Long-Term Treatment and Disposal of Landfill Leachate Year 1

Transcription

1 Long-Term Treatment and Disposal of Landfill Leachate Year 1 Debra R. Reinhart Department of Civil and Environmental Engineering College of Engineering and Computer Science University of Central Florida PO Box Orlando, FL July 31, 2005

2 TABLE OF CONTENTS LIST OF FIGURES...ii LIST OF TABLES...iv LIST OF ABBREVIATIONS, ACRONYMS AND UNITS OF MEASURE...v KEYWORDS...vii ABSTRACT...vii EXECUTIVE SUMMARY...viii 1.0 INTRODUCTION AMMONIA-NITROGEN REMOVAL REMOVAL OF RECALCITRANT ORGANICS PROJECT OBJECTIVES Ammonia-Nitrogen Removal Removal of Recalcitrant Organics EXPERIMENTAL METHODS IN-SITU AMMONIA REMOVAL Waste Acclimation Process Microcosm Studies Analytical Techniques CHEMICAL OXIDATION Leachate Collection and Characterization Molecular Weight Distribution Method Reagents Oxidation/Coagulation Experiments RESULTS AND DISCUSSION IN-SITU AMMONIA REMOVAL Waste Acclimation Process Microcosm Studies Ammonia Removal at 22 o C and 100% Oxygen in Both Acclimated and Unacclimated Wastes Impact of Oxygen on Ammonia Removal CHEMICAL OXIDATION Reaction Time Effect of ph Dose Nature of Intermediate Products CONCLUSIONS IN-SITU AMMONIA REMOVAL CHEMICAL OXIDATION SECOND YEAR OBJECTIVES REFERENCES...34 i

3 LIST OF FIGURES Figure 1. Waste acclimation process Figure 2. Waste acclimation process schematic...8 Figure 3. Microcosm bags...9 Figure 4. The ultrafiltration cell used for molecular weight measurements...14 Figure 5. The oxidation/coagulation experiments Figure 6. Ammonia-nitrogen removal in all microcosm studies at different oxygen concentrations Figure 7. Rate data for the 0.7, 4.0 and 100% oxygen tests fit to the double Monod equation..18 Figure 8. Rate of ammonia removal at different gas-phase oxygen concentrations...19 Figure 9. Effect of ph on organic removal from leachate using Fenton s treatment (a) measured as COD and (b) measured as DOC. Using a dose of 1g H 2 O 2 : g COD, and a molar ratio of 0.4 Fe 2+ : H 2 O Figure 10. Effect of ph on organics removal from leachate using ferrate treatment (a) measured as COD and (b) measured as DOC. Using a dose of 0.7 g Fe 6+ : g COD...23 Figure 11. Effect of iron dose on organics removal from leachate using Fenton s treatment (a) measured as COD and (b) measured as DOC. Using a ph of 4 and a dose of 1g H 2 O 2 : g COD...25 Figure 12. Effect of oxidant dose on organics removal from leachate using Fenton s treatment (a) measured as COD and (b) measured as DOC. Using a ph of 4 and a molar ratio of 0.4 Fe 2+ to H 2 O Figure 13. Effect of oxidant dose on organics removal from leachate using ferrate treatment (a) measured as COD and (b) measured as DOC. Using a ph of 4 and a dose of 0.7 g Fe 6+ : g COD...27 Figure 14. Molecular weight distribution before and after two treatments doses of each Fenton s reagent and ferrate...29 Figure 15. Dissolved organic matter COD/DOC ratio changes after ferrate and Fenton s reagent oxidation...29 ii

4 Figure 16. Changes in the BOD 5 /COD ratio for the 20 years old leachate after Fenton s oxidation using different oxidant doses...30 Figure 17. Changes in the BOD 5 /COD ratio for the 20 years old leachate after ferrate oxidation using different oxidant doses...30 iii

5 LIST OF TABLES Table 1. Values of leachate parameters for samples used in the oxidation experiments...20 iv

6 LIST OF ABBREVIATIONS, ACRONYMS AND UNITS OF MEASURE Abbreviation, Acronym or Unit of Measure Definition % percentage ε Extinction coefficient = 1150 M -1 cm -1 A Absorbance (at 510 nm) Amicon YM Brand of ultrafiltration membranes BOD Biochemical oxygen demand BOD 5 Five-day biological oxygen demand C Carbon C Concentration (M) CaCO 3 Calcium carbonate Cl - chloride C N Ammonia-nitrogen concentration COD Chemical oxygen demand C p Instantaneous solute concentration in permeate at time t C ro the initial concentration of the solute molecules with an apparent molecular weight smaller than the membrane cutoff. D Diameter DI deionized DOC Dissolved organic carbon ECD Electron capture detector F 1 - (V r /V o ) = the fractional reduction in retenate volume at time t Fe 2+ ferrous Fe 6+ Ferrate iron FeCl 3 Ferric chloride FeSO 4.7H 2 O Ferrous sulfate g grams GC Gas chromatograph H + Hydrogen H 2 O water H 2 O 2 Hydrogen peroxide hrs hours HS - Sulfide k Specific rate constant kg kilograms K I Inhibition constant K O Oxygen half-saturation constant K s Ammonia-nitrogen half-saturation constant L Liters L Cell path length = 1 cm m meters M Molar v

