model resulted in a high F-value (the second and third highest F-values were the Monod model alone, cases 26 and 27), as the classic Monod model is the most significant term in the multiplicative Monod model.

The model chosen to most significantly describe the impact of ammonia concentration, oxygen, temperature and pH did not result in the highest F-value (Case

#3). Although this model did not have the highest F-value, it contains the terms that are mechanistically required to describe biological nitrification. The oxygen inhibition term is included because there was some inhibition of nitrification at high oxygen levels

(100% at 22 and 45oC). Table C-3 contains the parameter values and their respective p- values for this model. As shown, all parameters are statistically significant (p-value <

0.05).

Table C- 3. Parameters of the Chosen Multiplicative Monod Model. Parameter Parameter p-value Value k 1.17 < 0.0001 Ks 167 < 0.0001 KO2 2.84 0.0089 KI 206 0.0317 t 33.81 < 0.0001 n 7.66 < 0.0001 z 6.13 < 0.0001

Because 100% oxygen will likely never be added to landfills, a model without an oxygen inhibition term was also evaluated (case #11). All parameters were found to be

204 significant, as presented in Table C-4. The Sigma Plot output are included and are located after this discussion.

Table C- 4. Parameters of the Chosen Multiplicative Monod Model Without Oxygen Inhibition. Parameter Parameter p-value Value k 0.899 < 0.0001 Ks 159 0.0031 KO2 2.46 0.0056 t 34.7 < 0.0001 n 4.56 0.0009 z 5.40 < 0.0001

205 Data Sheets

Case #1

206 Case #2

207

Case #3

208

Case #4

209

Case #5

210

Case #6

211

Case #7

212

Case #8

213

Case #9

214

Case #10

215

Case #11

216

Case #12

217

Case #13

218

Case #14

219

Case #15

220

Case #16

221

Case #17

222

Case #18

223

Case #19

224

Case #20

225

Case #21

226

Case #22

227

Case #23

228

Case #24

229

Case #25

230

Case #26

231

Case #27

232 Sigma Plot Output With No 100% Oxygen Data

233 234 235 236

237 Sigma Plot Output With ALL Data

238 239 240 241 242 243 244

245 References

Petruccelli, J., Nandram, B. and Chen, M. (1999). Applied Statistics for Engineers and Scientists, Prentice Hall, Upper Saddle River, New Jersey.

246 APPENDIX D

CONFIDENCE INTERVAL DETERMINATION

247 Introduction

Confidence interval estimations for both the complete Monod model and the model without oxygen and pH terms were conducted to determine whether or not there was a statistical difference between the environmental conditions evaluated. The following is a brief summary of the results.

Summary of Results

Confidence intervals. were calculated using equation D.1:

2

1 xpred − x + ( ) yypred =+− tα /2s1 + (D.1) nSSxx

Note: ta/2 is based on (n-1) degrees of freedom

where, ypred is the y value predicted at the confidence interval selected (the upper and lower bounds of the interval), ta/2 is the t-distribution value for the confidence level being tested, s is the standard error of the fit, n is the number of samples in each analysis, xpred is the predicted x from the Monod model, x is the average x in the data set, and SSxx is a statistic that quantifies the spread in x.

Confidence intervals were calculated at both the 75 and 80% levels to determine significance between the different oxygen concentrations and temperatures. Three separate cases were examined. All data sheets resulting from this analysis can be found on p. 251 – XXX.

248 Case #1: Comparison Between Oxygen Levels at Constant Temperature Using the Monod Model Without Temperature and pH Terms

All data were sorted by oxygen level and temperature. For this case, significance between all oxygen levels at each temperature was evaluated. The Monod model described in Chapter 4 in which temperature and pH factors are not included was used in this case so that significance due to changes in oxygen alone was evaluated. The oxygen value was assumed to remain constant (although the level changed slightly in the experiments) for each confidence level determination.

It was found that at both 22 and 45oC, there were some statistical differences among the oxygen concentrations evaluated in the experiments. At 22oC and a confidence level of 80%, the 100% oxygen experiments were found to be statistically different from those at 18 and 4.5%, while at 45oC the 18 and 100% studies were determined to be statistically different. At both temperatures, there was no statistical difference between the 18 and 4.5% oxygen experiments. At 35oC, however, there was no significance between oxygen levels (even at a 75% confidence level). Because the rate expressions at

22 and 45oC had higher oxygen half-saturation constants (5.7 and 9.4, respectively) than at 35oC (3.35), this result is not surprising.

Case #2: Comparison Between Oxygen Levels at Constant Temperature Using the Complete Monod Model (including temperature and pH terms)

Again, all data were sorted by oxygen level and temperature. All oxygen levels at each temperature were evaluated to determine whether there was a statistical difference among oxygen levels tested when temperature is accounted for in the Monod model. The

249 complete Monod model described in Chapter 4 was used. The oxygen values were assumed constant (although the level changed slightly in the experiments) for each interval. It was found that when temperature is present in the Monod model there is no statistical difference between all oxygen concentrations at every temperature.

Case #3: Comparison Between Temperatures at Different Oxygen Levels Using the Complete Monod Model (including temperature and pH terms)

For this case, all data were sorted first by temperature and then oxygen level. At each oxygen level, the different temperatures were compared to evaluate their significance. The complete Monod model described in Chapter 4 was used. Results indicated that at all oxygen levels there was a statistical difference between ammonia removal rates at 22 and 35oC (at the 80% level) and between 35 and 45oC (75% level), while there was no significant difference between the tests at 22 and 45oC.

Discussion of Results

Results from this analysis indicate that temperature is the most significant factor in the nitrification process and that operating at different temperature levels does in fact change the ammonia removal rates measured. The significance of temperature, however, may be influenced by the impact it has on the oxygen levels in the liquid-phase. Oxygen concentrations were found to have a significant impact on nitrification at 22 and 45oC.

The fastest removal rates were observed at 35oC. At this temperature, no statistical significance was observed at the different oxygen levels, indicating that

250 microbial activity was not influenced by the different oxygen levels. At 22 and 45oC, however, the removal rates are more influenced by gas-phase oxygen levels.

