&EPA United States FPA-R00/R-04/1 ^0 Environmental Protection bUUm U4/1vjU Agency September 2004 Evaluation of SCR Catalysts for Combined Control of NOx and Mercury ------- ------- EPA-600/R-04/130 September 2004 Evaluation of SCR Catalysts for Combined Control of NOx and Mercury by Ravi K. Srivastava Chun Wai Lee U.S. Environmental Protection Agency Office of Research and Development National Risk Management Research Laboratory Air Pollution Prevention and Control Division Research Triangle Park, NC 27711 ICCI Project Manager: Joseph Hirschi Illinois Clean Coal Institute 5776 Coal Drive, Suite 200 Carterville, IL 62918 EPA Project Officer: Ravi K. Srivastava National Risk Management Research Laboratory Air Pollution Prevention and Control Division Research Triangle Park, NC 27711 U.S. Environmental Protection Agency Office of Research and Development Washington, DC 20460 ------- Abstract In addition to reducing nitrogen oxides (NOx), selective catalytic reduction (SCR) catalysts can oxidize elemental mercury (Hg°) in flue gases of coal-fired boilers. Therefore, Hg° oxidation by SCR catalysts upstream of flue gas desulfurization (FGD) systems should create an opportunity for simultaneous and cost-effective control of mercury, NOx, and sulfur dioxide (S02) emissions at power plants. This is because, unlike Hg°, oxidized mercury (Hg2+) is highly soluble in aqueous solutions and, therefore, can be easily removed in FGD systems. A two-task, bench- and pilot-scale research study was conducted to investigate the effect of SCR catalysts on mercury speciation in Illinois and Powder River Basin (PRB) coal combustion flue gases. Task I studied Hg° oxidation by a SCR catalyst using a bench-scale reactor. Illinois and PRB coal flue gases were simulated in the bench-scale system. Very good oxidation of Hg° was observed in all the coal flue gases (approximately 90%). It was shown that hydrogen chloride (HC1) may be the critical flue gas component that causes the conversion of Hg° to Hg2+ under SCR reaction conditions. Since high HC1 concentrations are expected during combustion of all Illinois coals, firing of these coals should result in significant Hg° oxidation occurring across the SCR reactors in the field. Based on bench-scale results, an appropriate SCR catalyst was produced in a pilot-scale size by Cormetech, Inc. and installed in a pilot-scale combustor for Task II studies. Three different Illinois coals (from high to low sulfur and chlorine) and one PRB coal were combusted in the pilot-scale unit. For the high sulfur/low chlorine Illinois coals, about 84% to 68% Hg°, and 15% to 9% Hg2+ were measured at the inlet of the pilot-scale SCR reactor. The SCR catalyst induced high oxidation of Hg°, decreasing the percentage of Hg° at the outlet of the SCR to values below 16%. The low sulfur/high chlorine Illinois coal had a relative high amount (21%) of Hg2+ at the inlet of SCR. The Hg° content of this coal flue gas was decreased from about 73% at the inlet to about 4% at the outlet of the SCR reactor. The PRB coal tests indicated a low oxidation of Hg° by the SCR catalyst. The percentage of Hg° decreased from about 96% at the inlet of the reactor to about 80% at the outlet. 11 ------- Foreword The U.S. Environmental Protection Agency (EPA) is charged by Congress with protecting the Nation's land, air, and water resources. Under a mandate of national environmental laws, the Agency strives to formulate and implement actions leading to a compatible balance between human activities and the ability of natural systems to support and nurture life. To meet this mandate, EPA's research program is providing data and technical support for solving environmental problems today and building a science knowledge base necessary to manage our ecological resources wisely, understand how pollutants affect our health, and prevent or reduce environmental risks in the future. The National Risk Management Research Laboratory (NRMRL) is the Agency's center for investigation of technological and management approaches for preventing and reducing risks from pollution that threaten human health and the environment. The focus of the Laboratory's research program is on methods and their cost-effectiveness for prevention and control of pollution to air, land, water, and subsurface resources; protection of water quality in public water systems; remediation of contaminated sites, sediments and ground water; prevention and control of indoor air pollution; and restoration of ecosystems. NRMRL collaborates with both public and private sector partners to foster technologies that reduce the cost of compliance and to anticipate emerging problems. NRMRL's research provides solutions to environmental problems by: developing and promoting technologies that protect and improve the environment; advancing scientific and engineering information to support regulatory and policy decisions; and providing the technical support and information transfer to ensure implementation of environmental regulations and strategies at the national, state, and community levels. This publication has been produced as part of the Laboratory's strategic long-term research plan. It is published and made available by EPA's Office of Research and Development to assist the user community and to link researchers with their clients. Sally Gutierrez, Acting Director National Risk Management Research Laboratory 111 ------- EPA Review Notice This report has been peer and administratively reviewed by the U.S. Environmental Protection Agency and approved for publication. Mention of trade names or commercial products does not constitute endorsement or recommendation for use. This document is available to the public through the National Technical Information Service, Springfield, Virginia 22161. Disclaimer This report was prepared by Ravi K. Srivastava of U.S.EPA with support, in part, by grants made possible by the Illinois Department of Commerce and Economic Opportunity through the Office of Coal Development and the Illinois Clean Coal Institute. Neither Ravi K. Srivastava of U.S. EPA nor any of its subcontractors nor the Illinois Department of Commerce and Economic Opportunity, Office of Coal Development, the Illinois Clean Coal Institute, nor any person acting on behalf of either: (A) Makes any warranty of representation, express or implied, with respect to the accuracy, completeness, or usefulness of the information contained in this report, or that the use of any information, apparatus, method, or process disclosed in this report may not infringe privately-owned rights; or (B) Assumes any liabilities with respect to the use of, or for damages resulting from the use of, any information, apparatus, method or process disclosed in this report. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise, does not necessarily constitute or imply its endorsement, recommendation, or favoring; nor do the views and opinions of authors expressed herein necessarily state or reflect those of the Illinois Department of Commerce and Economic Opportunity, Office of Coal Development, or the Illinois Clean Coal Institute. Notice to Journalists and Publishers: If you borrow information from any part of this report, you must include a statement about the state of Illinois' support of the project. iv ------- Contents Section Page Abstract ii Foreword iii List of Figures vi List of Tables vi Acknowledgments vii Nomenclature viii Executive Summary 1 Objectives 5 Introduction and Background 7 Experimental Procedures 9 Results and Discussion 13 Task I: Bench-Scale Evaluation of SCR Catalyst for Hg° Oxidation in Simulated Illinois and PRB Coal Combustion Flue Gases 13 Task II: Pilot-Scale Evaluation of SCR Catalyst for Hg° Oxidation in the Combustion of Natural Gas, Illinois, and PRB Coal Combustion Flue Gases 15 Subtask I: SCR Catalyst Evaluation in Natural Gas Combustion 15 Subtask II: SCR Catalyst Evaluation in Illinois and PRB Coal Combustion 19 Galatia Coal Combustion 19 Turris Coal Combustion 22 Crown II Coal Combustion 23 PRB (Black Thunder) Coal Combustion 23 Conclusions and Recommendations 25 References 27 v ------- List of Figures Figure Page 1 Schematic of Bench-Scale SCR/Hg° Oxidation Reactor System 9 2 Schematic of the Pilot-Scale Combustor Used to Evaluate the SCR Catalyst 10 3 SCR Temperature Profiles for the Doped Natural Gas Test Day 06/17/03; Hg CEM (PSA) Sampling; Insulation Added Across the SCR System 17 4 Hg CEM (PSA) Measurement Results for the Doped Natural Gas Test Day 06/17/03 18 5 Hg CEM (PSA) Measurement Results for the Doped Natural Gas Test Day 06/18/03; IGS Probe Hg Speciation Bias Check and the Effect of HC1 on Hg° Oxidation 19 List of Tables Table Page 1 Proximate and Ultimate Analyses of Three Illinois Basin Coals and a PRB Coal ... 13 2 Summary of Simulated Flue Gas Compositions for Task I Studies 14 3 Summary of NOx Reduction Results for Task I Studies 14 4 Summary of Mercury Speciation Results for Task I 15 5 Hg Speciation Across the SCR Catalyst for the Natural Gas Test Day 06/05/03 .... 