7 mg N/L mg/l min ml MSW MW N N 2 Na 2 FeO 4 NaCl NaOCl NaOH NH 3 nm NO 2 NO 3 o C OCl - P PCM R r 2 SO 4 t TCD um V V o V r yrs Milligrams nitrogen per liter Milligrams per liter minute milliliters Municipal solid waste Molecular weight nitrogen Nitrogen gas Sodium ferrate Sodium chloride Sodium hypochlorite Sodium hydroxide Ammonia Nano-meters Nitrite Nitrate Degrees Celsius hypochlorite Permeation coefficient permeation coefficient model Rate of ammonia removal Correlation coefficient sulfate Time Thermal conductivity detector Micrometer Volume Initial volume used Volume of retenate at time t years vi

8 KEYWORDS: Bioreactor landfills, in situ nitrification, denitrification, chemical oxidation, leachate treatment, Fenton s reagent ABSTRACT: Operating the landfill as a bioreactor is a very advantageous solid waste management option. However, when operating bioreactor landfills, challenges remain, including the persistence of ammonia-nitrogen in the leachate and remaining complex organics. It is likely that these parameters will determine when the landfill is biologically stable and when post-closure monitoring may end. By addressing and finding solutions to existing challenges, bioreactor landfill technology can be furthered, becoming an even more advantageous waste management option. Laboratory-scale studies were conducted to evaluate removal of both ammonia (via in situ nitrification and denitrification) and remaining complex organics (via chemical oxidation techniques). In-situ nitrification and denitrification were evaluated by operating two laboratory studies: a waste acclimation process and microcosm studies. The waste acclimation process was dual purpose, to provide an acclimated waste source for parallel batch microcosm studies and to demonstrate the efficacy of in-situ nitrification and denitrification. The microcosm studies were smaller scale experiments conducted to evaluate the kinetics of ammonia removal. Waste from the waste acclimation process is the waste source used in the microcosm studies. The microcosm experiments were aimed at determining nitrification and denitrification rates under different environmental conditions (i.e. oxygen levels and waste acclimation). The treatment of recalcitrant organics portion of the project consisted of laboratory-scale studies aimed at determining a set of oxidation conditions (time, ph, and dose) for both Fenton s reagent and ferrate that yielded the maximum organics removal from mature leachate for each treatment method. Additionally, an evaluation of changes in the nature of intermediate organics produced by oxidation using various gross organic parameters was conducted. In-situ nitrification experiments were conducted demonstrating that ammonia removal via nitrification and denitrification is feasible in decomposed solid waste environments at various oxygen concentrations. Monod expressions were derived from ammonia removal data to describe removal in both acclimated and unacclimated wastes. Additionally, a relationship depicting the effect of oxygen on ammonia removal was found and shows that the highest rates of ammonia removal will occur at levels just below atmospheric oxygen levels, which may be typically found in older waste environments. Results also suggest that nitrification and denitrification may occur simultaneously in one aerobic landfill cell (under low biodegradable conditions), rather than requiring two separate cells containing two different in-situ environments, which is significant when developing field-scale guidance for implementation of such processes. In the treatment of refractory organics portion of this work, two oxidation methods (Fenton and ferrate) were tested for treatment of non-biodegradable leachate organics. Results from this experiment include optimum oxidation conditions for both treatment methods. The results also indicated that both Fenton and ferrate had high organics removal efficiencies. However, Fenton was superior in removing both COD and DOC and in increasing the biodegradability of organics remaining in the leachate, which could be seen in the higher BOD 5 /COD produced. vii

9 EXECUTIVE SUMMARY July 2004 to July 2005 PROJECT TITLE: Long-Term Treatment and Disposal of Landfill Leachate Year 1 PRINCIPAL INVESTIGATOR: Debra R. Reinhart AFFILIATION: University of Central Florida COMPLETETION DATE: July 31, 2005 OBJECTIVES: This research investigated in-situ nitrification and denitrification processes in solid waste environments and the treatment of refractory organics in mature landfill leachate. In-situ nitrification kinetics studies were conducted to determine ammonia-nitrogen removal kinetics under different environmental conditions to aid in developing a field-scale implementation strategy. Further, this project explored the treatment of refractory organics in mature landfill leachate (mainly humic materials and xenobiotic organic compounds) by means of combined external partial chemical oxidation followed by biological treatment within the landfill. Laboratory experiments were conducted to evaluate and optimize oxidation techniques (Fenton s reagent and ferrate) in preparation for the evaluation of the in-situ biodegradation of oxidized leachate. METHODOLOGY: These objectives were met by conducting laboratory studies. In-situ nitrification and denitrification were evaluated by operating two laboratory studies: a waste acclimation process and microcosm studies. The waste acclimation process was dual purpose, to provide an acclimated waste source for parallel batch microcosm studies and to demonstrate the efficacy of in-situ nitrification and denitrification. The microcosm studies were smaller scale experiments conducted to evaluate the kinetics of ammonia removal. Waste from the waste acclimation process is the waste source used in the microcosm studies. The microcosm experiments were carried out in 3 and 10-L containers aimed at determining nitrification and denitrification rates under different environmental conditions (i.e. oxygen levels and waste acclimation). The treatment of recalcitrant organics portion of the project consisted of laboratory-scale studies aimed at determining a set of oxidation conditions (time, ph, and dose) for both Fenton s reagent and ferrate that yielded the maximum organics removal from mature leachate for each treatment method. Additionally, an evaluation of changes in the nature of intermediate organics produced by oxidation using various gross organic parameters was conducted. RATIONALE: Operating the landfill as a bioreactor is a very advantageous solid waste management option. However, when operating bioreactor landfills, challenges remain, including the persistence of ammonia-nitrogen in the leachate and remaining complex organics. It is likely that these parameters will determine when the landfill is biologically stable and when postclosure monitoring may end. By addressing and finding solutions to existing challenges, viii