251 Data Sheets

Comparison Between Oxygen Levels at Constant Temperature Using the Monod Model Without Temperature and pH Terms

252 253 254 255

Comparison Between Oxygen Levels at Constant Temperature Using the Complete Monod Model (including temperature and pH terms)

256 257 258

259

Comparison Between Temperatures at Different Oxygen Levels Using the Complete Monod Model (including temperature and pH terms)

260 261 262 263

APPENDIX E

MODEL CODE

264 !! This program calculates and tracks the removal of ammonia in a landfill !! by modeling the landfill as several CSTRs in series. Each CSTR volume is !! based on the hydraulic conductivity of the waste. Thus, each CSTR has a depth !! equivalent to the distance the leachate would flow in a give time period. !! The Monod parameters found during laboratory microcosm studies are used !! as input to the model. Multiplicative Monod models with and without with an Andrew's inhibition term !! can be simulated. Most of the input is read from a file. Temperature, oxygen and pH !!are inputted to give a treatment time and hight of waste.

Real k,Kn,Ko,Pd,Qing,Co,r,Vmc,Clf,Dt,Tf,Depth,DeltX,Vlf,Vlq,Ki Integer m,B,Z,FF,PT,PTStop,NT DIMENSION C(1000),COT(1000),tstps(100), V(10000)

CHARACTER*80 INPUTFL,OUTFL PRINT *,' ENTER THE NAME OF THE INPUT FILE NAME.' READ(*,5) INPUTFL PRINT *,' ENTER THE NAME OF THE OUTPUT FILE NAME.' READ(*,5) OUTFL Print*,"Please answer the following questions:" Print*,' ' Print*,' ' Print*,"Would you like to simulate conditions with the oxygen inhibition term?" Print*,"Type 1 for yes and 2 for no." Read*,B If (B.EQ.1) then Print*,"Please enter the inhibition constant." Read*,Ki end if Print*,"What is the oxygen concentration in the landfill(%):" Read*,O Print*,"What is the temperature in the landfill (degrees Celcius):" Read*,tem Print*,"What is the average pH in the landfill:" Read*,ph Print*,"At what time would you like leachate recirculation to stop (min)? Please enter in an increment of 30." Read*,PT

5 FORMAT(A80) OPEN(UNIT=15,FILE=INPUTFL, STATUS='UNKNOWN') OPEN(UNIT=16,FILE=OUTFL, STATUS='UNKNOWN') READ(15,*)HY,Pd,Qing,Co,r,Vmc,Clf,Dt,Tf,depth Read(15,*)k,Kn,Ko,fc Read(15,*)ntstps Read(15,*)(tstps(I),I=1,ntstps) Write(16,*) Write(16,*)" MODEL INFORMATION" Write(16,*)"********************************************************" Write(16,*)"The following input parameters were used in this model:" WRITE(16,*)"Hydraulic conductivity (cm/sec)=",HY

265 WRITE(16,*)"Dry density of waste (g/L):",Pd WRITE(16,*)"Leachate addition flowrate per well (L/day):",Qing WRITE(16,*)"Ammonia concentration in the leachate being recirculated (mg/L-N):",Co WRITE(16,*)"Radius of influence around the well (ft):",r WRITE(16,*)"volumetric moisture content of the waste (%)",Vmc WRITE(16,*)"Initial ammonia concentration in the landfill:",Clf WRITE(16,*)"Gas-phase oxygen concentration in the landfill:",O WRITE(16,*)"Temperature in the landfill:",tem WRITE(16,*)"Average pH in the landfill:",ph WRITE(16,*)"The following Monod parameters were used in this simulation:" WRITE(16,*)"k (mg N/g dry waste*day):",k WRITE(16,*)"Kn (mg/L-N):",Kn WRITE(16,*)"Ko2 (%):",Ko Write(16,*)"Was this simulation done using inhibition?",B If (B.EQ.1) then Write(16,*)"The inhibition constant is:",Ki end if WRITE(16,*)"The time step used for this simulation is(min):",Dt WRITE(16,*)"The simulation was run for (days):",Tf Write(16,*)"Pumping of leachate will stop at this time (min):",PT WRITE(16,*)"The total depth of landfill simulated is:",Depth Write(16,*) "The time steps chosen to see results at are:",(tstps(I),I=1,ntstps)

!!Calculate and print initial parameters DELTX=HY*(60)*Dt*0.03281 Vlf=(3.14*(r**2)*DeltX)*28.3 Area=(3.14*(r**2))

Write(16,*) Write(16,*) WRITE(16,*)"The following are some initial parameters calculated:" WRITE(16,*)"The height of each time step is:",deltX WRITE(16,*)"The volume of each CSTR in this simulations is:",Vlf

!!Begin initializing concentrations

COT(1)=CO

!!Calculate the number of time and space steps

NT=(Tf*24*60)/Dt M=(Depth/DELTX)

!!Print output at time = 0 and the total number of time steps as a check Write(16,*)"The number of time steps in this simulation:",Nt Write(16,*) Write(16,*) Write(16,*) "MODEL OUTPUT:"

Write(16,*)"******************************************************************"

266 Write(16,*)"Initial Information:" Write(16,*)"Node Number, Depth, Concentration (mg/L-N)"

If (B.EQ.2) then

!!Loop to assign and print initital conditions DO 10 I=2,M COT(I)=CLF V(I)=(Qing*Dt)/(24*60)

Xcord=DeltX*(I-1)

Write(16,1000)I,XCORD,COT(I)

10 CONTINUE

!!Initialize the print counter

JJ=1

!!Loop with time to start the simulation

DO 50 IT=1,NT

TIME=((IT)*Dt)/(60*24) tma=tstps(JJ)/(60*24) tmb=(tstps(JJ)+Dt)/(60*24)

if(TIME.ge.tma .and. TIME.le.tmb) then WRITE(16,*)'Output at time step= ',IT WRITE(16,*)'Output at time= ',TIME end if