16 6 Hg Speciation Across the SCR Catalyst for the Doped Natural Gas Combustion Test Days; Comparison between OH and PSA 18 7 Combustion Conditions for the Illinois Turris, Galatia, and Crown II Coals and the Black Thunder PRB Coal 20 8 Galatia Coal Combustion Test, OH Sampling Results Measured at the Inlet of the SCR Reactor 21 9 Galatia Coal Combustion Test, OH Sampling Results Measured at the Outlet of the SCR Reactor 21 10 Turris Coal Combustion Test, OH Sampling Results Measured at the Inlet and Outlet of the SCR Reactor 22 11 Crown II Coal Combustion Test, OH Sampling Results Measured at the Inlet and Outlet of the SCR Reactor 23 12 First PRB Coal Combustion Test, OH Sampling Results Measured at the Inlet and Outlet of the SCR Reactor 24 13 Second PRB Coal Combustion Test, OH Sampling Results Measured at the Inlet and Outlet of the SCR Reactor 24 vi ------- Acknowledgments The authors would like to acknowledge the support for this work provided by the Illinois Clean Coal Institute. This work was completed under the ICCI project number 02-1/2.2A-1. Mr. Joe Hirschi, the ICCI project manager, provided many helpful suggestions. Cormetech, Inc. provided the SCR reactor and catalyst used in this work. Dr. Thomas W. Hastings, Mr. Frank M. Stevens, and Mr. Mark A. Conger of Cormetech, Inc. provided SCR catalyst-related technical guidance and support. Finally, research support provided by Dr. B. Ghorishi, Dr. W. Jozewicz, and Mr. Jarek Karwowski of ARCADIS under EPA contract 68-C-99-201 is also appreciated. vii ------- Nomenclature Term Definition APPCD Air Pollution Prevention and Control Division CAA Clean Air Act CEM continuous emission monitoring CO carbon monoxide co2 carbon dioxide EPA U.S. Environmental Protection Agency FGD flue gas desul-furization HC1 hydrogen chloride Hg mercury Hg° elemental mercury Hg2+ ionic mercury Hgp Hg associated with particulate matter HgT total mercury ICCI Illinois Clean Coal Institute IFR innovative furnace reactor IGS isokinetic/inertial gas sampling nh3 ammonia NOx nitrogen oxides OH Ontario Hydro PM particulate matter PRB Powder River Basin PSA PS analytical SCR selective catalytic reduction SO, sulfur dioxide viii ------- Evaluation of SCR Catalysts Executive Summary The Air Pollution Prevention and Control Division (APPCD) of the U.S. Environmental Protection Agency (EPA), ARCADIS, Inc., and Cormetech, Inc. conducted an Illinois Clean Coal Institute (ICCI) co-funded research to evaluate the viability of a mercury control technology utilizing oxidation of elemental mercury (Hg°) to ionic mercury (Hg2+) for subsequent removal by conventional flue gas desul- furization (FGD) systems. Mercury speciation modi- fication was implemented via selective catalytic reduction (SCR) catalysts, currently used for nitrogen oxides (NOx) control. In December of2001, the EPA determined that control of mercury emissions from coal-fired power plants is appropriate and necessary. For Illinois coal-fired utilities, control of mercury emissions may be needed to comply with future regulations. Oxidation of Hg° to Hg2+ prior to an FGD system is a viable option for controlling mercury emissions because FGD systems are known to effi- ciently remove Hg2+. This research generated valu- able data, which may be used by utilities using Illinois coals to cost-effectively control mercury, NOx, and sulfur dioxide (S02) emissions using SCR/FGD systems. Currently there are no commercial technologies for modifying mercury speciation prior to an FGD unit. This study evaluated the viability of SCR technology for mercury speciation modification in Illinois and Powder River Basin (PRB) coal combustion pro- cesses. Since SCR and FGD are increasingly being used at power plants to control NOx and S02 emis- sions, respectively, their combination has the poten- tial to provide effective mercury control. The objectives of the research were to conduct bench- and pilot-scale studies in order to demonstrate Illinois and PRB coal combustion conditions that favor conversion of Hg° to Hg2+ by a Cormetech, Inc. formulated SCR catalyst. These objectives were accomplished initially through a systematic bench- scale study of the effect of the simulated flue gas compositions (three different Illinois and one PRB coals) on the oxidation of Hg° to Hg2+ in the presence of the SCR catalyst. Hg° oxidation activity of the SCR catalyst was established during the bench-scale testing phase of this study. A pilot-scale version of the Cormetech SCR catalyst was installed in a pilot- scale combustor to evaluate and establish Hg° oxidation capabilities of this catalyst during combustion of three different Illinois coals (from high to low sulfur/chlorine) and one PRB coal (very low sulfur and chlorine). The research was performed over a 12-month period (from September 2002 to August 2003) and consisted of two main tasks. Task I involved bench-scale tests of a Cormetech SCR catalyst formulation to character- ize oxidation of Hg° to Hg2+ in the presence of coal combustion/SCR flue gas species such as NOx, ammonia (NH3), S02, and hydrogen chloride (HC1). The ranges of NOx, NH3, HC1, S02, and water (H20) concentration employed during Task I were those expected to be found in three representative Illinois coals and one PRB coal combustion flue gases. The Illinois coals tested include Galatia [low sulfur (1.13 wt%) and high chlorine (0.29 wt%)], Turris [medium to high sulfur (3.11 wt%) and low chlorine (0.17 wt%)], and Crown II [high sulfur (3.48 wt%) and low chlorine (0.13 wt%)]. The PRB coal tested (from Black Thunder) is a Southern PRB coal with very low sulfur (0.29 wt%) and chlorine content (about 0.01 wt%). Task I results showed that HC1 is the critical flue gas component for converting Hg° to Hg2+ under SCR emission control conditions. Hg2+ was measured as the predominant species at the outlet of the SCR 1 ------- Evaluation of SCR Catalysts catalyst in the four simulated coal tests (PRB, Galatia, Turns, and Crown II). The HC1 concentra- tion (8 ppm) used for PRB test was much lower than those used for the Illinois coal combustion tests (100-200 ppm). However, it was still much higher than that of the doped Hg° (19 ppb) in the simulated flue gas and may be adequate for converting most of the Hg° to Hg2+. Hg° was found to be the predominant species at the exit of the SCR catalyst for only one test, PRB coal that had no HC1 present in the flue gas. This is believed to be due to the lack of a chlorine source in this test. The final result of Task I was the confirmation of the SCR catalyst formulation to be used in Task II. Task II was performed in two subtasks. In Subtask I, a SCR catalyst was installed in a vertical downward flow configuration in the post-combustion region of a pilot-scale 34.9 kW (150,000 Btu/h), refractory lined, down-fired cylindrical furnace. This furnace, termed the innovative furnace reactor (IFR), has the capability of firing natural gas or pulverized coal. In Subtask I, flue gas from natural gas combustion was doped with appropriate amounts of Hg°, HC1, NOx, NH3, and S02. It was determined that the catalyst contribution to the oxidation of Hg° was very low (only about 8% increase). It was hypothesized that the absence of particulates (fly ash) might have a bearing on Hg° oxidation reactions. Indeed, the coal combustion tests performed later indicated a very good Hg° oxidation by the SCR catalyst. The natural gas combustion test indicated that the IFR could be operated consistently, and most of the injected mercury could be recovered at the outlet of the SCR catalyst. Furthermore, it established reliable operation of the pilot-scale combustion/SCR system. In Subtask II of Task II, three different Illinois coals were combusted in the IFR facility. The objective of Subtask II was to characterize Hg° oxidation for a spectrum of Illinois coals and to establish any addi- tional increases of Hg2+ resulting from the SCR catalyst used in Task I and Subtask I of Task II. Since combustion of PRB coals will result in the production of Hg°-dominated flue gases, the effect of the pilot- scale SCR catalyst on oxidation of Hg° in a typical PRB coal combustion flue gas was also studied. Mercury speciation of the flue gas (Hg°, Hg2+, and Hg associated with particulate matter, Hgp) across the SCR catalyst was measured using the Ontario Hydro (OH) method. Attempts were made to measure gas- phase Hg (Hg° and Hg2+) by using a specially de- signed dual isokinetic/inertial gas sampling (IGS) probe. The Hg speciation samplings were performed at the inlet of the baghouse where high particulate (fly ash) loading on the filter of the isokinetic OH sam- pling train may change the speciation of mercury. Deposition of white powder was observed occurring on the walls of the first impinger connected to the IGS probe. Comparison of the data measured by using the OH method and those using the IGS probe showed the total Hg in solid and gaseous phases measured by the OH method was significantly higher than the gas- phase Hg. The white ammonium salt aerosols were formed near the inlet of the first impinger connected to the IGS probe as the ammonia-containing flue gas cooled down. The aerosols may retain gas-phase Hg and change in Hg speciation may result. Further work is needed for the development of the IGS probe for sampling ammonia-containing flue gas in SCR sys- tems. NOx conversions by the Cormetech-formulated SCR catalyst for all these coals were around 90%. Hg° was found to be the predominant species in the flue gases measured at the inlet of SCR catalyst for the three Illinois coals tested. The combustion of the low sulfur/high chlorine Galatia coal produced a relative high amount (21%) of Hg2+ at the inlet of SCR. The Hg° content of this coal flue gas was decreased from about 73% at the inlet to about 4% at the outlet of SCR reactor. Turris and Crown II coal combustion tests exhibited similar mercury speciation and oxidation behavior. These two coals are very similar in their properties. During the combustion of these coals about 84% to 68% Hg°, and 15% to 9% Hg2+ were measured at the inlet of the pilot-scale SCR reactor. The SCR catalyst induced high oxidation of Hg°, dropping the percentage of Hg° at the outlet of the SCR to below 16%. These results suggest that the use of SCR systems in boilers firing Illinois coal and 2 ------- Evaluation of SCR Catalysts equipped with FGD systems may result in about 85% to greater than 90% control of mercury. The PRB coal combustion tests showed that almost all the Hg measured at the inlet was Hg°. The test results also indicated a very low oxidation of Hg° by the SCR catalyst. Based on the measurements con- ducted in this study, it appears that SCR applications on PRB coal-fired boilers may not result in significant increase in Hg2+ content of flue gas. Therefore, such boilers equipped with wet FGD systems may not achieve increased mercury removal resulting from SCR applications. 