10 bioreactor landfill technology can be furthered, becoming an even more advantageous waste management option. RESULTS AND CONCLUSIONS: In-situ nitrification experiments were conducted demonstrating that ammonia removal via nitrification and denitrification is feasible in decomposed solid waste environments at various oxygen concentrations. Monod expressions were derived from ammonia removal data to describe removal in both acclimated and unacclimated wastes with specific rates of removal of and mg N/day-g dry waste and half-saturation constants of 59.6 mg N/L and 147 mg N/L for acclimated and unacclimated wastes, respectively. Based on the ammonia removal rates measured in these studies, unacclimated solid waste environments do not appear to be an issue, as ammonia will still be removed at high rates. Additionally, a relationship depicting the effect of oxygen on ammonia removal was found and shows that the highest rates of ammonia removal will occur at levels just below atmospheric oxygen levels, which are typically found in older waste environments. Results also suggest that nitrification and denitrification may occur simultaneously in one aerobic landfill cell (even under low biodegradable C:N conditions), rather than requiring two separate cells containing two different in-situ environments (i.e. anoxic and aerobic), which is significant when developing field-scale guidance for implementation of such processes. Also, demonstrating denitrification can occur in older portions of the landfill provides valuable insight as to where nitrification and denitrification can occur in landfills. In the treatment of refractory organics portion of this work, two oxidation methods (Fenton and ferrate) were tested for treatment of non-biodegradable leachate organics. Results from this experiment include optimum oxidation conditions for both treatment methods. For ferrate a time of 60 min, a ph range of 2 to 5, and a dose of 0.7 g Fe:g COD were found to be optimal. For Fenton, a time of 60 min, a ph range of 2 to 5, and a dose of 1 gh 2 O 2 :g COD were optimal. The results also indicated that both Fenton and ferrate had high organics removal efficiencies. However, Fenton was superior, removing 69 to 79% of the COD and 65 to 78% of the DOC compared to 54 to 56% of the COD and 59% of the DOC removed by ferrate. Fenton was also superior to ferrate in increasing the biodegradability of organics remaining in the leachate, which could be seen in the higher BOD 5 /COD produced by Fenton (0.23) to that produced by ferrate (0.08). Coagulation that occurred simultaneously with oxidation in both ferrate and Fenton was shown to remove almost 65 to 70% of the total COD removed by either of the methods. Finally, Fenton showed that it causes more partial oxidation than ferrate, which had more total oxidation properties. ix

11 1.0 INTRODUCTION A new and promising trend in solid waste management is to treat the landfill as a bioreactor. Bioreactor landfills are controlled systems in which moisture control and/or air injection are used as enhancements to create a solid waste environment capable of actively degrading the readily biodegradable organic fraction of the waste. Moisture control, usually accomplished via leachate recirculation, can be an effective leachate treatment method. When recirculating leachate, the leachate is treated in situ, often resulting in a rapid reduction of both biodegradable organic compounds and heavy metals, thus serving as a viable and potentially long-term treatment method. However, treatment challenges remain, including the persistence of ammonia-nitrogen and humic and xenobiotic compounds in the leachate. These challenges must be addressed before bioreactor landfills are an accepted means for long-term leachate management. Ammonia-nitrogen levels found in leachate from bioreactor landfills are much higher than those found in conventional landfills. Although recirculating leachate results in decreasing organic compounds, it results in an increase in ammonia-nitrogen concentrations. Recirculating leachate stimulates microbial activity, increasing the rate of ammonification (release of ammonia from the solid to the liquid phase). Additionally, an accumulation of higher levels of ammonia-nitrogen concentrations occurs as leachate with already high ammonia concentrations is reintroduced to the landfill. Refractory organics are also problematic in older leachates. Although the biodegradability of leachate organic compounds declines with time in bioreactor landfills, complex organic compounds such as humic substances and xenobiotic compounds, remain in solution. Thus, leachate, with time, will require a complex sequence of biological, physical, and chemical treatment processes to reach discharge limits. It is likely that ammonia-nitrogen, humics, and xenobiotic compounds will determine when the landfill is biologically stable and when post-closure monitoring may end. Treating these compounds effectively in-situ would be very advantageous, potentially resulting in both environmental and economic advantages. As a result of removal of ammonium-nitrogen and humic and xenobiotic compounds, post-closure monitoring would be reduced (resulting in economic benefits), as would the potential impact to the environment. Additionally, by treating these compounds in-situ, a long-term treatment method for landfill leachate results, allowing for landfill operators to continually treat their leachate on-site, as well as operating their landfills as bioreactors and realizing the numerous advantages associated with bioreactors. This research project involves conducting laboratory-scale research examining the capacity of the landfill to provide complete in-situ treatment of landfill leachates, focusing on the in situ removal of ammonia-nitrogen and combined external partial chemical oxidation of refractory organics followed by biological treatment within the landfill. 1