!!Initialize parameters for the space loop

PTStop=PT/Dt Vin=0 Vout=0 Vwlf=0 Vfc=(((fc/(1-fc))*Pd*(Vlf))/1000)

!!Loop in space DO 30 J=2,M

!!Define some important relationships about volume

If (IT.le.PTStop.and.J.eq.2)Then Vin=(Qing*Dt)/(24*60) Vwlf=Vin+Vfc+V(J)

267 If (Vwlf.ge.Vfc) then Vout=Vin else Vout=0 end if V(J)=Vwlf-Vout-Vfc end if

If (IT.le.PTStop.and.J.gt.2)Then Vin=Vout Vwlf=Vin+(Vfc)+V(J)

If (Vwlf.ge.Vfc) then Vout=Vin else Vout=0 end if V(J)=Vwlf-Vout-Vfc end if

If (IT.gt.PTStop.and.J.eq.2) Then Vin=0 Vwlf=Vin+Vfc+V(J) Write(*,*)'LINE 209' Vtest=Vin+V(J)+Vfc If (Vwlf.gt.Vfc) then

Vout=Vtest*.3 else Vout=0 end if V(J)=Vwlf-Vout-Vfc end if

If (IT.gt.PTStop.and.J.gt.2)then Vin=Vout Vwlf=Vin+(Vfc)+V(J) Vtest=Vin+V(J) If (Vwlf.gt.Vfc) then Vout=Vtest*.3 else Vout=0.0 end if V(J)=Vwlf-Vout-Vfc end if

!!Monod Calculations, these are without oxygen inhibition

268 COT(1)=CO R=((k*COT(J))/(Kn+(COT(J))))*(O/(Ko+O))*(((tem/34.7)*exp((1- (tem/34.7))))**4.56)*(1/(1+(10**(5.4-ph))))*Pd*Vlf DC=(Dt)*((((Vin/dt)*COT(J-1))-((Vout/dt)*(COT(J)))-(R/(24*60)))/(Vwlf)) !Write(16,*)'DC=,',DC C(J)=COT(J)+DC XCORD=(J-1)*DeltX

!!Statement to determine when to print based on specified times if(TIME.ge.tma .and. TIME.le.tmb) then WRITE(16,1000)J,XCORD,Cot(J),Vin,Vout,Vwlf,Vfc,V(J),DC,C(J)

end if

1000 FORMAT(I8,9D12.3)

!!Becasue data is not saved, this statement makes the current concentrations !!the old ones for use during the next time step

COT(J)=C(J)

30 continue

!!This print statement is there just to show that the model is running Print*,"Hi!!!!!!"

!!This increases the counter for the printing decisions if(TIME.ge.tma .and. TIME.le.tmb) then JJ=JJ+1 end if

50 CONTINUE

else !!Loop to assign and print initital conditions DO 15 Z=2,M COT(Z)=CLF V(Z)=(Qing*Dt)/(24*60)

Xcord=DeltX*(Z-1)

Write(16,1000)Z,XCORD,COT(Z)

15 CONTINUE

!!Initialize the print counter

FF=1

!!Loop with time to start the simulation

269

DO 70 IT=1,NT

TIME=((IT)*Dt)/(60*24) tma=tstps(FF)/(60*24) tmb=(tstps(FF)+Dt)/(60*24)

if(TIME.ge.tma .and. TIME.le.tmb) then WRITE(16,*)'Output at time step= ',IT WRITE(16,*)'Output at time= ',TIME end if

!!Initialize parameters for the space loop

PTStop=PT/Dt Vin=0 Vout=0 Vwlf=0 Vfc=(((fc/(1-fc))*Pd*(Vlf))/1000)

!!Loop in space DO 60 Z=2,M

!!Define some important relationships about volume

If (IT.le.PTStop.and.Z.eq.2)Then Vin=(Qing*Dt)/(24*60) Vwlf=Vin+Vfc+V(Z) If (Vwlf.ge.Vfc) then Vout=Vin else Vout=0 end if V(Z)=Vwlf-Vout-Vfc end if

If (IT.le.PTStop.and.Z.gt.2)Then Vin=Vout Vwlf=Vin+(Vfc)+V(Z) If (Vwlf.ge.Vfc) then Vout=Vin else Vout=0 end if V(Z)=Vwlf-Vout-Vfc end if

If (IT.gt.PTStop.and.Z.eq.2) Then Vin=0

270 Vwlf=Vin+Vfc+V(Z) Write(*,*)'LINE 209' Vtest=Vin+V(Z)+Vfc If (Vwlf.gt.Vfc) then Vout=Vtest*.3 else Vout=0 end if V(Z)=Vwlf-Vout-Vfc end if

If (IT.gt.PTStop.and.Z.gt.2)then Vin=Vout Vwlf=Vin+(Vfc)+V(Z) Vtest=Vin+V(Z) If (Vwlf.gt.Vfc) then Vout=Vtest*.3 else Vout=0.0 end if V(Z)=Vwlf-Vout-Vfc end if

!!Monod Calculations, these are with oxygen inhibition

Ra=((k*COT(Z))/(Kn+(COT(Z))))*(O/(Ko+O+((O**2)/Ki))) Rb=(((tem/33.8)*exp((1-(tem/33.8))))**7.66)*(1/(1+(10**(6.13-ph))))*Pd*Vlf R=Ra*Rb DC=(Dt)*((((Vin/Dt)*COT(Z-1))-((Vout/Dt)*(COT(Z)))-(R/(60*24)))/(Vwlf)) C(Z)=COT(Z)+DC XCORD=(Z-1)*DELTX

!!Statement to determine when to print based on specified times if(TIME.ge.tma .and. TIME.le.tmb) then

WRITE(16,1001)Z,XCORD,C(Z),Vin,Vout,Vwlf,Vfc,V(Z) end if

1001 FORMAT(I8,7D12.3)

!!Becasue data is not saved, this statement makes the current concentrations !!the old ones for use during the next time step

COT(Z)=C(Z)

60 continue

!!This print statement is there just to show that the model is running Print*,"Hi!!!!!!"

