EPRI Impact of Air Emissions Controls on Coal Combustion Products

Impact of Air Emissions Controls on Coal Combustion Products

E c o n o m ize r F u rn a c e E c o n o m ize r A s h C o lle c tio n

F G D A b s o rb e r

A ir H e a te r P artic u la te C o lle c to r B o tto m A s h /S la g C o lle c tio n

F ly As h C o lle c tio n

F G D G yp s u m C o lle c tio n

S tac k

Impact of Air Emissions Controls on Coal Combustion Products 1015544

Technical Report, October 2008

Cosponsor Ameren Corporation One Ameren Plaza 1901 Choteau Avenue St. Louis, MO 63166-6149 Project Manager P. Pike

EPRI Project Manager K. Ladwig

ELECTRIC POWER RESEARCH INSTITUTE 3420 Hillview Avenue, Palo Alto, California 94304-1338 • PO Box 10412, Palo Alto, California 94303-0813 • USA 800.313.3774 • 650.855.2121 • [email protected] • www.epri.com

DISCLAIMER OF WARRANTIES AND LIMITATION OF LIABILITIES THIS DOCUMENT WAS PREPARED BY THE ORGANIZATION(S) NAMED BELOW AS AN ACCOUNT OF WORK SPONSORED OR COSPONSORED BY THE ELECTRIC POWER RESEARCH INSTITUTE, INC. (EPRI). NEITHER EPRI, ANY MEMBER OF EPRI, ANY COSPONSOR, THE ORGANIZATION(S) BELOW, NOR ANY PERSON ACTING ON BEHALF OF ANY OF THEM: (A) MAKES ANY WARRANTY OR REPRESENTATION WHATSOEVER, EXPRESS OR IMPLIED, (I) WITH RESPECT TO THE USE OF ANY INFORMATION, APPARATUS, METHOD, PROCESS, OR SIMILAR ITEM DISCLOSED IN THIS DOCUMENT, INCLUDING MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE, OR (II) THAT SUCH USE DOES NOT INFRINGE ON OR INTERFERE WITH PRIVATELY OWNED RIGHTS, INCLUDING ANY PARTY'S INTELLECTUAL PROPERTY, OR (III) THAT THIS DOCUMENT IS SUITABLE TO ANY PARTICULAR USER'S CIRCUMSTANCE; OR (B) ASSUMES RESPONSIBILITY FOR ANY DAMAGES OR OTHER LIABILITY WHATSOEVER (INCLUDING ANY CONSEQUENTIAL DAMAGES, EVEN IF EPRI OR ANY EPRI REPRESENTATIVE HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES) RESULTING FROM YOUR SELECTION OR USE OF THIS DOCUMENT OR ANY INFORMATION, APPARATUS, METHOD, PROCESS, OR SIMILAR ITEM DISCLOSED IN THIS DOCUMENT. ORGANIZATION(S) THAT PREPARED THIS DOCUMENT URS Corporation

NOTE For further information about EPRI, call the EPRI Customer Assistance Center at 800.313.3774 or e-mail [email protected]. Electric Power Research Institute, EPRI, and TOGETHER…SHAPING THE FUTURE OF ELECTRICITY are registered service marks of the Electric Power Research Institute, Inc. Copyright © 2008 Electric Power Research Institute, Inc. All rights reserved.

CITATIONS This report was prepared by URS Corporation P.O. Box 201088 Austin, TX 78720 Principal Investigators G. Blythe J. Paradis

This report describes research sponsored by the Electric Power Research Institute (EPRI) and Ameren Corporation. The report is a corporate document that should be cited in the literature in the following manner: Impact of Air Emissions Controls on Coal Combustion Products. EPRI, Palo Alto, CA: 2008. 1015544.

iii

REPORT SUMMARY

Coal combustion products (CCPs) have been extensively studied and well characterized over the last 30 years. However, new air emissions control technologies at power plants will change the characteristics of some existing CCPs. These changes may affect the selection of appropriate management methods for high-volume CCPs with respect to both disposal and use. This report examines evolving air emissions controls and their likely impact on CCPs. Background Extensive environmental characterization of the four high-volume CCPs—fly ash, bottom ash, boiler slag, and flue gas desulfurization (FGD) solids—extends back to the 1970s, and accelerated following the initial enactment of the Resource Conservation and Recovery Act (RCRA) governing nonhazardous and hazardous waste disposal. While a significant database has been accumulated during this time period, recent changes in air emissions controls are resulting in changes to CCP engineering and environmental characteristics. Combined with tightening regulation of trace constituents such as mercury and arsenic, these changes have the potential to impact disposal practices and potential use opportunities. Objectives To summarize power plant environmental control systems, the most significant likely changes in CCP characteristics resulting from these technologies, and potential impacts on CCP management. Approach Researchers evaluated the range of air emissions control technologies currently being considered for power plants. These were grouped as controls on nitrogen oxides (NOX), sulfur trioxide/sulfuric acid (SO3/H2SO4), and mercury. These technologies are described, and their potential impacts on the four high-volume CCPs are discussed. In addition, ancillary solid wastes arising from the generation and management of coal combustion products are described. Results NOX control consists primarily of four technology types: furnace combustion modifications, which for the purposes of this report include the use of low-NOX burners; rich reagent injection (RRI), which is implemented in the furnace; selective noncatalytic reduction (SNCR), which is implemented in the upper furnace; and selective catalytic reduction (SCR), which is implemented between the economizer and air heater. The primary impacts of these control technologies are an increase in unburned carbon, which can have an adverse impact on fly ash use in concrete, and addition of ammonia to the fly ash.

v

Increased ammonia could cause odor issues that affect both ash use and disposal operations, although there are methods available to mitigate this problem. SO3 is formed in the furnace and in downstream SCR systems, and the most tangible manifestation of the resulting sulfuric acid in the flue gas is stack plume visibility. Consequently, SO3 controls can be implemented anywhere in the flue gas path from the furnace to just upstream of the stack. The most common technologies for mitigating SO3 plumes involve addition of alkali materials, such as magnesium compounds, sodium carbonate, trona, lime, limestone, ammonia, or sodium bisulfite, in the flue gas. The primary impacts on the fly ash are increases in alkali content, increases in pH, and addition of soluble salts of sodium, magnesium, calcium, and sulfate. These could render the ash unusable in concrete due to the higher alkali content, and may require additional ash pond treatment and disposal site engineering measures due to the higher soluble salt content and changes in trace constituent mobility at the higher pH. Mercury removal can affect both fly ash and FGD solids. Mercury captured in the fly ash is relatively stable at normal temperatures and is not expected to pose a significant environmental or use concern. However, the use of activated carbon may affect the ability to use the ash in concrete due to greater need for air-entraining agents. Alternatives are non-carbon sorbents, treated carbon sorbents, and fly ash beneficiation to remove or passivate the carbon. Additionally, re-release of mercury at high temperatures in cement kilns may limit fly ash use in some cement applications. Mercury also may be captured in the FGD system. Research is currently underway on the partitioning of mercury in the absorber and on evaluating the effect of increased mercury concentrations on FGD solids used in wallboard and agricultural applications. EPRI Perspective Both power plant control technologies and environmental regulations are continually evolving. A key component of EPRI’s research is to understand the impact of emerging control technologies on CCP characteristics, and to develop methods to preserve or enhance CCP management opportunities. This information can be used by power companies to make informed decisions with respect to implementation of control technologies, to assess the impacts that they will have on CCP management, and to assess alternative CCP management strategies. Keywords Coal combustion products Characterization Use Disposal Emissions controls

vi

CONTENTS

1 INTRODUCTION ....................................................................................................................1-1 2 HIGH-VOLUME COAL COMBUSTION PRODUCTS ............................................................2-1 Fly Ash ..................................................................................................................................2-2 Flue Gas Conditioning for Fly Ash Capture ......................................................................2-3 Fly Ash Resistivity Conditioning...................................................................................2-3 Fly Ash Cohesivity Conditioning ..................................................................................2-4 Bottom Ash............................................................................................................................2-5 Boiler Slag .............................................................................................................................2-5 FGD Solids ............................................................................................................................2-6 Wet FGD Systems............................................................................................................2-6 Inhibited or Low Natural Oxidation FGD Systems .......................................................2-8 Forced Oxidation or High Natural Oxidation Systems .................................................2-9 Dry FGD Systems...........................................................................................................2-11 Spray Drying Systems ...............................................................................................2-11 Dry Injection Systems ................................................................................................2-12 Furnace Sorbent Injection (FSI) Systems ..................................................................2-12 References ..........................................................................................................................2-13 3 NOX CONTROL TECHNOLOGY ............................................................................................3-1 Combustion Modifications/Low-NOX Burners ........................................................................3-1 Technology Description ....................................................................................................3-1 Impact on CCPs ...............................................................................................................3-1 Rich Reagent Injection (RRI).................................................................................................3-2 Technology Description ....................................................................................................3-2 Impact on CCPs ...............................................................................................................3-2 Selective Non-Catalytic Reduction (SNCR) ..........................................................................3-3 Technology Description ....................................................................................................3-3

vii

Impact on CCPs ...............................................................................................................3-3 Selective Catalytic Reduction (SCR).....................................................................................3-5 Technology Description ....................................................................................................3-5 Impact on CCPs ...............................................................................................................3-5 References ............................................................................................................................3-6 4 SO3 CONTROL TECHNOLOGY.............................................................................................4-1 Fuel Additives........................................................................................................................4-2 Technology Description ....................................................................................................4-2 Impact on CCPs ...............................................................................................................4-3 Furnace Injection of Magnesium-Based Sorbents ................................................................4-3 Technology Description ....................................................................................................4-3 Impact on CCPs ...............................................................................................................4-4 Magnesium Oxide Injection Upstream of the Air Heater .......................................................4-5 Technology Description ....................................................................................................4-5 Impact on CCPs ...............................................................................................................4-5 TM

CleanStack Condensing Air Heater ....................................................................................4-6 Technology Description ....................................................................................................4-6 Impact on CCPs ...............................................................................................................4-7 Sodium Based Sorbent (SBS) Injection ................................................................................4-8 Technology Description ....................................................................................................4-8 Impact on CCPs ...............................................................................................................4-9 Ammonia Injection at the Air Heater Outlet .........................................................................4-10 Technology Description ..................................................................................................4-10 Impact on CCPs .............................................................................................................4-12 Dry Sorbent Injection at the Air Heater Outlet .....................................................................4-12 Hydrated Lime Powder Injection.....................................................................................4-12 Technology Description .............................................................................................4-12 Impact on CCPs.........................................................................................................4-13 Trona Injection................................................................................................................4-14 Technology Description .............................................................................................4-14 Impact on CCPs.........................................................................................................4-15 Wet ESP..............................................................................................................................4-15 Technology Description ..................................................................................................4-15 Impact on CCPs.........................................................................................................4-16

viii

References ..........................................................................................................................4-16 5 MERCURY REMOVAL...........................................................................................................5-1 Mercury Co-Removal with Fly Ash ........................................................................................5-1 Combustion Modifications ................................................................................................5-1 Technology Description ...............................................................................................5-1 Impact on CCPs...........................................................................................................5-3 Activated Carbon Injection (ACI) ......................................................................................5-3 Technology Description ...............................................................................................5-3 Impact on CCPs...........................................................................................................5-4 ®

®

Toxecon and Toxecon II ................................................................................................5-6 Technology Description ...............................................................................................5-6 Impact on CCPs...........................................................................................................5-7 Non-Carbon Sorbent Injection ..........................................................................................5-8 Technology Description ...............................................................................................5-8 Impact on CCPs...........................................................................................................5-8 Mercury Co-Removal with FGD By-Products........................................................................5-8 Chemical Addition.............................................................................................................5-9 Technology Description ...............................................................................................5-9 Impact on CCPs.........................................................................................................5-10 Mercury Oxidation Catalysts...........................................................................................5-11 Technology Description .............................................................................................5-11 Impact on CCPs.........................................................................................................5-12 Other Mercury Oxidation Technology.............................................................................5-12 Scrubber Additives .........................................................................................................5-13 Technology Description .............................................................................................5-13 Impact on CCPs.........................................................................................................5-13 Mercury Co-Removal in Wet ESPs .....................................................................................5-14 References ..........................................................................................................................5-14 6 ANCILLARY SOLID WASTES ASSOCIATED WITH THE GENERATION AND MANAGEMENT OF COAL COMBUSTION PRODUCTS .........................................................6-1 Fabric Filters .........................................................................................................................6-1 Spent Selective Catalytic Reduction (SCR) Catalysts...........................................................6-1 Mercury Oxidation Catalysts .................................................................................................6-2

ix

Urea Wastes .........................................................................................................................6-2 FGD Wastewater Treatment Wastes ....................................................................................6-3 Wastewater Treatment Solids ..........................................................................................6-3 Spent Carbon Filtration Beds ...........................................................................................6-4 Demolition Materials..............................................................................................................6-4 References ............................................................................................................................6-5

x

LIST OF FIGURES

Figure 2-1 Typical Coal-Fired Power Plant Layout ...................................................................2-1 Figure 2-2 Typical Dewatering System for Low Oxidation FGD System...................................2-9 Figure 2-3 Typical Dewatering System for Forced Oxidation Wet FGD..................................2-10 Figure 3-1 Illustration of Typical Locations for NOX Control Technologies................................3-2 Figure 4-1 Typical Locations for Installation of SO3 Control Technologies ...............................4-2 Figure 4-2 Example Process Flow Diagram for Furnace Injection of Magnesium Hydroxide Slurry.................................................................................................................4-4 Figure 4-3 Illustration of a CleanStack Installation....................................................................4-7 Figure 4-4 Simplified Process Flow Diagram for the SBS Injection Process ............................4-8 Figure 4-5 Simplified Process Flow Diagram for Ammonia Injection (typical).........................4-11 Figure 4-6 Simplified Process Flow Diagram for Hydrated Lime Powder Injection.................4-13 Figure 5-1 Illustration of where Mercury Control Technologies are ..........................................5-2 Figure 5-2 Photograph of an ACI System, Showing the Activated Carbon Storage Silo (Center of Photo)................................................................................................................5-4 Figure 5-3 Illustration of the Toxecon Process Concept ...........................................................5-6 Figure 5-4 Illustration of the Toxecon II Process Concept ........................................................5-7 Figure 5-5 Example Mercury Oxidation Catalyst Block...........................................................5-11 Figure 5-6 Example Mercury Oxidation Pilot-Scale Catalyst Module (0.5 MW)......................5-12

xi

LIST OF TABLES

Table 2-1 Sorbents and Products of Wet and Dry FGD Systems .............................................2-7

xiii

1 INTRODUCTION

High-volume coal combustion products (CCPs)—fly ash, bottom ash, boiler slag, and flue gas desulfurization (FGD) solids—have been generated at power plants for decades and are well characterized. Utility CCP managers understand the characteristics of these high-volume CCPs and how they impact selection of the most appropriate management methods. However, new air emissions control technologies at power plants will change the characteristics of some existing CCPs. These changes may affect the selection of appropriate management methods for highvolume CCPs with respect to both disposal and use. In addition, appropriate management methods have to be determined for any new, lower-volume CCPs generated as a result of these air emissions controls. The objective of this report is to summarize possible modifications to power plant environmental control systems, including new technologies; the possible changes in characteristics of CCPs resulting from those modifications; and potential impacts on CCP management. This report is organized into six sections, including this introduction. Section 2 describes the four high-volume CCPs—fly ash, bottom ash, boiler slag, and FGD solids. A general description is provided for each, including where in the boiler or flue gas path it is produced, physical and chemical characteristics, and typical management practices. Sections 3 through 5 describe air emissions control technologies that may impact one or more of the four high-volume CCPs, or that may produce new, lower-volume CCPs. Each technology is described in terms of where in the boiler or flue gas path it is applied, its principle of operation, what impacts it may have on any of the four high-volume CCPs, and whether or not it will produce a new CCP stream. Where possible, the potential changes in high-volume CCP quantity or composition are briefly described. Section 3 describes nitrogen oxides (NOX) control, Section 4 describes sulfur trioxide (SO3)/sulfuric acid (H2SO4)control, and Section 5 describes mercury (Hg) control. Section 6 describes ancillary solid wastes arising from the generation and management of coal combustion products.

1-1

2 HIGH-VOLUME COAL COMBUSTION PRODUCTS

According to the American Coal Ash Association (ACAA), high-volume CCP production for 2006 totaled over 124 million tons, with approximately 54 million tons being used in a variety of construction and manufacturing applications (ACAA, 2007). High-volume CCPs are broken down into four main types: fly ash, bottom ash, boiler slag, and FGD solids. Each of these categories is defined by where the product is removed from the system, as well as its physical properties, chemical composition, and uses. Each is briefly described below. A typical power plant layout and the locations of solids collection are shown in Figure 2-1.