12 1.1 AMMONIA-NITROGEN REMOVAL The nitrogen content of municipal solid waste (MSW) is less than 1 %, on a wetweight basis (Tchobanoglous et al., 1993), and is composed primarily of the proteins contained in yard wastes, food wastes, and biosolids (Burton and Watson-Craik, 1998). As the proteins are hydrolyzed and fermented by microorganisms, ammonia-nitrogen is produced. This process is termed ammonification. Researchers report that as much as 20% of the nitrogen will leach from the waste (Institute Of Waste Management Sustainable Landfill Working Group, 1999), yielding concentrations ranging from less than detection levels to over 5,000 mg/l (Reinhart and Townsend, 1998; Ehrig, 1989; Grosh, 1996; Qasim and Chiang, 1994). Leachate composition is quite variable, depending on waste composition, moisture content of the waste, and age of the landfill. Removal of ammonia-nitrogen from leachate to low levels is necessary because of its aquatic toxicity and oxygen demand in receiving waters. Bioreactor landfills result in higher ammonia-nitrogen concentrations than found in conventional landfills because leachate recirculation increases the rate of ammonification and results in accumulation of higher levels of ammonia-nitrogen concentrations. Currently, ammonia treatment in leachate removed from landfills is primarily practiced ex-situ via biological co-treatment at publicly owned treatment works or on site treatment via biological nitrification/denitrification processes. The additional costs associated with ex-situ treatment of ammonia have made in-situ removal techniques attractive alternatives. A few in-situ, or partially in-situ studies have been conducted, however, the data required to enable adequate implementation of such processes at fieldscale bioreactor landfills are lacking. These include external nitrification and in-situ denitrification (Barlaz, 2001; Jokela et al., 2002; Aljarallah and Atwater, 2000; Burton and Watson-Craik, 1999; Price, 2001; and waste management, 2001), incidental treatment in aerobic or semi-aerobic landfills (Hanashima, 1999), and establishing in-situ dedicated treatment zones (Onay and Poland, 1998; Onay, 1995). Burton and Watson-Craik (1999) operated a landfill test cell designed to denitrify externally nitrified leachate. Nitrate returned to the landfill cell was efficiently consumed under the anoxic/anaerobic landfill conditions, confirmed using labeled isotopic nitrate. Both Waste Management (2001) and Aljarallah and Atwater (2000) have completed similar studies at field and laboratory-scale, respectively. Aljarallah and Atwater (2000) noted that denitrification was feasible in a bioreactor landfill, however it hindered waste degradation and thus gas production. Additionally, it was noted that poor leachate quality resulted (high organic strength). Jokela et al. (2002) conducted a similar laboratory study demonstrating in-situ denitrification is possible and can result in total oxidized nitrogen concentrations below detection limits. Barlaz (2001) and Price (2001) conducted similar studies and found that fresh waste contained enough organic carbon to denitrify, while older waste required the addition of an external carbon source. Youcai et al. (2002) conducted a study in which a biofilter consisting of old waste (8 to10 years old) was used to treat leachate. Aerobic portions existed at the top and 2

13 bottom of the system, while the middle of the system was anaerobic. An incidental removal of 99.5% of the ammonia in leachate was observed, as were elevated concentrations of nitrate and nitrite, indicating the ammonia was converted biologically. Additionally, 20-30% of total nitrogen was removed suggesting in-situ nitrogen removal can occur in solid waste systems. The most advantageous method evaluated is complete in-situ removal of nitrogen using dedicated zones. Onay and Pohland (1998) completed an in-situ nitrification/denitrification laboratory study in which a three-component system was used to facilitate the process. Reactors containing anoxic, anaerobic, and aerobic conditions were operated. When utilizing leachate recirculation among the zones, approximately 95% nitrogen removal was achieved, suggesting in-situ nitrification and denitrification readily occurred in-situ. 1.2 REMOVAL OF RECALCITRANT ORGANICS This project will also explore the treatment of refractory organics in mature landfill leachate (mainly humic materials and xenobiotic organic compounds) by means of combined external partial chemical oxidation followed by biological treatment within the landfill. Removal of refractory organics is important because humic materials increase the mobility of groundwater contaminants such as heavy metals cadmium, nickel, and zinc (Christensen, et al., 1996). Also, humic materials are considered to be the principal organic precursors for trihalomethanes (potential carcinogens in drinking water, Reckhow, (1990)) and many xenobiotic compounds are suspected carcinogens (Chiou et al., 1986). Removal of refractory organics requires the use of a chemical method. During the last stage of landfill stabilization the biodegradability of the leachate decreases, as organics in leachate are mostly recalcitrant. Indicators of this decrease in biodegradability include low 5-day biological oxygen demand (BOD 5 )/chemical oxidation demand (COD) ratios (below 0.1), high molecular weight (MW, >1000dalton), and high concentrations of humic-like material (60% of dissolved organic carbon (DOC), Kjeldsen et al., 2002). Chemical oxidation has been used in previous studies to increase the biodegradability of organics extracted from different types of wastewaters and landfill leachates, as well as to increase the overall biodegradability of each waste stream. Humic acids extracted from leachate has been converted to more biodegradable compounds (BOD/COD increased from to 0.28) using an advanced chemical oxidation method (gamma-ray irradiation). During the oxidation, gamma-ray irradiation produced hydroxyl free radicals, the element responsible for all advanced oxidation reactions (Yamazaki et al., 1983). Amador et al. (1989) found that advanced oxidation also increased microbial mineralization and decreased molecular size of complex molecules consisting of synthetic organic compounds and humic acids, similarly to those likely to occur in leachate. Also, Wang (1992) showed that chemical oxidants such as hydrogen peroxide, potassium permanganate, and ozone increased anaerobic biodegradability (measured as biochemical methane potential) of phenolic compounds. Studies using chemical oxidation 3