271

!!This increases the counter for the printing decisions if(TIME.ge.tma .and. TIME.le.tmb) then FF=FF+1 end if

70 CONTINUE end if

stop end

272 APPENDIX F

PLOTS OF ALL MICROCSM DATA

273

Microcosms at 22oC

Acclimated Waste

274 Ammonium 60 Nitrite Nitrate ) 50 N g

m 40

es (

eci p 30 S n e

g 20 o r t i N f 10

0 Mass o 0123 4567 Time (Days)

N2 g 10

m 9 N2O

(

O 8 2 7 6 nd N

N) 5 a 2 4 N 3 of

s 2 s

a 1 M 0 0246 810 T ime (Days)

0.2500

) e t y 0.2000 da Ra - l e a t

v 0.1500

s o a

m w 0.1000 y Re a dr i

g 0.0500 n /

o m g N 0.0000 m ( Am 0 50 100 150

Ammonia Concentration (mg N/L)

Figure F- 1. Microcosm results from the test at the following conditions: 22oC, 500 mg N/L, 100% Oxygen and acclimated: Set 1.

275 70.00 Am m onia Nitrite

) 60.00 Nitrate N

g m 50.00

es ( 40.00 eci p

S

n 30.00 e g o r

t 20.00 i N

f 10.00

Mass o 0.00 0246 81012 Time (Days)

N2 g 12

m N2O

( 10 O

2

N 8 d N)

an 6

2

N 4 f 2

ass o

M 0 024 681012 Time (Days)

0.2000

e t a ay)

d 0.1500 R e-

val o ast 0.1000 m w e

y r a R d 0.0500 i g n / o N

m g 0.0000 m m ( A 0 100 200 300 400 Ammonia Conc entration (mg N/L)

Figure F- 2. Microcosm results from the test at the following conditions: 22oC, 500 mg N/L, 100% Oxygen and acclimated: Set 2.

276 120 Ammonium Nitrite ) 100

N Nitrate g m 80 ( s

ie c 60 e

Sp 40 N f

o 20 s

s a 0 M 0 5 10 15 20 25 30 Ti me (Days)

g 28 N2

m ( 24

O N2O 2 20

16 nd N N) a

2 12

N 8

of s 4 s a

M 0 0 5 10 15 20 25 30 T ime (Days)

0.2

e t a ay)

d 0.15 R e-

val o ast 0.1 m w e

y r a R d 0.05 i g n / o N

m g 0 m m ( A 0 200 400 600 Ammonia Conc entration (mg N/L)

Figure F- 3. Microcosm results from the test at the following conditions: 22oC, 1000 mg N/L, 100% Oxygen and acclimated: Set 1.

277 25.00 Ammonia )

N Nitrate g 20.00 m

( Nitrite s

ie 15.00 c e

Sp 10.00 N f O

s 5.00 s a M 0.00 024 6810 Ti me (Days)

N2 g 6

m N2O (

O 5

N2 4 d

n N)

a 3

N2 2 f

o

s 1

s a

M 0 02468 Time (Days)

12.0

) e t y 10.0

a d Ra l

e- 8.0 a

v o ast 6.0

m y w

r 4.0 Re a d i g n / 2.0

o N m g 0.0 m ( Am 0 5 0 100 150

Ammonia Concentration (mg N/L)

Figure F- 4. Microcosm results from the test at the following conditions: 22oC, 200 mg N/L, 100% Oxygen and acclimated: Set 1.

278 60 NH4 ) 50 N NO3 g m

( NO2 40

s e i

ec 30

p S 20 N f 10

ass o M 0 0246 81012 Ti me (Days)

) N2 60 N2O g N

m 50

( O

2 40

nd N 30

a 2 20

N

of 10 s

s a

M 0 024 6810 T ime (Days)

0.100 0.090 N) g 0.080 m 0.070 ( O 0.060

N2 0.050 d n 0.040 a 0.030 N2 f 0.020 o

s 0.010 s a 0.000 M 0 100 200 300 400 500 600 Ammonia Concentration (mg N/L)

Figure F- 5. Microcosm results from the test at the following conditions: 22oC, 500 mg N/L, 0.7% Oxygen and acclimated: Set 1.

279 60 NH4 50 NO3 N) g NO2

m 40 ( s e 30 eci p 20

N S f 10

ss o

a 0

M 0123456 Time ( Days)

N2 28 N) N2O g 24 m ( O 20

N2 16

d n

a 12

N2

f 8

o s 4 s a

M 0 024 6810 Ti me (Days)

0

g

m 0 ( e t y) a a d R - 0

l e a

v o ast

m 0 y w r Re

d a i

g 0 / n o N m 0 Am 0 100 200 300 400 500 600 Ammonia Co ncentration (mg N/L)

Figure F- 6. Microcosm results from the test at the following conditions: 22oC, 500 mg N/L, 5% Oxygen and acclimated: Set 1.

280 180 NH4 160 NO3 ) 140

N g 120 NO2

m ( 100 80 ecies p 60 S 40 N f

o 20 s s

a 0 M 0123 456 Time (Days)

N2 18 N2O 16

) N 14 g m

( 12

O 2 10

N d 8 an 2 6 N

f 4 2 Mass o 0 024 6810 T ime (Days)

0

0

g

m 0

( e t

a ay) 0 d R

l e-

a 0 t

s a 0 w mov e y

r 0 R a d g ni / 0 N

mo 0

m

A 0 100 200 300 400 500 Ammonia Concent ration (mg N/L)

Figure F- 7. Microcosm results from the test at the following conditions: 22oC, 500 mg N/L, 5% Oxygen and acclimated: Set 2.

281 250 NH4 NO3

N) 200 g

m NO2 150

es ( i ec 100 p

N S

f 50

0 ass o M 012 3456 Time (Days)

70 N2

) 60

N N20

g 50 m (

es 40

eci

p 30

S N

f 20 o 10

Mass 0 0123456 Time ( Days)

0.60

e t ay) Ra d

0.40 - l e a v st o a m

w y

r 0.20 Re d a

i g n / o N

m g 0.00 m

( Am 0 200 400 600 800 1000 Ammonia Concen tration (mg N/L)

Figure F- 8. Microcosm results from the test at the following conditions: 22oC, 1000 mg N/L, 5% Oxygen and acclimated: Set 1.

