Dedication For Anna Konrad Taylar Trevar
Preface Over the course of our careers, the practice of combustion engineeri...
256 downloads
1919 Views
9MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
Dedication For Anna Konrad Taylar Trevar
Preface Over the course of our careers, the practice of combustion engineering applied to solid fuels has changed dramatically. Agents of change have included significant advances in knowledge concerning fuel characteristics and characterization methods, advances in knowledge concerning the processes and mechanisms of combustion, commercial deployment of new combustion processes including bubbling and circulating fluidized bed systems, new methods for pollution control (e.g., selective catalytic reduction or SCR for reducing oxides of nitrogen emissions), and a host of new methods for using computers to enhance the combustion engineering process. Our knowledge of fuels includes increased focus on statistics—inherent variability among fuel deposits or sources. It includes more scientific knowledge on petrography, and the relationship of fuel origins to fuel characteristics. In addition to traditional analyses—proximate and ultimate analysis, ash elemental analysis, etc.—there is increasing emphasis and information on structure and reactivity both of the organic fuel matrix and the inorganic constituents of the various ranks of coal and on other solid fuels ranging from biomass and peat to waste coal and petroleum coke. Methods for analysis now include measurements of pyrolysis and char oxidation kinetics, inorganic reactivity, and more. Techniques for analysis now include use of drop tube reactors and laser-based instrumentation, scanning electron microscopy (SEM) frequently governed by computer controls (CCSEM), Carbon 13 Nuclear Magnetic Resonance (13C NMR), Thermo Mechanical Analysis (TMA) for ash fusion characteristics, and a host of other analytical technologies. These technologies assist in both fuel characterization and combustion mechanism analysis. All of these are additions to—not substitutes for—traditional analyses of coal. New combustion processes include the family of fluidized bed systems: bubbling bed combustors, circulating fluidized bed combustors, and spouted bed combustors. These may be operated at atmospheric pressure or at elevated pressures. New combustion processes also include xix
xx
Preface
oxygen-enhanced combustion for NOx control and pure oxygen-based combustion to facilitate capture and sequestration of carbon dioxide, a greenhouse gas. Many new combustion techniques are driven by regulatory requirements—particularly for NOx control or greenhouse gas capture and sequestration. The computer has made dramatic impacts on combustion engineering in a host of different ways. Computer controls in the form of DCS systems have virtually replaced the older bench boards and hard stations. PI systems and their counterparts now accomplish data archiving, replacing the old circular charts, strip charts, and extensive manual data logging. PI systems and their counterparts also permit data manipulation, making test programs more feasible, and more sophisticated. Computers have been used extensively for modeling. Such modeling efforts include computational fluid dynamics (CFD) modeling of furnaces, boilers, windboxes, ducts, precipitators, etc. Modeling of solids flow in conveying systems and through both bunkers and silos also provides a means for understanding—in real time—the characteristics of the solid fuels being burned. More simplified zoning models and thermodynamic models also are employed. The modeling can be employed successfully because of better characterization of the fuel or fuels being burned. Computer applications have other significant contributions to combustion engineering. These permit more detailed understanding of fuel blending processes and the fuel behavior of blends. Such processes facilitate fueling boilers and kilns that can no longer access their “design fuel.” Computer analyses also provide the basis for technology selection. The driving forces behind many of these improvements include changing economics and increasing environmental regulations. From an economic perspective, solid fuels—particularly the various coals—have become increasingly attractive relative to petroleum and natural gas. Within the various coals, eastern bituminous coal prices have risen faster than subbituminous coals typically found in the Powder River Basin (PRB) and other western coal fields. Consequently PRB coal mines now supply over 400 million tons to the U.S. economy—generating over 20 percent of the electricity used in the United States. Other countries are experiencing similar phenomena. Because petroleum and natural gas are increasingly expensive for power generation and industrial applications, and because of rising coal prices, opportunity fuels such as petroleum coke, tire-derived fuel, wood waste, and other such products have become increasingly attractive to electricity-generating stations and industrial organizations. This, too, has impacted combustion engineering. Economically, a major shift in design approach has also occurred. Twenty-five years ago plants were designed to a “design fuel,” and a long-term
Preface
xxi
contract was signed with a mine. As a consequence, the design was specific to that coal. Today designs consider a range of coals, and potentially opportunity fuels from waste coals to petroleum coke to biomass. This causes significant changes in the combustion engineering approach. Existing units also are modified to handle a range of coals as the availability of the “design coals” recedes. The environmental drivers, not totally divorced from the economic drivers, include more stringent regulations of criteria pollutants: particulates, sulfur dioxide, and oxides of nitrogen. More attention is being given to other pollutants as well: mercury, arsenic and other trace metals, sulfur trioxide and sulfuric acid mist (SO3 and H2SO4), halogen-based emissions (HCl, HF), and more. Greenhouse gases—principally CO2 but also methane (CH4)—are also driving many environmental decisions. The above inventory does not include increasingly stringent water and solid waste concerns, along with the management of hazardous wastes. Because these concerns impact environmental permitting, they impact the selection of fuels and the design and operation of solid fueled installations—hence solid fuel combustion engineering. Combustion engineering is also impacted, increasingly, by policy and political decisions that may or may not have a technical foundation. These decisions—made at the federal, state, and local level as well as at the private industry level—determine to a significant extent how combustion engineering is practiced. Despite all of the dramatic advances in the past 25 years, firing of solid fuels fundamentally involves the following steps: acquiring (and understanding) the fuel or fuels to be burned; preparing the fuels, largely through pulverization; burning the fuels; and capturing the benefits (heat transfer) and managing the consequences (airborne emissions, solid wastes) of the combustion process. The advances and changes that have occurred can be considered within that context. This text attempts to capture many of the major forces and trends within solid fuel combustion engineering. It brings together authors from the academic community, from the consulting world, from manufacturing and system supply industry, and from the solid fuel utilization community. These authors provide a level of expertise essential in their given areas of specialty. This diversity of backgrounds among authors, along with the expertise represented, attempts to reflect the varying perspectives on combustion engineering, as well as the major topics. This book could not have been assembled without the diligent efforts of its many authors, along with the patience and perseverance of many colleagues throughout the combustion community. These colleagues, from all aspects of the combustion community, provided ideas and concepts for
xxii Preface
inclusion; they also provided extensive dialogue concerning topics to include in the book. Family members and friends provided constant encouragement and support for the entire project. With that, we have prepared the book: Combustion Engineering Issues for Solid Fuels. Bruce G. Miller Associate Director Energy Institute The Pennsylvania State University University Park, PA David A. Tillman Chief Engineer – Fuels and Combustion Foster Wheeler North America Clinton, NJ
List of Authors Dao N.B. Duong Performance Engineer, Foster Wheeler, NA Sharon Falcone Miller Research Associate, Energy Institute, The Pennsylvania State University Bruce C. Folkedahl Senior Research Manager, University of North Dakota Energy and Environmental Research Center N. Stanley Harding President, N.S. Harding & Associates Christopher Higman Chief Consultant, Syngas Consultants Ltd. Joe Hoffman Staff Engineer, ECG Consultants Donald Kawecki Vice President – Engineering, Foster Wheeler, NA Jason D. Laumb Research Manager, University of North Dakota Energy and Environmental Research Center Peter Marx President, ACT, Inc. Melanie McCoy General Manager, Wyandotte Municipal Utilities xxiii
xxiv
List of Authors
Bruce G. Miller Associate Director, Energy Institute, The Pennsylvania State University Jeffrey Morin Staff Engineer, ACT, Inc. David Nordstrand Subject Matter Expert – Emissions Control, DTE Energy Michael Santucci President, ECG Consultants James Scavuzzo Vice President, ECG Consultants David A. Tillman Chief Engineer – Fuels and Combustion, Foster Wheeler NA Anthony Widenman Director – Fuels Laboratory, DTE Energy Christopher J. Zygarlicke Deputy Associate Director, University of North Dakota Energy and Environmental Research Center
CHAPTER
1
Introduction David A. Tillman
Chief Engineer – Fuels and Combustion Foster Wheeler NA
1.1 Overview What is combustion engineering? An operational definition would be the application of engineering disciplines—principally mechanical and chemical engineering—to the conversion of fuels into useful forms of energy through the use of combustion processes. It involves the design, construction, and operation of utility and industrial power plants, process industry kilns and furnaces, and a host of similar facilities designed to supply and use fuels. Combustion is the dominant means for converting the potential energy, typically measured in Btu or kilocalories, contained in solid, liquid, and gaseous fuels into useful energy forms. The heat released from combustion can be used directly in thermal applications. More commonly, it is used to raise steam which, when driving turbines or steam engines, can be converted into shaft power or electricity. Combustion engineering applied to solid fuels has a long and illustrious history with practitioners including Thomas Newcomen, Sadi Carnot, James Watt, Lord Kelvin, George Stephenson, Allen Stirling, and many more. These individuals developed the origins of the theory and applications associated with combustion engineering. Combustion engineering for solid fuels involves a diverse collection of disciplines and activities, and it requires understanding of a variety of issues. These issues include an historical perspective concerning combustion of solid fuels, a basic understanding of the chemistry and physics 1
2 Combustion Engineering Issues for Solid Fuel Systems
involved in combustion, and a consideration of the elements of the combustion system from fuel receiving and management from fuel preparation through burning fuel in the boiler system to post-combustion pollution controls. All these considerations come under the umbrella of combustion engineering for solid fuels.
1.1.1 A Perspective on Solid Fuel Utilization Coal, lignite, petroleum coke, wood waste, trees and crops grown as fuel, tire-derived fuel, municipal solid waste, animal wastes, and a host of industrial byproducts and wastes; these are the solid fuels and, collectively, these are the dominant energy sources for electricity generation and much of process industry. Combustion of solid fuels, historically, has been the primary source of energy for industrialization throughout the world. Further, economic development in the 20th and 21st centuries has virtually paralleled the electrification of industrial, commercial, and residential activity (see Figure 1-1) [1]. While all other stationary uses of energy have exhibited virtually no growth, electricity generation continues to increase in its use of all fossil and nuclear fuels, hydroelectric resources, and such renewable resources as wind power. In the United States, this growth in electricity has dramatically increased the use of solid fuels—principally coals of all ranks—as coal, wood waste, urban refuse, and other solid fuels provide the driving force for over 55% of the electricity generated and consumed. Solid fuels also dominate energy supply for such process industries as steel, cement, pulp and paper (including internally generated solid fuels such as
Energy Consumption (1012 Btu)
45000
Electricity
40000 Industrial - Primary Energy 35000 30000 25000 20000 15000
Transportation Residential - Primary Energy
10000 5000 Commercial - Primary Energy 0 1970 1975 1980 1985
1990 Year
1995
2000
2005
2010
FIGURE 1-1 Consumption of fuel in the United States for electricity generation and all other uses (Source: [1]).
Introduction
3
hogged wood waste and spent pulping liquor), and some aspects of mineral ore processing and refining. Intensification of the use of electricity in the economy is but one of several reasons why projections of future energy consumption show increased consumption of coal and other solid fuels on both a relative and an absolute basis, as shown in Figure 1-2 [1]. Increased use of coal—and of all solid fuels (e.g., peat, petroleum coke, biomass, waste-based fuels)— is based on their relative abundance along with the ability to convert these fuels into all useful forms of energy: electricity, liquid and gaseous fuels, and process and space heat. Coal, petroleum coke, wood waste, and other biomass fuels are used in a wide variety of applications. The relative worldwide resources and reserves of coal, peat, wood and other biofuels, and other solid fuels dwarf the worldwide resources and reserves of liquid and gaseous fuels; the long-term future of energy supply remains with the solid fuels. Major deposits of various ranks of coal, peat, and other solid fuels are found throughout the United States, Canada, Russia, China, South Africa, Australia, Germany, Poland, Colombia, and other countries. Estonia operates its economy largely on the use of oil shale as a solid fuel and has reserves for the long-term future. Scandinavia and other regions of the world rely heavily on peat, wood, and other renewable biomass fuels. Peat bogs also supply significant energy resources to Ireland
Annual Energy Consumption (1015 Btu)
60 Petroleum (Liquid Fuels) 50
40
Coal
30 Natural Gas
20 Nuclear 10
0 2005
Renewables 2010
2015
2020
2025
2030
Year
FIGURE 1-2 Projected consumption of coal, petroleum, natural gas, nuclear, and renewable energy sources in the United States in the year 2030. Note that current consumption of coal is about 231015 (quads)/yr. Coal consumption is projected to increase more rapidly and significantly than any other primary energy source (Source: [1]).
4 Combustion Engineering Issues for Solid Fuel Systems
and other nations as well. The vast forests of Siberia, Brazil and the Amazon Basin, and other regions of the globe also must be considered as solid fuel energy resources. Countries such as Brazil are capable of growing agricultural crops for energy supplies on an economically advantageous basis. Some industrial and post-consumer wastes also must be included in the available solid fuels; these include petroleum coke, mixed municipal solid wastes, tire-derived fuel, and a host of other products. Petroleum coke has long been used as a priority solid fuel for some select industries. The pulp and paper industry derives a significant share of its energy from the combustion of spent pulping liquors in chemical recovery boilers. Other specific industrial, pre-consumer, and post-consumer wastes have found niche markets or have found use in generating electricity as a byproduct of incineration.
1.1.2 Fuels and Combustion Technology Development The discovery of fire is often considered to be the most important discovery of mankind [2]; this discovery is commonly considered on a par with the invention of the wheel [2–4]. Taming of fire meant productively using combustion—particularly of solid fuels. Consequently, combustion has been with societies—both the most primitive and most sophisticated—virtually forever. Despite the long traditions in the productive application of combustion and solid fuels, knowledge of this arena has virtually exploded in the past 30 years. During this time, a plethora of new technologies have been developed and commercially deployed. These technologies include (not exhaustive): (1) circulating fluidized-bed combustion, (2) supercritical boilers (boilers with typical main steam conditions of 3500 psig/1000 F) and ultra supercritical boilers (boilers with steam conditions approximating 5000 psig/1200 F main steam), (3) low-NOx firing technologies, (4) integrated gasification–combined cycle (IGCC) technology, (5) advanced post-combustion treatments for pollution control, (6) oxygen-enhanced combustion, and more. New technologies including combustion supported by pure oxygen and oxygen/flue gas mixtures are also under development and will emerge in the coming years. Analytical technologies also have advanced significantly including online, real-time analysis of coal; computer modeling of combustion including computational fluid dynamics (CFD) modeling, carbon 13 and nitrogen 15 nuclear magnetic resonance (13C NMR and 15N NMR) analysis of coal, the application of lasers to coal and combustion analysis, and many more analytical tools and approaches. Software and control technologies have also advanced, including the development of data acquisition systems and computer control technologies. Combustion, as a scientific and engineering arena, has received renewed vigor in recent years. Given the recent advances in combustion science and engineering, several questions merit attention: (1) what are the solid fuels? (2) what is
Introduction
5
the process of solid fuel combustion? and (3) what is involved in the combustion system, applicable to combustion engineering? These questions will be surveyed here and addressed in significantly greater detail in subsequent chapters.
1.2 Solid Fuels Used in Electricity Generation and Process Industry Applications Solid fuels used in electricity generation and process industrial applications include a wide array of materials: anthracite; various ranks and types of bituminous coals; various ranks and types of subbituminous coals, lignites, and brown coals; numerous types of petroleum cokes, oil shale, wood wastes (including spent pulping liquors); and the array of biomass fuels, tire-derived fuel, municipal wastes, and their derivatives; and a host of industrial residues and byproducts. Of these, the coals are the dominant solid fuels. Chapter 2 details the characteristics of coal, and Chapter 3 provides similar information concerning biomass and other opportunity fuels from petroleum coke to tire-derived fuels and agriculturally based biomass. Chapter 4 provides critical information concerning the behavior of inorganic constituents of solid fuels. For purposes of introduction, some salient characteristics of these fuels are summarized next.
1.2.1 Characteristics of Solid Fuels Fuel characteristics typically center on physical and chemical properties. These properties form the basis of subsequent chapters describing various lignites and coals, and various opportunity fuels. Of the analytical values, the standard chemical analyses—proximate and ultimate analyses—are most commonly used to characterize the composition of specific fuels. For the solid fuels, characterizations also include ash elemental analyses to define the basic properties of the inorganic constituents in such fuels. Table 1-1 provides representative proximate and ultimate analyses for various fuels including selected biomass fuels, lignites and various coals, a representative petroleum coke, tire-derived fuel, and mixed municipal waste. Table 1-2 provides representative ash elemental analyses for selected biomass fuels, representative coals, and other solid fuels. Note that the increase in rank from biomass fuel to bituminous coal or petroleum coke involves deoxygenation of the solid fuel and an increase in carbon content. The consequence is a decrease in the atomic hydrogen/carbon ratio and a decrease in the atomic oxygen/carbon ratio as shown in Figure 1-3 [9]. The data in Table 1-1 and Figure 1-3 show that, on balance, as solid fuels increase in rank, reactivity as measured by hydrogen/carbon atomic ratios and oxygen/carbon atomic ratios decreases. These phenomena are
6 Combustion Engineering Issues for Solid Fuel Systems TABLE 1-1 Proximate and Ultimate Analysis of Representative Solid Fuels
Fuel
Weathered Switchgrass Fresh sawdust Beulah Lignite Black Thunder Caballo Rojo Shoshone Crown II (Illinois Basin) Pittsburgh Seam (washed) Pittsburgh Seam Shot Petroleum Coke
Proximate Analysis (wt %, oven dry)
Moisture (% A.R.)
Ultimate Analysis (wt %, oven dry)
HHV (Btu/lb)
VM
FC
Ash
C
H
O
N
S
13.7
81.8
14.8
3.4
49.4
5.9
40.6
0.4
0.3
8,150
42.1 37.0 27.9 31.1 14.7 14.0
80.0 47.8 45.2 45.3 42.6 38.1
19.0 44.9 49.1 48.2 51.3 50.3
1.0 7.5 5.5 6.5 6.1 11.5
49.2 65.1 69.1 70.0 73.1 68.5
6.0 4.1 4.8 4.9 5.4 4.7
43.7 21.6 19.1 23.6 19.1 7.3
0.1 0.6 0.8 1.0 1.4 1.5
— 1.0 0.6 0.6 0.9 1.7
8,400 10,800 12,200 11,900 12,500 12,450
7.1
34.0
55.5
10.5
76.6
5.0
5.0
1.4
1.6
13,600
11.0 6.5
30.6 16.0
55.7 81.5
13.7 2.5
73.6 87.8
4.7 3.7
4.9 0.9
1.3 1.3
1.6 3.9
13,000 14,800
Source: [5]
TABLE 1-2 Ash Elemental Analysis of Representative Solid Fuels Ash Elemental Analysis (wt %) Fuel
SiO2
Al2O3
TiO2
Fe2O3
CaO
MgO
Na2O
K2O
P2O5
Switchgrass Sawdust Utah Bituminous Illinois Basin coal East Ky coal Montana PRB coal Wyoming PRB coal N. Dakota Lignite Petroleum Coke*
71.98 20.38 47.53 44.32 54.69 33.82 32.24 29.8 21.22
1.43 4.05 10.89 19.33 29.92 17.93 16.37 10.0 11.31
0.14 0.06 0.48 0.93 1.37 1.17 1.25 9.0 6.66
0.60 4.57 6.60 22.17 6.23 6.08 5.11 0.4 4.51
9.65 32.6 19.30 3.19 1.80 14.22 21.77 19.0 7.54
3.12 3.51 2.68 0.66 0.99 4.01 5.14 5.0 2.13
0.14 0.85 0.34 0.41 0.48 6.84 1.55 5.8 1.13
3.57 19.9 0.67 2.09 2.31 0.99 1.01 0.49 1.58
2.55 2.49 0.13 0.13 0.29 0.38 1.18 — 1.54
*Vanadium content was 17.8% of ash Source: [5, 6, 7]
explored more completely in Chapter 2. The data in Table 1-2 show that the lower rank fuels—biomass, lignites, and subbituminous coals—have more alkali metal and alkaline earth inorganic constituents. These constituents are more likely to create slagging and fouling issues in their use.
Introduction
7
3.00
Oxygen/Carbon Atomic Ratio
Increasing Rank of Coal 2.50
2.00
1.50
1.00
0.50
0.00 0.00
1.00
4.00 2.00 3.00 Hydrogen/Carbon Atomic Ratio
5.00
6.00
FIGURE 1-3 Coal rank as a function of hydrogen/carbon and oxygen/carbon atomic ratios (Source [9]).
Further, the data available show that the lower-rank coals and biomass fuels have more readily available and reactive inorganic constituents; this further contributes to the slagging and fouling potential of such fuels [10, 11]. In addition to the basic chemical characteristics discussed previously, the solid fuels exhibit physical properties, which must be understood in terms of the use of these fuels. Such properties include porosity, specific gravity and bulk density, and grindability or the ability to reduce the fuel particles to their most usable particle sizes. Thermal properties of significance include heat capacity and thermal conductivity (the ability to conduct heat from the surface of the particle to its center). Additional chemical properties include chemical structure plus pyrolysis and char oxidation kinetic parameters. All these are considered in more detail in Chapters 2 and 3. Virtually all solid fuels are porous structures. Further, most solid fuels consist of aromatic clusters with 1–10 fused aromatic rings linked together by aliphatic bridges (see Figure 1-4) [12]. The aromatic structures provide the backbone of the fuel. These structures include functional groups and heteroatoms linked either to the aromatic structures or to the bridges. The number of fused aromatic rings provides critical insights into the rank of the fuel. Biomass fuels and lignites typically contain single aromatic rings linked by open aromatic structures. Subbituminous coals and bituminous
8 Combustion Engineering Issues for Solid Fuel Systems
H O
H S
H
H
H
H H2 H2 H
H C H HH
O NH2 H2 HH HH H C H H
H H2H H2 H H C H H C H H S H H C H H C H
H
C
H
H
O H
H H H2
C
H C
H
HH
H2
OH H
S C O O H H
S
O
N
O
H C
H
H H O
H
H O
N
H
H
O
H
H C H
H C
C
C
CH3 C
O
H
S H
O
H H
H2 H
H
O H2
H
H H
HH
H O H C H H H H 2 C H H H2 H2
H2
H
O
H
H2
C HH
H H
H
H O
H
H
H
H H
H
H H
H H C H H H O
O H H
H H2
H
C
H
H
H
H2
FIGURE 1-4 Representative structure of subbituminous coal (Source [12]).
coals contain 2–4 fused aromatic rings on average. Anthracites and petroleum cokes contain many fused aromatic rings and far fewer bridges, functional groups, and heteroatoms.
1.2.2 Some Economic Considerations of Solid Fuels The properties of various types of solid fuels determine many of the economic considerations in their usage. For example, the woody biomass fuels with 40–50% moisture, bulk densities of 15–20 lb/ft3, and higher heating values of 4,000–5,000 Btu/lb have a limited transportation radius [7]. Typically this is considered to be 35–50 miles [7]. The biomass fuels such as switchgrass, with 5–10% moisture and 6–10 lb/ft3 have even a more limited effective transportation radius despite higher heating values of 5,500– 7,000 Btu/lb [7]. Lignites are also limited in transportation distance, while subbituminous coals can be moved over 1000 miles by rail or ship, and bituminous coals can be moved even longer distances. Because of the economics of transportation, typical biomass fuel installations are restricted to 500 106 – 700 106 Btu/hr heat input (nominally 50–70 MWe). Lignite and coal-fired plants—including those that burn petroleum coke in the fuel blend—can range in size from the smaller industrial installations to >3,000 MWe (see Figure 1-5). Size begets efficiency to a
Introduction
9
FIGURE 1-5 The Monroe Power Plant of DTE Energy, in Monroe, MI. This 3100 MWe (net) station is fueled by a blend of Powder River Basin and Central Appalachian Coals, blended in the coal yard shown in the upper left area of the picture.
significant extent—and as capacities increase the attention to efficiency also increases. This includes both the combustion system and heat recovery system. It is only the larger units (e.g., >500 MWe) that employ supercritical steam cycles, single and double reheat cycles, extensive use of feedwater heaters, and similar devices. The confluence of size and efficiency can be seen through historical data, as presented in Figures 1-6 and 1-7 [13]. Figure 1-6 shows the increase in boiler size as a function of time, starting in the post World War II era. Figure 1-7 shows the increase in main steam pressure over this period of time. By 1960, virtually all boilers were installed with 1000 F main steam temperatures—the limit of metallurgy at that time. Note that there was one experiment with an ultra supercritical unit (5,000 psig/1,200 F/1,000 F) in the mid 1950s [13]. Since metallurgy could not support this development, the unit was returned to a 3,500 psig/1,000 F/1,000 F configuration. It should also be noted that reheat cycles, invented in the 1920s, became popular after World War II. By 1950–1955, 75% of the boilers installed by the electric utility industry were reheat boilers. By 1960, virtually all boilers
10
Combustion Engineering Issues for Solid Fuel Systems Average MW per Boiler 750.00 700.00 650.00
Megawatts (Name plate)
600.00 550.00 500.00 450.00 400.00 350.00 300.00 250.00 200.00 150.00 100.00 50.00 0.00 1940-1945 1946-1950 1951-1955 1956-1960 1961-1965 1966-1970 1971-1975 1976-1980 1981-1985 1986-1990 1991-1996
Time period
FIGURE 1-6 The growth in the size of utility boilers installed since World War II (Source: [13]).
Increase in Main Steam Pressure over Time 5500
Main Steam Pressure (psig)
5000
Maximum Main Steam Pressure
4500 4000 3500 3000 2500 2000 1500 1000 Average Main Steam Pressure
500 0 1940-1945 1946-1950 1950-1951 1955-1960 1961-1965 1966-1970 1971-1975 1976-1980 1981-1985 1986-1990 1991-1995
Time Period (5-year Increments)
FIGURE 1-7 The increase in main steam pressure in utility boilers since World War II (Source: [13]).
Introduction
11
installed by electric utilities were reheat boilers. However, supercritical boilers received some—but less—acceptance relative to the reheat trends. Consequently, many boilers designed in the late 1970s and early 1980s were installed using subcritical (drum) designs with steam conditions of 2,400 psig/1,000 F/1,000 F. Currently there is resurgence in coal-fired boiler orders. This resurgence, however, focuses on supercritical issues due to advances in metallurgy and techniques to overcome the problems associated with the early supercritical units. The trends in utility boilers contrast with industrial boilers and kilns. These installations are significantly smaller. Reheat is not commonly employed in industrial installations (e.g., pulp and paper industry boilers). Further, this resurgence includes more significant interest in and application of circulating fluidized-bed (CFB) technology. Despite the fact that industrial boilers were not of sufficient size to justify reheat units, they did justify increased attention to system efficiency. Industrial installations joined the parade of units employing extensive preheating of combustion air, attention to the temperatures of gaseous combustion products exiting the combustion system, minimizing the use of excess air, and other approaches to system efficiency. In the cement industry, for example, attention was given to the design of precalcination towers and preheating of combustion air to improve fuel efficiency; use of oxygen-enhanced combustion also came to the cement industry to improve fuel effectiveness and increase system capacities.
1.3 The Combustion Process for Solid Fuels Combustion of solid fuels, which involves both physical and chemical processes, is generally summarized by the following chemical equation: Ca Hb Oc Sd þ (a þ d þ b=4 c=2)O2 ! aCO2 þ 1=2bH2 O þ dSO2 þ heat [1-1] Because solid fuels typically contain some nitrogen and, potentially, chlorine, the equation can be modified to reflect their reactions; specifically, the oxidation of some but far less than all the fuel nitrogen to NOx and the behavior of chlorine as an oxidant, competing with oxygen for available hydrogen. Additionally, most solid fuels contain inorganic matter that may pass through the combustion system largely unreacted, may oxidize, and may undergo phase changes to form liquid matter that becomes slag or fouling deposits. Additionally, the oxygen normally comes from air, which contains 3.76 moles of nitrogen/mole of oxygen, and contains varying amounts of moisture depending on the humidity of the air. All these contribute to the products of combustion, including the amount of heat released that can be captured in useful form.
12
Combustion Engineering Issues for Solid Fuel Systems
Given the general characteristics of solid fuel utilization systems and general properties of these fuels, it is useful to posit an overall combustion mechanism or framework. This framework can then be used to provide an overall method for evaluating the fuels, their uses, and the combustion systems—equipment and processes—used to harness solid fuels as positive economic forces.
1.3.1 Combustion Mechanism Overview The process of combustion involves multiple activities, which are best viewed in terms of reactions of a single particle. The combustion process, as frequently described, is complex [14, 15]. The generally accepted mechanism involves, initially, particle heating and drying. Dried particles then pyrolyze or devolatilize, yielding an array of volatile species and chars. Volatile species further react, as pyrolysis is typically a two-stage mechanism. Subsequently, volatiles are oxidized in a series of free radical reactions. Chars are also oxidized in a series of complex reactions. Subsets of these reactions include behaviors of inorganic constituents including both major inorganics (e.g., silica, alumina, iron, calcium) and trace metals. These overall reaction sequences, shown in Figures 1-8 through 1-10 [15], lead to the operator’s requirements for combustion: time, temperature, and turbulence.
1.3.2 Heating and Drying The solid fuel particle enters the boiler, furnace, or kiln containing moisture. Further, it enters at a relatively low temperature (e.g., room temperature to 150 F). If the particle is being fired in a stoker or cyclone boiler, it is crushed Noncondensible Volatiles
O2, N2
CO2 H2O N2
H2O
Solid Coal Particle
Condensible volatiles
Dry Coal Particle Heat
Heat
Heat
Char (with ash)
FIGURE 1-8 Overall combustion mechanism schematic.
O2, N2
CO2 SO2 Ash
Introduction
S
H
H2
H2
H
H2 H
CH3
H H2 HH H 2H C H H C H H S H H C H H C H
H
H C HH
H H H2
H
H2 HH
H O
H H H2
H
H
H2 H H2
2
CH3
H O
N
H O
H
O C
H H2 H2
H
H H
H
H H
H
H
H
H
O
CH3 H H H 2
CH3 H2
H S
H2 C
H
H O
CH3 N
OH
H
H2 H2
H CH3 O
OH
H H
H H
H O
H
H2 H2 H
H
H O H C
H
H H
H H2
OH H
H
H O
N
H
CH3
H O
O
H
H
H H
H
H
O
CH3
CH3 CH3
CH3
O
H H C H H H
H
H C H H O O H C H HH H C H H H
H
N
H
O
H
H2
OH H
H H2 H
H H C H CH3
H C H H H H2
SH
H2 H
H2
H C H
H2 C H
H H
H
13
H C H
H H
H H
H
H
H H
H
O H H
H
C
H
H
H2 H2
FIGURE 1-9 Representation of the mechanisms involved in the first stage of coal pyrolysis (Source: [15]).
but not pulverized. For stoker firing, the particle can be 3/4" 0"; for the cyclone boiler, the particle is typically 3/8" 0". If the particle is fired in a pulverized boiler, it is typically <200 mesh (74 mm) in particle size. When the particle enters the combustor—boiler, kiln, or process heater—it is heated to reactor temperature. Any moisture in the particle evolves, leaving a heated and dry fuel particle ready for further reaction. Because most solid fuel particles are porous, and many such as low-rank coals and biofuels contain moisture in the pore structure, the process of
14
Combustion Engineering Issues for Solid Fuel Systems
CH4
CH4
H2 NH3
SH
CH4 CH4
H2 H2
H2
C2H4
H2
S H2 C H H
H2O
H2O H2O N
CH4
H2O
CH4
O
H
H2O
CH4
CH4
CH4
H2O
CH4
O N
H2O
C H H C H H H H
HO
H2O
OH
Tar N
S H
C
H O
H CH4
Tar
C C
H
HO
FIGURE 1-10 Representation of reactions in the second stage of coal pyrolysis (Source: [15]).
drying may explode the single fuel particle into multiple fuel particles. Further, the process of heating and drying holds down the particle temperature until the moisture has evolved. The process of heating and drying, being endothermic, is governed by heat transfer to the particle, heat transfer within the particle, and by the heat capacity of the particle. Because drying occurs at temperatures of about 220 F (105 C), with the steam evolving providing cooling of the particle, heat transfer within the particle is of considerable significance. The general formula governing heat transfer within the particle is as follows: q ¼ kA[(T1 T2 )=x]
[1-2]
where q is the flow of heat, k is the thermal conductivity of the fuel, A is the surface area of the fuel particle, T1 is the temperature of the particle
Introduction
15
at the surface, and T2 is the temperature at the center of the fuel particle. Assuming that the particle is a sphere, x is the radius of the fuel particle [15]. Heat capacities limit the rate of temperature rise within the particle. Typical fuel heat capacity values for pure, dry fuel matter are as follows (in Btu/lb-R or cal/g-K): coal, 0.26–0.37; coke, 0.265–0.403; charcoal, 0.242; cellulose, 0.204; and silica (representing coal ash), 0.191 [15]. If one takes the heat capacity of coal on a moisture and ash free (MAF) basis at 0.26 depending on chemical composition according to equation [1-3], uses 1.0 to represent the heat capacity of water, and 0.191 to represent the heat capacity of coal ash, then the heat capacity of any coal can be approximated by equation [1-4] as discussed by Tillman [15]: Cpðmaf coalÞ ¼ 0:189 C þ 0:874 H þ 0:491 N þ 0:360 O þ 0:215 S [1-3] where C, H, N, O, and S are the concentrations of carbon, hydrogen, nitrogen, oxygen, and sulfur, respectively, expressed on a decimal basis. Then CpðcoalÞ ¼ Cpðmaf coalÞ F þ Cpðcoal ashÞ I þ M
[1-4]
where Cp(coal) is the heat capacity of the as-received coal, F is the fraction of MAF fuel in the total coal supplied, Cp(coal ash) is the heat capacity of the coal ash (taken as silica), I is the fraction of inorganic matter in the total coal supplied, and M is the fraction of moisture in the total coal supplied [15]. Similar equations can be created for other solid fuels including biomass, petroleum coke, and the array of opportunity fuels.
1.3.3 Pyrolysis or Devolatilization Devolatilization is the first chemical process occurring as part of solid fuel combustion. Devolatilization occurs by pyrolysis—the heating of the fuel compounds in the absence of sufficient oxygen to cause oxidation of the carbon, hydrogen, and sulfur within the fuel matrix. Pyrolysis of solid fuels, an endothermic series of reactions, typically involves a two-stage mechanism. In the first stage, bridges between aromatic clusters are broken resulting in the formation of both tars and char. In the second stage, functionalities and atoms are stripped from the aromatic cluster forming non-condensable volatile matter. The second stage can occur separately from, or simultaneously with, the first stage. The specific mechanisms of pyrolysis depend on the composition and characteristics of the solid fuel being burned. Of particular interest are such considerations as the number of aromatic rings per fused aromatic structure, the specific composition of bridges between aromatic clusters, the number and composition of functional groups attached to the aromatic structures,
16
Combustion Engineering Issues for Solid Fuel Systems
and the number and location of such heteroatoms as nitrogen and sulfur. Additionally, the percentage and nature of inorganic constituents is of some importance. Inorganic matter that is included in the fuel absorbs heat that would otherwise be available to drive the pyrolysis reactions, slowing the rate of pyrolysis. Representative mechanisms for devolatilization, based on the subbituminous coal shown in Figure 1-4, have been shown previously in Figures 1-9 and 1-10. Typical volatiles produced are both compounds and radicals, and include CO, CO2, CH4, C2H2-6, C3H3-8, H2O, OH, CH3, and OCH3, among others. The yield of volatile matter from pyrolysis or devolatilization is a function both of the chemical composition and structure of the fuel, and of the temperature at which it is reacted. Maximum volatile yields are considerably higher than those measured in proximate analysis when they occur at typical combustion temperatures of 2,500 F–3,500 F, depending on firing method. Volatile matter yields can be expressed either as a percentage (commonly on a moisture and ash free, or MAF, basis) or as volatile matter/fixed carbon (VM/FC) ratios. In burner design, the latter measure is typically shown as the inverse: the FC/VM ratio. Maximum volatile yields for a range of solid fuels including biofuels, lignite, several coals, and a petroleum coke have been measured in a drop tube furnace at typical combustion system temperatures (2,700 F–3,100 F). The values are listed in Table 1-3 and are reported on a dry fuel basis to include the influence of ash on the volatile yields. The specific relationship between the two measures of reactivity is as follows: MVY(%) ¼ 0:697 (VM=FC)
0:18
100
[1-5]
where MVY is maximum volatile yield (%). The r2 for this equation is 0.95 [5]. TABLE 1-3 Maximum Volatile Yield for a Suite of Solid Fuels Fuel Sawdust Weathered Switchgrass Beulah Lignite Black Thunder (PRB) Cordero Rojo (PRB) Shoshone Freeman Crown Pittsburgh Seam (washed) Pittsburgh Seam (as-received) Petroleum Coke Source: [7]
Maximum Volatile Yield (wt %)
Temperature of Maximum Volatile Yield ( C)
92.4 92.5
1000 1500
71.3 71.2 70.2 68.9 62.2 68.4
1700 1700 1600 1600 1500 1705
53.8
1700
50.2
1500
Introduction
17
Aromaticity vs Volatile/Fixed Carbon Ratio from Drop Tube Reactor 18.000
VM/FC Ratio from DTR
16.000 14.000 12.000 10.000 8.000 2 6.000 y = 19.063x2 – 35.275x + 17.642 R = 0.9785 4.000
2.000 0.000 0.00% 10.00% 20.00% 30.00% 40.00% 50.00% 60.00% 70.00% 80.00% 90.00% Aromaticity (%)
FIGURE 1-11 The relationship between aromaticity and fuel volatility (Source: [5]).
The kinetics of devolatilization are a function of the fuel composition—particularly the aromaticity and average number of aromatic carbons per cluster, as shown in Figure 1-11. Fuel rank is a function of aromaticity to a large extent, driving this relationship. Representative Arrhenius relationships for solid fuel pyrolysis kinetics are shown in Table 1-4. Note that the two-stage pyrolysis relationships are shown for woody biomass and a subbituminous coal while a single relationship exists for the petroleum coke tested. The interactions between volatility, aromaticity, and kinetics of devolatilization are shown in equations [1-6] and [1-7] [5]: VM=FCdtr ¼ 19:063A2 35:275A þ 17:642
[1-6]
Eact ¼ 14:425A þ 1:145
[1-7]
where VM/FCdtr is volatile matter/fixed carbon ratio measured at maximum volatile matter production from the drop tube reactor, A is aromaticity of the fuel, and Eact is activation energy (kcal/mol). The coefficients of correlation (r2) for these equations are 0.975 and 0.852, respectively. The evolution of heteroatoms in volatile form is of particular significance. Nitrogen is considered to evolve, initially, largely in the form of HCN; for some of the more reactive coals, lignites, and biofuels, the volatile nitrogen may evolve in one or more amine species as well as in the HCN form [15]. Nitrogen volatile evolution, like total volatile evolution, is a function of the rank of fuel—particularly the aromaticity [5]. As the
18
Combustion Engineering Issues for Solid Fuel Systems
TABLE 1-4 Pyrolysis Kinetics for 10 Representative Solid Fuels Reactivity Parameters* Fuel Weathered Switchgrass Fresh sawdust Beulah Lignite Black Thunder Caballo Rojo Shoshone Crown II (Illinois Basin) Pittsburgh Seam (washed) Pittsburgh Seam Shot Petroleum Coke
Testing Temperatures (oC)
Maximum Volatile Yield (%)
A (1/sec)
E (kcal/mol)
400–1700
90.7
1.91
1.35
400–600/600–1,000 600–1,000/1,000– 1,700 800–1,700 600–1,000/1,000– 1,700 800–1,700 600–1,000/1,000– 1,500 800–1,700
92.4 71.3
1.17/5.74 1.87/142
0.68/3.42 1.87/12.4
70.0 69.6
59.1 5.33/89.8
9.53 3.68/10.8
68.9 62.2
36.2 5.88/338
8.06 4.08/14.6
68.4
89.5
10.7
52.9 50.2
66.2 104
10.3 11.5
1,000–1,700 1,000–1,700
*The two-stage reactivity constants conform to the two-stage temperatures as shown. Source: [5]
total fuel rank increases from biomass to anthracite and petroleum coke, nitrogen evolution in volatile form as opposed to char form decreases. Sulfur atoms tend to become part of the char matrix; however, some sulfur evolves in volatile form, in such compounds as H2S. Char is the solid product of pyrolysis and is the fuel matter that does not get released as volatile matter during the pyrolysis process. Char is a carbon-rich product which may include residual amounts of oxygen, sulfur, nitrogen, and inorganic matter. Little, if any, hydrogen remains in the char product. The char produced is a highly porous material available for heterogeneous gas-solids oxidation reactions. Char yields are inversely impacted by temperature; lower temperature devolatilization increases char yields, whereas higher temperatures of pyrolysis decrease char yields.