3 ------- 4 ------- Evaluation of SCR Catalysts Objectives The obj ective of this research program was to investi- gate Hg° oxidation across selective catalytic reduc- tion (SCR) systems in Illinois coal combustion processes. Additional tests were also performed to characterize SCR catalyst activity in a PRB coal com- bustion environment. The specific goals of the pro- posed research were to identify the effect of SCR catalysts, flue gas compositions, and combustion conditions (time-temperature history and concentra- tions of gas-phase species) on the oxidation of Hg° upstream of the flue gas desulfurization (FGD) units. This work was synergized with similar collaborative bench-scale research being conducted by U.S. EPA and Cormetech, Inc. This research was performed over a 12-month period and consisted of two tasks. Task I involved bench- scale testing of a promising SCR catalyst formulation to characterize oxidation of Hg° to Hg2+ in the pres- ence of coal combustion flue gas constituents such as NOx, NH3, S02, and HC1. Previous EPA investiga- tions with synthetic and actual fly ashes have indi- cated that certain transition metals are very active in catalyzing conversion of Hg° to Hg2+ in the presence of HC1 and NOx. SCR catalysts contain vanadium (a transition metal) active sites. In Task I, the role of this transition metal on Hg° oxidation was identified. The NOx, NH3, HC1, S02, and H20 concentration ranges employed during Task I were those expected to be found in three representative Illinois coals (low to high sulfur and chlorine) and one PRB coal com- bustion flue gases. The final result of Task I was the confirmation of the SCR catalyst formulation to be used in Task II. Task II was performed in two subtasks. In Subtask I, a pilot-scale configuration of the active SCR catalyst was installed in a vertical downward flow configura- tion in a pilot-scale 34.9 kW (150,000 Btu/h), refrac- tory lined, down-fired cylindrical furnace. This furnace, termed the IFR, has the capability of firing natural gas or pulverized coal. In Subtask I, natural gas combustion flue gas was doped with appropriate amounts of Hg°, HC1, NOx, NH3, and S02. Mercury speciation of the flue gas (Hg° and HgCl2 content) across the SCR catalyst was measured using the Ontario Hydro (OH) method. Once the reliable operation of the pilot-scale combustion system was established, three different Illinois coals were com- busted in Subtask II of Task II. The objective of Subtask II was to characterize Hg° oxidation for a spectrum of Illinois coals and to establish any addi- tional increases of Hg2+ using the SCR catalyst characterized in Task I and Subtask I of Task II. The effect of the active pilot-scale SCR catalyst on oxidation of Hg° in a typical PRB coal combustion flue gas was also studied. 5 ------- 6 ------- Evaluation of SCR Catalysts Introduction and Background Metallic air toxics, such as mercury, are found in coal combustion flue gases in elemental or various oxi- dized forms and exist in the solid, aerosol, or vapor state. These compounds originate from the raw coal and are typically enriched in the fine particles (Markowski and Filby, 1985). Mercury speciation may be potentially modified by SCR catalysts cur- rently used for NOx control. In December of 2001, the EPA determined that control of mercury emis- sions from coal-fired power plants is appropriate and necessary. For utility boilers fired with Illinois coal, control of mercury emissions may be needed to comply with future regulations. Oxidation of Hg° to Hg2+ prior to an FGD system is a viable option for controlling mercury emissions, becauseFGD systems easily remove Hg2+. Research data is needed to assess the contribution of SCR catalysts to the oxidation of Hg° prior to the FGD systems. Preliminary EPA laboratory results (Ghorishi and Gullett, 1998) have indicated the effectiveness of alkaline sorbents (dry FGD units) in removing Hg2+ (but not Hg°) from flue gases. Previous investigations have also indicated that many flue gas and fly ash parameters determine the speciation of mercury (Ghorishi, 1998; Ghorishi etal., 1999). The distribution of various forms of mercury in flue gases is a function of system-specific properties. For example, various concentrations of chlorine or sulfur have been shown to significantly affect the expected equilibrium product distribution (Linak and Wendt, 1993). Ghorishi et al. (1999) have shown that, in combustion flue gas with the presence of active simulated or actual fly ashes, Hg° is readily oxidized by HC1 and NOx at temperatures typical of air pollu- tion control systems (150-250 °C). The presence of these constituents in the flue gas can therefore shift the ratio of Hg2+/Hg° to a higher value and result in better removal of mercury in F GD systems. However, quantitative information on speciation of mercury in the SCR systems in the presence of coal combustion flue gases at typical SCR operating temperatures (340-370 °C) is lacking at this time. The control of mercury emissions has been found to strongly depend on the form of mercury; these results were observed at the EPRI High Sulfur Test Center's 4 MWe pilot-scale wet FGD system and at municipal waste combustor (MWC) tests (Volland, 1991). The presence of HC1 and NOx in the combustion flue gases shifts the mercury speciation toward its oxi- dized form (Hall et al., 1991, Ghorishi, 1998). The type of coal has also been found to affect different levels of mercury capture (Felsvang et al., 1992; 1993). The chlorine present in coal is capable of forming Hg2+, which is less volatile and more wa- ter-soluble than Hg°, and is, therefore, thought to be an important factor in the capture of mercury in coal combustion flue gas (Felsvang et al., 1993; DeVito et al., 1993). Therefore, an important scientific issue that is being investigated is the speciation of mercury in coal combustion flue gases. Knowledge about the transformation of Hg° to HgCl2 is absolutely crucial in this regard. Senior et al. (1997) has reviewed in detail the state of knowledge of mercury speciation in coal-fired processes. Their review and an investiga- tion by Krishnan et al. (1994) concluded that the assumption of gas-phase equilibrium for mercury species in coal-fired flue gases is not valid, and major reaction pathways for mercury oxidation in coal combustion flue gas need to be investigated. These investigations should include heterogeneous reactions on active surfaces of particles in the flue gas as well as gas-phase reactions (Senior et al. 1997). A preliminary investigation was performed at EPA 7 ------- Evaluation of SCR Catalysts laboratories (Ghorishi, 1998) to provide some in- sights into these questions by studying mercury speciation in simulated coal combustion flue gases. Gas-phase studies indicated that the in-flight, post- combustion oxidation of Hg° in the presence of HC1 is very slow and proceeds at measurable rates only at temperatures greater then 700 °C and high HC1 con- centrations (200 ppm). The presence of S02 and H20 in the simulated flue gas significantly inhibited the gas-phase oxidation of Hg° in the presence of HC1. These results indicate that conversion of Hg° to Hg2+ needs to be mediated by the presence of an active surface such as fly ash (or SCR catalyst, as in this study). The effects of coal fly ash components and compositions were investigated using a fixed-bed of model (simulated) fly ashes. The primary focus was on evaluating the catalytic Hg° oxidation activity of major mineral constituents of coal fly ashes: alumina (A1203), silica (Si02), iron (III) oxide (Fe203), copper (II) oxide (CuO), and calcium oxide (CaO). Copper and iron oxides exhibited significant catalytic activity toward a surface-mediated oxidation of Hg°. SCR has been used extensively to control NOx emissions from hundreds of utility and industrial boilers in Japan, Europe, and the United States, and a large number of SCR systems are under construc- tion or planned to be installed in the United States. A recent study (Cichanowicz and Muzio, 2001) esti- mated that, by the year 2004, in excess of 100 GW of coal-fired capacity in the United States will be equipped with SCR to meet the requirements of Title I of the Clean Air Act (CAA). The study describes how today's U.S. applications of SCR have evolved from two prior generations of design in response to changing application conditions. The first commer- cial SCR application in Japan occurred in the late 1970s, and a subsequent "wave" of installations took place in Europe (predominantly Germany) in the mid-1980s. As a result of this increased SCR market, present-day SCR catalysts have evolved to feature thinner walls, improved mass transfer and activity, and better poison resistance than earlier generations. Limited field data obtained from European coal-fired boilers equipped with SCR systems suggest that SCR catalysts promote the formation of Hg2+ (Gutberlet et al., 2000). Results of recent pilot-scale screening tests on the impact of SCR on mercury speciation also indicate that SCR systems could promote the forma- tion of ionic Hg (EPRI, 2000). The screening test results reflected that the impact of SCR on mercury speciation appears to be highly coal-type dependant. For the coals tested (three bituminous coals and one subbituminous PRB coal), only high-sulfur bitumi- nous coal showed significant increase in oxidized mercury in the outlet of the SCR. The low-sulfur (also low chloride) PRB coal showed very little impact of SCR on mercury speciation. The other two bituminous coals' results were between the two extreme cases. Results of the recent field tests con- ducted in the United States showed there was an increase in Hg oxidation across the SCR catalysts for plants firing eastern bituminous coals with sulfur contents ranging from 1.1% to 4.2% (Laudal et al., 2003). The increase in oxidation of Hg° to ionic Hg across the SCR reactor could result from the changes in flue gas chemistry caused by the SCR catalysts. The SCR catalysts (which include oxides of vanadium, tita- nium, and tungsten) have the potential to catalyze the formation of S03 (from S02) and Cl2 (from HC1) in coal combustion flue gases. Both S03 (Galbreath and Zygarlicke, 2000) and Cl2 (Ghorishi etal., 1999)have been suggested to react with Hg° to cause its oxida- tion. However, not much is known on the effect of the flue gas species, chemical reactions, and reaction conditions on Hg° oxidation across SCR reactors. 8 ------- Evaluation of SCR Catalysts Experimental Procedures Task I studies were performed in a bench-scale SCR/Hg° oxidation reactor system. A schematic of the reactor system is shown in Figure 1. The system consists of preheating and premixing sections, a mercury generation unit, a SCR reactor, and an on-line reactor effluent measurement unit. Flue gas components including carbon dioxide (C02), S02, air, and nitrogen (N2) were mixed and preheated to 350 °C and then mixed with another preheated stream of NOx and HC1 at the main heating section. Water, for simulating the moisture content in the flue gas, was also pumped into this section at a calibrated rate and mixed with the other flue gas components. Ammonia was preheated and added into a section equipped with static mixing hardware. The gas mixture flowed into a Pyrex reactor. The alkaline NH3 reacts with the acid gas components in the gas mixture to form ammo- nium salts at temperatures lower than the SCR reaction temperatures, so the simulated flue gas mixture was preheated by electrical furnaces and maintained at 350 °C by temperature-controlled electrical heating tapes. A mercury generation unit consisting of a mercury permeation tube surrounded by a temperature-controlled water bath was used to generate Hg° vapor for the oxidation experiments. The Hg° vapor was carried by an N2 stream and mixed with flue gas mixture near the top of the reactor (4 cm in inner diameter and 35 cm in length). The flue gas mixture containing Hg° and NH3 passed through a honeycomb flow straightener to produce a uniform laminar flow before passing through the honeycomb SCR catalyst. The effluents exiting the SCR reactor passed through a drying tube for remov- ing moisture in the flue gas. The dried gas then passed to an online ultraviolet S02 analyzer (Bovar Engineering, Inc., model 721AT2; accuracy of about ±5%) for measuring S02. An online chemilumines- centNOx analyzer (Advanced Pollution Instrumenta- tion, Inc., model 200AH; accuracy of about ±5%) downstream of the S02 analyzer was used for mea- suring NOx. Sampling Port Heating Tape Water Pump ^ Thermocouple —2- Thermocouple Heating Tape - w Electrical Heating . Furnaces Static Mixiny Section Heating Tape Mass Flow Controller -cw- -mn- -mK- ¦£> NO HCI Air (Nj, 02) co2 S02 Nj nh3 Honeycomb Flow — Straightener SCR Catalyst Thermocouple Mass Flow Controllers =LJ=LJ=C3=E3= Exhaust NOx Analyzer so2 Analyzer Mercury Analyzer Water Removal System Mercury 1 Generation System Mass Flow Controller N2 Electrical Heating Furnace Thermocouple Figure 1. Schematic of Bench-Scale SCR/Hg° Oxidation Reactor System. 