282 140 )

N 120 g

m 100

es ( NH4 80 eci NO3 p 60 S NO2 N

f 40

o s 20 s

a

M 0 0246

Time (Days)

20 N2 18

) 16 N20 N

g 14 m 12 10 ecies ( p 8 S N

f 6 o 4

Mass 2 0 0123 456 Time (D ays)

0

y r

d 0

g N/ 0 g m

( 0 e t y) a

d

Ra 0 - l e a

v st o a 0 w m

Re 0 a i

n

o 0

m 0 Am 0 100 200 300 400 500 Ammonia Concentr ation (mg N/L)

Figure F- 9. Microcosm results from the test at the following conditions: 22oC, 500 mg N/L, 17% Oxygen and acclimated: Set 1.

283 180 NH4 160 ) NO3

N 140 g NO2 m 120 (

s 100 ie

c 80 e p 60 S N

40 f o 20 s

s a 0 M 0123456 Time (Day s)

70 N2

) 60

N N20

g 50 m (

s

e 40 i ec

p 30 S N

f 20

o

ss 10

a M 0 01234 56 Time (D ays)

0.40

g 0.35 m ( ) e 0.30 t y da Ra - 0.25 l e a t v

s

o 0.20 a

m w

y 0.15

Re a i

g dr 0.10 / n o N

m 0.05

Am 0.00

0 200 400 600 800 1000 Ammonia Conce ntration (mg N/L)

Figure F- 10. Microcosm results from the test at the following conditions: 22oC, 1000 mg N/L, 17% Oxygen and acclimated: Set 1.

284

3.0 es 2.5 A mmonia eci p Nitrate S 2.0 n ) e g N o g 1.5 r t

(m

Ni 1.0 f

0.5 ass o

M 0.0

0 5 10 15 Tim e (Days)

16 N2

) N2O

N g m 12 (

N2O

d 8

N2 an f 4

ass o M 0

0 5 10 15 Tim e (Days)

Figure F- 11. Microcosm results from the test at the following conditions: 22oC, no ammonia, 100% oxygen: Biotic Control.

285

g 90.00

m

( 80.00

s e

i 70.00

c e 60.00 p

S 50.00

n Ammonium e N) 40.00

og 30.00 Nitrate r

t i 20.00 Nitrite

N 10.00

of s

s 0.00 a

M 0 5 1 0 15 20 25 Time (Days)

16 ) 14 Nitrogen Gas

(mg N 12 s e i c 10 e p 8

n S 6

oge r t i 4 N 2 of s

s 0 a M 0 5 10 15 20 25

T ime (Days)

Figure F- 12. Microcosm results from the test at the following conditions: 22oC, 500 mg N/L, 100% oxygen: Abiotic Control.

286

g 80.00

70.00 es (m 60.00 eci p 50.00 S

en Ammonia

N) 40.00

g o

r 30.00 Nitrate t i

N 20.00 Nitrite f 10.00

ass o 0.00

M 0 5 10 15 20 Ti me (Days)

) 10 N

g 9 m 8

ies ( 7 Nitrogen Gas ec 6 p S 5 n e

g 4 o r t i 3 N f 2 1

Mass o 0 0 5 10 15 20

Time (Days)

Figure F- 13. Microcosm results from the test at the following conditions: 22oC, 500 mg N/L, 100% oxygen: Abiotic Control, Set 2.

287

g 80.00

m (

s 70.00 Nitrate e i

c 60.00 Ammonia e

p 50.00 S n

N) 40.00 oge

r 30.00 t i

N 20.00

of 10.00 s

s a 0.00 M 0 5 10 15 20 Ti me (Days)

)

N 10

g m

(

s 8

e i c

e p 6

n S 4 oge r

t i 2 N Nitrogen Gas f

o s

s 0 a

M 0 5 10 15 20

Time (Days)

Figure F- 14. Microcosm results from the test at the following conditions: 22oC, 500 mg NO3-N/L, 100% oxygen: Nitrate Abiotic Control, Set 1.

288 50.00 g m

( Nitrate

s 40.00 e

i Am m onia c

e

p 30.00

n S N) 20.00

oge r t i

N 10.00

f o

s

s 0.00

a M 0 5 10 15 20 Ti me (Days)

20 ) 18 g N 16 m

( 14

s

e Nitrogen Gas i 12 c e p 10

n S 8 6 oge

r t i 4 N 2

of s

s 0

a

M 0 5 10 15 20 Ti me (Days)

Figure F- 15. Microcosm results from the test at the following conditions: 22oC, 500 mg NO3-N/L, 100% oxygen: Nitrate Abiotic Control, Set 2.

289

Microcosms at 22oC

Unacclimated Waste

290 70.00

) 60.00 Ammonium N

g m 50.00 NO3 (

s e i NO2 ec 40.00

p

n S 30.00 oge r t

i N f 20.00 o

s s

a

M 10.00

0.00 0123456789101112131415 Time (days)

10

) N N2

ss (mg

Ma s

a G

n e g

o r t i

N

0 0123456789101112131415 Time (days)

0.1

) te y

a 0.08 da R - l e a

t 0.06 v s a mo w

e 0.04

y R

a

g dr 0.02 ni /

g N 0 mmo

(m A 0 100 200 300 400 500

Ammonia Concentration (mg N/L)

Figure F- 16. Microcosm results from the test at the following conditions: 22oC, 500 mg N/L, 100% oxygen: Unacclimated waste, Set 1.