1.3.4 Volatile Oxidation Reactions Volatile oxidation involves a series of free-radical reactions including chain initiation, chain propagation, and chain termination. Chain initiation occurs with the volatile species and fragments evolving from the solid fuel matrix. Of particular significance is the concentration of hydroxyl radicals (OH). These are among the most reactive species in the combustion process, promoting both chain initiation and chain propagation reactions. Volatile oxidation reactions occur extremely rapidly.
Introduction
19
Any number of equations can be used to represent the volatile oxidation process including, for example, the following chain initiation reactions [16]: C2 H6 þ M ! CH3 þ CH3 þ M
[1-8]
CH4 ! CH3 þ H
[1-9]
H2 O ! OH þ H
[1-10]
M represents heat removal matter such as inorganic constituents in the fuel. Any number of such reactions can be posited. Once chain initiation has commenced, with formation of numerous highly reactive free radicals such as the hydroxyl radical, chain propagation commences at a vigorous rate. A few examples of such reactions are as follows [16]: H þ O2 ! HOO
[1-11]
HOO ! OH þ O
[1-12]
CH3 þ O2 ) H3 COO
[1-13]
H3 COO ) H3 CO þ O
[1-14]
The potential for such reactions is limitless; consequently, one can posit hundreds of reactions comparable to those shown in the preceding equations. Chain termination completes the process and a few typical examples of such reactions can be shown as follows: CO þ O þ M ) CO2 þ M
[1-15]
OH þ O þ M ) H2 O þ M
[1-16]
1.3.5 Char Oxidation Reactions Char oxidation is a highly complex process. Mechanistically, char oxidation commences when volatile release is largely completed, and when oxygen from the combustion environment can reach the surface of the char particle. At that point, char oxidation commences. Char oxidation is largely diffusion controlled. Generally, char oxidation rates are governed by partial pressure of oxygen in the furnace as a whole and at the surface of the char particle, and char particle temperature [15]. At temperatures
20
Combustion Engineering Issues for Solid Fuel Systems
above about 1,650oF, the rate of oxygen consumption—of char oxidation— is totally governed by diffusion of oxygen to the surface of the char particle. At such temperatures, activation energies are on the order of 4–5 kcal/ mole, as would be anticipated by a mass transfer (diffusion controlled) model. At lower temperatures, chemical kinetics play an increasing role, with activation energies existing as a function of the type of solid fuel being burned. The more reactive solid fuels, with lower aromaticities and with fewer aromatic carbons per cluster, generally exhibit lower activation energies for char oxidation. Any number of char oxidation reactions have been proposed by researchers including those shown by Bradbury and Shafizadeh [17, 18] and Mulcahy and Young [19], as shown here: C þ O2 ! C(O) ! C(O)m ! CO þ CO2
[1-17]
C þ O2 ! C(O)s ! CO2
[1-18]
and
where C(O)m is a mobile C(O) site and C(O)s is a stable C(O) site. Additionally, char oxidation can proceed by reaction with the hydroxyl radical: 2OH þ C ! CO þ H2 O
[1-19]
Reaction [1-19] is particularly interesting because it illustrates the impact of hydroxyl radicals on the heterogeneous gas-solids reactions of char oxidation. Gasification reactions also contribute to char oxidation, particularly at high temperatures. These include the following [15]: C þ CO2 ! 2CO
[1-20]
C þ H2 O ! CO þ H2
[1-21]
2C þ H2 O ! C(O) þ C(H2 )
[1-22]
Any number of other reactions can be proposed to describe char oxidation; these are simply the most commonly proposed simplified reactions. Essentially, char oxidation proceeds along pathways that produce combustible gaseous compounds that then oxidize completely. The dominant reactions are those where oxygen penetrates to the surface of the char particle producing, and releasing, CO for subsequent oxidation. The gasification reactions, although present, contribute significantly less to char oxidation [15].
Introduction
21
TABLE 1-5 Char Oxidation Kinetics Parameters for a Suite of Solid Fuels Fuel
A (1/sec)
E (kcal/mol)
Switchgrass Sawdust Beulah Lignite Black Thunder (PRB) Cordero Rojo (PRB) Shoshone Freeman Crown Pittsburgh Seam (washed) Pittsburgh Seam (as-received) Petroleum Coke
1.26Eþ9 1.63Eþ5 1.30Eþ5 7.61Eþ4 1.83Eþ4 3.24Eþ5 1.72Eþ7 3.72Eþ5 3.27Eþ9 9.26Eþ8
35.0 25.7 27.0 27.4 26.2 29.0 35.7 32.3 40.4 42.2
Source: [7]
Char oxidation for high ash fuels—lignites, some biomass fuels, some waste fuels—is influenced by the presence of inorganic compounds. These compounds absorb heat available to drive reactions. Penn State’s Energy Institute measured char oxidation kinetics, and the results are shown, for 10 solid fuels, in Table 1-5.
1.3.6 Formation of Airborne Emissions Combustion reactions also produce airborne emissions. Among the volatile oxidation reactions of most significance are those promoting or inhibiting the formation of pollutants such as carbon monoxide (CO), sulfur dioxide (SO2), and oxides of nitrogen (NOx). CO formation largely occurs in the absence of sufficient oxygen for complete combustion of the fuel, and it occurs when CO ! CO2 reactions are quenched by insufficient temperatures available to support complete combustion. SO2 is directly formed as a combustion product, as the sulfur oxidizes virtually completely. Some may be captured by calcium contained in the inorganic matter of the fuel, however. This is typically no more than 10–13% depending on the type of fuel. Some SO2 is further oxidized to SO3. The SO3 may contribute to blue haze as a visible pollutant. Alternatively, it may become a plant problem if it reacts with moisture and cools to a condensed phase as sulfuric acid. This can attack expansion joints, electrostatic precipitators, and other elements of the combustion system. NOx represents a class of compounds including NO, NO2, N2O, and other compounds. Of these, NO is, by far, the dominant product of combustion. However, it can oxidize in the environment to NO2. NOx formation may occur either by oxidation of nitrogen from the fuel or by oxidation of nitrogen contained in the air. The former, known as fuel NOx, can be managed by creating the volatile nitrogen in an oxygen-deficient environment (staged combustion). The latter, known as thermal NOx, requires very high
22
Combustion Engineering Issues for Solid Fuel Systems
temperatures to promote the Zeldovich mechanism. The Zeldovich mechanism is typically represented by the following reversible reactions [16]: N2 þ O $ NO þ N
[1-23]
N þ O2 $ NO þ O
[1-24]
N þ OH $ NO þ H
[1-25]
Of these, reaction [1-23] has a very high activation energy and is generally considered to be the rate-determining step. Reaction [1-24] is governed by oxygen availability [16]. Additionally, thermal NOx can be formed by reaction with hydrocarbon fragments produced during solid particle pyrolysis with the initial reactions being as follows [15]: CH þ N2 $ HCN þ N
[1-26]
C þ N2 $ CN þ N
[1-27]
Note that volatile nitrogen radicals and compounds are formed that can readily be oxidized according to extended Zeldovich mechanism reactions or by conventional volatile oxidation reactions. Additional NO is formed by the “prompt NO” mechanism elucidated by Fenimore [16].
1.3.7 Reactions of Inorganic Matter All solid fuels contain varying amounts of inorganic matter including the major inorganic constituents: silicon, aluminum, titanium, iron, calcium, magnesium, sodium, potassium, and phosphorus. Typically, in fuel analyses, these are expressed as oxides (see Table 1-2). However these minerals come in a variety of forms (e.g., silicon found in quartz; iron found in pyrites). Additionally, the solid fuels contain varying concentrations of a host of trace metals including (not exhaustive): antimony, arsenic, barium, cadmium, chromium, copper, lead, mercury, nickel, strontium, tin, vanadium, and zinc. It has been said that if a metal exists in the periodic table it can be found in one or another deposit of coal or in many of the biomass fuels. The behavior of the inorganic materials in the solid fuels is driven by temperature and the susceptibility of compounds to temperature. The issues associated with inorganic matter in solid fuels are discussed extensively in Chapter 4. It is of critical importance in the design and operation of solid fuel combustion systems. In simplistic form, however, it should be noted that the basic inorganics are the primary creators of deposition—slagging and fouling— and of particulate emissions. On the acid side are silica, alumina, and titania; on the basic side are iron oxide, calcium oxide, magnesium oxide, sodium oxide, and potassium oxide. The base acid ratio (B/A), shown mathematically
Introduction
23
in equation 1-28, is historically a primary determinant of slagging and fouling behavior and of the viscosity of slag measured as the T250 temperature (the temperature at which the slag exhibits 250 poise). These historically have been primary determinants of design and operation of pulverized coal (PC), cyclone, and stoker-firing systems but, as discussed in Chapter 4, other techniques are being developed as well. B=A ¼ (%Fe2 O3 þ %CaO þ %MgO þ %Na2 O þ %K2 O)=(%SiO2 þ %Al2 O3 þ %TiO2 )
[1-28]
In addition to the B/A ratio, calcium and iron both act as fluxing agents, depressing the temperatures at which inorganic matter softens and becomes slag. Further, these fluxing agents can create eutectics to further enhance the slag formation. The CaO/Fe2O3 ratio is a clear measure of the potential for forming eutectics to enhance the fluxing activity of the coal [14, 15]. Several additional measures obtained from chemical fractionation and other advanced characterization techniques further refine the understanding of the behavior of inorganic matter in solid fuels. Trace metals can exist in the fuel matrix either as discrete inorganic minerals within the porous fuel structure (e.g., arsenopyrites) or bonded to the overall fuel structure as metal chelates. In waste-based fuels or materials, trace metals can exist as free metal compounds (e.g., in discarded batteries). Specific forms of metals in the solid fuels depend heavily on the metal and on the specific fuel. For coals this may vary as a function of coal rank and of individual coal deposit. For biomass fuels such as wood, the concentration and form of metals can be a function of the growing environment and the presence or absence of specific metals in the available atmosphere, in the fertilizer used to grow the wood, or in other available sources of such elements [20–22]. For some waste-based products, the metals exist as a consequence of processing. For example, zinc is found in significant concentrations in tirederived fuel. Zinc is an accelerant in the vulcanizing process used to make the synthetic rubber used in tires. Vanadium and nickel are found in petroleum cokes, in concentrations depending on the source of crude oil. Their concentration results from the coking process; since these are not highly volatile metals, they concentrate in the solids remaining from coke production. Trace metals can be classified either as chalcophiles (with affinity for sulfur) or lithophiles (affinity for oxygen). Arsenic, cadmium, lead, and zinc are representative examples of chalcophiles, and beryllium, chromium, nickel, and vanadium are representative examples of lithophiles. As a general rule, chalcophiles tend to volatilize and produce products in the vapor phase for subsequent reactions. Lithophiles tend to melt and concentrate more either in bottom ash or equally in bottom ash and fly ash. Some trace metals, upon entering the combustion environment, volatilize completely (e.g., mercury). Others such as arsenic, cadmium, and lead
24
Combustion Engineering Issues for Solid Fuel Systems
volatilize to a very large extent and are enriched in the fly ash. Still others such as manganese are distributed equally between bottom ash and fly ash in combustion systems [20]. Volatilization behavior is largely a function of the type of metal, combustion temperature, and vapor pressures of the metal. Table 1-6 presents some volatility temperatures for some representative metal oxides and metal sulfides. It should be noted that Table 1-6 assumes no chlorine in the combustion system, save for one example with lead. Chlorine reduces the volatility temperatures for many metals (e.g., lead), whereas sulfur can increase the volatility temperatures of selected metals. Table 1-7 presents representative trace metal products of combustion [20]. It should be noted that trace metals produce very small quantities of emissions. Estimates made for the worldwide annual trace metal emissions from coal combustion in electric utilities include the following: arsenic, 0.2–1.6 thousand tons; cadmium, 0.1–0.4 thousand tons; chromium, 1.2– 7.8 thousand tons; mercury, 0.2–0.5 thousand tons; nickel, 1.4–9.3 thousand tons; lead, 0.8–4.7 thousand tons; vanadium, 0.3–4.7 thousand tons; and zinc, 1.1–7.8 thousand tons [23]. It is recognized that these emissions were estimated almost 20 years ago; however, they represent the magnitude of trace metal emissions annually.
1.3.8 Combustion and Heat Release The overall purpose of combustion, particularly with solid fuels, is to produce useful heat that can be used to raise high-pressure steam for power generation, low or medium pressure steam for process applications (e.g., power boilers in the pulp and paper industry), or direct heat for process applications such as firing cement kilns or ore processing kilns. The TABLE 1-6 Some Metal Volatility Temperatures for Some Metal Oxides and Sulfides Temperature Metal Arsenic Beryllium Cadmium Chromium Lead
Mercury Nickel Vanadium Zinc
Compound As2O3 BeO BeS CdO Cr2O3 PbO PbO2 PbS PbCl4 Hg NiO V2O5 ZnO
o
o
170 3,500 2,820 1,120 5,400 1,730 570 930 4.4 57 2,700 1,850 1,930
77 1,927 1,549 604 3,000 943 299 499 15 14 1,485 1,010 1,057
F
C
Introduction
25
TABLE 1-7 Some Representative Metal Products of Combustion in the Absence or Presence of Chlorine Metal
No Chlorine in System
10% Chlorine in System
Arsenic
As2O3/As2O5/calcium arsenates/Ca3(AsO4)2 Ba(OH)2 Be(OH)2 Cd/CdO CrO2/CrO3/Cr2O3 PbO/PbO2/Pb Hg/HgO Ni(OH)2/NiO ZnO
As2O3/As2O5/calcium arsenates/ Ca3(AsO4)2 BaCl2 Be(OH)2 Cd/CdO CrO2/CrO3/Cr2O3 PbCl4/PbCl2/PbO Hg/HgO NiCl2/NiCr2O4/NiO ZnO
Barium Beryllium Cadmium Chromium Lead Mercury Nickel Zinc Source: [20]
efficiency of heat release is a function of flame temperature, as well as the effectiveness of the heat transfer method and equipment. Adiabatic flame temperatures can be calculated for any fuel; alternatively, theoretical flame temperatures can be calculated incorporating CO2 dissociation and other similar reactions, using Gibbs Free Energy minimization techniques. Theoretical flame temperatures and equilibrium products of combustion can be calculated using the CET-89 code of Sanford Gordon and Bonnie McBride [24], or other Gibbs computer codes. Using these codes, flame temperature approximation equations can be calculated for bituminous coals as follows:
Tf ( F) ¼ 3852 þ 0:066(HHV) 1087(SR) þ 0:61(Ta )
[1-29]
where HHV is the higher heating value of the coal in Btu/lb, SR is the stoichiometric ratio, and Ta is the temperature of the combustion air above 77 F [15]. Alternatively, for biomass, the equation is as follows:
Tf ( F) ¼ 3870 15:6(MC) 130:4(EO2 ) þ 0:59(Ta )
[1-30]
where MC is the moisture content of the material on a percentage basis, and EO2 is the percentage of excess oxygen in the flue gas (total basis) [15]. Driving combustion to maximum flame temperatures is not necessarily the most desirable practice, however, due to the influence of high temperatures on NOx formation. The practice of generating useful heat depends on the combustion system and the heat transfer approach. Consequently, numerous firing systems have been developed, as discussed in Chapter 7, including grate firing, suspension firing, cyclone firing, and fluidized-bed combustion.
26
Combustion Engineering Issues for Solid Fuel Systems
1.4 The Combustion System Harnessing the combustion process—and creating useful energy from solid fuels—has practical design and operational consequences. These consequences manifest themselves in the design and operation of combustion systems. Given the diversity of fuels—coals, lignites, biomass fuels, petroleum coke, tire-derived fuel, waste-based fuels, and more—and given the diversity of applications, combustion systems can become highly complex. Combustion engineering, by itself, is of little practical use until it is applied to such applications as electricity generation, process steam raising for such industries as pulp and paper, process heat raising for such applications as cement kilns, and incineration of waste materials. Combustion systems go well beyond the burners and boilers. Combustion systems start with the receipt of the fuel or solid material to be burned, and understanding fuel or combustible material quality. From there, combustion engineering considerations include fuel management and then fuel preparation (e.g., pulverizers for coal, petroleum coke, and other suspension-fired fuels). From fuel preparation, the combustion system then includes the fuel firing systems themselves—pulverized fuel burners, cyclone burners, spreaderstokers, or fluidized-bed systems. Incorporated in the combustion systems are certain elements of the boiler, specifically sootblowers. Finally, combustion systems include post-combustion controls including electrostatic precipitators, baghouses or fabric filters, selective catalytic reduction (SCR) systems for NOx control along with combustion-based NOx reduction systems (e.g., selective non-catalytic reduction [SNCR], reburn systems, separated overfire air). The relationships between the fuel and its preparation, the management of the combustion process, and controlling the products of combustion (heat, bulk gaseous combustion products, pollutants) all are a part of combustion engineering. Gasification systems are specialized combustion-related systems that merit consideration by combustion engineers. Understanding the combustion process involves both combustion controls and combustion modeling. Combustion controls, and associated software such as data acquisition systems, provide sufficient information for the operation of combustion processes. Combustion modeling provides mechanisms for optimizing the performance of combustion systems—frequently in close association with combustion controls. While Chapters 7 and 8 provide critical insights into these systems, major elements are summarized in the following sections.
1.4.1 Fuel Quality and Fuel Management Fuel quality issues commence when the solids to be burned are brought to the site of the combustion system (e.g., the electricity-generating station). With increasing competition among electricity generators as a consequence of the Public Utilities Regulatory Policies Act (PURPA) and, more recently,
Introduction
27
deregulation of electricity generation, knowledge of fuel and fuel quality is increasingly important. This involves laboratory analysis of fuels and, to a limited extent, on-site analysis of fuel quality. Issues include not only heat content (Btu/lb or kcal/kg) and moisture content, but also chemical analysis of the fuel and the ash contained in the fuel. Fuel management increasingly involves fuel blending to control the quality of the fuel being fired in the boiler or kiln. For plants such as Monroe Power Plant of DTE Energy, this includes online analysis of the coal, and determining the appropriate blend based on the properties of the coal being sent to the plant. It also includes careful consideration of the influence of blending on fuel properties including impacts of blending on fuel volatility and impacts of blending on ash slagging and fouling behavior based on eutectics [25, 26]. Blending also considers certain physical properties such as Hardgrove grindability index (HGI). The HGI, along with moisture, substantially determines the available capacity of pulverizing equipment. The blending system itself is of considerable importance. Its ability to accurately blend, through the use of variable speed conveyors and accurate scales, is one approach. A bucket of this coal and a bucket of that coal is mixing, but it is not blending and its consequences can be deleterious to plant operations. Blending can go beyond varying the types of coal being fired in a boiler, or the incorporation of multiple forms of biomass and waste-based fuels in the feed to an industrial facility. Blending can involve dissimilar fuels such as cofiring wood waste or switchgrass in a pulverized coal or cyclone boiler, or incorporating petroleum coke and/or tire-derived fuel into the feed for a utility operation [7, 8, 27, 28]. Again, the system needs careful control over the proportion of each fuel in the blend, and it requires some detailed knowledge of the fuels, and their blended properties, in the firing system.
1.4.2 Fuel Preparation Fuel preparation depends on the firing system. For stokers and fluidized-bed boilers, preparation involves crushing to size. In the case of biomass fuels— woody materials, crop wastes, biomass fuel crops, and the like—preparation involves largely chopping and screening. This is particularly the case when the firing system is a stoker or grate-fired boiler, or a fluidized-bed boiler. Suspension firing systems such as pulverized coal (PC) boilers and kilns require the use of mills or pulverizers to reduce the fuel particle sizes to appropriate levels. Typically, suspension firing requires a fuel pulverized to <½% retained on a 50 mesh screen (300 mm), and >70% passing through a 200 mesh screen (74 mm). For initial preparation, crushing and grinding machines are employed. These reduce coals to such sizes as 3/8" 0". If the coal is fired in a cyclone, that is sufficient preparation. However, if suspension firing is the system employed, then mills or pulverizers are included in the process.
28
Combustion Engineering Issues for Solid Fuel Systems
For biomass fuels and most waste-based fuels, screening and coarse shredding or chopping are normally the only preparation employed. In rare circumstances, producing a very fine particle is necessary. Pulverized coal (and petroleum coke) firing is the most common form of solid fuel combustion; consequently, it merits the most attention. There are numerous types of pulverizers including roll-wheel pulverizers, bowl mills, ball-and-race mills, horizontal ball mills, deep-set hammermills (Atritta mills), and more. Each has advantages and disadvantages depending on specific application [14]. Pulverizers accomplish both particle size reduction and drying of the fuel. They consume primary air both to achieve the drying function and to serve as the transport medium conveying the pulverized fuel to the burners. Pulverizers are among the most critical elements associated with combustion engineering.
1.4.3 Burners and the Combustion Systems Burner design depends on the specific configuration of the combustion system involved. For pulverized coal boilers, burners may be designed as wall-fired burners installed either on the front wall or on both the front and back walls. Wall-fired burners may be designed as swirl-stabilized burners, or designed with impellers to spread the fuel in the furnace. Each burner creates its own flame, and there is limited physical interaction between the flames of the wall-fired burners. Commonly, large wall-fired suspension burners are sized to support ~25 MWe of electricity generation. Alternatively, suspension firing may involve tangential or corner firing. Such burners are fundamentally different from wall-fired burners. To some extent, the primary furnace may be considered as a single burner with multiple injection points for fuel, since a single fireball is created. Air supply systems are critical to suspension burners, and consequently the combustion engineer must be mindful of the forced draft (FD) fan systems. Air is supplied both as primary (transport) air and as secondary air for wall-fired and tangentially fired boilers. For many systems, separated overfire air systems are also employed to stage the combustion mechanisms and reduce the formation of NOx. Pulverizers typically support 4–8 burners. Pulverized firing systems typically generate flame temperatures of 2,500– 2,900 F (1,370–1,590 C). In PC firing, nominally 80% of the inorganic matter becomes fly ash and is removed by the air pollution control system. In PC firing, sootblowing is of critical importance for the management of deposits of ash and slag on furnace walls and boiler tubes. Management of the combustion system to minimize or eliminate slagging and fouling deposits becomes of great significance. This management process involves fuels knowledge, fuels management, and combustion or firing system control. PC firing is used not only in boilers, but also in process industry kilns for the direct generation of process heat. Cyclone firing systems employ larger burners, typically supporting 35–50 MWe. Cyclone barrels are commonly 9–12 feet in diameter and
Introduction
29
10–14 feet in length. In the cyclone barrel, temperatures commonly reach 3,500 F (1,930 C). Heat release rates are on the order of 500,000 Btu/hr-ft3 inside the cyclone barrel. Heat release rates for the total primary furnace, however, are comparable to those associated with PC firing. In cyclone firing, 12–15% of the air is supplied as primary air in the scroll section of the cyclone barrel, and 85–88% of the air is supplied as secondary air swirling around the barrel and maintaining the cyclonic action. Cyclone firing is designed for slagging fuels, and it is designed to manage the slagging properties of coal or other fuels by generating a slag and removing it in liquid phase through slag taps in the bottom of the boiler. Ideally, some 70% of the inorganic matter is removed as slag, and only 30% becomes fly ash. Again, attention is paid to slagging and fouling deposits formed on boiler tubes. Stoker firing, fluidized-bed firing, and gasification are alternatives used in specific applications. Stokers are commonly used for waste-based fuels and biomass fuels where minimal fuel preparation is preferred and where a premium is placed on fuel flexibility. Stokers may be spreaderstoker fired on traveling grates or, for smaller systems, gravity fed onto either moving or fixed grates. Fluidized-bed boilers and combustors also exhibit fuel flexibility. These may be either bubbling bed or circulating fluidized-bed (CFB) systems. There is also effort to utilize pressurized fluidized beds as well as atmospheric fluidized beds. Recently, significant advances have been made in increasing the capacity of CFB systems and incorporating reheat steam cycles into these designs. Gasification is becoming interesting as an alternative to direct combustion in the generation of power and also the providing of fuel for process heat applications. Gasification coupled with combustion turbine technology and combined cycle technology in integrated gasification-combined cycle installations such as Polk County, Florida, shows promise for dramatically increasing the efficiency of electricity generation with solid fuels while reducing airborne emissions dramatically.
1.4.4 Post-Combustion Controls Post-combustion controls exist for particulates, NOx, and SO2; many of these technologies also provide measures of control for such trace metals as arsenic and mercury. Some of these technologies, such as SCR, can also be negatively influenced by such trace metals as arsenic. There are a host of technologies, as mentioned previously. What is becoming increasingly critical in combustion engineering is the recognition of the influence of fuel quality and combustion practices on the effectiveness of such technologies. Inorganic composition (e.g., percent SiO2, Al2O3) has a dramatic impact on the performance of precipitators, particularly as the combined percentage of those compounds approaches and exceeds 80% of the ash. When blended fuels are used, the threshold for impact can be lower.
30
Combustion Engineering Issues for Solid Fuel Systems
Firing practices significantly influence NOx formation, hence the performance of the SCR and its alternatives. Fuel selection and management, and firing practices, are the first line of defense in controlling NOx emissions. Scrubber technologies available include conventional wet scrubbers and, as alternatives, wet-dry scrubbers and dry scrubbers. Scrubber system considerations also include the chemistry of synthetic gypsum production. All these issues—fuel quality, fuel management, fuel preparation, firing, and post-combustion controls—are parts of combustion engineering. They are supported by combustion controls and combustion modeling. All these issues are dictated by federal and state policies impacting combustion systems through regulatory requirements for the design, installation, and operation of combustion systems.
1.5 Organization of This Book This book provides significant information on the issues associated with combustion engineering applied to solid fuels. Chapters 2–4 are designed to provide key information on fuel quality, including recent developments in fuel and ash characterization. Chapter 5 deals with a specific topic in fuel management— fuel blending. Through this discussion, the interplay between fuel quality issues and combustion issues is elucidated. Chapters 6–9 deal with the equipment associated with fuel preparation and combustion, and the performance of associated equipment. Chapter 10 provides insights into techniques for combustion modeling and the application of those techniques. Chapter 11 discusses gasification as an alternative to direct combustion. Chapter 12 considers how the combustion engineer responds to policy issues. In providing this approach, this text attempts to discuss the total interplay between the fuel issues of selection, knowledge, management; the firing issues of technique and combustion management; the concerns of managing the consequences of combustion; the concerns of optimizing combustion through the application of many modeling techniques; and the issues associated with response to policies creating opportunities or limiting choices for the combustion engineer.
1.6 References 1. Energy Information Administration. 2007. Monthly Energy Review. U.S. Department of Energy. May 24. Also see EIA. Annual Energy Review. July 2006. 2. Thirring, H. 1956. Energy for Man. New York: Harper. 3. Tillman, D.A. 1978. Wood as an Energy Resource. New York: Academic Press. 4. Glesinger, E. 1949. The Coming Age of Wood. New York: Simon and Schuster.
Introduction
31
5. Tillman, D.A., B.G. Miller, D.K. Johnson, and D.J. Clifford. 2004. Structure, Reactivity, and Nitrogen Evolution Characteristics of a Suite of Solid Fuels. Proc. 29th International Technical Conference on Coal Utilization and Fuel Systems. Coal Technology Assn. Clearwater, FL. pp. 563–574. 6. Stultz, S.C., and J.B. Kitto (eds). 1992. Steam: Its Generation and Use, 40th ed. Barberton, OH: Babcock & Wilcox. 7. Tillman, D.A. 2001. Final Report: EPRI-USDOE Cooperative Agreement: Cofiring Biomass with Coal. Clinton, NJ: Foster Wheeler. Contract No. DEFC22-96PC96252. 8. Tillman, D.A. 1999a. Biomass Cofiring: Field Test Results. Palo Alto, CA: Electric Power Research Institute. Report TR-113903. 9. U.S. Geological Survey. 2006. U.S. Coal Resource Databases (USCOAL). Washington, D.C. 10. Baxter, L.L. 2007. Advanced Chemical Analyses as Indicators for Coal Fouling and Slagging. Proc. 21st Annual ACERC Conference. Brigham Young University, Provo, UT. February 28. 11. Miller, S.F., and B.G. Miller. 2006. The Occurrence of Inorganic Elements in Various Biofuels and Its Effect on Ash Chemistry and Behavior and Use in Combustion Products. Proc. Fuel Quality Impacts Conference. Snowbird, UT. October 28–November 1. 12. Sliepcevich, C.M. et al. 1977. Assessment of Technology for the Liquefaction of Coal. Washington, D.C.: National Research Council, National Academy of Sciences. 13. Utility Data Institute. 1996. Power Plant Equipment Directory, 2nd ed. Washington, D.C.: Utility Data Institute (UDI 2055-96). 14. Miller, B.G. 2005. Coal Energy Systems. Burlington, MA: Elsevier–Academic Press. 15. Tillman, D.A. 1991. The Combustion of Solid Fuels and Wastes. San Diego, CA: Academic Press. 16. Palmer, H.B. 1974. Equilibria and Chemical Kinetics in Flames. In Combustion Technology: Some Modern Developments (Palmer, H.B., and J.M. Beer, (eds.). New York: Academic Press. pp. 1–33. 17. Bradbury, A.G.W., and F. Shafizadeh. 1980. Combustion and Flame. 37:85–89. 18. Bradbury, A.G.W., and F. Shafizadeh. 1980. Carbon. 18:109–116. 19. Mulcahy, M.F.R. and C.C. Young. 1975. Carbon. 13:115–124. 20. Tillman, D.A. 1994. Trace Metals in Combustion Systems. San Diego, CA: Academic Press. 21. Huffman, G.P., F.E. Huggins, N. Shah, and J. Zhao. 1993. Speciation of Critical Trace Elements in Coal and Combustion Ash by XAFS Spectroscopy. Proc. Trace Element Transformations in Coal-Fired Power systems. EPRI and USDOE. Scottsdale, AZ. April 19–22. 22. Baes, C.F., and S.B. McLaughlin. 1986. Trace Metal Uptake and Accumulation in Trees as Affected by Environmental Pollution. Oak Ridge, TN: Oak Ridge National Laboratory. Publication 2571.
32
Combustion Engineering Issues for Solid Fuel Systems
23. Clark, L.B., and L.L. Sloss. 1992. Trace Element Emissions from Coal Combustion and Gasification. IEA Coal Research. London: Gemini House. 24. Gordon, S., and B.J. McBride. 1976. Computer Program for Calculation of Complex Chemical Equilibrium Compositions, Rocket Performance, Incident and Reflected Shocks, and Chapman-Jouget Detonations. Interim Revision. NASA SP-273. NASA Lewis Research Center. March, 1976. Updated with Interim Revisions in 1989 and 1993. 25. Raask, E. 1985. Mineral Impurities in Coal Combustion: Behavior, Problems, and Remedial Measures. Washington, D.C.: Hemisphere Publishing Corporation. 26. Duong, D., D. Tillman, and B. Miller. 2006. Characterizing Blends of PRB and Central Appalachian Coals for Fuel Optimization Purposes. Proc. 31st International Technical Conference on Coal Utilization and Fuel Systems. Coal Technology Assn. Clearwater, FL. 27. Tillman, D., A. Dobrzanski, D. Duong, and P. Dezsi. 2006. Fuel Blending with PRB Coals for Combustion Optimization: A Tutorial. Proc. 31st International Technical Conference on Coal Utilization and Fuel Systems. Coal Technology Assn. Clearwater, FL. 28. Tillman, D., A. Dobrzanski, D. Duong, J. Dosch, K. Taylor, R. Kinnick, and T. Ellsworth. 2006. Optimizing Blends of Powder River Basin Subbituminous Coal and Bituminous Coal. Proc. PRB Users Group. Atlanta, GA.
CHAPTER
2
Coal Characteristics Bruce G. Miller
Associate Director, Energy Institute The Pennsylvania State University and
David A. Tillman Chief Engineer – Fuels and Combustion Foster Wheeler NA
Coal is the dominant solid fuel source worldwide. In this chapter, coal formation, coalification, and characteristics, with an emphasis on the organic/ combustible materials in coal, are discussed. (Note that a detailed discussion of the characteristics and behavior of the inorganic constituents is provided in Chapter 4.) Coal resources and reserves are discussed, with an emphasis on those located in North America and the United States. Coal petrography, traditional methods for analyzing coal, and evaluation of coal properties by nontraditional characterization methods (e.g., 13C nuclear magnetic resonance and drop-tube furnace characterization) are presented with an emphasis on how they are applied in industrial applications. Relationships between coal structure and behavioral characteristics are discussed.
2.1 Introduction to Coal Coal is a sedimentary rock composed of both organic and inorganic material. Coal is composed of macerals, discrete minerals, inorganic elements held molecularly by the organic matter, and water and gases contained in submicroscopic pores. Organically, coal consists primarily of carbon, hydrogen, and oxygen, and lesser amounts of sulfur and nitrogen. Inorganically, coal 33
34
Combustion Engineering Issues for Solid Fuel Systems
consists of a diverse range of ash-forming compounds distributed throughout the coal. The inorganic constituents can vary in concentrations from several percentage points to parts per billion of the coal. Coal is the most abundant fossil fuel in the United States as well as the world. At the end of 2004, recoverable coal reserves in the United States, which contains the largest reserves in the world, totaled 267 billion short tons compared to approximately 1 trillion short tons of recoverable coal reserves worldwide [1, 2]. From an energy standpoint, world coal reserves are approximately 2.5 times greater than those from estimated world oil reserves. As of January 1, 2006, there were 1,292.5 billion barrels of oil reserves in the world [2]. Using an average world oil density of 8.020 bbl oil/short ton and an average energy content of 5,884 103 Btu/bbl [3] results in 7,605 quadrillion (i.e., 1015) Btu of available world oil reserves. By comparison, coal, with reserves of 1,000.9 billion short tons [2] has 18,417 quadrillion Btu of world reserves when using an average coal heating value of 18,131 103 Btu/short ton [3]. The importance of coal as a major energy source today is apparent, and coal reserves/resources are discussed in more detail in a subsequent section.
2.1.1 Coal Formation and Coalification Coal is formed from the accumulation of vegetative debris that has undergone physical and chemical changes over millions of years. These changes include decaying of the vegetation, deposition, and burying by sedimentation, compaction, and transformation of the plant remains in the organic rock found today. Coal formation began during the Carboniferous Period, which is known as the first coal age [4], and major coal deposits formed in every geological period since the Upper Carboniferous Period, 270–350 million years ago, with the main coal-forming periods shown in Figure 2-1 [5]. Figure 2-1 shows the relative ages of the world’s major coal deposits. The considerable diversity of various coals is due to the differing climatic and botanical conditions that existed during the main coal-forming periods along with subsequent geophysical actions. Layers of plant debris were deposited in wet or swampy regions under conditions that limited exposure to air and complete decay as the debris accumulated, thereby resulting in the formation of peat. As the peat became buried by sediment, it was subjected to higher temperatures and pressures resulting in chemical and physical changes that, over time, formed coal. Cycles of plant debris accumulation and deposition were followed by diagenetic (i.e., biological) and tectonic (i.e., geological) actions, and, depending on the extent of temperature, time, and forces exerted, formed the different ranks of coal observed today. The geochemical process that transformed plant debris to coal is called coalification and is expressed as follows: Peat ! Lignite ! Subbituminous Coal ! Bituminous Coal ! Anthracite
Coal Characteristics Era Period
Palaeozoic Carboniferous
Permian
Mesozoic Jurassic
Triassic
Cretaceous
35
Cenozoic Tertiary Quaternary
Lignite EASTERN USA UNITED KINGDOM GERMANY POLAND/CZECH REPUBLIC CIS CHINA AUSTRALIA INDIA SOUTH AFRICA WESTERN CANADA WESTERN USA COLOMBIA/VENEZUELA INDONESIA Age
350
300
250
200
150
100
50
Million Years
FIGURE 2-1 Comparison of the geological ages of the world’s hard coal and lignite deposits (Source: [5]).
Coalification can be described geochemically as consisting of three processes: the microbiological degradation of the cellulose of the initial plant material; the conversion of the lignin of the plants into humic substances; and the condensation of these humic substances into larger coal molecules [6]. The kind of decaying vegetation, conditions of decay, depositional environment, and the movements of the earth’s crust are important factors in determining the nature, quality, and relative position of the coal seams [7]. Variations in the chemical composition of the original plant material contributed to the variability in coal composition [7, 8]. The vegetation of various geologic periods differed biologically and chemically. The process of coalification, impacting the organic constituents in coal, involves several sequential reactions from the base material—the plant life which forms the basis for coal. Highly reactive matter (e.g., some extractives, hemicelluloses) can undergo low-grade oxidation or volatilization processes in the presence of heat and pressure. This material can “leave” the fuel matrix. The lignins, which are the “glues” that hold the biomass material together, remain. The lignins contain aromatic structures as single units. Under pressure and temperature, lignins and other remaining biomass matter undergo condensation to form peat and then the various
36
Combustion Engineering Issues for Solid Fuel Systems
coals. The condensation processes produce coals with fused aromatic rings ranging in number from 2–6 depending on coal rank. Coalification also involves changes in functional groups attached to the fused aromatic structures, or coal backbones. As rank increases, the number and type of functional groups decreases. For example, methoxy (-OCH3) functional groups are commonly found in lignites and some subbituminous coals—and in significant concentrations. As rank increases, the concentration of such functionalities decreases; bituminous coals do not contain such functional groups. The characteristics of oxygen, identified by van Krevelen and Schuyer [9], is a key to understanding the chemical changes in coal as a function of rank, caused by increasing coalification processes. In Table 2-1, note the higher total concentration of oxygen in the lower ranked coals, and also the presence of oxygen in highly reactive functional groups (OCOOH, OOCH3, and OOH). As rank increases, indicated by carbon content, the concentration of oxygen in the coal decreases, and the presence of oxygen in the most reactive functionalities decreases to nearly zero. The conditions under which the vegetation decayed are important. The depth, temperature, degree of acidity, and natural movement of water in the original swamp are important factors in the formation of the coal [7, 10]. During coalification, chemically there is a decrease in moisture and volatile matter (i.e., methane, carbon dioxide) content, as well as an increase in the percentage of carbon, a gradual decrease in the percentage of oxygen, and ultimately, as the anthracitic stage is approached, a marked decrease in the content of hydrogen [8, 11]. For example, carbon content (on a dry, mineral-matter free basis) increases from approximately 50% in herbaceous plants and wood, to 60% in peat, 70% in lignite, 75% in subbituminous coal, 80–90% in bituminous coal, and >90% in anthracite [8, 12, 13, 14]. TABLE 2-1 Oxygen Content by Functional Group (%) Oxygen Content by Functional Group (wt %) Carbon Content (%) 65.5 70.5 75.5 81.5 85.5 87.0 88.6 90.3 Source: [9]
OCOOH
OOCH3
OOH
OC¼O
ONR
8.0 5.1 0.6 0.3 0.05 0.0 0.0 0.0
1.1 0.4 0.3 0.0 0.0 0.0 0.0 0.0
7.2 7.8 7.5 6.1 5.6 3.2 1.9 0.5
1.9 1.1 1.4 0.5 0.5 0.6 0.25 0.2
9.6 8.2 6.4 4.2 1.75 1.3 0.85 2.2
Coal Characteristics
37
During coalification, the presence and structure of heteroatoms also become significant. Nitrogen in the biomass typically exists in amine form; under coalification processes it becomes incorporated into ring structures and commonly exists as pyridine (6-membered ring accounting for 50–60% of the total nitrogen) or pyrrole (5-membered ring contributing 20–40% of the nitrogen). The remaining 0–20% nitrogen is in amine or quaternary forms, whose contribution increases with decreasing rank [15]. Only lignite exhibits some nitrogen in amine form. Sulfur, resulting largely from the environment, becomes incorporated into the coal matrix; often it is found in bridges between fused aromatic clusters. Chlorine, coming from the environment, may also become incorporated into the coal structure. The coalification process that governed the formation of eastern and midwestern United States coals is distinct and well understood. Coals coming from other countries and regions, however, have been produced with coalification processes that differ significantly from the processes governing North American bituminous coals. Particularly coals formed in the Pacific regions known as the “Ring of Fire” are influenced by the fact that volcanic activity supplied both heat and inorganic matter to the coalification process.