9 ------- Evaluation of SCR Catalysts A SCR catalyst with a vanadia/titania formulation and a honeycomb configuration, designated as Catalyst A, was used in Task I. A small piece (2.2 cm for both sides and 1.9 cm in length; calculated space velocity of 2609 hr"1) of the catalyst sample was placed into the catalyst compartment of the SCR reactor. The flows of the flue gas components were maintained at the desired levels by using mass flow controllers. The variability of the gas concentrations is about ±2.5%. A constant total flow rate of 400 cm3/min—at a standard temperature and pressure (STP) of 25 °C and 101.4 kPa, respectively—was used for all tests. The concentrations of S02 and NOx at the outlet of the reactor were monitored continu- ously for 4 hours by using S02 and NOx analyzers to ensure that NOx reduction reached a steady state. Then the two gas analyzers were disconnected from the outlet of the reactor, which was subsequently connected to a mercury sampling train. The mercury speciation measurement method known as the OH method was used for measuring Hg° and Hg2+ in this study. In Task II, tests were performed in a pilot-scale 34.9 kW (150,000 Btu/h), refractory lined, down-fired cylindrical furnace fired with coal or natural gas (Figure 2). The furnace, termed the IFR, has an inner diameter of 15.2 cm and overall length of about 4 m. The IFR is used to simulate and generate a coal combustion environment and quench rate conditions similar to those upstream of FGD units in coal-fired utility boilers. View and injection/probe points tra- verse the length of the furnace for testing flexibility. The IFR system was retrofitted with two vertical sections, an upward and a downward flow. The two-stage, pilot-scale SCR reactor based on the bench-scale results was installed in the downward section. Figure 2 illustrates the dimensions of this two-stage SCR system. A sampling port was installed between the two stages. The first stage of the SCR catalyst is equipped with a flow straightener, a soot blower, and a honeycomb catalyst. The second stage has a soot blower and a honeycomb catalyst. The two honeycomb catalysts are identical and have a pitch of 8.2 mm, wall thickness of 0.83 mm, and length of Natural Gas 16ft J Gate Valve ° Va' Temperature Port O 3" OH/CEM Port Air • o»o«O *0 Coal 3ft 5" id Duct 2th Mezzanine 7.3 ft 1 SCR Reactor includes honeycomb, flow straightener, and soot blower SCR By-pass Vertical Furnace Refractory Section 8ft 1 ft Sampling Joint 1 Mezzanine 6.8 ft 2"*-'SCR Reactor includes honeycomb and soot blower 9ft 23 ft 3ft Collection Bin Floor Baghouse Figure 2. Schematic of the Pilot-Scale Combustor Used to Evaluate the SCR Catalyst. 10 ------- Evaluation of SCR Catalysts 1250 mm; they consist of 8x8 cells (a total of 64). The calculated space velocity of the two stages when the IGS probe is operational was 2943 hr"1. This space velocity is comparable to the 2609 hr"1 used in bench-scale tests and to SCR space velocities ob- served in the field. Sampling ports along the horizon- tal ducts at the inlet and the outlet of the vertical SCR reactor allow gas and particle monitoring. The IFR is equipped with a continuous emission monitoring (CEM) system for 02, C02, CO, NOx, and S02. The IFR is also equipped with 12 thermocouples in order to accurately determine the time-temperature history of the combustor. Isokinetic OH trains were used for making mercury speciation measurements at the inlet and the outlet of the SCR reactor. Two inertial gas sampling (IGS) probes (Mott Co.) were also used in conjunction with the isokinetic OH trains at the inlet and the outlet of the SCR reactor for Hg measurements. The IGS probe has the capability of separating gas from particulate matter without the need for a filter, thus minimizing the potential Hg speciation biases created by the filter of the isokinetic OH train. The IGS probes were equipped with venturi flow meters for accurate axial flow measurement. The focus of the isokinetic trains for mercury sampling was only on total mercury data (Hgx), since these samplings were performed at the inlet of the baghouse with signifi- cant particulate loading in the flue gas. The IGS probe data are only for the gas-phase elemental and ionic mercury. The difference between Hgx results obtained by the isokinetic OH train and the gas-phase Hg° and Hg2+ obtained by the IGS probe is the amount of Hg associated with particulate matter (Hgp). The IFR is also equipped with a baghouse and a scrubber to treat the flue gas before releasing it into the atmosphere. 11 ------- Evaluation of SCR Catalysts 12 ------- Evaluation of SCR Catalysts Results and Discussion Task I: Bench-Scale Evaluation of SCR Catalyst for Hg° Oxidation in Simulated Illinois and PRB Coal Combustion Flue Gases Bench-scale testing of the SCR catalyst was per- formed in the simulated combustion flue gases of three different Illinois coals and one PRB coal. These coals were also tested in the pilot-scale combustor during Task II studies. Table 1 summarizes the characteristic of the obtained Illinois coals (Turris, Crown II, and Galatia) and the PRB coal (Black Thunder). Turris coal sample contains medium amounts of sulfur and chlorine. Crown II is a high sulfur/low chlorine coal sample. Galatia contains high amounts of chlorine and low amounts of sulfur. The ultimate analysis results of these coals were used to determine the composition of the simulated flue gas in each test case shown in Table 2. The base flue gas mixture consisted of 350 ppm NOx, 315 ppm of NH3, 15% C02, 3.5% of 02, 5.3% of H20, 19 ppb of Hg°, and balance in N2. The nominal substoichio- metric NH3/NOx ratio of 315/350 was used in the all tests in order to achieve a nominal 90% NOx reduc- tion. The PRB/no HC1 test represents a situation where the CI released from the combustion of PRB coal reacts with Ca resulting in no HC1 in the flue gas. This simulation was designed to determine the effect of HC1 on the oxidation of Hg°. Results of the NOx reduction for these tests are summarized in Table 3. NOx reductions of 85% to 88%) were observed for all the tests. These levels of NOx reduction are similar to those achieved in the field, suggesting that the reactor system used in the Table 1. Proximate and Ultimate Analyses of Three Illinois Basin Coals and a PRB Coal. Content PRB Black Thunder Illinois Turris (medium S/Cl) Illinois Crown II (high S/low CI) Illinois Galatia (low S/high CI) Moisture, % 14.00 16.99 16.07 11.33 Ash, % 5.92 9.26 7.34 6.29 Volatiles, % 37.33 33.89 37.05 34.16 Fixed C, % 42.76 39.85 39.55 48.55 Heating Value (Btu/lb) 9903 10,531 10,877 12,179 C, % 59.71 59.00 60.48 68.31 H, % 3.83 4.32 4.70 4.50 N, % 0.82 1.19 1.07 1.50 S, % 0.29 3.11 3.48 1.13 0, % 15.44 5.96 5.73 6.94 CI, % NAa 0.17 0.13 0.29 Ha. ppmw NA 0.07 0.07 0.09 aNA = not available. 13 ------- Evaluation of SCR Catalysts Table 2. Summary of Simulated Flue Gas Compositions for Task I Studies. „ PRB PRB/no HC1 Galatia Turris Crown II Concentration HC1, ppm 8 0 204 134 98 S02, ppm 280 280 934 2891 3116 NOx, ppm 350 350 350 350 350 NH3, ppm 315 315 315 315 315 co2, % 15 15 15 15 15 02, % 3.5 3.5 3.5 3.5 3.5 h20, % 5.3 5.3 5.3 5.3 5.3 H2°. DDb 19 19 19 19 19 Table 3. Summary of NOx Reduction Results for Task I Studies. NOx Results PRB PRB/no HC1 Galatia Turris Crown II Outlet NOx Concentration, ppm 44 52 44 43 41 NO, Reduction. % 87 85 87 88 88 present study closely simulated the SCRNOx emis- sion control conditions in the field. The mercury speciation results are summarized in Table 4 (± indicates duplicate test). Two mercury speciation samples were taken at the inlet of the SCR catalyst by using the OH method after the PRB and Galatia tests. Two OH samples were also taken at the outlet of the mercury generation unit at the beginning of the test program. The total mercury concentrations (19.3±1.0 ppb) with very little Hg2+ (0.5 ppb) mea- sured near the inlet of the catalyst are very close to those measured at the outlet of the mercury genera- tion unit (19.2±0.1 ppb Hgx with 0.5 ppb Hg2+). The two inlet OH speciation results obtained in the presence of two different simulated flue gas mixtures (PRB and Galatia tests) showed Hg° was the only mercury species. The consistent inlet results suggest that the presences of HC1 and S02 with different concentrations in the simulated flue gas mixtures do not cause gas-phase oxidation of Hg° at 350 °C prior to the SCR catalyst. The speciation results shown in Table 4 suggest that HC1 has a significant effect on conversion of Hg° to Hg2+ under SCR/NOx emission control conditions. All tests except PRB/no HC1 showed mostly Hg2+ at the outlet of the SCR catalyst. As shown in Table 2, the gas mixtures used for PRB and PRB/no HC1 tests were identical except that 8 ppm of HC1 was added to the simulated flue gas mixture for the PRB test. Almost all the mercury measured in the PRB test was Hg2+, but very little Hg2+ was measured in the PRB/no HC1 test. The total mercury concentration at the outlet of the SCR catalyst measured for the PRB/no HC1 test (13.1 ppb) is about 60% of that measured at the inlet (19.3 ppb). The results of the two outlet replicate tests are similar to each other, indicating good precision. One possible explanation for the discrepancy in inlet and outlet results is that the SCR catalyst might have adsorbed