Citation
Hydrogen Sulfide Emissions and Control Strategies at Construction and Demolition Debris Landfills

Material Information

Title:
Hydrogen Sulfide Emissions and Control Strategies at Construction and Demolition Debris Landfills
Creator:
XU, QIYONG
Copyright Date:
2008

Subjects

Subjects / Keywords:
Adsorption ( jstor )
Chemicals ( jstor )
Diffusion coefficient ( jstor )
Drywall ( jstor )
Flasks ( jstor )
Gypsum ( jstor )
Landfills ( jstor )
Pollutant emissions ( jstor )
Sandy soils ( jstor )
Soils ( jstor )
City of Gainesville ( local )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Qiyong Xu. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
12/31/2007
Resource Identifier:
496174524 ( OCLC )

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

HYDROGEN SULFIDE EMISSIONS AN D CONTROL STRATEGIES AT CONSTRUCTION AND DEMOLITION DEBRIS LANDFILLS By QIYONG XU A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2005

PAGE 2

Copyright 2005 by Qiyong Xu

PAGE 3

To my parents and loving wife.

PAGE 4

ACKNOWLEDGMENTS I would like to express my sincerest gratitude and appreciation to the chairman of my committee, Dr. Timothy G. Townsend, for his guidance, support and advice throughout my graduate study in University of Florida. I would also like to thank the other members of my committee, Dr. Angela Lindner, Dr. Gabriel Bitton, and Dr. Madeline Rasche for their participation, guidance and valuable ideas. I wish to thank the Florida Center for Solid and Hazardous Waste Management (FCSHWM) for funding the research. And I thank Mr. John Cook in Waste Management for his support and help on the field work at the Keene Road Landfill. I would like to thank Pradeep Jain for sharing his idea and assistance in this research. I would like to thank Aaron Jordan, Hwidong Kim, and Murat Semiz for their assistance in the field work. I also thank Kimberly Cochran, Judd Larson and Steve Musson for reviewing some parts of this dissertation. My thanks are extended to the follow graduate students in the solid and hazardous waste group, for their kindness and cooperation. Special thanks go to my parents, to my aunts and uncles for their continued love and support throughout my studies. Finally, the greatest thanks go to my wife, Jiaoju Ge, for her understanding, encouragement, patience, and love. iv

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES ...........................................................................................................viii LIST OF FIGURES ...........................................................................................................ix ABSTRACT......................................................................................................................xii CHAPTER 1 INTRODUCTION........................................................................................................1 1.1 Background and Problem Statement..................................................................1 1.2 Research Objectives...........................................................................................4 1.3 Research Approach............................................................................................6 1.4 Outline of Dissertation.......................................................................................7 2 INHIBITION OF HYDROGEN SULFIDE GENERATION FROM GYPSUM DRYWALL USING CHEMICAL INHIBITORS......................................................11 2.1 Introduction......................................................................................................11 2.2 Mechanisms for Inhibition of H 2 S Generation................................................12 2.3 Materials and Methods.....................................................................................15 2.3.1 Chemical Inhibitor Selection.................................................................15 2.3.2 Flask Experiment...................................................................................16 2.3.3 Inhibition Column Experiment..............................................................16 2.3.4 Sample Collection and Analysis............................................................17 2.4 Results and Discussion....................................................................................18 2.4.1 H 2 S Generation from Gypsum Drywall.................................................18 2.4.2 Effect of Sodium Molybdate on H 2 S Generation..................................19 2.4.3 Effect of Ferric Chloride on H 2 S Generation.........................................20 2.4.4 Effect of Hydrate Lime on H 2 S Generation...........................................22 2.4.5 Environmental and Industrial Implication.............................................24 2.5 Summary..........................................................................................................27 v

PAGE 6

3 ATTENUATION OF H 2 S AT CONSTRUCTION AND DEMOLITION DEBRIS LANDFILLS USING ALTERNATIVE COVER MATERIALS..............................35 3.1 Introduction......................................................................................................35 3.2 Materials and Methods.....................................................................................36 3.2.1 Field Study.............................................................................................37 3.2.1.1 Cover materials and plot construction...........................................37 3.2.1.2 Field sampling...............................................................................38 3.2.2 Laboratory Experiments.........................................................................40 3.3 Results and Discussion....................................................................................40 3.3.1 H 2 S Emissions from the Testing Plots...................................................40 3.3.2 Soil Vapor H 2 S Concentrations in the Testing Plots.............................41 3.3.3 H 2 S Removal by the Cover Materials....................................................43 3.3.4 Inhibition of Underlying H 2 S Production..............................................45 3.4 Summary..........................................................................................................47 4 H 2 S EMISSION RATE AND TEMPORAL CHANGE AT CONSTRUCTION AND DEMOLITION DEBRIS LANDFILLS............................................................53 4.1 Introduction......................................................................................................53 4.2 Materials and Methods.....................................................................................54 4.2.1 Field Experiment....................................................................................54 4.2.2 Laboratory Experiments.........................................................................55 4.3 Results and Discussion....................................................................................56 4.3.1 H 2 S Emission Rate in the Field Study...................................................56 4.3.2 Temporal Variation of H 2 S Emission Rate............................................57 4.3.3 Effect of H 2 S Concentration on H 2 S Emission Rate..............................59 4.3.4 Effect of Soil Moisture on H 2 S Emissions.............................................60 4.3.5 Effect of Soil Moisture on Soil Vapor H 2 S Concentration....................62 4.4 Summary..........................................................................................................65 5 MODELING OF H 2 S MIGRATION THROUGH LANDFILL COVER MATERIALS.............................................................................................................73 5.1 Introduction......................................................................................................73 5.2 Theory..............................................................................................................74 5.3 Model Simulation and Laboratory Experiment...............................................76 5.3.1 Model Simulation...................................................................................76 5.3.2 Laboratory Column Experiment............................................................77 5.3.2.1 Column construction.....................................................................77 5.3.2.2 Cover materials.............................................................................77 5.3.2.3 Column operation and gas sampling.............................................78 5.3.2.4 Laboratory experiments.................................................................79 5.4 Results and Discussion....................................................................................80 5.4.1 Effect of Initial H 2 S Concentration on H 2 S Migration in the Cover Soils.........................................................................................................80 5.4.2 Effect of Advection Velocity on H 2 S Migration in the Cover Soils......81 vi

PAGE 7

5.4.3 Effect of Effective Diffusion Coefficient on H 2 S Migration in the Cover Soils..............................................................................................82 5.4.4 Effect of H 2 S Adsorption Coefficient on H 2 S Migration in the Cover Soils.........................................................................................................83 5.4.5 Model Application and Limitations.......................................................83 5.5 Summary..........................................................................................................86 6 SUMMARY AND CONCLUSIONS.........................................................................97 6.1 Summary..........................................................................................................97 6.2 Conclusions......................................................................................................99 6.3 Future Work...................................................................................................100 APPENDIX A SUPPLEMENTAL LEACHATE DATA FOR THE INHIBITION COLUMN EXPERIMENT.........................................................................................................102 B CONSTRUCTION PROCEDURE OF THE TESTING AREA AT A CONSTRUCTION AND DEMOLITION DEBRIS LANDFILL............................103 C SUPPLEMENTAL DATA IN THE FIELD STUDY..............................................106 D CALCULATION OF HYDROGEN SULFIDE RATE USING THE FLUX CHAMBER METHODS..........................................................................................119 LIST OF REFERENCES.................................................................................................120 BIOGRAPHICAL SKETCH...........................................................................................127 vii

PAGE 8

LIST OF TABLES Table page 2-1. Concentrations of chemical inhibitors in the flask experiment..................................28 2-2. H 2 S concentrations in the flasks with addition of sodium molybdate........................28 2-3. pH change of ferric chloride solutions after mixing with gypsum drywall................28 3-1. Summary of cover materials used in the field study..................................................48 3-2. Results of pH, temperature, and particle size of cover materials used in the field study.........................................................................................................................48 4-1. Influencing factors investigated in the laboratory experiment...................................66 4-2. Comparison of reported landfill gas emission rates...................................................66 4-3. The change of soil vapor H 2 S concentrations under various soil moistures in the laboratory experiment..............................................................................................66 5-1. Parameters values used for the sensitivity analysis....................................................88 5-2. Summary of the characteristics of cover materials in the column experiment...........88 5-3. Summary of experimental operation conditions in the column experiments.............88 5-4. Calculation values of the D, v, and of the cover materials......................................88 A-1. Results of leachate parameter in the inhibition columns.........................................102 C-1. H 2 S emission rate change in the sandy soil testing plot during the 28-hour continuous monitoring............................................................................................116 C-2. Temperature change at different depths of the cover soils during the 28-hour continuous monitoring............................................................................................117 C-3. Soil vapor H 2 S concentration change at different depths of the cover soil during the 28-hour continuous monitoring........................................................................118 C-4. Soil moisture content change during the 28-hour continuous monitoring...............118 viii

PAGE 9

LIST OF FIGURES Figure page 1-1. Annual gypsum drywall products in the U.S. from 1991 to 2005................................9 1-2. Hydrogen sulfide health and safety hazards as a function of concentration..............10 2-1. Schematic of flask experiment apparatus...................................................................29 2-2. Schematic of laboratory inhibition column experiment.............................................29 2-3. Comparison of the average H 2 S concentration from gypsum drywall in the flask and column experiments...........................................................................................30 2-4. The inhibition effect of Na 2 MoO 4 on H 2 S generation. A) The flask experiment results. B) The column experiment results...............................................................31 2-5. The inhibition effect of FeCl 3 on H 2 S generation. A) The flask experiment results. B) The column experiment results...........................................................................32 2-6 The inhibition effect of Ca(OH) 2 on H 2 S generation. A) The flask experiment results. B) The column experiment results...............................................................33 2-7. Change of average H 2 S concentrations and average leachate pH in the lime columns....................................................................................................................34 3-1. Schematic of the testing plot. A) Top view. B) Side view.........................................49 3-2. Schematic of the H 2 S emission rate measurement system in the field study.............50 3-3. Monthly change of H 2 S emission rate at the sandy soil testing plot..........................50 3-4. H 2 S concentration profiles of the testing plots on September 10 th 2004....................51 3-5. Average H 2 S concentration underneath different cover materials from August 2004 to January 2005...............................................................................................51 3-6. Comparison of H 2 S removal by cover materials in laboratory serum bottle experiment................................................................................................................52 3-7. Average sulfur content of the cover soils at different depths in the field study.........11 4-1. Schematic of the laboratory buchner funnels experiment..........................................67 4-2. The change of H 2 S emission rates and bottom concentrations in the sandy soil plot67 4-3. Temporal change of H 2 S emission rates at daytime in the sandy soil plot.................68 ix

PAGE 10

4-4. Temporal change of A) H 2 S emission rate; B) soil moisture content and soil temperature; C) H 2 S concentrations in the 28-Hour continuous field study............69 4-5. The effect of soil H 2 S concentrations on H 2 S emissions in the field study................70 4-6. The change of H 2 S emission rate with different bottom H 2 S concentrations (laboratory experiment)............................................................................................70 4-7. The effect of soil moisture on H 2 S emissions in the field study.................................71 4-8. The effect of soil moisture and temperature on H 2 S emission rate in the laboratory experiment................................................................................................................71 4-9. Schematic of the relationship among H 2 S emissions and possible influencing factors.......................................................................................................................72 5-1. Schematic of H 2 S migration in landfill cover soils....................................................89 5-2. Schematic of laboratory cover soil column................................................................89 5-3. The effect of H 2 S initial concentration on H 2 S migration. A) Modeling results. B) Experiment results....................................................................................................90 5-4. The effect of H 2 S advection velocity on H 2 S migration. A) Modeling results. B) Experiment results....................................................................................................91 5-5. The effect of effect diffusion coefficient on H 2 S migration. A) Modeling results. B) Experiment results...............................................................................................92 5-6. The change of H 2 S diffusion coefficient vs. soil moisture content and soil compaction...............................................................................................................93 5-7. The effect of adsorption coefficients of cover soils on H 2 S migration. A) Modeling results. B) Experiment results..................................................................94 5-8. Flow chart for designing the depth of landfill cover..................................................95 5-9. Comparison of modeling results and experimental data............................................96 B-1. Construction procedure of the testing area in the field study..................................104 B-2. Picture of the testing area in the field study.............................................................105 B-3. Picture of the H 2 S emission measurement system in the field study.......................105 C-1. H 2 S concentration profile in the testing plots on March 2 nd 2004...........................106 C-2. H 2 S concentration profile in the testing plots on March 9 th 2004............................106 x

PAGE 11

C-3. H 2 S concentration profile in the testing plots on March 30 th 2004..........................107 C-4. H 2 S concentration profile in the testing plots on April 20 th 2004............................107 C-5. H 2 S concentration profile in the testing plots on April 25 th 2004............................108 C-6. H 2 S concentration profile in the testing plots on May 3 rd 2004...............................108 C-7. H 2 S concentration profile in the testing plots on May 10 th 2004.............................109 C-8. H 2 S concentration profile in the testing plots on June 15 th 2004.............................109 C-9. H 2 S concentration profile in the testing plots on June 30 th 2004.............................110 C-10. H 2 S concentration profile in the testing plots on July 6 th 2004.............................110 C-11. H 2 S concentration profile in the testing plots on July 7 th 2004.............................111 C-12. H 2 S concentration profile in the testing plots on July 12 th 2004...........................111 C-13. H 2 S concentration profile in the testing plots on July 20 th 2004...........................112 C-14. H 2 S concentration profile in the testing plots on July 28 th 2004...........................112 C-15. H 2 S concentration profile in the testing plots on July 29 th 2004...........................113 C-16. H 2 S concentration profile in the testing plots on August 27 th 2004.......................113 C-17. H 2 S concentration profile in the testing plots on September 10 th 2004.................114 C-18. H 2 S concentration profile in the testing plots on December 2 nd 2004...................114 C-19. H 2 S concentration profile in the testing plots on January 20 th 2005......................115 xi

PAGE 12

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy HYDROGEN SULFIDE EMISSIONS AND CONTROL STRATEGIES AT CONSTRUCTION AND DEMOLITION DEBRIS LANDFILLS By Qiyong Xu December 2005 Chair: Timothy G. Townsend Major Department: Environmental Engineering Sciences Hydrogen sulfide (H 2 S) gas generation and emissions from construction and demolition (C&D) debris landfills have become a major environmental concern in North America. The emission of H 2 S not only causes odor complaints, but also poses a potential health and safety threat to people living or working near these facilities. Research was conducted to investigate potential engineering methods to address H 2 S problems at C&D debris landfills, including inhibition of H 2 S generation and attenuation of H 2 S emissions. Three chemical inhibitors, including sodium molybdate, ferric chloride, and hydrated lime, were evaluated for their potential to inhibit H 2 S generation from gypsum drywall. Flask and column experiments were conducted to assess the necessary inhibition concentration of each inhibitor and to evaluate long-term inhibition effects on H 2 S generation, respectively. Results indicated that 10 mM sodium molybdate and 500 mM of ferric chloride effectively inhibited H 2 S generation over six-month period. However, due to the extremely low pH and large amount of chemical needed, ferric chloride may not be xii

PAGE 13

practical in the field. Low concentrations of sodium molybdate provide a potential method for reducing H 2 S generation from gypsum drywall. A 12 m by 18 m testing area was constructed at a C&D debris landfill to evaluate H 2 S attenuation by six different materials, such as sandy soil, fine concrete, compost, and lime-amended sandy soils. Results showed that the H 2 S emission rate was only detected from the testing plot using sandy soil as cover, with an average emission rate of 4.67 -6 mg m -2 s -1 over a 10-month period. The alternative cover materials effectively attenuated H 2 S emission by reducing H 2 S generation and adsorbing H 2 S. The temporal variation of H 2 S emission rate resulted from the effect of many parameters, such as soil moisture, temperature, H 2 S concentration, and barometric pressure. Finally, the migration of H 2 S in the cover soil was investigated by developing a mathematical model using an advection-diffusion equation and a gas flow equation. The laboratory experiment data were comparable with the simulation results. The migration model provides a potential method to estimate the H 2 S adsorption coefficient values of cover materials and to predict the depth of landfill cover soils based on a laboratory column experiment. xiii

PAGE 14

CHAPTER 1 INTRODUCTION 1.1 Background and Problem Statement Construction and demolition (C&D) debris represents one of the largest contributors of solid waste in the U.S. and includes waste material produced from the construction, renovation and demolition of buildings, roadways, and bridges. Typical C&D debris components include wood, gypsum drywall, concrete, metal, roofing material, and soil (EPA 1998). It has been estimated that C&D debris comprises approximately 23 percent of the municipal solid waste collected in Florida (FDEP 2000). The EPA estimated that approximately 136 million tons of building-related C&D debris was generated in 1996 and that the most common management practice for C&D debris is landfilling (EPA 1998). Historically, C&D debris has not received the same degree of attention as other solid waste streams (e.g., household waste, hazardous waste). It was once thought C&D debris landfills would generate little or no landfill gas because of the lack of readily biodegradable organic material in C&D debris (Flynn 1998). However, since the middle of the 1980s, cases where residents living near C&D debris landfills have complained about odorous gas emitted from these facilities have been noted (Gypsum Association 1992a and 1992b; Flynn 1998; Johnson 1986; Lee 2000). Although many gas compounds can contribute to malodorous conditions, hydrogen sulfide (H 2 S) has been identified as the major contributor to odor problems at C&D debris landfills (Lee 2000; Francoeur 1

PAGE 15

2 1993). At ten C&D debris landfills in Florida, H 2 S was found to range in concentration from less than 0.003 ppm to as high as 12,000 ppm (Lee 2000). It is well known that H 2 S formation can result from a series of biological reactions that reduce the sulfate in anaerobic environments and that gypsum drywall provides a large source of sulfate in C&D debris landfills (Gypsum Association 1992a and 1992b; Fairweather and Barlaz 1998; Townsend et al. 2000). Gypsum drywall, or wallboard, is extensively used as a construction material in the U.S. and is a major C&D debris component, typically ranging from 17% to 27% (Lee 2000). Approximately 30 million tons of gypsum drywall are manufactured in North America each year, and it is estimated that 10-12% of the drywall used for new construction is wasted (Musick 1992). The industry rule of thumb is that one pound of gypsum drywall scrap is generated for every square foot of constructed floor area, which translates to one ton per average house (Musick 1992). The production of gypsum drywall in the U.S. has been increasing over the past decade (Figure 1-1). Waste gypsum drywall can be recycled, but most currently ends up disposed in landfills. When gypsum drywall (approximately 90% CaSO 4 H 2 O and 10% paper) becomes wet in landfills, sulfate-reducing bacteria (SRB) flourish and utilize sulfate as an electron acceptor, producing H 2 S (Gypsum Association 1992a and 1992b; Fairweather and Barlaz 1998; Townsend et al. 2000 and 2005). Since C&D debris landfills are not normally required to have gas collection systems, as is the case with many MSW landfills, the generated H 2 S is not typically collected, resulting in H 2 S emissions at C&D debris landfills.

PAGE 16

3 Previous research showed ambient H 2 S concentrations at C&D debris landfills (measured at the surface of the landfill) to be variable, ranging from below 0.003 ppm to above 50 ppm (Lee 2000). Because of its distinctive “rotten egg” odor and low detectable odor threshold (reported as low as 0.5 ppb; Godish 1991), H 2 S emitted from C&D debris landfills can result in odor complaints from residents living around those facilities. In addition to the odor problem, as a toxic colorless air pollutant, H 2 S poses adverse impacts on human health at high concentrations (Flynn 1998; WHO. 2000; Selene and Chou 2003; Campagna et. al. 2004). Exposure to low concentrations of H 2 S may cause difficulty in breathing for some asthmatics and irritation to the eyes, nose, and throat. As H 2 S concentrations increase beyond 100 ppm, it quickly paralyzes the olfactory senses and begins to affect the whole body. At 500 ppm and higher, H 2 S can cause convulsions, respiratory arrest, coma, and even death (Figure 1-2). High concentrations of H 2 S (up to 80,000 ppm) have been reported in C&D debris landfill gas (Lee 2000; Flynn 1996), which means caution must be taken for those who are involved in excavation and pipe installation jobs in C&D debris landfills. Although ambient H 2 S concentrations in C&D debris landfills are much lower than that of landfill gas, long-term exposure to low levels of H 2 S can cause an increased risk of eye irritation, cough, headache, nasal blockage, impaired neurological function and pulmonary function compared to unexposed residents (WHO 2000; ATSDR 2004). The Sunset Sand Mine and Landfill, a C&D debris landfill in Central Florida, was closed in 1995 largely as a result of high concentrations of H 2 S in the surrounding neighborhoods. Nearby residents were evacuated twice. They reported eye, throat and lung irritation, as well as headaches, chest pain, and asthma (Crosson 1995).

PAGE 17

4 Some landfills greatly increase the dumping or tipping fee because of the problems associated with H 2 S production at C&D debris landfills (Nelson 1990). Some even refuse to accept waste gypsum drywall (EPA 1995). However, with the increasing use of drywall products (Figure 1-1), more and more waste gypsum drywall is produced. Gypsum drywall recycling is not readily available in most areas of the U.S. because of logistic and economical issues surrounding drywall collection, processing and marketing (Townsend et al. 2005). Despite the problems resulting from the land disposal of gypsum drywall, the majority of this waste stream continues to be managed by landfilling. Therefore, the control of H 2 S problems associated with landfilling gypsum drywall has become a large issue and challenge facing many C&D debris landfills throughout North America. 1.2 Research Objectives The general objective of this work was to investigate possible engineering and microbiological methods of reducing the H 2 S problems in C&D debris landfills. Since the H 2 S problem is the result of H 2 S generation from disposed gypsum drywall and subsequent emissions of H 2 S from landfills, this doctoral research explores the solutions from two aspects: preventing H 2 S generation and attenuating H 2 S emissions. In addition to exploring engineering methods for controlling H 2 S problems, research was conducted to better understand H 2 S movement through landfill cover soil. Four specific objectives were established in this research. The first objective was to evaluate the use of chemical inhibitors for inhibiting H 2 S generation from gypsum drywall. Since H 2 S generation in C&D debris landfills is the result of biological conversion of disposed gypsum drywall by sulfate-reducing bacteria

PAGE 18

5 (SRB), H 2 S generation may be reduced by inhibiting the activity of SRB with some chemicals, such as sodium molybdate and lime. The second objective was to evaluate the attenuation of H 2 S emissions by using various alternative cover materials. Landfill cover, utilized by landfill operators to control litter, odors, and fires, is applied at most C&D debris landfills. Previous laboratory research found that some materials available at landfills, such as concrete and compost, can more effectively remove H 2 S from gas streams than soil (Yang 2000; Townsend et al. 2000; Plaza 2003). Utilization of alternative cover materials may provide a cost-effective method to attenuate H 2 S emissions at C&D debris landfills. The third objective was to investigate the emission rate of H 2 S and the factors influencing H 2 S emissions at C&D debris landfills. Several studies have been performed to investigate landfill gas emissions from MSW landfills, but most of them focused on methane and carbon dioxide (Boeckx et al. 1996; Borjesson and Svensson 1997). In previous research, H 2 S emissions were different from site to site and were affected by some factors such as rainfall (Reinhart et al. 2005; Xu and Townsend 2004). The fourth objective was to mathematically simulate H 2 S movement in the cover soils by developing a migration model. After generation, H 2 S tends to migrate upward through landfill covers and into the atmosphere due to the concentration gradient and pressure difference, which result in H 2 S emissions from landfills. Therefore, the emissions of H 2 S, to a large extent, depend on H 2 S migration in the cover soils. Since the migration is influenced by many factors, such as soil type, particle size, as well as climate variables, understanding the migration of H 2 S in the cover soils can lead to better management strategies for reducing H 2 S emissions at C&D debris landfills.