14 to increase the overall biodegradability of leachate have also been conducted. Geenens et al. (2000) used ozonation to increase BOD/COD of landfill leachate from 0.06 to Also, Gonze et al. (2003) applied high frequency ultrasound, increasing BOD/COD of leachate from 0 to Other studies, such as Beltran et al. (1999) and Benitez et al. (1999), reported similar results. Removing refractory organics by means of combined external partial chemical oxidation followed by biological treatment within the landfill is expected to promote safe discharge of leachate into natural bodies by removing humic and fulvic acids and xenobiotic organics. As an in-situ treatment technique, this method is expected to reduce the cost and environmental risk of transportation and ex-situ treatment of leachate. Additionally, because recalcitrant organics are removed, post-closure costs associated with monitoring of groundwater, surface water, and air will decline, as will long-term environmental risks from leachate contamination. Finally, there is a potential for a methane gas production increase in landfills due to biological degradation of previously recalcitrant organics. This technique would be applied at a time when methane generation is low because waste has been biostabilized. Assuming the gas is collected and utilized, economic advantages may be realized. 1.3 PROJECT OBJECTIVES This research project proposes to conduct laboratory-scale research to examine the capacity of the landfill to provide complete in-situ treatment of landfill leachates so that they may be released to the environment without adverse impact. There are two main components of this project: the evaluation of (1) ammonia-nitrogen removal and (2) refractory organics destruction Ammonia-Nitrogen Removal Currently, ammonia treatment in leachate removed from landfills is primarily practiced ex-situ using biological nitrification/denitrification processes. Nitrification and denitrification processes are advantageous nitrogen removal processes as complete destruction of nitrogen results. Because there are additional costs associated with ex-situ treatment of ammonia, the development of an in-situ nitrogen removal technique would be an attractive alternative, potentially yielding both economic and environmental advantages such as eliminating the need for continuous leachate treatment and allowing the landfill to become a more sustainable system. Thus, this research will investigate insitu nitrification processes in solid waste environments, allowing for a more informed approach to designing and operating bioreactor landfills by addressing the following: Determination of optimal environmental conditions and expected efficiencies for in-situ nitrification and denitrification Development of an implementation strategy for in-situ nitrification and denitrification at field-scale, including an economic comparison with ex-situ treatment approaches 4

15 With a main goal of this portion of the project to develop field-scale guidance for implementation of in-situ nitrogen removal, specific data necessary for field-scale implementation will be determined during this study. Operational parameters such as waste age, the amount of oxygen necessary, allowable leachate recirculation rates, and any potential needs for ph control will be determined from the experiments. Additionally, particular attention will be paid to potent greenhouse gas emissions, such as N 2 O, as production will contribute to the global climate change problem. The results from the laboratory studies will be used to develop a strategy for a subsequent pilot-scale study at the Florida Bioreactor Landfill Demonstration at the New River Regional Landfill, as well as for future field-scale implementation. Specific tasks completed during year one of this project include: Commencement of waste acclimation process (serves as waste source for kinetic studies) Evaluation of nitrification kinetics at room temperature Evaluation of effect of oxygen concentrations on ammonia removal at room temperature Data analysis and evaluation of nitrification kinetic studies Development of a preliminary operation strategy to implement in-situ nitrification in full-scale bioreactor landfills Removal of Recalcitrant Organics Treatment of refractory organics in mature landfill leachate by means of combined external partial chemical oxidation followed by biological treatment within the landfill will be investigated. Objectives of this research include providing data on mature leachate oxidation by ferrate and Fenton used on two leachates (aged 12 and 20 years). These data will present a comparison between advanced oxidation and direct oxidation techniques and will also explore the nature of oxidation intermediates produced by each of the oxidants. The approach proposed is an in situ treatment technique. Therefore there will be no need to transport the leachate to an external treatment plant, which reduces cost and potential environmental impacts. It is anticipated that dosages required for partial oxidation would be a fraction of that needed for complete oxidation and that minimal equipment and structures would be required to implement the treatment process in landfills. Results of this research will facilitate the design of the full-scale application of the suggested technology and will include an economic analysis of the treatment scheme. Specific tasks completed during year 1 of this project include: 5

16 Determination of a set of oxidation conditions (time, ph, and dose) that will yield maximum organics removal from mature leachate using Fenton s reagent. Determination of the set of conditions that will yield maximum organics removal from mature leachate using ferrate Comparison between the efficiencies of Fenton and ferrate in terms of COD and DOC removal Evaluation of the changes in the nature of intermediate organics produced by oxidation using indicators including: BOD 5 /COD ratio, DOC/COD ratio and organics molecular weight (MW) distribution 6

17 2.0 EXPERIMENTAL METHODS 2.1 IN-SITU AMMONIA REMOVAL Two types of experiments were conducted: a waste acclimation process and microcosm studies. The waste acclimation process was dual purpose, to provide an acclimated waste source for parallel batch microcosm studies and to demonstrate the efficacy of in-situ nitrification and denitrification. In this study, acclimated waste is defined as waste that has been exposed to a nitrifying microbial population and is capable of removing ammonia concentrations as high as 1000 mg N/L. The microcosm studies were smaller scale experiments conducted to evaluate the kinetics of ammonia removal. Waste from the waste acclimation process is the waste source used in the microcosm studies Waste Acclimation Process A 133-L reactor (see Figure 1) was designed to allow for leachate draining and recirculation, air addition, and gas sampling. To prevent clogging of the leachate drain, a layer of gravel was placed at the bottom of the reactor. Digested municipal solid waste (MSW) was obtained from an aerobic MSW compost facility located in Sumter County, Florida, USA. A mixture of 14 kg of digested MSW (i.e. compost) and 0.6 kg of approximately 4-cm long wood chips was added to the reactor to promote air distribution throughout the matrix. Initially, 6 L of deionized (DI) water and 2 liters of mixed liquor from a local nitrifying wastewater plant were added to the system to initiate leachate production and aid in developing a nitrifying microbial population. All liquid was immediately drained and 3.95 liters of the leachate were recirculated. Moisture saturated air was added to both the bottom and middle of the compost matrix at a total rate of 2.77 L/min. The air was saturated with moisture prior to introduction to the reactor to replenish any water lost due to evaporation and was added continuously throughout the duration of the study. After two days, leachate was drained and then recirculated every two to three days. Samples were routinely removed and analyzed for ph, chemical oxygen demand (COD), alkalinity, sulfate, nitrate, nitrite, and ammonia-nitrogen to evaluate whether nitrification or other nitrogen removal processes were occurring. Two liters of the collected leachate were recirculated; if needed, tap water was also added to bring the recirculated volume to two liters. Digested MSW samples were periodically removed from the reactor and used in parallel kinetic studies. Figure 2 presents a schematic depicting how the waste acclimation process is operated. Additional digested MSW was added to the reactor as needed to maintain the same waste mass. 7