2.2 Coal Classification Coals differ throughout the world in the degree of metamorphism or coalification (rank of coal), in the kinds of plant materials deposited (type of coal), and in the range of impurities included (grade of coal). In this section, coal rank, type, and grade are discussed along with the U.S. classification system by rank.
2.2.1 Coal Rank The degree of coal maturation is known as rank of coal and is an indication of the extent of metamorphism the coal has undergone. Heat and pressure converted the organic material (accumulated plant debris) progressively to a substance that more resembles graphite. Coals are ranked according to how severely they were metamorphosed. Because vitrinite precursors (see Section 2.2.2 for a discussion on maceral types), such as humates and humic acids, were the major constituents in peat, the extent of metamorphism is often determined by noting the changes in the properties of vitrinite [11]. Some of the more important properties of vitrinite that did change with metamorphism are listed in Table 2-2 [11]. Metamorphism did affect the other macerals, but the relationship between the severity of metamorphism and magnitude of change are different from those of vitrinite. Several useful rank-defining properties are elemental carbon content, volatile matter content, moisture-holding capacity, heating value, and microscopic reflectance of vitrinite [11]. Figure 2-2 illustrates the relationship between rank and fixed carbon content [16], the one chemical property
38
Combustion Engineering Issues for Solid Fuel Systems
TABLE 2-2 Properties of Vitrinite Affected Progressively by Metamorphism Increase with Increasing Rank Reflectance (microscopic) and optical anisotropy Carbon content Aromaticity, fa ¼ Caromatic/Ctotal Condensed-ring fusion Parallelization of molecular moieties Heating value (a slight decrease at very high rank) Decrease with Increasing Rank Volatile matter (especially oxygenated compounds) Oxygen content (especially as functional groups) Oxidizability Solubility (especially in aqueous alkalis and polar hydrocarbons) Increase Initially to a Maximum, then Decrease Hardness (minor increase at very high rank) Plastic properties Hydrogen content Decrease Initially to a Minimum, then Increase Surface area Porosity (and moisture-holding capacity) Density (in helium) Source: [11]
most used to express coal rank. Figure 2-2 also shows the comparison between heating value, another property widely used as a measure of rank. Note that the heating value increases with rank but begins to decrease with semianthracite and higher rank coals due to the significant decrease in volatile matter. Porosity decreases with increasing level of metamorphism, thereby reducing the moisture-holding capacity of the coal. Moisture-holding capacity is also affected by the functional group characteristics, and in coals where cationic elements replace protons on acid functional groups, moisture-holding capacity is reduced [11]. Vitrinite reflectance is another import rank-measuring parameter. The advantage of this technique, discussed in Section 2.4, is that it measures a rank-sensitive property on only one petrographic constituent; therefore, it is applicable even where the coal type is atypical [11].
2.2.2 Coal Type As mentioned previously, coal is composed of macerals, discrete minerals, inorganic elements held molecularly by the organic matter, and water and gases contained in submicroscopic pores. Macerals are organic substances derived from plant tissues that have been incorporated into the sedimentary strata, subjected to decay, compacted, and chemically altered by
Coal Characteristics
39
16,000
2000
A
E
D
C
A
Meta-anthracite
E
R
B
Meta-anthracite
Anthracite
Semianthracite
E
T
R R
Low-volatie bituminous
M
A T
U
Medium-volatie bituminous
E
T
High-volatile A bituminous
Lignite B
X
L
High-volatile C bituminous
I
A T
L
Subbituminous A
Subbituminous C
F
40
20
O
Lignite A
V
Subbituminous B
60
I
S
High-volatile B bituminous
M
80
Percent
Anthracite
0 100
B
Semianthracite
I
Low-volatile bituminous
Lignite B Lignite A
6000 4000
Subbituminous B
Subbituminous C
Btu/Ib
8000
High-volatile A bituminous
High-volatile C bituminous
O
10,000
High-volatile B bituminous
Subbituminous A
12,000
Medium-volatile bituminous
14,000
O
N
FIGURE 2-2 (A) Comparison of heating values (on a moist, mineral matter-free basis) and (B) proximate analyses of coals of different ranks (Source: [16]).
geological processes. This organic matter is extremely heterogeneous, and a classification system has been developed to characterize it [11, 17, 18]. Classifying the coal, known as petrography, was primarily used to characterize and correlate coal seams and resolve questions about coal diagenesis and metamorphism, but it also has a role in coal utilization (see Section 2.4). All macerals are classified into three maceral groups—vitrinite, liptinite (sometimes also referred to as exinite), and inertinite—and they are characterized by their appearance, chemical composition, and optical properties. Each maceral group includes a number of macerals and other subcategories; however, only the three maceral groups are introduced here because extensive discussions of petrography can be found elsewhere [11, 12, 17, 18]. In most cases, the constituents in the coal can be traced back to specific components of the plant debris from which the coal formed [11, 12, 17, 18]. This is illustrated in Figure 2-3, which is a simplified overview of
40
Combustion Engineering Issues for Solid Fuel Systems Source Material
Peat Swamp
Coal Components
Humified “Decomposed”
Vitrinite
“Charred”
Fusinite
Waxy Exines
Incorporated
Exinite
Resins
Incorporated
Resinite
“Wood”
Micrinite
“Chelated” Inorganic Ions Mineral Grains
Precipitated
Minerals
Incorporated
Minerals
FIGURE 2-3 Plant materials that accumulated along with inorganic materials in the peat swamp retain their identity as distinctive macerals in the coal (modified from [11]).
the various substances that accumulated as peat deposited and the components they represent in coal [11]. Adding to the complexity of the source materials were inorganic substances (also shown in Figure 2-3) that entered the environment as mineral grains and dissolved ions. Many of the dissolved ions either combined with the organic fraction or were precipitated in place to form discrete mineral grains [11]. Vitrinite group macerals are coalification products of humic substances originating from woody tissues and can either possess remnant cell structures or be structureless [9, 17]. Vitrinites contain more oxygen than the other macerals at any given rank level, and are characterized by a higher aromatic fraction. Liptinite group macerals are not derived from humifiable materials but rather from relatively hydrogen-rich plant remains such as resins, spores, cuticles, waxes, fats, and algal remains, which are fairly resistant to bacterial and fungal decay [9, 17]. Liptinites are distinguishable by a higher aliphatic (i.e., paraffin) fraction and a correspondingly higher hydrogen content, especially at lower rank [17]. The inertinite group macerals were derived mostly from woody tissues, plant degradation products, or fungal remains. While they were derived from the same original plant substances as vitrinite and liptinite, they have experienced a different primary transformation [17]. Inertinite group macerals are characterized by a high carbon content that resulted from thermal or biological oxidation, as well as low hydrogen content and an increased level of aromatization [9, 17].
Coal Characteristics
41
2.2.3 Coal Grade The grade of a coal establishes its economic value for a specific end use. Grade of coal refers to the amount of mineral matter that is present in the coal and is a measure of coal quality. Sulfur content; ash fusion temperatures, i.e., measurement of the behavior of ash at high temperatures; and quantity of trace elements in coal are also used to grade coal. Although formal classification systems have not been developed around grade of coal, grade is important to the coal user.
2.2.4 Coal Classification Since the rank of the coal is most important for the coal industry, almost every coal-producing country has its own economic coal classification, which is based mainly on rank parameters [17]. An excellent discussion of the many classification systems, scientific as well as commercial, is provided by van Krevelen [7]. There are two primary commercial classification systems in use—the American Society for Testing and Materials (ASTM) system used in the United States/North America and an international Economic Commission for Europe (ECE) Codification system developed in Europe. The classification systems used commercially are primarily based on the content of volatile matter [7]. In some countries, a second parameter is also used, and in the United States, for example, this is the heating value (see Figure 2-2). For many European countries, this parameter is either the caking or coking properties. 2.2.4.1 ASTM Classification System The ASTM classification system (ASTM D388) distinguishes between four coal classes, each of which is subdivided into several groups and is shown in Table 2-3. High-rank coals, i.e., medium volatile bituminous coals or those of higher rank, are classified based on their fixed carbon and volatile matter contents, expressed on a dry, mineral matter-free (dmmf) basis, whereas low-rank coals are classified in terms of their heating value (expressed on a moist, mineral matter-free [mmmf] basis). 2.2.4.2 International Classification System In 1998, the ECE Coal Committee developed a new classification system for higher rank coals [7]. This classification system, which in reality is a system of codes, is better known as a Codification System. The Codification System for hard coals, combined with the International Organization for Standardization (ISO) Codification of Brown Coals and Lignites, provides a complete codification for coals in the international trade. The ISO Codification of Brown Coals and Lignites is given in Table 2-4 [7]. Total moisture content of run-of-mine coal and tar yield are the two parameters coded.
42
Combustion Engineering Issues for Solid Fuel Systems
TABLE 2-3 ASTM Coal Classification by Rank Class/Group
Fixed Carbona (%)
Volatile Matterb (%)
>98 92–98 86–92
<2 2–8 8–14
78–86 69–78 <69
14–22 22–31 >31
Anthracitic Metaanthracite Anthracite Semianthracite Bituminous Low volatile Medium volatile High volatile A High volatile B High volatile C Subbituminous Subbituminous A Subbituminous B Subbituminous C Lignitic Lignite A Lignite B
Heating Valueb (Btu/lb)
>14,000 13,000–14,000 10,500–13,000c 10,500–11,500c 9,500–10,500 8,300–9,500 6,300–8,300 <6,300
a Calculated on dry, mineral matter-free coal; correction from ash to mineral matter is made by means of the Parr formula: mineral matter ¼ 1.08[percent ash þ 0.55(percent sulfur)]. Ash and sulfur are on a dry basis. b Calculated on mineral matter-free coal with bed moisture content. c Coals with heating values between 10,500 and 11,500 Btu/lb are classified as high volatile C bituminous if they possess caking properties or as subbituminous A if they do not. Source: [12]
TABLE 2-4 Codification of Brown Coals and Lignites Parameter:
Total Moisture Content (run-of-mine coal)
Tar Yield (dry, ash free)
1
2
Digit: Coding:
Code 1 2 3 4 5 6
Weight % 20 >20–30 >30–40 >40–50 >50–60 >60
Code
Weight %
0 1
10 >10–15
2 3 4
>15–20 >20–25 >25
Source: [7]
The ECE International Codification of Higher Rank Coals is much more complicated and is listed in Table 2-5. Eight basic parameters define the main properties of the coal, which are represented by a 14-digit code number. The codification is commercial; includes petrographic, rank, grade, and environmental information; is for medium- and high-rank coals
TABLE 2-5 International Codification of Higher Rank Coalsa Parameter:
Maceral Group Composition (mmf) Vitrinite Reflectance (mean random)
Characteristics of Reflectogramb
Inertinitec
Liptinite
1, 2
3
4
5
Digit: Coding:
Parameter:
Code
Rrandom%
Code
02 03 04 — 48 49 50
0.2—0.29 0.3—0.39 0.4—0.49 — 4.8—4.89 4.9—4.99 5.0
0 1 2 3 4 5 —
a
1 >0.10.2 >0.2
—
no gap no gap no gap 1 gap 2 gaps >2 gaps —
Type
Code
Vol. %
Code
Vol. %
Seam coal Simple blend Complex blend Blend with 1 gap Blend with 2 gaps Blend with >2 gaps —
0 1 2 — 7 8 9
0—<10 10—<20 20—<30 — 70—<80 80—<90 90
1 2 3 — 7 8 9
0 to <5 5 to <10 10 to <15 — 30 to <35 35 to <40 40
Crucible Swelling No.
Volatile Matter,d daf
Ash, dry
Total Sulfur, dry
Gross Calorific Value, daf
6
7, 8
9, 10
11, 12
13, 14
Digit: Coding:
Standard deviation
Code
Number
Code
Wt.%
Code
Wt.%
Code
Wt.%
Code
MJ/kg
0 1 2 — 7 8 9 — — —
0—0.5 1—1.5 2—2.5 — 7—7.5 8—8.5 9—9.5 — — —
48 46 44 — 12 10 09 — 02 01
48 46 to <48 44 to <46 — 12 to <14 10 to <12 9 to <10 — 2 to <3 1 to <2
00 01 02 — 20 — — — — —
0 to <1 1 to <2 2 to <3 — 20 to <21 — — — — —
00 01 02 — 29 30 — — — —
0—<0.1 0.1—<0.2 0.2—<0.3 — 2.9—<3.0 3.0—<3.1 — — — —
21 22 23 — 37 38 39 — — —
<22 22 to <23 23 to <24 — 37 to <38 38 to <39 39 — — —
Petrographic Tests
Technological Tests
43
Higher rank coals are coals with gross calorific value (maf) 24 MJ/kg and those with gross calorific value (maf) <24 MJ/kg provided mean random vitrinitic reflectance 0.6%. To convert from MJ/kg to Btu/lb, multiply by 429.23. b A reflectogram as characterized by code number 2 can also result from a high-rank seam coal. c It should be noted that some of the inertinite may be reactive. d Where the ash content of the coal is more than 10%, it must be reduced, before analysis, to below 10% by dense medium separation. In these cases, the cutting density and resulting ash content should be noted. Source: [7]
44
Combustion Engineering Issues for Solid Fuel Systems
only; is for blends and single coals; is for raw and washed coals; and is for all end-use applications [7]. The major drawback of this system is that it is complicated.
2.3 Coal Reserves/Resources Coal deposits are broadly categorized into resources and reserves. Resources refer to the quantity of coal that may be present in a deposit or coalfield but may not take into account the feasibility of mining the coal economically. Reserves generally tend to be classified as proven or measured and probable or indicated, depending on the level of exploration of the coalfield. The basis for computing resources and reserves varies between countries, which makes it difficult because direct comparisons and totals, for the same year, can vary between different government agencies. The definitions typically used when assessing coal resources and reserves are: Total resources—Coal that can currently, or potentially may, be
extracted economically; Measured resources—Quantity of coal that has been determined to
a high degree of geologic assurance; Indicated resources—Quantity of coal that has been determined to a
moderate degree of geological assurance; Inferred resources—Quantity of coal that has been determined with
a low degree of geologic assurance; and Recoverable reserves—Coal that can be recovered economically
with technology currently available or in the foreseeable future. This section discusses the recoverable reserves of coal worldwide with an emphasis on the United States. A detailed discussion of the major coal regions can be found elsewhere [19].
2.3.1 World Coal Reserves Coal is available in almost every country worldwide, with recoverable reserves in around 70 countries [19]. Total recoverable reserves of coal around the world are estimated at 1,001 billion short tons [2]. At current production levels, these reserves will last approximately 180 years. However, this could be extended still further by the discovery of new reserves through ongoing and improved exploration activities, and by advances in mining techniques, which will allow previously inaccessible reserves to be reached. Historically, estimates of world recoverable coal reserves have declined from 1,174 billion short tons in 1990 to 1,001 billion short tons in 2003 [2]. The most recent assessment of world coal reserves includes a
Coal Characteristics
45
significant downward adjustment for Germany, from 73 billion short tons of recoverable coal reserves to 7 billion short tons. The reassessment reflects more restrictive criteria for various parameters. This downward trend is also observed for other countries as well. Coal reserves are nearly double that of oil and natural gas combined. Proven oil and gas reserves are equivalent to about 41 and 67 years, respectively [20]. Coal reserves are also more widely distributed throughout the world, as shown in Figure 2-4. All major regions of the world contain coal, except for the Middle East. Over 68% of oil and 67% of natural gas reserves are concentrated in the Middle East and Russia [20]. A detailed breakdown of EIA’s estimated recoverable world coal reserves of 1,001 billion short tons is provided in Table 2-6 [21]. Table 2-6 classifies the recoverable coal reserves in two major categories—recoverable anthracite and bituminous coal (i.e., hard coal) and recoverable lignite and subbituminous coal—for the major regions and countries of the world. Although coal deposits are widely distributed, 67% of the world’s recoverable reserves are located in four countries: the United States (27%; 271 billion short tons), Russia (17%; 173 billion short tons), China (13%; 126 billion short tons), and India (10%; 102 billion short tons). Figure 2-5 shows the 10 countries with the largest recoverable coal reserves. Approximately 70 countries contain recoverable coal; however, the 10 shown in Figure 2-5 contain about 918 billion short tons, or more than 90% of the world’s total.
Africa 55,486 million short tons North America 270,506 million short tons
Asia,Oceania, and Middle East 322,294 million short tons
Central and South America 21,929 million short tons Western Europe 36,489 million short tons
Eastern Europe and Former U.S.S.R. 279,778 million short tons
FIGURE 2-4 Distribution of recoverable coal reserves in the world.
46
Combustion Engineering Issues for Solid Fuel Systems
TABLE 2-6 World Estimated Recoverable Coal Reserves (million short tons) Region/Country
Anthracite and Bituminous
Lignite and Subbituminous
North America Canada Greenland Mexico United States Total
3,826 0 948 125,412 130,186
3,425 202 387 145,306 149,320
7,251 202 1,335 270,718 270,506
Central & South America Brazil Chile Colombia Peru Other Total
0 34 6,867 1,058 529 8,489
11,148 1,268 420 110 494 13,439
11,148 1,302 7,287 1,168 1,023 21,928
Western Europe Germany Greece Turkey United Kingdom Yugoslavia Other Total
202 0 306 243 10 810 1,571
7,227 4,299 4,308 0 18,279 806 34,918
7,428 4,299 4,614 1,653 18,288 243 36,489
4 2,308 218 31,031 15,432 24 54,110 17,939 1,102 0 122,1706
2,406 3,812 3,482 3,448 0 520 118,964 19,708 3,307 1,960 157,607
2,411 6,120 3,700 34,479 15,432 545 173,074 37,647 4,409 1,960 279,778
Africa Botswana South Africa Zimbabwe Other Total
44 53,738 553 959 55,294
0 0 0 192 192
44 53,738 553 1,151 55,486
Middle East, Asia, & Oceania Australia China India Indonesia
42,549 68,564 90,302 816
43,982 57,651 2,601 4,661
86,531 126,215 101,903 5,476
Eastern Europe & Former U.S.S.R. Bulgaria Czech Republic Hungary Kazakhstan Poland Romania Russia Ukraine Uzbekistan Other Total
Total
(continued)
Coal Characteristics
47
TABLE 2-6 World Estimated Recoverable Coal Reserves (million short tons) (continued) Region/Country
Anthracite and Bituminous
Lignite and Subbituminous
Total
331 0 0 1,166 212,727 530,438
331 3,362 1,493 918 114,999 470.475
661 3,362 1,4938 2,084 322,394 1,000,912
Korea, North Pakistan Thailand Other Total World Total Source: [21]
United States Russia China
Anthracite/Bituminous Coal
India
Subbituminous Coal/Lignite
Australia South Africa Ukraine Kazakhstan Yugoslavia Poland 0
50
100
150 200 Billion Short Tons
250
300
FIGURE 2-5 Top 10 countries with the largest recoverable coal reserves.
2.3.2 United States Coal Resources and Reserves Figure 2-6 illustrates the coal resources (as of 1997) and reserves (as of 2004) in the United States. The United States has a total of nearly 4,000 billion short tons of coal resources with approximately 18 billion short tons classified as recoverable reserves at active mines out of 271 billion short tons that are economically recoverable. The coal reserves of the United States are the largest of any country in the world, with about 271 billion short tons as of January 1, 2004 [2]. Recoverable coal reserves are found in 32 of the states, with the major coalfields
48
Combustion Engineering Issues for Solid Fuel Systems
Recoverable Reserves at Active Mines (18.1)
270.7 494.5
Estimated Recoverable Reserves Demonstrated Reserve Base (Measured and Indicated, Specified Depths and Thickness) Identified Resources (Measured, Indicated, and Inferred)
1,730.9 Total Resources (Identified and Undiscovered)
3,968.3
FIGURE 2-6 United States coal resources and reserves in billion short tons (Source: [21], [22]).
shown in Figure 2-7 [16]. United States coal reserves, by state and type, are listed in Table 2-7. The 10 states with the largest recoverable coal reserves are listed in Table 2-8, and they contain approximately 89% of the total coal in the United States [1]. The top five states contain more than 70% of the total recoverable coal reserves in the United States. Estimated low sulfur recoverable coal reserves make up the largest portion of the total, at 36% [22]. Low-sulfur coal is defined as less than 0.8% and 0.5% by weight (as received) sulfur for high-grade bituminous coal and high-grade lignite, respectively. These sulfur contents are a quantitative rating and have been correlated with United States sulfur emissions regulations from coal-fired power plants and the various stages of control that are required [22]. Estimated medium- (0.8–2.2% for bituminous coal and 0.5–1.3% for lignite) and high- (>2.2 and >1.3% for bituminous coal and lignite, respectively) sulfur recoverable reserves account for 31 and 33% of the total, respectively.
2.4 Coal Production This section discusses worldwide coal production with an emphasis on the major coal-producing countries. Specific emphasis is given to coal production in the United States.
49
Coal Characteristics
WA
NORTHERN GREAT PLAINS REGION
MT
ME
ND
MI
CA
BIGHORN BASIN
OR
VT
NY
NH
SD POWDER RIVER BASIN
W JD GREE N RIVE BASIN R UT
MN IA
MI IN OH ILLINOIS BASIN
DENVER BASIN CO
SAN JUAN BASIN
AZ
PA MD
MO TN
AR
NJ DE
WV
VA NC
AP
TX RATON MESA FIELD
WESTERN INTERIOR BASIN
PAL AC HIA N
KS
BLACK MESA FIELD
RI CT
WI
NE
WY
UNITA BASIN
MA
BA SIN
COO S FIEL BAY D
AL
SC
GA
100
NM
0
100
200
SCALE IN MILES
MS LA
FL
AK
RANK Anthracite
200
0
200
FIELD
SMALL FIELD OR ISOLATED OCCURRENCE A
Bituminous Coal
B
Subbituminous Coal
S
Lignite
L
400
SCALE IN MILES
FIGURE 2-7 Major coal-bearing areas of the United States (Source: [16]). TABLE 2-7 United States Coal Reserves (million short tons) by State and Type as of January 1, 2004 State Eastern U.S. Alabama Georgia Illinois Indiana Kentucky Eastern Western Maryland Michigan Mississippi North Carolina Ohio Pennsylvania Bituminous Anthracite
Reserves at Active Mines
Estimated Recoverable Reserves
Demonstrated Reserve Base
341 0 796 398 1,129 823 306 17 0 Wa 0 318 614 592 22
2,806 2 38,019 4,080 15,004 5,960 9,044 366 59 0 5 11,507 11,822 11,062 760
4,242 4 104,529 9,534 30,225 10,671 19,554 652 128 0 11 23,342 27,597 20,397 7,200 (continued)
TABLE 2-7 United States Coal Reserves (million short tons) by State and Type as of January 1, 2004 (continued) Reserves at Active Mines
Estimated Recoverable Reserves
Demonstrated Reserve Base
Tennessee Virginia West Virginia Total Eastern U.S.
26 250 1,518 5,407
462 1,022 18,104 103,258
779 1,740 33,230 236,003
Western U.S. Alaska Arizona Arkansas Colorado Idaho Iowa Kansas Louisiana Missouri Montana New Mexico North Dakota Texas Utah Washington Wyoming Total Western U.S.
Wa W 0 415 0 0 W W 0 1,140 1,304 1,191 546 317 0 7,053 11,983
3,291 5 228 9,798 2 1,127 681 316 3,847 74,989 6,934 6,935 9,578 2,750 681 41,804 164,053
6,112 7 417 16,293 4 2,189 973 427 5,990 119,280 12,172 9,090 12,442 5,445 1,341 64,325 258,447
Grand Total U.S.
18,122
267,311
494,450
State
a Withheld to avoid disclosure of individual company data Source: [1]
TABLE 2-8 Top 10 States with the Largest Estimated Recoverable Coal Reserves as of January 1, 2004 State Montana Wyoming Illinois West Virginia Kentucky Pennsylvania Ohio Colorado Texas North Dakota Total Percentage of U.S. Total U.S. Total Source: [1]
50
Reserves (million short tons) 74,989 41,804 38,019 18,104 15,004 11,822 11,507 9,798 9,578 6,935 237,560 88.9% 267,311
Coal Characteristics
51
2.4.1 World Coal Production World coal production, in 2004, totaled approximately 392.8 million short tons anthracite, 974.1 million short tons lignite, and 4,648.2 million short tons bituminous coal [4]. Tables 2-9 through 2-11 list the top 10 producing countries for each coal type, respectively. Note that subbituminous coal is included with the bituminous coal production because, internationally, subbituminous coal is often considered a hard coal and lignite is considered a soft, or brown coal. Tables 2-12 through 2-14 contain characteristic coal analyses for selected international anthracite, lignite/brown coals,
TABLE 2-9 World Anthracite Production Country China Vietnam Ukraine Russia Spain South Korea Germany United Kingdom United States South Africa Others (France, Poland, Peru) Total Global Production
Production (million short tons) 392.8 18.1 15.1 14.1 4.3 3.5 2.9 2.1 1.7 1.4 0.3 456.3
Source: [4]
TABLE 2-10 World Lignite Production Country Germany China United States Greece Russia Australia Poland Turkey Serbia & Montenegro Romania Total Global Production Source: [4]
Production (million short tons) 200.5 87.3 83.5 79.5 77.5 76.6 67.5 48.2 44.7 34.8 974.2
52
Combustion Engineering Issues for Solid Fuel Systems TABLE 2-11 World Bituminous Coal Production Country
Production (million short tons)
China United States India Australia South Africa Russia Indonesia Poland Kazakhstan Czech Republic Total Global Production
1,676.4 1,026.9 411.4 314.3 266.3 217.3 142.3 110.0 91.4 67.6 4,648.2
Source: [4]
and bituminous coals, respectively. These analyses are from several of the top 10 producing countries with the exception of the United States’ coals. Analyses of coals from the United States are provided in Section 2.4.2. In 2004, China produced over 86% of the anthracite worldwide. Vietnam, Ukraine, and Russia combined contributed another 10% of the global total, with nine other countries producing the remaining 4%. Lignite was produced by 32 countries in 2004 with the top 10 countries listed in Table 2-10 producing over 82% of the lignite worldwide. Germany was the leading lignite producer with over 200 million short tons produced, which is over 20% of the global production. Bituminous (and subbituminous) coal was produced in 53 countries in 2004. The top 10 countries listed in Table 2-11 produced 93% of the hard coal worldwide, with two countries, China and the United States, producing over 58% of the global hard coal.
2.4.2 United States Coal Production Enormous quantities of coal are mined in the United States, and 1.13 billion short tons of coal were mined in 2005 [23]. Coal flow in the United States for 2004 is summarized in Figure 2-8 [21]. The trend in United States coal production for the last 115 years is illustrated in Figure 2-9. Coal production by region and state is listed in Table 2-15 for 2005 [23]. Coal production has shifted from mainly underground mines to surface mines, as illustrated in Figure 2-10. Also, coal resources west of the Mississippi River, especially those in Wyoming, underwent tremendous development (see Figure 2-11). Coal was produced in 26 states in 2005, with 15 of the states located in the western United States.
TABLE 2-12 Properties of Selected International Samples of Anthracite [Modified from 13] Rank
53
Proximate Analysis (%, as-received) Ash Moisture Volatile Matter Fixed Carbon Ultimate Analysis (%, moisture, ash free) Carbon Hydrogen Oxygen Nitrogen Sulfur Higher Heating Value (Btu/lb) Ash Fusion Temperatures (reducing conditions, oF) Initial Deformation Softening Hemispherical Fluid Ash Composition (%) SiO2 Al2O3 Fe2O3 CaO MgO Na2O K2O TiO2 P2O5 SO3 Hardgrove Grindability Index a
Not available
Germany Anthracite
Spain Anthracite
Russia Anthracite
Russia Semi-Anthracite
China Anthracite
China Anthracite
Korea Anthracite
4.0 3.0 7.2 85.8
31.9 3.1 7.1 57.9
11.3 2.0 3.0 83.7
13.4 4.0 9.9 72.7
15.0 2.3 7.6 75.1
9.9 1.1 8.5 80.5
31.5 3.4 4.8 60.3
91.8 3.6 2.6 1.4 0.7 15,440
87.2 2.5 8.0 0.9 1.4 14,090
94.6 1.8 1.8 1.0 0.8 14,590
90.0 4.2 2.1 1.5 2.2 15,380
– – – – – 15,360
– – – – – 14,790
12,980
–a – – –
2,160 2,420 N/A 2,640
– – – –
– – – –
– – – –
– – – –
– 2,220 – –
– – – – – – – – – – –
59.0 20.5 7.5 3.5 1.9 0.9 4.0 1.3 N/A 2.0 54
– – – – – – – – – – –
– – – – – – – – – – –
– – – – – – – – – – –
– – – – – – – – – – –
65.0 20.1 5.3 1.0 1.0 0.5 5.0 0.9 N/A N/A 59
54
Combustion Engineering Issues for Solid Fuel Systems
TABLE 2-13 Properties of Selected International Samples of Lignite/Brown Coal [Modified from 13] Rank
Germany Brown
Germany Lignite
Proximate Analysis (%, as-received) Ash 14.0 16.0 Moisture 56.0 10.0 Volatile Matter 16.5 38.5 Fixed Carbon 13.5 35.5 Ultimate Analysis (%, moisture, ash free) Carbon 68.3 74.0 Hydrogen 5.0 5.5 Oxygen 27.5 14.5 Nitrogen 0.5 1.4 Sulfur 0.5 4.6 Higher Heating 11,340 12,780 Value (Btu/lb) Ash Fusion Temps. (reducing conditions, oF) Initial Deformation –b – Softening – – Hemispherical – – Fluid – – Ash Composition (%) SiO2 – – Al2O3 – – Fe2O3 – – CaO – – MgO – – Na2O – – K2O – – TiO2 – – – – P2O5 SO3 – – Hardgrove – – Grindability Index
Greece Lignite
F.Y.a Brown
Turkey Brown
Poland Lignite
Russia Lignite
15.0 62.0 14.3 8.7
17.8 49.7 16.9 15.6
16.0 55.0 19.4 9.6
7.3 53.9 23.6 15.2
6.9 37.0 24.5 31.6
60.5 6.2 30.6 1.3 1.4 10,510
61.9 4.3 29.8 0.9 3.1 9,450
61.4 5.1 29.6 0.8 5.1 10,190
71.4 5.8 20.2 0.6 2.0 12,510
72.2 4.3 22.1 1.1 0.3 11,190
– – – –
2,700þ 2,700þ N/A 2,700þ
– – – –
– – – –
– – – –
– – – – – – – – – – –
21.6 3.1 6.0 45.6 2.5 0.4 0.2 0.1 0.4 15.1 58c
– – – – – – – – – – –
– – – – – – – – – – –
– – – – – – – – – – –
a
Former Yugoslavia Not available c At 36% moisture b
The shift toward surface-mined coal, especially west of the Mississippi River, came about because of the technological improvements in mining, and thick coal seams are located near the surface. Coal in the eastern United States generally occurs in seams that tend to be less than 15 feet thick. Thicker coalbeds are common in the western United States, particularly Wyoming, where coal seams average 65 feet. Coal seams that are more than 200 feet under the surface are mined by underground methods. Most underground mines are less than 1,000 feet deep; although several reach
Coal Characteristics
55
depths of about 2,000 feet [24]. The largest coal-producing western mines are surface mines. The most important coal deposits in the eastern United States are in the Appalachian Region, an area that encompasses more than 72,000 square miles and parts of nine states (see Figures 2-7 and 2-12). This region contains large deposits of low- and medium-volatile bituminous coal and the principal deposits of anthracite. Historically, this region has been the TABLE 2-14 Properties of Selected International Samples of Bituminous Coal [Modified from 13]
Rank
Poland
Russia
Russia
Russia
Russia
China
China
hvA
Low Vol.
hvA
hvB
hvC
Low Vol.
Med. Vol.
Proximate Analysis (%, as-received) Ash 11.5 16.0 19.0 Moisture 8.8 3.0 5.0 Volatile Matter 29.6 13.0 24.3 Fixed Carbon 50.1 68.0 54.7 Ultimate Analysis (%, moisture, ash free) Carbon 82.9 88.0 83.0 Hydrogen 5.2 4.5 5.1 Oxygen 9.9 2.9 5.6 Nitrogen 1.0 1.5 1.5 Sulfur 1.0 3.1 4.8 Higher Heating 14,550 15,400 15,140 Value (Btu/lb) Ash Fusion Temps. (reducing conditions, F) Initial 2,160 – – Deformation Softening 2,210 – – Hemispherical –a – – Fluid 2,460 – – Ash Composition (%) SiO2 46.8 – – Al2O3 21.8 – – Fe2O3 9.6 – – CaO 5.8 – – MgO 3.5 – – Na2O 0.8 – – K2O 3.1 – – TiO2 0.7 – – 0.3 – – P2O5 SO3 6.6 – – Hardgrove 49 – – Grindability Index
14.7 8.0 30.2 47.1
19.8 12.0 30.0 38.2
13.2 1.2 18.4 67.2
12.7 0.9 26.7 59.7
82.0 5.5 8.5 1.5 2.5 14,600
77.0 5.6 12.3 1.6 3.5 13,900
– – – – – 15,940
– – – – – 15,530
–
–
–
–
– – –
– – –
– – –
– – –
– – – – – – – – – – –
– – – – – – – – – – –
– – – – – – – – – – –
– – – – – – – – – – –
(continued)
56
Combustion Engineering Issues for Solid Fuel Systems
TABLE 2-14 Properties of Selected International Samples of Bituminous Coal [Modified from 13] (continued)
Rank
China
China
India
India
S. Africa
Australia
Australia
hvA
hvB
Med. Vol.
hvB
Med. Vol.
hvA
hvB
Proximate Analysis (%, as-received) Ash 16.6 10.2 38.9 Moisture 1.2 3.1 1.1 Volatile 36.3 41.2 15.2 Matter Fixed Carbon 45.9 45.5 44.8 Ultimate Analysis (%, moisture, ash free) Carbon –a – 83.6 Hydrogen – – 4.5 Oxygen – – 9.9 Nitrogen – – 1.3 Sulfur – – 0.7 Higher 15,530 13,610 14,635 Heating Value (Btu/lb) Ash Fusion Temps. (reducing conditions, oF) Initial – – 2,250 Deformation Softening – – 2,670 Hemispherical – – – Fluid – – 2,700þ Ash Composition (%) SiO2 – – 65.9 Al2O3 – – 23.7 – – 6.0 Fe2O3 CaO – – 1.1 MgO – – 0.6 Na2O – – 0.1 – – 1.4 K2O TiO2 – – 2.2 P2O5 – – – – – 0.3 SO3 – – 63 Hardgrove Grindability Index
31.6 6.9 22.9
13.9 2.9 28.1
16.5 4.1 35.5
26.0 7.3 31.0
38.6
55.1
43.9
35.7
– – – – 1.8 13,500
84.6 4.9 7.2 2.2 1.1 14,070
81.2 6.1 11.0 1.1 0.6 14,630
77.4 5.5 15.3 1.2 0.6 13,490
2,600
2,410
2,280
2,280
2,700þ – 2,700þ
2,430 – 2,700
2,500 – 2,700þ
2,480 – 2,700þ
60.8 24.8 6.8 2.1 0.5 0.1 1.0 – – 1.7 60
41.5 30.8 4.8 8.7 2.2 0.2 0.3 1.0 1.0 7.6 46
56.1 25.9 3.8 5.0 2.1 0.6 0.7 1.1 0.5 3.2 48
54.5 24.0 6.7 3.7 2.3 0.3 1.0 1.8 0.2 3.3 45
a
Not available
major source of United States coal, with approximately 75% of the total annual production as recently as 1970 [25]. Now the region produces less than 50% of the United States’ total, with 396 million short tons mined in 2005, with the reduction due to the increased coal production in the western United States (see Figures 2-12 and 2-13). This region however, still
Coal Characteristics Residential Exports 0.5 48.0
57
Commercial 3.8 Industrial 84.9
Surface 744.0
Bituminous Coal 546.6
Production 1,111.5 Subbituminous Coal 479.6 Underground 367.5
Consumption 1,104.3
Electric power 1,015.1
Lignite 83.5
Anthracite 1.7 Imports 27.3
Stock Change 18.2
Losses and Unaccounted for 4.7
FIGURE 2-8 Coal flow (in million short tons) in the United States in 2002 (Source: [21]).
1200
1000
Million Short Tons
800
600
400
200
0 1890
1910
1930
1950
1970
1990 2005
FIGURE 2-9 Coal production in the United States for the period 1890–2005 (Source: [23]).
58
Combustion Engineering Issues for Solid Fuel Systems TABLE 2-15 United States Coal Production in 2005 by Region and State Coal-Producing Region and State Appalachia Total Alabama Kentucky, Eastern Maryland Ohio Pennsylvania Anthracite Bituminous Tennessee Virginia West Virginia Interior Total Arkansas Illinois Indiana Kansas Kentucky, Western Louisiana Mississippi Missouri Oklahoma Texas Western Total Alaska Arizona Colorado Montana New Mexico North Dakota Utah Washington Wyoming Refuse Recovery U.S. Total
Million Short Tons 396.4 21.3 93.4 5.2 24.7 67.3 1.6 65.6 3.2 27.7 153.6 149.2 <0.05 32.1 34.4 0.2 26.4 4.2 3.6 0.6 1.8 45.9 587.0 1.5 12.1 38.5 40.4 28.5 30.0 24.5 5.3 406.4 0.7 1,133.3
Source: [5]
is the principal source of bituminous coal and anthracite and three of the top four coal-producing states (West Virginia, eastern Kentucky, and Pennsylvania) come from this region. (Note that eastern Kentucky is in the Appalachian Region, while western Kentucky is in the Interior Region.) These states, listed in Table 2-16, are among the top 10 producing states (of the 26 producing coal in 2005), which as a group produced over 85% of the coal in the United States. The Interior Region is composed of several separate basins located from Michigan to Texas and produced approximately 149 million short tons
Coal Characteristics
59
800
Million Short Tons
600
Surface
400 Underground 200
0 1950
1960
1970
1980
1990
2000
FIGURE 2-10 Coal production by mining method in the United States (Source: [24]).
800 East of Mississippi River
Million Short Tons
600
400
200 West of Mississippi River 0 1950
1960
1970
1980
1990
2000
FIGURE 2-11 Coal production by location in the United States (Source: [24]).
in 2005. Four states—Texas, Indiana, Illinois, and western Kentucky— produce the majority of the coal in this region. These coals range in rank from high-volatile bituminous coal in the northern part of the region to lignite in Texas. The Western Region of the United States has several coal basins that contain all ranks of coal. Over half of the coal produced in the United States comes from this region, with four of the states listed in Table 2-16 (Wyoming, Montana, Colorado, and North Dakota) producing more than
60
Combustion Engineering Issues for Solid Fuel Systems
Western 587.0 ian
Appalach 396.4 Interior 149.2
FIGURE 2-12 Coal production in 2005 by coal-producing region (million short tons) (Source: [23]).
1200 U.S. Total
Million Short Tons
1000 800 600
Western
Appalachian 400 200 Interior
0 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
FIGURE 2-13 Coal production by region (1993–2005) (Source: [23]).
Coal Characteristics
61
TABLE 2-16 Top 10 Coal-Producing States in 2005 State Wyoming West Virginia Kentucky Eastern Western Pennsylvania Texas Montana Colorado Indiana Illinois North Dakota Total Percentage of U.S. Total
Production (million short tons) 406.4 153.6 119.8 93.4 26.4 67.3 45.9 40.4 38.5 34.4 32.1 30.0 968.4 85.4%
45% of the total coal. Of these, Wyoming is by far the largest coal mining state in the United States and produces more than 35% of the country’s total coal. Lignite is mined in North Dakota and Montana; subbituminous coal is mined in southeastern Montana and northeastern Wyoming; and the principal bituminous coal mining production is in Utah, Colorado, and Arizona. Tables 2-17 through 2-19 contain selected characteristic anthracite, bituminous and subbituminous coals, and lignite analyses for United States coals. These analyses illustrate the range of analyses of United States coals with an emphasis on the major coal-producing regions and seams.