PAGE 19

6 1.3 Research Approach Laboratory experiments, field studies, and mathematical modeling were used in this research to meet the objectives described above. Objective 1 Evaluate the inhibition effect of chemical inhibitors on H 2 S generation from gypsum drywall. Approach: Three chemicals were used to inhibit H 2 S generation from gypsum drywall. They were sodium molybdate, ferric chloride, and calcium hydroxide. The necessary inhibition concentration of each chemical and its long-term inhibition effect on H 2 S generation were evaluated. First, laboratory flask experiments were conducted to quickly examine the inhibition effect of each chemical at various concentrations. Then a column experiment was designed to investigate the long-term inhibition effect of the specific concentrations obtained from the flask experiment. Objective 2 Evaluate attenuation of H 2 S emissions by using various alternative cover materials. Approach: A field study was conducted to evaluate H 2 S attenuation by several alternative cover materials at a C&D debris landfill. A 12 m by 18 m testing area was constructed in the landfill and six different cover materials, including sandy soil, fine concrete, compost, and various lime amended sandy soils, were used as cover soils to attenuate H 2 S emissions. Soil vapor H 2 S concentrations and H 2 S emission rates were measured to evaluate the attenuation of H 2 S by these cover materials. Also, laboratory experiments were conducted to support field observations and to investigate possible reasons for the attenuation of H 2 S.

PAGE 20

7 Objective 3 Investigate temporal changes in H 2 S emission rates at C&D debris landfills and the factors influencing H 2 S emissions. Approach: H 2 S emission rates were measured using a 65-L flux chamber from the sandy soil testing plot constructed in Objective 2. A field study was performed for 10 months and changes of H 2 S emissions were monitored. It was observed that H 2 S emissions changed with time throughout the experimental period and were influenced by some factors, such as soil moisture. Laboratory experiments were also conducted to examine the possible factors affecting H 2 S emissions such as soil moisture and soil vapor H 2 S concentrations. Objective 4 Simulate H 2 S movement in cover soils by developing a migration model and evaluating the factors influencing H 2 S migration in cover soil. Approach: Based on a one dimensional gas diffusion-advection equation, a mathematical model of H 2 S migration in landfill covers was developed. A sensitivity analysis was undertaken to evaluate the effects of four key model parameters on the migration of H 2 S in the cover soils: concentration (C 0 ), diffusion coefficient (D), advection velocity (v), and H 2 S adsorption by cover soil (). A laboratory column experiment was conducted to compare the results from modeling simulations and to estimate the H 2 S adsorption coefficient of the cover materials. A potential method for designing the depth of landfill cover was provided using the migration model and the laboratory column experiment results. 1.4 Outline of Dissertation The dissertation is organized into six chapters. Chapter 1 provides background information regarding H 2 S problems in C&D debris landfills, as well as the objectives and approaches for this research. Chapter 2 investigates the inhibition effect of chemical inhibitors on H 2 S generation. Chapter 3 explores the attenuation of H 2 S by various

PAGE 21

8 alternative cover soils in a field study. Chapter 4 provides the information about H 2 S emission rate and the temporal change in a C&D debris landfill. Chapter 5 presents a mathematical modeling of H 2 S migration in landfill covers. Chapter 6 presents the summary and conclusions of this dissertation, followed by the bibliography.

PAGE 22

9 Year 199019921994199619982000200220042006 Gypsum drywall (million square feet) 1600018000200002200024000260002800030000320003400036000 Figure 1-1. Annual gypsum drywall products in the U.S. from 1991 to 2005

PAGE 23

10 0.011010.110,000(1%)1,000100100,000(10%)Odor Threshold (0.0005~0.1 ppm)Offensive Odor: Rotten Egg Smell Loss of Sense of Smell (100 ppm)Throat & Eye IrritationHeadaches & NauseaLower Explosive Limit(43,000 ppm)DEATHImmediate Collapse with Respiratory ParalysisStrong Nervous System StimulationImminent Life Threat Conjunctivitis & Respiratory Tract Irritation OlfactoryHydrogen Sulfide Concentration (ppm) Figure 1-2. Hydrogen sulfide health and safety hazards as a function of concentration

PAGE 24

CHAPTER 2 INHIBITION OF HYDROGEN SULFIDE GENERATION FROM GYPSUM DRYWALL USING CHEMICAL INHIBITORS 2.1 Introduction A potential environmental problem of construction and demolition (C&D) debris landfills is hydrogen sulfide (H 2 S) gas generation (Nelson 1990; Gypsum Association 1992a and 1992b). H 2 S results from the biological reduction of sulfate from disposed gypsum drywall (CaSO 4 .2H 2 O), one of the major components of C&D debris (Flynn 1998; Townsend et al. 2000). A C&D debris landfill provides an ideal environment for H 2 S generation: the drywall provides a sulfate source; the environment tends to be anaerobic, moist, and warm; and wood, vegetation and paper provide a carbon source (Reinhart et al. 2005). High concentrations of H 2 S have been reported in previous studies in the gas from C&D debris landfills (Flynn 1998; Lee 2000). The generation of H 2 S at these facilities not only causes odor problems, but may also pose a severe health threat to those who might be exposed to the gas through waste excavation or pipe installation. In addition, H 2 S can be corrosive to landfill equipment and can be converted to SO 2 in the atmosphere, an air pollutant that has its own environmental detriments. Chemical agents such as sodium molybdate and ferric compounds are known to have the ability to reduce biological H 2 S production, and as such, several researchers have investigated these chemicals for preventing H 2 S generation in marine sediments, anaerobic digestors, and the petroleum industry (Clancy et al. 1992; Okabe et al. 1999; 11

PAGE 25

12 Nemati et al. 2001; Picot et al. 2001; Isa and Anderson 2005). The hypothesis of this research was that chemical addition may provide a potential biotechnological method to reduce the odor problems associated with H 2 S generation at C&D debris landfills. The inhibition of H 2 S generation from gypsum drywall using several chemical additives under simulated landfill conditions was investigated. Both flask and column experiments were performed. The flask experiments provided a quick evaluation of the effect of chemical inhibitors in various concentrations upon H 2 S generation. Based on the flask experiment results, the column experiment investigated a longer-term inhibition effect of these chemicals at specific concentrations using simulated landfill columns. 2.2 Mechanisms for Inhibition of H 2 S Generation Under anaerobic conditions such as occur in a landfill, sulfate-reducing bacteria (SRB) utilize leached sulfate from gypsum drywall as an electron acceptor to generate H 2 S (Equation 2-1) (Postgate 1984; Hao et al. 1996): 3222422HCOSHOCHSOSRB (2-1) Since it is a biological process, H 2 S generation can be prevented by inhibiting the growth of sulfate-reducing bacteria. SRB are anaerobic bacteria and widespread in anaerobic environments and have been observed to occur in soil, water, sewage, oil and natural gas wells. Postgate (1984) reported that in order to cultivate SRB, the redox potential of the environment must start around -100 mV and the best and cheapest inhibitor of SRB is oxygen. In the secondary production of petroleum, it is common to inject aerated water into reservoirs for inhibiting SRB growth (Montgomery 1990). But the injection of air or aerated water is not practical in C&D debris landfills.

PAGE 26

13 SRB prefer an environment around a pH of 7 and are usually inhibited at pH values lower than 5.5 or higher than 9 (Hao et al. 1996). Changing the pH value by adding acid or alkali was reported as a method of inhibiting SRB in industrial plants (Zobell 1958; Postgate 1984). H 2 S generation was reported to be reduced when gypsum drywall was co-disposed with alkaline concrete waste (Townsend et al. 2000; Yang 2000). In terms of temperature, the optimal temperature range for SRB growth is between 30 0 C and 37 0 C (Widdel 1986) and most species of SRB die rapidly at temperatures above 45 0 C (Postgate 1984; Hao et al. 1996). In addition to providing unfavorable environments, the activity of SRB can also be suppressed by some chemicals. Chemical inhibitors of SRB were comprehensively reviewed and summarized by Saleh et al. (1964), including antibiotics, detergents, dyes, mercurials, metal ions, complexes, nitrocompounds, phenolic substances, sulfate analogues, sulphoamides, and other miscellaneous substrates. The most commonly used chemical inhibitor of SRB is molybdate (MoO 4 2) in the form of sodium molybdate (Na 2 MoO 4 ). It has been extensively used as a selective inhibitor in lake, marine sediments, and anaerobic digesters (Parkes et al. 1989; Clancy et al. 1992; Teer et al. 1997; Ranade et al. 1999). Biological sulfate reduction by SRB includes three major steps: sulfate activation (Equation 2-2), sulfate reduction to sulfite (Equation 2-3) and sulfide formation (Equation 2-4) (Barton and Plunkett 2002): PPAPSATPSOesulfurylasATP 24 (2-2) HAMPHSOHAPSSO3224 (2-3) OHHSHHSO22333 (2-4)

PAGE 27

14 Since sulfate ion is very stable, it cannot be reduced without first being activated by ATP (Equation 2-2). This reaction is catalyzed by the enzyme ATP sulfurylase which catalyzes the attachment of the sulfate ion to a phosphate of ATP. As a stereo chemical analog of sulfate (SO 4 2), molybdate (MoO 4 2) can inhibit the ATP sulfurylase, resulting in the interruption of sulfate reduction (Taylor and Oremland 1978). Another commonly used inhibition method is stimulating growth of a competing group of anaerobic bacteria with the addition of chemicals such as ferric iron and nitrate. The addition of ferric iron can stimulate the growth of iron reducing bacteria (IRB) which can outcompete SRB by maintaining concentrations of substrates at levels lower than thresholds required by SRB (Lovley 1991). Lovley and Phillips (1986) found that the addition of ferric iron to sediment inhibited sulfate reduction by 86 to 100%. In addition, ferric iron is an effective odor control chemical and has been widely used in wastewater treatment plants to reduce H 2 S odor (Hobson and Yang 2000). It acts as a precipitant that oxidizes and precipitates dissolved sulfides in wastewater to produce an insoluble iron sulfide by the reaction in Equation 2-5: HClFeSSSHFeCl623223 (2-5) Nitrate has also been used to prevent H 2 S generation in oil industries (McInerney et al. 1992). The addition of nitrate can not only increase redox potential (Poduska and Anderson1981), but it also stimulates the growth of nitrate-reducing bacteria (NRB) which have a large thermodynamic advantage over the SRB for the same substrates (Equation 2-6 and 2-7) (Eckford and Fedorak 2002): -495kJ/molG 4410385223323OHNHCOHNOCOCH (2-6)

PAGE 28

15 -47kJ/molG 232423HSHCOSOCOCH (2-7) The potential environmental problem for nitrate addition is groundwater contamination. Biocides (e.g., glutaraldehyde) have been used to prevent H 2 S generation as well, but it is expensive and not very effective because of growth of SRB in protected niches (Myhr et al. 2002; Eckford and Fedorak 2002). 2.3 Materials and Methods 2.3.1 Chemical Inhibitor Selection To evaluate inhibition potential of different chemicals, H 2 S was produced by placing crushed gypsum drywall into simulated landfill conditions. The gypsum drywall was purchased from local hardware stores. For the flask experiment, the drywall was ground into fine powder to improve the surface area using Urschell mill grinder (Fritsch, Germany). For the column experiment, the drywall was cut into 5 cm by 5 cm drywall blocks. Three chemical inhibitors were evaluated for their inhibition effect on H 2 S generation from gypsum drywall. They were sodium molybdate (Na 2 MoO 4 ), ferric chloride (FeCl 3 ), and hydrated lime (Ca(OH) 2 ). For each inhibitor, four different concentration solutions were tested using laboratory experiments conducted in 1-L flasks (Table 2-1). The chemical solutions were prepared by dissolving appropriate amounts of the solid chemicals into 1-L DI water. Based on previous research (Smith and Klug 1981), the concentration of sodium molybdate solutions were chosen from 2 mM to 20 mM to inhibit SRB. A high range of ferric chloride concentrations, from 5 mM to 500 mM, was tested in this study due to the lack of references. Based on the flask test

PAGE 29

16 results, one concentration of each chemical was tested using laboratory-scale simulated landfill columns. 2.3.2 Flask Experiment The flask experiment was designed to quickly assess an appropriate inhibition concentration of each chemical over 30-day period. As shown in Figure 2-1, the experimental apparatus consisted of a 1-L glass flask (Cole-Parmer Inc.), a #9 rubber stopper (FisherSci. Inc.), two different lengths of glass tube, and two plastic valves (model 30600-01, Cole-Parmer). The glass tubes penetrated through the rubber stopper into the flask at different depths. Two plastic valves were connected to the tubes for flushing nitrogen and taking samples, respectively. One hundred grams of the drywall powder was put into each flask and 100 ml inhibitor solution was sprayed on the drywall powder to a moisture content 50%. The rubber stopper was tightly capped and pure nitrogen gas was flushed into the flask for at least 5 minutes to remove air. The valves were then immediately closed to keep the flask under anaerobic conditions. The flask was placed into an incubator (FisherSci. Inc.) with a constant temperature of 35 0 C. Gas samples were collected using a glass syringe from the gas sampling port and transferred into Tedlar bags for gas dilution and H 2 S concentration analysis. 2.3.3 Inhibition Column Experiment Column experiments were used to evaluate the longer-term inhibition effect of each chemical in the specific concentration which was obtained from the flask experiment. Similar methodology (laboratory columns) has been previously used to evaluate the effect of co-disposed C&D debris on H 2 S generation and to evaluate H 2 S removal by cover materials in simulated landfill conditions (Yang 2000; Plaza 2003). In the present study, eight columns were constructed from PVC pipe; all of the columns had a height of

PAGE 30

17 50 cm and a diameter of 10 cm. As shown in Figure 2-2, a 10-cm slip cap was fitted on the end of each column and a MPT hose labcock valve (Asahi-America) was installed in the slip cap for leachate drainage. A 10-cm deep layer of gravel was loaded into each column to serve as a leachate reservoir. A geotextile was then placed over the gravel layer to separate it from the waste layer. The drywall blocks were pretreated by immersing them in the inhibitor solutions over night. Approximately 750 g of pretreated gypsum drywall blocks were loaded into each column, forming a 30 cm waste layer with a bulk density of 318 kg m -3 . A 10-cm slip cap was fitted on the top of each column with a final headspace height of 10 cm. To collect gas samples from the headspace, another MPT hose labcock valve (Asahi-America) was installed on the top slip cap of each column. After adding 400 ml of DI water to the top valve to each column, pure nitrogen gas was flushed for 10 minutes from the bottom to remove air. Finally, all of the valves were closed to maintain the columns under anaerobic conditions. The columns were monitored for 6 months. To avoid changing the inhibitor concentration, no water was added during the experiment. 2.3.4 Sample Collection and Analysis In the flask experiment, gas samples were collected three times per week over the 30-day period. In the column experiment, gas samples were collected from the columns twice per week during the 6 months. Gas samples were analyzed for H 2 S concentration using a Jerome 631-X H 2 S analyzer (Arizona Instruments, AZ). The Jerome meter’s internal pump pulls gas samples over a gold film sensor whose electrical resistance is proportional to the concentration of H 2 S in the gas samples (Arizona Instrument 2003). The Jerome meter has a detection range of 0.003 ppm to 50 ppm and is factory calibrated annually using methods traceable to NIST (National Institute of Standards and

PAGE 31

18 Technology). Before analyzing the gas samples, the Jerome meter was checked using a 25 ppm standard H 2 S gas (Air Liquid America Corp) and laboratory air. In the column experiment, leachate samples were collected monthly by draining the leachate by gravity from the bottom of the columns. Six leachate samples were collected over the experimental period. The leachate samples were analyzed for pH, sulfate, molybdenum, iron, and calcium. The methods used for pH were equivalent to Standard Method 4500-H. Sulfate was analyzed using a Dionex DX 500 ion chromatography system according to EPA SW-846 Method 9056. Metals analysis was carried out using the EPA method 6010B by inductively coupled plasma atomic emission spectroscopy (ICP-AES). Blanks, replicates, and calibration check samples were performed as appropriate. 2.4 Results and Discussion In this section, H 2 S generation from gypsum drywall is discussed first, followed by the inhibition of H 2 S generation by each chemical inhibitor. The environmental and industrial implications of these inhibitors are then discussed. 2.4.1 H 2 S Generation from Gypsum Drywall Figure 2-3 represents the change of average H 2 S concentration over a 30-day period from the flask and column experiments. The generation of H 2 S in the flask experiment was more rapid than in the column experiment. In the flask experiment, the concentration of H 2 S increased from 1.1 ppm at day 1 to 11,550 ppm at day 8 and continued to increase throughout the experimental period, reaching a maximum concentration 67,500 ppm at day 28. In the column experiment, gas sampling was started from day 6 with a concentration of 0.5 ppm; by day 30, the H 2 S concentration reached 11,300 ppm. The quick generation of H 2 S in the flask experiment resulted from the warm temperature

PAGE 32

19 (35 0 C) and fine gypsum drywall particle size, which stimulated SRB to rapid grow in a short period. It has been reported that biological sulfate reduction rates strongly depend on temperature (Hao et al. 1996). Both experiments illustrated that gypsum drywall alone can generate large concentration of H 2 S. Although biological sulfate reduction requires an organic carbon source, the paper facing and backing from the gypsum drywall (approximately 10% the mass of the drywall) provided enough organic carbon to produce large H 2 S concentrations (Townsend et al. 2000; Plaza 2003). Since H 2 S concentrations in the range of 500 ppm to 1,000 ppm can be lethal to humans (Flynn 1998), these concentrations are certainly of environmental significance. C&D debris landfills provide such a good environment for H 2 S generation, and high concentrations of H 2 S (12,000 to 80,000 ppm) were reported in previous research (Lee 2000; Flynn 1998), which indicates that proper personal protection should be taken for those who involved in excavation activities and operation of gas collection systems at C&D debris landfills. 2.4.2 Effect of Sodium Molybdate on H 2 S Generation The inhibition results of sodium molybdate in the flask and column experiments are presented in Figure 2-4 A and B, respectively. Compared to the generation of H 2 S from untreated gypsum drywall, H 2 S generation from the treated drywall by sodium molybdate was low and the concentration remained at approximately 10 ppm over the experimental period, approximately three orders of magnitude lower than that in the control flask. The highest concentration was 26.3 ppm detected in the 2 mM Na 2 MoO 4 flask at day 30. The results illustrated that H 2 S generation from gypsum drywall was effectively inhibited by sodium molybdate solutions. The average H 2 S concentration in the 2 mM, 4 mM, 10 mM, and 20 mM Na 2 MoO 4 flask were 8.6 ppm, 5,5 ppm, 2.9 ppm, and 4.8 ppm, respectively

PAGE 33

20 (Table 2-2). Based on the results, 10 mM sodium molybdate was selected for the following inhibition column experiment. In the column experiment, gypsum drywall was treated by immersing them into the 10 mM sodium molybdate solution over night. The generation of H 2 S from the treated gypsum drywall was effectively inhibited, resulting in an average H 2 S concentration of only 0.27 ppm (Figure 2-4 B). In the control columns, the H 2 S concentration was approximately 20,000 ppm after the first month, four to five orders of magnitude higher than that in the inhibition columns. The average leachate pH values and sulfate concentration from the control and the inhibition columns were similar. In the control columns, the sulfate concentration was 1,300 mg/L with average pH of 6.7, and the sulfate concentration in the inhibition column was 1,260 mg/L with average leachate pH 6.6. However, the average leachate molybdenum concentration was 23.6 mg/L in the inhibition columns, while only 0.07 mg/L in the control columns. The results indicated that the inhibition of H 2 S generation from gypsum drywall resulted from the inhibitor, sodium molybdate, not from the leachate pH. 2.4.3 Effect of Ferric Chloride on H 2 S Generation The inhibition results of ferric chloride in the flask and column experiments are presented in Figure 2-5 A and B, respectively. Although all ferric chloride solutions inhibited the generation of H 2 S, the inhibition effects on H 2 S generation were different. At day 14, for example, H 2 S concentration was 2,900 ppm in the 5 mM FeCl 3 flask and 110 ppm at the 10 mM FeCl 3 flask, while the concentrations were only 6.65 ppm and 0.21 ppm in the 100 mM and 500 mM FeCl 3 flask, respectively. In the flasks with low concentrations of ferric chloride (5 and 10 mM), H 2 S concentrations initially lagged behind those in the control flask, but eventually reached similar concentration levels at

PAGE 34

21 the end of the experiment. However, with the addition of high ferric chloride (100 and 500 mM), the generation of H 2 S was effectively inhibited throughout the 30-day period, and H 2 S concentrations in these flasks fluctuated near 1 ppm. Since higher ferric chloride concentration had better inhibition effect, 500 mM ferric chloride was selected for the inhibition column experiment. By treating gypsum drywall with the 500 mM ferric chloride solution, the generation of H 2 S from the treated gypsum drywall was almost totally inhibited. The average H 2 S concentration was only 0.08 ppm in the inhibition columns, compared to approximately 20,000 ppm in the control columns. The leachate average sulfate concentration in the FeCl 3 columns was 1,500 mg/L, slightly higher than that in the control columns. However, the leachate pH was much lower (pH=1.9) than that from the control columns. Due to the acidic characteristic of ferric chloride, the acidic leachate pH resulted from the adsorption of ferric chloride solution by the gypsum drywall blocks. Although ferric iron can affect biological sulfate reduction by stimulating growth of iron reducing bacteria (IRB) which can outcompete SRB by maintaining concentrations of substrates at levels lower than thresholds required by SRB (Lovley 1991), the inhibition of H 2 S generation by the high concentration of ferric chloride may not be attributed to the same mechanism because the pH range for IRB growth was reported from 4.8 to 8.2 (optimum 6.3-6.5) (Wiegel et al., 2003). The pH range for SRB was reported from 5.5 to 9.0 and the SRB and IRB live in similar environments. Therefore, the activity of SRB and IRB were possible both suppressed by the acidic leachate from the gypsum drywall which was treated by the high concentration of ferric chloride. The