18 Figure 1. Waste acclimation process. MLSS Ammonium Bicarbonate Off Gases NH 3 Microcosms Air Injection Digested MSW ~15 kg Acclimated Waste Leachate COD, ph, NO 2, NO 3, NH 3, SO 4 Figure 2. Waste acclimation process schematic. 8

19 2.1.2 Microcosm Studies Microcosm experiments were conducted at various oxygen levels (100, 16.6, 4.5, 4.0 and 0.7%), at room temperature (22 o C) and on acclimated waste. One set of experiments were conducted on unacclimated waste to determine the impact of the waste acclimation process on ammonia-nitrogen removal. Ammonia removal kinetics were calculated for each test and fit to the Monod equation. Microcosm experiments were conducted in foil gas sampling bags (SKC, Inc., Pennsylvania) modified to permit addition of solid components to the bag, as shown in Figure 3. Bag size was dependent on the type of experiment being conducted. Tests with 100% oxygen were conducted in 3-L bags (Figure 3a), while tests with lower oxygen concentrations were conducted in 10-L bags (Figure 3b). A gas sampling port was located at the top of each bag. Each microcosm was loaded with 200 g of acclimated, digested MSW taken from the aerobic reactor along with 20 g of wood chips. A volume of 15 ml of ammonium bicarbonate solution (concentration varied depending on target ammonia level) was added to each system resulting in moisture levels at field capacity (approximately 63% on a wet-weight basis). After each bag was loaded with digested MSW and ammonia, the bags were purged with helium and then filled with the appropriate oxygen concentration. Studies were conduced at average gas-phase oxygen concentrations of 100, 16.6, 4.5, 4.0 and 0.7%. Gas mixtures consisted of oxygen and helium. Each bag was sealed to allow for all nitrogen products to be controlled, captured, and measured. Both liquid- and gas-phase samples were taken from the bags over time and percent recoveries of nitrogen species were calculated. The microcosms were inverted once each day to simulate leachate recirculation. (a) Three-liter microcosm bag Figure 3. Microcosm bags. (b) Ten-liter microcosm bag 9

20 Three series of experiments were conducted; the first series was conducted on acclimated waste at constant temperature (22 o C), with 100% oxygen initially in the headspace, and with varying ammonia loadings (200, 500 and 1000 mg/l-n). Ammonia removal rates were quantified for each experiment. The second series of experiments was conducted to evaluate the effect of waste acclimation on ammonia removal rates. The experiments were conducted using unacclimated waste at 22 o C, with 100% oxygen initially in the headspace and with an ammonia concentration of 500 mg/l-n for comparison. The third set of experiments was conducted with acclimated waste at 22 o C and 500 mg/l-n with varying oxygen concentrations (0.7, 4.0, 4.5 and 16.6%). Each microcosm set consisted of several different bags operated in a batch mode and each bag destructively sampled over time. Each bag was destroyed during each sample time because there was not sufficient leachate volume for repetitive samples and because the potential for sorption precluded the analysis of leachate alone. After disassembling each microcosm, a mild extraction procedure was conducted to desorb any ammonianitrogen or nitrate sorbed on the waste. In all bags, both liquid- and gas-phase parameters were measured, including ammonia-nitrogen (leachate and sorbed masses), nitrate (leachate and sorbed masses), nitrite, sulfate, ph, nitrogen and nitrous oxide gases, and gas-phase oxygen. Nitrogen and nitrous oxide were measured because there may be micro-anoxic areas in which denitrification is occurring, resulting in the production of nitrous oxide and nitrogen gas. Nitrous oxide was not measured in all studies because of initial procedural issues Analytical Techniques COD, ph, and alkalinity were measured using methods found in Standard Methods (1995). Ammonia-nitrogen in the leachate was measured using an ion-specific electrode (Fisher Scientific, Inc.) via the known-addition method. The form of ammonianitrogen present in solution is dependent on the solution ph. In all studies conducted, the ph was below 8.0, indicating the dominant form of ammonia is ammonium. All anions, including nitrite, nitrate, and sulfate, were measured using a DX-120 ion chromatograph (Dionex, Inc.) equipped with an AS-14 column and a bicarbonate/carbonate eluent. Prior to anion analysis, all samples were centrifuged and filtered using a 0.45-micron nitrocellulose filter. In the microcosm studies, a mild extraction procedure was used to measure any ammonia-nitrogen and nitrate that was sorbed to the waste matrix. After disassembling the microcosms, 210 ml of DI water were added to each system and the microcosms were subsequently drained. The drained leachate was analyzed for ph, nitrite, nitrate, and sulfate. Three hundred ml of a 0.5-M sodium sulfate were added to each system to desorb ammonia-nitrogen and nitrate. The system was shaken at 60rpm for 1.5-hours and then subsequently drained and ammonia and nitrate concentrations were measured. Preliminary work was completed to ensure the volume and concentration of sodium sulfate and shaking time were sufficient. Ammonia was measured using an ion specific 10