2.5 Traditional Coal Characterization Methods and Their Industrial Application In this section, traditional coal characterization methods and their application to industry are discussed. Special tests specific to some companies and nontraditional techniques will be discussed here, while test methods that pertain to the inorganic portion of the coal, such as deposition indices, will be presented in Chapter 4. Standard coal methods have been developed under the jurisdiction of the American Society for Testing and Materials (ASTM) Committee D-5 on Coal and Coke. These are referred to as ASTM methods and are commonly used in the United States and Canada. Standard coal methods have also been developed under the jurisdiction of the International Organization for Standardization (ISO), Technical Committee 27, Solid Mineral
62 TABLE 2-17 Properties of Selected United States Anthracites Rank
Anthracite
Anthracite
Anthracite
Sample Description
Mammoth Seam, PA
Buck Mtn. Seam, PA
#8 Leader Seam, PA
Prox. Anal. (%, as-rec.) Moisture Ash Volatile Matter Fixed Carbon Ult. Anal. (%, dry, ash free) Carbon Hydrogen Nitrogen Sulfur Oxygen HHVa (Btu/lb) Maceral Group Anal. (vol. % min. freeb) Vitrinite Inertinite Liptinite Vitrinite Reflectance (mean-maximum % in oil) Ash Fusion Temps. (Red, oF) Initial Deform. Softening Hemispherical Fluid
SemiAnthracite
SemiAnthracite
SemiAnthracite
PA #8 Seam
Gunnison Co., CO
L. Spadra Seam, AR
3.1 23.4 3.9 69.6
2.5 13.6 5.6 78.3
1.3 8.2 7.0 83.5
1.0 25.1 8.2 65.7
1.1 7.0 10.2 81.7
0.8 6.3 12.1 80.8
93.5 1.9 1.2 1.0 2.4 10,860
90.8 2.6 0.8 0.6 5.2 12,603
91.3 3.9 0.6 1.2 3.0 14,125
89.6 3.8 1.4 1.6 3.6 11,194
91.5 3.4 1.6 0.8 2.7 14,098
92.2 3.9 2.1 0.8 1.0 14,372
89.5 10.5 0.0
96.3 3.7 0.0
97.5 2.5 0.0
70.8 29.2 0.0
92.4 7.6 0.0
87.7 12.3 0.0
5.72
4.17
2.82
2.15
3.10
2.47
2,700þ 2,700þ 2,700þ 2,700þ
–c – – –
– – – –
– – – –
– – – –
2,150 2,250 2,485 2,560
Ash Comp. (%) SiO2 Al2O3 TiO2 Fe2O3 MgO CaO Na2O K2O P2O5 SO3 Trace Elements (ppm whole coal, dry) Chlorine Chromium Mercury Nickel Lead Strontium Vanadium a
Higher heating value White light Not available Source: [26] b c
56.5 28.4 1.77 6.10 1.32 0.84 0.29 2.70 0.25 0.70
54.9 32.4 3.82 3.29 1.09 1.53 0.12 1.22 0.06 2.10
56.7 32.5 1.81 4.55 0.41 0.58 0.16 2.30 0.09 0.20
58.8 32.3 1.27 4.18 0.64 0.29 0.28 2.99 0.06 0.10
37.4 25.3 0.80 12.8 2.02 10.8 1.25 1.20 1.69 7.69
34.4 26.2 1.01 13.8 4.96 7.19 1.80 1.50 0.66 9.00
1,300 41 – 21 – 82 58
100 44 0.05 27 11 17 24
100 27 0.08 15 <3 11 22
400 71 – 53 19 5 76
100 4 – 4 – 178 11
100 16 0.03 29 10 116 20
63
64
Combustion Engineering Issues for Solid Fuel Systems
TABLE 2-18 Properties of Selected United States Bituminous Coals Rank Sample Description
Low Vol.
Low Vol.
Med Vol.
Med. Vol.
L. Kittanning Seam, PA
Elk Lick Seam, WV
U.Freeport Seam, PA
Dutch Creek Seam, CO
Prox. Anal. (%, as-rec.) Moisture 2.0 1.3 Ash 10.1 18.3 Volatile Matter 17.4 16.0 Fixed Carbon 70.5 64.4 Ult. Anal. (%, dry, ash free) Carbon 88.8 87.2 Hydrogen 4.7 4.7 Nitrogen 1.6 1.5 Sulfur 1.6 3.3 Oxygen 3.2 3.3 HHVa (Btu/lb) 14,025 12,535 Maceral Group Anal. (vol. % min. freeb) Vitrinite 90.0 92.7 Inertinite 9.9 7.3 Liptinite 0.1 0.0 Vitrinite Reflectance (mean-maximum 1.73 1.63 % in oil) Ash Fusion Temps. (Red, oF) Initial Deform. 2,700þ 2,530 Softening 2,700þ 2,550 Hemispherical 2,700þ 2,600 Fluid 2,700þ 2,670 Ash Comp. (%) SiO2 42.7 50.8 Al2O3 39.0 27.5 TiO2 1.86 1.55 Fe2O3 10.7 14.5 MgO 0.57 0.80 CaO 1.71 1.15 Na2O 0.43 0.49 K2O 1.61 2.40 P2O5 0.94 0.39 SO3 0.40 0.82 Trace Elements (ppm whole coal, dry) Chlorine 1,300 700 Chromium 26 32 Mercury –c – Nickel 10 15 Lead – 8 Strontium 203 152 Vanadium 37 38
Med. Vol.
High Vol. A
Sewell Seam, WV
Pittsburgh #8, PA
6.1 10.1 22.9 60.9
1.1 5.3 26.9 66.7
1.5 4.2 24.6 69.7
2.4 10.0 35.2 52.4
87.0 5.5 1.7 2.9 2.9 14,006
87.4 5.9 1.7 0.7 4.3 14,889
88.2 5.0 1.5 0.7 4.6 14,871
83.3 5.7 1.4 1.3 8.3 13,532
88.5 11.5 0.0
94.4 5.6 0.0
77.2 18.0 4.8
83.0 8.9 8.1
1.24
1.28
1.35
0.87
2,020 2,130 2,200 2,250
2,750 2,800þ 2,800þ 2,800þ
2,225 2,330 2,390 2,445
39.2 27.5 1.05 18.7 0.34 4.06 0.11 0.91 0.84 5.00
24.8 22.7 0.82 10.8 3.52 9.40 2.84 0.33 0.90 13.7
52.8 33.3 1.02 4.59 0.97 1.78 0.70 2.29 1.16 0.20
55.8 25.8 1.21 6.37 0.91 3.20 0.49 2.19 0.56 2.10
100 19 0.16 20 13 172 26
700 3 – <1 – 200 6
2,600 11 – 11 – 104 11
2,300 21 – 11 – 111 7
– – – –
Coal Characteristics
65
TABLE 2-18 Properties of Selected United States Bituminous Coals (continued) Rank Sample Description
High Vol. A
High Vol. A High Vol. A High Vol. A High Vol. B High Vol. B
L. U. Kittanning Blind Kittanning Seam, WV Canyon Seam, WV Seam, UT Prox. Anal. (%, as-rec.) Moisture 1.8 1.5 4.7 Ash 11.8 10.4 5.6 Volatile 33.6 32.1 42.4 Matter Fixed Carbon 52.8 56.0 47.3 Ult. Anal. (%, dry, ash free) Carbon 84.5 85.2 81.3 Hydrogen 5.6 5.5 6.2 Nitrogen 1.4 1.5 1.6 Sulfur 1.0 2.0 0.4 Oxygen 7.5 5.8 10.5 HHVa (Btu/lb) 13,272 13,678 13,923 Maceral Group Anal. (vol. % min. freeb) Vitrinite 52.5 89.0 69.1 Inertinite 30.3 7.6 13.6 Liptinite 17.2 3.4 17.3 Vitrinite Reflectance (mean0.91 1.07 0.66 maximum % in oil) Ash Fusion Temps. (Red, oF) Initial Deform. 2,700þ 2,310 1,900 Softening 2,700þ 2,400 2,020 Hemispherical 2,700þ 2,460 2,090 Fluid 2,700þ 2,510 2,240 Ash Comp. (%) SiO2 53.0 46.1 50.3 Al2O3 34.9 29.5 15.3 TiO2 2.28 1.28 0.96 Fe2O3 5.49 15.0 6.88 MgO 0.70 0.82 1.26 CaO 0.89 2.59 12.0 Na2O 0.16 0.24 6.91 K2O 2.14 1.87 0.59 P2O5 0.08 0.21 0.38 SO3 1.00 1.30 4.80 Trace Elements (ppm whole coal, dry) Chlorine 600 3,700 1,200 Chromium 31 18 6 Mercury –c – – Nickel 18 6 <0.4 Lead – – – Strontium 39 62 85 Vanadium 28 7 9
L. Elkhorn Seam, KY
Hazard #5 Seam, KY
Hiawatha Seam, UT
1.3 9.7 30.9
3.8 9.1 34.6
4.9 7.1 39.5
58.1
52.5
48.5
86.5 5.5 1.4 0.6 6.0 13,897
82.8 5.5 1.6 0.8 9.3 13,369
80.6 5.7 1.4 0.9 11.4 13,264
71.2 16.8 11.9
60.3 21.1 18.7
89.2 9.6 1.2
1.03
0.80
0.58
2,700þ 2,700þ 2,700þ 2,700þ
2,700þ 2,700þ 2,700þ 2,700þ
2,110 2,210 2,260 2,360
58.2 28.3 1.58 3.37 1.10 0.69 0.56 3.74 0.08 0.50
54.1 31.1 1.81 3.86 0.83 2.09 0.32 2.08 0.42 1.20
49.5 21.2 1.12 4.19 0.97 8.09 4.91 0.21 0.88 8.00
900 21 0.08 15 7 73 25
1,100 31 – 24 10 297 34
300 19 – 24 8 262 8
(continued)
66
Combustion Engineering Issues for Solid Fuel Systems
TABLE 2-18 Properties of Selected United States Bituminous Coals (continued) Rank Sample Description
High Vol. B
High Vol. B
Kentucky #9 Illinois #9 Seam, KY Seam, IL Prox. Anal. (%, as-rec.) Moisture 6.8 5.3 Ash 11.4 10.3 Volatile Matter 38.4 37.0 Fixed Carbon 43.4 47.4 Ult. Anal. (%, dry, ash free) Carbon 79.1 79.9 Hydrogen 5.8 5.6 Nitrogen 1.4 1.6 Sulfur 4.8 4.6 Oxygen 8.9 8.3 HHVa (Btu/lb) 12,687 12,633 Maceral Group Anal. (vol. % min. freeb) Vitrinite 85.7 87.9 Inertinite 6.6 12.1 Liptinite 7.7 0.0 0.56 0.33 Vitrinite Reflectance (meanmaximum % in oil) Ash Fusion Temps. (Red, oF) Initial Deform. 1,755 2,090 Softening 1,900 2,125 Hemispherical 2,000 2,185 Fluid 2,040 2,285 Ash Comp. (%) SiO2 41.2 – Al2O3 15.6 – TiO2 0.75 – Fe2O3 22.0 – MgO 0.70 – CaO 8.08 – Na2O 0.66 – K2O 1.89 – P2O5 0.04 – SO3 9.40 – Trace Elements (ppm whole coal, dry) Chlorine 1,600 – Chromium 23 – Mercury –c – Nickel 8 – Lead – – Strontium 21 – Vanadium 19 – a
Higher heating value White light c Not available Source: [26] b
High Vol. C Wadge Seam, CO
High Vol. C
High Vol. C
Illinois #5 Seam, IL
Indiana #6 Seam, IN
9.5 6.4 38.0 46.1
9.5 9.5 36.4 44.6
11.5 13.5 33.5 41.5
77.5 5.5 1.8 0.6 14.6 12,762
78.4 5.2 1.4 4.9 10.1 12,722
79.7 4.9 1.6 3.4 10.4 12,292
89.2 8.7 2.1 0.60
87.5 10.6 1.9 0.62
89.0 8.7 2.3 0.43
2,700þ 2,700þ 2,700þ 2,700þ
1,945 2,065 2,160 2,245
2,035 2,080 2,365 2,425
48.7 31.1 0.92 3.60 1.47 6.22 0.92 0.84 1.03 3.90
43.1 16.0 0.78 24.2 0.81 5.66 0.45 1.76 0.10 6.10
50.1 18.8 0.89 20.2 0.86 2.16 0.57 2.57 0.15 2.30
100 4 – <2 – 175 7
800 10 – 13 – 17 14
500 14 – 6 – 24 19
Coal Characteristics
67
TABLE 2-19 Properties of Selected United States Low-Rank Coals Rank
Subbit. B
Subbit. B
Subbit. B
Subbit. B
Lignite A
Sample Description
Wyodak Seam, WY
Rosebud Seam, MT
Black Thunder
Antelope
Beulah Seam, ND
L. Wilcox Seam, TX
11.0 5.6 38.4 45.0
33.4 6.4 37.4 22.8
28.5 15.3 44.2 12.0
75.1 5.2 0.9 0.3 18.5 11,758
73.1 4.5 1.0 0.8 20.6 11,062
72.3 5.2 1.4 0.9 20.2 9,882
– – – –
– – – –
73.8 19.5 6.7 0.35
90.0 6.9 3.1 0.44
– – – –
– – – –
2,000 2,070 2,100 2,140
2,310 2,360 2,500 2,540
31.4 14.0 1.6 6.4 4.5 21.2 1.0 0.5 1.2 18.2
35.5 17.1 1.02 5.41 4.45 18.1 1.39 0.40 2.22 9.8
11.5 11.1 0.57 9.29 9.94 27.4 9.39 0.59 0.49 18.2
47.9 22.2 1.75 3.51 2.79 14.0 0.37 0.55 0.11 7.00
– – – – – – – – – –
– – – – – – – – – –
– – 800 5 – <0.4 – – 772 14
– – 1,100 17 – 5 – – 158 34
Prox. Anal. (%, as-rec.) Moisture 26.3 19.8 Ash 5.6 7.3 Volatile Matter 33.1 32.3 Fixed Carbon 35.0 41.6 Ult. Anal. (%, dry, ash free) Carbon 75.5 75.3 Hydrogen 6.1 5.1 Nitrogen 1.0 1.1 Sulfur 0.5 0.9 Oxygen 16.9 17.6 HHVa (Btu/lb) 12,145 11,684 Maceral Group Anal. (vol. % min. freeb) Huminite 85.7 79.8 Inertinite 9.7 18.8 Liptinite 4.6 1.4 0.29 0.51 Huminite Reflectance (meanmaximum % in oil) Ash Fusion Temps. (Red, F) Initial Deform. 2,100 2,145 Softening 2,130 2,170 Hemispherical 2,160 2,205 Fluid 2,185 2,310 Ash Comp. (%) SiO2 31.7 31.3 Al2O3 16.1 16.8 TiO2 1.27 0.79 Fe2O3 4.84 4.78 MgO 4.64 5.21 CaO 23.5 17.1 Na2O 1.80 0.20 K2O 0.40 0.10 P2O5 0.89 0.24 SO3 12.7 19.3 Trace Elements (ppm whole coal, dry) Arsenic 1 – Cadmium <0.2 – Chlorine –c 200 Chromium 3 7 Mercury 0.12 – Nickel 2 7 Lead 5 5 Selenium 2 – Strontium 164 225 Vanadium 11 5 a
Higher heating value White light c Not available Source: [26–28] b
26.3 5.6 33.1 35.0 75.5 6.2 1.0 0.4 16.9 12,220
Lignite
68
Combustion Engineering Issues for Solid Fuel Systems
Fuels. The ISO standards have been developed by ISO member nations for international trade, and, in many cases are similar to the ASTM standards [11]. This section will discuss selected ASTM standards and is not inclusive of all coal-related standards. As previously discussed, two analyses are used in classifying coal, one of which is a chemical analysis (for high-ranking coals) and one is a calorific determination (for lower ranking coals). The chemical analysis that is used in the proximate analysis (discussed later) gives the relative amounts of moisture, volatile matter, ash, i.e., inorganic material left after all the combustible matter has been burned off, and, indirectly, the fixed carbon content of the coal. The second analysis, the calorific value (also discussed later), also known as-heating value, is a measure of the amount of energy that a given quantity of coal will produce when burned. Note that when using the international coal classification system, additional analyses are performed. Since moisture and mineral matter (or ash) are extraneous to the coal substance, analytical data can be expressed on several different bases to reflect the composition of as-received, air-dried, or fully water-saturated coal or the composition of dry, ash-free (daf), or dry, mineral-matter free (dmmf) coal. Descriptions of the most commonly used bases are [7] as follows: As-received—Data are expressed as percentages of the coal with
the moisture. This is also sometimes referred to as as-fired and is commonly used by the combustion engineer to monitor operations and for performing calculations, since it is the whole coal that is being utilized; Dry basis (db)—Data are expressed as percentages of the coal after the moisture has been removed; Dry, ash-free (daf) basis—Data are expressed as percentages of the coal with the moisture and ash removed; Dry, mineral matter-free (dmmf) basis—The coal is assumed to be free of both moisture and mineral matter and the data are a measure of only the organic portion of the coal. This is used for coal classification; Moist, ash-free (maf) basis—The coal is assumed to be free of ash but still contains moisture; and Moist, mineral matter-free (mmmf) basis—The coal is assumed to be free of mineral matter but still contains moisture. This is used for coal classification.
Formulas developed by S.W. Parr in 1906 [29] to correct for the mineral matter and to determine the volatile matter, fixed carbon, and heating value on a mineral matter-free basis (for classification purposes) are
Coal Characteristics
69
provided in equations [2-1] through [2-6]. Equations [2-1] through [2-3] are Parr Formulas, and equations [2-4] through [2-6] are approximate forms of the Parr Formulas. Parr Formulas: Dry; mineral matter-free FCð%Þ ¼
FC 0:15S 100 [2-1] 100 ðM þ 1:08A þ 0:55SÞ
Dry; mineral matter-free VMð%Þ ¼ 100 Dry; mineral matter-free FC [2-2] Moist; mineral matter-free Btu ðper lbÞ ¼
Btu 50S 100 [2-3] 100 ð1:08A þ 0:55SÞ
Approximation Formulas: Dry; mineral matter-free FCð%Þ ¼
FC 100 100 ðM þ 1:1A þ 0:1SÞ
[2-4]
Dry; mineral matter-free VMð%Þ ¼ 100 Dry; mineral matter-free FC [2-5] Moist; mineral matter-free Btu ðper lbÞ ¼
Btu 100 100 ð1:1A þ 0:1SÞ
[2-6]
where Btu ¼ British thermal units (heating value) per lb; FC ¼ percent of fixed carbon; VM ¼ percent of volatile matter; M ¼ percent of bed moisture; A ¼ percent of ash; and S ¼ percent of sulfur.
2.5.1 Proximate Analysis The proximate analysis, determined by ASTM D 5142, gives the relative amounts of moisture, volatile matter, ash, i.e., inorganic material left after all the combustible matter has been burned off, and, indirectly, the fixed carbon content of the coal. Moisture determination is important because the moisture content of coals varies widely, with the moisture content the highest for the low-rank coals. The moisture content of a coal can vary after mining and is affected
70
Combustion Engineering Issues for Solid Fuel Systems
by processing or the elements (e.g., air drying or rain/snow). Moisture levels provide an indication of the drying required in the handling and pulverizing portions of the boiler coal feed systems [29]. Volatile matter is that portion that is driven off in gas or vapor form (exclusive of moisture). The main constituents of volatile matter are hydrogen, oxygen, carbon monoxide, methane, other hydrocarbons, and that portion of moisture that is formed by chemical combination during thermal decomposition of the coal [13]. Volatile matter is used to determine coal rank, an indicator of ease of ignition and if a supplemental fuel will be necessary for flame stabilization, and used as a basis for selling and purchasing of coal [13, 29]. In addition, its rate of release from the coal and its composition at various stages of release can be used in assessing NOx emissions and burner design and is discussed further in Section 2.6. Fixed carbon is the combustible residue after driving off the volatile matter. It is not all carbon and represents that portion of the coal that must be burned in a solid state. Fixed carbon is used to determine coal rank and is a guide in the choice of fuel-firing equipment [13]. The ASTM definition of ash is the inorganic residue remaining after ignition of combustible substances. The weight of ash is usually slightly less than that of the mineral matter originally present before burning; however, for high calcium-containing coals, the ash can be higher than the mineral matter due to the retention of the oxides of sulfur [13]. The ash content indicates the load under which the collection system for boiler bottom ash and fly ash must operate and is used for assessing shipping and handling costs [29].
2.5.2 Ultimate Analysis The ultimate analysis, determined by ASTM D 5373 (carbon, hydrogen, nitrogen) and ASTM D 4239 (sulfur), gives the amounts of carbon, hydrogen, nitrogen, sulfur, and oxygen comprising the coal. Oxygen is determined by difference, i.e., subtracting the total percentages of carbon, hydrogen, nitrogen, and sulfur from 100 because of the complexity in determining oxygen directly. However, this technique does accumulate all the errors in determining the other elements into the calculated value for oxygen. The ultimate analysis is used with the heating value of the coal to perform combustion calculations including the determination of coal feed rates, combustion air requirements, weight of products of combustion to determine fan sizes, boiler performance, and sulfur emissions [13, 29].
2.5.3 Heating Value The calorific value, also known as heating value, of the coal is determined using an adiabatic bomb calorimeter per ASTM D 5865 and is a measure of
Coal Characteristics
71
the amount of energy that a given quantity of coal will produce when burned. Heating value is used in determining the rank of coals. It is also used to determine the maximum theoretical fuel energy available for the production of steam. It is used to determine the quantity of fuel that must be handled, pulverized, and fired in the boiler [29].
2.5.4 Sulfur Forms Sulfur forms are determined by ASTM D 2492, which measures the quantity of sulfate sulfur, pyritic sulfur, and organically bound sulfur in the coal. The sulfur forms test measures sulfate and pyritic sulfur and determines organic sulfur by difference. Pyritic sulfur is an indicator of potential coal abrasiveness and is used for assessing coal cleaning because sulfur in the pyritic form is readily removed using physical coal cleaning methods.
2.5.5 Chlorine Chlorine analysis is determined using ASTM D 4208-02e1. Chlorine analysis is used to assess potential corrosion issues in all temperature regions of the boiler systems.
2.5.6 Grindability The Hardgrove grindability test, ASTM D 409, is an empirical measure of a coal’s pulverization characteristics. Grindability is determined using a Hardgrove machine that has been calibrated using standard reference samples of coal, having Hardgrove grindability indices (HGI) of 40, 60, 80, and 100. Standard coals are prepared by and obtained from The Pennsylvania State University, the only supplier of the ASTM standards worldwide. Pulverizer manufacturers have developed correlations relating HGI to pulverizer capacity at desired levels of fineness. Boiler operators use HGI when considering coal switching.
2.5.7 Ash Composition Elemental ash analysis is conducted using a coal ash sample from the ASTM D 5142 procedure. The elements present in the ash are determined by atomic adsorption (ASTM D 4208) and reported as oxides. The results of the ash analysis permit calculations of fouling and slagging indices, slag viscosity versus temperature relationships, determination of mineralogy, thermodynamic modeling, and other deposition indices. This is important because coal mineralogy can affect the ability to remove minerals during coal preparation/cleaning; coal combustion and conversion (i.e., production of liquid fuels or chemicals) characteristics; and metallurgical coke properties. The influence of the inorganic constituents on coal utilization is discussed in detail in Chapter 4.
72
Combustion Engineering Issues for Solid Fuel Systems
2.5.8 Trace Element Characterization Trace element analysis is becoming increasingly more important as emission regulations are either being implemented, such as for mercury, or under consideration, such as chromium or arsenic. Trace elements (excluding arsenic, selenium, and mercury) are determined by atomic absorption similarly to the elemental ash analysis. Arsenic and selenium are determined using graphite furnace atomic absorption spectroscopy (GFAAS) using EPA Methods 7060 and 7740, respectively. The samples are first dissolved via acid microwave digestion (EPA Method 3051—Microwave Assisted Acid Digestion of Sediments, Sludges, Soils and Oils). The samples are not ashed, since arsenic and selenium can be volatized during heating and are lost prior to analysis. Arsenic and selenium are generally reported on an as-fired (i.e., whole coal) elemental basis (ppm). Mercury is analyzed using cold-vapor atomic absorption spectroscopy (CVAAS) as outlined in EPA Method 7470 (Mercury in Liquid Waste Manual Cold-Vapor Technique). The sample is digested, not ashed, prior to analysis due to the volatile nature of mercury. The total mercury measured in the digest (mg/L) is converted to a mass basis. Mercury is reported on an as-fired elemental basis (ppm). It may not be legitimate to convert the mercury to an ash basis given that a significant percentage of the mercury would not necessarily remain in the solid phase during combustion but would be present in the gas phase. Arsenic and selenium are generally partitioned between the solid and gas phases; therefore, it might be presumptuous to assign all these elements to the solid phase.
2.5.9 Ash Fusion Ash fusibility has been long used as a tool for measuring the performance of coals related to slagging and deposition. Coal ash fusion temperatures are determined from cones of ash prepared and heated in accordance with ASTM D 1857. The temperatures at which the cones deform to specific shapes are determined in oxidizing and reducing atmospheres. Fusion temperatures provide ash melting characteristics and are used in determining furnace volumes to keep the furnace exit gas temperature to a level that minimizes furnace slagging. Indices correlating ash fusion temperatures with slagging/deposition are empirical and have little theoretical basis.
2.5.10 Free-Swelling Index (FSI) The free-swelling index, determined by ASTM D 720, is used to measure the relative caking properties of a coal and its extent of oxidation. This test is often used as a rough check to determine whether a coal is oxidized [13].
Coal Characteristics
73
2.5.11 Petrography/Coal Reflectance Petrographic analysis has many uses. Initially, it was primarily used to characterize and correlate seams and resolve questions about coal diagenesis and metamorphism, but later it influenced developments in coal preparation (i.e., crushing, grinding, and removal of mineral constituents) and conversion technologies [12]. Industrially, petrographic analysis can provide insight into the hardness of a coal (i.e., its mechanical strength) as well as a coal’s thermoplastic properties, which is of significant importance in the coking industry. Coal reflectance, which is performed according to ASTM D 2797-72 and ASTM 2798-72, provides a rapid and accurate index of coal rank [30]. Reflectance is defined as the proportion of normally incident light that is reflected by a plane, polished surface of the substance under consideration [30]. For most purposes, it is the reflectance of the vitrinite component that is determined, for the following reasons [30]: Vitrinite is the primary maceral in most coals; Vitrinite often appears homogenous under the microscope; Particles of vitrinite are usually large enough to permit measure-
ments to be made easily; and In the application of petrographic techniques to the industrial uses
of coal, particularly in carbonization, interest is focused on the behavior of vitrinite, which is the maceral primarily responsible for the plastic and agglutinating properties of coal. The basic use for which determination of vitrinite reflectance is applied is the measurement of the degree of metamorphism or rank of coals and the organic components of sediments. When one uses established curves, it is possible to relate the reflectance of a coal to parameters such as carbon content, volatile matter, and calorific value. Several laboratories use petrographic analyses to predict properties of coke, such as strength, which would be produced from the subject coal. Other uses include the estimation of the chemical properties of fresh coals from the reflectance of weathered specimens [29].
2.6 Nontraditional Characterization Methods and Their Industrial Application New approaches to coal characterization have been developed and applied to enhance knowledge of this fuel and how it burns or gasifies. These approaches include structural analysis, reactivity analysis, and volatile evolution pattern analysis.
74
Combustion Engineering Issues for Solid Fuel Systems
2.6.1 Coal Structure In recent years, it has become evident that the traditional means for characterizing coals as discussed previously are necessary, but not sufficient, to characterize the performance characteristics of low-rank coals and coals developed with coalification processes that differ from those associated with Eastern and Midwestern bituminous coal deposits. New analytical tools and techniques such as Carbon 13 Nuclear Magnetic Resonance (13C NMR), Nitrogen 15 NMR (15N NMR), drop-tube reactors (DTRs), analytical pyrolysis techniques, thermogravimetric analysis (TGA), differential thermal analysis (DTA), and the like have contributed significantly to the structural analysis of coal. Today, coals can be characterized in terms of aromaticity (percentage of total carbon atoms existing in aromatic structures), average number of aromatic atoms—or rings—per cluster, number and types of functional groups, locations of oxygen atoms, location and types of heteroatoms, and similar measures. Speculative coal structures can be postulated today with significantly more accuracy. These structures lead to a better understanding of utilization processes such as combustion.
2.6.2 Coal Reactivity Coal reactivity is a function, and a consequence, of coal structure. Reactivity can be measured in several ways including (not exhaustive) the following: Fixed carbon/volatile matter ratio (or its inverse) from the proxi-
mate analysis; Maximum volatile yield at some temperature; Activation energy for coal devolatilization or pyrolysis; and Activation energy for char oxidation.
These measures of reactivity complement the traditional approaches to coal characterization to yield a more complete picture of the chemistry of coal. Reactivity, as measured by maximum volatile yield, recognizes that volatile yield is a function of the temperature at which the fuel is reacted. For coals with low aromaticity, maximum volatile yield may occur at relatively low temperatures (e.g., 1800 F or 1000 C). For fuels with a larger quantity of aromaticity, the maximum volatile yield may not occur until 3090 F (1700 C) is reached. Such data have been developed for numerous coals, and other solid fuels, as previously presented in Chapter 1. A general correlation can be developed between maximum volatile yield and proximate analysis (air-dried basis). The relationship is shown by the following equation [27]:
Coal Characteristics
MVY% ¼ 0:645ðVMPA Þ þ 39:9
75
[2-7]
Where MVY% is maximum volatile yield and VMPA is the volatile matter expressed in an air-dried proximate analysis. The correlation coefficient, r 2, for this equation, over a suite of coals and opportunity fuels, is 0.942. Volatility can be related directly to oxygen content, and to the oxygencontaining functional groups as previously discussed. Green developed a mathematical relationship relating oxygen content to fuel volatility [31]. Using the principles elucidated by Green, one can posit the following equation relating oxygen to volatility: VMdaf ¼ 0:2252 þ 1:5435ðOÞ 0:3709ðO2 Þ
[2-8]
Where VMdaf is volatile matter in the proximate analysis on a dry, ash-free basis, and O is the oxygen content in the ultimate analysis also on a dry, ash-free basis. All values are expressed as decimals, not percentages. The r 2 for this equation, using a suite of solid fuels from petroleum coke through all ranks of coal to selected biomass fuels, is 0.953. Activation energies, associated with kinetics measurements, are another significant measure of reactivity. Activation energies for pyrolysis or devolatilization of a suite of coals and opportunity fuels were measured by Penn State’s Energy Institute as part of an extensive biomass cofiring program [32]. The technique used to measure devolatilization kinetics is described in detail elsewhere [27, 33, 34]. A sample of the devolatilization activation energy measured by Penn State is shown in Figure 2-14 for Black Thunder PRB coal. 1700 C
Reactivity R(1/sec)
1500 C
E = 9.53 kcal/mol A = 59.1 1/sec
1000 C 1
0.1 0.4
800 C
0.5
0.6
0.7 1/T ⫻ 1000, 1/K
0.8
0.9
1
FIGURE 2-14 Activation energy for devolatilization of Black Thunder coal (Source: [32]).
76
Combustion Engineering Issues for Solid Fuel Systems
In general, lower activation energies for devolatilization are associated with more reactive coals. For example, the measured activation energy for a high volatile Pittsburgh Seam bituminous coal was 10.8 kcal/mol, the activation energy for an Illinois Basin bituminous coal was 14.6 kcal/mol, and the activation energy for Black Thunder Powder River Basin (PRB) subbituminous coal was 9.53 kcal/mol. Similarly, the activation energies for char oxidation are indicitave of char reactivity, with the lower activation energies representing the more reactive chars. Overall, coal structure is the best indicator of reactivity, and it ties together the other reactivity measures [27]. Aromaticity, the number of aromatic carbons per aromatic cluster, and selected oxygen-containing functional groups provide the best quantification of coal reactivity. The links between structure and reactivity measures are numerous. Figures 2-15 and 2-16 illustrate two such linkages: fixed carbon/volatile matter ratio as a function of aromaticity, and activation energy as a function of aromaticity, respectively. With respect to the data in Figure 2-16, a somewhat more precise equation can be constructed as follows: Eact ¼ 1:91 þ 17:23ðArÞ 2:00ðAC=CIÞ
[2-9]
Where Eact is activation energy (kcal/mol), Ar is aromaticity, and AC/Cl is the number of aromatic carbons/cluster. The r2 for this equation is 0.867.
2.000 1.800
Aromaticity vs Fixed Carbon/Volatile Matter Ratio for Biomass and Various Ranks of Coal
1.600 [basis is proximate analysis] FC/VM Ratio
1.400 1.200 1.000 0.800
y = 0.1587e3.1241x R2 = 0.955
0.600 0.400 0.200 0.000 0.00
10.00
20.00
30.00 40.00 50.00 Aromaticity (%)
60.00
70.00
80.00
FIGURE 2-15 Fixed carbon/volatile matter ratio as a function of aromaticity (Source: Adapted from [27]).
Coal Characteristics
77
Pyrolysis Activation Energy (kcal/mol)
16 Devolatilization Activation Energy vs Fuel Aromaticity Beulah Lignite Black Thunder
14 12 10
Petroleum Coke
8
Sawdust
6
y = 14.425x + 1.1448 R2 = 0.852
4 2
Switchgrass 0 0
10
20
30 40 50 60 Aromaticity (Percent)
70
80
90
FIGURE 2-16 Pyrolysis (devolatilization) activation energy as a function of aromaticity (Source: [27]).
Essentially, the size of the aromatic clusters, on average, is used to modify the basic plot shown in Figure 2-16. Any number of other reactivity measures can be posited including using the volatile matter and fixed carbon values derived from the droptube reactor. Alternatively, one can employ the slope of the TGA plot during devolatilization—particularly the slope at the steepest point—as a measure of reactivity; steeper slopes indicate more reactive fuels. The alternative TGA measure sometimes used is the temperature of devolatilization initiation; lower temperatures indicate more reactive fuels. All these can be, and have been, used to measure coal reactivity. All these relate back to coal structure. For inorganic matter, reactivity can be measured in a number of ways; one of the most useful is chemical fractionation. This technique and results of its analysis are discussed in Chapters 4 and 8.
2.6.3 Volatile Matter Evolution Patterns Volatile matter evolution patterns provide yet another series of insights into the characteristics of coal and related solid fuels. The rate of total volatile matter evolution and the rate of evolution for specific volatiles (e.g., nitrogen volatiles) are of particular significance in burner design and in setting boiler operations strategies. The rate of volatile matter evolution can change combustion and heat release patterns. In evaluating such patterns, volatile matter evolution can complement more traditional coal burning profiles in assessing the patterns of combustion in a furnace and
78
Combustion Engineering Issues for Solid Fuel Systems
boiler. Such data can be very useful in evaluating where fuels burn in a rotary kiln, and consequently the temperature profile in that kiln. Such profiles are important in process industries such as pulp and paper (lime kiln for calcining) and cement manufacturing. Management of volatile matter evolution patterns also is of significance in controlling NOx emissions from fuel nitrogen sources. Baxter et al. demonstrated that the pattern of nitrogen volatile evolution played a significant role in the formation of NOx emissions from fuel nitrogen [35]. Nitrogen evolving rapidly, frequently under reducing conditions, does not oxidize but converts to N2. Nitrogen volatiles that evolve more slowly can exist in an oxidizing environment; NOx formation is more likely from these fuel nitrogen sources. This principle, widely understood, is one of the principles behind staged combustion as well as oxygen-enhanced combustion approaches where the oxygen is injected at the root of the flame, increasing volatile nitrogen evolution; however, the quantity of oxygen used is insufficient to convert the reducing environment at the base of the flame to an oxidizing environment. Figure 2-17 presents volatile evolution patterns measured for three different coals: Beulah Lignite, Black Thunder (southern PRB) subbituminous coal, and Pittsburgh Seam bituminous coal. The figure compares nitrogen volatile evolution with carbon volatile evolution. For the lignite and PRB coal, total volatile evolution is also shown. Note that the volatile matter evolution patterns are most favorable for the Black Thunder coal. The volatile nitrogen evolution lags significantly behind volatile carbon evolution for the lignite and for the bituminous coals. Both the lignite and
Percent Nitrogen and Carbon Volatile Yields
80.00 Lignite
70.00
Total Volatile Evolution
60.00 50.00 Carbon Volatile Evolution
40.00 30.00
Nitrogen Volatile Evolution
20.00 10.00 0.00 0
500
1000
1500 2000 Temperature (8F)
2500
3000
3500
FIGURE 2-17 Volatile matter evolution patterns for lignite, southern PRB subbituminous coal, and Pittsburgh Seam bituminous coal (Source: [26]).
Coal Characteristics
79
90.00 Percent Nitrogen or Carbon Released as Volatile Matter
Black Thunder 80.00 Nitrogen Volatile Evolution 70.00 60.00
Total Volatile Evolution
50.00 40.00 30.00
Carbon Volatile Evolution
20.00 10.00 0.00
Percent of Nitrogen or Carbon Evolved as Volatile Matter
0
500
1000
1500 2000 Temperature (8F)
2500
3000
3500
80.00 Pittsburgh Seam 70.00 60.00 50.00 Carbon Volatile Evolution 40.00 30.00 20.00 Nitrogen Volatile Evolution 10.00 0.00 0
500
1000
1500 2000 Temperature (⬚F)
2500
3000 3250
FIGURE 2-17 (continued)
bituminous coals generally produce more NOx when burned in utility or industrial boilers than the PRB coal [26]. Similar volatile evolution patterns can be developed for other elements including both organic and inorganic matter. These patterns provide combustion engineers with significant additional data with which models can be refined and with which understanding of coal combustion can be improved.
80
Combustion Engineering Issues for Solid Fuel Systems
2.7 References 1. EIA. 2005, November. Annual Coal Report 2004. Washington, D.C.: U.S. Department of Energy, Energy Information Administration. p. 31. 2. EIA. 2006, June. International Energy Outlook 2006. Washington, D.C.: U.S. Department of Energy, Energy Information Administration. 3. EIA. 2006, May. International Energy Annual 2004. Washington, D.C.: U.S. Department of Energy, Energy Information Administration. 4. World Coal Institute. 2005. The Coal Resource: A Comprehensive Overview of Coal. London: World Coal Institute. 5. Walker, S. 2000. Major Coalfields of the World. London: IEA Coal Research. 6. Tatsch, J.H. 1980. Coal Deposits: Origin, Evolution, and Present Characteristics. Sudbury, MA: Tatsch Associates. p. 5. 7. Van Krevelen, D.W. 1993. Coal: Typology—Physics—Chemistry—Constitution, 3rd ed. Amsterdam: Elsevier Science Publishers. 8. Schobert, H.H. 1987. Coal: The Energy Source of the Past and Future. Washington, D.C.: American Chemical Society. 9. Van Krevelen, D.W., and J. Schuyer. 1957. Coal Science: Aspects of Coal Constitution. Amsterdam: Elsevier Science Publishers. 10. Mitchell, G. 1997. Basics of Coal and Coal Characteristics, Iron & Steel Society, Selecting Coals for Quality Coke Short Course. 11. Elliott, M.A. (ed). 1981. Chemistry of Coal Utilization Second Supplementary Volume. New York: John Wiley & Sons. 12. Berkowitz, N. 1979. An Introduction to Coal Technology. New York: Academic Press. 13. Singer, J.G. (ed). 1981. Combustion: Fossil Power Systems. Windsor, CT: Combustion Engineering, Inc. 14. Miller, B.G., S. Falcone Miller, R. Cooper, J. Gaudlip, M. Lapinsky, R. McLaren, W. Serencsits, N. Raskin, and T. Steitz. 2003. Feasibility Analysis for Installing a Circulating Fluidized Bed Boiler for Cofiring Multiple Biofuels and Other Wastes with Coal at Penn State University. U.S. Department of Energy, National Energy Technology Laboratory, DE-FG26-00NT40809, Appendix J. 15. Mitchell, S.C. 1998. NOx in Pulverized Coal Combustion. London: IEA Coal Research. 16. Averitt, P. 1975. Coal Resources of the U.S., January 1, 1974, U.S. Geological Survey Bulletin No. 1412, (reprinted 1976), 131 pp. 17. Taylor, G.H., M. Teichmu¨ller, A. Davis, C.F.K. Diessel, R. Littke, and P. Robert. 1998. Organic Petrology. Berlin: Bebru¨der Borntraeger. 18. Bustin, R.M., A.R. Cameron, D.A. Grieve, and W.D. Kalkreuth. 1983. Coal Petrology: Its Principles, Methods, and Applications. Geological Association of Canada. 19. Miller, B.G. 2005. Coal Energy Systems. Oxford, United Kingdom: Elsevier. 20. World Coal Institute. Coal Facts 2005. London: World Coal Institute, www. worldcoal.org (accessed July 12, 2006).