PAGE 35

22 low H 2 S generation from the columns was probably due to the inhibition of SRB by the acidic leachate, not to the competition with IRB. However, the competition of IRB might play a role in inhibiting H 2 S generation when low concentrations of ferric chloride was used in the flask experiment. Although all four concentrations of ferric chloride solutions used in this study were acidic, after mixing with the ground gypsum drywall, the pH of the drywall-ferric chloride mixtures increased (Table 2-3). For example, the pH of 5 mM ferric chloride was 2.58, while the pH of the drywall-ferric chloride (5 mM) mixtures was 6.91. Under this pH conditions, both IRB and SRB can growth and compete for the same organic carbon. Since the organic carbon source is limited (only 10% backing paper in gypsum drywall), IRB can outcompete SRB, resulting in H 2 S generation was initially inhibited (Figure 2-5 A). After all ferric iron from the ferric chloride was reduced to ferrous iron by IRB, the activity of SRB may become dominant and more H 2 S are produced, increasing H 2 S concentration in the flask (Figure 2-5 A). Based on these results, although low concentration ferric chloride may inhibit H 2 S generation by stimulating IRB growth, H 2 S generation may resume after ferric iron is consumed by IRB. By treating gypsum drywall with high concentrations of ferric chloride, H 2 S generation can be effectively inhibited in a long-period. But the high concentration of ferric chloride may lead to some potential environment problems due to the large amount of chemical needed and acidic leachate. 2.4.4 Effect of Hydrate Lime on H 2 S Generation The inhibition results of hydrated lime in the flask and column experiments are presented in Figure 2-6 A and B, respectively. It was noticed that the inhibition of H 2 S generation by low concentrations of hydrated lime (1 mM and 14 mM) was very limited,

PAGE 36

23 and H 2 S concentrations inside these flasks were slightly lower than that in the control flasks. At day 16, the average H 2 S concentration was 19,750, 18,125, and 16,875 ppm in the control, the 1 mM and 14 mM Ca(OH) 2 flasks, respectively. However, higher concentrations of hydrated lime (40 mM and 68 mM) did inhibit H 2 S generation effectively. At day 16, the concentration of H 2 S was 44 ppm and 1.1 ppm in the 40 mM and 68 mM Ca(OH) 2 flasks, respectively. It was observed that H 2 S concentration started to increase in the 40 mM Ca(OH) 2 flask, from 48 ppm at day 21 to 420 ppm at day 30, which indicated the inhibition effect became weak at the end of the experiment. Based on the results, 68 mM hydrated lime solution (0.5% of hydrated lime) was selected for the inhibition column experiment. As shown in Figure 2-6 B, the generation of H 2 S from the gypsum drywall treated by hydrated lime was obviously different, which resulted in H 2 S concentration changing over the experimental period. The average concentration of H 2 S was 0.2 ppm during the first 40 days, but increased to 5,277 ppm during the period from day 61 to day 170. It means that H 2 S generation was initially inhibited in the first month, but for some reasons, the inhibition became limited, resulting in resumed H 2 S generation. The average sulfate concentration of leachate from the control and the hydrated lime columns was similar, and was 1,300 and 1,250 mg/L, respectively. But the leachate pH in the hydrated lime columns decreased throughout the experimental period. Figure 2-7 presents the change of leachate pH value and H 2 S concentration in the column experiment. With the decreasing of leachate pH from 12.05 at day 6 to 6.91 at day 171, H 2 S concentration increased from 0.071 to 9,350 ppm. The results indicated that the increasing of H 2 S generation might result from the change of leachate pH. The initial

PAGE 37

24 high pH leachate probably suppressed the activity of SRB, resulting in low H 2 S concentration in the first month. However, the generated H 2 S tends to dissolve into the leachate and release H + (Equation 2-7) HSHHSSHSHpKpKliqg2292.1299.6)(2)(221 (2-7) In the initial alkaline leachate, the major dissolved sulfides are HS and S 2and more H + can be released. The released H + chemically neutralizes OH in the leachate, resulting in leachate pH decrease. With leachate pH going down, the activity of SRB resumed and more H 2 S was generated, which further lowered the leachate pH and increased H 2 S generation (Figure 2-7). At the end of the experiment, the leachate pH became neutral and the H 2 S concentration in the inhibition correspondingly increased to over 10,000 ppm. The results showed that hydrate lime could be used to inhibit H 2 S generation from gypsum drywall, but the inhibition effect would decrease over a long-term period. A similar phenomenon was observed in previous research. When drywall was co-disposed with wood, the H 2 S generation was initially inhibited because the organic acid (i.e., fulvic and humic) in wood lowered the leachate out of the ideal pH range for SRB, but the H 2 S generation resumed after the organic acid was diluted (Yang 2000). 2.4.5 Environmental and Industrial Implication Based on the experiment results, all three chemical inhibitors, to some extent, can inhibit H 2 S generation from gypsum drywall. This provides a potential method to address the odor problems associated with H 2 S generation in C&D debris landfills. However, in addition to considering the duration of the inhibition effect, other issues, such as cost

PAGE 38

25 effectiveness and environmental impacts, should also be considered when applying the chemical inhibitors onto landfills. The column experiment showed that the inhibition effect of hydrated lime could ultimately be neutralized by the generated H 2 S, but because of its low cost and availability, hydrated lime could be used to treat gypsum drywall waste at C&D debris landfills. It has been recommended that prior to disposing gypsum drywall in landfills, powdered lime could be added to mix with gypsum drywall to reduce H 2 S generation (Johnson 1986). Another possible application of lime at a landfill would be spraying appropriate lime solution on gypsum drywall, making lime well attached on gypsum drywall before being buried. In terms of ferric chloride, the results showed that high concentrations of ferric chloride can effectively prevent H 2 S generation from gypsum drywall by providing extremely acidic conditions for SRB. However, the application of high concentrations of ferric chloride solutions may not be practical in the landfill operation because of its extremely acidic, corrosive characteristic and the large amount needed (135,000 mg FeCl 3 /kg drywall in this study). Sodium molybdate, however, exhibited an effective inhibition on H 2 S generation at a relatively low concentration (10 mM) in this study. Approximately 336 mg molybdenum (Mo) was adsorbed by 750 g gypsum drywall in the column experiment, which was equal to 448 mg-Mo/kg-drywall. It was just slightly higher than the Soil Cleanup Target Levels in residential area standard, 440 mg Mo/kg waste, but was much lower than the standard in industrial areas, 11,000 mg Mo/kg waste. Since the landfills should be considered as industrial areas, it may be possible to treat gypsum drywall by

PAGE 39

26 sodium molybdate at C&D debris landfills to reduce the generation of H 2 S. As for the leachate quality, the average molybdenum (Mo) concentration was 23.6 mg/L, and was higher than the groundwater standard of 35 g/L. However, in the column experiment, the leachate was directly drained out from the treated drywall blocks without passing through any buffer layers, such as sandy soil layers. Molybdenum was reported to be strongly adsorbed in soils by Fe and Al oxides (Brinton 2000; Bradl 2005). Considering the adsorption by soil layer and water dilution by groundwater, the molybdenum concentration may be much lower than the leachate concentration observed in this column experiment. Therefore, it may be possible to use sodium molybdate for inhibiting the H 2 S generation from gypsum drywall. There are two possible ways to apply sodium molybdate to inhibit H 2 S generation. At C&D debris landfills, waste gypsum drywall can be treated by spraying them with appropriate sodium molybdate solution before being buried. Another potential way is to directly add sodium molybdate as an additive with other ingredients, such as starch, sugar, and cellulose, during the drywall manufacturing process. In this way, sodium molybdate can be well mixed with sulfate, greatly reducing the possibility of H 2 S generation from gypsum drywall. But the content of sodium molybdate in the drywall must be well controlled to avoid impacting the quality of drywall and causing health problems. Although molybdenum is an essential trace element for human, excessive intake of molybdenum poses adverse impacts on human health, causing a physiological copper deficiency and increasing blood uric acid concentrations (Vyskocil and Viau 1999). Molybdenum is not classifiable as to carcinogenicity in humans by EPA, but a high

PAGE 40

27 incidence of weakness, fatigue, headache, and gout-like symptoms, such as joint and muscle pain, deformities and edema, have been reported among people exposed to molybdenum mine and processing plant (Opresko 1993). Therefore, further study should be conducted to investigate the potential health impacts of molybdenum before industrial application of sodium molybdate for inhibiting H 2 S generation. 2.5 Summary In this study, three chemical inhibitors were used to inhibit H 2 S generation from gypsum drywall. In the flask experiments, different concentrations of each inhibitor were evaluated to choose an appropriate concentration for the column experiment, while the column experiment was conducted to evaluate the long-term effect of each inhibitor on the H 2 S generation. Results show that the H 2 S generation was reduced by these chemical inhibitors. Sodium molybdate reduced H 2 S generation by inhibiting the first enzyme of sulfate reduction; ferric chloride and hydrated lime, however, affect H 2 S generation by providing acidic or alkaline conditions for SRB. Although high concentration of ferric chloride can effectively inhibit H 2 S generation, it may not be applicable in a real landfill because of its extreme acidic characteristic and large amount chemical needed. As regards lime addition, it exhibited a limited long-term inhibition effect on H 2 S generation because the initial alkaline conditions were gradually neutralized by the generated H 2 S. Sodium molybdate effectively inhibited H 2 S generation over a 6-month period. As a potential engineering method to solve the H 2 S problems in C&D debris landfills, a cost-effective analysis should be conducted and further research should be performed to test lower concentrations of sodium molybdenum on H 2 S generation and to investigate the potential environmental and health impacts.

PAGE 41

28 Table 21 . Concentrations of chemical inhibitors in the flask experiment Amount of chemicals Chemical inhibitors concentration (mM) mg (chemical) /kg drywall Sodium molybdate 2; 4; 10; 20 484; 968; 2,420; 4,840 Ferric chloride 5; 10; 100; 500 1,350; 2,700; 27,000; 135,000 Hydrated lime 1; 14; 40; 68 100; 1,000; 3,000; 5,000 Table 2-2. H 2 S concentrations in the flasks with addition of sodium molybdate Sodium Molybdate Concentrations H 2 S concentration (ppm) 0 mM 2 mM 4 mM 10 mM 20 mM Average 27,304 21,365 8.6 7.2 5.5 3.7 2.9 1.6 4.8 2.7 Minimum 0.6 0.9 0.7 0.5 0.4 Maximum 67,500 26.3 11.1 4.3 10.4 Table 2-3. pH change of ferric chloride solutions after mixing with gypsum drywall FeCl 3 concentration 5 mM 10 mM 100 mM 500 mM pH of FeCl 3 solution 2.58 2.34 1.62 0.84 pH of the mixture of FeCl 3 and drywall 6.91 6.84 5.53 1.77

PAGE 42

29 N2Gas sampling C&Ddebris N2Gas sampling C&Ddebris Gypsum dr y wall Figure 2-1. Schematic of flask experiment apparatus 10 cm Gypsum drywall 30 cm10 cm 10 cm Headspace Gravel Leachate sampling port Geotextile Geotextile Gas sampling port 10 cm Gypsum drywall 30 cm10 cm 10 cm Headspace Gravel Leachate sampling port Geotextile Geotextile Gas sampling port Figure 2-2. Schematic of laboratory inhibition column experiment

PAGE 43

30 Day 051015202530 H2S concentration (ppm) 100101102103104105 Flask experiment Column experiment Figure 2-3. Comparison of the average H 2 S concentration from gypsum drywall in the flask and column experiments

PAGE 44

31 Day 0102030 H2S concentration (ppm) 10-1100101102103104105106 Control 2 mM Na2MoO4 4 mM Na2MoO4 10 mMNa2MoO4 20 mMNa2MoO4 A. Results of the flask experiment Day 020406080100120140160180 H2S concentration (ppm) 10-210-1100101102103104105 Control column #1 Control column #2 10 mM Na2MoO4 column #1 10 mM Na2MoO4 column #2 B. Results of the column experiment Figure 2-4. The inhibition effect of Na 2 MoO 4 on H 2 S generation. A) The flask experiment results. B) The column experiment results.

PAGE 45

32 Day 0102030 H2S concentration (ppm) 10-210-1100101102103104105 Control 5 mM FeCl3 10 mM FeCl33 100 mM FeCl33 500 mM FeCl33 A. Results of the flask experiment Day 020406080100120140160180 H2S concentration (ppm) 10-210-1100101102103104105 B. Results of the column experimentControl column #1 Control column #2 500 mM FeCl3 column #1 500 mM FeCl3 column #2 Figure 2-5. The inhibition effect of FeCl 3 on H 2 S generation. A) The flask experiment results. B) The column experiment results.

PAGE 46

33 Day 0102030 H2S concentration (ppm) 10-1100101102103104105 A. Results of the flask experimentControl 1 mM Ca(OH)2 14 mM Ca(OH)2 40 mM Ca(OH)2 68 mM Ca(OH)2 Day 020406080100120140160180 H2S concentration (ppm) 10-210-1100101102103104105 B. Results of the column experimentControl column #1 Control column #2 68 mM Ca(OH)2 column #1 68 mM Ca(OH)2 column #2 Figure 2-6. The inhibition effect of Ca(OH) 2 on H 2 S generation. A) The flask experiment results. B) The column experiment results.

PAGE 47

34 Day 020406080100120140160180200 H2S concentration (ppm) 020004000600080001000012000 pH 57911134681012 H2S concentration pH of leachate Figure 2-7. Change of average H 2 S concentrations and average leachate pH in the lime columns

PAGE 48

CHAPTER 3 ATTENUATION OF H 2 S AT CONSTRUCTION AND DEMOLITION DEBRIS LANDFILLS USING ALTERNATIVE COVER MATERIALS 3.1 Introduction Hydrogen sulfide (H 2 S) is a toxic, corrosive air pollutant with a distinctive “rotten egg” odor at low concentrations (ATSDR 2004). H 2 S emissions to the environment results from many industrial processes, such as wastewater treatment, kraft paper manufacturing, oil refining, and petroleum coke manufacturing. Another recently recognized source of anthropogenic H 2 S is construction and demolition (C&D) debris landfills, or other landfills that dispose of C&D debris. It is known that H 2 S can be generated as a result of biological conversion of gypsum drywall, one of the major components of C&D debris (Gypsum Association 1992a and 1992b; Flynn 1996 and 1998). The emissions of H 2 S from such facilities have become increasing environmental and health concerns in many parts of North America (Johnson 1986; Musick 1992; Flynn 1998; Townsend et al. 2000 and 2005). Many different methods have been developed to remove H 2 S from industrial emissions; these include adsorption on activated carbon, scrubbing with water or chemical solutions, reaction with heavy metals, oxidation by ozone, incineration, and biofiltration (Chung et al. 1998; Burgess et al. 2001). Application of these techniques to C&D debris landfills is difficult, however, as C&D debris landfills do not typically capture and collect landfill gases. Recent research suggests that effective use of cover materials at C&D debris landfills may provide a low-cost, effective technique for 35

PAGE 49

36 controlling the H 2 S emissions from landfills (Townsend et al. 2005;Yang 2000). For example, Plaza (2003) used laboratory columns with different cover soil and reported that the efficiency of H 2 S emission reduction (when compared to control columns with no cover) was as high as 99% using soil amended with 5% hydrated lime, but only 30% using unamended sandy soil. Although some laboratory experiments have been conducted, a large-scale evaluation of these materials has not been performed in a real C&D debris landfill environment. In addition, other materials naturally available at most C&D debris landfills, such as concrete and compost (which has been used to treat H 2 S in wastewater treatment plant; Nicolai and Janni, 2001; Yang 1992), have not been investigated for removing H 2 S emissions from C&D debris landfills. This paper presents research conducted to evaluate the attenuation of H 2 S by various alternative cover materials in the field. The alternative materials used in this study were sandy soil, compost, fine concrete, and various lime-amended sandy soils. The results provide insight into H 2 S removal by cover soils. The utilization of these cover materials to attenuate H 2 S emissions at C&D debris landfills might prove a useful alternative to more expensive gas control techniques such as landfill gas collection system. 3.2 Materials and Methods A field study and along with a complementary laboratory experiment were conducted to evaluate the attenuation of H 2 S by various cover materials. Field work focused on H 2 S concentrations and emissions from a field testing area using different materials as cover soils, while laboratory experiments were performed to support field observations and to investigate H 2 S removal by these cover materials on a laboratory-scale.

PAGE 50

37 3.2.1 Field Study 3.2.1.1 Cover materials and plot construction The impact of alternative cover soil materials on H 2 S emissions was studied at a landfill that accepted predominantly C&D debris in central Florida. This site was included in a previous study that investigated odor problems associated with H 2 S emissions at C&D debris landfills (Reinhart et al. 2005). Based on a survey conducted in 2004, ambient H 2 S concentrations at this landfill ranged from 3 ppb to 350 ppb. An area on top of an inactive portion of the landfill (not receiving waste at time of the experiment) was selected as the location for different cover materials. A 12 m by 18 m area was excavated to a depth of 0.6 m. The testing area was divided into six 6 m by 6 m testing plots; the cover materials investigated were sandy soil, compost, fine concrete, sandy soils amended with various amounts hydrated lime (Ca(OH) 2 ) (1% and 3%), and sandy soil amended with 10% agricultural lime (CaCO 3 ). Sandy soil and compost were available for use at the landfill and fine concrete was collected from a local concrete recycling facility. Since hydrated lime (5%) amended sandy soils have been reported to successfully removed H 2 S in laboratory simulated-landfill columns (Plaza, 2003), low percentages (1% and 3%) of hydrate lime amended sandy soils were used to investigate the H 2 S removal ability in this field study. Agricultural lime (CaCO 3 ) is commonly used to amend soil in agriculture, but the H 2 S removal ability of agricultural lime is not as effective as that of hydrate lime. Therefore, 10% agricultural lime amended sandy soils was used in this study. Both types of lime were purchased from a local agricultural supply operation. The amount of cover material used in each testing plot is listed in Table 3-1.

PAGE 51

38 Due to the heterogeneous character of C&D debris, H 2 S generation is variable (Lee, 2000). In order to avoid the variability of H 2 S generation in the testing area, approximately 10 m 3 of gypsum drywall was collected, crushed and placed into the bottom of the testing area to promote H 2 S generation. The crushed gypsum drywall formed a 5-cm deep layer at the bottom. Two layers of geocomposite (geonet with geotextile) were placed on the gypsum drywall layer to separate the gypsum drywall from the cover materials and to distribute the H 2 S gas. A gas distribution system was designed to reduce the pressure build-up underneath the cover soils. It was composed of gas distribution pipes and vents. The gas distribution pipes were placed between the two layers of geocomposite and connected to the vents which were open to the atmosphere. Before cover materials were loaded into testing plots, water was sprayed to wet the gypsum drywall for stimulating H 2 S production. Each cover material was then loaded into a separate 6 m by 6 m testing plots to form 0.3 m depth of landfill cover (Figure 3-1). A detailed description of the construction procedure is provided in Appendix B. 3.2.1.2 Field sampling The field study was conducted from March 2004 to January 2005. To compare H 2 S attenuation by the cover materials, the H 2 S emission rates and soil vapor H 2 S concentrations from the testing plots were measured. During the 10-month field study, the testing area was visited 24 times. In most of the visits (19 out of 24), both parameters were measured in all six testing plots. Other visits focused on the temporal change of the H 2 S emission rate. Therefore, only the emission rate was measured in the sandy soil testing plot during those visits. The H 2 S emission rates were measured using a 65-L acrylic resin flux chamber (Odotech Inc., Canada). The flux chamber method is the most widely used and proven

PAGE 52

39 technique for measuring gas emission rates (Borjesson and Svensson 1997; Liao and Chou 1998; Lin et al. 2002; Park and Shin 2001). As shown in Figure 3-2, the flux chamber consists of a cylindrical enclosure and a half-dome, with a diameter of 0.5 m and an overall height of 0.4 m. When the H 2 S emission rate was measured, the flux chamber was placed on the ground with the skirt buried approximately 6 cm deep and a bentonite slurry was applied to avoid gas leakage. A nitrogen flow was swept through the flux chamber using a perforated plastic tube configured as a loop along the interior circumference. The nitrogen flow was controlled by a flow meter (Cole-Parmer, IL) connected between the nitrogen tank and the flux chamber and the flow rate ranged from 5 L/min to 10 L/min. The emitted H 2 S from the covered area was mixed with the nitrogen in the chamber. The outlet mixed gases were analyzed for H 2 S concentration by a Jerome meter with a detection range from 0.003 ppm to 50 ppm. The H 2 S emission rate was calculated by the following equation: A CvF (3-1) where F is H 2 S emission rate (mg m -2 s -1 ); v is the flow rate of the sweeping nitrogen (m 3 min -1 ); A is the covered area by the flux chamber (m 2 ); C is the H 2 S concentration of outlet gas (ppm). Because the lowest recommended nitrogen flow rate for the flux chamber was 5L/min, the detection limit of H 2 S emission rate using the flux chamber method was 1.77 -6 mg m -2 s -1 . When H 2 S concentration was below detection limit of 0.003 ppm, half of the detection limit of 0.0015 ppm was used to calculate the emission rate (Berthoues and Brown 1994). To measure the soil vapor H 2 S concentrations, a series of different lengths of plastic probes were installed in all testing plots at depths of 2.5 cm, 7.5 cm, 15 cm,