Economizer Furnace Economizer Ash Collection

FGD Absorber

Air Heater Particulate Collector Bottom Ash/Slag Collection

Fly Ash Collection

FGD Gypsum Collection

Stack

Figure 2-1 Typical Coal-Fired Power Plant Layout

Other solids are currently generated at some plants as a result of specific air pollution control technologies installed and operating at those plants. These include spent catalysts, spent FGD wastewater treatment media, and other maintenance and demolition materials associated with CCP collection and management. These wastes are addressed in Section 6 of this report.

2-1

High-Volume Coal Combustion Products

Fly Ash Fly ash is part of the non-combustible portion of coal that, due to its small size, is entrained with the flue gas after combustion. It is carried along with the flue gas until it drops out or is removed by a particulate control device. A small percentage of fly ash drops out in the economizer section of the boiler, which is the last convection section in the boiler before entering the air heater. Particulate control devices to remove the remaining fly ash from the flue gas include cold- or hot-side ESPs, fabric filters, and wet particulate scrubbers. These particulate control devices are capable of removing over 99% of fly ash in a flue gas stream. Cold-side ESPs are the most common particulate control device in existing plants and are located downstream of the air heater. They typically operate at temperatures of 275°F to 325°F. Hot-side ESPs are located upstream of the air heater and are typically operated at temperatures around 700°F. ESPs remove fly ash using an electrostatic charge between discharge electrodes and metal collection plates. The electrostatic charge causes the migration of fly ash to the plates. A rapping system removes the fly ash from the plates into a hopper system below for collection of the fly ash. Fabric filters are always located downstream of the air heater because of limitations on bag fabric operating temperatures. Fabric filters collect fly ash from the flue gas on the bag surface as the gas stream passes through the bags. The bags are periodically cleaned with a reverse flue gas flow, shaking of the bags, or with a pulse of compressed air, and the collected fly ash falls into a hopper below each bag compartment. Wet particulate scrubbers are also located downstream of the air heater. They rely on high-energy collisions between fly ash particles and scrubber slurry droplets to capture the fly ash in the droplets. The fly ash is removed by blowing down a portion of the scrubber recirculating slurry to a dewatering system. Fly ash is a fine powdery material, comprised mostly of silica, alumina, iron, calcium, other alkaline earth metals, sulfate, and a number of trace metals (EPRI, 1987). It is tan to gray in color, and made up of silt- to clay-sized spherical particles, making it similar in feel to talcum powder. Because of these properties fly ash is a relatively valuable commodity. Fly ashes contain a high proportion of silica and alumina, and exhibit cementitious properties in the presence of water and calcium hydroxide (pozzolanic properties), which make them useful as a substitute for portland cement in concrete products. In fact, some properties of concrete made with fly ash can exceed those of concrete made with only portland cement. Some fly ash is also used directly as raw feed in the production of portland cement. Because of its composition and size distribution, fly ash is also useful as filler in asphalt, road bases, and soil and waste stabilization. It also has wide application as a structural fill to support land development, and as a mine fill to reclaim surface mines, reduce subsidence in deep mines, and help control acid mine drainage. In plants that have wet FGD systems that produce a calcium sulfite sludge, fly ash is sometimes mixed with the FGD solid, lime, and other pozzolanic material to form a stabilized material for landfill, or for use as structural fills or mine fills.

2-2


Fly ash generally must be collected and handled dry to be used in high-value applications such as cement replacement. Fly ash that is collected dry and then transported by a wet sluice system, and fly ash that is removed in a wet particulate scrubber rather than an ESP or baghouse, is transported as a wet slurry that is managed in an ash pond, with more limited reuse possibilities. Fly ash is the highest-volume CCP, with more than 72 million tons produced in 2006 (ACAA, 2007). Nearly 45 percent of the fly ash was utilized, with concrete products and cement representing the most important application, at over 19 million tons. A small portion of the fly ash consists of hollow spheres, called cenospheres, that float in ash ponds and are easily harvested and used in a variety of high-value applications. Approximately 11 million pounds of cenospheres were harvested and sold in 2006 (ACAA, 2007). Flue Gas Conditioning for Fly Ash Capture Flue gas conditioning (FGC) is typically implemented to modify fly ash properties so as to improve collection by cold-side ESPs. Two different types of conditioning are typically employed. One is to lower fly ash resistivity (a measure of its ability to accept an electrical charge) and the other is to make the ash more cohesive, such as to avoid re-entrainment losses of low-resistivity material. Each type of conditioning is briefly discussed below. Fly Ash Resistivity Conditioning Fly ash resistivity conditioning is often required when units designed for high-sulfur coal are switched to lower sulfur, compliance coals. Fly ash itself is composed of minerals containing silica, alumina, iron, and calcium, among other species. These minerals typically have an inherently high resistivity, which can limit ESP performance. However, in high-sulfur coals, the ash resistivity is modified by the presence of SO3 and moisture in the flue gas. These two vaporphase species combine to form vapor-phase sulfuric acid, which adsorbs to the fly ash particles. The adsorption of sulfuric acid results in a lower resistivity layer at the particle surface, which improves ESP performance. With lower-sulfur coals, and particularly with alkaline ashes, this effect is greatly diminished, and ESP performance may be poor because of high ash resistivity. One solution is to inject SO3 into flue gases that have inherently low SO3 concentrations. Several different approaches are employed to produce the SO3, such as burning elemental sulfur, oxidizing pure SO2 at high temperature across a catalyst, or passing a small amount of hot flue gas across a catalyst to convert its SO2 content to SO3. Although most resistivity conditioning systems involve injection of SO3 into the flue gas, in some circumstances the ash resistivity is modified just by evaporating water into the flue gas between the air heater and ESP. This lowers the flue gas temperature and increases its relatively humidity. In these cases adsorption of humidity and/or dilute sulfuric acid from the flue gas is adequate to lower the ash resistivity.

2-3


Resistivity conditioning systems typically have only minor impacts on the fly ash, increasing the fly ash sulfate and/or moisture content by small percentages. An exception is for plants that use (or will use) activated carbon to remove mercury in the flue gas by injecting the carbon into the flue gas upstream of the ESP. In this case, SO3 injected for resistivity conditioning can adsorb on the carbon and make the carbon less effective at adsorbing mercury, such that greater carbon injection rates are required. Thus, SO3 injection can indirectly cause an increase in the ash carbon content for plants that also inject activated carbon for mercury control. This effect is discussed further in the section on mercury control technology. Fly Ash Cohesivity Conditioning In some cases fly ash has a resistivity that is low enough for the particles to accept a charge and migrate to the collecting plates, but too low to result in strong “clamping forces” to hold the particles to the plates. In such cases ESP performance is limited by particle re-entrainment off of the plates and back into the flue gas stream. Re-entrainment losses can occur during plate rapping, or can occur just because of the viscous forces of flue gas flowing past the particles on the plates. One way to reduce re-entrainment losses is to make the fly ash more cohesive, so that collected particles tend to stick to one another rather than be held in place only by electrostatic forces. A common way to achieve this is to inject ammonia into the flue gas in approximately a 1:1 molar ratio compared to the flue gas SO3 content. The SO3 content can be due to SO3 injection, or due to the native flue gas SO3 concentration. This type of conditioning is commonly called “dual conditioning” because both SO3 and ammonia are involved. By adding ammonia in a 1:1 molar ratio, the reaction product is ammonium bisulfate (NH4HSO4), also known as ABS, which has a melting point of 296oF and thus tends to be sticky at cold-side ESP collection temperatures. If the ammonia injection rate is not carefully controlled and the ammonia-to-SO3 molar ratio approaches 2:1, the reaction product is ammonium sulfate ((NH4)2SO4), which is stable up to 455oF and does not increase fly ash cohesivity. Dual conditioning can affect fly ash disposal or reuse, depending on the amount of ammonia added to the fly ash. Ammonia concentrations up to about 250 ppm have been observed in fly ash due to FGC (EPRI, 2001). If the fly ash is wetted (e.g., wet sluicing) the ammonium salts will dissolve. Under alkaline conditions some ammonium ion converts to ammonia, which can then be released into the atmosphere. Under more acidic conditions, the ammonium ion remains in the water phase. Ammonia injection for FGC should not have any effects on downstream byproducts (e.g., FGD gypsum) unless the ammonia injection rate exceeds a 2:1 molar ratio compared to SO3 in the flue gas. The effects of ammonia injection on fly ash and FGD solids are discussed in more detail in Sections 3 and 4 under ammonia-based systems for NOX control and ammonia injection for SO3 control, respectively. Other proprietary additives have been developed that are not based on ammonia injection to improve ash cohesivity. The impacts of these additives on ash disposal or reuse are not well documented in the public domain. 2-4


Bottom Ash Bottom ash is the non-combustible part of coal that, due to its larger size, does not get entrained with the flue gas after combustion. Instead, bottom ash collects on the furnace walls or in ash hoppers at the bottom of the furnace. Like fly ash, bottom ash is comprised mostly of silica, alumina, iron, calcium, other alkaline earth metals, sulfate, and trace metals. Bottom ash is produced from boilers that operate with their furnace temperatures generally below the ash fusion temperature. The ash fusion temperature is the point where the ash mineral matter begins to melt. The ash fusion temperature is different for every fuel and depends on the concentrations of typical coal ash components; however, the range of melting points for typical coal ash components is between 1610°F and 5070°F (Singer, 1991). If the furnace operates at temperatures below the ash fusion temperature, the bottom ash remains as a solid-phase material and is relatively porous and free flowing. Bottom ash is a product of “dry bottom” boilers, which infers that the bottom ash comes out as a dry (not molten) free-flowing solid material. Bottom ash has an angular and porous surface, with grain sizes ranging from fine sand to fine gravel. It can be used as an aggregate in road construction material. Because of its relatively wide size distribution, it generally does not need to be blended with fine aggregates to meet gradation requirements. But due to its high surface porosity, it is less durable than conventional aggregate material and is mostly used in base and shoulder rather than wearing surface mixtures. It also finds considerable use in concrete and structural fill applications, soil stabilization, snow and ice control on roads, and blasting grit/roofing granules. Total production of bottom ash in 2006 was nearly 19 million tons (ACAA, 2007). About 45% of this total was used, with structural fills, road base, concrete products, and cement production accounting for more than half of the applications. Mill rejects are a special class of CCP that are often managed with bottom ash at power plants. Mills are used for coal size reduction prior to injection into the furnace. Mill rejects are hard rock fragments, minerals, etc., that cannot be pulverized to the desired size, and are therefore rejected from the mills. The term “pyrites” is often used interchangeably with mill rejects, although not all mill rejects contain pyrite. Characterization data and guidance for the management of mill rejects was published in a report jointly sponsored by EPRI and the Utility Solid Waste Activities Group (USWAG) (EPRI, 1999).

Boiler Slag Boiler slag is, like bottom ash, part of the non-combustible part of coal that due to its larger size does not get entrained with the flue gas after combustion. In contrast to bottom ash, boiler slag is produced from boilers that operate their furnace above the ash fusion temperature. If the furnace operates at temperatures well above the ash fusion temperature, the portion of the ash not entrained in the flue gas comes out of the bottom of the boiler as a molten mass.

2-5


Boiler slag is a product of primarily older, higher-temperature coal furnaces. Higher furnace temperatures have fallen out of favor because these high temperatures also favor the formation of NOX in the flue gas. Boilers that produce slag are often of the cyclone firing configuration, and are often called slag tap or wet bottom boilers. In cyclone boilers, the coal feed to the cyclone burners or “barrels” is relatively coarse. This results in less of the ash getting entrained with the flue gas and in production of boiler slag as the majority ash product. In contrast, pulverized coal boilers (most of which are dry bottom) typically produce more fly ash than bottom ash. The molten slag is periodically drawn from the bottom of the furnace and is quenched with water, which causes it to fracture and form pellets. These pellets are hard and angular, with a glassy appearance similar to obsidian. Boiler slag, once quenched, is a durable material that is resistant to surface wear, which makes it desirable for asphalt and roofing applications. It also permanently maintains its black color, another desirable quality for asphalt. Boiler slag is marketed under the name “Black Beauty®.”1 The supply of boiler slag is diminishing, though, as the older boilers that produce this slag are being phased out of service. The total quantity of boiler slag produced in 2006 was about 2 million tons (ACAA, 2007). Most (83 percent) of this material was used, primarily for blasting grit or roofing granules.

FGD Solids FGD solids are produced by the removal of SO2 from the flue gas. Because there are many different types of FGD processes available, different types of FGD solids are produced, varying from wet sludges to dry powders. In this section, the types of solid FGD materials produced are described. Since SO2 control will increase substantially over the next ten to twenty years, a relatively detailed discussion of FGD systems and their solid products is provided. FGD systems can generally be categorized as “wet” or “dry.” Wet FGD systems are often called scrubbers, and contact the flue gas with an aqueous slurry or solution. The solid produced is typically in the form of slurry that must be dewatered. As might be expected, dry FGD systems produce a dry product that does not require dewatering. The primary dry FGD technologies include the spray drying process, which typically uses slaked lime (Ca(OH)2) as the reagent, and true dry injection processes, which generally use powdered sodium-based alkalis as a reagent. There are several variations on spray drying technology, such as circulating fluid bed absorbers, but these produce solids that are very similar to those from spray drying. Wet FGD Systems Most wet scrubbers use slaked lime or limestone (CaCO3) as the alkaline reagent. Some limebased FGD systems use lime with a high magnesia (MgO) content; the high magnesia content 1

Registered trademark of Reed Minerals

2-6


leads to increased liquid-phase alkalinity in the scrubber, and can enhance SO2 removal. Magnesia-enhanced limes are called Mg-lime in this report. The FGD solid is either calcium sulfite hemihydrate (CaSO3•½H2O), gypsum (CaSO4•2H2O), or a mixture of the two. Table 2-1 shows the primary sorbents and products for wet and dry FGD systems. Table 2-1 Sorbents and Products of Wet and Dry FGD Systems Process

Wet

Dry (Spray Dryer)

Oxidation Mode

Inhibited Oxidation

Natural Oxidation

Forced Oxidation

Natural Oxidation

Primary Sorbents

Lime, Mg-Lime, Limestone

Lime, Mg-Lime, Limestone

Limestone, Lime (some low-sulfurcoal systems)

Lime, Alk. Fly Ash, Na-alkali

Calcium Sulfite Hemihydrate

Calcium Sulfite Hemihydrate + FGD Gypsum

Primary Products

CaSO3•½H2O

CaSO3•½H2O + CaSO4•2H2O

FGD Gypsum CaSO4•2H2O

Calcium Sulfite Hemihydrate + Gypsum + Fly Ash + unspent sorbent CaSO3•½H2O + CaSO4•2H2O + Fly Ash

The design and operation of the wet FGD system determine which type of solid is formed. Calcium sulfite hemihydrate would be the only solid product formed in the absence of oxygen in the flue gas. However, since all flue gases have some oxygen content, a portion of the sulfite is oxidized to the sulfate form. If the oxidation is less than 15%, all of the sulfate gets incorporated into the calcium sulfite crystal lattice. So even though the solids might contain up to 15% sulfate, they look and behave as 100% calcium sulfite hemihydrate. As the oxidation exceeds 15%, gypsum begins to form as a separate solid phase. FGD systems tend to be most reliable if they operate such that they produce all calcium sulfite hemihydrate (including up to 15% sulfate) or all gypsum. Thus, it is most desirable to control the sulfite oxidation at 15% or less, or near 100%. Some FGD systems naturally operate in these regimes, but most require that the oxidation percentage be controlled, either by inhibiting or forcing oxidation. Each oxidation percentage control approach is discussed below. Some wet FGD systems use performance additives to enhance the amount of liquid-phase alkalinity available in the scrubbing slurry. Typical additives include adipic acid, dibasic acid (DBA, a by-product that is a mixture of adipic, glutaric, and succinic acids), formic acid, sodium formate, and magnesium. Most of these additives can be used in either inhibited or forced oxidation FGD systems, but magnesium (e.g., Mg-lime) is generally used only in inhibited or low natural oxidation systems, as magnesium sulfite is the alkaline species formed in the FGD 2-7


liquor at high magnesium concentrations. In an “in situ” forced oxidation system (forced oxidation within the FGD reaction tank) this sulfite would all become oxidized to the sulfate form and there would be no benefit from the magnesium. These performance additives can impact the solid by-products from wet FGD system, either directly through their presence (e.g., organic material content of the dewatered by-product) or potentially by modifying the solid byproduct crystal habit and/or dewatering properties (Radian, 1995). Inhibited or Low Natural Oxidation FGD Systems Inhibited oxidation systems generally add sodium thiosulfate or elemental sulfur (which reacts with sulfite to form thiosulfate) to the FGD system. The thiosulfate ion acts as a sulfite oxidation inhibitor. Some high-sulfur FGD systems naturally operate at low sulfite oxidation percentages. The ratio of sulfur dioxide to oxygen in the flue gas is important in controlling oxidation percentages, so high-sulfur systems tend to operate at lower natural oxidation percentages than low-sulfur systems. Low oxidation systems produce calcium sulfite hemihydrate, mixed with a small amount of excess lime or limestone and some inert material (fly ash and inert minerals from the dissolved lime or limestone) as the absorber slurry. The blow down slurry from the absorber vessels is usually dewatered in thickeners followed by a centrifuge or vacuum filter, although some systems blow down directly from the absorbers to a pond. In plants that have dewatering equipment, the water separated from the solids is usually returned to the FGD absorbers. Even though many inhibited (or low natural oxidation) systems have two stages of dewatering, the by-product still contains 30 to 40% moisture. The resulting wet sludge is thixotropic and difficult to handle, so it is often mixed with fly ash, lime, and sometimes other additives in a pug mill to produce a relatively dry, stabilized material. This stabilized mixture, sometimes referred to as fixated scrubber sludge, is often landfilled, but can also be used as structural fill or mine fill. A typical low oxidation dewatering system is shown in Figure 2-2. There are relatively few established uses for calcium sulfite hemihydrate. In 2006, 16 million tons of calcium sulfite hemihydrate were produced, and only 0.9 million tons (6 percent) were used (ACAA, 2007). It is unlikely that many new inhibited or low natural oxidation systems will be installed, while it is likely that some of the existing units will be converted to forced oxidation.