21 electrode. Nitrate in the extract was measured using a nitrate-selective electrode (Fisher Scientific, Inc.). The electrode was used in place of the ion chromatograph for measuring nitrate in the extract because of an interference with sulfate. A known-addition method ensured no interferences from the sample matrix influenced the measured concentration. Oxygen, nitrogen and methane in the headspace were analyzed using a gas chromatograph (GC) (Shimadzu, Inc) equipped with a TCD detector and two packed 13X molecular sieve columns (Alltech Associates, Inc.) in series (to achieve adequate separation of the oxygen and nitrogen peaks). The injector, detector, and oven temperatures were 25 o C. Nitrous oxide was measured with a GC (Varian, Inc.) equipped with an electron capture detector (ECD) in conjunction with an AT-Q 30-m capillary column (Alltech Associates, Inc); the injector, detector, and oven temperatures were 110 o C, 110 o C, and 60 o C, respectively. In each microcosm, gas volume in the headspace was measured using a water displacement technique. Each bag was submerged in water and the volume of water displaced measured. The gas volume was equal to the displaced water volume corrected the volume of the waste and liquid. The pressure applied to the bags was minimal and thus neglected. Solid-phase organic nitrogen was measured using a modified macro-kjeldahl method (Standard Methods, 1995). Modifications to the Standard Method include the use of a larger volume of digestion reagent (100 ml) and a longer acid hydrolysis step (6-hrs) during which ground solid samples were shaken in 200 ml of solution prior to digestion; 100 ml of digestion solution and 100 ml DI. Moisture content was measured by drying solids in an oven at 105 o C for 24 hours. Volatile solids were measured by heating 1 g ground waste samples at 550 o C for 2 hours. 2.2 CHEMICAL OXIDATION Leachate Collection and Characterization Leachate was collected from two Florida landfills that had older, lined cells (12 and 20 years old). Samples were collected from manholes of leachate collection systems and were kept in one L amber glass bottles with no head space at 4 o C until used. Samples collected were characterized for COD, DOC, BOD 5, ph, alkalinity, ammonia, and chloride according to Standard Methods (1995). Organic compound MW was determined using ultrafiltration according to Logan and Jiang (1990) Molecular Weight Distribution Method Ultrafiltration was used to determine the MW distribution of dissolved organic matter. An Amicon model 8050 stirred cell (D = 44.5 mm, V = 50 ml), which can be 11

22 seen in Figure 4 was used with YM membranes of 1000, 10,000, 30,000, and 100,000 dalton membrane cut offs. These MW measurements were conducted in the parallel mode with a sample size of 50 ml for every membrane cut off. Membranes used in ultrafiltration are designed to keep all molecules larger than a certain size (membrane cut off) from permeating through. However this does not necessarily mean that all other smaller molecules will directly permeate through the membrane. What usually happens is that flow of the smaller molecules will be retarded due to accumulation of solute molecules at the membrane surface. To account for this phenomenon, Logan and Jiang (1990) suggested a permeation coefficient model (PCM) which is given in Equation 1. This model is obtained from a mass balance over the pressurized ultrafiltration cell. C p = PC ro F p 1 (1) Where, C p = Instantaneous solute concentration in permeate at time t P = Permeation coefficient C ro = the initial concentration of the solute molecules with an apparent molecular weight smaller than the membrane cutoff. F = 1 - (V r /V o ) = the fractional reduction in retenate volume at time t V r = Volume of retenate at time t V o = Initial volume used The goal here was to estimate C ro for every molecular weigh membrane cut off and that was done by converting Equation 5 to the linear form shown in Equation 2 below. lnc p = ln( PC ) + ( p 1) ln F (2) ro For a single point in a MW size distribution, an array of F versus Cp was determined experimentally by running the ultrafiltration unite and measuring the permeate DOC at three V r values (40, 30, and 20 ml). Finally, ln F and ln C p are calculated and plotted to get a linear line with a slope equals to (p-1) and an intercept equal to PC ro Reagents The sodium ferrate used in this study was prepared using a wet oxidation method (Thompson et al. 1951). In this method, ferrate is produced from the oxidation of ferric by hypochlorite in a strong basic solution as shown in Equation 3. Industrial grade reagents were used and were obtained from local companies. Sodium hypochlorite (13.9% OCL - by weight) was obtained from Odyssey manufacturing (Tampa, FL) and was kept at 4 o C in the dark until use. Sodium hydroxide (50% by weight) and ferric chloride (40% by weight) were obtained from Brentag Mid-South, Inc. (Tampa, Fl) and were kept at room temperature. 12