PAGE 53

40 22.5 cm, 30 cm, and between two geocomposite layers (Figure 3-1 B). A 10-ml plastic syringe was used to extract soil vapor samples from these sampling probes. Soil vapor samples were slowly extracted by pulling a syringe plunger, with the content being transferred into a Tedlar bag for dilution and analysis for H 2 S concentration. H 2 S concentration was analyzed using a Jerome 631-X H 2 S Analyzer (Arizona Instruments, AZ) with a detection range from 0.003 ppm to 50 ppm. 3.2.2 Laboratory Experiments A laboratory experiment was designed to examine the H 2 S adsorption abilities of cover materials using serum bottles. The serum bottle made by Wheaton (Millville, NJ) had a 250-ml volume. The bottles were made of borosilicate glass with a diameter of 6 cm and a height of 158 cm. The mouths of the bottles had a 1.5-cm inner diameter and a 3 cm outer diameter. Ten grams of cover material were put into the bottle. It was then sealed with a rubber septa and aluminum crimp seal. Fourteen ml of 5000 ppm H 2 S gas was injected into the bottle using a 20-ml glass syringe to make the initial H 2 S concentration approximately 250 ppm. Then, a 10-ml glass syringe was used to take gas samples from the bottle at intervals. The gas samples were then transferred to a 500-ml Tedlar bag for dilution, which was analyzed for H 2 S concentration using the Jerome meter. Controls were also used to examine H 2 S change in the bottles without cover materials. 3.3 Results and Discussion 3.3.1 H 2 S Emissions from the Testing Plots During the 10-month field study, detectable H 2 S emissions were only encountered in the sandy soil testing plot. No H 2 S emissions were detected from other testing plots over the 19 measurements, indicating that the alternative cover materials did attenuate

PAGE 54

41 H 2 S emissions to a greater extent than the sandy soil (the normal cover used at this site). Figure 3-3 presents the monthly average emission rate of H 2 S at the sandy soil testing plot during the field study. The emission rate ranged from below detection limit to 1.24 -5 mg m -2 s -1 , and the average emission rate was 4.67 -6 mg m -2 s -1 . The emission rates of H 2 S changed during the 10-month period, with lower emission rates measured in the first three months and compared to higher emissions in the remaining months. In addition to changing in the H 2 S emission rate over time, changes within a day were also noted. The temporal variation of H 2 S emission rate is discussed in next chapter. The present chapter focuses on the attenuation of H 2 S by the cover materials. 3.3.2 Soil Vapor H 2 S Concentrations in the Testing Plots Soil vapor samples were extracted from different depths to assess the change of H 2 S concentrations in the cover soils and thus provide an assessment of the ability of the materials for H 2 S attenuation. Based on the soil vapor concentrations, H 2 S concentration profiles in the testing plots were plotted. Figure 3-4 presents typical concentration profiles based on the concentrations measured on September 10 th , 2004. The complete concentration profiles documented in this study are presented in Appendix C. In general, all concentration profiles were similar in that H 2 S concentration decreased from the bottom to the surface of cover soils. This indicated that H 2 S was gradually removed by the cover soils as it diffused through the cover soils. At the same depth, the H 2 S concentration in the sandy soil plot was always higher than that of other testing plots. For example, the concentration of H 2 S was as high as 530 ppm at 22.5-cm depth in sandy soil plot, while the concentration in the CaCO 3 -amended sandy soil was 26 ppm, and the concentration was 10.6 ppm in the 1% Ca(OH) 2 amended sandy soil. The lowest concentration of 0.55 ppm was in the compost plot (Figure 3-4). High soil vapor H 2 S

PAGE 55

42 concentration in the sandy soil resulted in concentration profiles distinct from other concentration profiles. Another observation illustrated in Figure 3-4 was the difference of the bottom H 2 S concentrations among the test plots. Sandy soil had the highest bottom concentration (3,300 ppm), while the bottom concentration in the fine concrete plot was only 150 ppm. Similar observations were found throughout the field study (though H 2 S concentrations in this layer did increase when the soil became saturated; discussed in Chapter 4). As mentioned before, a 5-cm gypsum drywall layer was placed at the bottom of the testing area to promote H 2 S generation and two layers of geocomposite placed over the drywall in an effort to evenly distribute the generated H 2 S. The intent was to make H 2 S concentrations uniform underneath the different cover soil plots. However, as shown in Figure 3-5, the average bottom H 2 S concentrations for the last six months in the testing area were different. The sandy soil testing plot had the highest average concentration (3,425 ppm). The lowest bottom concentration was detected in the fine concrete plot, only 327 ppm. According to the Fick’s law (Equation 3-2): zCDq (3-2) where q is the flux of H 2 S and C is the concentration of H 2 S; z refers to the depth of the cover soils. At the same cover soil depth, high bottom H 2 S concentrations results in high H 2 S emission rates. The high bottom H 2 S concentration in the sandy soil plot, to some extent, explains the high emission rate detected from that area. Based on the bottom concentration results shown in Figure 3-5, two hypotheses were formulated to explain the difference of the bottom concentrations. The first

PAGE 56

43 hypothesis was that the generated H 2 S was being extracted from the vapor phase by interactions (e.g., adsorption by the cover materials). The second hypothesis was that H 2 S generation underneath the cover soil layers was reduced by the alternative materials. For example, the lime amended soils could have leached high pH water (e.g., higher than 9) which could have suppressed the activity of the SRB. 3.3.3 H 2 S Removal by the Cover Materials To test the first hypothesis that H 2 S was removed by the cover materials, a laboratory H 2 S adsorption experiment was conducted to examine the adsorption of H 2 S by the cover materials used in the field study. Each cover material and the control were tested in duplicate. Figure 3-6 represents the change of average H 2 S concentration of each cover material over time. Fine concrete exhibited the fastest H 2 S adsorption rate, reducing approximately 90% H 2 S within 5 minutes. It took about 10 minutes for the sandy soils amended with hydrated lime to adsorb 90% H 2 S. Sandy soil had the lowest H 2 S adsorption compared to other alternative cover materials, reducing only 60% H 2 S in 60 minutes. However, the concentrations of H 2 S in the control bottle almost maintained the initial concentration during the experiment. The results showed that H 2 S was adsorbed by the cover materials with different adsorption rates. In the field study, the concentrations of H 2 S decreased from the bottom to top of the cover soils in all testing plots, indicating H 2 S was gradually removed by the cover materials (Figure 3-4). Due to different characteristics of the cover materials, the mechanisms of H 2 S removal are supposed to be different. It was observed in field that the color of bottom cover soil in the sandy soil plot was changed from yellowish to black. Similar black color changes were observed in other testing plots, except the compost plot (because of the dark color of compost). Right after

PAGE 57

44 adding hydrochloric acid (HCl), the black materials quickly changed its color back to yellow and lots of gas bubbles appeared with a strong “rotten egg” smell which was identified as H 2 S. It has been reported that H 2 S reacts with trace metals (MO x ) in the soils to form black sulfide compounds (MS x ), as shown in Equation 3-3 (Gumerman 1968; Gumerman and Carlson 1969): OxHMSSxHMOxx22 (3-3) Some metal sulfide minerals, such as FeS and MnS, were found in the black materials using a XRD analysis. With the addition of HCl, the sulfide minerals reacted with the acid to produce H 2 S, as shown in Equation 3-4: xxMClSxHxHClMS222 (3-4) Compared to the sandy soil, the fine concrete and lime amended sandy soils contain more metals, making more H 2 S could be removed by forming the sulfide minerals. Another observation was the acidification of the compost. The pH value of the compost dropped from 7.35 in March 2004 to 6.28 in January 2005. Instead of being removed by the trace metals, H 2 S was probably utilized by various microorganisms growing on compost as an electron donor. In the biological removal process, H 2 S was first converted into elemental sulfur (S 0 ) and then oxidized to sulfate with H + production, resulting in the acidification of the compost, as shown in Equation 3-5: HSOSSHOO240222 (3-5) In addition to chemical removal (Equation 3-3) and biological oxidation (Equation 3-5), H 2 S can be removed by dissolution into the soil water, as shown in Equation 3-6: HSHHSSHSHpKpKliqg2292.1299.6)(2)(221 (3-6)

PAGE 58

45 With a considerable water solubility (approximately 3,018 mg/L at 27 0 C; Chwirka 1990), a large amount of H 2 S can be removed when H 2 S migrates through the wet bottom portion of the cover soils. With the addition of lime into the soils, the OH can chemically neutralize the dissolved H + , leading to more dissolved H 2 S into the soil water of the alkaline cover materials. 3.3.4 Inhibition of Underlying H 2 S Production To test the second hypothesis, laboratory soil tests were performed to analyze the soil pH and particle size of all the cover materials. Soil temperatures at the bottom of the cover materials were also measured in July 2004. The results of these parameters are summarized in Table 3-2. Except the sandy soil and compost, the other cover soils were alkaline, with pH values higher than 9. Since the SRB prefer an environment with a pH around 7 and usually are inhibited at pH values lower than 5.5 or higher than 9 (Widdel 1988), the alkaline cover materials provided an adverse environment for SRB, which might inhibit the SRB activity and reduce H 2 S generation from the bottom gypsum drywall layer. In the compost plot, although the pH values were around neutral, the average temperature underneath the compost was higher than 50 0 C. Since most species of SRB were reported to die rapidly at temperatures above 45 0 C (Hao et al. 1996), the high soil temperature may have reduced the H 2 S generation in the compost plot. In addition, compost had much larger particle size than any other cover materials and could not be well compacted during the construction procedure. Due to the downward diffusion of air, the oxic zone in the compost plot may expand and expose the anaerobic SRB to aerobic conditions, inhibiting the activity of SRB. Compared to other testing plots, the sandy soil plot provided a better environment for SRB growth, as a result, the bottom H 2 S

PAGE 59

46 concentrations increased from 140 ppm in March to approximately 3,000 ppm in July 2004. In addition to the soil pH analysis, soil samples were also taken in duplicate from different layers of the sandy soil and fine concrete plots to analyze for total sulfur content. Figure 3-7 presents the average total sulfur content of the duplicate soil samples at different depths of the sandy soil and fine concrete plots. In both cover materials, the sulfur contents decreased from the bottom to the top layer (Figure 3-7), corresponding to the decreasing of H 2 S concentrations in the cover soils (Figure 3-4). As mentioned in previous serum bottom H 2 S adsorption experiment, fine concrete has stronger ability to adsorb H 2 S than sandy soil. In other words, under the same conditions, fine concrete can adsorb more H 2 S than sandy soil. However, the sulfur contents of the fine concrete samples were obviously lower than that of the sandy soils (Figure 3-7). For example, the average sulfur content of the bottom fine concrete was only 0.056%, while it was 0.17% in the sandy soils. The results indicated that the H 2 S generation was inhibited by the fine concrete, because even fine concrete has higher H 2 S adsorption ability, the sulfur content in the fine concrete was lower than sandy soil. Compared to the control samples (not exposed to H 2 S gas), the soil samples from the testing plots contained more sulfur because of the adsorption of H 2 S by the soils (Figure 3-7). The top cover soil, however, never changed in color to black, even though it was also exposed to H 2 S. This is because the black sulfide compounds are not chemically stable in the presence of oxygen and can be readily oxidized to elemental sulfur (S 0 ), as shown in Equation 3-7 (Bohn et al. 1989) 02222xSMOxOMSxx (3-7)

PAGE 60

47 It was observed that when the black sulfide minerals exposed to air, the black color gradually faded in a few hours, but under anaerobic conditions, the black sulfide minerals can remain the black color for a few months. 3.4 Summary Landfill cover is generally used to attenuate H 2 S emissions in most C&D debris landfills. Six different cover materials were tested as landfill cover to attenuate H 2 S emissions in a field study. The field results indicated that the alternative cover materials could attenuate H 2 S emissions more effectively than sandy soils and H 2 S emissions were only detected from the sandy soil plot. In this chapter, the attenuation of H 2 S by the cover materials was discussed, and the next chapter will focus on the emissions of H 2 S at C&D debris landfills. The attenuation of H 2 S by the alternative cover materials is attributed to H 2 S adsorption and inhibition of H 2 S generation. Laboratory experiments indicated that although H 2 S was removed by all cover materials, the alternative cover materials were more effective than the sandy soils. When H 2 S diffuses through landfill cover materials, it can be removed in various ways. These mechanisms include physical dissolution, chemical reaction, and biological oxidation. The generation of H 2 S was inhibited by the alternative cover materials by providing an adverse environment for SRB growth, by increasing soil pH or increasing the soil temperature.

PAGE 61

48 Table 3-1. Summary of cover materials used in the field study Testing Plot Cover Material 1 Sandy soil 22,000 kg 2 Fine concrete 25,000 kg 3 Compost 11.5 m 3 4 10% CaCO 3 200 kg CaCO 3 + 2000 kg Sandy soil 5 1% Ca(OH) 2 22 kg Ca(OH) 2 + 2200 kg Sandy soil 6 3% Ca(OH) 2 66 kg Ca(OH) 2 + 2200 kg Sandy soil Table 3-2. Results of pH, temperature, and particle size of cover materials used in the field study Testing plots pH Soil temperature ( 0 C) Percentage passed 2 mm sieve Sandy soil 6.98 0.18 26.2 1.1 94.8 % Fine concrete 11.22 0.27 36.1 1.9 79.6 % Compost 6.79 0.41 51.3 1.8 54.5 % 10% CaCO 3 9.71 1.23 25.5 0.9 96.6 % 1% Ca(OH) 2 11.68 0.76 26.5 1.0 97.9 % 3% Ca(OH) 2 12.59 0.48 26.3 1.3 96.0 %

PAGE 62

49 Buffer areaTubing sampling ports 6 m 3.6 m 6 m Testing areaVent 3.6 m 1.2 m 1.8 m Air pipe (a) Top view of a testing plot Buffer areaTubing sampling ports 6 m 3.6 m 6 m Testing areaVent 3.6 m 1.2 m 1.8 m Air pipe Buffer areaTubing sampling ports 6 m 3.6 m 6 m Testing areaVent 3.6 m 1.2 m 1.8 m Air pipe (a) Top view of a testing plot5 cm Gypsum drywall Cover soilTubing sampling ports 30 cm 22.5 cm 15 cm 7.5 cm 2.5 cm Geocomposite 0.3 m 3.6 m 1.2 m 1.2 m Bottom. Air pipe Vent (b) Side view of a testing plot5 cm Gypsum drywall Cover soilTubing sampling ports 30 cm 22.5 cm 15 cm 7.5 cm 2.5 cm Geocomposite 0.3 m 3.6 m 1.2 m 1.2 m Bottom. Air pipe Vent 5 cm Gypsum drywall Cover soilTubing sampling ports 30 cm 22.5 cm 15 cm 7.5 cm 2.5 cm Geocomposite 0.3 m 3.6 m 1.2 m 1.2 m Bottom. Air pipe Vent (b) Side view of a testing plot A. Top view of a testing plot B. Side view of a testing plot Figure 3-1. Schematic of the testing plot. A) Top view. B) Side view.

PAGE 63

50 Sampling portFlux chamber N2inN2tank Flow meter H2SLandfillN2+H2S Figure 3-2. Schematic of the H 2 S emission rate measurement system in the field study Month MarAprMayJunJulAugSepOctNovDecJanFeb Monthly average H2S emission rate (mg m-2 s-1) 02x10-64x10-66x10-68x10-610x10-612x10-614x10-6 Annual average 4.67x10-6mg m-2 s-1Detection limit 1.77x10-6mg m-2 s-1 Figure 3-3. Monthly change of H 2 S emission rate at the sandy soil testing plot

PAGE 64

51 H2S Concentraton (ppm) 0.0010.010.1110100100010000 Depth of Cover Soil (cm) 051015202530 Sandy soil Fine concrete Compost 10% CaCO3 1% Ca(OH)2 3% Ca(OH)2 Figure 3-4. HS concentration profiles of the testing plots on September 10 2004 2 th Bottom H2S concentration (ppm) 01000200030004000 Sandy SoilFine concreteCompost3% Ca(OH)2 1% Ca(OH)2 10 % CaCO3 Figure 3-5. Average H 2 S concentration underneath different cover materials from August 2004 to January 2005

PAGE 65

52 Minute 0102030405060 C/C0 0.10.30.50.70.90.00.20.40.60.81.0 Fine concrete 3% hydrated lime 1% hydrated lime Compost 10% agricultural lime Sandy soil Control Figure 3-6. Comparison of H 2 S removal by cover materials in laboratory serum bottle experiment Total sulfur in the cover soils (%) 0.000.050.100.150.200.25 Bottomlayer Sandy Soil Fine concrete MiddlelayerToplayerControl Figure 3-7. Average sulfur content of the cover soils at different depths in the field study

PAGE 66

CHAPTER 4 H 2 S EMISSION RATE AND TEMPORAL CHANGE AT CONSTRUCTION AND DEMOLITION DEBRIS LANDFILLS 4.1 Introduction The odor problems associated with H 2 S emissions have become a major environmental and health concern at many C&D debris landfills in North America (Johnson 1986; Flynn 1998; Townsend et al. 2000 and 2005). Compared to the H 2 S concentration of 0.05 ppb in the background air, ambient H 2 S concentrations at C&D debris landfills were reported ranging from lower than 3ppb to over 50 ppm (Lee 2000). The high concentrations of H 2 S are attributed to H 2 S emissions from C&D debris landfills, resulting in odor complaints from the residents living around these facilities (Crosson 1995; Xu and Townsend 2004). In order to address H 2 S problems, it is necessary to investigate H 2 S emission rate at C&D debris landfills. Although some research has been performed to measure landfill gas emissions from MSW landfills, few studies have been conducted to quantify the emissions of H 2 S at C&D debris landfills. From 2003, research was started to investigate H 2 S emissions from C&D debris landfills (Reinhart et al. 2004). H 2 S emission rates were surveyed from 100 locations at five different C&D debris landfills in Florida. Results showed that the emission rates of H 2 S changed form site to site, ranging from zero to 2.8 3 mg m -2 day -1 (3.24 -2 mg m -2 s -1 ). The variation of H 2 S emissions was attributed to the heterogeneous C&D debris and to different landfill operation among these landfills (Xu and Townsend 2004). As part of the previous research, the attenuation of H 2 S emissions by various cover materials was evaluated in a field study. It was found 53

PAGE 67

54 that even in the same spot, H 2 S emission rates were not constant, but varied with time, even through the course of the day (Chapter 3). Since H 2 S emissions from C&D debris landfills directly affect the H 2 S concentrations in the ambient air around these facilities, understanding the emissions of H 2 S from C&D debris landfills is important from both an environmental and regulatory perspective. Therefore, further research was conducted to investigate and to quantify H 2 S emission rates from a testing plot using sandy soil as cover soil at a C&D debris landfill in Florida. In addition, laboratory experiments were performed to explore the factors influencing H 2 S emissions such as soil moisture. The results should provide beneficial to better understand H 2 S emissions and to evaluate the H 2 S odor problems at C&D debris landfills. 4.2 Materials and Methods Both field study and laboratory experiments were performed in this research. In the field study, the emission rates of H 2 S were monitored in a 10-month period at the sandy soil testing plot. Laboratory experiments were conducted to verify field observations and to investigate the factors influencing H 2 S emissions. 4.2.1 Field Experiment A field testing area was constructed at a C&D debris landfill to measure the emissions of H 2 S from different cover materials. The cover materials included sandy soil, fine concrete, compost, and sandy soils amended by 10% CaCO 3 , or 1% Ca(OH) 2, or 3% Ca(OH) 2 . The construction procedure of this testing area is described in Chapter 3 and Appendix B. During the 10-month experimental period, H 2 S emissions were measured 19 times from all six testing plots, but detectable H 2 S emissions were only encountered in the sandy soil testing plot and the emissions changed with time. In order to investigate the

PAGE 68

55 temporal variation of H 2 S emissions, continuous monitoring was conducted by placing the flux chamber on the same site with continuous nitrogen gas flushing. Gas samples were taken from the outlet gas stream at intervals for H 2 S concentrations analysis. Three daytime continuous monitoring were performed on August 11 th , September 10 th , and October 1 st 2004. Another overnight continuous monitoring was conducted from October 20 th to October 21 st 2004 to investigate the temporal change of H 2 S emissions over a 28-hour period. 4.2.2 Laboratory Experiments Since the H 2 S emission rates underwent temporal change in the field study, laboratory experiments were conducted to investigate the possible factors that could affect H 2 S emissions. As shown in Figure 4-1, the test device was composed of a Buchner funnel with glass fritted disc (Trenton, FL), a Masterflex pump (Cole-Parmer, IL), a #12 rubber stopper, and an 80-L Tedlar bag. The Buchner funnel was constructed with heavy-walled stems of 6-cm diameter and a 5-cm long stem. The height above the glass disc was 9 cm. H 2 S gas was pumped into the funnel by the Masterflex pump to simulate H 2 S diffusion in landfills and 200-g of sandy soil was loaded to form a 5-cm depth of simulated landfill cover. A layer of geotextile was placed on the fritted discs to avoid sandy soil passage through the disc. The cover soils were open to the atmosphere and the H 2 S flow rate was controlled at 5 ml min -1 by the Masterflex pump. Sampling was performed by capping the Buchner funnel with a #12 rubber stopper and using a 10-ml glass syringe to take a gas sample from the stainless steel tube inserted through the stopper. After sampling, the stopper was removed and the cover soils were again open to the atmosphere.