2-8


Figure 2-2 Typical Dewatering System for Low Oxidation FGD System

Forced Oxidation or High Natural Oxidation Systems Forced oxidation systems pump air into the absorber reaction tank (often called the mix tank or recycle tank) through distribution devices, typically called spargers. This increases the effective oxygen-to-SO2 ratio, and is controlled to achieve essentially 100% sulfite oxidation. Some very low-sulfur coal FGD systems also achieve 100% or nearly 100% oxidation without forced oxidation systems. The product resulting from forced oxidation systems is a relatively pure gypsum, referred to as FGD gypsum. FGD gypsum settles and dewaters much more effectively than calcium sulfite hemihydrate or calcium sulfite hemihydrate/gypsum mixtures (FGD systems with 15% to 90% sulfite oxidation). While inhibited oxidation or low natural oxidation FGD systems tend to have thickeners for primary dewatering, forced oxidation systems generally have hydrocyclones for primary dewatering and centrifuges, vacuum drums, or vacuum belt filters for secondary dewatering. A typical forced oxidation wet FGD dewatering system is shown in Figure 2-3.

2-9


Figure 2-3 Typical Dewatering System for Forced Oxidation Wet FGD

FGD gypsum can be dewatered to produce a moist (7 to 12% moisture) but free-flowing solid that is easily handled. FGD gypsum is a high-value product that is most often used as feedstock for wallboard production, or less commonly as a soil amendment. In 2006, over 12 million tons of FGD gypsum were produced, and 79% of that total was used, primarily as feedstock for wallboard manufacture (ACAA, 2007); FGD gypsum represents approximately 30% of the U.S. wallboard plant feedstock. Since limestone forced oxidation systems will be the dominant SO2 control technology over the next two decades, the production of FGD gypsum is expected to increase substantially. FGD gypsum used for wallboard production typically has relatively strict product specifications imposed by the wallboard manufacturers. Specifications include free moisture content, gypsum purity, particle size distribution, and maximum combustibles, chloride, total soluble salt, and/or ammonia content. Washing is often required during the secondary dewatering step to meet chloride and ammonia specifications. Some power plants are not located near wallboard plants and cannot cost effectively ship their FGD gypsum to a wallboard plant, or produce FGD gypsum that may not meet wallboard specifications. Such gypsum may be landfilled after dewatering, or it may be “stacked.” FGD systems that use a gypsum stack typically do not have primary or secondary dewatering, but send their absorber blow down to a landfill area where the solids settle to a relatively low free moisture content and the water is reclaimed to return to the FGD system, or is treated and 2-10


discharged. This area is managed so that with time, the settled gypsum level raises above grade, producing a stack of settled gypsum solids. Most wet scrubbers are preceded by an ESP or baghouse for primary particulate control. However, some older wet scrubbers, built in the 1970s and 1980s, are used for primary particulate control instead of an ESP or baghouse. These scrubbers typically operate at high pressure drop (15 to 25 in. H2O) to provide the energy required for efficient particle collection. These scrubbers may or may not have lime or limestone added to result in high SO2 removal efficiency as well. Those that do add an alkaline reagent to produce a mixture of fly ash and calcium sulfite hemihydrate and/or gypsum. The absorber slurry is typically either blown down to a pond, or dewatered and the dewatered solids are sent to a landfill. Dry FGD Systems Dry FGD technologies include spray drying, which typically uses slaked lime as the reagent, and true dry injection systems, which most often use powdered sodium-based alkalis as a reagent. These FGD systems produce a dry product that does not require dewatering. In 2006, an estimated 3.6 million tons of solids were produced by dry scrubbers (EPRI, 2007). Only about 9 percent were used, with mining applications accounting for most of the utilization (ACAA, 2007). Spray Drying Systems Spray dryer-based FGD is the most common “dry” FGD technology used in the U.S. However, it is somewhat of a misnomer to call this technology “dry” since a slaked-lime-based aqueous slurry is actually fed to the absorber(s). The slurry is finely atomized in the absorber, and heat from the flue gas evaporates all of the moisture to produce a dry (typically less than 1% moisture) solid product. In most spray dryer FGD systems, the spray dryer vessel is located upstream of the primary particulate control device (fabric filters are most commonly used due to their SO2 removal benefits compared to an ESP). Thus, the spray-dried by-product is mixed with the normal fly ash produced by the unit. It is possible to put a small ESP upstream of the spray dryer vessels to collect most of the fly ash, retaining it as a high-value product, but this configuration is rarely used in the U.S. (EPRI, 2007). The spray drying process solid is a mixture of calcium sulfite hemihydrate, gypsum, excess slaked lime, and calcium carbonate, and it is most commonly captured along with fly ash in either an ESP or fabric filter. The solids collected at the bottom of the spray dryer contain approximately 70% fly ash, 13% calcium sulfite, 7% gypsum, 5% unreacted lime, and 5% water. The solids collected in the particulate control device contain approximately 72% fly ash, 15% calcium sulfite, 9% gypsum, 1% unreacted lime, and 3% water (EPRI, 2007). For some lowsulfur-coal applications, the amount of fly ash in the solid can exceed the amount of calciumbased salts. It generally produces a high pH when wetted due to the excess slaked lime and the alkaline nature of the fly ash (when the process is applied to western coal). The dry product is 2-11


mostly landfilled, but can see limited use in waste stabilization, road base, or solidification applications. Most spray dryer FGD systems have what is called a “two point” solids discharge, as some solids drop out at the bottom of the spray dryer vessel while the majority are entrained with the flue gas leaving the vessel and collected in the fabric filter or ESP. The solid material that drops out at the bottom of the spray dryer tends to be wetter and coarser than the solids collected in the fabric filter or ESP. The two streams are generally combined when sent to disposal or reuse. Some spray dryer FGD systems employ some percentage of recycle of the spray-dried solids with the fresh hydrated lime slurry feed to the atomizers. This recycle improves overall lime utilization and improves slurry drying characteristics. In systems that employ recycle, it is generally the fabric filter or ESP catch solids that are recycled, not the spray dryer dropout material, which is more difficult to handle. Dry Injection Systems Although not widely used, there are some plants that employ truly dry FGD processes, employing dry injection of finely ground sodium alkali, such as trona (sodium sesquicarbonate), into the flue gas between the air heater and ESP or fabric filter. Dry injection processes are sometimes mass transfer limited for SO2 removal. Thus, an excess of the sodium-based alkali is often injected to increase the particle surface area available for absorbing and reacting with flue gas SO2. The dry injection by-product is typically a mixture of sodium sulfate and excess sodium alkali, usually in the form of sodium carbonate. This material is collected along with fly ash in either an ESP or a baghouse. Because of the highly water-soluble sodium salts present, dry injection solids are mostly landfilled and not reused. Because an excess of water-soluble, alkaline reagent is typically injected, ponding of the solid by-product is typically not practical. Leachates from landfilling of this material can be very alkaline and have high dissolved salt content. Furnace Sorbent Injection (FSI) Systems Another dry FGD technology that was developed in the 1970s and 1980s involves injection of limestone or hydrated lime into the upper furnace; this process is often called furnace sorbent injection. The injected sorbent is collected along with the fly ash in the existing ESP or fabric filter. The technology is rarely applied in the U.S. today, but may see increased usage in the future on small boilers if the value of SO2 allowances remains high. The utilization of the injected calcium-based sorbents is typically low (10 to 25%), so a large excess of reagent is required. Thus, the technology typically adds a significant amount of calcium sulfate, calcium carbonate, lime (CaO), and/or hydrated lime (Ca(OH)2) to the fly ash stream, and results in a solid that produces a high pH when wetted.

2-12


References ACAA, 2007. 2006 Coal Combustion Product (CCP) Production and Use Survey. American Coal Ash Association. www.ACAA-USA.org, 2007. EPRI, 1987. Chemical Characterization of Fossil Fuel Combustion Wastes. EPRI, Palo Alto, CA: 1987. EA-5321. EPRI, 1999. Guidance for Comanagement of Mill Rejects at Coal-Fired Power Plants. EPRI, Palo Alto, CA: 1999. TR-108994. EPRI, 2001. Characterization of Ammonia Leaching from Coal Fly Ash. EPRI, Palo Alto, CA: 2001. 1005163. EPRI, 2007. A Review of Literature Related to the Use of Spray Dryer Absorber Material: Production, Characterization, Utilization Applications, Barriers, and Recommendations. EPRI, Palo Alto, CA: 2007. 1014914. Radian, 1995. The Effects of Flue Gas Desulfurization (FGD) Additives on Solid By-Products. EPRI, Palo Alto, CA: 1995. TR-102367. Singer, 1991. Combustion Fossil Power. Combustion Engineering, Inc., Connecticut.

2-13

3 NOX CONTROL TECHNOLOGY

NOX control technology consists primarily of four technology types: furnace combustion modifications, which for the purposes of this report include the use of low-NOX burners; rich reagent injection (RRI), which is implemented in the furnace; selective non-catalytic reduction (SNCR), which is implemented in the upper furnace; and selective catalytic reduction (SCR), which is implemented between the economizer and air heater. Figure 3-1 illustrates where in the flue gas path each of these technologies would typically be located. The technologies and their impacts on coal combustion products are discussed below.

Combustion Modifications/Low-NOX Burners Technology Description Combustion modifications and/or low-NOX burners control the formation of NOX during the fuel combustion process, generally by limiting the availability of oxygen as fuel nitrogen is released. They typically involve some form of staged combustion, wherein certain areas of the furnace operate fuel rich, then additional air is added to complete combustion. Overfire air (OFA), or separate overfire air (SOFA), is often added above the burner level to stage combustion conditions. A challenge that sometimes limits the effectiveness of combustion modifications and low-NOX burners is to lower NOX emissions while still ensuring complete fuel combustion. Increased flue gas carbon monoxide (CO) concentrations and/or increased loss on ignition (LOI) in the fly ash can result from low-NOX firing conditions. Impact on CCPs The primary impact of low-NOX burners to the CCPs is higher LOI levels in the fly ash. Higher LOI adversely affects the air-entraining properties of concrete made with the fly ash, which can in turn adversely affect fly ash sales for concrete if the LOI exceeds ASTM C-618 specification (6 percent LOI) or contractual limits due to increased need for air entraining. Higher carbon content may also slightly increase the concentration of some trace elements in the fly ash due to its greater adsorption capacity, but also may reduce their leachability. Because combustion modifications and low-NOX burners impact firing conditions in the furnace, it is also possible that bottom ash characteristics could be slightly altered.

3-1

NOX Control Technology

Economizer

SNCR

Combustion Modifications, RRI

SCR

Furnace

Air Heater

Particulate Collector

FGD Absorber

Stack

Figure 3-1 Illustration of Typical Locations for NOX Control Technologies

Finally, as the LOI content of the fly ash increases, it is possible that unburned carbon can penetrate a cold-side ESP, as the carbon has a low electrical resistivity and has a tendency to be re-entrained. Consequently, this technology could lead to an increase in the amount of carbon captured in the wet FGD system and increase the carbon content of a gypsum by-product.

Rich Reagent Injection (RRI) Technology Description RRI involves injecting urea or ammonia into fuel-rich portions of the furnace to reduce NOX formation. The technology was originally developed for NOX control for cyclone-fired coal boilers, but can be applied to other boiler types. It involves staged combustion and overfire air as the means of producing a fuel-rich zone in the furnace. The technology can be combined with low-NOX burners and/or with SNCR, which is discussed in the next subsection, to result in greater overall NOX reduction. Impact on CCPs RRI technology would primarily impact the fly ash, where it would have similar impacts as described above for combustion modification/low-NOX burners. Although the RRI technology also involves injecting urea or ammonia into the furnace, these are reportedly destroyed at high efficiency within the furnace so ammonia slip is essentially zero. Thus, little or no contamination of the fly ash with ammonia is expected if RRI alone is implemented on a unit (RJM, 2007). 3-2


Selective Non-Catalytic Reduction (SNCR) Technology Description SNCR is used to chemically reduce NOX formed in the furnace, by oxidation/reduction reactions between ammonia and NOX within a certain temperature range. The name for this technology comes from the fact that the ammonia reactions are selective for NOX in the target temperature range, but no catalyst is used. SNCR is typically implemented in the upper furnace. Urea is often used as the ammonia source, as urea decomposes to form ammonia at elevated temperatures. Urea is often preferred over pure ammonia because of health and safety issues associated with handling ammonia, and the fact that the trajectory and penetration of urea injected as an aqueous solution into an upper furnace can be better controlled than when injecting gaseous ammonia. Chemical equations showing the conversion of urea to ammonia and the reduction of NOX by ammonia are shown below (Bhattachara at al., 2003).

NH2CONH + (1+x)H2O → NH4COONH2 + (x)H2O → 2 NH3 + CO2 + (x)H2O 4 NO + 4 NH3 + O2 → 4 N2 + 6 H2O 2 NO2 + 4 NH3 + O2 → 3 N2 + 6 H2O

Because SNCR relies solely on gas-phase reactions (no catalyst is used), an excess of urea/ammonia is typically required to get higher NOX reduction levels (e.g., 40 to 60% NOX reduction levels). While most of the excess ammonia is destroyed at upper furnace temperatures, ammonia “slip” (excess ammonia in the flue gas leaving the economizer of the boiler) is common with SNCR installations. It is generally desirable to control ammonia slip to 2 ppm or less in the flue gas entering an air heater, to avoid excessive formation of ABS within the air heater and resulting air heater pressure drop increases. However, many SNCR applications exceed this level of ammonia slip. Impact on CCPs The biggest issue for ammonia injection is its effect on fly ash disposal/reuse; it will not affect bottom ash or boiler slag. Ammonia can be collected as ABS with the fly ash in the downstream particulate control device, or even penetrate the particulate control device and impact the FGD system. At 2 ppm of ammonia slip, the ABS formed would amount to about 300 ppm as ammonia (NH3) if it all reports to the fly ash. Since many SNCRs exceed 2 ppm slip, higher ammonia concentrations on fly ash can be expected. Ammonia concentrations as high as 900 ppm have previously been reported (EPRI, 2001). At this ammonia concentration there may be issues with ammonia in fly ash in some disposal settings and reuse applications. In general, fly ashes from subbituminous coals (e.g., Powder River Basin) adsorb less ammonia than fly ashes 3-3


from bituminous coals (EPRI, 1999). Ammonia that is not adsorbed on the fly ash in a system that has no FGD system will exit the unit out the stack. Ammonia can be released in vapor form if the fly ash is contacted with water, especially when the ash pH is naturally alkaline (pH > 7.0). Ammonia odors will result if alkaline ash is sluiced to an open pond or if dry ash is wetted in uncovered portions of a landfill. Ammonia vapor will also be released if the fly ash is used as an admixture for cement or concrete, or if the ash is mixed with FGD by-products and lime for stabilization, due to high pH conditions in both of these applications. Also, if caustic is added to a pond or landfill leachate to raise its pH to within the range allowed by the plant NPDES permit, a substantially higher ammonia vapor pressure could result in the vicinity of the caustic addition. The amount of ammonia released and the odor level in any of these scenarios depend on the amount of ammonia adsorbed on the ash and the pH. Several methods have been advanced for the beneficiation of ammoniated fly ash for eventual use. These include chemical treatment using the Ammonia Slip Mitigation (ASM™) technology (EPRI, 2005a) and removal using humid air (EPRI, 2002). Ammonia salts are highly soluble and are readily released to the water phase as ammonium ion. As much as 70 percent of the ammonia adsorbed to the fly ash is released almost immediately on contact with water in batch leaching tests (EPRI, 2001). Ash sluiced to ponds will have relatively low ammonia concentrations (<10 ppm) due to the large volume of sluice water, and the ponds are effective at removing ammonia through biologic uptake and volatilization (EPRI, 2003). In landfills, the much lower volumes of contact water from rainfall can result in much higher ammonium concentrations in solution, again depending on the initial adsorbed ammonia and the pH (EPRI, 2001). The presence of the ammonia can also impact the leaching of trace constituents from the fly ash to some degree, by altering the solution pH or by complexing with some of the metals. Depending on the direction of the pH change, leaching of some trace constituents may increase, while others may decrease. Ammonia complexation is only likely to significantly affect metals concentrations at alkaline pH and very high ammonia levels (>1000 mg/L), much higher than would occur in a ponded environment (EPRI, 2005b; Wang et al., 2007; EPRI, 2008). These high concentrations potentially may be attained in localized areas of a landfill for short periods for fly ash exposed to very high ammonia slip levels, but are not expected to be typical. Most of the ammonia slip is expected to be retained in the fly ash, however some may also be captured in the FGD solids. This situation is more likely when ammonia is used for SO3 mitigation. The impact of ammonia on FGD solids will be discussed in more detail in the next section under SO3 control.