Coal Characteristics
81
21. EIA. 2005, August. Annual Energy Review 2004. Washington, D.C.: U.S. Department of Energy, Energy Information Administration. 22. EIA. 1999, February. U.S. Coal Reserves: 1997 Update. Washington, D.C.: U.S. Department of Energy, Energy Information Administration. p. 5, Appendix A. 23. Freme, F., 2006. U.S. Coal Supply and Demand: 2005 Review. Washington, D.C.: U.S. Department of Energy, Energy Information Administration. 24. National Mining Association. 2006, May. Facts at a Glance. Washington, D.C.: National Mining Association, www.nma.org (accessed July 2006). 25. EIA. 1995, February. Coal Data: A Reference. Washington, D.C.: U.S. Department of Energy, Energy Information Administration. 26. Penn State Coal Sample Bank. www.energy.psu.edu/copl/index.html. 27. Tillman, D.A., B.G. Miller, D.K. Johnson, and D.J. Clifford. 2004. April. Structure, Reactivity, and Nitrogen Evolution Characteristics of a Suite of Solid Fuels. Proc. 29th International Technical Conference on Coal Utilization & Fuel Systems. Coal Technology Association, Gaithersburg, MD. 28. Miller, B.G., unpublished data, 2005. 29. Stultz, S.C., and J.B. Kitto (eds). 1992. Steam, Its Generation and Use, 40th ed. Barberton, OH: The Babcock & Wilcox Company. 30. Karr, J.R. (ed). 1978. Analytical Methods for Coal and Coal Products, vol. I. New York: Academic Press. 31. Green, A. 2003. Overview of Co-Utilization of Domestic Fuels. Proc. of International Conference on Co-Utilization of Domestic Fuels. University of Florida, Gainseville, FL. 32. Tillman, D.A., 2001. Final Report: EPRI-USDOE Cooperative Agreement Cofiring Biomass with Coal. Clinton, NJ: Foster Wheeler. Contract No. DEFC22-96PC96252. 33. Johnson, D.K., D.A. Tillman, B.G. Miller, S.V. Pisupati, and D.J. Clifford, 2001, September. Characterizing Biomass Fuels for Cofiring Applications. Proc. 2001 Joint International Combustion Symposium. American Flame Research Committee, Kauai, Hawaii. 34. Tillman, D.A., D.K. Johnson, and B.G. Miller. 2003, March. Analyzing Opportunity Fuels for Firing in Coal-Fired Boilers. Proc. 28th International Technical Conference on Coal Utilization and Fuel Systems. Coal Technology Association, Gaithersburg, MD. 35. Baxter, L.L., R.E. Mitchell, T.H. Flectcher, and R.H. Hurt, 1996. Energy & Fuels. 10(1): 188–196.
CHAPTER
3
Characteristics of Alternative Fuels N. Stanley Harding President N.S. Harding & Associates
3.1 Introduction The term “opportunity fuels” is typically defined by listing combustible resources that are outside the mainstream of commercial fuels, but that can be used productively in the generation of electricity or the raising of process steam and space heat in industrial and commercial applications. Under certain circumstances, materials once defined as opportunity fuels can emerge as mainstream and highly popular energy resources. Letheby [1, 2], for example, contends that Powder River Basin coals originally were opportunity fuels that emerged as a dominant commercial energy source when industry learned how to handle and use these materials, and when regulatory and economic pressures provided a highly receptive marketplace. Because opportunity fuels are outside the mainstream of fuels of commerce, the most common types are residues or low-value products from other processes. These can include, for example, petroleum coke, sawdust, hogged wood waste (mixtures of sawdust, hogged bark, planer shavings, and other solid wood products’ residues), spent pulping liquor, rice hulls, oat hulls, animal fats and proteins from meatpacking and rendering plants, wheat straws and other straws from agricultural activities, unusable hays, and a host of other solid and liquid materials [3, 4]. In addition, blast furnace gas, coke oven gas, refinery off-gases, and like products also can be considered as opportunity fuels and used primarily on-site in process 83
84
Combustion Engineering Issues for Solid Fuel Systems
industries. Because this book focuses on solid fuels, these liquid and gaseous fuels will not be discussed here, but several are presented in other sources such as Tillman and Harding [5]. Of this entire range of fuels, petroleum coke and wood wastes are the most widely used opportunity fuels, with the wood fuels alone supplying about 3 EJ (or quads) to the U.S. economy and about 20 EJ worldwide [6].
3.1.1 Typical Alternative Fuel Applications Opportunity fuels are used in a wide variety of applications, including electric utility boilers, industrial boilers and kilns, and institutional energy systems (e.g., prisons, hospitals, colleges, and universities). Frequently, these fuels are used to supplement traditional fossil energy sources such as coal; alternatively, they may be used as the sole fuel for a given boiler or kiln. 3.1.1.1 The Use of Alternative Fuels in Electric Utility Boilers The use of opportunity fuels in electric utility applications is particularly significant, given that electricity generation is the most significant growth sector in the energy arena, as shown in Figure 3-1. Fuel consumption for all other applications—in aggregate—has remained constant, whereas fuel consumption for electricity generation has continued to grow. Further, given the role that coal plays in electricity generation now and the fact that the U.S. Energy Information Agency projects coal to supply over half of
Fuel Consumption by Use 100.0
Non-Electricity Uses
90.0
Electricity Generation
Fuel Consumption, %
80.0 70.0 60.0 50.0 40.0 30.0 20.0 10.0 0.0 1950
1960
1970
1980
1990
2000
Year
FIGURE 3-1 Fuel consumption for electricity generation and non-electricity uses in the United States (Source: [7]).
Characteristics of Alternative Fuels
85
the electricity generated in the United States for the next 20 years [7], opportunity fuels have a significant potential in this arena. Despite the recent growth in natural gas-fired combined cycle combustion turbines (CCCT), wall-fired pulverized coal (PC) boilers, tangentially fired PC boilers, and cyclone boilers are the dominant sources of electricity generation. Recently, circulating fluidized-bed boilers also have entered the electricity-generating arena as well, firing crushed coal and opportunity fuels. Typical applications of opportunity fuels in utility boilers are as supplements to the main fuel—commonly eastern or midwestern bituminous coal, or Powder River Basin coal. This approach—cofiring an opportunity fuel with the base fuel—has received increasing attention in recent years, largely in response to the changing forces influencing electric utility behavior. Utility deregulation has been the first driver, forcing increased attention on fuel cost, and the total cost associated with using one or another coal or coal substitute. Deregulation has forced utilities to broaden their fuel slate to include “green power,” or environmentally friendly power. This power, typically sold at a premium, results from generation of electricity from renewable resources in an environmentally sensitive manner. Deregulation also has come with a variety of regulations promoting the use of certain alternative energy sources including biomass. These regulations typically take the form of Portfolio Standards, and numerous states have such laws mandating that utilities generate and/or sell a specific portion of their power based on renewable technologies including wind, solar energy, and—for some states—biomass. Certain environmental regulations also have promoted the cofiring of opportunity fuels such as wood waste, herbaceous materials, landfill gas, wastewater treatment gas, sewage sludge, and the like. The use of opportunity fuels to reduce SO2, NOx, and trace metal emissions has been a powerful driving force encouraging their use. In Europe, response to the potential for global climate change and global warming has resulted in regulations mandating the use of biomass opportunity fuels. Environmental regulations governing landfills and waste disposal have promoted the use of tire-derived fuels (TDF), refuse-derived fuels, and other opportunity fuels based on post-consumer wastes. At the same time, environmental regulations and regulatory interpretations have been less favorable to opportunity fuels such as petroleum coke and OrimulsionW. Driving forces to use opportunity fuels in electric utility boilers, then, range from reducing total fuel costs to improving boiler performance to addressing customer needs to achieving environmental gains. Consequently, many electricity-generating utilities in the United States and in Europe regularly investigate, and commonly fire, opportunity fuels. With all these pressures, Letheby [1] concludes that the baseload coalfired power plant is among the most promising markets for opportunity fuels. These plants can utilize up to 400,000 tonnes/yr at blend rates up
86
Combustion Engineering Issues for Solid Fuel Systems
to 20% (heat input basis), which is the common limit for using opportunity fuels in existing PC and cyclone boilers. At 20%, the opportunity fuels can significantly impact fuel cost, environmental performance, and other technical and economic parameters of such units. 3.1.1.2 Cofiring Alternative Fuels in Process Industries and Independent Power Producers Numerous process industries consume their own residuals as opportunity fuels. Pulp and paper mills consume bark and wood waste, spent pulping liquors, and pulp mill sludge as fuel. Spent pulping liquors are fired first for chemical recovery and secondarily for the generation of process steam or for power generation and process steam. Some utilize post-consumer waste paper in the form of recycle sludge. Steel mills consume blast furnace gas, coke oven gas, and other internally generated combustibles along with fossil fuels to be more cost effective. Petroleum refineries consume petroleum coke and refining off-gas in the production of refined oil products. Many additional industries consume opportunity fuels in their production process. Cement kilns commonly fire TDF and light liquid hazardous wastes in the production of clinker. Expanded aggregate kilns and other ore roasting kilns utilize the same hazardous wastes in their operations. These companies use opportunity fuels strictly as a means to reduce production costs in highly competitive environments. Independent power producers (IPPs), created initially by the passage of the Public Utilities Regulatory Policies Act (PURPA), are another group dependent heavily on opportunity fuels. Unlike electric utilities and process industries, however, IPPs tend to use opportunity fuels as 100% of the feed to any boiler. In eastern Pennsylvania, IPPs have constructed numerous circulating fluidized-bed (CFB) boilers fired with anthracite culm—a waste generated by the mining of anthracite. In California, IPPs built over 800 MWe of capacity firing vineyard prunings, orchard prunings, and a wide variety of agricultural materials and wood wastes. While some of this capacity has been mothballed, over 500 MWe is currently in operation. In Modesto, California, and elsewhere, IPPs have built generating stations fired totally with waste tires. Despite the use of opportunity fuels in standalone operations, however, the most common use—and the most promising use—remains in cofiring applications [8]. This results from the number of coal-fired boilers in the existing fleet of units; the potential benefits of fuel costs, emissions management, and other technical improvements resulting from cofiring wood waste; and the ability to manage the risk of using opportunity fuels by firing them in situations in which they can be removed if problems occur. This chapter will focus on four of the most commonly used alternative fuels in industry: petroleum coke, wood wastes, tire-derived fuel, and herbaceous crops.
Characteristics of Alternative Fuels
87
3.2 Petroleum Coke Petroleum coke is a solid byproduct of petroleum refining, useful in the production of electrodes used as carbon anodes for the aluminum industry and graphite electrodes for steel making, as fuel in boilers used to generate electricity, or as fuel for cement kilns. Currently, North America produces a very high percentage of all petroleum coke, and coking capacity continues to increase. Petroleum coke is of increasing importance in the cement industry [9], and is growing in use within the electricity-generating and industrial boiler communities also. Further, as utilities shift their focus toward lower cost generation including solid fuel supercritical boilers [10] and continue to investigate integrated gasification-combined cycle (IGCC) combustion turbine generating stations, petroleum coke will continue to grow in importance as a solid fuel. Petroleum coke is the dominant fuel used by JEA at its new circulating fluidized-bed (CFB) boilers; petroleum coke also provides over half of the fuel for the Polk County IGCC of Tampa Electric Company. Already, in the United States, over 1.5 million tonnes (1.68 106 tons) of petroleum coke are used by major utilities, as is shown in Table 3-1. The primary reason is the low cost of petroleum coke compared to coal.
TABLE 3-1 Petroleum Coke Consumption by U.S. Electric Utilities in the Year 2000 Utility Central Illinois Public Service Jacksonville Electric Authority Lakeland Dept of Water and Elect. Manitowoc Public Utilities Michigan South Central Power NIPSCO (NiSources) Northern States Power Ohio Edison Co. Owensboro, City of Pennsylvania Power San Antonio, City of Tampa Electric Co. Union Electric Co. (AMERON) Wisconsin Electric Power Co. Wisconsin Power & Light (Alliant Energy) TOTAL (*) *Totals may not add due to rounding. Source: [11]
Use (tonnes)
Use (tons)
Delivered Cost ($/GJ)
Delivered Cost ($/106 Btu)
23,400 400,000 1,800 32,400 1,800 156,800 198,200 7,200 8,100 183,000 8,100 190,000 111,700 132,400 62,200
26,000 444,000 2,000 36,000 2,000 174,000 220,000 8,000 9,000 203,000 9,000 211,000 124,000 147,000 69,000
0.86 0.58 0.41 0.45 1.02 0.62 0.31 0.70 0.51 0.70 0.40 0.48 0.58 0.66 0.45
0.91 0.61 0.43 0.47 1.07 0.65 0.33 0.74 0.54 0.74 0.42 0.51 0.61 0.70 0.47
1,516,200
1,683,000
0.55
0.58
88
Combustion Engineering Issues for Solid Fuel Systems
3.2.1 Petroleum Coke Production Processes Petroleum coke production processes, including delayed coking and fluid coking, are reviewed by Bryers [12, 13]. Delayed coking is a batch process in which residual components of crude oil are heated to about 475–520 C (890–970 F) in a furnace and then confined in a coke drum for thermal cracking reactions. The products of the coking process are gas, gasoline, gas oil, and coke. In a typical delayed coking arrangement, two coke drums are used; one is being filled and reactions are proceeding while the produced coke is being removed by high-pressure water. The delayed coking process produces various types of petroleum coke including needle coke, granular coke, sponge coke, and shot coke. Sponge and shot cokes are commonly found in the fuel market. Sponge coke is highly porous and anisotropic in nature. Shot coke, which appears like an assembly of small balls or beebees, also exhibits anisotropic characteristics; however, it is less porous and much harder to crush. Shot coke often is the result of upset conditions in the delayed coker. Fluid coke is produced in a fluidized-bed reactor where the heavy oil feedstock is sprayed onto a bed of fluidized coke. The oil feedstock is cracked by steam introduced into the bottom of the fluidized-bed reactor. Vapor product is drawn off the top of the reactor while the coke descends to the bottom of the reactor and is transported to a burner where a portion of the coke is burned to operate the process. Fluid coke reactors are operated at about 510–540 C (950–1,000 F). Flexicoke is a variant on fluid coke, where a gasifier is added to the process to increase coke yields. Fluid coker installations tend to have yields that are lower than delayed coker installations, whereas flexicoker installations have yields that can be significantly greater than delayed coker installations. Fluid cokers produce layered and nonlayered cokes. Both delayed and fluid coke installations produce amorphous, incipient, and mesophase cokes with the amorphous cokes having higher volatility and mesophase cokes having the lowest volatility. The attractiveness of fuel grade petroleum coke extends beyond price characteristics. It is typically high in heat content (typically >32.68 MJ/kg or 14,000 Btu/lb) and low in ash content (typically <1% ash). These characteristics favor its primary fuel use—as a fuel to be cofired with coal [14]. High sulfur concentrations and, for many petroleum cokes, high vanadium and nickel concentrations, remain as the less desirable characteristics of this alternative fuel. Periodically, low Hardgrove grindability index (HGI) values also inhibit the use of some petroleum cokes. In reality, even within fuel grade petroleum coke, there are substantial differences in fuel characteristics depending on source of crude oil and coking method.
3.2.2 Fuel Characteristics of Petroleum Coke Traditional analyses of solid fuels include proximate, ultimate, and ash elemental analysis along with calorific value and, increasingly, trace metal
Characteristics of Alternative Fuels
89
concentrations. These values are presented and discussed in following sections; however, additional fuel characterization discussions can be found in Tillman and Harding [5]. 3.2.2.1 Proximate and Ultimate Analysis of Petroleum Coke There is a significant body of literature concerning the traditional characteristics of petroleum coke [e.g., 12, 13, 15–17], and the influence of coking processes and conditions on such properties [e.g., 12, 13, 18]. Table 3-2 presents typical characteristics of various petroleum cokes as a function of coking method. Note that the flexicoke has the lowest volatile content and also the lowest calorific value. However, all the petroleum cokes are high in calorific value. On occasion, off-specification petroleum coke is also supplied to generating stations. A survey of tests conducted by others shows that petroleum cokes with 6–8% volatile matter are common; in one case, the Bailly Generating Station cofiring tests, the petroleum coke had a volatile matter content of 14% [19, 20]. The supply of petroleum coke with 12– 15% volatile matter is not uncommon in the use of this alternative fuel. The data in Table 3-2 can be converted into certain empirical performance characteristics as shown in Table 3-3.
TABLE 3-2 Typical Fuel Characteristics of Petroleum Coke Fuel Type Analysis Proximate Analysis (wt %) Fixed Carbon Volatiles Ash Moisture Ultimate Analysis (wt %) Carbon Hydrogen Oxygen Nitrogen Sulfur Ash Moisture Higher Heating Value MJ/kg Btu/lb HGI
Delayed Coke (Sponge)
Shot Coke
Fluid Coke
Flexicoke
80.2 4.48 0.72 7.60
89.59 3.07 1.06 6.29
91.50 4.94 1.32 2.24
94.9 1.25 0.99 2.86
81.12 3.60 0.04 2.55 4.37 0.72 7.60
81.29 3.17 0.93 1.60 5.96 0.76 5.69
84.41 2.12 0.82 2.35 6.74 1.32 2.24
92.00 0.30 0.00 1.11 2.74 0.99 2.86
33.18 14,298 54
33.34 14,364 39
32.53 14,017 35
32.43 13,972 55
Note: Some values in table depend on crude oil properties. Source: [12, 13]
90
Combustion Engineering Issues for Solid Fuel Systems
TABLE 3-3 Typical Empirical Performance Parameters of Petroleum Coke Fuel Type Analysis Volatility Measures Fixed Carbon/Volatiles Ratio Volatiles/Fixed Carbon Ratio H/C Atomic Ratio O/C Atomic Ratio Pollutant Measure (kg/GJ) Sulfur (as SO2) Nitrogen Ash Pollutant Measure (lb/106 Btu) Sulfur (as SO2) Nitrogen Ash
Delayed Coke
Shot Coke
Fluid Coke
Flexicoke
17.86 0.056 0.53 0.0004
29.18 0.034 0.46 0.0086
18.52 0.054 0.30 0.0073
75.92 0.013 0.04 0.00
2.63 0.76 0.22
3.57 0.48 0.23
4.14 0.72 0.40
1.69 0.34 0.31
6.11 1.78 0.50
8.30 1.11 0.53
9.62 1.68 0.94
3.92 0.79 0.71
With these performance characteristics, the subtle differences between types of petroleum cokes begin to emerge. The volatility differences are significant—despite the low volatility of all petroleum cokes represented in the table. The sulfur and nitrogen differences are a function of crude oil source as well as coking technology. Lower sulfur petroleum cokes, however, are more frequently used in nonfuel applications; petroleum cokes used for fuel purposes are typically high in sulfur content. 3.2.2.2 Ash Characteristics of Petroleum Coke Petroleum coke is, inherently, a low ash fuel, as shown previously in Tables 3-2 and 3-3. However, the ash chemistry of petroleum coke remains significant due to issues of slagging and fouling, and trace metal emissions. Bryers [12, 13] has reported general ash characteristics for petroleum cokes as a function of coking method. These data are shown in Table 3-4. Note the high concentrations of vanadium and nickel in all but the fluid coke. As a practical matter, the concentration of vanadium, nickel, and other inorganic matter in petroleum coke is largely a function of the source of the crude oil, as is shown in Table 3-5. “Sweet” (low sulfur) crude oils not only contain less ash, but can contain lower concentrations of vanadium, nickel, and other deleterious metals (see data from [12]). The vanadium content in the Venezuelan crude oil shown in Table 3-5 is particularly high, whereas the vanadium content in the Canadian crude oil is quite low. Mercury concentrations are high in the sample of crude oil from California, although mercury is unlikely to remain in the petroleum coke solid product due to the refinery processes.
Characteristics of Alternative Fuels
91
TABLE 3-4 Typical Ash Characteristics of Petroleum Coke Fuel Type Analysis
Delayed Coke
Shot Coke
Fluid Coke
Flexicoke
10.1 6.9 0.2 5.3 2.2 0.3 1.8 0.3 0.8 12.0 58.2
13.8 5.9 0.3 4.5 3.6 0.6 0.4 0.3 1.6 10.2 57.0
23.6 9.4 0.4 31.6 8.9 0.4 0.1 1.2 2.0 2.9 19.7
1.6 0.5 0.1 2.4 2.4 0.2 0.3 0.3 3.0 11.4 74.5
1,599 1,599 1,599 1,599
1,436 1,599 1,599 1,599
1,378 1,386 1,439 1,474
1,538 1,538 1,538 1,538
1,374 1,425 1,431 1,433
1,259 1,429 1,471 1,471
1,095 1,155 1,183 1,224
749 790 853 1,311
Elemental Composition (wt %) SiO2 Al2O3 TiO2 Fe2O3 CaO MgO Na2O K2O SO3 NiO V2O5 Ash Fusion Temperatures ( C) Reducing Initial Def. Spherical Hemispherical Fluid Oxidizing Initial Def. Spherical Hemispherical Fluid Source: [12]
TABLE 3-5 Trace Metal Concentrations in Crude Oil from Various Locations (mg/kg oil) Source of Crude Oil Metal As Co Cr Cu Hg Mn Ni Sb Se V Zn Source: [21, 22]
California
Libya
Venezuela
Alberta (Canada)
0.655 13.5 0.640 0.93 23.1 1.20 98.4 0.056 0.364 7.5 9.76
0.077 0.032 0.0023 0.19 — 0.79 49.1 0.055 1.10 8.2 62.9
0.284 0.178 0.430 0.21 0.027 0.21 117 0.303 0.369 1100 0.692
0.0024 0.0027 — — 0.084 0.048 — — 0.0094 0.682 0.670
92
Combustion Engineering Issues for Solid Fuel Systems
TABLE 3-6 Melting and Decomposition Temperatures of Nickel and Vanadium Compounds Potentially Formed from Petroleum Coke Combustion Compound Nickel oxide Nickel sulfate Vanadium trioxide Vanadium tetraoxide Vanadium pentoxide Sodium metavanadate Sodium pyrovanadate Sodium orthovanadate Sodium vanadyl vanadates Sodium vanadyl vanadates Nickel sulfides
Nickel pyrovanadate Nickel orthovanadate Ferric metavanadate Ferric vanadate
Symbol NiO NiSO4 V2O3 V2O4 V2O5 Na2OV2O5 2Na2OV2O5 3Na2OV2O5 5Na2OV2O45V2O5 5Na2OV2O411V2O5 NiS Ni2S4 Ni2S 2NiOV2O5 3NiOV2O5 Fe2O3V2O5 Fe3O42V2O5
Melting and Decomposition Temperature ( C) 2,090 Decomp 840; ! NiO 1,970 1,970 675–690 630 640 850 625 535 1,832 800 645 900 900 860 855
Source: [12]
Because nickel and vanadium are of primary concern, it is useful to note the potential nickel and vanadium compounds that can be formed during combustion of petroleum coke. Both nickel and vanadium are lithophiles and are typically found in porphyrin complexes [13]. Table 3-6 identifies such compounds along with their melting and decomposition temperatures. Note the wide array of potential compounds that can be formed depending on temperature and combustion conditions. The vanadium trioxide (V2O3) and vanadium tetraoxide (V2O4) compounds are typically associated with approximate unburned carbon levels in the fly ash 10%. This is consistent with the porphyrin structure of the vanadium in the crude oil and in the petroleum coke. Nickel can occur as a sulfide, an oxide, or in a vanadate (e.g., nickel pyrovanadate) depending on combustion condition. For example, reducing conditions promotes formation of sulfide compounds.
3.2.3 Petroleum Coke Utilization in Cyclone Boilers Cyclone boilers comprise about 9% of the coal-fired boiler capacity in the United States, and are also used in large industrial complexes such as the Eastman Kodak manufacturing park in Rochester, New York. Cyclone boilers are typically fired as baseload units. Typically, they generate significant NOx emissions, on the order of 0.52–0.65 kg/GJ (1.2–1.5 lb/106 Btu)
Characteristics of Alternative Fuels
93
unless NOx controls are added such as separated overfire air (SOFA), reburn technology, or selective catalytic reduction (SCR) technology. The high NOx emissions, coupled with developments that facilitated firing slagging coals in pulverized coal boilers, ultimately terminated further development of this technology. Because cyclone boilers were developed for a crushed fuel rather than a pulverized fuel, and because cyclone boilers were designed to remove inorganic matter as slag, they have become known for fuel flexibility. At the same time, the requirement to have sufficient ash in the fuel to form a slag layer—typically >5% ash in the coal—limits the use of alternative fuels in these combustion systems to cofiring applications. Further, cyclones require significant volatile matter concentrations to promote ignition and combustion within the cyclone barrel. Petroleum coke has been cofired in several such units [1, 23]. The impact of petroleum coke cofiring in cyclone boilers has been detailed elsewhere [5].
3.2.4 Cofiring Petroleum Coke in Pulverized Coal Boilers PC boilers are the most common large combustion systems for the generation of electricity in the United States and the industrialized economies of the world. PC boilers include both wall-fired boilers and tangentially fired (T-fired) boilers; wall-fired boilers include both front-wall and opposed-wall configurations. Like cyclone boilers, they have the potential to use petroleum coke as a fuel provided that a sulfur dioxide scrubber has been installed on the unit. The vast majority of the petroleum coke fired in the United States is burned in PC boilers due to their dominance of the industry. Again the low volatility in the petroleum coke limits its use in PC boilers; the typical cofiring percentage is on the order of 20–30% (calorific value basis). Cofiring petroleum coke with coal in PC boilers has the economic potential associated with reducing fuel costs identified previously. However, there are certain limitations with PC boilers. These units frequently sell fly ash as pozzolanic material under the American Society for Testing and Materials (ASTM) Specification C-618. ASTM C-618 limits the unburned carbon content in the fly ash. The ability to sell fly ash—or the inability to sell fly ash—can impact plant economics by >$1–3 million annually. Few cyclone boiler operators sell fly ash. Further, low NOx firing has been significantly advanced for PC boilers capitalizing on high volatile release at the base of the flame—and capitalizing on air-staged fuel combustion. While these concepts have been developed significantly for cyclone firing, they have been sufficiently advanced for PC firing that several units using PRB coals have achieved NOx emissions <64.5 g/GJ (0.15 lb/106 Btu) [24]. The constraints on PC cofiring of petroleum coke are more significant than in the case of cyclone firing. Consequently, the results of firing petroleum coke in PC boilers are mixed as discussed elsewhere [5].
94
Combustion Engineering Issues for Solid Fuel Systems
Petroleum coke can have certain economic advantages in PC boilers, particularly in terms of improving boiler efficiency and reducing fuel costs. Further, in many boilers, there will be no impact on main steam or reheat steam temperatures and consequently no impact on turbine efficiency. However, the impact of petroleum coke on operational issues in PC boilers is quite site-specific. Care must be taken to evaluate these impacts on any given location. The impact of petroleum coke cofiring on fly ash may or may not be significant, depending on whether or not the fly ash is sold as pozzolanic material. The impact of petroleum coke cofiring on airborne emissions from PC boilers is not as favorable as for cyclone boilers. While there are numerous reports of units where NOx emissions did not increase when petroleum coke was introduced with coal, there is potential for this outcome. That potential is consistent with the fuel characteristics of petroleum coke. Such potential may or may not be significant depending on whether a boiler has an associated SCR system for NOx control. Sulfur emissions also can increase, although this can be managed by scrubber technology. Nevertheless, the many PC boilers cofiring petroleum coke with coal speak to its advantages as an alternative fuel.
3.2.5 Petroleum Coke Utilization in Fluidized-Bed Boilers Circulating and bubbling fluidized-bed boilers have proven readily adaptable to the combustion of many opportunity fuels including petroleum coke. Their long residence times and high rates of heat transfer to the fuel particles makes them suitable for low volatile solids. At the same time, their use of limestone in the bed, which subsequently calcines to calcium oxide or free lime (CaO), provides a mechanism for capturing the sulfur during the combustion of high sulfur fuels. The free lime reacts with SO2 formed during the combustion process to produce calcium sulfate (CaSO4). In the past decade, circulating fluidized-bed (CFB) combustion has rapidly dominated this field. At the same time, low combustion temperatures of about 815 C–870 C (1,500 F–1,600 F) not only optimize the calcining and sulfation reactions, but also work with staged combustion processes in the CFB systems to minimize NOx formation. Petroleum coke can be fired in combination with coals or heavy oils in CFB boilers; alternatively, it can be fired as the sole fuel. Examples of petroleum coke firing in fluidized-bed boilers are shown in Table 3-7. Other examples of CFB boilers firing petroleum coke alone or in combination with other fuels include installations at Gulf Oil (now Chevron) in California; Oriental Chemical Industries in Korea; General Motors in Michigan; the Petrox refinery in Chile; a paper mill in Kattua, Finland; and numerous other sites. The principal benefits associated with combusting petroleum coke in fluidized-bed boilers include high boiler efficiencies and availabilities, and
Characteristics of Alternative Fuels
95
TABLE 3-7 Representative Circulating Fluidized-Bed Boilers Firing Petroleum Coke or Petroleum Coke/Coal Blends
Owner Purdue University City of Manitowoc (100% pet coke) Fort Howard Paper #3 (85% pet coke) Fort Howard Paper #5 (100% pet coke) Hyundai Oil, Korea (100% pet coke) University of N. Iowa (50–70% pet coke) NIPSCO (2 reheat boilers, 100% pet coke) JEA, Jacksonville, FL (2 300 MWe boilers firing coal and/or pet coke)
Steam Capacity (kg/sec)
Steam Capacity (103 lb/hr)
Main Steam Conditions (bar/C)
Main Steam Conditions (psig/ F)
25.2 25.2
200 200
45/440 67/485
650/825 975/905
40.4
320
103/510
1,500/950
40.4
320
103/510
1,500/950
30.3
240
108/520
1,565/970
13.2
104.5
52/332
750/630
104.0
825
112/540/540
1,625/1,005/1,005
251.4
1,993.6
172/538/538
2,500/1,000/1,000
Source: [25–27]
control of airborne emissions. The principal issues associated with fluidized-bed combustion of petroleum coke involve ash chemistry—management of limestone addition for optimized sulfur capture with minimum limestone cost—and management of vanadium-limestone interactions that cause agglomeration and fouling of heat transfer surfaces [28, 29]. Conn [29] has identified many of the solid products of CFB combustion, as shown in Table 3-8. The low melting points of some of these compounds, including TABLE 3-8 Possible Ash Constituents Formed during CFB Combustion of Petroleum Coke Compound Calcium sulfate Nickel oxide Sodium sulfate Vanadium trioxide Vanadium tetraoxide Vanadium pentoxide Calcium metavanadate Sodium metavanadate Nickel pyrovanadate Ferric metavanadate Source: [29]
Symbol
Melting Point C
Melting Point F
CaSO4 NiO Na2SO4 V2O3 V2O4 V2O5 CaOV2O5 2Na2OV2O5 2NiOV2O5 Fe2O3V2O5
1,450 2,090 880 1,970 1,970 680 780 630 900 860
2,642 3,794 1,616 3,578 3,578 1,274 1,432 1,166 1,650 1,580
96
Combustion Engineering Issues for Solid Fuel Systems
sodium sulfate, vanadium pentoxide, and calcium and sodium metavanadate, indicate some of the agglomeration issues that can exist. Petroleum coke and other petroleum-based fuels have made their way into large utility and industrial boilers, and into process industries such as cement manufacture and steel production. The characteristics of petroleum coke indicate a fuel that is high in calorific value and carbon content; however, it is not highly reactive as indicated either by the H/C and O/C atomic ratios, or by advanced analytical techniques such as the kinetics of devolatilization and char oxidation and maximum volatile yield [5]. Petroleum coke can be highly useful as a supplementary fuel in cyclone, PC, and fluidized-bed boilers. It also can be the exclusive fuel for CFB boilers. Alternatively, petroleum coke can be used in IGCC applications or as an additive in the production of metallurgical coke. Petroleum coke cofiring in conventional coal-fired boilers can improve boiler efficiency and make modest reductions in parasitic power loads of a generating station. When used in cyclone boilers, it can be effective in reducing NOx emissions along with trace metal emissions; the NOx reduction advantage does not exist in cofiring petroleum coke with coal in PC boilers.
3.3 Woody Biomass The first fuel of importance in pre-industrial economies, including the U.S. economy, was wood. Until 1850, wood was the dominant energy source, fueling steam engines, riverboat and railroad boilers, and metal smelting operations such as iron furnaces [30–32]. The lumber industry in Maine utilized steam boilers to reduce dependence on river water, and factories organized according to the principles of textile magnate Francis Lowell began employing steam boilers to provide mechanical energy for looms and other machinery [31]. With the advent of wood pulping technology, however, a higher and better use for wood emerged. Mechanical pulping was invented in Germany in 1844 and brought to the United States in 1867. The soda pulping process was invented in England in 1851 and brought to the United States in 1855; the sulfite process was invented in the United States and commercialized in 1866–1867. Kraft or sulfate pulping was invented in 1884 in Germany and introduced rapidly throughout the industrialized world [31]. The naval stores industry, with a rich tradition, formed one of the bases for the chemicals industry, and with the advent of dissolving pulp processes in the period 1916–1920, chemicals from woody biomass reached over 2 million tonnes [31]. Wood distillation products—e.g., methanol, acetone, acetate—also were significant components of the early chemicals industry. Wood and cotton were the main feedstocks of the synthetic materials industry until the 1930s [33]. During the period 1850–1900, energy consumption in the United States and throughout the industrializing world increased dramatically,
Characteristics of Alternative Fuels
97
and coal became the dominant fuel. Petroleum production, initiated in Titusville, Pennsylvania, in 1844, provided a glimpse of future energy demand; however, solid fossil fuels used in boilers and in coke ovens dominated energy use in all industrialized economies. By 1950, wood and biomass energy declined both absolutely and relatively; in the United States wood energy consumption decreased to about 1.26 EJ (1.2 1015 Btu) per year [5]. However during the last half of the 20th century, woody biomass became increasingly important, contributing about 3.5 EJ/year to the U.S. energy supply, along with supporting about 7,000 MWe of generating capacity. In selected European countries such as Finland and Sweden, wood fuel consumption assumed increasing importance, as is shown in Figure 3-2. The overwhelming percentage of biomass used is woody biomass. Wood energy use in the United States is dominated by woody biomass consumption for energy purposes, especially in the forest products industry. Wood is being consumed in standalone electricity generation installations as well—either in independent power producer (IPP) stations or in utility-owned generating installations. In the organization of forest industry activity, typically logs are debarked and either sawn for lumber, peeled for veneer and plywood, or sent to pulp mills. Chips from the sawmill and veneer production processes are sent to pulp mills along with logs, or sent to fiberboard plants. Sawdust and bark are sent to fuel applications—either as hog fuel to boilers or as feedstock to charcoal plants. While most logging residue is left in the forest, some logging residue has been reclaimed as “barky chips” either for pulping or for fuel applications.
Biomass Use by Country 20.0 18.0
Biomass Use (Percent)
16.0 14.0 12.0 10.0 8.0 6.0 4.0 2.0 0.0 Finland Sweden Austria Denmark
Italy
Germany
USA
Country
FIGURE 3-2 Percent of total energy consumption attributed to biomass consumption in selected industrialized economies (Source: [34])
98
Combustion Engineering Issues for Solid Fuel Systems
3.3.1 Types of Woody Biomass Fuels Today, in the United States and throughout the industrialized world, several types of woody biomass fuels are used in boilers, in kilns, and in cofiring applications. Cofiring of woody biomass in utility boilers has, in reality, become a new and potentially large market for these opportunity fuels. Wood fuels available today include, primarily, residues from the wood processing industry. This industry includes sawmills, veneer and plywood mills, particleboard plants, medium density fiberboard (MDF) mills, and pulp mills of various types. It also includes secondary processing industries such as furniture, flooring, cabinetry, and related product manufacturing operations. The residues from the processing facilities are bark, sawdust, planer shavings, sander dust, chips, and mixtures of the above known as “hog fuel.” The name “hog fuel” refers to the fact that much of this material is processed through hammermills, also known as “hammer hogs” or “hogs.” Spent pulping liquor from pulp and paper mills—including black liquor from kraft mills and red liquor from sulfite mills—provides a special type of wood waste burned only within the pulp and paper community in special chemical recovery boilers designed to return sodium and sulfur to the Kraft process and provide steam back to pressure turbines and process needs. In addition to mill residues, forestry produces significant in-forest residues. In some cases these have been recovered for fuel applications, although such cases are rare due to the relatively high cost of such fuels. The 800 MW wood-to-electricity industry in California during the late 1980s and 1990s was based partly on recovery of such materials. It is frequently proposed that trees be grown specifically as an energy crop, and such proposals have been made for at least the past 30 years [35–41]. Environmental benefits of such short rotation woody crop concepts are cited to promote such concepts. To date, this source of wood fuel has not been commercialized in the United States and other industrialized economies due to the high costs of such materials. The final class of woody biofuels available is the urban wood wastes. Urban wood waste has long been viewed as a potential source of fiber, and recycling programs have made this source potentially promising. Urban wood wastes include manufacturing residues (e.g., residues from the production of manufactured homes or recreational vehicles); construction, demolition, and land-clearing materials; wood from pallet processors; pallets and other dunnage from the commercial transportation industry; and related “clean” or untreated woody materials. Treated wood wastes also are included in the urban wood waste category. These include used railroad ties, used utility poles, lumber treated with copper-chromium-arsenate (CCA) or pentachlorophenol (PCP) for outdoor applications, and related products. Treated wood waste is a special class of materials, distinct and separate from “clean” or untreated wood waste.