23 2FeCl 3 + 3NaOCl + 4NaOH Na 2 FeO 4 + 3NaCl + Cl - + H 2 O (3) Since ferrate is an unstable iron product, it is important to measure the concentration of the ferrate in the final produced solution. For this reason, ferrate concentration was measured in every batch produced and was used within an hour of making to minimize ferrate dissociation. Ferrate was measured using a spectroscopic technique. The absorbance of ferrate was measured at wave length of 510 nm then equation 4 was used to calculate the concentration of ferrate. Where, A = Absorbance (at 510 nm) ε = Extinction coefficient = 1150 M -1 cm -1 L = Cell path length = 1 cm C = Concentration (M) A C = (4) ε l All other reagents were obtained from Fisher scientific. For Fenton s reagent, analytical grade 40% H 2 O 2 and FeSO 4.7H 2 O were used Oxidation / Coagulation Experiments The oxidation experiments were conducted following a five-step process using 600 ml glass beakers (see Figure 5). The contents were mixed using stir plates at room temperature (22 o C) and atmospheric pressure. To remove suspended solids and to achieve more repeatability leachate was filtered using a 0.45 um filter. The ph of the leachate was adjusted to the target value. The oxidant was then added and the reaction was allowed to process. The ph was brought back to 7 and the leachate was filtered (0.45-um) to separate the organics removed by coagulation from the dissolved organics. A COD sample was taken from the initial filtered leachate, from the oxidized-unfiltered samples, and from the oxidized filtered samples to determine organics removal percentages by oxidation and by coagulation. However DOC could only be measured in two locations (after the initial filtration and after the final filtrations). Since oxidant residual will interfere with COD readings, residual H 2 O 2 was measured using hydrogen peroxide strips obtained from Fisher scientific to make sure the peroxide is all gone before the COD test. On the other hand, ferrate residual was determined to be completely gone by the color change from purple (color of ferrate) to brown (color of ferric). After the required oxidation, ph and dose of oxidants were selected. The intermediate organic compounds produced by oxidation were characterized by measuring the COD, DOC, organic compounds MW, and BOD 5 before and after oxidation. 13

24 Figure 4. The ultrafiltration cell used for molecular weight measurements. Figure 5. The oxidation/coagulation experiments. 14

25 3.0 RESULTS AND DISCUSSION 3.1 IN-SITU AMMONIA REMOVAL First, a literature review was conducted to determine and evaluate all potential nitrogen removal and transformation pathways that may occur in bioreactor landfills. Results of this literature review are in the form of a journal article accepted for publication in the journal Critical Reviews in Environmental Science and Technology (Berge et al, 2005a). Experiments were then conducted evaluating both the fate of nitrogen removal and the kinetics of ammonia removal in solid waste environments. The following sections describe the results obtained Waste Acclimation Process Detailed results describing the waste acclimation process have been published elsewhere (Berge et al., 2005b). This section is a summary of the important results obtained from this process. The waste acclimation process was operated for over 700 days and was periodically spiked with ammonium bicarbonate to ensure an active nitrifying and/or denitrifying microbial population was present. After each spike, ammonia-nitrogen readily disappeared, followed by an increase in nitrate and, at times, an increase in nitrite. Sulfate concentrations also increased after each ammonia-nitrogen spike. Although nitrate concentrations were measured, they were never as high as were stoichiometrically expected if nitrification was solely occurring, suggesting that both nitrification and denitrification were occurring within the system. Mass balances could not be conducted on this reactor because of inefficient gas capture. Additionally, although air was continuously added to the system, a sulfide odor was detected in the leachate, suggesting the presence of anaerobic pockets in the reactor. Additionally, each ammonia-nitrogen spike results in a nitrate increase coupled with an increase in sulfate concentration, suggesting a portion of nitrate removal may be attributed to autotrophic denitrification. Autotrophic denitrification follows reaction 5 and is favored in environments with a low C:N ratio in the presence of inorganic sulfur compounds, such as hydrogen sulfide (Koenig and Lui, 1996). 1/5 NO /8 HS - + 3/40 H + 1/10 N 2 + 1/8 SO /10 H 2 O (5) Onay and Pohland (2001) also suspected autotrophic denitrification was occurring in their laboratory study and was the reason for sulfate production. In the waste acclimation process, conditions were favorable to the autotrophic denitrifiers, as the COD of the leachate remained stable, suggesting the remaining concentration was recalcitrant, 15

26 and the BOD was low, around 1 mg/l. On average, during each spike, the sulfate concentrations measured suggest conversion of 10 to 15% of the nitrate via autotrophic denitrification Microcosm Studies Ammonia Removal at 22 o C and 100% Oxygen in Both Acclimated and Unacclimated Wastes Detailed results describing microcosm experiments conducted at 22 o C with 100% oxygen and in both acclimated and unacclimated wastes can be found elsewhere (Berge et al., 2005b). In all microcosm studies, ammonia readily disappeared. During each study, nitrate, nitrite, nitrogen gas, and nitrous oxide were detected, all indicating that simultaneous processes (nitrification and denitrification) were occurring, contributing to the large removal of ammonia-nitrogen. The occurrence of denitrification was not surprising because the presence of micro-anoxic areas within the digested MSW was expected and this phenomenon was observed during the waste acclimation process. Nitrous oxide concentrations in the gas-phase were the result of either partial nitrification or denitrification (Mummey et al., 1994; Venterea and Rolston, 2000). The mass of nitrogen added that was converted to nitrous oxide was significant at times, reaching levels as high as 20% of the total nitrogen added. Nitrous oxide production is a concern because it is a potent greenhouse gas. In each study, the ammonia-nitrogen removal rates were calculated using a central difference method of analyses and then fit to the Monod equation, which is used to describe kinetics in biological systems. The Monod equation follows equation 6: R kc K + C N = (6) S N where, R is the ammonia removal rate (mg N/g dry waste-day), K s is the halfsaturation constant (mg/l-n), C N is the total ammonia-nitrogen concentration (mg/l-n), and k is specific rate of removal of ammonia (mg N/day-g dry waste). The data were normalized by the dry mass of waste present in each microcosm test and plotted against the total ammonia concentration. Studies were conducted to evaluate the effect of the waste acclimation process on the ammonia removal rate. Unacclimated waste is defined as waste in which acclimation to ammonia-nitrogen did not occur prior to running the microcosm studies. During the unacclimated studies, the ammonia removal rate appeared to reach a maximum level at high ammonia concentrations, similar to the acclimated studies, 16