291 70.00

60.00 Ammonium ) N

g 50.00 NO3 m

s ( ie 40.00 NO2

ec p

S 30.00 N f

o s

s 20.00

a M 10.00

0.00 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Time (days)

20

) Nitrogen Gas g N 15 m ( s e i c e p 10 S of N

s s a

M 5

0 024681012141618 Time (days)

0.1

e ) t y

a 0.08 R da

- l e a

t 0.06

s a mov

w

e 0.04 y R a

g dr 0.02 / 0 mmoni (mg N A 0 50 100 150 200 250 300

Ammonia Concentration (mg N/L)

Figure F- 17. Microcosm results from the test at the following conditions: 22oC, 500 mg N/L, 100% oxygen: Unacclimated waste, Set 2.

292

Microcosms at 35oC

293 160

) 140 N g

m 120

es ( 100 Ammonium eci Nitrite p

S 80 Nitrate

n e g 60 o r

t i

N 40 f o 20 ass

M 0

0123456 Time (Days)

36 N2

O

2 32 N2O 28

24) and N 20 N 2

N 16 (mg 12

ss of 8 a 4 M 0 0246 Ti me (Days)

0.60

) 0.50

day e* t 0.40 s

a 0.30 y w

dr g

/ 0.20

g N 0.10 m

( e t

a 0.00 R 0 50 100 150 200 250 300 350 Ammonia Co ncentration (mg/L-N)

Figure F- 18. Microcosm results from the test at the following conditions: 35oC, 500 mg N/L, 100% oxygen: Set 1.

294 70

) 60 Ammonium g N

m Nitrite

( 50 Nitrate s

e i c

e 40 p

n S 30

oge r t i 20

N

of s 10 s

a M 0 0123 456 Time (Days)

20 N2

O

2 N2O 16

12) and N N 2 N

(mg 8

ss of 4 a

M 0 0246 T ime (Days)

0.80 0.70

ay) d

* 0.60

e t

s 0.50

a 0.40 y w dr 0.30 g

/

N 0.20 g

m 0.10 (

e t a 0.00 R 0 50 100 150 200 250 300 Ammonia Concentration (mg N/L)

Figure F- 19. Microcosm results from the test at the following conditions: 35oC, 500 mg N/L, 100% oxygen: Set 2.

295 80

70 N) Ammonium g Nitrite m 60 ( Nitrate s e i 50 c

e p S

40

en g

o 30 r

t i

N f 20

o

ss 10 a M 0 00.51 1.522.5 T ime (Days)

0.60

y) 0.50

a d e* 0.40 st

a 0.30 y w r g d / 0.20

N g 0.10

m (

e t

a 0.00 R 0 100 200 300 400 500

Ammonia C oncentration (mg/L-N)

Figure F- 20. Microcosm results from the test at the following conditions: 35oC, 500 mg N/L, 100% oxygen: Set 3.

296 200 180 ) N 160 NO3 g m 140 NH4 NO2 es ( 120 eci

p 100

S

80 N

f o 60

s s a 40 M 20 0 00.511 .522.533.5 Time (Days)

N2

) 12

N N2O g 10 m (

O 8 2 N

d 6

an 2 4

N of 2

ass

M 0 012 34 T ime (Days)

0.7000

y) 0.6000 a d 0.5000 e* st a 0.4000 y w

r 0.3000 d g / 0.2000 N

g

m 0.1000

( e t

a 0.0000 R 0 100 200 300 400 500 600 700

Ammonia Conc entration (mg/L-N)

Figure F- 21. Microcosm results from the test at the following conditions: 35oC, 1000 mg N/L, 100% oxygen: Set 1.

297 160 g m ( 140 NO3 s NH4 120

asse NO2

M 100 e it

r t N) i 80

N d 60 a an i 40 n

o m 20

m A 0 00.5 11.522.5 Time (Days)

g 10 N2 m

( N2O O 8 2 N 6

N) and

2 4

N 2

ss of a 0 M 01 23 Time (Days)

0.8000

) 0.7000

day

* 0.6000

e

st 0.5000

a 0.4000 y w

dr 0.3000 g /

N 0.2000 g

m 0.1000 ( e

t

a 0.0000

R 0 200 400 600 800 1000 Ammoni a Concentration (mg N/L)

Figure F- 22. Microcosm results from the test at the following conditions: 35oC, 1000 mg N/L, 100% oxygen: Set 2.

298 80 70 N)

g 60 m ( 50 NH4 es 40 NO3

eci

p 30 NO2 20 N S

f

o 10

ss 0 a M 01 234 Time (Days)

30 N2

) N20

N 25

g

m 20

ies (

c 15 e p

S 10 N

f 5 ss o a

M 0 012 3456 Tim e (Days)

0.40

g / 0.35

g N 0.30

(m

te 0.25 a ay) d R l

e- 0.20

a v o ast 0.15 m

e y w r R 0.10 d

a

ni o 0.05

m

m 0.00

A 0 100 200 300 400 500 Ammonia Conc entration (mg N/L)

Figure F- 23. Microcosm results from the test at the following conditions: 35oC, 500 mg N/L, 5% oxygen: Set 1.

299 80 NH4 70 NO3 N)

g 60

m 50 NO2

es ( 40

eci

p 30

20 N S

f 10 0 ass o M 01 234 Time (Days)

35 N2 )

N 30 N20 g

m 25 (

ies 20

ec

p 15

S

N 10 f

o 5 ss

a

M 0

0123456 Ti me (Days)

1

1 (mg ) te y a 0 da R -

l e a t

s 0 a mov w

e y

R 0

a g dr / N 0

mmoni

A 0 0 100 200 300 400 500 Ammonia Concentration (mg N/L)

Figure F- 24. Microcosm results from the test at the following conditions: 35oC, 500 mg N/L, 5% oxygen: Set 2.

300 180 NH4 160 NO3 140 NO2 120 100 80 60 40 20 0

Mass of N)N of Species (mg Mass 01234

Time (Days)

25 N2

20 N20

15

10

5

Mass Species of N (mg N) 0 0123456 Time (Days)

0.80 0.70

0.60 0.50 0.40

0.30

dry waste-day) 0.20

0.10

0.00 Ammonia Removal Rate (mgN/g 0 200 400 600 800 1000

Ammonia Concentration (mg N/L) Figure F- 25. Microcosm results from the test at the following conditions: 35oC, 1000 mg N/L, 5% oxygen: Set 1.