Hg° in the absence of HC1 in the flue gas mixture during the PRB/no HC1 test. Evidence of Hg° adsorption was observed during the start-up of the SCR reactor system. When a gas mixture with 3% 02, 10% C02, 14 ------- Evaluation of SCR Catalysts Table 4. Summary of Mercury Speciation Results for Task I „ . ,. Inlet PRB PRB/no HC1 Galatia Turris Crown II Concentration Hg°, ppb3 18.8±1.0 0.7±0.1 12.6±0.4 0.73 3.3 2.0±0.1 Hg2+, ppb 0.5±0.0 17.8±0 0.46±0.0 16.2 29.4 17.9±0.5 HgT, ppb 19.3±1 18.4±0.2 13.1±0.4 16.9 32.7 20.0±0.3 HS2+. % 2.6 96.7 3.5 95.9 89.9 89.5 a 1 ppb =8.3 |ig/dscm 240 ppm of NO, and 30 ppb of Hg° (balance N2) was passed through the SCR catalyst at 350 °C, very little Hg° was measured at the outlet of catalyst by using a UV mercury analyzer. However, this adsorption of Hg° by the SCR catalyst requires further research. Table 4 also shows that Hg2+ was the predominant mercury species measured at the outlet of the catalyst for all the Illinois coal simulation tests. It appears that the relatively high HC1 concentrations present in these simulated flue gases provide adequate chlorine for conversion of Hg° to Hg2+. The total Hg concen- trations (Hgx) measured at the outlet of the catalyst for the Galatia and Crown II tests (16.9 and 20.0±0.3 ppb, respectively) are comparable to that measured at the inlet (19.3 ppb). However, much higher outlet total Hg concentration (32.7 ppb) was measured for the Turris Test. It is possible that, during this test, a spike in the generation of Hg° might have occurred. However, this shows that the SCR catalyst was active enough to oxidize the relatively high spike of Hg°. It appears that HC1 is the critical flue gas component for converting Hg° to Hg2+ under SCR emission control conditions. Hg2+ was measured as the pre- dominant specie at the outlet of the SCR catalyst for the four simulated coal tests even though these coals had widely different HC1 and S02 concentrations. The HC1 concentration (8 ppm) used for the PRB test was much lower than those used for the Illinois coal combustion tests. However, it is still much higher than that (19 ppb) of the Hg° in the flue gas and appears to be adequate for converting most of the Hg° to Hg2+ in the SCR catalyst. Hg° was found to be the predominant species for the only test that had no HC1 present in the flue gas. No Hg° oxidation observed for this test may be due to the lack of a chlorine source. Task II: Pilot-Scale Evaluation of SCR Catalyst for Hg° Oxidation in the Com- bustion of Natural Gas, Illinois, and PRB Coal Combustion Flue Gases Task II, Subtaskl: SCR Catalyst Evaluation in Natural Gas Combustion Natural gas tests were used to shakedown the system, to determine IFR temperature profiles, and to assess the feasibility of operating the pilot-scale SCR reactor at temperatures around 350 °C. Another objective of Task II, Subtask I was to assess the change in Hg speciation across the SCR in a particu- late-free environment. Natural gas tests were also used to determine whether mercury speciation in the IFR duct could be biased by use of the IGS probes. Initial temperature profile measurements across the SCR catalyst indicated that, at the nominal IFR firing rate of 120,000 Btu/hr, atemperature of 350 °C could be easily achieved in the first section of the SCR reactor. The axial temperature gradient along the SCR was about 80 °C, and it was determined that this axial temperature gradient needed to be reduced. Therefore, a layer of insulation was added to the SCR reactor wall, and all the flanges and exposed surfaces were insulated. The axial temperature gradient for the insulated system was 50 °C, which was deemed to be ------- Evaluation of SCR Catalysts appropriate for the operation of the SCR reactor. Ontario Hydro sampling for the natural gas tests was performed on 06/05/03. The new insulation was not in place during this test, and the axial temperature profile along the SCR reactor was 80 °C. Natural gas was combusted at a firing rate of 118,674 Btu/hr. The excess air was calculated to be 20%. The 02 concen- tration measured at the outlet of the SCR was 4.8% (predicted to be 3.8%), and the measured C02 con- centration was 8.9% (predicted to be 9.6%). Compar- ison between the measured and the calculated values indicated a 7% in-leakage based on C02 and a 6% in-leakage based on 02. The flue gas compositions measured at the inlet and outlet of the SCR reactor were identical, indicating no in-leakage between the inlet and the outlet of the SCR system. The total flow (including the in-leakage) was calculated to be 29.3 scfm (0.829 scm/min). NOx and S02 were doped into the flue gas in the vertical section, near the 2nd tem- perature location, at 600-ppm levels. NH3 and HC1 were doped at 500- and 50-ppm levels (calculated), respectively. HC1 was injected with S02 and NOx. NH3 was inj ected at the entrance of the top horizontal duct before the SCR system. The injected NH3 level corresponded to an NH3/NOx ratio of 0.83. The SCR catalyst was conditioned with S02 and HC1 for 4 hours prior to the test. A 90% NOx removal across the SCR catalyst was observed. Hg° was introduced in the vertical furnace section of the IFR via a sepa- rate injection nozzle. Simultaneous isokinetic OH sampling was performed at the inlet and the outlet of the SCR catalyst. IGS probes were not used since there were no particulates during the natural gas combustion tests. Mercury speciation results are shown in Table 5. Table 5. Hg Speciation Across the SCR Catalyst for the Natural Gas Test Day 06/05/03. Location Hg° Hg2+ Hgl (lg/dscm |og/dscm (ig/dscm % SCR Inlet 722 L86 9*08 20 SCR Outlet 6.35 2.48 8.83 28 Very good total Hg recovery was obtained across the SCR catalyst. The catalyst showed very little contri- bution to the oxidation of Hg° (only an 8% increase). It was hypothesized that the newly installed catalyst may need more aging in order to achieve a higher Hg° oxidation rate. The absence of particulates (fly ash) might also have a bearing on the catalytic Hg° oxida- tion reactions. A decision was made to continue with natural gas and Illinois coal combustion tests to induce aging in this catalyst. The natural gas combus- tion tests indicated that the IFR can be operated consistently and most of the injected mercury can be recovered at the outlet of the SCR catalyst. The doped natural gas combustion test was repeated on 06/17/03 after the additional SCR reactor insula- tion was installed. A mercury CEM known as PS analytical (PSA) was used to measure Hg speciation on this test day. Natural gas was combusted at a firing rate of 127,630 Btu/hr. The excess air was calculated to be 16%). The 02 concentration measured at the SCR outlet was 4.8% (calculated to be 3.3%) and the measured C02 concentration was 9.1% (calculated to be 9.9%). Comparison between measured and calcu- lated values indicated an in-leakage of 8% based on C02 and 7% based on 02 (insignificant). There was no in-leakage between the inlet and the outlet of the SCR. The total flow (including the in-leakage) was calculated to be 31.1 scfm (0.883 scmm). NOx and S02 were doped at 600- and 700-ppm levels, respec- tively. NH3 and HC1 were doped at 500- and 50-ppm levels (calculated), respectively. SCR catalyst was conditioned for 4 hours prior to the test with S02 and HC1. A 94%) NOx removal across the SCR catalyst was observed. Furthermore, the IGS probe was biased-checked by connecting combustion gas monitors to the sampling arm of the IGS probe. No difference in measurement of 02, C02, S02, andNOx was observed, indicating no leak in the IGS probe. This is the standard operating procedure for the bias-check of the IGS probe. SCR temperatures (top, middle, and bottom) are shown in Figure 3. As mentioned before, the more thorough reinsulation of the SCR reactor caused a 16 ------- 300 H 1 1 1 1 1 1 t 1 1 1 1 1 ¦ 1 1 1 1 t 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1— ^ >J CN O M CI O sj CM O Xj CM O Xl CM CD S| CM O s| CM O vj CI CD Sj CM O "-J CM O Nj CM CD vj CM CD sj CM FncomwiNoj^^^dddcnmaicocdcoN^N^uj'Xiiriifjint^^cQCQfrjcNCNoi^^ c^ir»o,^cNCQ^rir)o^ojcoco^rinoT-(MCQ^rmo^-cMco'^ir:oT-cNcQ^]-ir)o^csJcoxr ¦ • CI> 0"» CD CD CD CD CD CD *—¦ " t— *—¦ CM CM CM CM CM CM CO CO CO CO CO CO vj Nj vj vj vj vj Iri U"J llrj lO IjTj -C _ — — — — — — — — Time Figure 3. SCR Temperature Profiles for the Doped Natural Gas Test Day 06/17/03; Hg CEM (PSA) Sampling; Insulation Added Across the SCR System. decrease in the axial temperature gradient from 80 to 50 °C (377 °C at the top and 327 °C at the bottom). The axial temperature gradients in the first and the second sections were 15 and 35 °C, respectively. The mercury CEM, PSA, was connected to the IFR duct downstream of the SCR catalyst and just up- stream of the baghouse. Since no particulate matter (PM) was present during the natural gas combustion, the PSA was not connected to the IGS probe. PSA results are shown in Figure 4. NH3 was added around the 13:20 time frame. Addition of ammonia caused an increase in Hg° and, thus, total Hg. This was observed before during the bench-scale reactor shakedown tests. Probably, addition of NH3 caused desorption of previously adsorbed Hg°. The PSA data before the addition of NH3 (excluding the outlier) can be com- pared to the OH tests performed on 06/05/03. PSA results as compared to the simultaneous isokinetic OH samplings performed at the inlet and the outlet of SCR catalyst are shown in Table 6. Considering experimental variability and the fact that OH and PSA sampling were performed on two different test days, a relatively good agreement was obtained. As discussed, this shows that the SCR catalyst did not contribute to the oxidation of Hg° in natural gas firing tests. Moreover, the PSA showed a large variability in the measurement of Hg even when the outlier result (see Figure 4) was not included. The reason is unknown and may be related to the sample- conditioning unit of the PSA. Before proceeding to the coal tests, the potential biases of the IGS probe toward changes in Hg specia- tion needed to be investigated (an obj ective of natural gas test, Subtask I of Task II). Hg speciation in the IGS probe must be preserved in order to accurately quantify the Hg speciation in the duct of the IFR. A doped natural gas combustion test was performed on 17 ------- Evaluation of SCR Catalysts E o ) 5 3) =L rii 80 70 60 f> n 40 30 20 10 PSA results downstream of the SCR on 06/17/03; doped natural gas combustion; catalyst was conditioned with HCI and S02 for 4 hours. outlier (total Hg) Addition ot MH3 at 5UU ppm < — ~ .. i natural gas doped with NO, (S00 ppm), S02 (700 ppm), HCI (50 ppm), and Hg < *¦ total Hg elemental Hg ~~~ ~~~ +++ ~~~ 7:12 8:24 0:36 10:48 12:00 Time 13:12 14:24 15:36 16:48 Figure 4. Hg CEM (PSA) Measurement Results for the Doped Natural Gas Test Day 06/17/03. Table 6. Hg Speciation Across the SCR Catalyst for the Doped Natural Gas Combustion Test Days; Comparison between OH and PSA. Location Hg° Hg2+ HgT Hg2+ |Xg/dscm (lg/dscm |ig/dscm % 7.22 1.86 9.08 20 6.35 2.48 8.83 28 7.8±1.4 13.4±3.0 12-60 SCR inlet, OH sampling, 06/05/03 SCR outlet, OH sampling, 06/05/03 SCR outlet. PSA sampling. 