PAGE 69

56 Based on the field observations, the effects of H 2 S concentration, soil moisture, and soil temperature on H 2 S emissions were investigated using the experimental apparatus, respectively. Concentrations of H 2 S gas were prepared by mixing pure H 2 S with nitrogen gas with various ratios in the 80-L Tedlar bags. DI water was sprayed on the sandy soil as simulated rainfall to make various soil moisture contents. To simulate the change of soil temperature, a heating tape was wrapped around the funnel to heat the cover soil and a thermometer was inserted into the soil to monitor soil temperature. Table 4-1 summarizes laboratory experiments conducted in this study. 4.3 Results and Discussion In this section, the results of H 2 S emission rates in the 10-month period are first discussed, followed by the temporal variation of H 2 S emissions in the continuous monitoring study. Finally, the laboratory results are presented to discuss the effects of the factors on H 2 S emissions. 4.3.1 H 2 S Emission Rate in the Field Study Figure 4-2 presents the results of H 2 S emission rates and H 2 S concentrations underneath the cover soils during the 10-month period in the sandy soil plot. The emission rates ranged from zero to 1.24 -5 mg m -2 s -1 , and the average emission rate was 4.67 -6 mg m -2 s -1 . The concentrations of H 2 S increased from 140 ppm in March to 3,000 ppm in July and remained at approximately 3,500 ppm at the rest of months. Higher emission rates were observed when H 2 S concentrations underneath the cover soils were relatively high. To put the results into perspective, the H 2 S emission rates measured in this study were compared to other reported landfill gas emission rates (Table 4-2). The average emission rate, 4.67 -6 mg m -2 s -1 found in this study, was similar to that in previous

PAGE 70

57 research, 2.07 -6 mg m -2 s -1 . However, the range of H 2 S emissions in this study was much smaller than that in previous research (from zero to 3.24 -2 mg m -2 s -1 ; Reinhart et al. 2004). It was because some “hot spots” were detected in the previous research. In these spots, large amount of H 2 S gas was emitted, resulting in high concentrations of H 2 S in the ambient air. In a laboratory column experiment, the emission rates of H 2 S ranged from 9.55 -4 to 0.26 mg m -2 s -1 (Plaza 2003), which was one to four orders of magnitude higher than the emissions shown in this study. The high H 2 S emission rates in the column experiment were attributed to high concentrations of H 2 S (as high as 124,000 ppm) underneath the cover soil and thin cover soil layer (15 cm), compared to approximately 3,500 ppm H 2 S concentrations under 30 cm depth of cover soil in this field study. When compared with methane emissions from MSW landfills, H 2 S emission rates in this field study were three to five orders of magnitude lower. The high emissions of methane resulted from large quantities of organic waste accepted by MSW landfills, but there is much less amount of organic waster at C&D debris landfills. Although the emission rates of H 2 S are relatively low at C&D debris landfills, it greatly increases H 2 S concentrations in the ambient air, posing adverse health effects on people living or working around these facilities (Selene and Chou 2003; Townsend et al. 2000 and 2005). 4.3.2 Temporal Variation of H 2 S Emission Rate During the field study, a large fluctuation in the emission rate was observed, even in two consecutive days (Figure 4-2). For example, the emission rate changed from 8.85 -6 mg m -2 s -1 on July 20 th to 1.06 -5 mg m -2 s -1 on July 21 st . Two questions have been raised: why does the H 2 S emission rate change? And does the emission rate change throughout the day? To answer these questions, H 2 S emission rates were continuously

PAGE 71

58 measured from morning to afternoon on three different days. The results showed that the H 2 S emission rates in the three days changed with time and had a similar pattern, increasing from the morning to the afternoon (Figure 4-3). Most emission rates were below the detection limit in the morning time, but high emission rates were observed in the afternoon, with the maximum emission rate of 1.24 -5 mg m -2 s -1 detected at 13:30 on September 10 th . To investigate the H 2 S emissions at nighttime, H 2 S emission rates were monitored in a 28-hour period, from 10 am on October 20 th to 2 pm on October 21 st , 2004. In addition to monitoring the H 2 S emission rate, other parameters, such as soil temperature, soil moisture content, and soil vapor H 2 S concentration, were also measured in this period. The results of this 28-hour continuous field study are presented in Figure 4-4. Completed raw data are provided in Appendix C. The H 2 S emission rates followed the same pattern previously observed during the daytime, increasing from morning to afternoon (Figure 4-4 A). The average emission rate of the period was 3.17 -6 mg m -2 s -1 . The maximum emission rate was detected at 3 pm on October 20 th , 1.16 -5 mg m -2 s -1 . Right after a 15-minute thunderstorm, the emission rate dropped from the maximum emission to zero, and then gradually increased at night. The H 2 S emission rate reached 5.67 -6 mg m -2 s -1 at 2:30 am on October 21 st . The emission gradually dropped through dawn and then started increasing near 10 am on October 21 st . The soil moisture content changed inversely with the changing of soil temperature over time (Figure 4-4 B). When soil temperature peaked around 32 0 C, the maximum H 2 S emission rate (3.17 -6 mg m -2 s -1 ) was detected with the lowest soil moisture content of

PAGE 72

59 9.7%. After the thunderstorm, cover soil moisture content greatly increased to 12.3%, and continued to slowly increase at night as soil temperature decreased. The highest soil moisture content was 14.3% at 7am on October 21 st . At daytime, the moisture content decreased with increasing the soil temperature. The soil vapor H 2 S concentration at a depth of 2.5 cm was decreased after the thunderstorm from 90 ppb to 20 ppb and gradually increased at night, reaching the maximum concentration of 2100 ppb at 6 am on October 21 st (Figure 4-4 C). While sampling at midnight, a strong H 2 S odor was noted on the testing plot. Ambient H 2 S concentrations at this time fluctuated around 50 ppb at midnight, compared to only a few ppb H 2 S during the daytime. Based on the field results, the H 2 S emissions were supposed to be affected by three main factors: soil H 2 S concentration, soil moisture and temperature. The laboratory experiments were then conducted to investigate the effects of these factors on H 2 S emissions. 4.3.3 Effect of H 2 S Concentration on H 2 S Emission Rate As mentioned above, H 2 S concentrations underneath the cover soils changed with time. Based on the concentration change, the 10-month experiment was divided into two periods. The H 2 S concentrations increased throughout the first period, from March to June 2004; in the second period (July 2004 to January 2005), the concentrations were relatively high and stable. Figure 4-5 presents the comparison of the average H 2 S emission rates in the two periods. In the first period, the average concentrations of H 2 S underneath cover soil were 1,369 ppm and the average H 2 S emission rate was 1.32 -6 mg m -2 s -1 . In the second period, with a high average H 2 S concentration (3,320 ppm), the average H 2 S emission rate was 6.68 -6 mg m -2 s -1 , approximately five times higher than that in the first period (Figure 4-5).

PAGE 73

60 In the laboratory experiment, three concentrations of H 2 S (200 ppm, 400 ppm, and 100 ppm) were introduced into the Buchner funnel in series to simulate the change of H 2 S concentration underneath the cover soils. As presented in Figure 4-6, 200 ppm H 2 S was first pumped into the funnel to simulate generated H 2 S in landfills. After diffusing through the sandy covers, the emission rate was approximately 0.0011mg m -2 s -1 . When H 2 S concentration increased to 400 ppm, the emission rate of H 2 S also increased to 0.0016 mg m -2 s -1 . Finally, immediately following the change from 400 ppm to 100 ppm H 2 S, the emission rate dropped, correspondingly. The results prove that H 2 S concentrations underneath the cover soil play a role in H 2 S emission rate. During the 28-hour continuous monitoring, it was observed that when the soil vapor H 2 S concentration increased at night, the H 2 S emission rate also increased (Figure 4-4). Based on the Fick’s law (Equation 4-1), dzdCDq (4-1) Increasing H 2 S concentrations underneath the cover soils would result in increasing of emission rates. At C&D debris landfills, the soil vapor H 2 S concentration has been reported to be extremely variable over a large range, from below 3 ppb to 12,000 ppm (Lee 2000), which may correspond to the large variety of H 2 S emission rates found in C&D debris landfills (Table 4-2). 4.3.4 Effect of Soil Moisture on H 2 S Emissions During the field study, it was noted that the H 2 S emission rates were low when the ground surface was wet, even with a high concentration of H 2 S underneath the cover soils. For example, on July 6 th , the H 2 S emission rate was 4.72 -6 mg m -2 s -1 with a bottom concentration of 2,400 ppm. With a heavy rainfall occurring on the afternoon of

PAGE 74

61 July 6 th , although the bottom H 2 S concentration increased to 3,250 ppm on July 7 th , the emission rate dropped to 2.78 -6 mg m -2 s -1 (Figure 4-2). Figure 4-7 presents the average emission rates of H 2 S under wet and dry surface conditions during the second period. During the second period, there were four rainfalls encountered, making the testing area surface wet. When the soil was wet, the average H 2 S emission was 3.06 -6 mg m -2 s -1 , while it increased to 8.00 -6 mg m -2 s -1 when the cover soil was in dry conditions. Since the soil moisture is highly depending on soil temperature, a laboratory experiment was conducted to investigate the effects of soil moisture and temperature on H 2 S emission using two Buchner funnels. In one funnel, the soil was heated by heating tape to increase the soil temperature to near 30 0 C, while soil in the other funnel was not heated and remained at 23 0 C. Both funnels were pumped with 200 ppm H 2 S from the bottoms. The emission rates from the funnels were continuous monitored over 34 hours (Figure 4-8). When the cover soil was maintained dry during the first 5 hours, there was no obvious difference in H 2 S emission rates of both funnels. After 5 hours, 10 ml of water was sprayed as simulated rainfall into each funnel. The top soil was quickly saturated, resulting in approximately a 95% drop in the emission rate of both funnels. As time progressed following the moisture addition, the emission rates of each funnel slowly increased. However, due to different water evaporation in the two funnels, after 24 hours the soil moisture content of the heated funnel and unheated funnel were 1.37% and 4.03%, respectively, and the emission rate in the heated funnel was higher than the unheated funnel. A second simulated rainfall at 29 hours caused the H 2 S emission rates of the two funnels to decrease sharply again and to gradually increase after that.

PAGE 75

62 The results confirm the field observation that soil moisture plays a role in affecting H 2 S emissions. The reasons for soil moisture effects upon H 2 S emissions include decreasing the H 2 S effective diffusion coefficient and the dissolving of H 2 S into the soil water. The effective diffusion coefficient represents the gas diffusion ability in a media and is defined by the Millington-Quirk Equation: 23/10)(TgasairDD (4-2) where D and D air are the effective diffusion coefficient of H 2 S in soil and air, respectively; T and gas refer to total porosity and air porosity in the cover soil. With the increasing of soil moisture, the air porosity ( gas ) decreases, reducing the effective diffusion coefficient (D) of H 2 S in soil, which means the diffusion of H 2 S in the cover soil will decrease. In other words, when H 2 S diffuses through cover soils, it tends to migrate through air pores of the cover soils. Increasing soil moisture reduces the amount of air pores, making H 2 S more difficult to diffuse through the cover soils, thus decreasing the H 2 S emission rates. In addition, with a considerable solubility in water (3,018 mg/L at 27 0 C; Chwirka 1990), more H 2 S could be dissolved in soil water as soil moisture increases, reducing the H 2 S emission rate. 4.3.5 Effect of Soil Moisture on Soil Vapor H 2 S Concentration Another field observation was that soil vapor H 2 S concentrations changed with soil moisture. As shown in Figure 4-4 B and C, as soil moisture increased from 9.7% to 14.3% from the afternoon thunderstorm to midnight, the soil vapor H 2 S concentration correspondingly increased from 0.02 ppm to 2.1 ppm. A similar observation was found

PAGE 76

63 on July 7th as H 2 S concentration underneath the cover soils increased from 2,400 ppm to 3,250 after a heavy rainfall on July 6th. Since both of the parameters can affect H 2 S emissions, another laboratory experiment was conducted to investigate the relationship between soil moisture and H 2 S concentration. The soil vapor samples were taken from the top, middle and bottom of the cover soils in the Buchner funnel under various soil moisture contents. When the cover soils were dry (soil moisture was 0%), the soil vapor H 2 S concentrations at the top, middle and bottom were 0.16 ppm, 8.6 ppm and 23 ppm, respectively. When the soil moisture increased to 5% by spraying 10 ml DI water on the dry soils, the concentrations of H 2 S at the three locations changed to 0.076 ppm, 10 ppm and 31 ppm, respectively. After adding another 10 ml DI water, the top H 2 S concentration further decreased to 0.063 ppm, while the bottom concentration increased to 35 ppm (Table 4-3). These results showed that with increasing soil moisture, H 2 S concentration underneath the cover soil increased, while the surface H 2 S concentration decreased. As discussed above (Equation 4-2), increasing soil moisture would retard the diffusion of H 2 S in cover soils, causing H 2 S accumulation in the bottom layers of the cover soils and reducing the surface H 2 S concentration. In landfill conditions, the increasing H 2 S concentrations after rainfall may result in more H 2 S generation, because the increasing soil moisture may stimulate the activity of SRB. In addition to the soil moisture, temperature, and H 2 S concentration, other parameters, such as air pressure, may affect H 2 S emissions as well. A significant inverse relationship was reported between CH 4 emission and atmospheric pressure (Czepiel et al. 2003). Increasing air pressure results in decline in landfill gas emissions, while a decrease

PAGE 77

64 in air pressure increases the emission rate. Figure 4-9 presents the possible the relationships between H 2 S emissions and the potential factors. For example, rainfall can increase the soil moisture (positive effect) and decrease the soil temperature (negative effect). Increasing soil moisture can directly reduce H 2 S emissions (negative effect) by decreasing H 2 S diffusion and dissolving H 2 S in the cover soils; however, it can also increase the H 2 S concentration, which may result in H 2 S emission rate increasing. As a result, the emission of H 2 S is determined by the predominant factor. As shown in Figure 4-5, the soil moisture might play a role in H 2 S emissions at daytime, while at night, H 2 S concentrations in the cover soils became the dominant factor. Based on the field study, the highest H 2 S emissions are most likely to occur at nighttime or afternoon at a landfill site because of the effect of soil moisture. However, it was noticed that most H 2 S odor complaints happened in the morning and the ambient H 2 S concentrations in the morning were usually higher than that in the afternoon. This is because the ambient H 2 S concentrations are not only dependent on H 2 S emissions from landfills, but also on H 2 S dispersion in the atmosphere. In the morning, due to the low temperature of the ground, the environmental lapse is inverted (temperature increases with height) and the atmosphere is relatively stable, which reduces the dispersion of the emitted H 2 S. In addition, the vertical convective mixing of the emitted H 2 S and ambient air is limited in the morning because of the low vertical temperature difference between ground and ambient air (Xu and Townsend 2004). Therefore, the emitted H 2 S can not be effectively dispersed into the atmosphere in the morning, resulting in the ambient H 2 S concentration increasing around the landfills. In the afternoon, the stability of the

PAGE 78

65 atmosphere reduces and the vertical convective mixing increase, which makes the emitted H 2 S can be quickly dispersed in the atmosphere, diluting the ambient H 2 S concentrations. 4.4 Summary In this chapter, the H 2 S emission rates in the sandy soil testing plot were monitored in a 10-month period. The emission rates of H 2 S were variable, ranging from zero to 1.24 10 -5 mg m -2 s -1 , with an average of 4.67 -6 mg m -2 s -1 . The emission rates of H 2 S were lower during the morning and higher in the afternoon. Laboratory experiments were conducted to investigate the effect of the soil moisture, temperature and soil H 2 S concentration on H 2 S emissions. Results showed that soil moisture plays an important role in affecting H 2 S emissions. It can reduce H 2 S emissions by retarding H 2 S diffusion in cover soil and dissolving H 2 S into soil water; but on the other hand, increasing soil moisture can result in H 2 S accumulation underneath the cover soil, which may increases H 2 S emissions. The H 2 S emission rate is not dependent on one single factor, since the factors influencing H 2 S emissions interact with each other and change throughout the day. The ambient H 2 S concentration depends not only on H 2 S emissions from landfills, but also on the dispersion of H 2 S in the atmosphere. Due to the limited vertical convective mixing and inverted environmental lapse, the emitted H 2 S can not be effectively dispersed in the atmosphere during the morning period. The poor dispersion of H 2 S, to some extent, results in the morning odor problems of H 2 S at C&D debris landfills.

PAGE 79

66 Table 4-1. Influencing factors investigated in the laboratory experiment Experiment Influencing Factors Range Comment 1 H 2 S concentration 100, 200, 400 ppm Mix H 2 S with N 2 at different ratios 2 Soil moisture 0%, 5%, 10% DI water spray 3 Soil temperature 23 0 C and 30 0 C Heated by heating tape Table 4-2. Comparison of reported landfill gas emission rates Landfill gas Range of Emission rates Sites Source H 2 S 0 1.24 -5 mg m -2 s -1 Field testing plot This study 2005 H 2 S 0 1.78 -4 mg m -2 s -1 The same landfill as this study H 2 S 0 3.24 -2 mg m -2 s -1 Five different C&D debris landfills Reinhart et al. 2004 H 2 S 1.9 535 ml day -1 (9.55 -4 0.26 mg m -2 s -1 ) Laboratory column experiment Plaza 2003 CH 4 2.2 230.8 g m -2 day -1 (2.55 -2 2.67 mg m -2 s -1 ) MSW landfill Boeckx et al. 1996 Table 4-3. The change of soil vapor H 2 S concentrations under various soil moistures in the laboratory experiment Soil moisture content Locations 0% 5% 10% Surface 0.16 ppm 0.076 ppm 0.063 ppm Middle 8.6 ppm 10 ppm 12 ppm Bottom 23 ppm 31 ppm 35 ppm

PAGE 80

67 Master Flex Pump H2S Sampling SandTedlarBagH2S 5 cm Rubber stopperGlass Buchner funnel Master Flex Pump H2S Sampling SandTedlarBagH2S TedlarBagH2S 5 cm Rubber stopperGlass Buchner funnel Figure 4-1. Schematic of the laboratory buchner funnels experiment Month Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb H2S emission rate (mg m-2 s-1) 02x10-64x10-66x10-68x10-610x10-612x10-6 Bottom H2S concentration (ppm) 010002000300040005000 H2S emission rate H2S concentration Detection Limit:1.77x10-6mg m-2 s-1 Figure 4-2. The change of H 2 S emission rates and bottom concentrations in the sandy soil plot

PAGE 81

68 Time 08:00 09:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 H2S emission rate (mg m-2 s-1) 02x10-64x10-66x10-68x10-610x10-612x10-6 Augest 11th September 10th October 1st Figure 4-3. Temporal change of H 2 S emission rates at daytime in the sandy soil plot

PAGE 82

69 Time 10:00 14:00 18:00 22:00 2:00 6:00 10:00 14:00 H2S concentration at 2.5 cm depth (ppb) 0500100015002000 Moisture content Soil temperature (0C) 2022242628303234 H2S emission rate (mg m-2 s-1) 02x10-64x10-66x10-68x10-610x10-612x10-614x10-6 ABC14%13%12%11%10%9%Moisture content Soil temperature Rainfall Figure 4-4. Temporal change of A) H 2 S emission rate; B) soil moisture content and soil temperature; C) H 2 S concentrations in the 28-Hour continuous field study

PAGE 83

70 H2S concentration (ppm) 010002000300040005000 H2S emission rate (mg m-2 s-1) 02x10-64x10-66x10-68x10-610x10-612x10-6 H2S concentration H2S emission rate First period(March 2004 June 2004) Second period(July 2004 January 2005) Figure 4-5. The effect of soil H 2 S concentrations on H 2 S emissions in the field study Time (hour) 02468 10 H2S emission rate (mg m-2 s-1) 0.00060.00080.00100.00120.00140.00160.0018 200 ppm H2S400 ppm H2S100 ppm H2S Figure 4-6. The change of H 2 S emission rate with different bottom H 2 S concentrations (laboratory experiment)

PAGE 84

71 H2S concentration (ppm) 010002000300040005000 H2S emission rate (mg m-2 s-1) 02x10-64x10-66x10-68x10-610x10-612x10-6 H2S concentration H2S emission rate Wet Dry Figure 4-7. The effect of soil moisture on H 2 S emissions in the field study Time (Hour) 0246810242628303234 F/F0 Simulated rainfall Simulated rainfall DryMC=1.37%MC=4.03%MC=5%100%80%60%40%20%0 Soil temperature 230C Soil temperature 300C Figure 4-8. The effect of soil moisture and temperature on H 2 S emission rate in the laboratory experiment

PAGE 85

72 Temperature Soil Moisture H2S Concentration Air PressureH2S Emission Rate Rain H2S Generation (+)(+)(+)(+)(-)(-)(-)(+)(-)(+): Positive effect(-): Negative effect(?): Unknown Other Factors (?) Temperature Soil Moisture H2S Concentration Air PressureH2S Emission Rate Rain H2S Generation (+)(+)(+)(+)(-)(-)(-)(+)(-)(+): Positive effect(-): Negative effect(?): Unknown Other Factors (?) Figure 4-9. Schematic of the relationship among H 2 S emissions and possible influencing factors

PAGE 86

CHAPTER 5 MODELING OF H 2 S MIGRATION THROUGH LANDFILL COVER MATERIALS 5.1 Introduction Hydrogen sulfide (H 2 S) emissions from C&D debris landfills are an increasing issue of concern in North America. The proper use of landfill cover can minimize H 2 S emissions by acting as a barrier to prevent H 2 S migration into the atmosphere (Koerner and Daniel 1997; Reinhart et al. 2004). Previous field research showed that as H 2 S migrated through landfill covers, H 2 S was gradually removed by cover soils, reducing H 2 S emissions (Chapter 3). Although lots of research has been conducted to investigate methane and carbon dioxide migration in MSW landfill cover systems, little research has been performed for H 2 S migration in landfill cover systems (El-Fadel et al. 1996; Kjeldsen and Fischer 1995; Mosher et al. 1996; Williams et al. 1999). As a reactive and reducing gas, H 2 S may be removed both chemically and biologically in cover soils (Bohn and Bohn 1988; Bohn et al. 1989; Cihacek and Bremner 1993; Gumerman and Carlson 1969). The migration of H 2 S in cover soils is a complicated process and is influenced by cover soil properties such as soil type, particle size, compaction, as well as climatic variables such as moisture and temperature (Chapter 3 and 4). Understanding H 2 S migration in landfill covers can lead to better management strategies for reducing H 2 S emissions at C&D debris landfills. The objective of this study was to mathematically simulate H 2 S migration in landfill systems. Using a migration model, a sensitivity analysis was performed to evaluate the factors influencing H 2 S migration in cover soils. A laboratory column experiment was 73

PAGE 87

74 conducted to compare the experimental data with model simulation results and to estimate the H 2 S adsorption coefficient of different cover materials. In addition, a possible method for designing the depth of cover soil was provided. 5.2 Theory To simplify, all equations in this study are given as one-dimensional (vertical) equations, from underneath the cover soil to the atmosphere. The mechanisms governing H 2 S migration through the cover soils include diffusion and advection (Figure 5-1). Diffusion is the tendency for H 2 S to move from a greater concentration to a lower concentration, while advection is the movement of H 2 S due to pressure gradient across the cover soil layer. After generation from disposed gypsum drywall, H 2 S will migrate through cover soil by both advection and diffusion and the flux of H 2 S in cover soil can be given by the advection-diffusion equation (Equation 5-1), zCDCq (5-1) where q is the flux of H 2 S; v is advection velocity of H 2 S in the vertical direction; C is H 2 S concentration; and D is effective diffusion coefficient of H 2 S in cover soil. The H 2 S vertical velocity is function of pressure and is determined from Darcy’s law (Tchobanoglous et al. 1993). Based on the equation of gas flow in porous media given by Bear (1972), H 2 S migration in cover soil can be expressed as follows: zqtCg (5-2) where g is the total porosity of the cover soil and is the generation rate of H 2 S in the cover soil. As described previously in this dissertation, H 2 S may be gradually removed by

PAGE 88

75 cover materials as it migrates through landfill cover (Chapter 3). H 2 S removal should thus also be considered in the migration equation. Assuming that H 2 S removal by the cover soils occurs as a first-order reaction: Cd t dC (5-3) where is the H 2 S adsorption coefficient in the first-order reaction. Assuming there is no H 2 S generation in the cover soils, the equation for H 2 S migration in the cover soil is expressed in Equation (5-4) CzCzCDtCg22 (5-4) In steady state, the H 2 S concentration in the cover soil layer does not change over time, 0tC (5-5) therefore, the H 2 S migration equation in the cover soils at steady state is simplified as Equation (5-6) 022CdzdCdzCdD (5-6) To solve the above equation, the H 2 S concentration underneath the cover soils was specified as C 0 (Equation 5-7), and the concentration gradient was assumed to be zero at the top of the cover soils (Equation 5-8). 0)0(CC (5-7) 0zdzdC (5-8) Therefore, in steady state, the H 2 S concentration at depth z of the cover soil is:

PAGE 89

76 ]2)(exp[)(0DzuCzC (5-9) where )41(2Du (5-10) Based on the solution (5-9), the H 2 S concentration (C (z) ) in a cover soil layer is a function of many parameters, including the initial H 2 S concentration, the effective diffusion coefficient, the H 2 S adsorption constant, cover soil type, compaction, moisture content and others. 5.3 Model Simulation and Laboratory Experiment In this study, both modeling simulation and laboratory experiments were performed to investigate the factors influencing H 2 S migration in the cover soils and to compare with each other. In addition, the experimental data were used to estimate the H 2 S adsorption coefficient () of four cover materials. 5.3.1 Model Simulation Based on the solution of H 2 S migration in landfill cover (Equation 5-9), a sensitivity analysis was undertaken to evaluate the sensitivity of model results to the following key model parameters: initial H 2 S concentration (C 0 ), effective diffusion coefficient (D), advection velocity (v), and H 2 S adsorption constant of cover soil (). In the sensitivity analysis, all four parameters were systematically changed (one parameter at a time) in a plausible range. The parameters values and range used in the model simulations are listed in Table 5-1.