3-4


Selective Catalytic Reduction (SCR) Technology Description Like SNCR, SCR chemically reduces NOX formed in the furnace by oxidation/reduction reactions with ammonia, but with SCR this occurs in the presence of a catalyst. The catalyst provides active surface area on which these reactions can take place, so nearly equimolar quantities of ammonia are added compared to the amount of NOX in the flue gas. The catalysts typically are made of a ceramic material that includes titanium oxide as a carrier and a vanadium oxide as the active species. Some catalysts also use an oxide of tungsten as part of the active species. Catalysts are most commonly installed in a honeycomb or plate configuration (DOE, 1997). The catalytic reactions proceed most rapidly at temperatures in the range of 600oF to 750oF, so SCR reactors are typically installed between the economizer and air heater (Figure 3-1). Ammonia slip remains an issue with SCR, but most well-controlled SCR systems limit ammonia slip to less than 2 ppm in the air heater inlet flue gas. Ammonia slip tends to increase with the age of the catalyst. Impact on CCPs The impacts of ammonia injection on CCPs will be similar to those previously described for SNCR systems. However, since SCR systems generally are well controlled and have much lower slip than SNCR systems, the fly ash contains much less ammonia. Accordingly, the adverse impacts on fly ash are less. As previously noted, at 2 ppm of ammonia slip, the ABS formed would amount to about 300 ppm as ammonia (NH3) if it all ends up in the fly ash. Most SCRs operate at 2 ppm or less slip. Concentrations less than 100 ppm ammonia in fly ash have been previously reported for operating SCR facilities (EPRI, 2001). Even at this ammonia concentration there may be issues with ammonia in fly ash in some reuse applications. Speciation of mercury has been shown to be affected by the SCR catalyst (Lee et al., 2004), and other redox sensitive constituents (As, Cr, Se) may also be affected. Speciation changes may affect their capture on CCPs. For example, elemental mercury can be oxidized to ionic mercury, increasing its capture in wet FGD systems. The combination of SCR and wet FGD is likely to play a significant role in mercury capture. This is discussed in more detail in Section 5 under mercury co-removal with FGD solids. Changes in the capture and leaching characteristics of other redox sensitive elements are unlikely to significantly change the overall character of the CCPs from a disposal perspective, but may be important in the treatment of ash pond discharge waters due to differences in treatability of the different species. Recent leaching research suggests that the release of chromium from fly ash may be increased by SCR operation (EPA, 2008).

3-5


References Bhattacharya et al., 2003. “Urea to Ammonia (U2A™) systems: Operation and Process Chemistry. Combined Power Plant Air Pollutant Control symposium. Washington D.C, 2003. DOE, 1997. www.fossil.energy.gov/programs/powersystems/publications/Clean_Coal_Topical_Reports/topic al9.pdf, July 1997.

EPA, 2008. “Characterization of Coal Combustion Residues from Electric Utilities Using Wet Scrubbers for Multi-Pollutant Control.” F.Sanchez, D. Kosson, R. Keeney, R. Delapp, L. Turner, and P. Kariher. EPA/600/R-08/077. 2008. EPRI, 1999. Investigation of Ammonia Adsorption on Fly Ash and Potential Impacts of Ammoniated Ash. EPRI, Palo Alto, CA: 1999. TR-113777. EPRI, 2001. Characterization of Ammonia Leaching from Coal Fly Ash. EPRI, Palo Alto, CA: 2001. 1005163. EPRI, 2002. Novel Ash Beneficiation Processes for Managing Unburned Carbon and Ammonia. EPRI, Palo Alto, CA: 2002. 1004395. EPRI, 2003. Fate of Ammonia in Ash Ponds; Site F Pilot Field Studies. EPRI, Palo Alto, CA: 2003. 1005410. EPRI, 2005a. Ammonia Removal from Fly Ash: Process Review; Headwaters Ammonia Slip Mitigation (ASM™) Technology. EPRI, Palo Alto, CA: 2005. 1010381. EPRI, 2005b. Effects of Ammonia on Trace Element Leaching from Coal Fly Ash. EPRI, Palo Alto, CA: 2005. 1010063. EPRI, 2008. The Effect of Ammonia on Mercury Partitioning in Fly Ash. EPRI, Palo Alto, CA: 2008. 1014014. Lee et al., 2004. Lee, CW, Srivastava, RK, Ghorishi, SB, Hastings TW, and Stevens, FM. 2004. “Investigation of Selective Catalytic Reduction Impact on Mercury Speciation under Simulated NOx Emission Control Conditions.” J. Air & Waste Manage. Assoc. 54:1560–1566. RJM, 2007. www.RJM.com/pdf/richreagent.pdf Wang, 2007. Wang, J, Wang, T, Mallhi, H, Liu, Y, Ban, H, Ladwig, K. 2007. “The Role of Ammonia on Mercury Leaching from Coal Fly Ash.” Chemosphere, in press.

3-6

4 SO3 CONTROL TECHNOLOGY

Coal combustion results in the liberation of sulfur primarily as sulfur dioxide, SO2. Sulfur trioxide (SO3) and/or sulfuric acid (H2SO4) is present in the flue gas from coal firing because of the further oxidation of a small percentage of the coal sulfur (typically 0.5 to 1.5%) to SO3 rather than SO2. As the flue gas cools across the air heater, the SO3 combines with flue gas moisture to form gas-phase sulfuric acid. At the inlet to a wet FGD absorber, a sub-micron diameter sulfuric acid mist is formed when the flue gas is humidified and cooled, which causes the gas-phase sulfuric acid to rapidly condense. This sub-micron acid mist behaves like small particles rather than as a gas in the wet absorber, and is not removed at high efficiency. The mist that exits the scrubber very effectively scatters light, which leads to visible plume opacity (sometimes referred to as blue plume). Other adverse effects can result from having SO3/sulfuric acid in the flue gas, such as ESP and duct corrosion, air heater plugging, or the need to increase air heater outlet gas temperatures to avoid such problems. Plume opacity issues due to SO3 mist have increased recently due to the installation of SCRs. Over the past decade, a number of coal-fired generating units have retrofitted SCR systems in their boiler flue gas path for NOX control. SCR catalysts are oxidation catalysts (the injected ammonia is oxidized, which correspondingly chemically reduces NOX species), and can oxidize other flue gas species such as SO2 and mercury. Typically, 0.5 to 1.5% of the flue gas SO2 is oxidized to SO3 across an SCR catalyst. The retrofitting of SCR systems can therefore significantly increase flue gas SO3 concentrations at the air heater inlet due to further oxidation of flue gas SO2 across the SCR catalysts. Wet FGD systems generally condense all of the SO3/sulfuric acid in a flue gas, making it more visible. After SCR and wet FGD systems have been retrofitted on units currently firing compliance coals, it may be cost effective to switch from low-sulfur to higher-sulfur coal. With the resulting increase in flue gas SO2 concentrations, along with SCR catalysts to oxidize a portion of the SO2 to SO3 and wet scrubbers to condense sulfuric acid mist, many units may experience a stack gas opacity increase, and sulfuric acid controls may be required. A number of units have already experienced these issues and have installed some form of SO3 control. This section describes currently available technologies and the effects of SO3 controls on coal combustion products. As described above, SO3 is formed in the furnace and in downstream SCR systems, and the most tangible manifestation of sulfuric acid in the flue gas is stack plume visibility. Consequently, SO3 controls can be implemented anywhere in the flue gas path from the furnace to just upstream of the stack. Figure 4-1 shows a number of SO3 control technologies, and illustrates where in the flue gas path they are likely to be implemented. Each of these technologies is described in the following subsections. The technologies are discussed below in approximately the same order as they are shown in the flue gas path illustration in Figure 4-1. 4-1

SO3 Control Technology

- SBS Injection?

Furnace Mg Injection Economizer

- Mg Injection

- SBS Injection - Hydrated Lime Injection - Condensing Air Heater

Furnace

Air Heater

- Fuel Additives

Wet ESP

Particulate Collector

- Ammonia Injection

- Hydrated Lime Injection - SBS Injection - Trona Injection

FGD Absorber

Stack

Figure 4-1 Typical Locations for Installation of SO3 Control Technologies

Fuel Additives Technology Description Fuel additives have long been used as an approach for SO3 mitigation for oil-fired units. The additives typically include some form of magnesium oxide or magnesium hydroxide. In oil-fired units, the magnesium-based additives primarily prevent SO3 formation by tying up vanadium in the ash in the form of unreactive magnesium vanadates. Fuel additives have typically been less effective on coal-fired units, where the primary control mechanism appears to be removal of SO3 after it is formed, through gas-solid reactions. These additives are typically added to the coal from a storage silo as the coal is conveyed to the bunkers. Since the additives are injected into the furnace with the fuel they can become “dead burned” or sintered by the high temperatures to which they are exposed, and thus less reactive with the SO3 formed. The injected sorbent is collected in the ESP or baghouse along with the fly ash. A portion of the additive may end up in the bottom ash stream. One proprietary fuel additive for coal-based applications is OmniClearTM. Based on information provided in company literature, the additive appears to be ground dolomite (MgCO3•CaCO3), sized to 75% minus 200 mesh, that is partially calcined then treated with a coating that helps prevent dead burning (OmniClear, 2003). The company literature reports an additive rate for a 3% sulfur coal application of approximately 0.5 to 1.0 wt% of the coal feed rate.

4-2


Impact on CCPs For a coal with 3% sulfur and 8% ash, and assuming a dry-bottom boiler with 80% of the ash leaving as fly ash, the additive rate above would correspond with an 8% to 17% increase in the amount of fly ash if it all reports to the fly ash rather than bottom ash. The added material would be a mixture of calcium and magnesium oxides, carbonates, and sulfates. Because a considerable excess of the alkaline additive is generally required, if the fly ash is disposed of in a pond or landfill, the pH of ash pond water or landfill leachates may be increased if this technology is implemented. The higher pH would also affect the leachability of trace constituents, possibly increasing leachability of some (e.g., oxyanions) and decreasing others (e.g., trace metals). The increased alkalis could also affect the use of the ash in concrete. No effects on downstream solids, such as FGD gypsum, are expected. Injection of the additives at high rates can adversely affect cold-side ESP performance, due to two effects. One is that the magnesium- and calcium-based materials tend to have a high electrical resistivity, and the other is that the removal of SO3 from the flue gas stream tends to increase fly ash resistivity. These effects could lead to fly ash and spent sorbent carryover into the FGD system. However, it is most likely that particulate emission and opacity limits would require that injection rates be lowered before such carryover from the ESP got to a level that could impact FGD solids.

Furnace Injection of Magnesium-Based Sorbents Technology Description As mentioned above, magnesium hydroxide or magnesium oxide has been injected into the furnaces of oil-fired power plants and, to a lesser extent, coal-fired power plants to achieve moderate levels of SO3 control for many years. More recent testing has focused on the injection of magnesium hydroxide slurries into the upper regions of coal-fired furnaces to achieve 90+% removal of furnace-formed SO3. The technology has been installed for SO3 control on at least two large coal-fired units with SCR. In this technology, magnesium hydroxide is typically fed as a slurry into the upper furnace, where temperatures are lower, to prevent dead burning of the sorbent (EPRI, 2002). When magnesium hydroxide slurry is injected into the furnace, the water evaporates and leaves a solid MgO particle. The MgO solid reacts with SO3 in the gas phase to form magnesium sulfate (MgSO4). Magnesium sulfate is water soluble and therefore unlikely to form hard deposits in the boiler. The magnesium sulfate solids are removed along with fly ash in the downstream particulate control device. Figure 4-2 is an example process flow diagram for how magnesium hydroxide injection in the furnace would typically be implemented.

4-3


Filter

Blower

LIC

Storage Tank

Day Tank

LIC

Atomizing Air PI

Truck or Barge Unloading

M Moyno Pump

PI

FI

VFD FIC

Unit Load

Injection Lance 1 of 12 per unit

Figure 4-2 Example Process Flow Diagram for Furnace Injection of Magnesium Hydroxide Slurry

Adding MgO in the boiler may modify coal ash fusion temperatures and slag properties, making slag deposits more friable (Benson, 2003). One vendor reports that they have optimized furnace injection of magnesium sorbents by the use of computational fluid dynamics (CFD) modeling to target injection towards key spots in the upper furnace. They call this technology targeted infurnace injection (TIFI™) 2 . They claim that they can reduce boiler slagging and fouling as well as effectively control stack gas SO3 concentrations (Fuel Tech, 2007). Impact on CCPs Magnesium hydroxide slurry is available as a commercial product, or it can be produced on site ® at plants equipped with a Thiosorbic lime FGD system. Required sorbent injection rates will depend on the coal sulfur, percent SO3 removal required, and whether or not the injection locations have been optimized. For a 3% sulfur coal, the amount of fly ash collected could increase by 2 to 8%. The increased mass in the fly ash would consist of a mixture of magnesium oxide, sulfate, and carbonate if commercial magnesium hydroxide is used. If by-product magnesium hydroxide is used, the amount in the fly ash would be increased by about one-third. The increased mass would be due to gypsum impurities in the magnesium hydroxide. Because a considerable excess of the alkaline additive is generally required, the pH of ash pond water or landfill leachates would likely increase if this technology is implemented on a unit that disposes of its ash. The higher pH would also affect the leachability of trace constituents, possibly increasing leachability of some (e.g., oxyanions) and decreasing others (e.g., trace

2

TIFI is a trademark of Fuel Tech.

4-4


metals). The increased alkalis could also affect the use of the ash in concrete. Little, if any, of the material injected into the upper furnace should end up in the bottom ash or boiler slag. No effects on downstream solids, such as FGD gypsum, are expected. Injection of magnesium hydroxide slurries at very high rates can adversely affect cold-side ESP performance, as described above, which could lead to fly ash and spent sorbent carryover into the FGD system. However, it is more likely that sorbent injection rates would be controlled at lower rates to avoid exceeding particulate emissions and opacity limits, and thus would not impact FGD solids.

Magnesium Oxide Injection Upstream of the Air Heater Technology Description Magnesium oxide or magnesium hydroxide slurry can be injected into the furnace as described above or post-combustion; this discussion is focused on injection downstream of the economizer and/or SCR and upstream of the air heater. Three versions of magnesium alkali injection upstream of the air heater for SO3 control are under development or are being applied commercially. Each is described below. One version uses a highly reactive magnesium oxide (MgO) powder that is typically injected pneumatically into the flue gas upstream of the air heater. Injection rates are generally low, ranging from about 0.5:1 to 1.5:1 molar ratio of magnesium oxide to SO3 (Martin Marietta, 2003). The MgO reacts with SO3 to form magnesium sulfate. Another technology, called targeted duct injection (TDI™) 3 , injects magnesium hydroxide slurry into the flue gas downstream of the SCR but upstream of the air heater. The technology vendor reports that through a combination of TIFI (discussed earlier) and TDI if needed, treatment rates to produce a clear stack range from about 2 to 6 lb of sorbent injected per ton of coal fired in the unit treated (Schulz, 2006). A DOE co-sponsored pilot-scale evaluation used by-product magnesium hydroxide slurry injection upstream of an air heater, combined with low air heater exit temperatures, to achieve as high as 80% mercury capture. The by-product magnesium hydroxide slurry was previously described in the furnace injection discussion. As part of this testing, it was determined that by o operating with an air heater outlet temperature as low as 220 F along with injecting the byproduct magnesium hydroxide, nearly complete capture of the flue gas SO3 was achieved as well (Winschel, 2005). Impact on CCPs Implementation of any of these three technologies would increase the fly ash mass rate by about 2 to 5%. The added material would be a mixture of magnesium sulfate, carbonate and hydroxide. 3

TDI is a trademark of Fuel Tech.