06/17/03 test day 06/18/03 to investigate potential Hg specia- tion biases created by the IGS probe. Combustion conditions were similar to those obtained on test day 06/17/03 (see above). The mercury CEM, PSA, was initially connected to the IFR duct downstream of the SCR catalyst and just upstream of the baghouse. After two hours of sampling from the duct (from 8:38 to 10:46), the PSA was connected to the IGS probe. The PSA results are shown in Figure 5. As seen in this figure, the Hg speciation was preserved through the IGS probe, indicating that the IGS probe does not affect the Hg speciation in the duct. It should be noted that the IGS probe temperature (measured at the surface) was maintained at 240-250 °F (116-121 °C). Previous unpublished results have shown that, at higher temperatures [390 °F (200 °C)], the stainless steel surfaces of the IGS probe causes catalytic oxidation of Hg° to Hg2+. Thus, it is crucial to main- tain the temperature of the IGS probe in the low temperature range mentioned above. During this test, 18 ------- Evaluation of SCR Catalysts PSA results on 06(18/03; doped natural gas combustion; catalyst was conditioned with HCI and S02 the night before; bias check of the IGS probe and the effect of HCI on Hg oxidation by the SCR catalyst. 40 35 30 E u t/J 25 i. 20 I 15 10 5 0 7:12 8:24 0:3G 10:48 12:00 13:12 14:24 15:36 16:48 Time Figure 5. Hg CEM (PSA) Measurement Results for the Doped Natural Gas Test Day 06/18/03; IGS Probe Hg Speciation Bias Check and the Effect of HCI on Hg° Oxidation. GuJier (Hg total) PSA sampling from IGS probe HCI increased by a factor of 2 ~ from 50 to 100 ppm nominally •« ~ PSA sarn pi ing from the duct 1 * ¦ total Hg (* * ~»» ~~~ ~' elemental Hg»4 * an attempt was also made to assess the effect of increased HCI concentration on Hg° oxidation by the SCR reactor. HCI concentration was increased from 50 to 100 ppm. Figure 5 shows that the increase in HCI concentration in the combustor had no effect on speciation of Hg, confirming that the SCR catalyst does not contribute to the oxidation of Hg° in natural gas firing tests. Note that the catalyst was conditioned with S02 and HCI for more than 24 hours. Task II, Subtask II: SCR Catalyst Evaluation in Illinois and PRB Coal Combustion Galatia Coal Combustion Galatia coal was tested under combustion conditions listed in Table 7 (in the Galatia column). Galatia coal was combusted at a firing rate of about 140,000 Btu/hr. Comparison between measured and calculated values of C02 and 02 indicated an insignificant in-leakage of air into the combustor (about 4%). There was no in-leakage between the inlet and the outlet of the SCR. A very good combustion of Galatia coal was observed as indicated by low CO concentra- tion (30±7 ppm) measured in the flue gas. The axial temperature gradient across the SCR catalyst was around 50 °C (change from 354 to 304 °C). Uncon- trolled levels ofNOx were around 850 ppm. NH3 was injected at a nominal NH3/NOx ratio of 0.9. NOx conversion around the SCR for the combustion of Galatia coal was about 86%. S02 concentration for this low sulfur Illinois coal was measured to be 929±63 ppm. The calculated level of S02 under the combustion conditions was 915 ppm, reflecting very good agreement with measured values. Simultaneous isokinetic and IGS probe OH sampling were performed at the inlet and outlet of the SCR reactor (i.e., upstream of the baghouse) for the Galatia test. The isokinetic trains measure all three forms of mercury, Hg°, Hg2+, and Hgp, and Hgx can be calculated from the isokinetic train measurement. When sampling upstream of the baghouse, fly ash 19 ------- Evaluation of SCR Catalysts Table 7. Combustion Conditions for the Illinois Turris, Galatia, and Crown II Coals and the Black Thunder PRB Coal. Parameter Turris Galatia Crown II PRB Coal feed rate lb/hr 14.0 11.5 13.4 15.7 IFR firing rate, Btu/hr 147,540 140,424 145,208 155,873 Total air flow, scfm 28.6 28.2 28.4 28.9 Excess air, % 11 18 11 5 Top SCR temperature, °C 373±8 354±1 363±4 380±2 Middle SCR temperature, °C 365±5 342±1 352±3 364±1 Bottom SCR temperature, °C 323±4 304±2 315±2 325±1 Measured CO, ppm (dry) 38±6 30±7 40±4 0±1 Uncontrolled NOx, ppm (dry) 960 850 650 525 NOx conversion across SCR, % 90 86 90 90 Measured S02, ppm (dry) 2921±49 929±63 2739±26 222±2 Calculated S02, ppm (dry) 3064 915 3006 274 Measured 02, % (dry) 2.7 4.0 4.4 4.1 Calculated 02, % (dry) 1.8 3.0 1.9 0.1 In-leakage based on 02, % 5 5 13 19 Measured C02, % (dry) 15.6 14.8 14.3 15.6 Calculated C02, % (dry) 16.0 15.4 15.7 18.0 In-leakage based on C02, % 3 3 9 13 Calculated flow based on 02 in-leakage, scm/min 0.879 0.896 0.944 1.008 Calculated flow based on C02 in-leakage, scm/min 0.858 0.853 0.911 0.856 Measured HC1, ppm (wet) NMa 246±23 NM NM Calculated HC1, ppm (wet) 141 208 96 7.9 SCR inlet PM concentrationb, mg/dscm 5863 3070±555 3946 2418±504 SCR outlet PM concentration13, mg/dscm 3506 1735±194 2560 1606±160 a NM = not measured. b Measured using the filter weight of the isokinetic OH method at the inlet and outlet of the SCR. collected in the filter of the OH train may cause catalytic reaction or retention of mercury, resulting in a change in mercury speciation. Therefore, the IGS probe can be used for sampling gas-phase Hg° and Hg2+, and the difference between Hgx obtained by the isokinetic train and the gas-phase Hg° and Hg2+ obtained by the IGS probe can be used to estimate the amount of HgP. During the initial coal combustion test, the total axial flow for the IGS probe installed at the inlet and the outlet of the SCR reactor dropped from between 7 and 8 to below 2.7 scfm. The total axial flow of the inlet of the IGS probe dropped from 5 to 3 scfm. This drop in the flow might have caused retention of particulate matter (PM, fly ash) in the probes. This might have created an unwanted packed bed in the IGS probe, thus positively biasing the HgP results, although the results of the 2nd test (see below) showed that this was not the case. The problem of drop in sampling flow was corrected by installing an air pump at the outlet for each of the IGS probes. The combination of the pump and the eductor (operated with compressed air) maintained the desired level of axial flow through the IGS probes. Biases created by packed beds of fly ash are not expected because the 20 ------- Evaluation of SCR Catalysts high axial flow prevents formation of such packed beds in the probe. The total axial flow in the IGS probes was maintained at around 7-8 scfm. Results of the simultaneous isokinetic and IGS probe OH sampling for the Galatia test are summarized in Tables 8 and 9. The flow through the IGS probes was maintained at around 8.7 scfm. Relatively good Hgx recovery was obtained across the SCR catalyst. The Hgx at the outlet was slightly lower than the inlet (7.2 vs. 9.9 |ig/dscm). This could be due to adsorption of Hg by the SCR catalysts. Comparison of the data obtained by using the iso- kinetic trains and those obtained by using IGS probes suggests that a considerable portion of the Hg in the Galatia coal combustion may be associated with the PM at the inlet of the SCR catalyst. However, the adsorption of mercury by fly ash is unlikely to be significant at the SCR inlet temperatures (354 °C), which is supported by recent field data that showed very little HgP at the inlet of various SCR systems firing different types of bituminous coals (Brickett et al., 2003). Deposition of white powder was observed on the walls of the first impinger connected to the IGS probes, indicating formation of fine particulate aerosols near the inlet of the impinger. Much more white powder was deposited in the impinger used for measuring the inlet of the SCR reactor than in the impinger connected to the outlet train. Ammonium salts formed near the inlet of the first impinger connected to the IGS probe as the NH3-containing flue gas cooled down, which could retain gas-phase Hg and result in changes in Hg speciation. It is well known that the reaction of NH3 with sulfuric acid (H2S04) condensed from S03 as the coal combustion flue gas cools down produces ammonium sulfate and bisulfate salts, which could cause severe air pre- heater plugging problems (Burke and Johnson, 1982). The capture of Hg° by the aerosols generated by the reactions of NH3 and S03 under flue gas cooling conditions has been demonstrated by a recent bench-scale study (Lee, et al., 2003). The NH3 concentration in flue gas decreased from about 850 Table 8. Galatia Coal Combustion Test, OH Sampling Results Measured at the Inlet of the SCR Reactor. Inlet OH Train Hg° Hg2+ HgP HgT Hg° Hg2+ HgP |i,g/dscm |i,g/dscm |i,g/dscm |i,g/dscm % % % Isokinetic 7.2 2.1 0.61 9.9 73 21 6 IGS probe 1.7 3.9 — — — — — Calculated inlet Ha SDCciatioiV 1.7 3.9 4.3 9.9 17 40 43 a Calculated assuming that isokinetic train is used for measuring Hgx and IGS probe is used for measuring Hg° and Hg2+. Table 9. Galatia Coal Combustion Test, OH Sampling Results Measured at the Outlet of the SCR Reactor. ^ ... T H§0 H§2+ HgP HgT Hg° Hg2+ HgP Outlet OH Train 6 , , ? 0, 0f [Xg/dscm p,g/dscm p,g/dscm p,g/dscm % % % Isokinetic 0.32 5.1 1.8 7.2 4 71 25 IGS probe 0.1 5.4 — — — — — Calculated outlet Hg SDeciationa 0.1 5.4 1.7 7.2 1 75 24 a Calculated assuming that isokinetic train is used for measuring Hgx and IGS probe is used for measuring Hg° and Hg2+. 