PAGE 90

77 5.3.2 Laboratory Column Experiment 5.3.2.1 Column construction Laboratory cover soil columns were constructed of 10-cm diameter and 75-cm long polyvinyl chloride (PVC) pipe (Figure 5-2). A 10-cm slip cap glued to the bottom of each column and a stainless steel Swagelok valve (model SS-4P4T4, Swagelok) was installed for introducing H 2 S from the bottom. A 15-cm layer of crushed glass was loaded at the bottom of each column for uniformly distributing the introduced H 2 S and to serve as a H 2 S reservoir. A geocomposite was placed on the top of the glass layer to prevent cover material falling through. Then a 50-cm depth of cover material was loaded into the column, and the final headspace depth of the cover soil column was kept at 10 cm. Six perforated plastic tubes were installed horizontally at different depths in each column, with 10-cm interval from the bottom to the surface of the cover material. The plastic tubes were used to collect soil vapor gas samples and were connected to six gas tight valves which were used as gas samples ports. The cover soils were open to the atmosphere, and a hood was installed on the top of the columns to collect the emitted H 2 S from the columns. 5.3.2.2 Cover materials Four different materials were used as cover soils in the column experiment. These were sandy soil, fine concrete, coarse concrete, and lime-amended sandy soil. Sandy soil was obtained from a C&D debris landfill in central Florida. Concrete was obtained from a concrete recycling facility and sieved by ASTM # 4 sieve (FisherSci, Inc.). The fine concrete had particle sizes less than 4.75 mm, while particle sizes of the coarse concrete was larger than 4.75 mm. The lime-amended sandy soil was prepared by mixing 1% hydrated lime (Ca(OH) 2 ) by weight with sandy soil. The moisture contents of these cover

PAGE 91

78 materials were measured and are listed in the Table 5-1. The air porosity ( g ) of the cover soil was calculated by (Equation 5-11): wTg (5-11) where w is the volumetric moisture content in the cover soil and T is the total porosity and can be obtained from: sbT1 (5-12) where b is the bulk density of the cover soil (Table 5-1) and s is the particle density of the cover soils. In this study, the particle density of all cover materials was assumed as 2.65 g/cm 3 . The effective diffusion coefficients for H 2 S of the cover soils were calculated by the Millington-Quirk Equation (Equation 5-13): TgairDD 3/10)( (5-13) where D air is the diffusion coefficient of H 2 S in air (1.85 -5 m 2 s -1 ; Chiang et al. 2000). The characteristics of these cover materials are summarized in Table 5-2. 5.3.2.3 Column operation and gas sampling Once the columns were filled with cover materials, all sampling ports were closed. H 2 S gas was introduced into the columns using a low flow rate pump (MasterFlex L/S Model 7419-10, ColePamer). The pump was connected between the Swagelok valve at the bottom of each cover column and an 80-L Tedlar bag containing H 2 S gas. The H 2 S gas was continuously pumped into the columns for a few days to initialize the system, and sampling was performed to monitor H 2 S concentrations in the cover soils. When the

PAGE 92

79 change of H 2 S concentrations remained within 10%, the system was assumed to be at a steady state. Gas samples were then taken from the sampling ports by a 10-ml glass syringe and were diluted in 500-ml Tedlar sampling bags by laboratory air. H 2 S concentration of the gas sample was then analyzed by a Jerome H 2 S analyzer with detection range from 0.003 ppm to 50 ppm (See Chapter 2 for more detail). 5.3.2.4 Laboratory experiments Based on Equation 5-9, a series of laboratory column experiments were conducted to investigate the effects of the four parameters on H 2 S migration in the cover soils, including: (1) initial H 2 S concentration; (2) advection velocity; (3) effective diffusion coefficient; (4) H 2 S adsorption coefficient. Table 5-3 lists the laboratory experiments performed in this study. Various concentrations of H 2 S were prepared by mixing pure H 2 S and nitrogen gas at different ratios in 80-L Tedlar bags which served as gas reservoirs. The H 2 S gas was introduced into the columns using a Masterflex pump to simulate the advection flow of H 2 S from a landfill into the cover soils. The advection velocity was changed by adjusting the flow rate of the pump and calculated by Equation 5-14: gAQv (5-14) where v is the advection velocity of H 2 S in the cover soil (m/s); Q is H 2 S flow rate provided by the pump; A is the column surface area, 78.5 cm 2 ; and the g is the air porosity of the cover soils. Since the fine concrete and coarse concrete have different diffusion coefficients (Table 5-2), they were used to investigate the effect of effective diffusion coefficient on H 2 S migration in the columns. In previous research, the adsorption of H 2 S by the sandy soil and the soil treated with lime were different

PAGE 93

80 (Chapter 3). Thus in this study, the two soils were used to investigate the effect of H 2 S adsorption coefficient on H 2 S migration. 5.4 Results and Discussion This section presents the effects of the four main parameters on H 2 S migration, followed by a discussion about a potential method for designing the depth of cover soil. 5.4.1 Effect of Initial H 2 S Concentration on H 2 S Migration in the Cover Soils In C&D debris landfills, because of the heterogeneous characteristic of C&D debris and how it is disposed in landfills, differences in cover soil types and practices among sites, H 2 S initial concentration underneath cover soils are expected to differ from site to site (Chapter 4). Previous research reported subsurface H 2 S concentrations to range from less than 50 ppm to 12,000 ppm (Lee 2000). As discussed in Chapter 4, H 2 S concentration underneath the cover soils plays a role in H 2 S emissions. Figure 5-3A and B represent H 2 S concentration results from the simulation and the laboratory experiment, respectively. The axis of H 2 S concentration was shown in log scale to illustrate the difference at lower concentrations. Three different initial concentrations of H 2 S were used to for both the model simulation (Table 5-1) and the laboratory experiment (experiment set 1, Table 5-3). The results indicated that with increasing initial H 2 S concentration, soil vapor H 2 S concentrations increased across the depth. For example, in the column experiment, when the initial H 2 S concentration was 2,800 ppm, the concentration at 0.4 m was 24 ppm, while, with the initial H 2 S concentration of 45 ppm, the concentration was only 0.21 ppm. Based on Fick’s law, with the same depth of cover soils, increasing H 2 S concentration means increasing the gradient of H 2 S diffusion across the cover soils, which would increase H 2 S emissions. It indicates that to eliminate the H 2 S emissions from the sites with high H 2 S concentrations at

PAGE 94

81 landfills, thicker depth of cover soils should be applied at these sites by reducing the H 2 S concentration gradient. 5.4.2 Effect of Advection Velocity on H 2 S Migration in the Cover Soils The advection of H 2 S in cover soils is the result of pressure difference inside and outside landfills. Because of landfill gas generation, the pressure inside landfills is normally higher than outside landfills, forcing H 2 S migration through the cover soils through advection. However, due to different generation rates of C&D debris, the pressures inside landfills are different from site to site, resulting in variation of the H 2 S advection velocity. Figure 5-4 A and B present the results of H 2 S concentration in the cover soils from the simulations and the laboratory experiment, respectively. In the laboratory experiment (experiment set 2, Table 5-3), the advection of H 2 S was simulated by pumping H 2 S. By adjusting the pump flow rate from 5 ml/min, 10 ml/min and 15 ml/min, the advection velocity of H 2 S in the columns were 2.11 -5 , 4.23 -5 , and 6.33 -5 m/s, respectively. The results showed that the higher advection velocity leads to larger H 2 S concentrations in soil vapor. As the advection velocity increased from 2.11 -5 to 6.33 -5 m/s in the column, the soil vapor H 2 S concentration increased from 0.72 ppm to 34 ppm at the 0.4 m of the sandy soil column. The increasing soil vapor H 2 S concentration resulted from less removal of H 2 S by adsorption processes in the cover soils. In previous serum bottle H 2 S adsorption experiments (Chapter 3), more H 2 S was removed by the cover materials with time. However, higher H 2 S advection velocity in the cover soils reduces contact time between H 2 S and cover soils, resulting in H 2 S concentration in the cover soils increasing.

PAGE 95

82 5.4.3 Effect of Effective Diffusion Coefficient on H 2 S Migration in the Cover Soils Figure 5-5 A and B demonstrate the effects of effective diffusion coefficient on H 2 S concentrations in the model simulation and the laboratory experiment, respectively. Both results showed that H 2 S concentrations were lower in the media with small diffusion coefficient than the media with large diffusion coefficient. Because of different particle size, the effective diffusion coefficients of fine concrete and coarse concrete were 6.53 -6 m 2 s -1 and 1.09 -6 m 2 s -1 , respectively (Table 5-2). In the fine concrete, the H 2 S concentration was only 0.003 ppb at the 0.4 m level, but was 100 ppm at the same level in the coarse concrete. Since the effective diffusion coefficient reflects the diffusion ability of H 2 S in the media, the media with low diffusion coefficient, such as fine concrete, can effective retard H 2 S migration in the media, extending the retention time of H 2 S gas in the cover soils, which results in soil vapor H 2 S concentration decreasing. In landfill conditions, because of the difference of soil type, particle size, moisture content, and compaction of cover soils, the H 2 S effective diffusion coefficients are expected to be different in landfill covers. In the field study, it was observed that H 2 S emission rates were reduced after rainfall (Chapter 4). One reason for that was high soil moisture content lowered H 2 S diffusion coefficient in cover soils. According to the Millington-Quirk Equation (Equation 5-14), increasing of moisture content or compacting the cover soils (increasing the bulk density b of the cover soil), the effective diffusion coefficient of H 2 S would decrease. Figure 5-6 presents the change of effective diffusion coefficient as a function of soil moisture and compaction, respectively. If the sandy soil is more compacted (compaction 120%) in this study, the effective diffusion coefficient would decrease from 6.09 -6 m 2 s -1 to 4.16 -6 m 2 s -1 . If the soil moisture

PAGE 96

83 increased from 4.27% in this study to 6%, the effective diffusion coefficient would decrease from 6.09 -6 m 2 s -1 to 5.31 -6 m 2 s -1 . The results indicate that during landfill cover operation, reducing cover soil particle size and effective compaction can help attenuate H 2 S emissions by decreasing the diffusion coefficient of H 2 S in cover soils. 5.4.4 Effect of H 2 S Adsorption Coefficient on H 2 S Migration in the Cover Soils As H 2 S migrated through the cover soils, it was gradually removed by physical adsorption/dissolution, chemical or biological reactions (Chapter 3). Due to different characteristics of the cover soils, H 2 S adsorption abilities of cover materials were different from one another. A pervious laboratory experiment showed that the lime-amended sandy soils had better H 2 S adsorption ability than the sandy soil (Chapter 3). Figure 5-7 A and B show the effect of H 2 S adsorption coefficient on H 2 S migration in the model simulation and the column experiment. The results showed that increasing the H 2 S adsorption coefficient leaded to soil vapor H 2 S concentration decreasing. For example, in the lime amended sandy soil column, soil vapor H 2 S concentration decreased to 0.003 ppm after 2800 ppm H 2 S migrated through 0.4 m cover soil. The H 2 S concentration in the sandy soil column, however, was as high as 24 ppm at the same depth. This indicates that materials with higher H 2 S adsorption coefficients may be used as cover soils to help attenuate H 2 S emissions from landfills. 5.4.5 Model Application and Limitations As discussed in Chapter 3, the emissions of H 2 S are highly dependent on the type of cover materials which have the different H 2 S adsorption abilities. Therefore, it is important to estimate the value of the H 2 S adsorption coefficient () of the cover materials to design landfill cover for attenuating H 2 S emission at C&D debris landfills.

PAGE 97

84 Based on the migration model, the H 2 S adsorption coefficient () of cover materials could be obtained using the laboratory column experiment. Figure 5-8 provides a possible flow chart to estimate the value of the H 2 S adsorption coefficient () using the migration model and the laboratory column experiment. First of all, the potential cover materials are collected to conduct the laboratory column experiment. Based on the experimental set up, the parameters of the cover materials, such as moisture content and bulk density, can be obtained, which can be used to calculate the effectively diffusion coefficient of H 2 S (D) according to the Millington-Quirk Equation. Then H 2 S gas (C 0 ) is introduced into the column containing the cover materials with a flow rate (Q) by a pump. Based on the gas flow rate, the advection velocity (v) can be calculated using Equation 5-15. In order to get H 2 S adsorption coefficient (), the soil vapor H 2 S concentrations at different depths (C i,i=1,2,.6 ) are plotted as function of the depth of cover soils (z). The slop of the best fit ( Duv2 ) can be obtained. Then, the H 2 S adsorption coefficient () of the cover materials can be calculated according to the Equation 5-10. According to the calculation procedure discussed above, the values of H 2 S adsorption coefficient () of the cover materials used in this study were calculated (Table 5-4). The fine concrete had the highest value (1.00 -2 s -1 ), followed by the lime-amended sandy soil with a value of 9.86 -3 s -1 . The sandy soil had a value of 1.65 -3 s -1 . The lowest value (1.23 -3 s -1 ) of the coarse concrete was attributed to its large particle size, resulting in high diffusion of H 2 S in the cover soils. High value of H 2 S adsorption coefficient () means that H 2 S could be removed more effectively by the cover materials. The adsorption coefficient () values from this study were also

PAGE 98

85 supported by the pervious serum bottle H 2 S adsorption experiment (Chapter 3). The fine concrete, with the highest value of H 2 S adsorption coefficient (), exhibited the most effective H 2 S removal ability compared to other cover materials. In the field study, although H 2 S can be removed in the sandy soil, because of its relatively low value, the sandy soil can not attenuate H 2 S emissions as effectively as the alternative cover materials, such as fine concrete and lime-amended sandy soil (Chapter 3). In addition to estimating the H 2 S adsorption coefficient of cover materials, the migration model could be used to design landfill cover depths. As shown in Figure 5-8, after obtaining the values of H 2 S effective diffusion coefficient (D), the advection velocity (v) of H 2 S in the cover soils, and H 2 S adsorption coefficient () of the cover materials, the depth of the cover soil (z) can be predicted, as shown in the Equation 5-15 )ln(ln2)(0zCCvuDz (5-15) where C (z) is the acceptable H 2 S concentration on the surface of the cover soils at C&D debris landfills and C 0 is the H 2 S concentration underneath the cover soils. For example, in a landfill site where the H 2 S concentration underneath the cover soils (C 0 ) was 1,000 ppm, the four cover materials used in this study were used to as landfill cover to reduce the concentration to an acceptable level (e.g., 0.003 ppm). Based on the parameter values obtained from the laboratory column experiment (Table 5-4), the depth of the cover soils would be 0.34 m if the fine concrete was used, 0.45 m if the lime amended sandy soil was used, 0.95 m for the sandy soil, and 1.10 m for the coarse concrete. The results indicated that using the fine concrete and the lime-amended sandy soil as cover soils could attenuate H 2 S emissions more effectively than the sandy soil. In other words, less cover materials (fine concrete and lime-amended sandy soil) would be

PAGE 99

86 required to achieve the same treatment results. This means that it might provide a potential cost-effective method to address the H 2 S odor problems at C&D debris landfills using the alternative cover materials with high values of H 2 S adsorption coefficient as landfill cover soils. Figure 5-9 presents the comparison results of H 2 S concentrations in the sandy soil from the model simulations and column experimental data. The model simulations and experimental data were comparable. The difference between the simulation and the experiment data may result from experimental error and model limitations. The H 2 S adsorption coefficient (), for example, was considered as a constant in the model. However, the H 2 S adsorption of cover soil could be decreased after long-term exposure to H 2 S because of the formation of sulfide minerals on the surface of cover soils (Gumerman 1968). In terms of the effective diffusion coefficient (D), it is temporal change at C&D debris landfills, with the changing of cover soil moisture content and soil temperature throughout the day (Chapter 4). In addition, other parameters, such as H 2 S viscosity, soil temperature, and effects of other landfill gases, may also affect the migration and should be considered in the migration model in the future. 5.5 Summary Using a one-dimensional advection-diffusion equation, a model was developed to investigate the migration of H 2 S in landfill covers. Four major parameters, including H 2 S initial concentration, advection velocity, effective coefficient, and adsorption coefficient, were considered in this model. The model simulations and the experimental data from the laboratory column experiment were comparable and showed that all four parameters play roles in affecting H 2 S migration in cover soils. H 2 S emissions can be reduced by

PAGE 100

87 decreasing the diffusion and advection of H 2 S in the cover soils or using alternative materials with high H 2 S adsorption coefficient as cover soils. In addition to providing a simple method to understand H 2 S migration, the migration model also provides a potential method to estimate the H 2 S adsorption coefficient values of cover materials and to predict the depth of landfill cover soils based on the laboratory column experiment.

PAGE 101

88 Table 5-1. Parameters values used for the sensitivity analysis Parameters Values Range Initial H 2 S concentration (C 0 ) 1000 ppm 100 – 10000 ppm Advection velocity () 4.23 -5 m s -1 4.23 -6 – 4.23 -4 m s -1 Effective Diffusion coefficient (D) 6.0910 -6 m 2 s -1 6.09 -7 – 6.09 -5 m 2 s -1 H 2 S adsorption coefficient () 1.6510 -3 s -1 1.65 -4 – 1.65 -2 s -1 Table 5-2. Summary of the characteristics of cover materials in the column experiment Cover Materials Moisture Content Amount (kg) Bulk Density (kg m -3 ) Air Porosity Effective Diffusion Coefficient (m 2 s -1 ) Sandy soil 4.27% 4.65 1.18 0.50 6.0910 -6 Fine concrete 1.15% 5.25 1.34 0.48 6.5310 -6 Coarse concrete 0.09% 3.20 0.82 0.68 1.0910 -5 1% lime amended sandy soil 4.15% 4.65 1.18 0.50 6.1610 -6 Table 5-3. Summary of experimental operation conditions in the column experiments Experimental Operation Conditions Set Influencing factors Introduced H 2 S concentration (ppm) Pump rate (ml/min) Cover materials 1 Initial H 2 S concentration 45, 900, and 2800 10 Sandy soil 2 Advection velocity 2800 5, 10, 15 Sandy soil 3 Effective Diffusion coefficient 2800 10 Fine concrete Vs. coarse concrete 4 H 2 S adsorption coefficient 2800 10 Sandy soil Vs. lime amended sandy soil Table 5-4. Calculation values of the D, v, and of the cover materials Parameters Sandy soil Fine concrete Coarse concrete Lime amended sandy soil D (m 2 s -1 ) 6.0910 -6 6.53 -6 1.09 -5 6.16 -6 v (m s -1 ) 4.2310 -5 4.42 -5 3.10 -5 4.21 -5 (s -1 ) 1.6510 -3 1.00 -2 1.23 -3 9.86 -3

PAGE 102

89 Landfill Cover SoilH2S Advection: vCDiffusion: Generation: Adsorption: -C z zCDEmitted H2S Landfill Cover SoilH2S Advection: vCDiffusion: Generation: Adsorption: -C z zCD Advection: vCDiffusion: Generation: Adsorption: -C z Generation: Adsorption: -C z zCDEmitted H2S Figure 5-1. Schematic of H 2 S migration in landfill cover soils TedlarbagH2S Glasses Cover soil Sampling ports Headspace 75 cm10 cm 10cm Plastictubing Pump 10cm 10cm 15cm 10cm 10cm 10cm Hood TedlarbagH2S Glasses Cover soil Sampling ports Headspace 75 cm10 cm 10cm 10cm Plastictubing Pump 10cm 10cm 10cm 10cm 15cm 10cm 10cm 10cm Hood Figure 5-2. Schematic of laboratory cover soil column

PAGE 103

90 H2S concentration (ppm) 10-210-1100101102103104 Depth of cover soil (m) 0.00.10.20.30.40.50.6 0.1 C0 (100 ppm) C0 (1000 ppm) 10 C0 (10000 ppm) Advection velocity (: 4.23x10-5 m/sDiffusion coefficient (D): 6.09x10-6m2/sAdsoprtion coefficient (: 0.00165 s-1A H2S concentration (ppm) 10-210-1100101102103104 Depth of cover soil (m) 0.00.10.20.30.40.50.6 C0=2800 ppm C0=900 ppm C0=45 ppm BH2S flow rate: 10 ml/minCover material: Sandy soil Figure 5-3. The effect of H 2 S initial concentration on H 2 S migration. A) Modeling results. B) Experiment results

PAGE 104

91 H2S concentration (ppm) 02004006008001000 Depth of cover soil (m) 0.00.10.20.30.40.50.6 0.1 (4.23x10-6 m/s) (4.23x10-5 m/s) 10 (4.23x10-4 m/s) H2S concentration (C): 1000 ppmDiffusion coefficient (D): 6.09x10-6m2/sAdsoprtion coefficient (: 0.00165 s-1A H2S concentration (ppm) 050010001500200025003000 Depth of cover soil (m) 0.00.10.20.30.40.5 Advection velocity =2.11x10-5 m/s Advection velocity =4.23x10-5 m/s Advection velocity =6.34x10-5 m/s BH2S concentration: 2800 ppmCover material: Sandy soil Figure 5-4. The effect of H 2 S advection velocity on H 2 S migration. A) Modeling results. B) Experiment results.