Characteristics of Alternative Fuels
99
3.3.2 Physical and Chemical Characteristics of Woody Biomass Fuels Woody biomass, either gymnosperms (softwoods) or angiosperms (hardwoods), is inherently anisotropic and hygroscopic; it is a porous material, with the porosity caused by the hollow fibers that make up the woody material. The porosity or void volume of wood exists as macropores—tracheids, rays, resin canals, and related structures useful in moving moisture and nutrients within the tree. Chemically, woody biomass is composed of cellulose, the hemicelluloses, one or another type of lignin, and extractives such as pinoresinol, catechin, pinosylvin, and related compounds. On an oven dry weight and volume basis, solid wood typically has a specific gravity of 0.4–0.7, with some species being slightly lower and others being slightly higher [42]. Table 3-9 provides a short table of specific gravities for representative species. Specific gravities of fast-growing woody biomass are typically lower than specific gravities of naturally grown trees. Short rotation woody crops can have specific gravities in the 0.28–0.33 range depending on species and the age at harvest. Moisture content in woody biomass fuels is both a function of the living and growing process, and a function of the manufacturing processes imposed on a given piece of wood. Because many debarkers and saws are water-cooled devices, much of the moisture in sawdust can be attributed to the processing activities. Heartwood in the hardwoods is typically 32–48% moisture before processing, and hardwood sapwood is typically 31–53% moisture depending on individual species [43]. Softwoods have typical heartwood moisture contents of 25–49%, and sapwood moisture contents of 52–71% depending on species [43]. As a practical matter, sawdust or hog fuel received for use under boilers is typically 40–50% moisture [44, 45]. The combination of specific gravity and moisture content leads to considerations of bulk density—critical in calculating conveying system requirements. For sawdust and hog fuel with moisture contents in the 35–45% range, typical bulk density measurements have been 0.205–0.256 kg/m3 (16–20 lb/ft3). For woody biomass fuels with lower moisture TABLE 3-9 Specific Gravities for Selected Species (Dry Basis) Species Douglas-fir W. Hemlock Loblolly Pine Pitch Pine White Pine Source: [42]
Specific Gravity
Species
Specific Gravity
0.45–0.5 0.42–0.45 0.47–0.51 0.47–0.52 0.34–0.35
American Elm Shagbark Hickory Black Maple White Oak Willow
0.46–0.50 0.64–0.72 0.52–0.57 0.60–0.68 0.56–0.69
100 Combustion Engineering Issues for Solid Fuel Systems
contents (e.g., 12% moisture) such as planer shavings or pallet wastes, bulk densities can be on the order of 0.128–0.154 kg/m3 (10–12 lb/ft3) [45, 46]. 3.3.2.1 Proximate and Ultimate Analysis of Woody Biomass Chemically, wood is composed of cellulose (C6H10O5); the hemicelluloses including xylan, galactoglucomannans, arabinoglucuronoxylan, arabinogalactan, and others in softwoods and glucuronoxylan and glucomannan and others in hardwoods; lignins (C9H10O3(OCH3)0.9-1.7); and extractives such as aliphatic compounds (fats and waxes), terpenes and terpenoids, and phenolic compounds [31, 47]. Additionally, woody biomass includes minor quantities of proteins and inorganic matter. On an extractive-free basis, softwoods typically contain 45–50% cellulose, 25–35% hemicelluloses, and 25–35% lignin, whereas hardwoods contain about 40–55% cellulose, 24–40% hemicelluloses, and 18–25% lignin. Extractives typically comprise 1–5% of the wood [44, 47]. Table 3-10 presents analyses for three wood types. From a fuels perspective, chemical composition is typically presented as proximate and ultimate analysis, shown in Table 3-11 for selected wood species. Note the differences between the virgin wood fuels and the urban wood waste. These differences go beyond lower moisture content and higher ash content, and include a very high concentration of fuel nitrogen. The fuel nitrogen comes from the urea formaldehyde and related glues used to manufacture plywood, particleboard, and paneling. On a kg/GJ (lb/106 Btu) basis, these fuels have the following fuel nitrogen concentrations: pine: 0.103 (0.24); red oak: 0.267 (0.62); mixed hardwood/softwood sawdust: 0.206 (0.48); and urban wood waste: 0.72 (1.67). 3.3.2.2 Inorganic Matter in Woody Biomass Inorganic matter, expressed as a percentage of the total fuel mass or in g/kJ (lb/106 Btu) is routinely low for wood fuels, particularly when compared to coals. Urban wood waste is much higher in ash content, as shown in Table 3-11; the difference is in the presence of dirt in the urban wood TABLE 3-10 Composition of Three Wood Species (Values in Weight %) Wood Species Component Cellulose Hemicelluloses Lignin Extractives Protein Inorganic Matter Source: [48]
Spruce
Pine
Birch
43 27 28.6 1.8 1.3 0.4
44 26 29 5.3 1.2 0.4
40 39 21 3.1 2.5 0.3
Characteristics of Alternative Fuels
101
TABLE 3-11 Typical Fuel Characteristics of Selected Wood Fuels Fuel Type Red Oak
Mixed Sawdust
Urban Wood Waste
Analysis
Pine
Moisture (%) Proximate Analysis (dry wt %) Fixed Carbon Volatiles Ash Ultimate Analysis (dry wt %) Carbon Hydrogen Oxygen Nitrogen Sulfur Ash Higher Heating Value (dry basis) MJ/kg Btu/lb
45.0
28.8
40.0
30.8
15.2 84.7 0.1
19.0 79.5 1.5
19.0 80.0 1.0
18.1 76.0 5.9
49.1 6.4 44.0 0.2 0.2 0.1
51.6 5.8 40.6 0.5 <0.1 1.5
49.2 6.0 43.0 0.4 <0.1 1.0
48.0 5.5 39.1 1.4 0.1 5.9
19.79 8,502
18.78 8,069
19.56 8,400
19.47 8,364
Source: [3, 6, 49]
waste, along with wood coatings (e.g., paints) and inorganic matter in many of the glues used for producing particleboard, plywood, paneling, furniture, and other products that end up in the urban wood waste pile. Tables 3-12 and 3-13 present bulk ash chemistries of various wood fuels obtained during various test programs. Table 3-12 focuses on virgin wood material, including sawdust and clean tree trimmings from the urban forest. Table 3-13 presents representative data from some urban wood wastes plus short rotation woody crops. Wide variations in ash compositions can be observed between the various woody fuel samples. To a large extent, the silica content reflects the amount of dirt in the sample—above about 5–10% silica in the inorganic constituents. Note the wide variation in calcium oxide and alumina among the samples shown. Also note the high concentration of potassium oxide in almost all samples shown. It can be readily observed that the woody biomass ash is almost always high in calcium oxide and potassium oxide; the sodium oxide can be relatively low although wood fuel derived from salt water-stored logs can have significant sodium concentrations in the inorganic constituents. 3.3.2.3 Trace Metal Concentrations Trace metal concentrations in woody biomass are typically lower than trace metal concentrations in coal [6, 22, 46]. Table 3-14 summarizes trace metal concentrations from wood fuel found at various locations [22, 50–52], and Table 3-15 summarizes trace metal concentrations in the urban wood waste cofired at the Bailly Generating Station.
102 Combustion Engineering Issues for Solid Fuel Systems TABLE 3-12 Ash Elemental Analysis of Typical Wood Fuels Fuel Type Analysis
Pine Wood
Elemental Composition (wt %) SiO2 57.2 Al2O3 13.4 1.16 TiO2 Fe2O3 5.94 CaO 8.75 MgO 3.35 Na2O 1.38 K2O 4.94 P2O5 1.44 0.05 SO3 Reducing Ash Fusion Temperatures ( C) Initial Def. — Spherical — Hemispherical — Fluid — Oxidizing Ash Fusion Temperatures ( C) Initial Def. — Spherical — Hemispherical — Fluid —
Hardwood Sawdust
Pine Bark
Mixed Hardwood, Softwood
0.40 0.30 0.00 0.20 40.6 5.10 0.30 26.5 11.5 3.0
23.5 5.10 0.10 2.10 33.6 5.10 0.20 12.0 4.80 1.60
1,246 1,414 1,417 1,424
1,375 1,504 1,506 1,507
1,230 1,240 1,245 1,290
1,397 1,406 1,408 1,414
1,340 1,525 1,650 1,650
1,210 1,225 1,250 1,275
23.7 4.10 0.36 1.65 39.9 4.84 2.25 9.81 2.06 1.86
Source: [6, 45, 46]
3.3.3 Using Woody Biomass in Dedicated Boilers Given the unique characteristics of woody biomass, it has been frequently used as an alternative fuel. It is a common fuel—the dominant fuel—in the wood products and pulp and paper industries. In such applications it is used in boilers designed specifically to fire wood waste. Large wood-fired boilers have been installed for both forest products industry applications and for selected utility applications. In industrial settings, wood-fired boilers can be sized anywhere from 3 kg/sec (25 103 lb/hr) saturated steam to over 88.3 kg/sec (700 103 lb/hr) superheated steam suitable for power generation. Largescale wood-fired boilers used in the pulp and paper industry for both process steam and power generation are on the order of 37.8 kg/sec (300 103 lb/hr) – 75.7 kg/sec (600 103 lb/hr) of steam typically produced at 86.2 – 100 bar (1,250 – 1,450 psig)/510 C (950 F). Such boilers are not reheat boilers; rather, the steam generated from these boilers passes through either a backpressure or noncondensing extraction turbine
Characteristics of Alternative Fuels
103
TABLE 3-13 Ash Elemental Analysis of Typical Wood Fuels Fuel Type
Analysis
Shredded Hybrid Poplar Willow Grown as Green Sawdust, Railroad Ties from Mine Short Rotation East Tennessee (8.58% ash, Reclamation (9.8% Woody Crop (1.82% (1.06% ash, o.d. basis)* ash, o.d. basis) ash, o.d. basis) o.d. basis)
Elemental Composition (wt %) SiO2 47.26 59.21 Al2O3 16.82 16.69 TiO2 1.25 0.67 12.34 8.20 Fe2O3 CaO 9.18 4.90 MgO 4.21 2.05 Na2O 2.73 0.31 2.15 8.10 K2O P2O5 0.64 0.62 SO3 0.65 0.09 Reducing Ash Fusion Temperatures ( C) Initial Def. 1,108 — Spherical 1,120 — Hemispherical 1,124 — Fluid 1,136 — Oxidizing Ash Fusion Temperatures ( C) Initial Def. 1,170 — Spherical 1,187 — Hemispherical 1,194 — Fluid 1,207 —
4.39 1.52 1.35 1.02 54.09 4.18 0.75 12.50 1.36 3.57
21.30 4.55 0.61 2.01 39.83 6.61 2.20 9.48 2.99 1.86
— — — —
1,392 1,397 1,399 1,403
— — — —
1,356 1,386 1,388 1,390
*% ash on an o.d. basis refers to the percent ash in the total fuel on an oven dry basis. Source: [45]
and then is exhausted typically at 3.4 bar (50 psig) – 10.3 bar (150 psig) for process applications. Wood-fired boilers of this type are typically grate-fired units, although CFB boilers are becoming increasingly popular for power boiler applications. With the advent of high pressure/high temperature CFB units, pulp mills have used them to overcome problems of variability in the fuel composition and moisture content. Further, several pulp mills have turned more to CFB boilers because they can accept wide variations in the fuel blend. The literature concerning such systems is replete with citations (see, for example, [22, 44, 54–59]). The typical grate-fired boiler for woody biomass can come in a number of firing system configurations including pile burner, inclined grate, and spreader-stoker designs. These boilers fire a range of solid fuels including bark, hogged wood waste, sawdust, orchard and vineyard prunings, municipal solid waste, refuse-derived fuel (from municipal solid waste), and blends of the above with such opportunity fuels as tire-derived fuel.
104 Combustion Engineering Issues for Solid Fuel Systems TABLE 3-14 Trace Metal Concentrations in Various Woody Biomass Fuels (mg/kg Wood Ash) Woody Biomass Fuel
Metal As Ba Be Cd Cr (total) Cr (þ6) Cu Pb Hg Mo Ni Se Ag V Zn
Big Valley Lumber, Bieber, CA
Burlington Electric, Burlington, VT
Georgia Pacific, Lyons, NY
Hog fuel from a Pulp Mill in Pacific Northwest
4.6 130 0.1 1.5 22 — 40 38 <0.05 14 11 5.0 <0.08 26 130
6.3 — — 16 25 — 70 70 — 3.0 — — — 27 560
— — 6.7 16.8 — 76.9 58.8 — — — — — — 310
0.475 51.5 BDL BDL 128.4 0.063 5.625 2.71 BDL — 137.3 — — — 99
Source: [22, 44, 50–53]
TABLE 3-15 Trace Metal Concentrations in the Urban Wood Waste Burned at Bailly Generating Station Metal Arsenic Chromium Lead Mercury Nickel Vanadium
Concentration (mg/kg) 2.145 6.57 2.922 0.0126 2.645 3.060
Source: [46]
Fuel for woody biomass-fired boilers—either grate fired or CFB fired— is typically sized depending on fuel type, with specifications for woody biomass being typically 25 mm 0 mm (1" 0"). In these systems, the maximum firing capacity of the boiler is typically around 630–740 GJ/hr or 175–206 MWth (600–700 106 Btu/hr) for inclined grate and spreaderstoker systems; pile-burning systems are typically much smaller. The maximum capacity is a function of both materials handling logistics and
Characteristics of Alternative Fuels
105
grate design. Typical combustion parameters for these systems are shown in Table 3-16. Studies of dedicated boilers have demonstrated that, under certain circumstances, significant deposition can occur either as slagging (in the furnace zone) or as fouling (in the convective passes of the boiler). These studies, led by Baxter et al. [60, 61], have shown that the alkalinity of the woody biomass ash—and the reactivity of the alkali metals and alkali earth elements—can be a significant problem. The problem is exacerbated when chlorine is present, resulting in the formation of alkali chlorides. This problem is not severe when sawdust or hog fuel from primary wood processing is used, unless that wood comes from salt water-stored logs. It is more severe when woody materials such as vineyard or orchard prunings are used as fuel, since these materials come from biomass grown in an agricultural setting. Commercially, dedicated woody biomass boilers are highly suited to the forest products industry, where the fuel is generated on-site and where both process steam and electricity—or shaft power—are required. However, the fact that woody biomass does not pulverize readily, as does coal, limits the firing systems to stokers and CFB boilers. The plant site logistics and low bulk density of woody biomass typically limits the boiler to <88.3 kg/sec (700 103 lb/hr) of steam. This, in turn, has placed economic limits on the steam conditions associated with woody biomass boilers. At the maximum, steam conditions in woody biomass boilers are typically 100 bar/510 C (1,450 psig/950 F); typical conditions for large-scale woody biomass boilers are on the order of 58.6–86.2 bar (850–1,250 psig) and 440–510 C (825–950 F). These flows, pressures, and temperatures do not economically support the use of reheat cycles in dedicated woody biomass-fired boilers [44]. Because woody biomass boilers are not installed with reheat cycles, the electricity-generating efficiencies of dedicated wood-fired systems are TABLE 3-16 Typical Parameters for Woody Biomass-Fired Boilers Boiler Type
Maximum fuel moisture (%) Grate heat release (MJ/m2-hr) Grate heat release (kW/m2) Grate heat release (103Btu/ft2-hr) Furnace heat release (MJ/m3-hr) Furnace heat release (kW/m3) Furnace heat release (103Btu/ft3-hr) Furnace gas residence time (sec) Typical combustion temp ( C) Typical combustion temp ( F) Source: [44]
Pile Burner
Inclined Grate
Spreader-Stoker
65 227–284 63–79 200–250 372 2.8 10 3–4 1,090 2,000
60 227–454 63–126 200–400 372–484 2.8–3.6 10–13 2–4 1,150 2,100
55 908–1,135 252–315 800–1,000 560–745 4.2–5.6 15–20 2–3 1,260 2,300
106 Combustion Engineering Issues for Solid Fuel Systems
significantly lower than those associated with coal-fired central generating stations. The typical wood-fired generating station has a net station heat rate (NSHR), at best, of 13.7 MJ/kWh (13,000 Btu/kWh). Typically, the modern wood-fired generating stations employing 86.2 bar/510 C steam generate electricity with an NSHR of 14.76 MJ/kWh (14,000 Btu/kWh) or higher [44, 62]. The efficiencies of modern single and double reheat coalfired boilers are typically 10.01–11.07 MJ/kWh (9,500–10,500 Btu/kWh). Modern natural gas-fired combined cycle combustion turbine generating stations can exhibit even more favorable NSHR values—on the order of 7.91 MJ/kWh (7,500 Btu/kWh) or better [63]. The largest wood-fired boilers, if dedicated to the generation of electricity, would support some 70–75 MWe of capacity. Typical woody biomass boilers installed throughout the United States have been 11–30 MWe in capacity. This contrasts with modern coal-fired generating stations averaging 550 MWe per boiler or more for all units installed in the United States since 1971 [63]. The consequence of this scale difference is reflected in operating and maintenance labor requirements. Woody biomass boilers typically generate 0.5–2 MWe/employee, with the largest units generating 2.3 MW/plant employee and 1.14 MW/total employees on the payroll [59]. Coal-fired generating stations can generate an order of magnitude more electricity per employee. Modern natural gas-fired combined cycle combustion turbine-based stations also can be built at scales of 500 MWe and greater. Economics do not favor standalone woody biomass electricitygenerating stations in modern industrial economies, except in niche applications or when incentives are applied.
3.3.4 Woody Biomass in Pulverized Coal Firing Applications Woody biomass has been tested and used as a supplementary alternative fuel in large-scale utility boilers generating >126 kg/sec (1 106 lb/hr) high pressure/high temperature steam. Typical steam conditions employed in the generating stations where cofiring has been deployed are on the order of 124–166 bar/538–551 C/538–551 C (1,800–2,400 psig/1,000–1,025 F/ 1,000–1,025 F). Some supercritical boilers have been employed in cofiring testing as well [6, 8]. Among PC boilers, woody biomass—either sawdust or urban wood waste—has been cofired in the 140 MWe Albright Generating Station boiler #3 of Allegheny Energy Supply Co., LLC, the Seward Generating Station of GPU Genco (now Reliant Energy), the Shawville Generating Station of GPU Genco (now Reliant Energy), the Greenidge Station of New York State Electric and Gas Co. (now AES), the Kingston and Colbert Fossil Plants of TVA, Plant Kraft and Plant Hammond of Southern Co., and several other units [6, 8, 64]. The experiments by Southern Co. at Plant Hammond and Plant Kraft were among the original tests of cofiring wood waste with coal [65].
Characteristics of Alternative Fuels
107
In addition to PC and cyclone boilers, cofiring in CFB boilers will be a significant part of the renewable program of JEA (formerly Jacksonville Electric Authority) in Jacksonville, Florida [66]. Co-gasification of woody biomass in an IGCC installation has been tested by Tampa Electric Company at its Polk Power Station Unit #1 [67], and gasification-based cofiring in a PC boiler has been commercialized in Lahte, Finland [68]. Cofiring in utility-owned PC units capitalizes on the large capacity, the more efficient reheat generating cycles, the more effective use of labor, and the generally significantly improved economics associated with large central station power plants. Cofiring of woody biomass in such coal-fired boilers, then, capitalizes on the overall infrastructure and traditions of the generating stations currently in place. Cofiring woody biomass with coal also provides benefits to these plants in terms of improved environmental performance, with particular attention to reduced airborne emissions. In some cases, cofiring of woody biomass also provides specific operational benefits to the coal-fired generating stations. Cofiring woody biomass in coal-fired boilers is performed differently depending on the type of boiler; in PC boilers, woody biomass is typically injected separately from the coal unless very low percentages of woody materials are used (e.g., <5% by mass, or <2% by heat input). In cyclone boilers the woody biomass is typically blended with the coal and introduced into the boiler with the coal [6]. Benefits from cofiring woody biomass in PC boilers can readily be seen. While there is a modest decrease in boiler efficiency, there is no loss of boiler capacity. As was shown at Seward Generating Station, capacity lost to wet coal could be recovered by separate injection cofiring. The overwhelming benefits, however, were in the area of emissions reduction. SO2, NOx, and mercury emissions have been consistently reduced by cofiring woody biomass in coal-fired PC boilers. These reductions are dramatic and readily quantified as a function of capitalizing on the fuel characteristics of woody biomass.
3.3.5 Cofiring Woody Biomass in Cyclone Boilers Cofiring woody biomass in cyclone boilers has involved both sawdust and urban wood waste. Initial programs involved the King Station of Northern States Power outside St. Paul, Minnesota, cofiring sawdust from the Andersen Windows plant in a 600 MWe supercritical boiler fired by 12 cyclone barrels. Cofiring wood waste in cyclone boilers also included the Allen Fossil Plant of Tennessee Valley Authority, the Bailly and Michigan City Generating Stations of Northern Indiana Public Service Co., the Gannon Station of Tampa Electric Company, and the Willow Island Generating Station of Allegheny Energy Supply Co., LLC. Unlike cofiring woody biomass in PC boilers, cofiring sawdust or urban wood waste in cyclone boilers virtually always involves blending
108 Combustion Engineering Issues for Solid Fuel Systems
the biomass with the coal on the way to the bunkers rather than separately injecting the sawdust into the boiler. Only the King Station system involves separate injection of dry woody biomass into the boiler. There have been studies of reburn using woody biomass in cyclone boilers [69, 70]; however, this has not been commercially applied to date. Woody biomass can readily be cofired in cyclone boilers as has been demonstrated in numerous locations. Because cyclone boilers do not utilize a pulverized fuel, the biomass can be blended with the coal going to bunkers, and then the blend can be fired directly in the cyclone barrels. Like the PC boilers, the cyclone boilers experience no capacity impacts associated with cofiring. Although they experience increased feeder speeds, such feeder limitations are rare in cyclone firing. Efficiency impacts are minor. Environmental impacts are significant. Cofiring does not increase deposition in the boiler—either as slagging or as fouling. Like cofiring in PC boilers, and firing woody biomass in dedicated combustion systems, there is no direct evidence of the effect of woody biomass firing on SCR catalyst poisoning or deactivation. However, there remain questions resulting from tests with herbaceous biomass materials. With cofiring in coal-fired boilers, the presence of 10–20% biomass on a mass basis has the net effect of reducing the ash loading in kg ash/GJ of fuel (or lb ash/106 Btu of fuel). While it is projected that the influence of cofired woody biomass on SCR catalyst would be minor, if at all, that remains to be proven by research, which is now ongoing [71].
3.3.6 Conclusions Regarding Using Woody Biomass as an Alternative Fuel Typically, wood fuels—including spent pulping liquor—have been used in dedicated boilers. These boilers in the pulp and paper industry commonly support cogeneration applications, generating 58.6–100 bar (850–1,450 psig) and 450–540 C (850–1,000 F) main steam and exhausting low pressure 3.5–10.3 bar (50–150 psig) steam from backpressure or automatic extraction turbines for use in process applications. Woody biomass supports the generation of over 7,000 MW of electricity for the U.S. economy and also considerable electricity for northern European economies. Dedicated wood-fired boilers are typically capacity limited by plant logistics as well as by volumetric heat release rates. The largest of these boilers typically generate 75.7–88.3 kg/sec (600,000–700,000 lb/hr) of steam for turbine and process applications. These capacity limitations have economically precluded the construction of reheat boilers being fired by woody biomass. Consequently, their efficiency in the generation of electricity through condensing turbine applications is limited; typical NSHRs for the largest and most efficient wood-fired power plants are on
Characteristics of Alternative Fuels
109
the order of 13.7–14.8 MJ/kWh (13,000–14,000 Btu/kWh). NSHR values for many wood-fired power plants are on the order of 18.4 MJ/kWh (17,500 Btu/kWh). Cofiring woody biomass in PC and cyclone boilers used by electricitygenerating utilities has been employed selectively as a means for capitalizing on the improved efficiencies of large, high-pressure, reheat boilers. Utility boilers employing cofiring—either in test or in commercial deployment applications—have ranged in size up to 600 MWe, and have commonly been in the 140–270 MWe range. In deploying woody biomass in cofiring applications, utilities have found that the fuel characteristics of this class of fuels can be used advantageously. The high reactivity is most beneficial. The low nitrogen content of sawdust and the low sulfur and mercury contents of all woody biomass fuels are also most beneficial. The use of woody biomass in cofiring applications has not been shown to cause a decrease in boiler capacity. In PC firing, it has been shown that cofiring can help recover lost capacity when the unit is pulverizer limited and firing wet coal under winter conditions. Woody biomass cofiring does cause modest efficiency losses; however, these are readily managed. Operationally, most tests have shown that base combustion temperatures have not been compromised; however, furnace exit gas temperatures (FEGT) values have decreased. At the same time, main steam and reheat steam temperatures have not decreased. Cofiring woody biomass has reduced airborne emissions resulting from coal combustion in central station power plants. It has been shown that every tonne of woody biomass fired reduces the emission of fossil CO2 by >1 tonne directly, and by about 3 tonnes of total fossil CO2 equivalent [6, 8]. Because of the low sulfur and mercury content of woody biomass, SO2 and mercury emissions are also decreased. NOx emissions are reduced as a function of the fuel volatility, the low nitrogen content when firing sawdust, and by reduced FEGT values. Further, the reduction in oxides of nitrogen occurs without loss-on-ignition (LOI) penalty. Cofiring also can be integrated with low NOx SOFA systems, further decreasing these emissions. Figure 3-3 presents a compilation of test data from all U.S. Department of Energy (DOE)-sponsored demonstrations. If the influence of cofiring were strictly the substitution of a low nitrogen fuel for coal, then the data points should be on the line indicated. The reality is that there is significant evidence that cofiring woody biomass with coal can dramatically reduce NOx emissions, particularly if the boiler operations are designed to achieve this objective. In conclusion, cofiring of woody biomass with coal has shown itself as a significant new approach to the use of this alternative fuel. Such an approach complements the use of woody biomass in dedicated boilers and related facilities. It provides additional markets for these energy sources in the coming years and decades.
110 Combustion Engineering Issues for Solid Fuel Systems 30
25 Percent NOx Reduction
Line Indicates 1% Cofiring (Heat Input Basis) = 1% NOx Reduction from Baseline Test
20
40 Tests Above the Line
15
10
5 18 Tests Below the Line 0
0
2
4 6 8 Percent Cofiring, Btu Basis
10
12
FIGURE 3-3 NOx reduction as a function of cofiring—summary of all U.S. DOEsponsored demonstrations (Source: [56]).
3.4 Tire-Derived Fuel (TDF) Used tires from automobiles, trucks, tractors, and other mobile equipment provide an energy resource of significant interest to many utilities. Tires— and tire-derived fuel (TDF)—have a high calorific value along with other favorable fuel characteristics. At the same time, they present material preparation and handling issues for fuel users [72, 73]. For environmental reasons, they are increasingly difficult and costly to dispose in landfills. In 1990, only 25 million tires, or 11% of the annually generated scrap tires in the United States, were utilized (recycled, retreaded, and burned for energy). In 1994, this number increased to 138 million tires, or 55% of the annually generated scrap tires, with the largest increase due to tires used for energy (101 million tires) [74]. The trend toward increased use of tires in material and energy applications continues its upward path. However, with an estimated number between 1 and 3 billion tires in stockpiles throughout the United States, this potential energy source is enormous [75–77]. Significant quantities of carbon black (97% pure carbon) are used to improve the properties of the tire rubber. Typically, this carbon black is made in furnaces firing highly aromatic oil feedstocks free of coke and low in asphaltene content [78]. Carbon black reinforcement of rubber was
Characteristics of Alternative Fuels
111
discovered in 1912 and has traditionally been used to impart strength, shape, load-carrying capabilities, and bruise and fatigue resistance to the tires. Tire cords originally were cotton. Subsequently, rayon, nylon, polyester, glass, and steel became cord materials of choice. In the United States, steel is the dominant cord material [79]. Modern tires, then, are combinations of synthetic rubber compounds, carbon black, and cord material brought together into a high-strength product capable of significant load support and significant resistance to stress. Because of the materials used, they are highly reactive, exhibit significant aromaticity as a consequence of the carbon black, and have substantial concentrations of such metals as zinc, resulting from the vulcanization process. The cords or belting material may or may not increase the inorganic matter in the tires, depending on the cord material of choice. When such products wear out, as a consequence of use, they are available for conversion into fuel, crumb rubber, playground tire chips, paving material, or other uses.
3.4.1 General Description of Tire-Derived Fuel Tire-derived fuel is broadly classified as a fuel feedstock derived from automobile, truck, off-road, and specialty tires. These tires are either used whole or chipped into sized pieces for blending with other more conventional fuels such as coal. These tires may contain a significant amount of steel used in reinforcement of the sidewall and rim. Depending on production requirements, this steel may need to be removed from the tire. The great advantage of TDF as an alternative fuel is its high calorific value, low ash (if the steel is removed), and low-to-moderate sulfur content. This makes its usage in boilers attractive if it can be prepared and fed into the boiler in an economic manner. There are three basic types of tire-derived fuel in use today: TDF with steel, without steel, and crumb rubber. Tires come from predominantly passenger automobiles and trucks; however, on a per mass basis, tractor, sport utility vehicle, and commercial truck tires provide a significant amount of TDF [80]. Tire-derived fuel with steel implies utilizing the entire tire as is or shredded into sized pieces. There is essentially no processing done on these tires except that the steel rim may be removed. This fuel has the highest ash content and the lowest calorific value of the TDF products. This is the most economical from a processing perspective but can result in utilization problems with the additional iron present in the ash. Other utilization problems occur with materials handling. Steel wire, randomly liberated during the shredding process or in materials handling, can cause damage to coal handling belts, primary and secondary crushers, and to other elements of the feed system. Further, steel “whiskers” on the tire chips can cause significant bridging in material handling systems—bridging
112 Combustion Engineering Issues for Solid Fuel Systems
that is difficult to break up due to the interweaving of fuel particles. This coarse, shredded TDF is typically produced in the largest particle sizes. Common sizes are <38–50 mm (<1.5–2 in.). Periodically, it is shredded to smaller particle sizes as well. To make the TDF fuel more compatible with energy production facilities, it is desirable to remove the steel rim and bead material. This is usually accomplished by removing the rim and steel bead from the tire sidewall and magnetically removing the steel. While most of the steel is removed, some still remains within the tire chip and may present handling issues. Manufacturing wireless chips requires additional machinery. However, such chips are periodically produced at particle sizes of <15–25 mm (0.6–1 in.). Crumb rubber is TDF essentially without any steel. This is usually material chipped/ground to less than 6 mm and passed through magnetic separators to remove any remaining steel. This material has the highest calorific value, lowest ash content, but is the most expensive to process. It is mainly used in specialty applications but not for energy production.
3.4.2 Fuel Characteristics of Tire-Derived Fuel Tire-derived fuel used as a fuel supplement comes in three main forms. In some cases, whole tires have been injected into a boiler and burned. Others require that tire crumb be produced without any steel and less than about 6 mm size. The most common size for utilizing TDF is between 25 mm and 50 mm squares with the steel removed. The economics of chipping to finer sizes and removing the steel are very important in determining the long-term potential for TDF usage. 3.4.2.1 Proximate and Ultimate Analysis of Tire-Derived Fuel Table 3-17 lists typical proximate, ultimate, and calorific value analyses of TDF samples [81]. As expected, those samples with steel had a lower heating value and higher ash content than those without the steel. However, the calorific value, even for those samples with steel, is higher, in most cases, than coal. This is one of the attractive features of the TDF fuel; because of its higher calorific content, it can replace coal with actually less mass being fed into the boiler. Even more pronounced is the low ash content of the samples without steel. Sulfur contents are typical of eastern bituminous coals; however, on a kg/GJ basis the sulfur is lower than most coals—potentially resulting in lower emissions. Using these typical analyses, one can calculate several fundamental derived parameters. These parameters are listed in Table 3-18 and are used to estimate combustion propensity and reactivity. The values determined are the volatile-to-fixed carbon ratio, hydrogen/carbon ratio, oxygen/carbon ratio, and the kg sulfur/GJ heat input values.
Characteristics of Alternative Fuels
113
TABLE 3-17 Typical Analyses of TDF Parameter
TDF with Steel
TDF w/o Steel
Moisture Volatile Matter Fixed Carbon Ash
0.75 54.23 21.83 23.19
1.15 63.50 29.28 6.22
Carbon Hydrogen Nitrogen Sulfur Oxygen Ash
67.0 5.81 0.25 1.33 1.64 23.19
81.70 7.18 0.56 1.62 2.65 6.29
kJ/kg Btu/lb
31,058 13,362
36,681 15,781
Source: [81]
TABLE 3-18 Fundamental Derived Values for TDF Samples Parameter VM/FC H/C Atomic O/C Atomic kg S/GJ kg SO2/GJ
TDF with Steel
TDF w/o Steel
2.48 1.083 0.018 0.43 0.86
2.17 1.07 0.024 0.44 0.88
Note that the VM/FC ratio exceeds 2.0 for both TDF with steel and TDF without steel, and the H/C atomic ratios exceed unity. This is consistent with the structure of SBR, which is low in aromaticity. The structure of SBR has a calculated aromaticity of 50%—well below that of even the highly volatile subbituminous coals. Recognizing that TDF contains significant concentrations of carbon black, and that the carbon black is highly aromatic, the actual aromaticity of TDF is probably around 65%. This is consistent with the aromaticity of many western U.S. subbituminous coals and North Dakota lignites. 3.4.2.2 Ash Constituents of TDF Table 3-19 lists some of the major ash constituents in various TDF samples. Of particular note is the very high zinc content of the ash resulting from the vulcanization processes. As mentioned earlier, the emissions of TDF compared to coal can be either the same or lower with the exception
114 Combustion Engineering Issues for Solid Fuel Systems TABLE 3-19 Ash Constituents for TDF Samples Parameter (wt % of ash) Al2O3 SiO2 TiO2 Na2O K2O Fe2O3 MgO CaO ZnO Other
TDF with Steel
TDF w/o Steel
1.12 4.13 0.26 0.16 0.20 80.10 0.38 1.80 8.96 0.21
7.85 20.95 7.40 0.81 0.72 23.05 0.99 3.61 25.70 1.62
Source: [79]
of zinc emissions. It is also important to note that the iron content in the ash is high, even for the wireless tires. This iron has the potential to flux slag formation in cyclone boilers, particularly if the iron is finely divided (e.g., not in bead wire). 3.4.2.3 Trace Element Emissions from TDF The EPA Office of Air Quality Planning and Standards completed a study on trace element emissions from TDF burning facilities [22, 82]. The data were from 22 industrial facilities that have used TDF: 3 kilns (2 cement and 1 lime) and 19 boilers (utility, pulp and paper, and general industrial applications). In general, the results indicate that properly designed existing solid fuel combustors can supplement their normal fuels (coal, wood, and combinations of coal, wood, oil, petroleum coke, and sludge) with 10–20% TDF and still satisfy environmental compliance emission limits. In fact, dedicated tires-to-energy facilities indicate that it is possible to have emissions much lower than produced by existing solid-fuel-fired boilers (on a heat input basis), when properly designed. In a laboratory test program on controlled burning of TDF, with the exception of zinc emissions, potential emissions from TDF are expected to be very similar to those from other conventional fossil fuels, as long as combustion occurs in a well-designed, well-operated, and well-maintained combustion device. Selected results from this study are shown in Table 3-20.
3.4.3 Cofiring Applications with Tire-Derived Fuel Cyclone boilers are well suited for TDF applications because they easily handle chipped material and the ash is mainly taken out in fluid form. Thus, the additional steel is less of a problem in cyclones than in conventional PC boilers.
Characteristics of Alternative Fuels
115
TABLE 3-20 Trace Element Emissions from TDF Facility Emissions Element Lead Cadmium Chromium Mercury Copper Manganese Nickel Tin Beryllium Zinc
g/MJ
lb/MMBtu
5.5E-7 1.6E-6 2.0E-6 2.9E-7 3.2E-6 6.9E-7 2.7E-6 1.8E-6 3.1E-5 6.0E-4
1.3E-6 3.7E-6 4.7E-6 6.7E-7 7.5E-6 1.6E-6 6.3E-6 4.2E-6 7.3E-5 1.4E-3
Source: [82]
Stoker boilers are another good match for TDF cofiring, especially traveling grate stokers [83]. The TDF chips are easily blended with the coal and completely combusted in the available time on the grate. Wire in the TDF is a concern because it may melt when the rubber has burned and potentially plug the grate keys. Like cyclone boilers, PC boilers have the potential to use TDF as a fuel provided that a separate feed system is installed for the TDF chips. The high calorific value and low-to-medium sulfur content make it an attractive alternative fuel to be blended in the 5–10% (energy basis) range with coal. The constraints on PC cofiring of TDF are more significant than with cyclone firing. TDF cannot easily pass through a pulverizer and must therefore be injected separately from the coal. Further, reduction of TDF to particle sizes appropriate for suspension firing is prohibitively expensive.
3.4.4 Summary Regarding TDF as an Alternative Fuel There is a sufficient supply of scrap tires in most areas of the United States; this warrants their use as an alternative fuel to be blended with a primary fuel such as coal for use in the energy production arena. TDF has been fired in cyclone, stoker, and fluidized-bed boilers and in limited instances in pulverized coal boilers. These boilers can use up to 20–25% TDF on an energy basis. Due to the size and potential make-up of the TDF chips, care must be taken to avoid plugging mills, transfer chutes, and other handling areas. Optimally, the steel should be removed from the TDF to reduce these handling concerns. As a final precaution, most users of TDF for cofiring applications perform TDF preparation on-site to monitor and maintain product consistency.
116 Combustion Engineering Issues for Solid Fuel Systems
TDF blends have not resulted in any major boiler operating problems if proper sized (<25 mm) chips were utilized. Full load was achievable at reasonable excess air levels, and boiler efficiencies were equal to or slightly greater with the TDF blends due to its higher calorific value and low moisture content. Gaseous emissions do not appear to be negatively impacted unless a low sulfur coal is being blended with the TDF; then the SO2 emissions might rise slightly. NOx emissions tend to be reduced with TDF cofiring, and opacity appears unchanged or slightly lower with the TDF. CO and hydrocarbon emissions can increase when cofiring due to insufficient time to completely burn out large particles. This can be mitigated to some extent by limiting the size of the tire chip. Zinc emissions are higher with the TDF blends but not above the regulated limits.
3.5 Herbaceous Crops Herbaceous biomass fuels—agricultural and agribusiness wastes, herbaceous crops such as switchgrass, and the like—are fundamentally different from woody biomass fuels and must be treated separately. Although bagasse resulting from cane sugar processing has long been a fuel source to that industry, these fuels are currently of modest consequence in the United States. They are becoming potentially significant sources of energy in numerous countries from Denmark to China. Further, herbaceous energy sources such as crop wastes and vineyard wastes have been used substantially in such states as California. Many of the herbaceous biofuels such as alfalfa residues can be very high in nitrogen [84]. Herbaceous crops, as fuel, can also contain significant concentrations of chlorine [84]. The herbaceous materials such as switchgrass can have very low bulk densities and can present serious material handling and logistics issues as well [8, 85]. The consequence of all these factors contributes to the conclusion that herbaceous materials are potentially significant as energy sources and that these are a separate class of alternative fuels.
3.5.1 Types of Herbaceous Biomass Fuels Several types of herbaceous biomass fuels are used, or proposed, as alternative fuels. These include crops grown specifically for fuel purposes [36, 86] and a potential long-term energy resource throughout the United States [87]. Utilities such as JEA, for example, have made investments in research designed to identify the most suitable herbaceous crops and to quantify the yields of such crops in energy terms [88]. China uses a substantial amount of herbaceous material for energy. Crop residues from the production of rice, corn, and wheat are currently used extensively for energy; over 40% of the biomass energy generated in China comes from herbaceous sources [89]. Further, this consumption of
Characteristics of Alternative Fuels
117
herbaceous crops is projected to grow significantly within the Chinese economy in the next decade [89]. Numerous researchers have documented the fundamental differences between herbaceous biomass and woody biomass (see, for example, WieckHansen et al. [90], Miles et al. [91], Tillman [6]). The herbaceous materials are significantly higher in ash than their woody counterparts; and this ash has significant slagging and fouling characteristics as noted by researchers at the Oak Ridge National Laboratory (ORNL) and other institutions (see, for example, [92–95]). The most prominent of these is switchgrass, although Giant Reed has also been discussed [88] along with miscanthus, Reed Canary grass, and other crops. Switchgrass is the prominent crop chosen for evaluation by ORNL researchers and by other institutions. Field crop residues also have been considered significantly. Field crop residues in Europe, such as straw, are commonly considered [90, 96] and used as energy sources; Denmark, particularly, focuses on straw burning. Field crop residues such as alfalfa stalks not used in animal feed production have been considered in the United States, along with corn stover, rice straw, and other residues from the growing and harvesting of grains [91, 97, 98]. These residues have long been evaluated as fuels despite difficulties in gathering, concentrating, and transporting them. Production materials and residues also have been long considered in the herbaceous fuels arena. These materials include a wide variety of potential fuels: bagasse from sugar cane processing, out-of-date corn seed, corn cobs, rice hulls, oat hulls, nut hulls (e.g., walnut hulls) and stone fruit pits (e.g., peach pits, olive pits), vineyard trimmings, orchard trimmings, tomato pummace from the manufacture of catsup, winery pummace from the squeezing of grapes, and a host of other locally available agribusiness wastes [98]. Cotton gin trash has also been used for energy purposes in the United States [99].
3.5.2 Sources and Uses of Herbaceous Materials Herbaceous materials can be obtained from farms growing crops—either as foodstuffs or as fuel—or from a variety of agribusinesses. These agribusinesses include food processors from the grain processing industry, from the animal food preparation industry, from fruit canneries and packagers, vineyards, vintners, processed food manufacturers, and a host of other organizations. In recent years there have been numerous efforts to gather farmers together into cooperatives and other organizations within the United States to produce sufficient material to support fuel applications. These organizations have promoted projects such as the proposed alfalfa stalk gasification efforts in Minnesota [100] and the ongoing efforts in Chariton Valley, Iowa, to grow switchgrass for cofiring in the Ottumwa Generating Station of Alliant Energy [101]. Similar efforts are also underway to promote
118 Combustion Engineering Issues for Solid Fuel Systems
the use of switchgrass for cofiring in the Southeast, principally at Southern Company generating stations [6, 87]. Since the production of these materials has been ongoing for decades, producers of herbaceous residues have developed other markets for some of the materials. Such markets include animal bedding, paper and low-grade board manufacturing, and opportunity fuel applications. Field crop wastes are frequently plowed under or burned, depending on local regulations. In developing countries, field crop wastes are gathered for use in domestic fuel applications as well as small industrial applications.
3.5.3 Fuel Characteristics of Switchgrass and Related Agricultural Biomass Materials Fuel characteristics of importance include fuel density, proximate and ultimate analysis, chemical structure and reactivity, and inorganic constituents and chemistry [102]. These characteristics are somewhat similar to—yet quite distinct from—the compositional aspects of woody biomass materials previously described. 3.5.3.1 Density of Switchgrass and Related Materials Switchgrass and the related herbaceous materials are a very low density, modest heat content set of fuels. For the field crops such as switchgrass, bulk densities are less than 0.064 kg/m3 (5 lb/ft3) loose, and about 0.103 kg/m3 (8 lb/ft3) when baled [6]. These values compare to typical coal densities on the order of 0.666 kg/m3 (52 lb/ft3). The values for switchgrass are comparable to loose and baled straws, hay, and other field materials. Densities of vineyard prunings, stone fruit pits, and related products are higher—approaching the densities of woody biomass [97]. Table 3-21 presents typical bulk densities of switchgrass and other agricultural materials.