4-5


The bottom ash should be unaffected by this technology, or any of the post-combustion technologies discussed later in this subsection. Because an excess of the alkaline additive is generally required, if the fly ash is disposed of in a pond or landfill, the pH of ash pond water or landfill leachates may be increased if this technology is implemented. The higher pH would also affect the leachability of trace constituents, possibly increasing leachability of some and decreasing others. The increased alkalis could also adversely affect the use of the ash in concrete. As with the previous magnesium-injection technologies discussed earlier, no effects on downstream by-products such as FGD gypsum are expected. Injection of magnesium oxide or magnesium hydroxide slurries at high rates can adversely affect cold-side ESP performance, which could lead to fly ash and spent sorbent carryover into the FGD system. However, it is likely that sorbent injection rates would be controlled at lower rates to avoid exceeding particulate emissions and opacity limits and would not impact FGD solids. Also, in the case of the DOE-cosponsored technology, the lower air heater outlet temperature would tend to offset adverse effects of the sorbent injection due to the decreased gas volume and increased gas relative humidity.

CleanStackTM Condensing Air Heater 4 Technology Description In the air heater, the primary mechanisms for the reduction of gas-phase SO3 are the adsorption and condensation of SO3/H2SO4. Condensation occurs when the temperature is lower than the dew point of sulfuric acid, which could occur in the cold end of a regenerative air heater. The sulfuric acid condenses both on ash particles in the flue gas and on cold air heater surfaces. Modeling by the University of North Dakota Energy and Environmental Research Center (EERC) has shown that sulfuric acid condensation occurs primarily on sub-micron-diameter ash particles (Hamel, 2003). CleanStack™ is a process that will enable the air heater to operate with a lower exit temperature, achieving high sulfuric acid capture and improving the plant heat rate. Injection of micronized limestone provides additional surface area for acid condensation, and helps neutralize the condensed acid. Coarsely ground limestone is micronized to a 2- to 3-micron median diameter with a jet pulverizer as it is injected (Bowes, 2006). Figure 4-3 illustrates a typical micronized limestone grinding and injection system. A full-scale implementation of the technology has reportedly been conducted on one of two air heaters on a 350-MW unit (Bowes, 2006). Micronized limestone was injected at a rate of 500 lb/hr upstream of the air heater. Assuming the unit was operating near full load and that one-half of the flue gas was treated, this injection rate corresponds with approximately a 6% increase in the fly ash mass rate. The added material would be a mixture of excess limestone (CaCO3) and calcium sulfate (CaSO4), and any impurities in the limestone.

4

CleanStack is a trademark of Marsulex Inc.

4-6


Impact on CCPs Because an excess of the alkaline additive is generally required, the pH of ash pond water or landfill leachates may be increased if this technology is implemented. However, because the excess reagent will be calcium carbonate rather than stronger alkalis like hydroxides or oxides, the pH should not rise much above 8. This could affect the leachability of some metals, but to a lesser degree than technologies using the stronger alkalis.

Figure 4-3 Illustration of a CleanStack Installation

Although the calcium-based sorbent would be expected to have a high resistivity, and removal of SO3 from the flue gas also increases ash resistivity, the lowered air heater exit temperature should tend to offset adverse effects on cold-side ESP performance. Therefore, this technology may not have an adverse effect on ESP performance and consequently on the FGD solids.

4-7


Sodium Based Sorbent (SBS) Injection Technology Description The SBS Injection process injects a clear, aqueous solution of sodium bisulfite [SBS] (NaHSO3), sodium sulfite (Na2SO3), and/or sodium carbonate (Na2CO3) into the flue gas. The injection location can be either upstream of the air heater (typically between an SCR and the air heater) or downstream of the air heater (between the air heater and cold-side ESP). Upstream injection is preferred by the process developers (Moser, 2007). Development work is also underway to determine if the SBS Injection process could be implemented upstream of an SCR. Figure 4-1 shows the typical SBS Injection process locations in the flue gas path, including possible injection upstream of an SCR. The reagent is injected as a solution through an array of air-atomized nozzles, and is fed at relatively low molar feed ratios, typically not much more than the theoretical amount required to react with all of the SO3 in the flue gas. The reaction products are salts containing sodium and sulfate, and are collected along with fly ash in the cold-side ESP, baghouse or other particulate control device. Figure 4-4 is a typical process flow diagram for the process.

INJECTION AIR: PLANT AIR OR COMPRESSOR

SBS DILUTION WATER

FILTER (SOFTENED)

SOLIDS WETTING PUMP

DILUTION WATER TANK

DUPLEX STRAINER

DILUTION WATER PUMP

DE FE

HOPPER TRUCK

REAGENT SOLUTION STORAGE TANK

FLUE GAS FE

INJECTION FEED PUMP

Figure 4-4 Simplified Process Flow Diagram for the SBS Injection Process

4-8

INJECTION LANCES


Initially it was thought that sodium sulfate (Na2SO4) was the primary reaction product. However, it is apparent that other sodium species are formed with sodium to SO3 molar ratios closer to 1:1 (e.g., sodium bisulfate [NaHSO4], which is sticky like ABS). In installations where injection occurs upstream of the air heater, the molar ratio of sodium in the reagent to SO3 in the flue gas is controlled near 2:1. This typically ensures high SO3 removal efficiency and facilitates the formation of sodium sulfate as the primary by-product. The SBS reagent can consist of commercially available sodium carbonate, sodium bisulfite, or sodium sulfite, or a by-product that is typically a mixture of sodium bisulfite and sulfite. The reagent can be received as a clear solution (typically 25 to 30 wt% dissolved solids) or as a dry powder. The dry powder must be mixed with softened plant water to obtain the desired reagent concentration. One additional benefit is possible when using the SBS process to remove SO3 at high efficiency upstream of the air heater, on plants equipped with SCR. It may be possible to operate an SCR at higher ammonia slip without ABS formation in the air heater if the SO3 concentration is controlled to very low levels (e.g., 2 ppmv or less). Once the ammonia concentration is higher than the SO3 concentration in the flue gas entering the air heater, ammonium sulfate rather than ABS forms. Higher ammonia slip can produce benefits such as allowing for longer catalyst life and/or higher NOX reduction percentages. However, operating in this mode would likely increase the ammonia concentration found in the fly ash. Another benefit the vendor claims is that by removing SO3 upstream of the air heater, unburned carbon in the fly ash can more effectively adsorb mercury. Adsorption of mercury by unburned carbon in the ash appears to occur as the flue gas is cooled in the air heater. SO3 is known to compete with mercury for adsorption sites on unburned carbon in fly ash. Thus, the mercury content of the fly ash may be increased if the SBS process is installed upstream of the air heater. Impact on CCPs The SBS injection process adds about 2 to 4 wt% to the fly ash mass collected in the particulate control device, as sodium sulfate. This corresponds with about 1 to 2 wt% added to the fly ash when reported as Na2O. The SBS process has been installed on 9600 MW of electric generating capacity (Moser, 2007), with 2680 MW coming on line in 2008 (Gray, 2008). Some of these installations sell their fly ash but most do not. It will be plant specific as to what, if any, impact the addition of approximately 1 to 2% sodium as Na2O in the fly ash has on the reuse of the fly ash from plants that use the SBS process. There is a supplementary optional chemical requirement in ASTM C-618 for use of Type C and Type F fly ash in concrete that limits available alkalis, as equivalent Na2O, to 1.5 wt%. It is possible that SBS injection could cause this specification to be exceeded. Because the SBS process typically is operated with high utilization of the injected alkali reagent, there should not be much excess alkali added to the fly ash. Thus, SBS injection should not increase ash pond water or landfill leachate pH to the extent of other sorbent injection technologies that operate with a large excess of reagent. The sodium in the reagent will be very 4-9


water soluble, so using the SBS process would likely increase the sodium and sulfate content of ash pond discharge water or landfill leachate. The SBS injection process uses sodium-based sorbents, and sodium can be used to condition fly ash resistivity in cold-side ESP applications, so no adverse effects on ESP performance are expected. Therefore, no impacts of the SBS process are expected downstream of the particulate control device, such as on wallboard-grade gypsum products.

Ammonia Injection at the Air Heater Outlet Technology Description Ammonia injection is commonly used along with native or injected SO3 in dual-flue gas conditioning systems for improved cold-side ESP performance, due to the cohesive properties of the ammonium salts that form. This process was described in Section 2. It has also been tested and/or used commercially for the express purpose of sulfuric acid control at a number of plants. Ammonia is injected between the air heater and ESP at NH3:SO3 molar ratios of 1:1 to 2:1 to achieve 90+% removal of sulfuric acid. Plants that have SCR systems already have a source of ammonia on site. The biggest issues in employing ammonia injection are its effect on fly ash disposal/reuse and the safe handling/storage of the reagent. The latter is less of an issue for plants where an existing urea-to-ammonia system for the SCR would be used for ammonia supply. Figure 4-5 is a simplified process flow diagram for an ammonia injection system. Ammonia/air mixtures are normally piped to an injection manifold where they are introduced into the flue gas duct via injection probes located just downstream of the air heater. The injection location is between the air heater and the particulate removal device.

4-10


Air Filter

Blower Unit Load Injection Air Heater

NH3 from U2A System

FIC

1 per unit

Air Heater

1 of 2 per unit

ESP

Figure 4-5 Simplified Process Flow Diagram for Ammonia Injection (typical)

Ammonia reacts very quickly with sulfuric acid vapor at flue gas temperatures less than 400°F. The reaction products are either ABS, ammonium sulfate, or a mixture of the two products. Because ABS has a relatively low melting point (296°F), it can be present as either a solid or a liquid when collected by the particulate control device. The low melting point makes the product stickier compared to ammonium sulfate, which has a much higher melting point (455°F). The relative formation of the two products depends on the ratio of injected ammonia to sulfuric acid vapor. The minimum molar ratio of NH3:SO3 required to remove almost all of the SO3 as ABS is 1:1. Higher ratios will produce more ammonium sulfate and less ABS, but will not substantially improve SO3 removal. In practice, the molar ratio of injected ammonia to SO3 is usually maintained between 1.0 and 2.0, depending on the particular site. Some plants operate at low ratios and produce mainly ABS in order to increase the cohesivity of their fly ash and minimize ammonia consumption costs. Other plants operate at higher ratios to avoid ash handling problems caused by the stickier ABS. Therefore, the optimum ratio is site specific. The ammonia injection rate should not exceed two times the molar SO3 rate in the flue gas because of the potential to form ammonium bisulfite through reaction with flue gas SO2. This can form as a sub-micron particle that, like condensed sulfuric acid mist, can penetrate FGD absorbers and produce a visible plume. Alternately, excess ammonia can be absorbed by the FGD system, leading to elevated ammonia concentrations in the FGD liquor and possible ammonia contamination of gypsum by-product. This effect is discussed further below. The published literature shows that the ESP collection efficiency generally improves or stays the same when ammonia is injected into the flue gas. The impacts of ammonia injection on the ESP include an increase in space charge effects, but a reduction in fly ash re-entrainment upon rapping. The space charge increases because the reaction of ammonia with SO3 produces a large 4-11


number of small solid particles that enter the ESP along with the existing fly ash. Fly ash reentrainment decreases because of the formation of “sticky” ABS solids. Impact on CCPs The impact of ammonia on fly ash was previously discussed in the section on SCRs and SNCRs for NOX control. However, while ammonia slip from SCR systems can result in small quantities of ammonia in the fly ash, using ammonia for sulfuric acid control can increase the amount in fly ash more than tenfold compared to injection of ammonia for SCR operation only. Most of the injected ammonia is converted to ABS or ammonium sulfate, which is collected along with the fly ash in the particulate control device. Ammonia injection for SO3 control on a plant with an SCR and firing a 3% sulfur coal would increase the fly ash mass rate by about 2 to 3%, and would add 0.5 to 1% ammonia to the ash. These higher ammonia concentrations could significantly increase the potential impacts on ash discussed for SCRs and SNCRs in Section 3. Little data have been published about the influence of ammonia injection on FGD system operation. However, as described above there is some concern that ammonia slip past the ESP will be removed in a downstream FGD system and will build in concentration over time. This could cause an issue with the ammonia content in wallboard-grade gypsum. A simple mass balance calculation was used to predict these levels for a high-sulfur coal case. With 10 ppm of ammonia slip past the ESP, even assuming the gypsum is washed as it is dewatered (90% efficiency) and that the wash water is purged from the FGD system, ammonia levels could be as high as 40 ppm in gypsum produced from a 3% sulfur coal.

Dry Sorbent Injection at the Air Heater Outlet Dry sorbent injection upstream of the ESP is part of the SO3 mitigation measures installed at several plants. Both hydrated lime injection and trona injection are being employed. Dry sorbents remove SO3 by a gas-solid reaction. The sorbent is pneumatically injected downstream of the air heater (in most cases) and upstream of the existing particulate control device. The reaction products and unreacted sorbent are removed along with the fly ash by the particulate control device. This section describes both hydrated lime (calcium-based) and trona (sodium-based) sorbent injection. Hydrated Lime Powder Injection Technology Description In this process, dry hydrated lime powder (Ca(OH)2) is injected between the air heater and ESP or baghouse. Another possible approach is to inject a portion of the hydrated lime upstream of the air heater. Figure 4-6 is a simplified process flow diagram for hydrated lime injection between the air heater and an ESP.

4-12


Signal to Storage Silo Blower/Pump

LIC

Day Silo

Unit Load Storage Silo

LIC

Ratio Controller

Filter Blower Filter Weigh Belt Blower

M Driflow Solids Pump or Equivalent

Truck Unloading

M Driflow Solids Pump or Equivalent

ESP Air Heater

1 of 2 per unit

Figure 4-6 Simplified Process Flow Diagram for Hydrated Lime Powder Injection

EPRI conducted pilot-scale tests of hydrated lime injection upstream of an ESP for SO3 control at the Environmental Control Technology Center (ECTC) in the early 1990s (Peterson, 1993). In these data, a lime injection rate of 2 to 4 lb/hr/kacfm (depending on the source of lime) was required to achieve 80% SO3 removal. These injection rates correspond to Ca:SO3 molar ratios of approximately 8:1 to 17:1, many times the stoichiometric (theoretical) amount. During those pilot tests, dry injection of calcium-based materials caused the ESP performance to degrade due to increased resistivity. The lime material coated the collector plates and acted as an insulator, thereby reducing the current flow from the wires to the plates. Hydrated lime injection caused strong sparking and reduced the operating current in the ESP. Some companies have reported the use of a high-specific-surface-area hydrated lime for SO3 control upstream of an ESP, with greatly reduced hydrated lime injection rates compared to conventional hydrated lime (Nolan, 2006). However, these high-specific-surface-area hydrated limes have only been tested in short-term tests at full scale in the U.S. Once this sorbent is available, SO3 removal may be achievable at lower hydrated lime injection rates than were indicated by the EPRI ECTC data. Impact on CCPs Hydrated lime injection affects the fly ash by-product in two ways. First, sorbent injection increases the quantity of fly ash collected by the particulate control device. For example, at a sorbent injection rate of 1.5 lb/hr/kacfm, sufficient for about 60% to 70% SO3 removal, and for a typical bituminous coal (~8% ash in the coal), the amount of solids removed by the ESP would 4-13


increase by about 12% compared to normal operation. This additional 12% of material would be a mixture of calcium sulfate, calcium hydroxide, calcium carbonate, and calcium sulfite. For some fly ash reuse scenarios, the additional mass and the addition of hydrated lime (calcium hydroxide) to the fly ash might be desirable. However, the increased alkalis could also adversely affect the use of the ash in concrete. The second effect is for plants that pond or landfill their ash. For a plant that ponds its fly ash, the presence of a large amount of excess lime in the ash may cause a significant increase in the sluice and pond water pH, and could require the addition of a large amount of acid to the pond to lower the discharge water pH into the range required by the plant’s NPDES permit. The higher pH sluice water would also affect the leachability of trace constituents, possibly increasing leachability of some and decreasing others. For plants that landfill their ash, hydrated lime injection would likely significantly increase leachate pH values, with corresponding change in leachability of trace constituents. For either situation, because of the excess hydrated lime that would be present, uncontrolled pH values upwards of 12 are possible. For land disposal, the additional mass would increase disposal costs and/or shorten landfill life. No significant impacts on FGD solids are expected from hydrated lime injection. Any excess lime collected in a wet FGD absorber would tend to be utilized as an FGD reagent, offsetting some of the normal lime or limestone reagent feed to the absorber. Some testing has been conducted where the hydrated lime is actually injected between the ESP and wet scrubber, rather than upstream of the ESP. Trona Injection Technology Description Trona is a hydrated sodium carbonate-bicarbonate compound (Na2CO3•NaHCO3•2H2O). When injected into the flue gas at air heater exit temperatures, the trona calcines to form soda ash. The soda ash then reacts with SO3 to form sodium sulfate. The excess sodium carbonate and sodium sulfate are collected along with fly ash in the ESP or baghouse. The process flow diagram for trona injection looks similar to that shown above for hydrated lime injection, although additional measures may be required to ensure the trona stays very dry so it maintains good handling properties. As a co-benefit, trona injection can enhance ESP performance. While the benefits of sodium injection are well known for treating sodium depletion in hot-side ESPs, sodium injection can also be used as a resistivity conditioning agent in cold-side ESPs. (Peterson, 1993; White, 1974). The results of short-term trona injection tests conducted at one plant were presented at the Mega Symposium in September 2004 (Ritzenthaler, 2004). At an injection ratio of about 0.5 lb/hr trona per 1000 acfm of flue gas, the ESP outlet SO3 concentration was reduced by about 62%. Reductions in ESP outlet SO3 concentration were increased to about 76% when the injection rate was increased to about 1.5 lb/hr/1000 acfm. Enhancements such as trona grinding just prior to injection may be employed to lower sorbent injection rate requirements. 4-14


Impact on CCPs Similar to hydrated lime, trona injection increases the quantity of ash collected, and can alter the pH of the ash pond and landfill leachate. The injection rates described above correspond with approximately a 4% to 12% increase in fly ash mass, respectively, for a typical bituminous coal. The extra material in the fly ash would be a mixture of sodium sulfate and sodium carbonate, and would add approximately 1 to 4 wt% sodium to the fly ash. This additional material could have adverse effects on fly ash disposal and re-use options. The addition of approximately 1 to 4% sodium as Na2O in the fly ash may negatively affect the performance of the ash in concrete and concrete product applications. There is a supplementary optional chemical guideline for available alkalis of 1.5 wt% (as equivalent Na2O) in ASTM C-618 for Type C and Type F fly ash used in concrete containing reactive aggregates. It is possible that trona injection could cause this specification to be exceeded. Because the trona injection process typically requires a significant excess of the injected alkali reagent to achieve moderate to high SO3 removal levels, there could be a significant excess of alkali added to the fly ash. Thus, like hydrated lime, trona injection could significantly increase the pH of ash pond water or landfill leachate in fly ash disposal scenarios. The higher-pH sluice water would also affect the leachability of trace constituents, possibly increasing leachability of some and decreasing others. The sodium salts are very water soluble, which will increase the sodium and sulfate content of ash pond discharge water or landfill leachate. No impacts on FGD solids would be expected, as excess trona and the sodium sulfate formed should be collected by an ESP at high efficiency, due to corresponding improvements in ESP performance during trona injection.