PAGE 105

92 H2S concentration (ppm) 02004006008001000 Depth of cover soil (m) 0.00.10.20.30.40.50.6 0.1 D (6.09x10-7 m2/s) D (6.09x10-6 m2/s) 10D (6.09x10-5 m2/s) H2S concentration (C): 1000 ppmAdvection velocity (: 4.23x10-5 m/sAdsoprtion coefficient (: 0.00165 s-1A H2S concentration (ppm) 050010001500200025003000 Depth of cover soil (m) 0.00.10.20.30.40.50.6 B Fine concrete D=6.53x10-6 m2/s Coarse concrete D=1.09x10-5 m2/s H2S concentration: 2800 ppmH2S flow rate: 10 ml/min Figure 5-5. The effect of effect diffusion coefficient on H 2 S migration. A) Modeling results. B) Experiment results.

PAGE 106

93 Effective diffusion coefficient D (m2/s) 2x10-64x10-66x10-68x10-610x10-6 Moisture content 0.000.020.040.060.080.10 Compaction 0.70.91.11.31.50.60.81.01.21.4 Soil moisture content Cover soil compaction 6.09 x10-6 m2/s Wet Compaction Figure 5-6. The change of H 2 S diffusion coefficient vs. soil moisture content and soil compaction

PAGE 107

94 H2S concentration (ppm) 02004006008001000 Depth of cover soil (m) 0.00.10.20.30.40.50.6 (0.000165 s-1) (0.00165 s-1) 10 0.0165 s-1) H2S concentration (C): 1000 ppmVertical velocity (: 4.23x10-5 m/sDiffusion coefficient (D): 6.09x10-6m2/sA H2S concentration (ppm) 050010001500200025003000 Depth of cover soil (m) 0.00.10.20.30.40.5 Sandy soil ( =0.00136 s-1) Lime amended sandy soil ( =0.00909 s-1) BH2S concentration: 2800 ppmH2S flow rate: 10 ml/min Figure 5-7. The effect of adsorption coefficients of cover soils on H 2 S migration. A) Modeling results. B) Experiment results.

PAGE 108

95 Collect cover material Build laboratory column Conduct column experiment Particle density: s Moisture content: MC Bulk density: b Volumetric moisture content: w Total porosity: TGas porosity: g Volumetric moisture content: w Total porosity: TGas porosity: g Effective diffusion coefficient (D) 210/3)(TgairDD Effective diffusion coefficient (D) 210/3)(TgairDD Effective diffusion coefficient (D) 210/3)(TgairDD Advection velocity (v) of H2S in the columns H2S flow rate Advection velocity (v) of H2S in the columns H2S flow rate H2S adsorption coefficient ()of the cover material H2S adsorption coefficient ()of the cover material H2S adsorption coefficient ()of the cover material Depth of cover soil (z) )ln(ln42)(02zCCvDvDz Depth of cover soil (z) )ln(ln42)(02zCCvDvDz Experimental procedure Experimental procedureData collection Data collection Calculation procedure Calculation procedure H2S concentration (C1,C2..C5) Ln(Ci)z Slopek H2S concentration (C1,C2..C5) Ln(Ci)z Slopek Figure 5-8. Flow chart for designing the depth of landfill cover

PAGE 109

96 H2S concentration (ppm) 0.0010.010.1110100100010000 Depth of cover soil (m) 0.00.10.20.30.40.5 Simulated result Experimental result Figure 5-9. Comparison of modeling results and experimental data

PAGE 110

CHAPTER 6 SUMMARY AND CONCLUSIONS 6.1 Summary Odor problems associated with hydrogen sulfide (H 2 S) generation and emissions at C&D debris landfills have become a significant environmental issue in North America. H 2 S is generated as a result of biological sulfate reduction of disposed gypsum drywall. H 2 S emissions not only lead to odor complaints, but may also pose a potential health and safety risk to people living or working near these facilities. Although the generation of H 2 S from landfill sites has been well documented in previous research (Lee 2000; Yang 2000), little research has been conducted to address control strategies for these problems. Research was performed to investigate potential engineering methods to solve H 2 S problems at C&D debris landfills, including inhibition of H 2 S generation and attenuation of H 2 S emissions. This dissertation was organized into four main studies. The first study evaluated the effects of three chemical inhibitors on H 2 S generation from gypsum drywall. They were sodium molybdate, ferric chloride, and hydrated lime. Various concentrations of each inhibitor were first tested using a flask experiment to determine the appropriate inhibition concentration for subsequent laboratory column experiment. The column experiment evaluated the longer-term effect of each inhibitor on H 2 S generation. All three inhibitors exhibited the ability to inhibit H 2 S generation. Sodium molybdate inhibited H 2 S generation by interrupting the biological sulfate reduction and hydrated lime inhibited H 2 S generation by providing unfavorable pH for SRB growth. Instead of stimulating a 97

PAGE 111

98 competing group of IRB, the high concentration of ferric chloride used in this study suppressed the activity of SRB by providing an extremely acidic environment. Both sodium molybdate and ferric chloride effectively inhibited H 2 S generation over a long period, but hydrated lime had a limited inhibition effect in the long term because the alkalinity it contributed was neutralized by the generated H 2 S. The next attempt of this research was to evaluate H 2 S attenuation by various cover materials in a field study. Six different cover materials, such as a sandy soil, fine concrete, compost, and various lime-amended sandy soils, were used in an attempt to attenuate H 2 S emissions. H 2 S concentration profiles illustrated that H 2 S was removed when it migrated through the cover soils. The results showed the alternative cover materials effectively attenuated H 2 S emissions from the testing area and detectable H 2 S emissions were only encountered in the sandy soil testing plot. Laboratory experiments showed the alternative cover materials removed H 2 S more effectively than the sandy soil by a series of reactions such as physical adsorption, chemical reaction, and biological oxidation. In addition, the alternative cover materials reduced H 2 S generation by providing an adverse environment for SRB, such as high pH and temperature. The third study investigated the emission rate of H 2 S and the factors influencing H 2 S emissions. During the 10-month field study, the emission rate of H 2 S changed with time, with an average emission rate of 4.67 -6 mg m -2 s -1 . The continuous emission monitoring results illustrated that the emission rates changed through the course of the day. Laboratory experiments were then conducted to investigate the factors influencing H 2 S emissions, including soil moisture, temperature, and H 2 S concentration. The results showed that increasing soil moisture could directly reduce H 2 S emissions by retarding

PAGE 112

99 H 2 S diffusion in the cover soils and dissolving H 2 S into soil water; however, increasing soil moisture could also increase soil H 2 S concentration, which might lead to an increase in H 2 S emission. The temporal variation of H 2 S emissions at C&D debris landfills is the result of the interaction of many factors which change throughout the day. In the final study, the migration of H 2 S was investigated by developing a mathematical model. Four major parameters: H 2 S initial concentration, advection velocity, effective diffusion coefficient, and H 2 S adsorption coefficient of cover soil, were considered in this model. Model simulations indicated that H 2 S migration can be decreased by reducing H 2 S diffusion and advection or using alternative cover soils with high H 2 S adsorption coefficient. Laboratory column experiments were conducted to evaluate the effect of these parameters and to compare the data from the simulations. A possible method for designing landfill cover depth was provided in this study. Based on the laboratory column experiment, the H 2 S adsorption coefficient of cover material can be estimated, which can be used to predict the depth of cover soil according to the H 2 S migration model. 6.2 Conclusions Based on this study, although H 2 S can be readily generated at C&D debris landfills, it may be possible to substantially control the H 2 S problems through proper engineering methods, landfill operation and management. More specifically, the conclusions of this research were as follows: H 2 S generation can be reduced by using chemical inhibitors. However, the inhibition effect is highly dependent on the chemical concentration. With respect to landfill application, potential environmental contamination resulting from the chemical inhibitors should be considered. Low concentrations of sodium molybdate effectively inhibited H 2 S generation from gypsum drywall over a long-term period.

PAGE 113

100 H 2 S generation was inhibited by hydrated lime, but over a long-term period (180 days), the inhibition effect was limited because the alkalinity it contributed was neutralized by the generated H 2 S. Instead of stimulating the growth of iron-reducing bacteria, the 500 mM ferric chloride solution used in the column study prevented H 2 S generation by providing unfavorable acidic environment for SRB. However, due to the extremely low pH of the chemical solution and large amount needed, it may not be practical in the field. Some alternative cover materials, such as fine concrete, compost, and lime-amended sandy soils, effectively attenuated H 2 S emissions from a C&D debris landfill. H 2 S emissions were only detected at the sandy soil testing plot. In other plots using these alternative cover materials, the emission rates of H 2 S were under the detection limit. Alternative cover materials can not only reduce H 2 S generation by providing an unfavorable environment for sulfate-reducing bacteria, but they also may remove H 2 S more effectively than sandy soil by increasing a series of reactions, such as chemical reaction or biological oxidation. The emission rates of H 2 S varied during the day, which resulted from the effects of many factors, such as soil moisture, temperature, barometric pressure, and soil vapor H 2 S concentrations. There are two different ways soil moisture affects H 2 S emission rates. Improving soil moisture can reduce H 2 S emission rate by retarding H 2 S migration and dissolving H 2 S into soil water; on the other hand, it stimulates H 2 S generation and increases soil vapor H 2 S concentration, resulting in an H 2 S emission rate increase. The migration of H 2 S in cover soils plays a role in affecting H 2 S emission from landfills and is affected by many factors, such as H 2 S concentration, H 2 S advection velocity, effective diffusion coefficient, and H 2 S adsorption coefficient. Based on the migration model, the value of H 2 S adsorption coefficient of the cover soil could be estimated using a laboratory column experiment. In addition, the migration model provides a potential method to design the depth of landfill cover soils. 6.3 Future Work The results from this research provide some potential methods for addressing the H 2 S problems at C&D debris landfills. Future work should be conducted to evaluate the inhibition effect of sodium molybdate on H 2 S generation in a large-scale field study, to assess the potential environmental impacts, and to analyze the cost-effectiveness. Other

PAGE 114

101 future work should investigate the effect of other alternative materials, such as coal ash and wood ash, on H 2 S attenuation. In terms of the migration model, although it could be used to design the depth of landfill cover soils, the effects of other parameters, such as soil temperature and H 2 S viscosity, the changing of H 2 S adsorption coefficient of cover soils due to a long-term exposure to H 2 S, and the temporal change of soil moisture should be considered in future model modification.

PAGE 115

APPENDIX A SUPPLEMENTAL LEACHATE DATA FOR THE INHIBITION COLUMN EXPERIMENT Table A-1. Results of leachate parameter in the inhibition columns Month Columns pH Sulfate (mg/L) Calcium (mg/L) Iron (mg/L) Molybdenum (mg/L) Sodium (mg/L) Control 6.34 890 474.9 27.0 0.1 41.7 Na 2 MoO 4 6.81 1246 338.2 3.6 22.2 245.2 FeCl 3 1.68 1352 2409.9 15497.6 0.1 42.4 March Ca(OH) 2 11.5 967 732.5 92.2 0.1 38.3 Control 6.48 1393 1274.8 25.9 0.0 181.4 Na 2 MoO 4 7.12 1024 734.7 13.3 15.0 384.9 FeCl 3 1.98 1564 4389.4 20625.9 0.0 64.9 April Ca(OH) 2 9.24 1428 1041.2 18.2 0.1 119.1 Control 6.89 1245 1191.8 1.0 0.0 78.3 Na 2 MoO 4 6.54 1332 814.0 7.1 34.6 434.1 FeCl 3 1.7 1720 5396.7 17394.0 0.3 101.6 May Ca(OH) 2 7.23 1147 628.0 16.5 0.8 44.6 Control 6.75 1267 1411.2 2.4 0.0 145.2 Na 2 MoO 4 6.43 1490 712.5 15.1 14.8 370.7 FeCl 3 1.85 1670 3999.8 18176.2 0.1 61.7 June Ca(OH) 2 7.14 1481 992.5 16.9 0.1 109.3 Control 6.77 1577 789.2 24.0 0.1 67.5 Na 2 MoO 4 6.68 1148 603.7 2.0 28.4 487.6 FeCl 3 2.12 1348 4491.9 25195.3 0.1 57.2 July Ca(OH) 2 6.99 1051 1494.7 4.6 0.1 71.7 Control 6.82 1479 887.3 0.4 0.1 107.4 Na 2 MoO 4 6.21 1107 625.4 3.4 26.9 580.5 FeCl 3 2.03 1327 2468.4 15851.4 0.3 63.0 August Ca(OH) 2 6.91 1420 1065.0 12.2 0.1 138.0 102

PAGE 116

APPENDIX B CONSTRUCTION PROCEDURE OF THE TESTING AREA AT A CONSTRUCTION AND DEMOLITION DEBRIS LANDFILL 103

PAGE 117

104 0.4 m 12 m 18 m 5-cm crushed gypsum drywall 5-cm crushed gypsum drywall First geocomposite First geocomposite Vent The second geocomposite Gas pipe Water Water Water0.3 m 6 m 6 m Sandy soil Fine concrete0.3 m 6 m 6 m Sandy soil Fine concrete Cover soil C&D debris landfill Excavation Load gypsum drywall Place geocomposite Install gas pipe and Place the 2ndgeocomposite Spay water Load cover materials Figure B-1. Construction procedure of the testing area in the field study

PAGE 118

105 Tubing sampling systemVent Compost Fine concrete 3% Ca(OH)2amended sandy soil 1% Ca(OH)2 amended sandy soil 10% CaCO3amendedSandy soil Sand 30 cm22.5 cm15 cm7.5 cm2.5 cm32 cm Figure B-2. Picture of the testing area in the field study Jerome Meter Flux Chamber Thermometer Flow Meter N2Gas Figure B-3. Picture of the H 2 S emission measurement system in the field study

PAGE 119

APPENDIX C SUPPLEMENTAL DATA IN THE FIELD STUDY H2S Concentraton (ppm) 0.0010.010.11101001000 Depth of Cover Soil (cm) 051015202530 Sandy soil Fine concrete Compost 10% CaCO3 1% Ca(OH)2 3% Ca(OH)2 March 2nd 2004 Figure C-1. H 2 S concentration profile in the testing plots on March 2 nd 2004 H2S Concentraton (ppm) 0.0010.010.11101001000 Depth of Cover Soil (cm) 051015202530 Sandy soil Fine concrete Compost 10% CaCO3 1% Ca(OH)2 3% Ca(OH)2 March 9th 2004 Figure C-2. H 2 S concentration profile in the testing plots on March 9 th 2004 106

PAGE 120

107 H2S Concentraton (ppm) 0.0010.010.1110100100010000 Depth of Cover Soil (cm) 051015202530 Sandy soil Fine concrete Compost 10% CaCO3 1% Ca(OH)2 3% Ca(OH)2 March 30th 2004 Figure C-3. H 2 S concentration profile in the testing plots on March 30 th 2004 H2S Concentraton (ppm) 0.0010.010.1110100100010000 Depth of Cover Soil (cm) 051015202530 Sandy soil Fine concrete Compost 10% CaCO3 1% Ca(OH)2 3% Ca(OH)2 April 20th 2004 Figure C-4. H 2 S concentration profile in the testing plots on April 20 th 2004

PAGE 121

108 H2S Concentraton (ppm) 0.0010.010.1110100100010000 Depth of Cover Soil (cm) 051015202530 Sandy soil Fine concrete Compost 10% CaCO3 1% Ca(OH)2 3% Ca(OH)2 April 25th 2004 Figure C-5. H 2 S concentration profile in the testing plots on April 25 th 2004 H2S concentration (ppm) 0.0010.010.1110100100010000 Depth of cover soil (cm) 051015202530 Sandy soil Fine concrete Compost 10% CaCO3 1% Ca(OH)2 3% Ca(OH)2 May 3rd 2004 Figure C-6. H 2 S concentration profile in the testing plots on May 3 rd 2004

PAGE 122

109 H2S Concentraton (ppm) 0.0010.010.1110100100010000 Depth of Cover Soil (cm) 051015202530 Sandy soil Fine concrete Compost 10% CaCO3 1% Ca(OH)2 3% Ca(OH)2 May 10th 2004 Figure C-7. H 2 S concentration profile in the testing plots on May 10 th 2004 H2S Concentraton (ppm) 0.0010.010.1110100100010000 Depth of Cover Soil (cm) 051015202530 Sandy soil Fine concrete Compost 10% CaCO3 1% Ca(OH)2 3% Ca(OH)2 June 15th 2004 Figure C-8. H 2 S concentration profile in the testing plots on June 15 th 2004

PAGE 123

110 H2S Concentraton (ppm) 0.0010.010.1110100100010000 Depth of Cover Soil (cm) 051015202530 Sandy soil Fine concrete Compost 10% CaCO3 1% Ca(OH)2 3% Ca(OH)2 June 30th 2004 Figure C-9. H 2 S concentration profile in the testing plots on June 30 th 2004 H2S Concentraton (ppm) 0.0010.010.1110100100010000 Depth of Cover Soil (cm) 051015202530 Sandy soil Fine concrete Compost 10% CaCO3 1% Ca(OH)2 3% Ca(OH)2 July 6th 2004 Figure C-10. H 2 S concentration profile in the testing plots on July 6 th 2004

PAGE 124

111 H2S Concentraton (ppm) 0.0010.010.1110100100010000 Depth of Cover Soil (cm) 051015202530 Sandy soil Fine concrete Compost 10% CaCO3 1% Ca(OH)2 3% Ca(OH)2 July 7th 2004 Figure C-11. H 2 S concentration profile in the testing plots on July 7 th 2004 H2S Concentraton (ppm) 0.0010.010.1110100100010000 Depth of Cover Soil (cm) 051015202530 Sandy soil Fine concrete Compost 10% CaCO3 1% Ca(OH)2 3% Ca(OH)2 July 12th 2004 Figure C-12. H 2 S concentration profile in the testing plots on July 12 th 2004

PAGE 125

112 H2S Concentraton (ppm) 0.0010.010.1110100100010000 Depth of Cover Soil (cm) 051015202530 Sandy soil Fine concrete Compost 10% CaCO3 1% Ca(OH)2 3% Ca(OH)2 July 20th 2004 Figure C-13. H 2 S concentration profile in the testing plots on July 20 th 2004 H2S Concentraton (ppm) 0.0010.010.1110100100010000 Depth of Cover Soil (cm) 051015202530 Sandy soil Fine concrete Compost 10% CaCO3 1% Ca(OH)2 3% Ca(OH)2 July 28th 2004 Figure C-14. H 2 S concentration profile in the testing plots on July 28 th 2004

PAGE 126

113 H2S Concentraton (ppm) 0.0010.010.1110100100010000 Depth of Cover Soil (cm) 051015202530 Sandy soil Fine concrete Compost 10% CaCO3 1% Ca(OH)2 3% Ca(OH)2 July 29th 2004 Figure C-15. H 2 S concentration profile in the testing plots on July 29 th 2004 H2S Concentraton (ppm) 0.0010.010.1110100100010000 Depth of Cover Soil (cm) 051015202530 Sandy soil Fine concrete Compost 10% CaCO3 1% Ca(OH)2 3% Ca(OH)2 August 27th 2004 Figure C-16. H 2 S concentration profile in the testing plots on August 27 th 2004

PAGE 127

114 H2S Concentraton (ppm) 0.0010.010.1110100100010000 Depth of Cover Soil (cm) 051015202530 Sandy soil Fine concrete Compost 10% CaCO3 1% Ca(OH)2 3% Ca(OH)2 September 10th 2004 Figure C-17. H 2 S concentration profile in the testing plots on September 10 th 2004 H2S Concentraton (ppm) 0.0010.010.1110100100010000 Depth of Cover Soil (cm) 051015202530 Sandy soil Fine concrete Compost 10% CaCO3 1% Ca(OH)2 3% Ca(OH)2 December 2nd 2004 Figure C-18. H 2 S concentration profile in the testing plots on December 2 nd 2004

PAGE 128

115 H2S Concentraton (ppm) 0.0010.010.1110100100010000 Depth of Cover Soil (cm) 051015202530 Sandy soil Fine concrete Compost 10% CaCO3 1% Ca(OH)2 3% Ca(OH)2 January 20th 2004 Figure C-19. H 2 S concentration profile in the testing plots on January 20 th 2005

PAGE 129

116 Table C-1. H 2 S emission rate change in the sandy soil testing plot during the 28-hour continuous monitoring Date Time N 2 gas flow rate (L/min) H 2 S concentration (ppm) H 2 S emission rate (mg m -2 s -1 ) 10/20/2004 12:00 5 0.003 1.77 -6 10/20/2004 13:30 5 0.005 2.95 -6 10/20/2004 14:30 6 0.003 2.13 -6 10/20/2004 15:00 8 0.012 1.16 -5 10/20/2004 15:30 10 0 1 0 10/20/2004 16:10 5 0.004 2.07 -6 10/20/2004 17:20 5 0.004 2.07 -6 10/20/2004 19:30 7 BDL 2 8.86 -7 10/20/2004 21:30 5 BDL 8.86 -7 10/20/2004 22:30 5 0.003 1.77 -6 10/20/2004 23:45 5 0.005 3.13 -6 10/21/2004 01:00 5 0.009 5.02 -6 10/21/2004 02:30 6 0.008 5.67 -6 10/21/2004 04:00 6 0.007 4.96 -6 10/21/2004 05:00 5 0.003 1.77 -6 10/21/2004 06:00 6 0.003 2.13 -6 10/21/2004 07:30 5.5 0.004 2.60 -6 10/21/2004 08:30 5 BDL 8.8610-7 10/21/2004 09:30 5 BDL 8.8610-7 10/21/2004 10:00 5 BDL 8.8610-7 10/21/2004 11:00 5 0.004 2.36 -6 10/21/2004 11:30 6 0.004 2.83 -6 10/21/2004 12:00 8 0.003 3.12 -6 10/21/2004 13:00 8.5 0.006 6.02 -6 10/21/2004 14:00 10 0.009 1.06 -5 1 : The reading of the Jerome meter is zero; 2 : The reading of the Jerome meter is 1 ppb or 2 ppb.