21 ------- Evaluation of SCR Catalysts ppm at the inlet of the SCR reactor to a few ppm at the reactor outlet as a result of the NOx reduction reactions that consumed NH3. Thus, the potential for deposition of ammonium salts in the sampling system used at the inlet of the SCR reactor is significantly higher than that used at the reactor outlet. The dis- crepancies in Hg speciation measured by the iso- kinetic OH train and the calculated speciation are much larger for the SCR inlet case (Table 8) than those for the outlet case (Table 9). The larger discrep- ancies for the inlet case are consistent with more white powder deposition found in the first impinger of the inlet IGS probe. During the initial shakedown tests of the bench-scale reactor (Task I), white pow- ders were found deposited near the inlet of an on-line mercury analyzer connected to the SCR outlet for continuously measuring Hg°, indicating aerosol for- mation. The white powder deposits were observed only when NH3 was added to the simulated flue gases, and the data measured by using the analyzer showed continuous drift as the deposit was accumu- lating during the test. Therefore, the OH sampling method was used for the Task I tests in order to avoid the measuring problem created by the aerosol forma- tion when using the mercury analyzer. It appears that the formation of ammonium salts in the Hg speciation sampling/measurement system in the presence of NH3 and under cooling conditions may cause significant bias of the speciation measurement. Further research is needed for applying the IGS probe to measure Hg speciation in SCR systems. Therefore, it was decided to not use IGS probe in this work. The OH sampling results show very good oxidation of Hg° across the SCR catalyst. Hg° was decreased from 73% to 4% and Hg2+ was increased from 21% to 7P/o across the SCR catalyst. The change in Hg speciation from predominantly Hg° at the inlet of the catalyst to predominantly Hg2+ at the outlet was observed in SCR systems on boilers firing bituminous coals (Brickett, et al., 2003). The isokinetic OH filter weights can be used to determine the PM concentra- tion in the IFR. The average PM concentrations for the Galatia coal combustion at the inlet and the outlet of SCR reactor are shown in Table 7. The lower PM concentration at the outlet may be due to the opera- tion of the IGS probe at the inlet. This probe draws a high flow from the duct and, thus, may preferentially take higher amounts of PM from the flue gas. Turris Coal Combustion Turris coal was combusted at a firing rate of about 148,000 Btu/hr. Combustor air in-leakage was about 4%. There was no in-leakage between the inlet and the outlet of the SCR. A very good combustion of Turris coal was observed as indicated by the low CO emissions (38 ppm). The axial temperature gradient across the SCR catalyst was around 50 °C (change from 373 to 323 °C). Slightly higher temperatures were obtained around the SCR catalyst as compared to the Galatia coal combustion test. Uncontrolled levels of NOx were around 950 (higher than Galatia). NH3 was injected at an NH3/NOx ratio of about 0.9. NOx conversion across the SCR was about 90%>. S02 concentration for this high sulfur Illinois coal was measured to be 2921±49 ppm. The calculated level of S02 under the combustion conditions was 3046 ppm, reflecting good agreement with measured values. Table 10 summarizes the results of the OH sampling Table 10. Turris Coal Combustion Test, OH Sampling Results Measured at the Inlet and Outlet of the SCR Reactor. Location Hg° Hg2+ Hgp H§T Hg° Hg2+ Hg" |Xg/dscm ng/dscm |lg/dscm |j,g/dscm % % % Inlet 4*7 06 L5 6*9 68 9 23 Outlet 03 M 04 3J 9 79 12 22 ------- Evaluation of SCR Catalysts for the Turns coal test at the inlet and the outlet of the SCR reactor. Only about half of the inlet Hgx was recovered at the outlet. This could be due to the adsorption of mercury by the catalyst. Very little Hg was associated with the particulate matter. The SCR catalyst exhibited a strong Hg° oxidation capability, dropping the amount of Hg° from 68% at the inlet to 9% at the outlet. The Hg2+ was increased from 9% to 79% at the two sampling locations. Such significant changes in Hg speciation are in agreement with those observed in the field for bituminous coal-fired SCR systems (Brickett, et al., 2003). Crown II Coal Combustion Crown II coal was combusted at a firing rate of about 145,000 Btu/hr. Combustor air in-leakage was about 11%). It appears that, as the combustor operated for a longer period of time, air in-leakage increased, but this level of air in-leakage is considered insignificant. There was no air in-leakage between the inlet and the outlet of the SCR. A very good combustion of Crown II coal was indicated by low measured CO emissions (40 ppm). The axial temperature gradient across the SCR catalyst was around 48 °C (change from 363 to 315 °C). SCR temperatures were within the range for those obtained during the other two Illinois coal tests. Uncontrolled levels of NOx were about 650 ppm, lower than those for the Galatia and Turris tests. NH3 was injected at a calculated NH3/NOx ratio of about 0.8. NOx conversion across the SCR was about 90%>. S02 emissions for this high sulfur Illinois coal tests measured 2739±26 ppm. The calculated level of S02 emission under these test conditions was 3006 ppm. Table 11 summarizes the results of the OH sampling for the Crown II coal combustion test at the inlet and the outlet of the SCR reactor. Relatively low Hgx recovery was obtained across the SCR catalyst. The Hgx concentration measured at the outlet was lower than that measured at the inlet (4.1 vs. 5.5 |ig/dscm). The lower HgT measured at the outlet of the SCR catalyst may be due to the adsorption of Hg by the catalyst. Very little Hg was associated with the PM. This was also found during the combustion of Turris coal. Crown II and Turris coals are very similar in their properties (see Table 1). Galatia was the only Illinois coal that generated relatively high HgP during its combustion. The SCR catalyst exhibited a very good Hg° oxidation capability. Hg° was reduced from 84%o at the inlet down to 12%> at the outlet, with the increase in Hg2+ from 15%> to 85%> across these two sampling locations. The changes in Hg speciation across the SCR catalyst observed for the Crown II coal combustion are consistent with those of the two other Illinois coal tests and also those of the bitumi- nous coal-fired SCR systems observed in field (Brickett, et al., 2003). PRB (Black Thunder) Coal Combustion Black Thunder PRB coal was combusted in duplicate runs. The combustion conditions were identical and are listed in Table 7 under the PRB column. PRB coal was combusted at a firing rate of about 156,000 Btu/hr. This firing rate was slightly higher than those used for the Illinois coal tests. In-leakage of air into the combustor was about 16%>. Similar to the Crown II test, as the combustor operated for a longer period of time, in-leakage seemed to increase, but the levels of air in-leakage are considered to be acceptable. There was no in-leakage between the inlet and the Table 11. Crown II Coal Combustion Test, OH Sampling Results Measured at the Inlet and Outlet of the SCR Reactor. Location Hg° Hg2+ Hgp H§T Hg° Hg2+ Hg" |Xg/dscm ng/dscm |lg/dscm |j,g/dscm % % % Inlet 4*6 08 01 ~ 84 15 1 Outlet 05 3J 01 4J 12 85 3 23 ------- Evaluation of SCR Catalysts outlet of the SCR. A very good combustion of the PRB coal was observed although no CO emissions were measured during the two combustion tests. The axial temperature gradient across the SCR catalyst was about 55 °C (change from 380 to 325 °C). SCR temperatures were slightly higher than those observed during Illinois coal tests, possibly due to slightly higher firing rate. Uncontrolled levels of NOx emis- sions were about 525 ppm, which are lower than those measured for the Illinois coal tests. NH3 was injected at a calculated NH3/NOx ratio of about 0.8. NOx conversion across the SCR was about 90%. S02 emissions for the combustion of this low sulfur coal was measured to be 222±2 ppm. The calculated level of S02 under the combustion conditions was 274 ppm, indicating good agreement with the measured values. Tables 12 and 13 summarize the results of the OH sampling for the first and second PRB coal combus- tion tests, respectively. A good Hgx recovery was obtained across the SCR catalyst during the second PRB test (5.1 |ig/dscm at the inlet vs. 4.6 |ig/dscm at the outlet), which was better than those obtained during the first test (7.2 |ig/dscm at the inlet vs. 5.1 |ig/dscm at the outlet). The inlet Hgx obtained during the first test (7.2 |ig/dscm) is also significantly higher than that measured at the same location for the second test, which may be an outlier. It appears that almost all Hg was in elemental form at the inlet with no appreciable Hg2+ or HgP. The comparison of data measured between inlet and outlet of the catalyst shows only a very small drop in Hg°, from 96% to 76%) for the second test and from 97% to 88% for the first test. This is equivalent to only a 20% oxidation of Hg° (9%> oxidation of Hg° for the first test). Field sampling conducted to date indicates the same Hg speciation profile as shown by this study, mostly Hg° at the inlet and very low Hg° oxidation. PRB coals in general have significantly lower chlorine and higher alkaline contents than those of the bituminous coals. The low oxidation of Hg° across the SCR catalyst observed for the two PRB coal tests supports the hypothesis that HC1, which serves as the chlorine source, is critical in Hg° oxidation across the SCR catalyst. The importance of HC1 in Hg oxidation across the SCR catalyst had been demonstrated in Task I. Table 12. First PRB Coal Combustion Test, OH Sampling Results Measured at the Inlet and Outlet of the SCR Reactor. Location Hg° |Xg/dscm Hg2+ |Xg/dscm HgP |Xg/dscm HgT |Xg/dscm Hg° % Hg2+ % HgP % Inlet 7.0 0.2 0.0 7.2 97 3 0 Outlet 4.5 0.5 0.1 5.1 88 10 2 Table 13. First PRB Coal Combustion Test, OH Sampling Results Measured at the Inlet and Outlet of the SCR Reactor. Location Hg° |Xg/dscm Hg2+ |Xg/dscm HgP |Xg/dscm HgT |Xg/dscm Hg° % Hg2+ % HgP % Inlet 4.9 0.2 0.0 5.1 96 4 0 Outlet 3.5 1.0 0.1 4.6 76 22 2 24 ------- Evaluation of SCR Catalysts Conclusions and Recommendations A two-task, bench- and pilot-scale research study was conducted to evaluate the viability of SCR-based mercury control technology utilizing oxidation of Hg° to Hg2+ and subsequent removal by conventional FGD systems. Mercury speciation modification was implemented through utilization of SCR catalysts, currently used for NOx control. The effect of SCR catalysts on mercury speciation was investigated in Illinois and PRB coal combustion flue gases. This research was designed to demonstrate the achievable Hg2+ levels in Illinois and PRB coal combustion processes utilizing SCR systems. Task I involved bench-scale testing of a SCR catalyst formulation to characterize oxidation of Hg° to Hg2+ in the presence of coal combustion/SCR flue gas species such as NOx, NH3, S02, and HC1. The NOx, NH3, HC1, and S02 concentration ranges employed during Task I were based on those expected to be encountered in flue gases resulting from firing three representative Illinois coals (low to high sulfur and chlorine) and