Wet ESP Technology Description Electrostatic precipitation is a well-established method for collecting fine particulate matter in many different industrial applications; however, sulfuric acid control at coal-fired power plants represents a relatively new application. An electrostatic precipitator works by charging particles in an electric field and moving them to a surface of opposite charge where they are deposited. In conventional dry ESPs, the deposited particles are removed by mechanical rapping. In wet ESPs, the collecting electrode is irrigated with water to remove particulate matter. Wet ESPs are typically located downstream of a wet scrubber, just before the flue gas goes up the stack; however, there are some installations where the last field of a dry ESP has been converted to a wet ESP. Wet ESP technology solves several limitations on SO3 control encountered by conventional dry ESPs. The wet environment helps ensure that the sulfuric acid is condensed and collectable by electrostatic forces, lowers particle resistivity, and allows high power input levels to enhance the removal of sub-micron-diameter mist. Wet ESPs can remove other flue gas components that 4-15


contribute to plume opacity, including fine fly ash particles, unburned carbon or soot, and wet FGD scrubber carryover. Particle resistivity problems typically encountered with dry ESPs are eliminated with a wet ESP since the water-saturated condition renders all particles conductive. Wet ESPs have also been demonstrated to remove oxidized mercury from flue gas at relatively high efficiency, and to remove elemental mercury at lower percentages. There are several ways to deliver the irrigation water to a wet ESP. In the most common method, a recycled water stream is sprayed onto the plates. Often, the last field of the wet ESP is washed with fresh water to prevent contaminated water from entering the cleaned gas stream. The plates are washed either periodically or continuously. A blow down stream is produced from this recycled water that must be treated. Impact on CCPs Since wet ESPs are typically installed at or near the end of the flue gas path after the wet scrubber, they generally do not impact other CCPs, with the possible exception of FGD gypsum. A wet ESP generates an aqueous slurry waste that must either be treated separately or added to the FGD system. This slurry waste would include all of the material collected by the wet ESP and washed from the plates, including fine particulate matter that penetrates the upstream particulate collector and wet FGD absorber; sulfuric acid; scrubber carryover slurry; and mercury captured in the wet ESP. If the slurry is treated separately, a filter cake solid waste is generated from the wastewater treatment. Adding this slurry stream to the FGD system would consume a small amount of FGD reagent to neutralize the acid, and would add sulfates and trace metals that could adversely affect FGD chemistry in low-sulfite oxidation systems. The mercury content of the FGD solids and/or liquor would also be increased by adding this stream.

References Benson, 2003. Benson, L.B., K.J. Smith, and R.A. Roden. “Control of Sulfur Dioxide and Sulfur Trioxide Using By-Product of a Magnesium-Enhanced Lime FGD System,” paper presented at the Institute of Clean Air Companies (ICAC) Forum 2003: Multi-Pollutant Emission Controls & Strategies. Nashville, TN, October 12-15, 2003. Bowes, 2006. Bowes, Edward, et al. “Sulfur Trioxide Reduction Technology Demonstrated at Dominion Generation’s Chesterfield Power Station,” Paper #25 presented at the Power Plant Air Pollutant Control Mega Symposium, Baltimore, MD, August 30, 2006. EPRI, 2002. Sulfuric Acid Removal Process Evaluation: Long-term Results. EPRI, Palo Alto, CA, U.S. Department of Energy, National Energy Technology Laboratory, Pittsburgh, PA, TVA, Chattanooga, TN, American Electric Power, Columbus, OH, and FirstEnergy, Shippingport, PA: 2002.1004165. Fuel Tech, 2007. www.fueltechnv.com, 2007. Gray, 2008. Conversation with Sterling Gray of URS Corporation on 4/25/08. 4-16


Hamel, 2003. Hamel, B., R. Rhodes, S.A. Benson, D.P. McCollor. “The Role of Combustion Air Preheater in an Effective Multipollutant Control Strategy,” paper presented at the U.S. EPA/DOE/EPRI Combined Power Plant Air Pollutant Control Mega Symposium, Washington, D.C., 19-22 May 2003. Martin Marietta, 2003. “SO3 Reduction Using Highly-Reactive Magnesium Oxide,” case study available for download at www.magspecialties.com, 2003. Moser, 2007. Robert E. Moser, “SO3’s impacts on plant O&M: Part III,” POWER magazine, April 2007. Nolan, 2006. Paul Nolan, Chemical Lime Company, and Lew Benson, Carmeuse Lime; presentations made at the DOE-NETL SO3 Control Conference, Pittsburgh, PA, May 18, 2006. OmniClear 2003. “OmniClear, A new alternative for sulfur trioxide control.” Document provided by Omni Materials, Inc., Maysville, KY. Peterson 1993. Peterson, J.R., A.F. Jones, F.B. Meserole, and R.G. Rhudy. “SO3 Removal from Flue Gas by Sorbent Injection, EPRI HSTC Phase II Test Results,” paper presented at the SO2 Control Symposium, Boston, MA, 24-27 August 1993. Ritzenthaler, 2004. Ritzenthaler, Douglas P. and John Maziuk. “Successful Mitigation of SO3 Emissions while Simultaneously Enhancing ESP Operation at the General James M. Gavin Plant in Chesire, Ohio by Employing Dry Sorbent Injection of Trona Upstream of the ESP.” Paper presented at the US EPA/DOE/EPRI Combined Power plant Air Pollutant Control Symposium: the Mega Symposium, Washington, D.C., August 30 – September 2, 2004. Schulz, 2006. Personal communication between David White of URS Corporation and Kent Schulz of Fuel Tech, Inc. May 2006. White, 1974. White, Harry J. “Resistivity Problems in Electrostatic Precipitation,” Journal of the Air Pollution Control Association, Volume 24, No. 4, April 1974, pp. 333-334. Winschel, 2005. Winschel, R.A., M.L. Fenger, K.H. Payette, and L.A. Brickett. “Control of Mercury Emissions by Absorption on Fly Ash – Experimental Results of the Consol/Allegheny Pilot Plant Program,” presentation at the DOE/NETL Mercury Control Technology R&D Program Review Meeting, Pittsburgh, PA, July 14, 2005.

4-17

5 MERCURY REMOVAL

From the standpoint of impacts on CCPs, mercury removal can take place in four areas: •

Co-removal with fly ash, as a result of boiler combustion modifications, injection of carbon5 6 based sorbents (ACI and MerCure™ ), or injection of non-carbon sorbents (MinPlus™ , 7 Amended Silicates™ , BASF mineral-based sorbents);

•

Co-removal with wet or dry FGD by-products, due to “native” FGD removal or because of add-on mercury oxidation technologies (KNX 8 , mercury oxidation catalysts) and/or wet FGD mercury re-emission control additives;

•

Co-removal in a wet ESP, whose mercury-containing by-product stream will either be combined with the wet FGD by-products or treated as a separate blow down; and/or

•

Removal in a new, separate CCP stream due to the implementation of EPRI-licensed, mercury-specific technologies such as Toxecon ® or Toxecon II®9 .

Figure 5-1 is an illustration of where each of these technologies would typically be implemented in the flue gas path. Like SO3 control, which was discussed in the previous section, mercury control can be implemented at a wide range of locations in the flue gas path, from the furnace to just upstream of the stack. These mercury control technologies are discussed below, including their expected impacts on other coal combustion products. The technology discussions are divided into the four mercury removal areas listed above.

Mercury Co-Removal with Fly Ash Combustion Modifications Technology Description The combustion modification process is aimed at modifying the fly ash to make it capable of physically adsorbing mercury. Adsorption is defined as a process where a gas, vapor, or dissolved matter is assimilated by the surface of a solid or liquid. To change the fly ash 5

MerCure is a trademark of Alstom. MinPlus is a trademark of MinPlus. 7 Amended Silicates is a trademark of Amended Silicates, LLC. 8 KNX is a trademark of Alstom. 9 Toxecon and Toxecon II are registered trademarks of EPRI. 6

5-1

Mercury Removal

properties, the combustion properties of the boiler are modified by changing the way the coal is burned in the furnace. This may include changing excess air levels, modifying the amount of air and fuel added at different locations in the boiler, and/or varying burner tilt.

Economizer MinPlus Injection

Furnace MerCure Sorbent Injection

Combustion Modifications

Toxecon II Sorbent Injection

Air Heater

Wet ESP Coremoval

FGD Absorber Particulate Collector

KNX Injection

-Activated Carbon Injection -Amended Silicates Injection -BASF Mineral-based Sorbent Injection

-Mercury Oxidation Catalysts -Toxecon Installation

-Wet FGD Coremoval Stack -Scrubber Additives

Figure 5-1 Illustration of where Mercury Control Technologies are

Most fly ash is composed of non-combustible minerals that are found in the coal. These minerals have little capacity to adsorb mercury. Boiler combustion modifications may be used to add unburned carbon (loss on ignition) to the fly ash. In some flue gases, unburned carbon has the capacity to adsorb mercury. The unburned carbon, including adsorbed mercury, is subsequently collected along with fly ash in the particulate control device. It has previously been observed that, for eastern bituminous coal-fired plants, significant mercury control can be achieved with fly ash having elevated LOI levels (Lu, 2007). Recent test programs on plants that fire eastern bituminous and Powder River Basin coal have shown that it is possible to increase the LOI content of the fly ash and optimize other flue gas variables to enhance mercury adsorption and capture with the fly ash (Romero, 2005, 2006).

5-2


Impact on CCPs This technology will primarily impact the fly ash, by increasing the LOI content and the mercury content. As discussed previously for NOX control, higher LOI adversely affects the air-entraining properties of concrete made with the fly ash, which can in turn adversely affect fly ash sales for concrete if the LOI exceeds ASTM C-618 specification (6 percent LOI) or contractual limits due to increased need for air entraining additives. Higher carbon content may also slightly increase the concentration of some trace elements in the fly ash due to greater adsorption capacity. Because combustion modifications impact firing conditions in the furnace, it is also possible that bottom ash characteristics could be slightly altered. Finally, as the LOI content of the fly ash increases, it is possible that unburned carbon can penetrate a cold-side ESP, as the carbon has a low electrical resistivity and has a tendency to be re-entrained. Consequently, this technology could lead to an increase in the amount of carbon captured in the wet FGD system and increase the carbon content of a gypsum by-product. This effect is further discussed in the next subsection. Activated Carbon Injection (ACI) Technology Description ACI is implemented by pneumatically injecting a fine powder of activated carbon into the flue gas duct upstream of the particulate device. In most cases the carbon is injected between the air heater and primary particulate control device, although some processes involve carbon injection between the economizer and air heater. An ACI system typically consists of a storage silo, solids feeders, blowers, transport lines, and injection lances. Figure 5-2 is a photograph of a temporary ACI system installed for a month-long full-scale ACI test. Activated carbon is made from coal, lignite, or other carbon-based material that is processed with heat and steam to produce a highly porous solid powder that has great capacity for adsorption. As described above for LOI, mercury in the flue gas can adsorb onto the highly porous carbon particles, and be collected along with the carbon in the unit’s particulate control device. Several different types of activated carbon are available, including one being developed specifically for hot-side ESP applications. Some activated carbons are treated with bromine addition to improve their performance in flue gas from low-chlorine-content coals, and for other applications. The performance of ACI for mercury control can be very site specific. Factors such as coal type, flue gas residence time in ductwork, particulate control device type and size, flue gas temperature, and concentrations of other species in the flue gas such as hydrochloric acid (HCl) and SO3 affect the performance of ACI. The type of particulate control device installed on a unit can, in particular, greatly impact the effectiveness of ACI; ACI is generally more effective with a fabric filter or baghouse than with an ESP, and generally less effective with a wet particulate scrubber.

5-3

Mercury Removal

Figure 5-2 Photograph of an ACI System, Showing the Activated Carbon Storage Silo (Center of Photo)

Impact on CCPs ACI injection rates are commonly expressed in units of pounds per million actual cubic feet of flue gas (lb/MMacf) treated. Typical rates range from less than 1 lb/MMacf to upwards of 10 lb/MMacf. For a typical bituminous coal with about 8% ash, and assuming 75-80% of the coal ash becomes fly ash, these rates correspond with approximately 0.5% to 5% of the fly ash mass being added as carbon. Even at the lowest carbon injection rate, fly ash uses for cement replacement/concrete admixture may be adversely affected, due to the effects of the activated carbon on air entrainment additives used in concrete. Air entrainment is required to give the concrete adequate freeze/thaw resilience, and additives are added to concrete to enhance air entrainment. The activated carbon has an affinity for the additive, so the presence of activated carbon in fly ash can greatly increase the amount of additive required for the concrete to get an acceptable “Foam Index Test” score. In some cases, it may not be possible to use enough of the additive to result in formation of a stable foam. Even if enough additive can be used to get an acceptable Foam Index Test score, the variability introduced in how much additive must be used as the activated carbon dosage in the fly ash varies is undesirable to concrete producers.

5-4


Some activated carbon vendors are developing proprietary technology to “passivate” sorbents so they do not significantly impact air entrainment additive requirements (Landreth et al, 2007). If these passivated sorbents are successfully developed and commercialized, this could remove a significant disadvantage of ACI for plants that sell their ash for concrete applications. In addition, methods are being developed to beneficiate fly ash containing activated carbon for use in concrete by ozonation (EPRI, 2005b) or re-combustion (Cooper, 2007). The use of fly ash in cement manufacture may be limited if ACI is used for mercury control. The mercury captured on the activated carbon can be re-released at the high temperatures found in typical cement kilns. For this reason, EPA is currently considering restricting the use of fly ash from units using ACI in the manufacture of cement (Barnett, 2007). Several recent studies have shown that the mercury is relatively stable at ambient temperatures, and not readily released from the fly ash/activated carbon (EPA, 2006; Gustin & Ladwig, 2004; EPRI, 2005a; UND EERC, 2007). Therefore the higher mercury concentrations should not significantly affect the ash from a disposal perspective. In addition to mercury, the activated carbon may also increase capture of other trace constituents such as arsenic and selenium (EPA, 2006). However, the leaching of these constituents was not found to be directly related to their total concentration in the fly ash. Chemically treated (e.g., brominated) carbon contains species such as bromine that could potentially leach from the fly ash in a landfill situation, or be thermally desorbed at elevated temperatures. Results of limited testing of ashes from brominated carbon injection trials by the synthetic groundwater leaching procedure (SGLP) suggest that bromine can be leached from the ash at high water-to-ash ratios (e.g., ponding) (Sjostrom, 2005a, 2005b). Laboratory testing has shown little potential for thermal desorption of bromine, though. Most of the carbon injected into a flue gas is expected to be captured with the fly ash in the ESP or baghouse. However, ACI could also adversely affect FGD gypsum quality, particularly if the unit has a relatively small ESP for particulate control. As mentioned above, carbon-based particles tend to have a low electrical resistivity, which reduces the electrostatic forces that hold particles to ESP collecting plates. This can lead to rapping and non-rapping re-entrainment losses that can allow carbon particles to penetrate the ESP and potentially be collected in the wet FGD absorber. In forced oxidation systems producing wallboard-grade gypsum, such particles can increase the combustible percentage in the gypsum product. Wallboard manufacturers typically have very tight limits for combustible content in their gypsum feedstock because much of their output is for “fire code” wallboard. One manufacturer reported that their limit on combustibles in gypsum is 0.2 wt%. For a plant that fires 3% sulfur coal and that injects 10 lb/MMacf of carbon, it would take a little over 10% of the carbon injected to penetrate the ESP and be collected in the FGD system to exceed this limit.