301 160 NH4 140 NO3 120 NO2 100 80 60

40 20 0 Mass of N Species (mg N) (mg N Species of Mass 01234 Time (Days)

35 N2

30 N20 25

20 15

10 5

Mass Species of N (mg N) 0 0 0.5 1 1.5 2 2.5 3 3.5 4 Time (Days)

0.80 y 0.70

0.60

0.50 0.40

0.30 waste-day) 0.20

0.10

Ammonia removal Rate (mg N/g dr 0.00 0 200 400 600 800 1000 Ammonia Concentration (mg N/L)

Figure F- 26. Microcosm results from the test at the following conditions: 35oC, 1000 mg N/L, 5% oxygen: Set 2.

302 80 NH4

70 NO3 60 NO2 50

40 30 20 10 0

Mass of Species N (mg N) 01234 Time (Days)

14 N2

12 N20 10

8

6

4

2 Mass ofMass N Species (mg N) 0

00.511.522.533.54 Time (Days)

0.800

0.600

0.400

0.200

0.000

(mg N/g dry waste-day)) dry N/g (mg Ammonia Removal Rate Rate Removal Ammonia 0 100 200 300 400 500 Ammonia Concentration (mg N/L))

Figure F- 27. Microcosm results from the test at the following conditions: 35oC, 500 mg N/L, 17% oxygen: Set 1.

303 70 NH4 60 NO3 50 NO2 40

30 20 10

0 Mass of N Species (mg N) (mg N Species of Mass 01234 Time (Days)

18 N2 16 N20 14 12 10 8 6 4 2

Mass ofMass N Species (mg N) 0 0123456 Time (Days)

0.800

0.600

0.400

0.200

(mg N/g dry waste-day) dry N/g (mg 0.000 Ammonia Removal Rate 0 100 200 300 400 500 Ammonia Concentration (mg N/L)

Figure F- 28. Microcosm results from the test at the following conditions: 35oC, 500 mg N/L, 17% oxygen: Set 2.

304 200 NH4 180 NO3 160 140 NO2 120 100 80 60 40 20

0

Mass N) of (mg Species N 01234

Time (Days)

45 N2 40 N20 35 30 25 20

15

10

Mass of N Species (mg N) 5 0 00.511.522.533.54 Time (Days)

1.000

0.750

0.500

0.250

dry waste-day) 0.000 0 200 400 600 800 1000

Ammonia Removal Rate (mg N/g (mg N/g Rate Ammonia Removal Ammonia Concentration (mg N/L)

Figure F- 29. Microcosm results from the test at the following conditions: 35oC, 1000 mg N/L, 17% oxygen: Set 1.

305

120

100

80

60

40 Ammonium Nitrite Nitrate 20

Mass of Nitrogen Species (mg N) (mg Species Nitrogen of Mass 0 0123456 Time (Days)

12

10

N2 8 N2O 6

4 2

0 Mass of N2 and N2O (mg N) (mg N2O N2 and of Mass 0246 Time (Days)

Figure F- 30. Microcosm results from the test at the following conditions: 35oC, 500 mg N/L, 17% oxygen: Abiotic Control, Set 1.

306

120 Ammonium 100 Nitrite 80 Nitrate

60

40

20

Mass of Nitrogen Species (mg N) 0 01234567 Time (Days)

4

3 N2

2 N2O

1

0 Mass of N2 and N2O (mg N) (mg N2O N2 and of Mass 0246 Time (Days)

Figure F- 31. Microcosm results from the test at the following conditions: 35oC, no ammonia, 100% oxygen: Biotic Control, Set 1.

307

180 160 140

120 100 Ammonium 80

60 Nitrite

Nitrate 40 20

Mass ofMass NitrogenSpecies (mg N) 0 01234567 Time (Days)

8 7

6 N2 5 N2O 4

3

2 1

Mass of N2 and N2O (mg N) Mass of and (mg N2 N2O 0

0123456 Time (Days)

Figure F- 32. Microcosm results from the test at the following conditions: 35oC, 500 mg NO3-N/L, 100% oxygen: Nitrate Biotic Control, Set 1.

308

Microcosms at 45oC

309 100 450

400 80 NH4 350 60 300 NO3 250 40

200

Ammonia Mass (mg N) (mg Mass Ammonia 20 N) Mass (mg Nitrate 150

0 100 0123456 Time (Days)

4 N2 N2O

) 3

2

1

Mass of N2 and N2O (mg N (mg N2O N2 and of Mass 0 0246

Time (Days)

0.250

0.200

0.150

0.100

0.050 dry waste-day) dry

0.000 0 100 200 300 400 Ammonia Removal Rate (mg N/g N/g (mg Rate Removal Ammonia Ammonia Concentration (mg/L-N)

Figure F- 33. Microcosm results from the test at the following conditions: 45oC, 500 mg N/L, 100% oxygen: Set 1.

310 100 400 NH4

80 NO3 350

60 300 40

Ammonia Mass (mg N) (mg Mass Ammonia 250 20

0 200 02468 Time (Days)

16 N2 N2O 12

8

4

Mass of N2 and N2O (mg N) (mg N2O and N2 of Mass 0 0246810 Time (Days)

0.250

0.200

0.150

0.100

0.050 dry waste-day) dry

0.000

0 100 200 300 400 Ammonia Removal Rate (mg N/g N/g (mg Rate Removal Ammonia Ammonia Concentration (mg/L-N)

Figure F- 34. Microcosm results from the test at the following conditions: 45oC, 500 mg N/L, 100% oxygen: Set 2.