PAGE 130

117 Table C-2. Temperature change at different depths of the cover soils during the 28-hour continuous monitoring Temperature ( 0 C) Date Time 2.5 cm 7.5 cm 15 cm 22.5 cm 30 cm 32 cm 10/20/2004 11:30 28.6 26.7 25.8 25.7 26.0 28.0 10/20/2004 12:00 31.4 28.8 26.8 26.0 26.2 38.1 10/20/2004 13:40 32.7 31.3 29.4 27.0 26.7 38.2 10/20/2004 14:30 33.4 31.8 30.2 27.8 27.4 37.8 10/20/2004 15:30 31.4 31.5 30.6 28.2 27.7 27.8 10/20/2004 16:10 30.7 30.7 30.4 28.7 28.2 27.0 10/20/2004 17:00 29.4 29.9 30.0 28.9 28.5 25.4 10/20/2004 19:00 25.9 26.7 27.8 28.6 28.6 22.4 10/20/2004 21:00 25.3 26.0 27.0 28.1 28.2 22.8 10/20/2004 22:30 24.9 25.4 26.4 27.6 27.8 22.2 10/20/2004 23:45 24.0 24.8 25.9 27.1 27.4 21.1 10/21/2004 01:00 23.4 24.2 25.3 26.7 27.0 20.5 10/21/2004 02:30 22.8 23.6 24.7 26.2 26.6 19.9 10/21/2004 04:00 22.2 23.0 24.1 26.5 26.1 19.4 10/21/2004 05:00 21.8 22.5 23.7 25.3 25.8 19.5 10/21/2004 06:00 21.5 22.3 23.4 24.9 25.5 19.1 10/21/2004 07:30 21.7 22.1 23.1 24.6 25.1 21.0 10/21/2004 08:30 22.7 22.6 23.1 24.4 24.9 22.4 10/21/2004 09:10 23.5 23.0 23.3 24.3 24.9 23.8 10/21/2004 10:00 25.5 24.2 23.8 24.4 24.8 29.0 10/21/2004 11:00 28.3 26.0 24.7 24.6 24.9 35.0 10/21/2004 11:30 30.5 27.9 26.0 24.9 25.1 40.0 10/21/2004 12:00 31.1 28.6 26.5 25.1 25.2 40.0 10/21/2004 13:00 32.0 30.3 28.2 26.1 26.0 43.1 10/21/2004 14:00 31.8 30.7 28.8 26.6 26.3 41.2

PAGE 131

118 Table C-3. Soil vapor H 2 S concentration change at different depths of the cover soil during the 28-hour continuous monitoring Soil vapor H 2 S concentration (ppm) Date Time 2.5 cm 7.5 cm 15 cm 22.5 cm 30 cm 10/20/2004 11:30 0.002 0.11 1.08 280 840 10/20/2004 13:30 0.004 0.02 — 200 408 10/20/2004 14:30 0.014 0.09 5 440 1280 10/20/2004 16:00 0.019 0.08 0.11 408 600 10/20/2004 17:00 0.017 0.02 0.18 520 440 10/20/2004 19:30 0.01 0.035 0.04 300 — 10/20/2004 21:00 0.003 0.03 0.04 256 200 10/20/2004 22:30 0.023 0.043 0.07 292 424 10/20/2004 23:45 0.014 0.29 0.31 312 560 10/21/2004 01:00 0.06 0.42 0.45 312 720 10/21/2004 02:30 0.04 0.5 0.73 — 640 10/21/2004 04:00 0.057 0.8 0.88 284 800 10/21/2004 05:00 0.066 1.7 2.3 400 720 10/21/2004 06:00 0.044 2.1 2.7 380 700 10/21/2004 07:30 0.013 0.34 0.39 300 620 10/21/2004 09:00 0.004 0.3 0.37 360 720 10/21/2004 10:00 0.006 0.32 0.38 460 820 10/21/2004 11:00 0.01 0.28 0.32 480 780 10/21/2004 12:00 0.009 0.23 0.28 580 660 10/21/2004 13:30 0.012 0.26 0.25 540 780 10/21/2004 14:00 0.008 0.73 0.97 480 760 Table C-4. Soil moisture content change during the 28-hour continuous monitoring Date Time Soil Moisture Content 10/20/2004 11:30 11.6% 10/20/2004 13:40 10.4% 10/20/2004 15:00 9.7% 10/20/2004 17:00 12.3% 10/20/2004 21:00 12.5% 10/20/2004 23:45 12.8% 10/21/2004 01:00 13.5% 10/21/2004 03:00 13.6% 10/21/2004 05:00 13.8% 10/21/2004 07:00 14.3% 10/21/2004 09:00 12.8% 10/21/2004 11:00 12.1% 10/21/2004 12:00 11.7% 10/21/2004 13:00 11.7% 10/21/2004 14:00 11.4%

PAGE 132

APPENDIX D CALCULATION OF HYDROGEN SULFIDE RATE USING THE FLUX CHAMBER METHODS The H 2 S emission rate was calculated by the following equation: A CvF Where: F = H 2 S emission rate (mg m -2 s -1 ); v = the flow rate of the sweeping nitrogen gas (m 3 /s); A = the covered area by the flux chamber (m 2 ); C = the H 2 S concentration of outlet gas (mg/m 3 ). The cover area of the 65-L flux chamber used in this study is 0.196 m 2 . The flow rate of nitrogen gas ranged from 5 L/min to 10 L/min. The detection limit of the Jerome meter is 0.003 ppm. 3333/102.4345.24003.0003.0003.0mmgmmolmgmlmmolmmlmmlppm Therefore, the detection limit of the flux chamber method for H 2 S emission rate is 12633233 1077.1102.4196.0min105smmgmmgmmF For example, if the outlet H 2 S concentration is 7 ppb with sweeping 6 L/min nitrogen gas, H 2 S emission rate will be 1263233 1096.45.2434007.0196.0min106smmgmmgmmF 119

PAGE 133

LIST OF REFERENCES Agency for Toxic Substances and Disease Registry (ATSDR) (2004). Toxicological Profile for Hydrogen Sulfide. Department of Health and Human Services, ATSDR, Division of Toxicology, Atlanta, Georgia. Arizona Instrument LLC. (2003). Jerome 631-x Hydrogen Sulfide Analyzer Operation Manual. Tempe, AZ. Barton, L. L., and Plunkett, R. M. (2002). "Sulfate-Reducing Bacteria: Environmental and Technological Aspects." In: Encyclopedia of Environmental Microbiology , Gabriel Bitton, edition. John Wiley, New York. Bear, J. (1972). Dynamic of Fluids in Porous Media . American Elsevier, New York. Berthouex, P. M., and Brown, L. C. (1994). Statistics for Environmental Engineering. Lewis Publishers, Boca Raton, Florida. Boeckx, P., Cleemput, O.V., and Villaralvo, I. (1996). "Methane Emission from a Landfill and the Methane Oxidising Capacity of Its Covering Soil." Soil Biology and Biochemistry, 28: 1397-1405. Bohn, H. and Bohn, R. (1988). "Soil Beds Weed Out Air Pollutants." Chemical Engineering, 95: 73-76. Bohn, H. L., Yong, F., and Cheng, H. (1989). "Hydrogen Sulfide Sorption by Soils." Soil Science Society of America Journal, 53: 1914-1917. Borjesson, G. and B. H. Svensson (1997). "Seasonal and Diurnal Methane Emission from a Landfill and Their Regulation by Methane Oxidation." Waste Management and Research, 15: 33-54. Bradl, H.B. (2005). Heavy Metals in the Environment: Origin, Interaction and Remediation . ELSEVIER Academic Press, San Diego , CA. Brinton, S.R. (2000). Sorption of Molybdenum in Two Florida Soils Amended with Biosolids . Department of Agricultural and Biological Engineering. University of Florida, Gainesville. Master. Burgess, J. E., Parsons, S. A., and Stuetz, R. M. (2001). "Development in Odour Control and Waste Gas Treatment Biotechnology: a Review." Biotechnology Advances, 19: 35-63. 120

PAGE 134

121 Campagna, D., Kathman, S. J., Pierson, R., Inserra, S. G., Phifer, B. L., Middleton, D. C., Zarus, G. M., and White, M. C. (2003). "Ambient Hydrogen Sulfide, Total Reduced Sulfur, and Hopital Visits for Respiratory Diseases in Northeast Nebraska, 1998-2000." Journal of Exposure Analysis and Environmental Epidemiology, 14: 180-187. Chiang, H.-L., Tsai, J.-H., Chang, D.-H., and Jeng, F.-T. (2000). "Diffusion of Hydrogen Sulfide and Methyl Mercaptan onto Microporous Alkaline Activated Carbon." Chemosphere, 41: 1227-1232. Chung, Y.-C., Huang, C., Pan, J.R., and Tseng, C-P. (1998). "Comparison of Autotrophic and Mixotrophic Biofilters for H 2 S Removal." Journal of Environmental Engineering, 124: 362-367. Chwirka, J. D., and Satchell, T. T. (1990). " Hydrogen Sulfide Treatment Systems in Sanitary Sewers." Water Environment and Technology, 2: 48-53. Cihacek, L. J., and Bremner, J. M. (1993). "Characterization of the Sulfur Retained by Soils Exposed to Hydrogen-Sulfide." Communications in Soil Science and Plant Analysis, 24: 85-92. Clancy, P. B., Venkataraman, N., and Lynd, L.R. (1992). "Biochemical Inhibition of Sulfate Reduction in Batch and Continuous Anaerobic Digesters." Water Science and Technology, 25: 51-60. Crosson, J. (1995). "County Focuses on Mine Security: Hydrogen Sulfide Gas Problems at Sunset Mine and Landfill Prompt Concern." The Tampa Tribune. February 15, 1995. Pasco Section. Page 1 Czepiel, P. M., Shorter, J.H., Mosher, B., Allwine, E., jMcManus, J.B., Harriss, R.C., Kolb, C.E. and Lamb, B.K. (2003). "The Influence of Atmospheric Pressure on Landfill Methane Emissions." Waste Management, 23: 593-598. Eckford, R. E., and Fedorak, P. M. (2002). "Planktonic Nitrate-Reducing Bacteria and Sulfate-reducing Bacteria in Some Western Canadian Oil Field Waters." Journal of Industrial Microbiology and Biotechnology, 29: 83-92. El-Fadel, M., Findikakis, A. N., and Leckie, J. O. (1996). "Numerical Modeling of Generation and Transport of Gas and Heat in Landfills I. Model Formulation." Waste Management and Research, 14: 483-504. Fairweath, R.L. and Barlaz, M.A. (1998). "Hydrogen Sulfide Production during Decomposition of Landfill Inputs." Journal of Environmental Engineering, 124: 353-361. Florida Department of Environmental Protection (FDEP). (2000). "2000 Solid Waste Management Annual Report".

PAGE 135

122 Flynn, B. (1996). "Sulfide Treatment Technologies for Sour Landfill Gas." Waste Age, 27: 155-160. Flynn, B. (1998). "Invisible Threat: Odors and Landfill Gas from C & D Waste." Waste Age, 29: 91-97. Francoeur, C. M. (1993). "H 2 S Control for the Landfill Industry." SWANA 16 th Annual Landfill Gas Symposium, Louisville, KY, 57-73. Godish, T. (1991). Air Quality 2 nd . Lewis Publisher, Inc. Chelsea, MI. Gypsum Association. (1992a). "Treatment and Disposal of Gypsum Board Waste: Industry Position Paper." Construction Dimensions. February. 5-6, 29-30. Gypsum Association. (1992b). “Treatment and Disposal of Gypsum Board Waste: Technical Paper Part II.” Construction Dimensions. March, 58-63. Gumerman, R. C. (1968). Chemical Aspects of H 2 S Removal in Soil . Department of Civil Engineering, University of Washington. Ph.D. Gumerman, R. C., and Carlson, D. A. "Chemical Aspects of Odor Removal in Soil System." Cornell U. Conf. on Ag. Waste Management, Ithaca, NY, 292-302. Hao, J.O., Jin, M.C., and Li, H. (1996). "Sulfate-Reducing Bacteria." Critical Review in Environmental Science and Technology, 26: 155-187. Hobson, J. and Yang, G. (2000). "The Ability of Selected Chemicals for Suppressing Odour Development in Rising Mains." Water Science and Technology, 41:165-173. Isa, H. M., and Anderson, G. K. (2005). "Molybdate Inhibition of Sulphate Reduction in Two-Phase Anaerobic Digestion." Process Biochemistry, 40: 2079-2089. Johnson, B. (1986). "Gypsum Wallboard Creates Landfill Odor Problem." World Wastes, 29: 53-54. Kjeldsen, P., and Fischer, E. V. (1995). "Landfill Gas Migration-Field Investigations at Skellingsted Landfill, Denmark." Waste Management and Research, 13: 467-484. Koerner, R. M. and Daniel, D. E. (1997). Final Covers for Solid Waste Landfills and Abandoned Dumps . Reston, Virginia, ASCE Press. Lee, S. (2000). Landfill Gas Composition at Florida Construction and Demolition Debris Facilities . Department of Environmental Engineering and Sciences. University of Florida. Gainesville, Master. Liao, W. and Chou, F. S. (1998). Measurement of Methane Emission from a Landfill with Flux Chamber . 91 st Annual Meeting of Air and Waste Management Association, San Diego, California.

PAGE 136

123 Lin, T.-F. and Su, Y.-Y. (2003). "Performance of Gas Mixing in a Flux Chamber." Energy Sources, 25: 597-605. Lovley, D. R. (1991). "Dissimilatory Fe (III) and Mn (IV) Reduction." Microbiological Reviews, 55: 259-287. Lovley, D. R. and E. P. Phillips (1986). "Organic Matter Mineralization with Reduction of Ferric iron in Anaerobic Sediments." Applied and Environmental Microbiology, 51: 683-689. McInerney, M. J., Bhupathiraju, V., and Sublette, K. L. (1992). "Evaluation of a Microbial Method to Reduce Hydrogen Sulfide Levels in a Porous Rock Biofilm." Journal of Industrial Microbiology, 11: 53-58. Montgomery, A. D., McInerney, M. J., and Sublette, K.L. (1990). "Microbial Control of the Production of Hydrogen Sulfide by Sulfate-Reducing Bacteria." Biotechnology and Bioengineering, 35: 533-539. Mosher, B. W., Czepiel, P. C., Shorter, J., Allwine, E., Harriss, R. C., Kolb, C., and Lamb, B. (1996). "Mitigation of Methane Emission at Landfill Sites in New England, USA." Energy Conversion and Management, 37: 1093-1098. Musick, M. (1992). "Recycling Gypsum from C & D Debris." Biocycle, 33: 34-36. Myhr, S., Lillebo, B.L.P., Sunde, E., Beeder, J., and Torsvik, T. (2002). "Inhibition of Microbial H 2 S Production in an Oil Reservoir Model Column by Nitrate Injection." Applied and Environmental Microbiology, 58: 400-408. Nemati, M., Mazutinec, T. J., Jenneman, G. E., and Voordouw, G. (2001). "Control of Biological H2S Production with Nitrite and Molybdate." Journal of Industrial Microbiology and Biotechnology, 26: 350-355. Nelson, W. H. (1990). "Gypsum Waste Disposal: Land Vs. Sea or Recycling." Chemistry and Ecology, 4: 247-258. Nicolai, R. E., and Janni, K. A. (2001). "Biofilter Media Mixture Ratio of Wood Chips and Compost Treating Swine Odors." Water Science and Technology, 44: 261-267. Okabe, S., Satoh, H., Itoh, T., and Watanabe, Y. (1999). "Microbial Ecology of Sulfate-Reducing Bacteria in Wastewater Biofilms Analyzed by Microelectrodes and FISH (Fluorescent In Situ Hybridization) Technique." Water Science and Technology, 39: 41-47. Opresko, D.M. (1993). “Toxicity Summary for Molybdenum.” In: The Risk Assessment Information System ( RAIS): Molybdenum (7439-98-7) , Oak Ridge Reservation Environmental Restoration Program. Oak Ridge, Tennessee. http://risk.lsd.ornl.gov/tox/profiles/molybdenum_f_V1.shtml

PAGE 137

124 Park, J. W. and Shin, H. C. (2001). "Surface Emission of Landfill Gas from Solid Waste Landfill." Atmospheric Environment, 35: 3445-3451. Parkes, R. J., Gibson, G. R., Harvey, M., Buckingham, W. J., and Herbert, R. A. (1989). "Determination of the Substrates for Sulphate-Reducing Bacteria within Marine and Estuarine Sediments with Different Rates of Sulphate Reduction." Journal of General Microbiology, 135: 175-187. Plaza, C. (2003). Evaluation of Cover Materials for the Removal of Hydrogen Sulfide from Construction and Demolition Landfills. Department of Environmental Engineering and Sciences. University of Florida. Gainesville, Master. Picot, B., Paing, J., Toffoletto, L., Sambuco, J. P., and Costa, R. H. R. (2001). "Odor Control of An Anaerobic Lagoon with A Biological Cover: Floating Peat Beds." Water Science and Technology, 44: 309-316. Poduska, R.A. and Anderson, B.D. (1981). "Successful Storage Lagoon Odor Control." Journal of Water Pollution Control Federation, 53: 299-310. Postgate, J. R. (1984). The Sulphate Reducing Bacteria . Cambridge, Cambridge University Press. Ranade, D. R., Dighe, A.S., Bhirangi, S.S., Panhalkar, V.S., and Yeole, T.Y. (1999). "Evaluation of the Use of Sodium Molybdate to Inhibit Sulphate Reuduction during Anaerobic Digestion of Distillery Waste." Bioresource Technology, 68: 287-291. Reinhart, D., Townsend, T., Eun, S., and Xu, Q. (2004). Control of Odors from Construction and Demolition (C&D) Debris Landfills, Florida Center for Solid and Hazardous Waste Management. Saleh, A. M., Macpherson, R., and Miller, J. D. A. (1964). "The effect of Inhibitors on Sulfate Reducing Bacteria: a Compilation." Journal of Applied Bacteriology, 27: 281-293. Selene, C. H. and Chou, J. (2003). Hydrogen Sulfide: Human Health Aspects . Geneva, World Health Organization. Smith, R. L. and Klug, M. J. (1981). "Electron Donors Utilized by Sulfate-Reducing Bacteria in Eutrophic Lake Sediments." Applied and Environmental Microbiology, 42: 116-121. Taylor, B. F. and Oremland, R. S. (1978). "Depletion of Adenosine Triphosphate in Desulfovibrio by Oxyanions of Group VI Elements." Current Microbiology, 3: 101-103. Tchobanoglous, G., Theisen, H., and Vigil, S. (1993). Integrated Solid Waste Management: Engineering Principles and Management Issues . McGraw-Hill, New York.

PAGE 138

125 Teer, J. E., Leak, D. J., Dudency, A. W. L., Narayanan, A., and Stuckey, D. C. (1997). "The Treatment of Iron Oxalate Leach Liquors in a USAB with Sulfate Reduction." Water Science and Technology, 36: 383-390. Townsend, T., Chadik, P., Bitton, G., Booth, M., Lee, S., and Yang, K. (2000). "Gypsum Drywall Impact on Odor Production at Landfills: Science and Control Strategies." 00-09, Florida Center for Solid Waste and Hazardous Waste Management, Gainesville. Townsend, T., Jambeck, J., Jang, Y., Plaza, C., Xu, Q., and Clark, C. (2005). C & D Waste Landfill in Florida: Assessment of True Impact and Exploration of Innovative Control Techniques. Florida Center for Solid and Hazardous Waste Management. U.S. Environmental Protection Agency. (1995). " Construction and Demolition Waste Landfills." U.S. EPA 68-W3-0008. U.S. Environmental Protection Agency. (1998). "Characterization of Building-Related Construction and Demolition Debris in the United States." U.S. EPA 530-R-98-010. Vyskocil, A. and Viau, C. (1999). " Assessment of Molybdenum Toxicity in Humans." Journal of Applied Toxicology, 19: 185-192. Wiegel, J., Hanel, J., and Aygen, K. (2003). "Chemolithoautotrophic Thermophilic Iron(III)-Reducer". In: Biochemistry and Physiology of Anaerobic Bacteria . Ljungdahl, L., Adams, M., Barton, L., Ferry, J. and Johnson, M. Edition. 235-251. Widdel. F. (1986). Sulphate-reducing Bacteria and Their Ecological Niches. In: Anaerobic Bacteria in Habitats other than Man . Barnes. E.M and Mead. G.C Edition. 157-184. Oxford. Williams, G. M., Ward, R. S., and Noy, D. J. (1999). "Dynamics of Landfill Gas Migration in Unconsolidated Sands." Waste Management Research, 17: 327-342. W.H.O. (2000). Hydrogen Sulfide. Air Quality Guidelines for Europe, 2 nd Edition . World Health Organization Regional Office for Europe. Copenhagen, Denmark, WHO Regional Publications. Xu, Q., Townsend, T. (2004). “Hydrogen Sulfide Emission and Dispersion at Construction and Demolition Debris Landfills.” Proceedings of the nineteenth international conference on solid waste technology and management, Philadelphia, PA. 1263-1272. Yang, K. J. (2000). Hydrogen Sulfide Generation in Simulated Landfill Columns. Department of Environmental Engineering and Sciences. University of Florida. Gainesville, Master.

PAGE 139

126 Yang, Y. (1992). Biofiltration for Control of Hydrogen Sulfide . Department of Department of Environmental Engineering and Sciences. University of Florida.Gainesville. Ph.D. Zobell, C.E. (1958). "Ecology of Sulfate-Reducing Bacteria." Producers Monthly, 22: 12-29.

PAGE 140

BIOGRAPHICAL SKETCH Qiyong Xu was born in Chongqing, China, on March 6 th , 1975, to Xianwei Xiong and Daizhu Xu. He obtained his bachelor’s degree in environmental engineering from Nanchang Institute of Aeronautical Technology in 1996, and received his master’s degree in environmental sciences from Sichuan University in 2000. He was accepted as a graduate student in the Department of Environmental Engineering Sciences at the University of Florida in fall 2001 and started his doctoral research in solid and hazardous waste under the direction of Dr. Timothy Townsend. He was married to Jiaoju Ge on May 21 st , 2002. 127