5-5

Mercury Removal

Toxecon® and Toxecon II® Technology Description EPRI has developed a patented technology that addresses several shortcomings of conventional ACI for plants currently equipped with an ESP for particulate control. In this process, called Toxecon®, a very compact fabric filter is installed in the flue gas path downstream of an existing ESP. Activated carbon is injected into the flue gas downstream of the ESP and upstream of this compact fabric filter. The Toxecon concept is illustrated in Figure 5-3. The fabric filter serves as a contacting device between the activated carbon and the flue gas, so less activated carbon may be needed and/or higher mercury removal percentages may be achieved than through injection upstream of the ESP. A similar technology, called Toxecon II, offers some of the benefits of the Toxecon process but at lower capital cost. The intent of Toxecon II is to avoid impacting the bulk of the fly ash from a unit, by injecting activated carbon into the downstream fields of a multi-field ESP (e.g., injecting between Fields 2 and 3 of a four-field ESP). The majority of the fly ash, collected in the upstream fields, would have no carbon content and would not have elevated mercury content.

Economizer

Activated Carbon Injection

Furnace Air Heater Particulate Collector

FGD Absorber

Existing ESP

Toxecon Baghouse

Figure 5-3 Illustration of the Toxecon Process Concept

5-6

Stack


The Toxecon II technology would require that the ash handling system be modified to segregate the ash from the fields downstream of injection. The Toxecon II process concept is illustrated in Figure 5-4. Impact on CCPs A significant benefit of the Toxecon process is that the fly ash captured by the ESP (about 99 percent) is not affected by carbon injection, and thus fly ash sales are not impacted. However, a new CCP stream is generated by the process, a small amount of material rich in carbon, mercury, and fine fly ash particles that have penetrated the ESP. For a 500-MW plant, this stream would probably amount to 100 to 200 lb/hr (about 800 tons per year). Mercury concentrations in this material may be in the tens of ppm or higher, and other volatile trace elements may also be present in higher concentrations than normally found in fly ash. Because of the relatively low activated carbon injection rate required upstream of a baghouse compared to an ESP, and because the baghouse should capture activated carbon at high efficiency, no significant impacts on FGD by-products would be expected.

Boiler

Sorbent Injection

Re-usable Fly Ash

Activated Carbon + Fly Ash

Figure 5-4 Illustration of the Toxecon II Process Concept

5-7

Mercury Removal

Since Toxecon II involves injection within the ESP, limited residence time is available and relatively high carbon injection rates may be required, particularly compared to the Toxecon approach. Thus, using the example of a 500-MW plant, the carbon-rich stream from a Toxecon II application could be on the order of 500 to 1000 lb/hr (about 4000 tons per year). The material will be similar to the material from Toxecon, but with a higher fly ash content and lower mercury concentration. Also, since Toxecon II involves using only part of the ESP to collect the injected activated carbon, the issue of carbon carryover to the wet FGD system and increased combustible levels in the FGD gypsum are potential concerns, as described for ACI. Non-Carbon Sorbent Injection Technology Description Because carbon-based mercury sorbents can adversely affect the ability to reuse fly ash for beneficial use, a number of mineral-based sorbents are under development that should avoid such impacts on fly ash sales. These sorbents are not yet in wide use and their exact formulations are considered proprietary information. One such sorbent, MinPlus, is apparently comprised of calcium-, alumina- and silica-based minerals, and is injected into the upper furnace at injection rates on the same order of magnitude as those for activated carbon (Biermann, 2006). Another non-carbon sorbent (Amended Silicates) is a chemically amended silicate substrate that is injected between the air heater and particulate control device as would be activated carbon, and at similar feed rates to activated carbon (Butz, 2006). One sorbent developer is investigating several mineral-based sorbents, including fly ash that has metal sulfides applied through controlled deposition, and molecular-sieve-type minerals (Tran, 2006). These sorbents are also injected in the same location and at similar rates compared to how activated carbon is injected for mercury control. Impact on CCPs There has been little published research on the impacts of these alternative sorbents on fly ash. All three sorbent developers cite test results showing fly ash that includes their spent sorbent would not be adversely affected for use in concrete, and would readily pass TCLP analyses.

Mercury Co-Removal with FGD By-Products The effectiveness of mercury co-removal in an FGD system is generally dependent on the percentage of the mercury in the flue gas that is in an oxidized form. The oxidized form of mercury is water-soluble and can be removed in wet or dry scrubbers, while the other predominant form, elemental mercury, is not.

5-8


Several factors influence the percentage of the coal mercury that is in the oxidized form as the flue gas enters the FGD system. These include: •

Coal chloride content – higher coal chloride tends to promote mercury oxidation. The predominant form of oxidized mercury in coal flue gases is believed to be mercuric chloride (HgCl2). Low-chloride coals such as PRB tend to produce mostly elemental mercury in the flue gas (except units equipped with reverse-gas baghouses, as mentioned below).

•

SCR – SCR catalysts have been observed to oxidize mercury, particularly for coals that have significant chloride levels (typically greater than 100 ppm chloride in the coal). Also, it appears that high SO2 to SO3 conversion SCR catalysts are generally more effective at oxidizing mercury than low conversion catalysts.

•

Type of particulate control device – in some cases it has been observed that reverse-gas fabric filters can oxidize a significant percentage of the mercury in the flue gas, even for lowchloride coals like PRB.

Additional technologies have been developed to enhance the percentage oxidation of mercury in flue gas at the FGD inlet. Several technologies are discussed below. Mercury captured by wet FGD systems can be found in the liquor, in the FGD solids, or both. The factors that control how the mercury partitions between phases are not well understood at this time. In general, mercury in the solid phase tends to be concentrated in fine particles. For wet FGD systems that have hydrocyclones for primary dewatering, the fine solids in the overflow stream can have as much as ten times the mercury concentration of the larger particles in the underflow (EPRI, 2004). In contrast, since all of the slurry liquor is evaporated in spray-dryer-based “dry” FGD systems, in these systems all of the absorbed mercury will be found in the solids stream. The net capture of oxidized mercury in an FGD system can be limited by liquid-phase reactions between absorbed oxidized mercury and absorbed sulfur dioxide that cause re-emission of the mercury. In these reactions, the mercury is chemically reduced to elemental mercury, the insoluble form, and is released back into the flue gas. One approach for limiting the amount of re-emission that occurs is to use FGD additives to precipitate absorbed mercury as a fine salt before it has time to undergo such liquid-phase reactions. This approach is also described below. Chemical Addition Technology Description The effectiveness of this technology is based on increasing the percentage of the mercury in the flue gas that is in the oxidized form. As described above, eastern coals with higher chlorine content produce mostly oxidized mercury in their flue gases, while western coals such as PRB, with low chlorine content, tend to produce mostly elemental mercury. One way to increase the amount of mercury removed by wet scrubbers is to add halogens such as chlorine or bromine to the coal to increase the percentage of mercury in the flue gas that is oxidized. 5-9

Mercury Removal

Halogen salts such as calcium chloride (CaCl2) or calcium bromide (CaBr2) can be added to the coal as a means of supplementing the coal chlorine content. Salts added with the coal or injected into the furnace decompose at furnace temperatures to form vapor-phase acids (hydrochloric acid [HCl] or hydrogen bromide [HBr]) in the flue gas. These vapor-phase halogen acids and/or the small amount of halogens present (Cl2 or Br2) react with elemental mercury to produce oxidized mercury that can be removed by a downstream wet scrubber. Results typically show that calcium bromide can be injected at lower rates than calcium chloride to produce equivalent oxidation percentages. Most full-scale test results have shown that higher overall levels of mercury oxidation can be achieved with calcium bromide injection and that this reagent might be more cost effective than calcium chloride injection when the delivered costs of the salt solutions are considered. A patented bromine addition process goes by the trade name “KNX” (Alstom, 2007). The halogen acids added to the flue gas are scrubbed at high efficiency in wet scrubbers and form soluble salts that build to high concentrations in wet scrubber slurries. Chloride salts are known to produce severe pitting corrosion in metallic components unless impervious coatings are applied or high molybdenum or nickel content alloys are used. Bromine is typically not present in coal flue gases at significant concentrations, so the long-term effects of bromine injection on wet scrubber chemistry and wet scrubber materials of construction are not known. Even for calcium chloride injection, the injection of this salt can greatly increase the chloride content of the liquor circulating through the wet scrubber. Impact on CCPs Halogen injection to enhance mercury oxidation could have effects on both fly ash and FGD solids such as gypsum. For fly ash, the main impact would likely be an increase in the amount of mercury removed with the fly ash, as the higher halogen concentrations in the flue gas would tend to enhance mercury capture by unburned carbon in the fly ash. Significant halogen removal with the fly ash would not be expected. For FGD by-products, several effects might be expected. One is that higher mercury concentrations may result in the FGD liquor, FGD solids, or both. The second is that halogen concentrations will be increased in the FGD liquor, as described above. This can increase the amount of halogen material remaining in dewatered products such as gypsum, which may have specifications by the wallboard manufacturer on the amount of dissolved salts allowed. The increased halogen concentration may lead to increased liquor blow down rates from the FGD system to maintain halogen concentrations at manageable levels, which will in turn increase wastewater treatment requirements and wastewater treatment filter cake volumes (discussed in Section 6). Materials of construction issues or halogens in gypsum specifications may be the limiting factor with respect to allowable halogen concentrations. For spray-dryer-based dry FGD systems, the presence of elevated halogen levels in the flue gas will alter droplet drying characteristics, due to the deliquescent nature of the calcium salts of these halogen acids. While SO2 removal performance may be improved, droplet drying will slow and the spray dryer outlet temperature may have to be raised to compensate. The presence of 5-10


these halogen salts in the FGD solids can alter optimum moisture levels, strength development, permeability, and leachate characteristics (Blythe, 1985, 1991). Mercury Oxidation Catalysts Technology Description Catalysts are materials that participate in chemical reactions and allow them to proceed more rapidly, but are not consumed in the reaction. Catalysts can promote the oxidation of elemental mercury in flue gases from low-chlorine coals such as PRB coal. This increases the percentage of the mercury in the flue gas that is in the water-soluble oxidized form, and enhances mercury capture by a downstream wet scrubber. These oxidation catalysts are typically inserted into the flue gas ducts as “honeycomb” type structures with many small passages through which the flue gas flows. This provides a large surface area to which the catalyst materials are applied and on which the oxidation reactions can take place. A 500-MW plant could have 3000 to 4000 ft3 of mercury oxidation catalyst installed. Figure 5-5 shows an individual catalyst block and Figure 5-6 shows an example catalyst module ready to be installed in a pilot unit. A 200-MW-scale demonstration of the technology has been installed at a power plant that fires PRB coal in Texas, and has been in operation since May 2008 (Blythe, 2007a).

Figure 5-5 Example Mercury Oxidation Catalyst Block

5-11

Mercury Removal

Impact on CCPs Because installation is downstream of the ESP or baghouse, oxidation catalysts primarily affect FGD by-products. Increased mercury concentrations would be expected in the FGD liquor, FGD solids, or both. It is possible that these catalysts could impact the oxidation of other vapor-phase metals in the flue gas downstream of the particulate control device, such as arsenic and selenium, but this has not been measured by the process developers.

Figure 5-6 Example Mercury Oxidation Pilot-Scale Catalyst Module (0.5 MW)

Other Mercury Oxidation Technology Photochemical Oxidation or PCOTM is an alternate technology that uses ultraviolet light to selectively oxidize mercury in flue gas (Powerspan, 2007). A pilot-scale evaluation of the technology is currently being conducted on a unit that fires PRB coal. Like oxidation catalyst technology, PCO would primarily impact the FGD by-products through increased mercury concentrations.

5-12


Scrubber Additives Technology Description As described earlier in this section, wet scrubbers are capable of capturing the water-soluble oxidized mercury in the flue gas, but remove little or no elemental mercury. However, the net removal of oxidized mercury can be limited by a phenomenon called “re-emissions,” where oxidized mercury absorbed in the scrubber liquor is chemically converted back to the elemental form. Since the elemental mercury formed is not water soluble, it is released back into the scrubber outlet flue gas. Evidence of re-emissions is seen when the elemental mercury concentration in the flue gas at the outlet of a wet scrubber is higher than the inlet elemental mercury concentration. The chemistry of mercury re-emissions in wet scrubbers is not completely understood at this time. In the future, it may be possible to adjust scrubber operating conditions to eliminate or at least limit re-emissions. In the interim, an approach identified for controlling re-emissions is to use scrubber additives. Additives generally work by rapidly precipitating oxidized mercury from the scrubber liquor as a solid salt, so it will not react with other liquid-phase species such as absorbed sulfur dioxide and be converted back to the elemental form. Scrubber additives have been used in Europe to prevent re-emissions from wet scrubber systems installed on municipal waste burning plants, and have reportedly been used for this purpose in some coal-fired plants. Limited testing of wet scrubber additives has been conducted in the U.S. with varied success (Blythe, 2006, 2007b; Ghorishi, 2005). The scrubber additives are generally water soluble and contain some form of sulfur, often sulfides. Impact on CCPs The main impact of scrubber additives will be increased mercury concentrations in the FGD solids, if the additive is effective. The mercury precipitated with the additive should be concentrated in fine particles. For plants with hydrocyclones for primary dewatering of the FGD by-product, it is possible that the FGD gypsum mercury content can be lowered by separating mercury-rich fines for disposal. It is also speculated that mercury present in gypsum as a precipitate with a re-emissions additive will be more stable than mercury would be in the gypsum with no additive, but this has not been substantiated. In one case, gypsum produced from a power plant using a wet scrubber re-emissions additive was tested at a commercial wallboard plant, with no adverse effects on the wallboard production process being noted (Sanderson, 2006).

5-13

Mercury Removal

Mercury Co-Removal in Wet ESPs Wet ESP technology was previously discussed as an SO3 control technology. However, wet ESPs have also been documented to remove oxidized mercury and even some elemental mercury from the flue gas treated. The latter is speculated to occur because of oxidation of elemental mercury due to the corona electrical discharge in the wet ESP. Mercury removed from wet ESPs would be present in the wastewater from the process, which would either be added to the upstream wet FGD system or would be treated in a wastewater treatment process.