TABLE 3-21 Bulk Densities of Switchgrass and Other Agricultural Materials Material Baled switchgrass Cotton gin trash Peach pits Rice hulls Walnut shells Peanut shells Vineyard prunings Hybrid corn seed
Bulk Density (kg/m3)
Bulk Density (lb/ft3)
0.102 0.078 0.384 0.232 0.426 0.176 0.170 0.320
8.0 6.1 30.0 18.1 33.3 14.2 13.3 25.0
Note: The densities of all but the switchgrass are for fuels produced at <2 mm particle sizes. Source: [6, 97, 102]
Characteristics of Alternative Fuels
119
Note in Table 3-21 that a wide diversity of agricultural materials is considered. All these materials have been considered for fuel applications by industries, independent power producers (IPPs), and utilities in the United States and Europe. Many of these alternative fuels have been used in small power plants within the state of California or the Southeast and related agricultural areas. Like the woody biofuels, the herbaceous biomass forms are porous solids, with significant fractional void volumes. Unlike the woody biofuels, however, the agricultural materials contain significant concentrations of inorganic matter, making the calculation of porosity less precise. 3.5.3.2
Proximate and Ultimate Analysis of Switchgrass and Related Agricultural Materials The proximate and ultimate analyses for switchgrass and related herbaceous materials are shown in Table 3-22. Note that the moisture contents for such materials are typically quite low relative to most biomass fuels. As shown, there is significant variability in moisture, ash, and nitrogen content. The cotton gin trash, vineyard prunings, and rice hulls are particularly high in nitrogen content. The higher heating values of the residues shown in Table 3-22, with the exception of out-of-date hybrid corn seed, are substantially lower than those associated with woody biomass or coal, and those associated with switchgrass grown for energy purposes. Much of this can be attributed to the higher ash contents associated with herbaceous materials—particularly residues from agribusinesses. The density and heat content of the hybrid corn seed has made it a desirable herbaceous alternative fuel when conditions are favorable. The proximate and ultimate analyses lead to derived ratios and values that relate specifically to system performance, including measures of reactivity and measures of pollution potential. The measures of emission potential include kg nitrogen and sulfur (or SO2)/GJ (or expressed in lb/106 Btu), and kg ash or inorganic matter/GJ of fuel. These measures of pollution potential are shown in Table 3-23. Note that many of the herbaceous fuels are not totally sulfur free, and most are high in nitrogen. The high nitrogen content is expected based on the fertilization practices in the agricultural community. These practices also contribute to the inorganic matter compositions of herbaceous materials—particularly the concentrations of potassium (K) and phosphorous (P) in the ash of herbaceous material. The significant variability in the herbaceous and agricultural materials used as alternative fuels makes their use potentially more challenging. Not shown in these data are concentrations of chlorine; however, chlorine concentrations can be significant and problematical in the herbaceous materials. Typical values for switchgrass are 0.08–0.16% (800–1,600 mg/kg, or ppmw, dry basis) [91, 104], while rice straw can contain as much as 0.50% (5,000 mg/kg) [104] and hybrid seed corn can contain 0.078% chlorine (780 mg/kg) [104]. This chlorine can present significant operational problems in terms of corrosion and contribute to environmental concerns.
120 TABLE 3-22 Proximate and Ultimate Analyses for Switchgrass and Selected Herbaceous Materials as Fuel Herbaceous Fuel Parameter
Fresh Switchgrass
Weathered Switchgrass
Moisture % 15 15 Proximate Analysis (wt % dry basis) Volatiles 76.18 81.8 Fixed Carbon 16.08 14.8 Ash 7.74 3.4 Ultimate Analysis (wt % dry basis) Carbon 46.73 49.4 Hydrogen 5.88 5.9 Nitrogen 0.54 0.4 Sulfur 0.13 0.3 Oxygen 38.99 40.6 Ash 7.74 3.4 Higher Heating Value (dry basis) MJ/kg 18.04 18.97 Btu/lb 7,750 8,150 *High ash content due to embedded dirt in sample. Source: [97, 102–104]
Reed Canary Grass
Mulch Hay
Cotton Gin Trash
Rice Hulls
Vineyard Prunings
65.2
19.5
7–12
7–10
20–40
76.1 19.8 4.1
77.6 17.1 5.3
75.4 15.4 9.2
63.6 15.8 20.6
45.8 6.1 1.0 0.1 42.9 4.1
46.5 5.7 1.7 0.2 40.6 5.3
42.77 5.08 1.53 0.55 35.38 14.69
16.54 7,103
18.76 8,058
15.60 6,700
Hybrid Seed Corn
Bagasse
12
45
74.9 14.3 10.8*
78.32 20.15 1.53
86.62 11.95 2.44
38.30 4.36 0.83 0.06 35.45 21.00
47.99 5.65 0.86 0.08 39.61 5.81
46.00 6.13 1.84 0.18 44.32 1.53
48.64 5.87 0.16 0.04 42.82 2.44
14.90 6,400
16.81 7,220
19.14 8,222
19.01 8,165
Characteristics of Alternative Fuels
121
TABLE 3-23 Pollution Potential Measures from the Ultimate Analyses of Herbaceous Materials Pollution Potential Measure: kg/GJ (lb/106 Btu) Fuel
Fuel Nitrogen
Fuel Sulfur*
Fuel Ash
Fresh Switchgrass Weathered Switchgrass Reed Canary Grass Mulch Hay Cotton Gin Trash Rice Hulls Vineyard Prunings Hybrid Corn Seed Bagasse
0.300 (0.697) 0.211 (0.491) 0.605 (1.408) 0.907 (2.110) 0.982 (2.284) 0.558 (1.297) 0.512 (1.191) 0.962 (2.238) 0.086 (0.20)
0.072 (0.168) 0.158 (0.368) 0.061 (0.141) 0.107 (0.248) 0.353 (0.821) 0.040 (0.094) 0.048 (0.111) 0.094 (0.219) 0.017 (0.04)
4.294 (9.987) 1.794 (4.172) 2.482 (5.772) 2.828 (6.577) 9.428 (21.93) 14.11 (32.81) 3.460 (8.047) 0.800 (1.861) 0.55 (2.98)
*SO2 values in kg/GJ or lb/106 Btu are double the fuel sulfur values. Source: [6, 98, 100–102, 104]
3.5.3.3 Ash Chemistry for Herbaceous Biomass Fuels The ash chemistry associated with herbaceous biomass is particularly significant when considering these opportunity fuels. Research in the United States [91, 102, 105–107] and in Europe [90, 96] has demonstrated that these herbaceous fuel materials—high in ash content and high in alkalinity— present unique issues associated with utilization of these opportunity fuels. Ash deposition has been evaluated extensively in Europe as a consequence of programs designed to fire herbaceous biomass fuels—particularly straws, miscanthus, and related materials [90, 108–111]. When cofired in fluidized beds, herbaceous biomass fuels can cause bed agglomeration, hot spots, and potentially defluidization; in stokers such materials can create large deposits impeding effective air distribution through the grate, whereas in pulverized coal boilers such materials can cause excessive slagging and fouling [110]. Traditional ash analyses result in ash elemental analyses such as those shown in Table 3-24. Miles et al. [105] proposed a rule of thumb for evaluating the slagging and fouling potential of biomass fuels based on ash elemental analyses as shown previously. Fuels with alkali (Na2O and K2O) concentrations below 0.172 kg/GJ (0.4 lb/106 Btu) have only a slight opportunity to experience slagging and fouling, whereas fuels with alkali concentrations >0.344 kg/ GJ (0.8 lb/106 Btu) are certain to experience significant slagging and fouling. Table 3-25 provides such indices for selected herbaceous biomass materials. A more precise and definitive approach to biomass ash characterization has been developed and employed by researchers at the Energy Institute of Pennsylvania State University [106, 107], Sandia National Laboratories [91], and numerous commercial and industrial laboratories. This procedure involves successive leaching of the fuel sample in distilled water, ammonium acetate, and hydrochloric acid. The most reactive
122 Combustion Engineering Issues for Solid Fuel Systems TABLE 3-24 Ash Analyses of Switchgrass and Other Herbaceous Crops Fuel Parameter
Fresh Switchgrass
Weathered Switchgrass
Alfalfa Stems
Wheat Straw
Rice Straw
65.18 4.51 0.24 2.03 5.60 3.00 0.58 11.60 4.50 0.44 0.00 0.33
65.42 6.98 0.34 3.56 7.14 3.17 1.03 7.00 2.80 2.00 0.00 0.30
1.44 0.60 0.05 0.25 12.90 4.24 0.61 40.53 7.67 1.60 17.44 28.00
55.70 1.80 0.00 0.70 2.60 2.40 0.90 22.80 1.20 1.70 0.00 0.51
73.00 1.40 0.00 0.60 1.90 1.80 0.40 13.50 1.40 0.70 0.00 0.24
SiO2 Al2O3 TiO2 Fe2O3 CaO MgO Na2O K2O P2O5 SO3 CO2 Base/Acid Ratio Source: [20, 91, 111]
TABLE 3-25 Slagging and Fouling Index for Selected Herbaceous Biomass Fuels Fuel Wheat Straw Sunflower Hulls Alfalfa Stems Fresh Switchgrass Weathered Switchgrass Reed Canary Grass Rice Straw Bagasse*
Miles Slagging and Fouling Index kg/GJ (lb/106 Btu) of K2O and Na2O
Deposition Propensity
1.33 (3.10) 2.51 (5.83) 1.45–1.97 (3.38–4.59) 0.34–0.52 (0.80–1.22) 0.14–0.22 (0.33–0.51) 0.51 (1.18) 0.78 (1.82) 0.064 (0.148)
High High High High Low–Medium High High Low
*Bagasse is a washed product; all the most reactive ash elements are leached out in the sugar cane processing activities. Source: [84, 100, 105, 106, 111]
materials leach in distilled water and ammonium acetate. These include salts of potassium and sodium, which readily vaporize when subjected to high temperatures associated with combustion systems [108]. These also include highly reactive calcium compounds. Carbonates and sulfates are commonly found in the acid soluble fraction, while silicates and sulfides— quite unreactive materials—are typically found in the residual. The chemical fractionation can then be enhanced by a number of other techniques: scanning electron microscopy (SEM) or computer-controlled scanning electron microscopy (CCSEM), X-ray fluorescence (XRF), atomic
Characteristics of Alternative Fuels
123
absorption spectroscopy (AAS), and a variety of other techniques. Of these techniques, chemical fractionation has shown the most immediate benefit in analyzing the inorganic fraction of biomass fuels. Trace metal concentrations in herbaceous biomass materials are the final area of concern, particularly with the emphasis on mercury emissions management and the possible concern for SCR catalyst deactivation or poisoning as a result of cofiring straws and other herbaceous biomass fuels [90, 96]. The database for herbaceous materials is not extensive. Table 3-26 presents a general range of values. Trace metals in herbaceous crops are typically a consequence of fertilizer practices, and economic activities in the area. Commercial fertilizers can contain as much as 4,570 mg/kg Cr, 430 mg/kg Pb, and 3,000 mg/kg Zn [44, 112]. Further, the presence of a large source of airborne metals (e.g., a metals smelter) can cause elevated levels of trace metals in herbaceous biomass. Significantly, it is the arsenic concentration that can cause poisoning of the SCR catalyst, along with the possibility of catalyst blinding from the alkali metals. The arsenic concentrations in herbaceous biomass ash are typically lower than those associated with western low-rank coals and lignites [45].
3.5.4 Herbaceous Crop Summary The herbaceous alternative fuels are similar to, yet distinctly different from, the woody biomass fuels. They are commonly used as energy sources in Europe and in China; and they are less commonly used as energy sources in North America. They are lower in bulk density, higher in inorganic TABLE 3-26 Trace Metal Concentrations in the Ash from Agricultural Materials (mg/kg) Metal Antimony Arsenic Barium Beryllium Cadmium Chromium Cobalt Copper Lead Mercury Nickel Selenium Vanadium Zinc Source: [22, 45, 55]
Minimum
Maximum
10 3.4 41 0.01 0.36 11 2.9 14 21 BDL 4.4 BDL 11 40
10 12 220 0.06 1.1 20 14 31 55 BDL 5.8 BDL 20 190
124 Combustion Engineering Issues for Solid Fuel Systems
matter, and more prone to slagging and fouling than the woody biomass fuels. Further, it is suspected that these biomass fuels present more significant problems for SCR catalysts than the woody biomass fuels due to trace metal concentrations. Demonstrations of firing and cofiring herbaceous biomass fuels have been conducted extensively in both Europe and the United States. These demonstrations have generally shown the herbaceous biomass energy sources to be significantly more difficult and costly than other alternative fuels available; however they have demonstrated that herbaceous biomass firing can be accomplished in response to regulatory or incentive-driven markets.
3.6 References 1. Letheby, K. 2002. Utility Perspectives on Opportunity Fuels. Proc. 27th International Technical Conference on Coal Utilization and Fuel Systems. Clearwater, FL. March 4–7. 2. Letheby, K. 2002. Utility Utilization of Opportunity Fuels. Proc. International Joint Power Generation Conference. Phoenix, AZ. June 22–25. 3. Miller, B.G., S. Falcone Miller, R. Cooper, J. Gaudlip, M. Lapinsky, N. Raskin, T. Steitz, and J.J. Battista. 2003. Feasibility Analysis for Installing a Circulating Fluidized Bed Boiler for Cofiring Multiple Biofuels and Other Wastes with Coal at Penn State University Final Report. Work Performed Under Grant No. DE-FG26-00NT40809. 4. Miller, B.G., S. Falcone Miller, and A.W. Scaroni. 2002. Utilizing Agricultural By-products in Industrial Boilers: Penn State’s Experience and Coal’s Role in Providing Security for Our Nation’s Food Supply. Proc. 19th Annual International Pittsburgh Coal Conference, Pittsburgh, PA September 23–27. 5. Tillman, D.A., and N.S. Harding. 2004. Fuels of Opportunity: Characteristics and Uses in Combustion. Burlington, MA: Elsevier. 6. Tillman, D.A. 2002. Cofiring Technology Review, Final Report. National Energy Technology Laboratory, U.S. Department of Energy, Pittsburgh, PA. 7. U.S. Energy Information Agency. 2003. Annual Energy Review 2001. Washington, D.C.: USEIA. 8. Tillman, D.A. 2001. Final Report: EPRI-USDOE Cooperative Agreement: Cofiring Biomass with Coal. Contract No. DE-FC22-96PC96252. EPRI, Palo Alto, CA. 9. Roskill Consulting Group. 1999. The Economics of Petroleum Coke. Roskill Reports on Metals and Minerals. 10. Swanekamp, R. 2002. Return of the Supercritical Boiler. Power. 146(4): 32–40. 11. Federal Energy Regulatory Commission. 2001. FERC Form 423, Monthly Report of Cost and Quality of Fuels for Electric Plants. Washington, D.C. 12. Bryers, R. 1994. Utilization of Petroleum Coke and Petroleum Coke/Coal Blends as a Means of Steam Raising. Coal-Blending and Switching of Low-Sulfur Western Coals. New York: ASME. pp. 185–206.
Characteristics of Alternative Fuels
125
13. Bryers, R.W. 1995. Utilization of Petroleum Coke and Petroleum Coke/Coal Blends as a Means of Steam Raising. Fuel Processing Technology. 44: 121–141. 14. Narula, R.G. 2002. Challenges and Economics of Using Petroleum Coke for Power Generation. London: World Energy Council. 15. Heintz, E.A. 1996. Review: The Characterization of Petroleum Coke. Carbon. 34(6): 699–709. 16. Heintz, E.A. 1995. Effect of Calcination Rate on Petroleum Coke Properties. Carbon. 33(6): 817–820. 17. Lapades, D.N. (ed.). 1976. Encyclopedia of Energy. New York: McGraw-Hill. 18. Silva, A.C., C. McGreavy, and M.F. Sugaya. 2000. Coke Bed Structure in a Delayed Coker. Carbon. 38: 2061–2068. 19. Tillman, D., and P. Hus. 2000. Blending Opportunity Fuels with Coal for Efficiency and Environmental Benefit. Proc. 25th International Technical Conference on Coal Utilization and Fuel Systems. Coal Technology Association. Clearwater, FL. March 6–9. pp. 659–670. 20. Johnson, D.K., S.V. Pisupati, R.S. Wasco, and B.G. Miller. 2001. Pyrolysis and Char Oxidation Kinetics. Prepared for Foster Wheeler Development Corporation as a Subcontract to the Electric Power Research Institute Biomass Cofiring Project. Energy Institute, Pennsylvania State University, State College, PA. 21. Edwards, L.O. et al. 1980. Trace Metals and Stationary Conventional Combustion Sources, vol. 1: Technical Report. Austin, TX: Radian Corporation. For USEPA, Contract No. 68-02-2608. 22. Tillman, D.A. 1994. Trace Metals in Combustion Systems. San Diego: Academic Press. 23. Tillman, D.A. 2002. Petroleum Coke as a Supplementary Fuel for Cyclone Boilers: Characteristics and Test Results. Proc. International Joint Power Generation Conference. Phoenix, AZ. June 24–28. Paper 2002-26157. 24. Pearce, R., and J. Grusha. 2001. Tangential Low NOx System at Reliant Energy’s Limestone Unit #2 Cuts Lignite, PRB, and Pet Coke NOx. Proc. EPRI-DOE-EPA Combined Power Plant/Air Pollution Control Symposium. Clearwater, FL. August 20–24. 25. Abdulally, I.F., and K.A. Reed. 1994. Experience Update of Firing Waste Fuels in Foster Wheeler’s Circulating Fluidized Bed Boilers. Proc. Power-Gen Asia ’94. Hong Kong. August 23–25. 26. Castro, A.L., and P.K. Chelian. 1996. Application of Foster Wheeler CFB Boilers to Burn Vacuum Residuals for Low Cost Power with Low Emissions. Proc. Mexico Power ’96. Monterrey, Mexico. October 8–10. 27. Dyr, R.A., A.L. Compaan, J.L. Hebb, and S.L. Darling. 2000. The JEA Northside Repowering Project: Low Cost Power and Low Emissions with CFB Repowering. Proc. PowerGen-2000. Orlando, FL. November 14–16. 28. Anthony, E.J., A.P. Iribarne, J.V. Iribarne, R. Talbot, L. Jia, and D.L. Granatstein. 2001. Fouling in a 160 MWe FBC Boiler Firing Coal and Petroleum Coke. Fuel. 80: 1009–1014.
126 Combustion Engineering Issues for Solid Fuel Systems 29. Conn, R.E. 1995. Laboratory Techniques for Evaluating Ash Agglomeration Potential in Petroleum Coke Fired Circulating Fluidized Bed Combustors. Fuel Processing Technology. 44: 95–103. 30. Enzer, H., W. Dupree, and S. Miller. 1975. Energy Perspectives: A Presentation of Major Energy and Energy Related Data. Washington, D.C.: U.S. Department of the Interior. 31. Tillman, D. 1978. Wood as an Energy Resource. New York: Academic Press. 32. Walker, J.E. 1966. Hopewell Village: The Dynamics of a Nineteenth Century Iron-Making Community. Philadelphia, PA: University of Pennsylvania Press. 33. National Materials Advisory Board. 1975. Problems and Legislative Opportunities in the Basic Materials Industries. Washington, D.C.: National Research Council, National Academy of Sciences. 34. Energy Information Agency. 2002. Annual Energy Review. Washington, D.C.: U.S. Department of Energy. 35. Henry, J. 1979. The Silvicultural Energy Farm in Perspective. Progress in Biomass Conversion, vol. 1. New York: Academic Press. pp. 215–256. 36. College of Forest Resources, University of Washington. 1979. Energy from Wood: A Report to the Office of Technology Assessment, Congress of the United States. Energy from Biological Processes, vol. III.—Appendixes. Washington, D.C.: Office of Technology Assessment. 37. Committee on Renewable Resources for Industrial Materials. 1976. The Potential of Lignocellulosic Materials for the Production of Chemicals, Fuels, and Energy. Washington, D.C.: National Research Council, National Academy of Sciences. 38. Tillman, D. 1982. The Cost of Electricity from Silvicultural Fuel Farm-Based Power Plants. In Energy from Forest Biomass (Smith, W., ed.). New York: Academic Press. pp. 253–274. 39. Ostlie, L.D. 1993. Whole Tree Energy Technology and Pilot Test Program. Proc. Strategic Benefits of Biomass and Waste Fuels Conference. EPRI, Palo Alto, CA. March 30–April 1. 40. Robertson, T., and H. Shapouri. 1993. Biomass: An Overview in the United States of America. Proc. First Biomass Conference of the Americas. Burlington, VT. August 30–September 2. pp. 1–17. 41. Hughes, E. 2000. Biomass Cofiring: Economics, Policy and Opportunities. Biomass and Bioenergy. 19(6): 457–466. 42. Forest Products Laboratory. 1974. Wood Handbook: Wood as an Engineering Material. Washington, D.C.: US Government Printing Office. 43. Haygreen, J., and J. Bowyer. 1982. Forest Products and Wood Science. Ames, IA: Iowa State University Press. 44. Tillman, D. 1991. The Combustion of Solid Fuels and Wastes. San Diego, CA: Academic Press. 45. Prinzing, D. 1996. EPRI Alternate Fuels Database. Palo Alto, CA: EPRI. Report TR-107602. 46. Tillman, D.A. 1999. Biomass Cofiring: Field Test Results. Palo Alto, CA: Electric Power Research Institute. Report TR-113903.
Characteristics of Alternative Fuels
127
47. Sjostrom, E. 1981. Wood Chemistry: Fundamentals and Applications. New York: Academic Press. 48. Wenzl, H. 1970. The Chemical Technology of Wood. New York: Academic Press. 49. Johnson, D.K., D.A. Tillman, B.G. Miller, S.V. Pisupati, and D.J. Clifford. 2001. Characterizing Biomass Fuels for Cofiring Applications. Proc. Joint International Combustion Symposium. American Flame Research Committee. Kauai, Hawaii. September 9–12. 50. Tillman, D., and K. Hood. 1990. Incineration of Pulp Mill Sludge in High Efficiency Power Boilers. AIChE Annual Meeting, San Diego, CA. August 15–17. 51. Envirosphere Co. 1984. Final Biomass Ash Study. Prepared for the California Energy Commission. Contract No.: 500-81-037. 52. Greene, W.T. 1988. Wood Ash Disposal and Recycling Sourcebook. Beaverton, OR: OMNI Environmental Services, Inc. 53. Payette, K., T. Banfield, T. Nutter, and D. Tillman. 2002. Emissions Management at Albright Generating Station through Biomass Cofiring. Proc. 27th International Technical Conference on Coal Utilization and Fuel Systems. Coal Technology Association. Clearwater, FL. March 4–7. pp. 177–186. 54. Junge, D.C. 1975. Boilers Fired with Wood and Bark Residues. Research Bulletin 17. Corvallis, OR: Forest Research Laboratory, Oregon State University. 55. Junge, D.C. 1979. Design Guideline Handbook for Industrial Spreader Stoker Boilers Fired with Wood and Bark Residue Fuels. Corvallis, OR: Oregon State University Press. 56. Stultz, S.C., and J.B. Kitto (eds). 1992. Steam: Its Generation and Use, 40th ed. Barberton, OH: Babcock & Wilcox. 57. Tillman, D., A. Rossi, and W. Kitto. 1981. Wood Combustion: Principles, Processes, and Economics. New York: Academic Press. 58. Villesvik, G., and D.A. Tillman. 1983. Cofiring of Dissimilar Solid Fuels: A Review of Some Fundamental and Design Considerations. American Power Conference, April. 59. Irving, J. 1993. Wood Fired Power Plant Experience at the 50 MW McNeil Station. Proc. Strategic Benefits of Biomass and Waste Fuels. Washington, D.C. March 30–April 1. Palo Alto, CA: EPRI. 60. Baxter, L.L. et al. 1996. The Behavior of Inorganic Material in Biomass-Fired Power Boilers—Field and Laboratory Experiences: Vol II of Alkali Deposits Found in Biomass Power Plants. SAND96-8225 Volume 2 and NREL/TP433-8142. 61. Baxter, L.L. et al. 1996. The Behavior of Inorganic Material in Biomass-Fired Power Boilers: Field and Laboratory Experiences. Proc. Biomass Usage for Utility and Industrial Power. Engineering Foundation Conference. Snowbird, UT. April 28–May 3. 62. Wiltsee, G.A. 1994. Biomass Energy Fundamentals, Vol 2: System Characteristics. Palo Alto, CA: Electric Power Research Institute. 63. Bergesen, C., and J. Crass (eds). 1996. Power Plant Equipment Directory, 2nd ed. Washington, D.C.: Utility Data Institute.
128 Combustion Engineering Issues for Solid Fuel Systems 64. Battista, J., and E. Hughes. 2001. Biomass Cofiring in the United States: A Survey of Past Tests. Proc. 26th International Technical Conference on Coal Utilization and Fuel Systems. Clearwater, FL. March 5–8. pp. 227–236. 65. Boylan, D. 1993. Southern Company Tests of Wood/Coal Cofiring in Pulverized Coal Units. Proc. Strategic Benefits of Biomass and Waste Fuels. Washington, D.C. March 30–April 1. Palo Alto, CA: EPRI. 66. King, J. 2002. Presentation Concerning the Renewable Program of JEA. Biomass Interest Group Meeting, November 7-8. Washington, D.C. 67. Tampa Electric Company. 2002. Biomass Test Burn Report: Polk Power Station Unit 1. Report to NETL-USDOE. April. 68. Raskin, N., J. Palonen, and J. Nieminen. 2001. Power Boiler Fuel Augmentation with a Biomass Fired Atmospheric Circulating Fluid-Bed Gasifier. Biomass and Bioenergy. 20(1): 471–481. 69. Shafizadeh, F., and W. DeGroot. 1977. Thermal Analysis of Forest Fuels. In Fuels and Energy from Renewable Resources (Tillman, D.K., Sarkanen, and L. Anderson, eds). New York: Academic Press. pp. 93–114. 70. Harding, N.S., and B.R. Adams. 2000. Biomass as a Reburning fuel: A Specialized Cofiring Application. Biomass and Bioenergy. 19(6): 429–446. 71. Plasynski, S.I., R. Costello, E. Hughes, and D. Tillman. 1999. Biomass Cofiring in Full-Sized Coal-Fired Boilers. Proc. 24th International Technical Conference on Coal Utilization and Fuel Systems. ASME-FACT, USDOE, and Coal Technology Association. Clearwater, FL. March 8–11. pp. 281–292. 72. McGowin, C.R., 1991. Alternate Fuel Cofiring with Coal in Utility Boilers. EPRI Proceedings: 1991 Conference on Waste Tires as a Utility Fuel, EPRI GS-7538. 73. Winslow, J., J. Ekmann, S. Smouse, M. Ramezan, and N.S. Harding. 1996. Cofiring of Coal and Waste. International Energy Association Report, IEACR/90. 74. Harding, N.S., 2002. Cofiring Tire-Derived Fuel with Coal. 27th International Technical Conference on Coal Utilization and Fuel Systems, Clearwater, FL. 75. Niemeyer, S., 1977. Managing Wastes: Tires. Nebraska Cooperative Extension NF94-197. 76. Waste Tire Perspective. 1997. Louisiana Department of Environmental Quality. 77. Environmental Waste International. 2001. Scrap Tire Supply. Advanced Technology for Specialty Waste Streams. 78. Dannenberg, 1985. Kirk-Othmer Concise Encyclopedia of Chemical Technology, New York: John Wiley & Sons, Inc. 79. Skolnik, L., 1985. Tire Cords. In Concise Encyclopedia of Chemical Technology (Kirk-Othmer, ed.) New York: John Wiley & Sons, Inc. 80. Goodyear Tire and Rubber Company. 2002. Scrap Tire Recovery. 81. Granger, J.E., and G.A. Clark. 1991. Fuel Characterization of Coal/Shredded Tire Blends. EPRI Proceedings: 1991 Conference on Waste Tires as a Utility Fuel, EPRI GS-7538. 82. Reisman, J.I., and P.M. Lemieux. 1997. Air Emissions from Scrap Tire Combustion. EPA Report No. EPA-600/R-97-115. 83. Harding, N.S., and W.D. Owens. 1994. The Utilization of Waste Tire-Derived and Railroad-Tie-Derived Fuels in Coal-fired Stoker Boilers. 10th Annual
Characteristics of Alternative Fuels
84.
85.
86. 87.
88.
89.
90.
91.
92.
93.
94.
95. 96.
97.
129
Coal Preparation, Utilization and Environmental Control Contractors Conference, Pittsburgh, PA. Baxter, L.L. 2003. Biomass Combustion and Cofiring Issues Overview: Alkali Deposits, Flyash, NOx/SCR Impacts. International Conference on Co-Utilization of Domestic Fuels. Gainsville, FL. February 5–6. Amos, W.A. 2002. Summary of Chariton Valley Switchgrass Co-fire Testing at the Ottumwa Generating Station in Chillicothe, Iowa. National Renewable Energy Laboratory, Golden, CO. Szego, G.C., and C.C. Kemp. 1973. Energy Forests and Fuel Plantations. CHEMTECH. May. Bransby, D. 2003. Fuel Sources for Cofiring: A Case for Herbaceous Energy Crops in the United States. International Conference on Co-Utilization of Domestic Fuels. Gainsville, FL. February 5–6. Brubaker, G. 2003. Woody and Herbaceous Biomass Production on JEA’s Biomass Energy Research Farm. International Conference on Co-Utilization of Domestic Fuels. Gainsville, FL. February 5–6. Jianxiong, M. 2003. The Energy Structure and the Technology of Co-firing Biomass and Coal in China. International Conference on Co-Utilization of Domestic Fuels. Gainsville, FL. February 5–6. Wieck-Hansen, K., P. Overgaard, and O.H. Larsen. 2000. Cofiring Coal and Straw in a 150 MWe Power Boiler Experiences. Biomass and Bioenergy. 19(6): 395–410. Miles, T.R., T.R. Miles Jr., L.L. Baxter, R.W. Bryers, B.M. Jenkins, and L.L. Oden. 1995. Alkali Deposits Found in Biomass Power Plants: A Preliminary Investigation of Their Extent and Nature. Golden, CO: National Renewable Energy Laboratory. Walsh, M.E., and D. Becker. 1996. Biocost: A Software Program to Estimate the Cost of Producing Bioenergy Crops. Proc. Bioenergy ’96. Nashville, TN. September 15–20. pp. 480–486. Graham, R.L., W. Liu, H.I. Jager, B.C. English, C.E. Noon, and M.J. Daly. 1996. A Regional-Scale GIS-Based Modeling System for Evaluating the Potential Costs and Supplies of Biomass from Biomass Crops. Proc. Bioenergy ’96. Nashville, TN. September 15–20. pp. 444–450. Tolbert, V.R., J.D. Joslin, F.C. Thornton, and B.R. Bock. 1999. Biomass Crop Production: Benefits for Soil Quality and Carbon Sequestration. Proc. Fourth Biomass Conference of the Americas. Oakland, CA. August 29-September 2. pp. 127–132. Costello, R., and H.L. Chum. 1998. Biomass, Bioenergy, and Carbon Management. Proc. Bioenergy ’98. Madison, WI. October 4–8. pp. 11–18. Wieck-Hansen, K., P. Overgaard, and O.H. Larsen. 2000. Cofiring Coal and Straw in a 150 MWe Power Boiler Experiences. Biomass and Bioenergy. 19(6): 395–409. Rossi, A. 1985. Fuel Characteristics of Wood and Nonwood Biomass Fuels. Progress in Biomass Conversion, vol. 5. Orlando, FL: Academic Press. pp. 69–100.
130 Combustion Engineering Issues for Solid Fuel Systems 98. Sanderson, M.A., R.P. Egg, and A.E. Wiselogel. 1997. Biomass Losses During Harvest and Storage of Switchgrass. Biomass and Bioenergy. 12(2): 107–114. 99. Lalor, W.F. 1977. Use of Ginning Waste as an Energy Source. In Fuels and Energy from Renewable Resources. New York: Academic Press. pp. 257–274. 100. Folkedahl, B.C., C.J. Zygarlicke, P.N. Hutton, and D.P McCollor. 2001. Biomass for Energy—Characterization and Combustion Ash Behavior. Proc. Power Production in the 21st Century: Impacts of Fuel Quality and Operations. The Engineering Foundation. Snowbird, UT. October 28–November 2. 101. Cooper, J., M. Braster, and E. Woolsey. 1998. Overview of the Chariton Valley Switchgrass Project. Proc. Bioenergy ’98. Madison, WI. October 4–8. pp. 1–10. 102. Falcone Miller, S., and B.G. Miller. 2006. The Occurrence of Inorganic Elements in Various Biofuels and Its Effect on Ash Chemistry and Behavior and Use in Combustion Products. Proc. Impacts of Fuel Quality on Power Production. Snowbird, UT. October 28–November 2. 103. Tillman, D.A., B.G. Miller, and D.V. Johnson. 2002. Nitrogen Evolution from Biomass Fuels and Selected Coals. Proc. 19th Annual International Pittsburgh Coal Conference. Pittsburgh, PA. September 23–26. 104. Prinzing, D.E. 1996. EPRI Alternative Fuels Database. Palo Alto, CA: EPRI. Report TR-107602. 105. Miles, T.R., et al. 1993. Alkali Slagging Problems with Biomass Fuels. Proc. First Biomass Conference of the Americas. Burlington, VT. August 30–September 2. pp. 406–421. 106. Falcone Miller, S., B.G. Miller, and D. Tillman. 2002. The Propensity of Liquid Phases Forming During Coal-Opportunity Fuel (Biomass) Cofiring as a Function of Ash Chemistry and Temperature. Proc. 27th International Technical Conference on Coal Utilization and Fuel Systems. Clearwater, FL. March 4–7. 107. Falcone Miller, S., and B.G. Miller. 2002. The Occurrence of Inorganic Elements in Various Biofuels and Its Effect on the Formation of Melt Phases During Combustion. Proc. International Joint Power Generation Conference. Phoenix, AZ. June 24–27. (IJPGC2002-26177). 108. Unterberger, S., C. Lopez, and K.R.G. Hein. 2001. EU-Project: Prediction of Ash and Deposit Formation for Biomass Pulverized Fuel Co-Combustion. Proc. Power Production in the 21st Century: Impacts of Fuel Quality and Operations. The Engineering Foundation. Snowbird, UT. October 28– November 2. 109. Kaer, S.K. 2001. Modeling Deposit Formation in Straw-Fired Grate Boilers. Proc. Power Production in the 21st Century: Impacts of Fuel Quality and Operations. The Engineering Foundation. Snowbird, UT. October 28– November 2. 110. Korbee, R., J. Kiel, M. Zevenhoven, B. Skirifvars, P. Jensen, and F. Frandsen. 2001. Investigation of Biomass Inorganic Matter by Advanced Fuel Analysis and Conversion Experiments. Proc. Power Production in the 21st Century:
Characteristics of Alternative Fuels
131
Impacts of Fuel Quality and Operations. The Engineering Foundation. Snowbird, UT. October 28–November 2. 111. Frandsen, F.J., P.A. Jensen, W. Lin, and K. Dam-Johansen. 2001. Practical Experience with Ash and Deposit Formation in Danish Biomass-Fired Boilers. Proc. Power Production in the 21st Century: Impacts of Fuel Quality and Operations. The Engineering Foundation. Snowbird, UT. October 28– November 2. 112. NCASI. 1984. National Council of the Paper Industry for Air and Stream Improvement Technical Bulletin #447. New York.
CHAPTER
4
Characteristics and Behavior of Inorganic Constituents Jason D. Laumb Research Manager University of North Dakota Energy and Environmental Research Center,
Bruce C. Folkedahl
Senior Research Manager University of North Dakota Energy and Environmental Research Center, and
Christopher J. Zygarlicke Deputy Associate Director University of North Dakota Energy and Environmental Research Center
4.1 Introduction Ash produced during combustion of coal in conventional power systems is a major problem that results in decreased efficiency, unscheduled outages, equipment failures, and increased cleaning. The many ways in which the detrimental effects of ash manifest themselves in a boiler system include fireside ash deposition on heat-transfer surfaces, corrosion and erosion of boiler parts, poor slag flow, and production of fine particulates that are difficult to collect. Decades of research have been conducted to develop a better understanding of the chemical and physical processes of ash formation and deposition in combustion systems. Overviews of ash-related issues and compilations of work by many investigators can be found by referring to the work of Harding et al. [1], Mehta and Benson [2], Baxter and Desollar [3], Couch [4], Williamson and Wigley [5], Schobert [6], Benson 133
134 Combustion Engineering Issues for Solid Fuel Systems
et al. [7], Benson [8], Bryers and Vorres [9], Raask [10, 11], and Benson [12]. This work has led to a better understanding of ash formation and behavior in combustion systems as well as the development of predictive methods [13–15]. A general consensus of processes is required to minimize ash deposition problems in utility boilers. The chemical composition and physical characteristics of ash-forming or inorganic components of the fuel(s) fired have an influence on the following processes in the combustion systems: Firing methods such as cyclone, pulverized coal, and low-NOx
burners; Transformations of coal inorganic components to ash particulate
and vapor-phase species; Boiler design characteristics, including number of burners, radiant
section area, tube bank spacing, and access for cleaning; Ash transport to heat-transfer surfaces in utility boilers; Erosion wear and sticking; Ash deposit growth and impact on heat transfer; Ash blinding and plugging of selective catalytic reduction catalysts; and Ash deposit removability.
The ash in coal is usually measured in terms of ash quality (percent ash) and ash composition (ash mineral analysis). The ash (produced at 750 C [1,382 F]) is derived from the noncombustible inorganic fraction of the coal. The composition of the ash is listed as an ash mineral analysis or oxides of Na, Mg, Al, Si, P, S, K, Ca, Ti, and Fe. In the original coal, inorganic components are distributed in several forms, including organically associated inorganic elements; coal-bound, included minerals; and coal-free, excluded minerals. The types of inorganic components depend on the rank of the coal and the environment in which the coal was formed. Low-rank coals contain higher levels of organically associated cations. Low-rank subbituminous and lignite coals contain high levels of oxygen that can act as bonding sites for various cations such as sodium, calcium, magnesium, potassium, strontium, and barium. Higher-ranked coals do not contain high levels of organically associated inorganic elements because of the lower levels of oxygen. The primary mineral groups that are found in all coals consist of clay minerals, carbonates, sulfides, oxides, and quartz. The association of the inorganic components in the coal influences the interactions and transformations of these components during combustion. The chemical and physical transformations of the inorganic components to ash or slag during combustion depend on the design of the system, operating conditions, and fuel composition. During the combustion and gas-cooling process, the inorganic species undergo a complex series of
Characteristics and Behavior of Inorganic Constituents
135
chemical and physical transformations. The inorganic species are transformed into inorganic vapors, liquids, and solids during the initial combustion phase. These ash precursor materials range from vapors to solid particles. The ash materials are cooled as a result of heat transfer as they are transported with the bulk gas flow through the combustion system. The cooling process causes the vapor-phase inorganic components to condense and the liquid-phase components to solidify. Studies of the final ash product (fly ash) indicate a multimodal size distribution [16–19]. The submicron-size particles form as a result of homogeneous condensation of flame-volatilized species. Studies have indicated an intermediate-size mode of ash at about 1 mm that is produced when coal is combusted [20]. Flamevolatilized species may also condense heterogeneously on the surfaces of larger particles. The larger particles, sometimes referred to as residual ash, are largely derived from mineral grains. The composition and size distribution of the larger particles result from transformations or interactions between discrete mineral grains in higher-rank coals. In lower-rank coals, the interaction of the organically associated elements with mineral grains occurs as well as mineral–mineral interactions. A review was conducted on factors that impact the size and composition of ash produced upon combustion [7]. The characteristics of the intermediate inorganic species determine the extent of deposition and where deposits will form. As described previously, the intermediate inorganic species are in the form of inorganic gases, liquids, and solids. The inorganic species are transported to the heat-transfer surfaces by several mechanisms based on the state of the inorganic species [21]. Submicron particles that originate largely from flame-volatized species are transported to the surfaces by small-particle processes (diffusion, thermophoresis, and electrophoresis). Larger particles are transported to the surface by inertial impaction. The chemical and physical characteristics of deposits found in utility boilers vary widely. In general, deposits are microscopically very heterogeneous. Under a microscope, the deposits consist of pores, unreacted ash particles, a liquid-phase matrix material that bonds the deposits together, and some crystalline material. The types of phases that are responsible for the liquid-phase components are temperature-dependent because of the stability of the chemical species. For example, higher-temperature deposits, such as those found on the boiler walls, contain high levels of silicate phases that bond the deposit together. Lower-temperature deposits that form in the convective passes of utility boilers have high levels of sulfate phases that bond the particles together. The types of phases change with temperature and location in the boiler. Iron-rich phases including silicates, sulfides, oxides, and metallic iron also contribute to the formation of slag deposits. The advances made over the past several years in predicting ash behavior are possible as a result of more detailed and better analysis of coal and ash materials. These advanced techniques, such as computer-controlled scanning electron microscopy (CCSEM) and chemical fractionation (CHF),
136 Combustion Engineering Issues for Solid Fuel Systems
are able to quantitatively determine the chemical and physical characteristics of the inorganic components in coal [7]. Many of the mechanisms of ash formation, ash deposition, and ash collection in combustion systems are more clearly understood as a result of these new data. This understanding has led to the development of better methods of prediction that include advanced indices and phenomenological models.