one PRB coal (very low sul- fur/chlorine). Task I results showed that HC1 (with concentrations as low as 8 ppm) is the critical flue gas component for converting Hg° to Hg2+ under SCR emission control conditions. Hg2+ was measured as the predominant species at the outlet of the SCR catalyst for all the simulated Illinois and PRB coal tests. Task I tests established the Hg° oxidation activity of the formulated SCR catalyst. Pilot-scale Task II tests were performed using a SCR catalyst installed in a vertical down-ward flow configuration in the post-combustion region of a pilot-scale 34.9 kW (150,000 Btu/hr), refractory lined, down-fired cylindrical furnace. Three different Illinois coals and one PRB coal were combusted in the pilot-scale facility. For the high sulfur/low chlo- rine Illinois coals, about 84% to 68 % Hg°, and 15% to 9% Hg2+ were measured at the inlet of the pilot- scale SCR reactor. The SCR catalyst induced high oxidation of Hg°, dropping the percentage of Hg° at the outlet of the SCR to values below 16%. The low sulfur/high chlorine Illinois coal had a relative high amount (21%) of Hg2+ at the inlet of SCR. The Hg° content of this coal flue gas was decreased from about 73% at the inlet to about 4% at the outlet of SCR reactor. These results indicate that installation of SCR systems in Illinois coal-fired boilers that are equipped with a wet FGD system may achieve about 85%) to greater than 90% control of mercury. The PRB coal tests indicated a low oxidation of Hg° by the SCR catalyst. Based on the measurements conducted in this study, it appears that SCR applica- tions on PRB (Black Thunder) coal-fired boilers may not result in significant increase in Hg2+ content of flue gas. Therefore, such boilers equipped with wet FGD systems may not achieve increased mercury removal resulting from SCR applications. The following recommendation are made based on the results of this one-year study: • For most typical Illinois coals (high sulfur and medium to low chlorine), installation of SCR catalysts prior to FGD systems should be very beneficial. In addition to achieving a 90% reduc- tion in NOx, about 85% to greater than 90% reduc- tion in mercury may be possible. • A particular low sulfur/high chlorine Illinois coal (Galatia) showed relatively high Hg2+ content at the SCR inlet (21%); the rest was predominantly Hg° (73%>) and very little HgP. However, the Hg con- tents at the SCR outlet were changed to 71% Hg2+, 25%) HgP, and very little Hg°. Based on these data, 25 ------- Evaluation of SCR Catalysts installation of an SCR catalyst may results in greater than 90% reduction of Hg across the PM control and the wet FGD systems. • The benefit of installing a SCR catalyst for the tested PRB coal (Black Thunder) appears to be very limited. Installation of an SCR catalyst could only increase the mercury removal to a maximum of 20%. • Further tests are required to evaluate the long-term activity of the tested SCR catalysts in both Illinois and PRB coal combustion flue gases. • Tests should be conducted to evaluate the effect of SCR catalyst space velocity on Hg° oxidation in Illinois and PRB coal combustion flue gas. Higher space velocities (greater than 3000 hr-1) should be tested to demonstrate further cost-effectiveness of this technology. 26 ------- Evaluation of SCR Catalysts References Brickett, L., P. Chu, C.W. Lee, R.K. Srivastava, D. Laudal, J. Thompson, and C. Wocken, 2003."Impact of SCR on Mercury Speciation for Coal-fired Boil- ers." Presented at the Combined Power Plant Air Pollutant Control Mega Symposium, Washington, DC, May 19-22. Burke, J.M., and K.L. Johnson, 1982. "Ammonium Sulfate and Bisulfate Formation in Air Preheaters." EPA Report number EPA-600/7-82-025a (NTIS PB82-237025), April. Cichanowicz, J.E., and L.J. Muzio, 2001. "Twenty- Five Years of SCR Evolution: Implications For U.S. Application And Operation." Presented at the MEGA Symposium, Chicago, IL, August 20-23. DeVito, M.S., P.R. Tumati, R.J. Carlson, and N. Bloom, 1993. "Sampling and Analysis of Mercury in Combustion Flue Gas." Presented at the Second International Conference on Managing Hazardous Air Pollutants, Washington, D.C., July 13-15. EPRI, 2000. "Pilot-Scale Screening Evaluation of the Impact of Selective Catalytic Reduction for NOx on Mercury Speciation." EPRI Report No. 1000755, December. Felsvang, K., R. Gleiser, G. Julip, and K. Kragh Nielson, 1992. "Control of Air Toxics by Dry FGD Systems." Presented at the Power-Gen '92 Confer- ence, Orlando, FL, November 17-19. Felsvang, K., R. Gleiser, G. Julip, K. Kragh Nielsen, 1993. "Air Toxics Control by Spray Dryer." Pre- sented at the 1993 S02 Control Symposium, Boston, MA, August 24-27. Galbreath, K.C., and C.J. Zygarlicke, 2000. "Mercury Transformation in Coal Combustion Flue Gas." Fuel Process. Technol., 65-66, pp. 289-310. Ghorishi, S.B., C.W. Lee, and J.D. Kilgroe. 1999. "Mercury Speciation in Combustion Systems: Studies with Simulated Flue Gases and Model Fly Ashes." Paper # 99-651 Presented at the 92nd Annual Meet- ing of Air and Waste Management Association, St. Louis, MO, June 20-24. Ghorishi, S.B., 1998. "Fundamentals of Mercury Speciation and Control in Coal-Fired Boilers." EPA Report number EPA-600/R-98-014 (NTIS PB98- 127095), February. Ghorishi, S.B., andB.K. Gullett, 1998. "Sorption of Mercury Species by Activated Carbons and Cal- cium-Based Sorbents: Effect of Temperature, Mer- cury Concentration and Acid Gases." Waste Manage- ment & Research, 16 (6), pp. 582-593. Gutberlet, H., A. Schliiten, and A. Lieutal, 2000. "SCR impacts on Mercury Emissions on Coal-Fired Boilers." Presented at EPRI Workshop, Memphis, TN, April. Hall, B., P. Schager, and O. Lindqvist, 1991. "Chem- ical Reactions of Mercury in Combustion Flue Gases." Water, Air, and Soil Pollution, 56, 3-14. Krishnan, S.V., B.K. Gullett, and W. Jozewicz, 1994. "Sorption of Elemental Mercury by Activated Car- bons." Env. Sci. & Tech. 28 (8), pp. 1506-1512. Krishnan, S.V., H. Bakhteyar, and C.B. Sedman, 1996. "Mercury Sorption Mechanisms and Control by Calcium-Based Sorbents." Paper 96-WP64B.05 27 ------- Evaluation of SCR Catalysts presented at the 89th Air & Waste Management Association Annual Meeting, Nashville, TN, June 23-28. Laudal, D.L., C.R. Wocker, P. Chu, L.A. Brickett, and C.W. Lee, 2003. "Evaluation of the Effect of SCR on Mercury Speciation and Emissions." Pre- sented at the Air Quality IV Conference, Arlington, VA, September 22-24. Lee, Y.P., S.J. Khang, and T.C. Keener, 2003. "Mercury Removal from Flue Gas with Aerosols Generated by S03-NH3 Reactions." Paper 69342 presented at the 96th Annual Meeting and Exhibition, AWMA, San Diego, CA, June 22-26. Linak, W.P., and J.O.L. Wendt. 1993. "Toxic Metal Emissions from Incineration: Mechanisms and Con- trol." Progress in Energy & Combustion Science, 19, pp. 145-185. Markowski, G.R., and R. Filby. 1985. "Trace Ele- ment Concentration as a Function of Particle Size in Fly Ash from a Pulverized Coal Utility Boiler." Env. Sci. Tech., 19 (9), 796-804. Senior, C.L., E.L., Bool, G.P., Huffman, F.E., Huggins, N., Shah, A., Sarofim I., Olmez, and T. Zeng. 1997. "A Fundamental Study of Mercury Partitioning in Coal-Fired Power Plant Flue Gas." Paper 97-WP72B.08 presented at the 90th Air & Waste Management Association Annual Meeting, Toronto, Ontario, Canada, June 8-13. Volland, C. 1991. "Mercury Emission from Munici- pal Solid Waste Combustion." Paper # 91-35.1 presented at the 84th Annual Meeting and Exhibition, AWMA, Vancouver, BC, June 16-21. 28 ------- TECHNICAL REPORT DATA (Please read Instructions on the reverse before completing) 1. REPORT NO. EPA-600/R-04/130 2. 3. RECIPIENT'S ACCESSION NO. 4. TITLE AND SUBTITLE Evaluation of SCR Catalysts for Combined Control of NOx and 5. REPORT DATE September 2004 Mercury 6. PERFORMING ORGANIZATION CODE 7. AUTHORS R.K. Srivastava 8. PERFORMING ORGANIZATION REPORT NO. 9. PERFORMING ORGANIZATION NAME AND ADDRESS See Block 12. 10. PROGRAM ELEMENT NO. 11. CONTRACT/GRANT NO. In-house 12. SPONSORING AGENCY NAME AND ADDRESS U. S. EPA, Office of Research and Development 13. TYPE OF REPORT AND PERIOD COVERED Final; 09/01/02-08/31/03 Air Pollution Prevention and Control Division Research Triangle Park, North Carolina 27711 14. SPONSORING AGENCY CODE EPA/600/13 15. SUPPLEMENTARY NOTES The EPA Project Officer is Ravi K. Srivastava, mail drop E305-01, phone (919) 541-3444, e-mail srivastava.ravi@epa.gov. 16. ABSTRACT The report documents two-task, bench- and pilot-scale research on the effect of selective catalytic reduction (SCR) catalysts on mercury speciation in Illinois and Powder River Basin (PRB) coal combustion flue gases. In task I, a bench-scale reactor was used to study the oxidation of elemental mercury (Hg°) in simulated Illinois and PRB coal flue gases by a vanadium/titanium SCR catalyst. Elemental mercury oxidation on the order of 90% was observed in all the simulated coal flue gases. It was shown that hydrogen chloride (HCI) may be the critical flue gas component that causes the conversion of Hg° to oxidized mercury (Hg2+) under SCR reaction conditions. Since high HCI concentrations are expected during combustion of all Illinois coals, firing these coals should result in significant Hg° oxidation occurring across SCR reactors. Based on bench-scale results, an appropriate SCR catalyst was produced in pilot-scale and installed in a pilot-scale combustor for Task II studies. Three different Illinois coals (from high to low sulfur and chlorine) and one PRB coal were combusted in the pilot-scale unit. For the high sulfur/low chlorine Illinois coals, about 68%-84% Hg° was measured at the inlet of the pilot-scale SCR reactor and below 16% Hg° at the outlet of the SCR. Flue gas from the combustion of low sulfur/high chlorine Illinois coal had 21% Hg2+ at the inlet of the SCR, and the Hg° content was decreased from about 73% at the inlet of the SCR reactor to about 4% at the outlet. For the PRB coal tests, the percentage of Hg° decreased from about 96% at the inlet of the SCR reactor to about 80% at the outlet. 17. KEYWORDS AND DOCUMENT ANALYSIS a. DESCRIPTORS b. IDENTIFIERS/OPEN ENDED TERMS c. COSATI Field/Group Air Pollution Coal Combustion Exhaust Gases Mercury (metal) Oxidation Reduction Reactions Catalysis Pollution Control Stationary Sources 13B 08G, 21D 21B 14G 07 B 07C 07 D 18. DISTRIBUTION STATEMENT 19. SECURITY CLASS (This Report) Unclassified 21. NO. OF PAGES 36 Release to Public 20. SECURITY CLASS (This Page) Unclassified 22. PRICE EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDITION IS OBSOLETE forms/admin/techrpt.frm 7/8/99 pad ------- |