References Alstom, 2007. www.alstom.com, 2007. Barnett, K. 2007. Impacts of Portland Cement NESHAP on Use of Coal Fly Ash. Biermann, 2006. Biermann, Joep, Brian Higgins, Peter Hoeflich, and Bruce Ramme, “Mercury Reduction in Coal Fired Power Plants using MinPlus Sorbent through Furnace Sorbent Injection,” paper presented at the Power Plant Air Pollutant “MEGA” Symposium, Baltimore, MD, August 28-31, 2006. Blythe, 1985. Blythe, G.M. et al. Field Evaluation of a Utility Spray Dryer System. Electric Power Research Institute. Report No. CS-3954, May 1985. Blythe, 1991. Blythe, G.M., L.R. Lepovitz, and C.M. Thompson. Evaluation of a 2.5-MW Spray Dryer/Fabric Filter SO2 Removal System. Electric Power Research Institute. Report no. GS7449. August, 1991. Blythe, 2006. Blythe, Gary M. Field Testing of an FGD Additive for Enhanced Mercury Control, Task 2 and 4 Pilot-scale Test Results. Topical Report. Cooperative Agreement No. DEFC26-04NT42309, URS Corporation, Austin, Texas 78729. May 2006. Blythe, 2007a. Blythe, Gary, Brice Freeman, Bruce Lani, and Charles Miller. “Mercury Oxidation Catalysts for Enhanced Control by Wet FGD,” paper presented at the International Conference on Air Quality VI, Arlington, VA, September 24-27, 2007. Blythe, 2007b. Blythe, Gary M. Field Testing of an FGD Additive for Enhanced Mercury Control, Task 3 Full-scale Test Results. Topical Report. Cooperative Agreement No. DE-FC2604NT42309, URS Corporation, Austin, Texas 78729. May 2007. Butz, 2006. Butz, James, Thomas Broderick, Craig Turchi, and J. Michael Geers, “Trial of Amended Silicates™ for Mercury Control at Miami Fort Station,” presentation at the DOE_NETL Mercury Control Technology Conference, Pittsburgh, PA, December 11-13, 2006. Cooper, L., 2007, “Carbon Burnout – A State of the Art in Commercial Ash Beneficiation,” World of Coal Ash Conference, May 8-10, 2007, Covington, KY. 5-14


EPA, 2006. Characterization of Mercury-Enriched Coal Combustion Residues from Electric Utilities Using Enhanced Sorbents for Mercury Control. EPA-600/R-06/008. EPRI, 2004. Mercury Stability in FGD Byproducts. EPRI, Palo Alto, CA: 2004. 1004711. EPRI 2005a. Mercury in Coal Combustion Products. EPRI, Palo Alto, CA: 2005. 1010061. EPRI 2005b. Scale-up and Demonstration of Fly Ash Ozonation Technology. EPRI, Palo Alto, CA: 2005. 1012995 Ghorishi, 2005. Ghorishi, S. Behrooz, Scott Renninger, George Farthing, Chuck Teets, Jeanne Neureuter, Thomas Hastings, and Chris DiFrancesco. “Effects of SCR Catalyst and Wet FGD Additive on Speciation and Removal of Mercury within a Forced Oxidized Limestone Scrubber,” paper presented at the ICAC Clean Air Technologies and Strategies Conference 05, Baltimore, MD, March 7-10, 2005. Gustin & Ladwig, 2004. Gustin, M, Ladwig, K. 2004. “An Assessment of the Significance of Mercury Release from Coal Fly Ash.” J of Air & Waste Management, V54, pp. 320-330. Landreth, R., Nelson, S., Liu, X., Tang, Z., Overholt, A., Brickett, L., 2007, “Mercury Reduction Performance of Concrete-FriendlyTM C-PACTM Sorbent.” World of Coal Ash Conference, May 8-10, 2007, Covington, KY. Lu, 2007. Lu, Yongqi, Massoud Rostam-Abadi, Ramsay Chang, Carl Richardson, and Jennifer Paradis, “Characteristics of Fly Ashes from Full-Scale Coal-Fired Power Plants and Their Relationship to Mercury Adsorption,” Energy & Fuels, 2007, 21, pp. 2112-2120. Powerspan, 2007. www.powerspan.com, 2007. Romero, 2005. Romero, Carlos, Harun Bilirgen, Nenad Sarunac, and Edward Levy, “Impact of Modified Boiler Control Settings on Mercury Emissions,” presentation at DOE-NETL’s Mercury Control Technology R&D Program Review, Pittsburgh, PA, July 12-14, 2005. Romero, 2006. Romero, Carlos, Harun Bilirgen, Edward Levy, Atira Mabin, Melanie McCoy, Robert Serlin and Craig Tylenda, “Boiler Optimization for Multi-Pollutant Control: Mercury and NOX Emissions,” paper presented at the Power Plant Air Pollutant “MEGA” Symposium, Baltimore, MD, August 28-31, 2006. Sanderson, 2006. Sanderson, Jessica, Gary M. Blythe, and Mandi Richardson. Fate of Mercury in Synthetic Gypsum Used for Wallboard Production. Topical Report, Task 6 Wallboard Plant Test Results, DOE Cooperative Agreement No. DE-FC26-04NT42080, Pittsburgh, PA, December 2006.

5-15

Mercury Removal

Sjostrom, 2005a. Sjostrom, Sharon. Evaluation of Sorbent Injection for Mercury Control, Topical Report for Sunflower Electric’s Holcomb Station, Reporting Period: October 1, 2003 – June 30, 2005. DOE Cooperative Agreement No. DE-FC26-03NT41986, Pittsburgh, PA, June 28, 2005. Sjostrom, 2005b. Sjostrom, Sharon. Evaluation of Sorbent Injection for Mercury Control, Topical Report for AmerenUE’s Meramec Station Unit 2, Reporting Period: October 1, 2003 – September 30, 2005. DOE Cooperative Agreement No. DE-FC26-03NT41986, Pittsburgh, PA, September 23, 2005. Tran, 2006. Tran, Pascaline, PhD, Xiaolin Yang, PhD, Larry Shore, PhD, and William Hizny, “FA 100: Mineral Based Mercury Sorbents,” paper presented at the Power Plant Air Pollutant “MEGA” Symposium, Baltimore, MD, August 28-31, 2006. UND EERC, 2007. Mercury and Air Toxic Element Impacts of Coal Combustion By-product Disposal and Utilization. UND EERC, 2007, Grand Forks, ND: 2007.

5-16

6 ANCILLARY SOLID WASTES ASSOCIATED WITH THE GENERATION AND MANAGEMENT OF COAL COMBUSTION PRODUCTS

There are several ancillary solid waste streams associated with the generation and management of high-volume CCPs described in Section 2. This section discusses some of these materials.

Fabric Filters While coal-fired units have traditionally used ESPs as the primary particulate control device, more and more plants are being built with or retrofitted with fabric filters (baghouses) for particulate control. A number of fabric filter configurations are employed, including shakedeflate, reverse-gas, and pulse-jet cleaning styles. Filter bag materials vary with flue gas conditions and cleaning style, but can include fiberglass and polymers such as PPS (poly ρphenylene sulfide). Filter bags fail over time for a variety of reasons, such as excessive bag weight, low permeability (high pressure drop), and/or loss of physical integrity. Individual failed bags are replaced one at a time, but eventually a complete bag replacement is required. The frequency of bag replacements can vary widely with plant conditions, ranging from about once per year to once per decade. The bags typically develop what is called a “permanent dust cake,” which is a layer of fly ash that cannot be cleaned from the bag surface with conventional cleaning techniques. When fabric filter bags are replaced, the spent bags contain some amount of this permanent dust cake, which should be very similar in composition to the fly ash. As described above, for most U.S. applications the bags themselves are primarily made of fiberglass or PPS. Bags taken out of service are typically disposed of in a landfill or ash pond.

Spent Selective Catalytic Reduction (SCR) Catalysts SCR catalysts are installed in the flue gas path between the economizer and air heater, to promote reactions between injected ammonia and flue gas NOX to lower NOX emissions. With time (typically over several years) they lose their catalytic activity, and must either be regenerated or replaced for the plant to maintain high NOX removal efficiency. The issue of whether catalyst regeneration or replacement is more cost effective has not been fully resolved by U.S. plants that have SCR installed; some utilities have chosen to regenerate while others have replaced with fresh catalyst. In general, just one of several catalyst layers is regenerated or replaced at a given time. 6-1

Ancillary Solid Wastes Associated with the Generation and Management of Coal Combustion Products

The SCR catalysts can either be in a honeycomb or parallel plate configuration. The catalysts contain titanium, vanadium, and sometimes tungsten. The catalysts can also adsorb trace metals from the flue gas, particularly metals that are volatile at economizer exit temperatures, such as arsenic, mercury, and selenium. The spent catalysts may also contain fly ash particles that have deposited over time. EPRI (2003) provides estimates of the annual volumes of SCR catalysts that will need to be replaced or regenerated for coal-fired power plants: 17,000 tons of catalyst will need to be replaced annually by 2010, and 21,000 tons by 2020. Spent catalysts that are not regenerated can either be disposed of (landfill or incinerator) or recycled. Recycling options include use in producing new catalysts, recovery of metals, raw feed to a steel mill, and use as a slag fluidizer additive (EPRI, 2003).

Mercury Oxidation Catalysts Oxidation catalysts begin service at their maximum mercury oxidation activity, then lose activity over time due to poisoning effects of other species in the flue gas. With time, the oxidation of elemental mercury across these catalysts tends to decay. To maintain high oxidation percentages over time, either a large excess of catalyst, frequent catalyst replacement, or frequent catalyst regeneration may be required. Like SCR catalysts, spent mercury oxidation catalysts will eventually have to be replaced. Depending on what type of catalyst is used, there may be enough precious metal content (e.g., gold or palladium) to make it cost effective to recover the metal from the spent catalyst. In this case, the spent catalyst could have a net positive value. If not, it may have to be disposed of. In addition to mercury, after thousands of hours of flue gas exposure other metals may be adsorbed to the catalysts, especially volatile metals like arsenic and selenium. Other than the precious metals, which provide the active component of the catalyst, these catalysts are made from cordierite (aluminum magnesium silicate) and gamma alumina.

Urea Wastes The build-up of wastes in a urea-to-ammonia reactor is highly dependent on the purity of urea and water that are used in the reactor. Any nonvolatile impurities that enter the reactor will stay in the reactor until the reactor has a blow down cycle. Operators manage the build-up of impurities by blowing down the reactor to remove the impurities. Some urea suppliers have heavy metals in their product, which will build up over time in the reactor. Depending on how often the reactor is blown down, the buildup of heavy metals can cause the blown down solution to be classified as a hazardous waste. The level of purity of the urea solution to use in the reactor is an economic decision: lower purity urea will have a lower purchase price but increased waste disposal costs (Spencer, 2008).

6-2


FGD Wastewater Treatment Wastes Wastewater Treatment Solids Many existing FGD installations do not have wastewater treating systems. For example: 1) FGD systems in the west often operate with net water evaporation in ponds; 2) many sulfite-producing FGD systems purge enough water with the calcium sulfite hemihydrate by-product that further water blow down is not required; and 3) other systems blow down from ponds to surface waters subject to limited permit requirements on wastewater quality. However, most new FGD systems produce gypsum, and the amount of water discharged from the system with the dewatered gypsum is not adequate to control dissolved chloride levels. Thus, a blow down stream is required to prevent chloride concentrations from building to unacceptable levels. For new FGD systems, this blow down stream is subject to relatively strict regulations for parameters such as pH, suspended solids, and dissolved metal concentrations, and wastewater treatment systems are required. Before describing this wastewater treatment process, the term “blow down” must be better defined for wet FGD applications. There are two streams that are commonly called “blow down.” One is from the FGD absorber reaction tank(s). As the scrubber operates, lime or limestone reagent is continually added and SO2 is removed by forming calcium sulfite hemihydrate or FGD gypsum solids. With time, the weight percent (wt%) solids level in this slurry builds. To control this solids level, a portion of the slurry is intermittently or continuously blown down and replaced with fresh or reclaimed water. In some plants this blow down is sent directly to a pond or gypsum stack, but in most FGD systems this blow down is sent to the dewatering system. In the latter case, the absorber blow down is not literally leaving the FGD system and/or the power plant; only the dewatered solids leave. Most or all of the reclaimed liquor is returned to the FGD system. The other type of FGD blow down is the purge to control chlorides and/or gypsum fines. This blow down is a portion of the liquor (or dilute slurry) produced during product dewatering that is not returned to the FGD absorbers. Not all FGD system have this blow down stream. For example, in systems that produce calcium sulfite hemihydrate, enough water may leave the FGD system with the dewatered by-product (containing 30 to 40% moisture) that chlorides are naturally controlled to an acceptable concentration. These systems typically would not control fines either. The blow down liquor may contain 2-4% or more of dissolved solids. The major components are calcium, magnesium, sodium, chloride, and sulfate. Trace components include heavy metals from fly ash and limestone, iron from metallic corrosion and grinding media, and other chemicals in non-traditional makeup water sources (cooling tower blowdown, landfill leachate, coal pile runoff, etc.). Minimal amounts of organic compounds are present (other than carboxylic acids if organic acid performance additives are used), so treatment options typically focus on physical/chemical technology.

6-3


For systems that do have a chloride and/or fines purge, the first step in wastewater treatment is typically removal of suspended solids from the blow down stream. These solids include fine FGD product solids (gypsum or calcium sulfite hemihydrate), fly ash removed by the FGD systems, and/or fine particles of inert minerals dissolved from the FGD lime or limestone reagent. A filter press is sometimes used to separate the suspended solids from the liquor. The standard process for metals reduction is pH adjustment (with slaked lime), precipitation of metal hydroxides, settling, pH neutralization, and filtration. Metal hydroxide sludge is formed and removed in clarifiers. One limitation on this technology for FGD liquor is the high concentration of soluble magnesium often present due to the magnesium content of the lime or limestone reagent. Above pH 9, magnesium hydroxide (Mg(OH)2) begins precipitating. Although magnesium is typically not a regulated pollutant, removing it creates a relatively large volume of poorly dewatering sludge as a CCP from the water treating process. Sometimes ad hoc treatment is required for specific metals. For example, sulfide-containing materials are sometimes added in a separate step to precipitate mercury from FGD wastewater. The solids streams (fine suspended solids, metal hydroxides, and mercury precipitates) derived from the treatment process may be combined or may be handled separately. Some of the separated solids can be blended with the gypsum product or can be used in concrete mixes. In many cases the co-mingled filter cake may be disposed. There is also a process called Advanced Biological Metals Removal Process (ABMet®) 10 that claims to remove metals, metalloids, nonmetals, and inorganic compounds such as nitrate. The process uses microbes in a bioreactor to either precipitate elements or destroy contaminants. Each biological system is specifically designed for the contaminants present (GE Water, 2008). Spent Carbon Filtration Beds Carbon filtration beds are used at coal-fired power plants to treat water and wastewater for soluble organic material. Process water from ponds or lakes is treated before use at the plant, and wastewater from the plant is treated before discharge into nearby water systems. If the carbon filtration system is small, carbon is usually landfilled as a non-hazardous waste. Large amounts of carbon are usually sent back to the vendor for reactivation of the carbon (Behrens, 2008).

Demolition Materials When pollution control devices are upgraded or replaced there are many components of the system that are either sold for scrap metal, landfilled, or reused at the power plant. Any metal of value (e.g., ductwork, ESP collector plates) is usually cut into manageable pieces and sold for scrap metal (Curry, 2008). Materials such as insulation and fiberglass pipe are landfilled. It is sometimes required that these materials be washed before disposal if plant solids are still on the items being disposed of. Some components are saved for possible re-use at the plant, such as pumps, valves, piping, instrumentation, stairways, and grating. There is usually a “lay down 10

ABmet is a registered trademark of GE.

6-4


area” at the plant where piping, valves, and other items that can be re-used are stored until the plant has a need for them.

References Behrens, 2008. Conversation with Greg Behrens of URS Corporation on 04/25/08. Curry, 2008. Conversation with Lloyd Curry of FL Smidth on 04/28/08. EPRI, 2003. Recycling and Disposal of Spent Selective Catalytic Reduction Catalyst. EPRI, Palo Alto, CA: 2003. 1004888. GE Water, 2008. www.gewater.com/products/equipment/other_equipment/ABMet.jsp (4/7/2008). Spencer, 2008. Conversation with Herbert Spencer on 05/06/08.

6-5

Export Control Restrictions

The Electric Power Research Institute (EPRI), with major

Access to and use of EPRI Intellectual Property is granted with the

locations in Palo Alto, California; Charlotte, North Carolina; and

specific understanding and requirement that responsibility for ensur-

Knoxville, Tennessee, was established in 1973 as an independent,

ing full compliance with all applicable U.S. and foreign export laws

nonprofit center for public interest energy and environmental

and regulations is being undertaken by you and your company. This

research. EPRI brings together members, participants, the Institute’s

includes an obligation to ensure that any individual receiving access

scientists and engineers, and other leading experts to work

hereunder who is not a U.S. citizen or permanent U.S. resident is

collaboratively on solutions to the challenges of electric power. These

permitted access under applicable U.S. and foreign export laws and

solutions span nearly every area of electricity generation, delivery,

regulations. In the event you are uncertain whether you or your com-

and use, including health, safety, and environment. EPRI’s members

pany may lawfully obtain access to this EPRI Intellectual Property, you

represent over 90% of the electricity generated in the United States.

acknowledge that it is your obligation to consult with your company’s

International participation represents nearly 15% of EPRI’s total

legal counsel to determine whether this access is lawful. Although

research, development, and demonstration program.

EPRI may make available on a case-by-case basis an informal assessment of the applicable U.S. export classification for specific EPRI

Together...Shaping the Future of Electricity

Intellectual Property, you and your company acknowledge that this assessment is solely for informational purposes and not for reliance purposes. You and your company acknowledge that it is still the obligation of you and your company to make your own assessment of the applicable U.S. export classification and ensure compliance accordingly. You and your company understand and acknowledge your obligations to make a prompt report to EPRI and the appropriate authorities regarding any access to or use of EPRI Intellectual Property hereunder that may be in violation of applicable U.S. or foreign export laws or regulations.

Program: Coal Combustion Products - Environmental Issues

© 2008 Electric Power Research Institute (EPRI), Inc. All rights reserved. Electric Power Research Institute, EPRI, and TOGETHER...SHAPING THE FUTURE OF ELECTRICITY are registered service marks of the Electric Power Research Institute, Inc. Printed on recycled paper in the United States of America

1015544

Electric Power Research Institute 3420 Hillview Avenue, Palo Alto, California 94304-1338 • PO Box 10412, Palo Alto, California 94303-0813 USA 800.313.3774 • 650.855.2121 • [email protected] • www.epri.com

EPRI Impact of Air Emissions Controls on Coal Combustion Products

Recommend Documents