311 100 500.00 NH4

450.00

75 NO3

400.00

50 350.00

300.00 Nitrate Mass (mg N)

Ammonia Mass (mg N) (mg Ammonia Mass 25 250.00

0 200.00 012345 Time (Days)

10 N2 N2O 8

6

4

2

Mass and of (mg N2 N2O N) 0

0246 Time (Days)

0.450 0.400 0.350 0.300 0.250

0.200

0.150 waste-day) 0.100 0.050 0.000 0 50 100 150 200 250 300 Ammonia Removal Rate (mg N/g dry dry N/g (mg Rate Removal Ammonia Ammonia Concentration (mg/L-N)

Figure F- 35. Microcosm results from the test at the following conditions: 45oC, 500 mg N/L, 100% oxygen: Set 3.

312 150 350.00

125 300.00

250.00 100 NH4 200.00 75 NO3 150.00 50

100.00 NitrateMass (mg N) Ammonia (mg N) Mass 25 50.00

0 0.00 012345 Time (Days)

10 N2 N2O 8

6

4

2

Mass of N2 and N2O (mg N) N2O and N2 of Mass 0 0246 Time (Days)

0.450

0.400

0.350 0.300 0.250 0.200 0.150 0.100 0.050 0.000 Rate (mg N/g dry waste*day) N/g dry Rate (mg 0 200 400 600 800 1000 1200

Ammonia Concentration (mg/L-N)

Figure F- 36. Microcosm results from the test at the following conditions: 45oC, 1000 mg N/L, 100% oxygen: Set 1.

313 180 NH4 160 NO3 140 120 100 80 60 40 20 0 Mass of N Species (mg N) (mg N Species of Mass 01234 Time (Days)

30 N2 25 N20

20 15

10

5

Mass ofMass N (mg Species N) 0 0123456 Time (Days)

0.600

g 0.500

0.400

0.300

0.200 0.100

N/g dry waste-day) dry N/g 0.000

Ammonia Removal Rate (m Rate Removal Ammonia 0 100 200 300 400 500

Ammonia Concentration (mg N/L))

Figure F- 37. Microcosm results from the test at the following conditions: 45oC, 500 mg N/L, 5% oxygen: Set 1.

314 160

140 NH4 120 NO3 100

80 60 40

20 Mass ofMass N (mg Species N) 0 0246810 Time (Days)

20 N2 18 16 N20 14 12 10 8 6 4 2

Mass ofMass N Species (mg N) 0 0123456 Time (Days)

0.500

0.400

0.300

0.200

0.100

dry waste-day) 0.000

0 100 200 300 400 500 Ammonia Removal Rate (mg Ammonia Rate Removal N/g Ammonia Removal Rate (mg N/L)

Figure F- 38. Microcosm results from the test at the following conditions: 45oC, 500 mg N/L, 5% oxygen: Set 2.

315 300 NH4 250 NO3 200 150

100 50

0 Mass Species of N (mg N) 0123456 Time (Days)

25 N2 20 N20

15

10

5

Mass of N Species (mg N) 0 0123456 Time (Days)

0.600

0.400

N) 0.200

0.000 Mass of N2 and N2O (mg (mg N2O N2 and of Mass 0 250 500 750 1000

Ammonia Concentration (mg N/L)

Figure F- 39. Microcosm results from the test at the following conditions: 45oC, 1000 mg N/L, 5% oxygen: Set 1.

316 250

200

150 NH4

100 NO3

50

0 Mass of N Species (mg N) (mg N Species of Mass 01234 Time (Days)

8 N2

7 N20 6

5

4 3 2 1

Mass ofMass N Species (mg N) 0 0123456 Time (Days)

e 0.500

0.400

0.300

0.200

0.100

(mg N/g dry waste-day) 0.000 Ammonia Removal Rat Removal Ammonia 0 100 200 300 400 500 Ammonia Concentration (mg N/L)

Figure F- 40. Microcosm results from the test at the following conditions: 45oC, 500 mg N/L, 17% oxygen: Set 1.

317 100 250.00

80 200.00

60 150.00 NH4 40 NO3 100.00

Species (mg N) 20 50.00 Mass of Ammonia

Mass of Nitrate (mg N) 0 0.00 0246

Time (Days)

5 N2 4 N20

3

2

1

Mass of N Species (mg N) 0 0123456 Time (Days)

0.600

0.400

0.200

0.000

(mg N/g dry waste-day) dry N/g (mg Ammonia Removal Rate 0 100 200 300 400 500 Ammonia Concentration (mg N/L)

Figure F- 41. Microcosm results from the test at the following conditions: 45oC, 500 mg N/L, 17% oxygen: Set 2.

318 300

250 200 NH4 NO3 150

100

50

Mass of Species N (mg N) 0 01234 Time (Days)

7 N2 6 N20 5 4 3 2

1

Mass ofMass N (mg Species N) 0 0123456 Time (Days)

0.800

0.600

0.400

0.200

0.000 (mg N/g dry waste-day) Ammonia Removal Rate 0 100 200 300 400 500 600 700 800 900 100

0 Ammonia Concentration (mg N/L)

Figure F- 42. Microcosm results from the test at the following conditions: 45oC, 1000 mg N/L, 17% oxygen: Set 1.

319

100 90 80 70 60

50 40 Am m onium 30 Nitrite Nitrate 20 10

Mass of Nitrogen Species (mg N) (mg Species Nitrogen of Mass 0

01234567 Time (Days)

60 N2

50 N2O

40

30

20 10

Massand of N2 (mg N2O N) 0

0246 Time (Days)

Figure F- 43. Microcosm results from the test at the following conditions: 45oC, 500 mg N/L, 100% oxygen: Abiotic Control.

320

90 80 70 60 Am m onium 50 Nitrate 40

30

20 10

Mass of Nitrogen Species (mg N) Species Nitrogen of Mass 0 0123456 Time (Days)

6

5

4 N2

3 N2O

2

1 0 Mass of N2 and N2O (mg N) N2of (mg and N2O Mass 0246 Time (Days)

Figure F- 44. Microcosm results from the test at the following conditions: 45oC, no ammonia, 100% oxygen: Biotic Control.

321 APPENDIX G

ECONOMIC CALCULATIONS

322 Air Flowrate Calculations

323 324

325 Blower Power Requirement Calculations

326

327 Air Piping Calculations

328 Leachate Injection Pump Calculations

329

330