4.2 Inorganic Composition of Coal There is much confusion on how to describe the inorganic composition of coal. The inorganic components in coals have been referred to as mineral matter, minerals, inherent/extraneous ash, and other names by many individuals who work with and utilize coal. For the purposes of this report, the term “inorganic constituents” will be used to describe all ash-forming constituents including both organically associated inorganic species and mineral grains.
4.2.1 Distribution of Inorganic Constituents in Coal The inorganic elements in coal occur as discrete minerals, organically associated cations, and cations dissolved in pore water. The fraction of inorganic components that are organically associated varies with coal rank. Lower-ranked subbituminous and lignitic coals have high levels of oxygen. These groups act as bonding sites for cations such as sodium, magnesium, calcium, potassium, strontium, and barium (other minor and trace elements may also be associated in the coal in this form). In higher-ranked coals, bituminous and anthracite, the inorganic components consist mainly of minerals. Mineral grains are usually the most abundant inorganic component in coal. The major mineral groups found in coals include silicates, aluminosilicates, carbonates, sulfides, sulfates, and phosphates, as well as some oxides. The behavior of the mineral grains associated with coal during combustion can only be predicted from detailed information on the abundance, size, and association of mineral grains in the coal. In addition, the association of the mineral grain with the coal matrix must be determined and classified. A mineral associated with the organic fraction of a coal particle is said to be “included.” A mineral that is not associated with organic material is referred to as “excluded.” The behavior of the organically associated elements, those elements that are atomically dispersed in the coal matrix, must also be measured as to their abundance in the coal. The organically associated elements will react and interact with the other ash-forming constituents during combustion.
Characteristics and Behavior of Inorganic Constituents
137
4.2.2 Methods of Determining Inorganic Composition Methods to determine the inorganic composition of coal have evolved significantly over approximately the past 80 years. The early methods involved concentrating the inorganic components for chemical analysis using ashing or gravity separation techniques. These methods are fraught with errors and do not provide quantitative information on the association and abundance of inorganic components in coals. Recently, more advanced methods such as CCSEM and CHF are being used to more quantitatively determine the abundance, size, and association of inorganic components in coals. High-temperature ashing is currently used routinely to determine the abundance of inorganic constituents present in coal. This technique involves oxidizing the coal at 1,023 K (750 C [1,382 F]) followed by chemical analysis of the resultant ash. Several major errors are involved in this type of analysis. First, significant transformations of the inorganic components occur during the ashing process. The loss of mass due to water loss from clay minerals, carbon dioxide loss from carbonates, and sulfur loss from pyrite can significantly influence the determination of the inorganic content of coal. The organically associated inorganic elements such as the alkali and alkaline-earth elements absorb oxygen and SO2 during the ashing processes, further complicating the determinations. In addition, these species (H2O, CO2, and SO2) are usually included as volatiles in the proximate analysis, leading to erroneous results. The resultant ash is analyzed to determine the quantities of Na, Mg, Al, Si, P, S, K, Ca, Ti, and Fe in the ash samples. The elemental components are assumed to be present as oxides and are converted to their equivalent oxide, which includes Na2O, MgO, Al2O3, SiO2, P2O5, SO3, K2O, CaO, TiO2, and Fe2O3, even though, in reality, the ash consists of aluminosilicates, alkali and alkaline-earth aluminosilicates, sulfates, and oxides that do not represent what was present in the original coal. The technique that has shown the most promise for quantitative determination of the mineral portion of the inorganic components in coal is SEM and microprobe (energy-dispersive X-ray analysis) analysis. Over the past 20 years, this technique has been used much more rigorously to determine the mineral component in coal. To determine the size, abundance, and association of mineral grains in both high- and low-rank coals, CCSEM and automated image analysis (AIA) are the preferred techniques used to analyze polished cross-sections of coal epoxy plugs [22]. The CCSEM technique is used to determine the size, shape, quantity, and quantitative composition of mineral grains in coals [23, 24]. The key components of the CCSEM system that make it possible to image, size, and analyze inorganic particles are the backscatter electron detector, digital beam control, and the ultrathin window energydispersive X-ray detector. Backscatter electron imaging is used for CCSEM because the intensity of the backscattered electrons is a function of the
138 Combustion Engineering Issues for Solid Fuel Systems
average atomic number of the features on or near the surface. Since the mineral particles appear brighter relative to the lower atomic number background of the matrix, a distinction can be made between coal, mounting media, and mineral grains. The electron beam is programmed to scan over the field of view and locate the bright inclusions that correspond to mineral or ash species. On finding a bright mineral grain, the grain is sized, the center is located, and an energy-dispersive spectroscopy (EDS) is collected at that point for 2 seconds. The EDS provides the chemical composition of the grain. Software classifies the mineral grains based on the EDS elemental composition and size. The parameters used to identify the minerals are based on published compositions of known minerals. Quantification of the type and abundance of organically associated inorganic elements in lower ranked subbituminous and lignitic coals is currently performed by CHF [25]. Chemical fractionation is used to selectively extract elements from the coal based on solubility, which reflects their association in the coal. Briefly, the technique involves extracting the coal with water to remove water-soluble elements such as Na in sodium sulfate or those elements that were most likely associated with the groundwater in the coal. This is followed by extraction with 1 M of ammonium acetate to remove elements such as Na, Ca, and Mg that may be bound as salts of organic acids. The residue of the ammonium acetate extraction is then extracted with 1 M HCl to remove acid-soluble species such as Fe and Ca, which may be in the form of hydroxides, oxides, carbonates, and organically coordinated species. The components remaining in the residue after all three extractions are assumed to be associated with the insoluble mineral species such as clays, quartz, and pyrite. The bulk chemical properties of selected coal types from the United States are summarized in Table 4-1. The table includes ranges of characteristics of coals in terms of proximate, ultimate, heating value, ash composition, and CHF results. The CHF data provide a means of determining the level of inorganic elements associated with the organic matrix or minerals in coals. The results of chemical fractionation shown in Table 4-1 indicate that major portions of the sodium, calcium, and magnesium in the lower rank and lignite and subbituminous coals are largely extracted with ammonium acetate, indicating an organic association. The higher rank coals show some removal of these elements with ammonium acetate, but the levels of the alkali and alkaline-earth elements are very low relative to the western U.S. coals. The CCSEM mineral analysis for coal types in the United States is summarized in Table 4-2. The CCSEM analyses are provided on a weight percent mineral basis. The major minerals found in coals consist of quartz, kaolinite, calcite, illite (potassium aluminosilicate), pyrite, and gypsum. Minor amounts of other minerals such as montmorillonite, dolomite, and calcium aluminum phosphates are also present. The size, composition, and abundance of these minerals in the coal have a major impact on the fireside performance of these coals in utility boilers.
Characteristics and Behavior of Inorganic Constituents
139
TABLE 4-1 Coal Properties of Selected U.S. Coals Lignite Max., wt% Moisture Volatile Matter Fixed Carbon Ash Hydrogen Carbon Nitrogen Sulfur Oxygen Ash Btu/lb (kJ/kg) SiO2 Al2O3 Fe2O3 TiO2 P2O5 CaO MgO Na2O K2O SO3 Average CHF Analysis Silicon Aluminum Iron Titanium Phosphorus Calcium Magnesium Sodium Potassium
Min., wt%
Av, wt%
Std. Dev., wt%
Proximate/Ultimate Analysis 38.8 22.9 30.5 37.3 24.0 33.0 34.7 24.2 29.0 11.7 4.6 7.0 7.2 5.6 6.5 49.1 39.8 44.5 0.7 0.5 0.6 2.1 0.5 1.1 46.9 34.2 40.4 11.7 4.6 7.0 8,602 (19,960) 7,003 (16,250) 7,715 (17,900)
4.5 2.8 2.9 2.0 0.5 2.9 0.1 0.5 3.3 2.0 436 (1000)
39.0 15.9 16.9 1.0 1.1 24.9 12.2 9.6 0.9 30.5
XRF Analysis 20.0 5.5 4.7 0.3 0.2 12.1 6.1 1.0 0.3 11.9
26.7 9.5 8.3 0.6 0.5 19.0 8.8 4.6 0.5 21.7
5.8 2.9 2.5 0.2 0.3 4.0 1.7 3.0 0.2 4.4
Removed by H2O (%) 0 1 5 3 13 2 4 20 8
Removed by NH4OAc (%) 0 0 4 5 24 49 60 66 31
Removed by HCl (%) 0 15 4 2 56 48 32 8 6
Remaining (%)
Av, wt%
Std. Dev., wt%
100 84 87 91 7 1 4 6 55
Montana Subbituminous Max., wt% Moisture Volatile Matter Fixed Carbon Ash Hydrogen
23.7 33.0 40.7 13.3 6.7
Min., wt%
Proximate/Ultimate Analysis 16.6 29.8 34.5 4.1 5.8
21.6 31.9 38.0 8.5 6.2
2.9 1.1 2.3 3.1 0.3 (continued)
140 Combustion Engineering Issues for Solid Fuel Systems TABLE 4-1 Coal Properties of Selected U.S. Coals (continued) Montana Subbituminous
Carbon Nitrogen Sulfur Oxygen Ash Btu/lb (kJ/kg) SiO2 Al2O3 Fe2O3 TiO2 P2O5 CaO MgO Na2O K2O SO3 Average CHF Analysis Silicon Aluminum Iron Titanium Phosphorus Calcium Magnesium Sodium Potassium
Max., wt%
Min., wt%
Av, wt%
Std. Dev., wt%
65.9 0.8 0.8 33.6 13.3 9,688 (22,490)
49.8 0.6 0.5 22.1 4.1 8,767 (20,350)
54.8 0.8 0.7 29.0 8.5 9,279 (21,540)
5.6 0.1 0.1 4.4 3.1 306 (700)
50.5 28.4 7.1 2.2 1.1 21.3 7.1 2.2 0.9 15.4
XRF Analysis 36.0 20.2 1.9 0.7 0.1 3.5 3.3 0.4 0.3 0.0
42.0 22.1 4.8 1.3 0.6 13.8 4.8 1.2 0.5 8.8
5.3 3.1 1.7 0.5 0.3 5.9 1.5 0.7 0.2 6.0
Removed by H2O (%) 0 1 20 9 13 1 8 37 18
Removed by NH4OAc (%) 0 0 4 2 8 73 67 52 9
Removed by HCl (%) 0 4 19 6 40 23 14 1 1
Remaining (%) 99 95 57 82 38 3 12 10 71
Wyoming Subbituminous Max., wt%
Moisture Volatile Matter Fixed Carbon Ash Hydrogen Carbon Nitrogen Sulfur Oxygen Ash Btu/lb (kJ/kg)
Min., wt%
Av, wt%
Proximate/Ultimate Analysis 31.4 13.4 25.3 43.8 30.7 34.2 40.8 24.1 35.6 9.1 3.5 4.9 13.4 4.9 6.6 59.8 47.4 51.5 1.0 0.6 0.7 0.9 0.2 0.4 40.8 16.0 35.7 9.1 3.5 4.9 10,681 (24,790) 8,218 (19,070) 9,020 (20,930)
Std. Dev., wt%
3.9 3.0 2.7 1.0 1.2 2.9 0.1 0.1 4.3 1.0 531 (1,230) (continued)
Characteristics and Behavior of Inorganic Constituents
141
TABLE 4-1 Coal Properties of Selected U.S. Coals (continued) Wyoming Subbituminous
SiO2 Al2O3 Fe2O3 TiO2 P2O5 CaO MgO Na2O K2O SO3 Average CHF Analysis Silicon Aluminum Iron Titanium Phosphorus Calcium Magnesium Sodium Potassium
Max., wt%
Min., wt%
Av, wt%
Std. Dev., wt%
36.4 24.0 10.2 2.3 2.0 31.6 10.2 7.2 0.8 30.1
XRF Analysis 15.6 12.6 2.3 0.8 0.1 12.8 3.6 0.2 0.0 9.1
28.3 16.3 5.7 1.3 1.3 22.5 6.9 1.5 0.3 16.0
4.3 2.3 1.7 0.4 0.5 3.9 1.8 1.3 0.1 4.7
Removed by HCl (%) 0 30 59 7 62 33 20 4 6
Remaining (%)
Removed by H2O (%) 0 1 4 4 6 2 3 30 15
Removed by NH4OAc (%) 0 4 9 8 15 62 71 60 21
99 65 28 80 17 4 6 3 51
Western Bituminous Max., wt% Moisture Volatile Matter Fixed Carbon Ash Hydrogen Carbon Nitrogen Sulfur Oxygen Ash Btu/lb (kJ/kg) SiO2 Al2O3 Fe2O3 TiO2 P2O5 CaO MgO
Min., wt%
Av, wt%
Std. Dev., wt%
Proximate/Ultimate Analysis 12.0 1.7 6.3 42.9 40.4 42.1 50.1 45.4 48.3 8.9 6.0 7.3 5.7 5.2 5.4 73.5 68.2 71.2 1.6 1.1 1.4 0.6 0.5 0.6 16.0 13.0 14.2 8.9 6.0 7.3 12,567 (29,170) 11,330 (26,300) 11,962 (27,760)
5.2 1.4 2.5 1.5 0.3 2.7 0.3 0.1 1.6 1.5 619 (1,440)
XRF Analysis 28.3 13.3 7.1 0.7 0.2 2.3 0.9
9.2 8.4 7.3 0.3 0.4 11.6 4.1
45.9 29.2 20.9 1.2 1.0 25.0 8.3
38.6 19.7 12.7 0.9 0.5 12.4 3.6
(continued)
142 Combustion Engineering Issues for Solid Fuel Systems TABLE 4-1 Coal Properties of Selected U.S. Coals (continued) Western Bituminous
Na2O K2O SO3 Average CHF Analysis Silicon Aluminum Iron Titanium Phosphorus Calcium Magnesium Sodium Potassium
Max., wt%
Min., wt%
Av, wt%
Std. Dev., wt%
3.6 1.5 14.4
0.5 0.8 2.1
1.7 1.2 8.8
1.6 0.4 6.2
Removed by H2O (%) 0 3 3 6 9 2 4 3 3
Removed by NH4OAc (%) 0 1 0 0 8 82 67 62 4
Removed by HCl (%) 8 18 64 1 0 5 3 12 4
Remaining (%)
Av, wt%
Std. Dev., wt%
92 78 33 93 83 12 27 23 90
Central Bituminous Max., wt%
Moisture Volatile Matter Fixed Carbon Ash Hydrogen Carbon Nitrogen Sulfur Oxygen Ash Btu/lb (kJ/kg) SiO2 Al2O3 Fe2O3 TiO2 P2O5 CaO MgO Na2O K2O SO3 Average CHF Analysis Silicon Aluminum
Min., wt%
Proximate/Ultimate Analysis 10.5 1.9 7.0 36.4 28.8 34.4 53.6 40.6 47.2 14.7 6.8 11.4 8.9 5.1 5.8 74.9 4.3 56.9 65.4 1.0 10.4 4.3 0.6 2.9 17.8 1.6 11.5 14.7 6.8 11.4 13,686 (31,770) 10,732 (24,900) 11,850 (27,504)
3.3 2.6 5.3 3.3 1.4 24.0 24.3 1.4 5.5 3.3 1,081 (2500)
50.6 23.1 22.0 1.5 0.3 15.7 2.1 1.8 2.0 4.2
XRF Analysis 47.3 17.5 8.4 0.6 0.0 1.8 0.8 0.3 1.0 1.6
49.1 19.8 17.7 0.8 0.2 5.8 1.3 0.8 1.7 2.7
1.0 1.7 4.5 0.3 0.1 4.6 0.5 0.5 0.4 1.1
Removed by H2O (%) 0 0
Removed by NH4OAc (%) 0 0
Removed by HCl (%) 0 2
Remaining (%) 100 98 (continued)
Characteristics and Behavior of Inorganic Constituents
143
TABLE 4-1 Coal Properties of Selected U.S. Coals (continued) Central Bituminous
Iron Titanium Phosphorus Calcium Magnesium Sodium Potassium
Max., wt%
Min., wt%
Av, wt%
Std. Dev., wt%
11 0 4 30 8 37 3
0 0 4 55 8 20 1
14 6 12 1 4 3 2
74 94 80 14 80 41 94
Av, wt%
Std. Dev., wt%
6.1 35.1
2.0 2.0
50.0 8.8 5.4 70.8 1.4 1.3 12.6 8.5 12,542 (29,110)
2.4 1.7 0.2 2.0 0.1 0.7 2.1 1.6 332 (770)
53.1 28.7 9.4 1.2 0.1 1.6 1.1 0.5 2.6 1.6
3.6 4.0 5.2 0.3 0.1 1.3 0.4 0.2 0.5 1.3
Removed by HCl (%) 0 3 25 0 64 6 0 2 12
Remaining (%)
Central Bituminous Max., wt% Moisture Volatile Matter Fixed Carbon Ash Hydrogen Carbon Nitrogen Sulfur Oxygen Ash Btu/lb (kJ/kg) SiO2 Al2O3 Fe2O3 TiO2 P2O5 CaO MgO Na2O K2O SO3 Average CHF Analysis Silicon Aluminum Iron Titanium Phosphorus Calcium Magnesium Sodium Potassium
9.6 39.7
Min., wt% Proximate/Ultimate Analysis 3.3 33.3
55.3 11.7 5.6 72.4 1.5 2.7 16.1 11.3 12,871 (29,870) 56.2 33.3 19.8 1.5 0.3 4.4 1.7 0.9 3.5 4.2 Removed by H2O (%) 0 0 1 0 36 54 8 35 6
47.4 6.2 5.1 66.4 1.3 0.6 8.7 6.2 11,900 (27,620) XRF Analysis 46.0 20.4 4.5 0.8 0.0 0.7 0.7 0.3 2.1 0.6 Removed by NH4OAc (%) 0 0 0 1 0 25 13 27 0
100 97 74 99 0 15 79 36 82
144 Combustion Engineering Issues for Solid Fuel Systems TABLE 4-2 Distribution of Minerals in Selected U.S. Coals as Determined by CCSEM Analysis, wt%, Mineral Basis U.S. Lignite Mineral Quartz Iron Oxide Periclase Rutile Alumina Calcite Dolomite Ankerite Kaolinite Montmorillonite K Al-Silicate Fe Al-Silicate Ca Al-Silicate Na Al-Silicate Aluminosilicate Mixed Al-Silicate Fe Silicate Ca Silicate Ca Aluminate Pyrite Pyrrhotite Oxidized Pyrrhotite Gypsum Barite Apatite Ca Al-Phosphate KCl Gypsum/Barite Gypsum/Al-Silicate Si-Rich Ca-Rich Ca–Si-Rich Unclassified
Max., wt%
Min., wt%
Av, wt%
Std. Dev., wt%
40.4 1.8 0.0 1.2 0.5 12.7 0.1 0.0 45.9 5.9 3.9 0.4 6.0 0.2 4.1 0.3 0.0 0.3 0.0 58.9 3.0 2.7 3.5 7.0 0.0 5.6 0.0 0.1 1.0 3.0 1.6 0.0 7.1
12.4 0.4 0.0 0.1 0.0 0.1 0.0 0.0 3.7 0.1 0.1 0.1 0.2 0.0 0.2 0.0 0.0 0.0 0.0 1.8 0.1 0.1 0.0 0.7 0.0 0.4 0.0 0.0 0.1 1.3 0.0 0.0 4.4
22.7 1.1 0.0 0.7 0.1 3.4 0.0 0.0 18.9 2.0 2.0 0.2 2.0 0.1 1.7 0.1 0.0 0.1 0.0 27.0 0.8 0.8 1.2 4.3 0.0 2.1 0.0 0.1 0.4 2.0 0.5 0.0 5.7
10.6 0.7 0.0 0.4 0.2 5.4 0.0 0.0 16.6 2.3 1.7 0.1 2.3 0.1 1.5 0.1 0.0 0.1 0.0 21.0 1.3 1.1 1.4 2.3 0.0 2.1 0.0 0.0 0.3 0.6 0.7 0.0 1.0
U.S. Western Subbituminous Mineral Quartz Iron Oxide Periclase Rutile Alumina Calcite Dolomite Ankerite
Max., wt%
Min., wt%
Av, wt%
Std. Dev., wt%
60.9 1.7 0.0 4.9 0.2 11.2 0.2 0.0
17.3 0.0 0.0 0.1 0.0 0.0 0.0 0.0
33.3 0.5 0.0 1.1 0.0 1.7 0.0 0.0
14.6 0.5 0.0 1.4 0.1 3.3 0.1 0.0 (continued)
Characteristics and Behavior of Inorganic Constituents
145
TABLE 4-2 Distribution of Minerals in Selected U.S. Coals as Determined by CCSEM Analysis, wt%, Mineral Basis (continued) U.S. Western Subbituminous Mineral Kaolinite Montmorillonite K Al-Silicate Fe Al-Silicate Ca Al-Silicate Na Al-Silicate Aluminosilicate Mixed Al-Silicate Fe Silicate Ca Silicate Ca Aluminate Pyrite Pyrrhotite Oxidized Pyrrhotite Gypsum Barite Apatite Ca Al-Phosphate KCl Gypsum/Barite Gypsum/Al-Silicate Si-Rich Ca-Rich Ca–Si-Rich Unclassified
Max., wt%
Min., wt%
Av, wt%
Std. Dev., wt%
51.1 21.6 4.1 1.0 2.9 0.1 10.4 1.3 0.1 0.3 0.0 12.3 0.5 0.1 0.1 2.5 0.8 12.6 0.1 0.4 1.1 5.7 0.3 0.1 13.6
10.9 0.2 0.5 0.0 0.2 0.0 0.2 0.1 0.0 0.0 0.0 2.6 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.1 0.0 0.0 2.2
28.6 5.4 2.5 0.2 1.4 0.0 2.3 0.5 0.0 0.1 0.0 7.2 0.1 0.0 0.0 1.2 0.2 5.5 0.0 0.1 0.4 2.0 0.0 0.0 5.6
13.3 6.2 1.2 0.3 0.9 0.0 2.9 0.3 0.0 0.1 0.0 3.1 0.1 0.0 0.0 0.9 0.3 4.3 0.0 0.1 0.4 1.5 0.1 0.0 3.3
U.S. Eastern Bituminous Mineral Quartz Iron Oxide Periclase Rutile Alumina Calcite Dolomite Ankerite Kaolinite Montmorillonite K Al-Silicate Fe Al-Silicate Ca Al-Silicate Na Al-Silicate Aluminosilicate Mixed Al-Silicate
Max., wt%
Min., wt%
Av, wt%
Std. Dev., wt%
18.1 2.6 0.0 0.6 0.3 9.5 2.9 0.0 22.7 8.0 27.4 12.0 3.6 0.5 3.5 2.9
13.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 11.2 1.7 9.2 0.1 0.0 0.0 0.9 0.0
15.2 0.9 0.0 0.2 0.1 3.1 0.5 0.0 14.9 4.5 14.0 2.6 0.7 0.1 1.8 0.8
1.7 0.9 0.0 0.2 0.1 3.7 1.1 0.0 3.7 1.9 6.3 4.3 1.3 0.2 0.9 1.0 (continued)
146 Combustion Engineering Issues for Solid Fuel Systems TABLE 4-2 Distribution of Minerals in Selected U.S. Coals as Determined by CCSEM Analysis, wt%, Mineral Basis (continued) U.S. Eastern Bituminous Mineral Fe Silicate Ca Silicate Ca Aluminate Pyrite Pyrrhotite Oxidized Pyrrhotite Gypsum Barite Apatite Ca Al-Phosphate KCl Gypsum/Barite Gypsum/Al-Silicate Si-Rich Ca-Rich Ca–Si-Rich Unclassified
Max., wt%
Min., wt%
Av, wt%
Std. Dev., wt%
0.4 0.6 0.0 35.9 0.5 2.0 5.8 0.0 0.1 0.0 0.0 0.0 0.5 3.9 0.5 0.2 22.1
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.8 0.0 0.0 3.8
0.1 0.1 0.0 23.2 0.2 0.6 1.7 0.0 0.0 0.0 0.0 0.0 0.2 2.7 0.3 0.1 11.7
0.1 0.2 0.0 13.0 0.2 0.7 1.9 0.0 0.0 0.0 0.0 0.0 0.2 0.7 0.2 0.1 6.4
U.S. Anthracite Mineral Quartz Iron Oxide Periclase Rutile Alumina Calcite Dolomite Ankerite Kaolinite Montmorillonite K Al-Silicate Fe Al-Silicate Ca Al-Silicate Na Al-Silicate Aluminosilicate Mixed Al-Silicate Fe Silicate Ca Silicate Ca Aluminate Pyrite Pyrrhotite Oxidized Pyrrhotite Gypsum Barite
Max., wt%
Min., wt%
Av, wt%
Std. Dev., wt%
16.4 0.7 0.0 4.9 1.1 7.2 8.2 0.8 60.4 13.8 32.4 11.0 11.1 6.1 6.3 2.3 0.1 0.3 0.1 6.4 0.2 0.6 1.8 0.0
1.2 0.0 0.0 0.4 0.0 0.2 0.0 0.0 12.1 0.6 6.9 0.3 0.3 0.1 1.9 0.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
9.6 0.3 0.0 1.1 0.4 1.6 1.6 0.2 32.9 5.5 18.6 3.6 2.0 1.1 4.0 1.3 0.0 0.1 0.0 2.4 0.0 0.2 0.3 0.0
5.7 0.2 0.0 1.5 0.5 2.3 2.8 0.3 19.6 4.3 10.3 3.7 3.7 2.0 1.6 0.7 0.1 0.1 0.0 2.6 0.1 0.2 0.6 0.0 (continued)
Characteristics and Behavior of Inorganic Constituents
147
TABLE 4-2 Distribution of Minerals in Selected U.S. Coals as Determined by CCSEM Analysis, wt%, Mineral Basis (continued) U.S. Anthracite Mineral Apatite Ca Al-Phosphate KCl Gypsum/Barite Gypsum/Al-Silicate Si-Rich Ca-Rich Ca–Si-Rich Unclassified
Max., wt%
Min., wt%
Av, wt%
Std. Dev., wt%
0.6 0.1 0.1 0.0 0.3 4.7 2.5 0.1 16.9
0.0 0.0 0.0 0.0 0.0 0.3 0.2 0.0 5.8
0.3 0.0 0.0 0.0 0.2 1.8 0.9 0.0 10.2
0.2 0.1 0.0 0.0 0.1 1.3 0.8 0.0 3.4
In addition to the composition of the fuel, variability in the composition has a major impact on the performance of a coal-fired power plant. Unanticipated changes in coal characteristics can have a major impact on the performance of a power plant. Figure 4-1 shows the range of characteristics of delivered coal to a minemouth power plant. This figure shows the sequence of deliveries beginning on July 6 and ending on July 12, 1999. During this period, the majority of the coal delivered was from seams designated as the HA seam, followed by KC, and HB. Most of the deliveries alternated between KC and HA. Figure 4-1 shows several instances in which significant quantities
FIGURE 4-1 Variations in the delivered coal quality during a test burn.
148 Combustion Engineering Issues for Solid Fuel Systems
of KC coal were delivered July 6–9, 1999. These dramatic changes in coal composition have major impacts on all aspects of fireside performance from slag flow management, waterwall slagging, and convective pass fouling [26, 27].
4.2.3 General Coal Characteristics The following information is a summary of general coal characteristics by coal rank. The information is most specific to fouling/slagging behavior of the fuels. Keep in mind that the information is general in nature, and there certainly are fuels that do not behave as described. All ranks of coal (with the exception of anthracite) found in the United States are discussed. 4.2.3.1 Lignites Lignites, in general, are higher moisture, higher alkali, and higher reactivity fuels as compared to higher rank coals. Lignites from the Fort Union region and the Gulf Coast region are the two primary sources of lignite for use in power generation in the United States. Fort Union lignite tends to have high and highly variable sodium contents which can lead to significant fouling and slagging problems. Nearly all, if not 100%, of the sodium contained in Fort Union lignites is organically bound and volatilizes in the combustion process. This vapor-phase sodium is then available for reaction with other flue gas and ash constituents and can produce very low-meltingpoint phases and well-sintered deposits. The boiler design for using Fort Union lignites has centered on decreasing the furnace exit gas temperature (FEGT) by increasing furnace volume, increasing burner-to-arch height, increasing the number of sootblowers per 100 MW, and increasing convection tube spacing. Gulf Coast lignites generally have a lower heating value, higher ash content, and lower sodium content than Fort Union lignites. Gulf Coast lignites also have a high silica content that can be very abrasive. 4.2.3.2 Subbituminous Coals Subbituminous coals, in general, tend to have more alkali and alkaline-earth elements than bituminous coals, and the majority will not be associated in a mineral. This is evidenced by the high amount of these materials that are leached during the chemical fractionation analysis. These inorganic materials will become vapor phase in the flame, and as the temperature decreases below approximately 1,900 F [1,038 C], sulfur, phosphorus, sodium, and potassium become thermodynamically more stable in condensed form. These elements will then condense onto other ash particles or form submicron ash particles through heterogeneous and homogeneous nucleation. The condensation of these materials onto other ash particles can lower the melting temperature of the ash particle and allow deposits to form and develop strength more rapidly. The very small particles nucleated can have
Characteristics and Behavior of Inorganic Constituents
149
a very large surface area and will react with other elements in the backpass (primary superheater/economizer) regions of the system and form problematic fouling deposits. This tends to manifest itself as calcium sulfate-rich deposits occurring in the primary superheater and economizer sections of the boiler. Silica, aluminum, iron, and magnesium are not generally thermodynamically stable in a gas phase at the temperatures seen in the combustion system and so exist as solids in the boiler, unlike sulfur, phosphorus, sodium, and potassium. However, the alkali and alkaline-earth elements can react with the silicates and aluminosilicates in the system acting as a flux and lowering the melting temperature of the material. This enables the silicate phases to become softer and sticky, which can then aid in sintering and in the development of strength in the deposits formed, making them harder to remove. Powder River Basin (PRB) coals, in general, tend to have more alkali and alkaline-earth elements, and they are not associated in a mineral. This is evidenced by the high amount of these materials that are leached during the chemical fractionation analysis. These inorganic materials become either vapor phase in the flame, condensing out onto other ash particles, or form submicron ash particles through heterogeneous and homogeneous nucleation. These very small particles can have a very large surface area and will react with other elements such as sulfur in the backpass regions of the system and form problematic fouling deposits. 4.2.3.3 Bituminous Coals Bituminous coals are generally characterized by the high iron content and low alkali and alkaline-earth element content when compared to subbituminous and lignite coals. Typically, the iron content of a bituminous coal exceeds that of the sum of the alkali and alkaline-earth elements. Bituminous coals tend to have higher sulfur content than western subbituminous and lignite coals as well. Alkaline elements are also present in bituminous coals, but the percentage that is available to volatilize and react with other elements is much less than that of subbituminous and lignites. A portion of the alkali and alkaline-earth elements can be tied up in clay minerals, which do not break down readily in the combustion environment. Lignite and bituminous coal may have the same amount of sodium present in the coal, but the sodium that is available to react with other species will generally be higher in the lignite. This higher concentration of organically bound sodium will tend to increase fouling propensity and increase the strength of deposits that do form. High iron concentrations in coals can also provide fouling and slagging difficulties. Iron in bituminous coals occurs predominantly as pyrite, FeS2. During combustion, this may be reduced to FeS, lower oxides such as FeO, or even metallic iron—all of which have lower melting temperatures. This can increase their ability to form deposits and increase the
150 Combustion Engineering Issues for Solid Fuel Systems
strength development of the deposits formed. Additionally, these compounds may react with other species and form even lower melting-point phases. Sulfur may also combine with the alkali and alkaline-earth elements to form very low melting-point phases, which can also enhance the depositional process. Calcium in the form of limestone, as well as an organically bound mineral, can form very low melting-point phases. Limestone releases CO2 when heated above 80 C, leaving behind calcium and magnesium. Calcium and magnesium residue left after releasing CO2 will likely not react with other fly ash particles in the gas stream but will more likely flux other minerals in the same coal particle to produce low-melting-point phases. These low-melting-point phases are soft and will deform and adhere to the heat-transfer surfaces that they impact. Solid hard phases will rebound from the heat-transfer surface when they are impacted and reenter the gas stream. The adhesion of the softer, low melting-point phases allows the buildup of deposits on heat-transfer surfaces to grow and sinter at a much higher rate than higher melting-point phases. 4.2.3.4 World-Traded Coals There is a very mature worldwide market for steaming coal, currently amounting to about 350 million metric tons per annum. Until recently, the main coal exporting countries have been Australia, South Africa, Colombia, Poland, and the United States. More recently, the export markets in Russia, China, and Indonesia have been developed. The main importers of traded coals are developed countries with limited or no indigenous coal, i.e., western Europe and the Far East (Japan, Korea, and Taiwan). It should be noted that nearly all this market is for relatively high-quality bituminous coal, as it is not economical to transport lower rank coals large distances. The main exception is Indonesian coal that is mainly subbituminous in nature, although some high-volatile-content bituminous coals are produced. These coals are mainly exported to countries in the Pacific Rim but have occasionally been purchased by European utilities. The common feature of nearly all these coals is the low sulfur content—0.3% to 0.8% for bituminous coals. Although the United States has been a significant coal exporter in the past, nowadays this tends to be restricted to the higher sulfur content (>1%) coals. Because of SO2 emission legislation in the United States, the low-sulfur indigenous coals generally command a higher price in the U.S. market. For the same reason, there is increasing interest in importing low-sulfur world-traded coals into the United States. Because of the geographical proximity, the obvious source is South America (mainly Colombia, but also Venezuela), but U.S. utilities have even purchased subbituminous Indonesian coal. The commercial attractiveness of importing coal into the United States will be heavily dependent on freight rates, which fluctuate considerably. However, it is conceivable that, in addition to South American coals, South African or
Characteristics and Behavior of Inorganic Constituents
151
even Russian coal could be attractive to East Coast utilities; similarly, Australian coals could be attractive on the West Coast. As indicated previously, the low sulfur content of these coals implies a low pyrite and, therefore, iron content. This is certainly true for northern hemisphere coals, for which pyrite is the main source of iron, though siderite (FeCO3) and ankerite (a mixed calcium–magnesium–iron–manganese carbonate) also occur. In southern hemisphere coals, the position is reversed, with siderite predominating and pyrite normally the minor form of iron. However, the levels of siderite are still low. There are also differences in the nature of the aluminosilicate minerals. Quartz is found in all coals, but in northern hemisphere coals, the aluminosilicate is predominantly illitic clay, with some kaolin normally present. In southern hemisphere coals, kaolin dominates and illite is not normally found. As kaolin is a very refractory, pure aluminosilicate, the southern hemisphere coals will tend, therefore, to be even more inert from a slagging point of view. South African coals generally have high calcium contents—normally 7–10% CaO in the ash—compared to most bituminous coals. However, the iron content is normally very low, and slagging problems are not normally experienced when these coals are fired. It should be noted that the common use of the terms “southern” and “northern” hemispheres in this context are not necessarily geographically correct. More correctly, “southern” refers to the protocontinent of Gondwanaland and includes Colombia and China, as well as Australia and South Africa. The “northern” hemisphere coals are restricted to Europe and North America. Many Australian coals produce exceptionally refractory ashes with very high silica contents, up to about 75%. However, the problem is not simply the result of the high level of quartz but of the extremely fine size distribution in a few New South Wales coals. This results in the formation of highly reflective deposits on the furnace walls that can have a major impact on heat transfer.
4.3 Ash Formation: Transformation of Coal Inorganic Constituents The inorganic fraction of the coal will undergo complex chemical and physical transformations during combustion to produce intermediate ash species. These intermediate ash species will consist of vapors, liquids, and solids. The partitioning of the inorganic components during combustion to form ash intermediates depends on the manner in which the inorganic material is bound in the coal, the chemical characteristics of the inorganic components, the physical characteristics of the coal particles, the physical characteristics of the coal minerals, and the combustion conditions. To predict the effects of inorganic constituents on combustion systems, the mechanisms by which the size and composition of intermediate
152 Combustion Engineering Issues for Solid Fuel Systems
ash species are formed must be elucidated. The size, composition, and phase of the intermediate ash species directly influence slagging and fouling problems in combustion systems. The physical transformation of inorganic constituents depends on the inorganic composition of the coal and combustion conditions. The inorganic components can consist of water-soluble fractions, organically associated cations, mineral grains that are included in coal particles, and excluded mineral grains. There is a wide range of combinations of mineral–mineral, mineral–coal, mineral–cation–coal, and mineral–mineral–cation–coal associations in coal. These associations are unique to each coal sample. The physical transformations involved in fly ash formation include (1) coalescence of individual mineral grains within a char particle, (2) shedding of the ash particles from the surface of the chars, (3) incomplete coalescence due to disintegration of the char, (4) convective transport of ash from the char surface during devolatilization, (5) fragmentation of the inorganic mineral particles, (6) formation of cenospheres, and (7) vaporization and subsequent condensation of the inorganic components upon gas cooling. As a result of these interactions, the resulting ash has a bimodal size distribution. The submicron component is largely a result of the condensation of flame-volatilized inorganic components. These volatilized inorganic components generally come from the fraction of inorganics in the coal that are either associated with water in the coal or from the organically associated cation fraction. The mass mean diameter of the larger particles is approximately 12 to 15 mm, depending on the coal and combustion conditions. The larger size particles have been called the residual ash by some investigators [28] because these ash particles resemble, to a limited degree, the original minerals in the coal. However, many processes affect the composition and form of the resultant ash, such as ash mineral coalescence, partial coalescence, ash shedding, and char fragmentation during char combustion and mineral fragmentation. They all play an important role in the size and composition of the final fly ash. Loehden et al. [16], Hurley et al. [29], and Zygarlicke et al. [30] indicate that three potential modes for fly ash generation can be used to describe fly ash particle-size and composition evolution for inorganic minerals contained within the coal particles or included minerals. The first, “fine limit,” assumes that each mineral grain forms a fly ash particle and that the organically associated elements form fly ash particles less than 2 mm. The second, “total coalescence,” assumes one fly ash particle forms per coal particle. The third, “partial coalescence,” suggests that the fly ash composition and particle size evolve because of partial coalescence. All three processes come into play during the combustion process. Inorganic minerals that are liberated from the coal during comminution or are detrital to the coal will tend to form discrete ash particles that will have the same general composition as the inorganic minerals had prior to the combustion processes. These excluded inorganics will have little interaction with the other inorganic fractions of the coal. The transformations of excluded
Characteristics and Behavior of Inorganic Constituents
153
minerals are dependent on the physical characteristics of the mineral. Excluded minerals such as quartz (SiO2) can be carried through the combustion system with its angular structure still intact. Excluded clay minerals can fragment during dehydration, melt, and form cenospheres. The behavior of excluded pyrite depends on its morphology. Some of the pyrite may be present as framboids. Framboidal pyrite may fragment more easily than massive pyrite particles. In addition, the decomposition of pyrite is very exothermic, and it transforms to pyrrhotite and oxidizes to FeO, Fe3O4, and Fe2O3 during combustion.
4.4 Ash Deposition Formation 4.4.1 Deposition Phenomena in Utility Boilers The characteristics of a deposit depend on the chemical and physical characteristics of the intermediate ash species, geometry of the system (gas flow patterns), gas temperature, gas composition, and gas velocity. Figure 4-2 illustrates the ash deposition phenomena in utility boilers, identifying the different areas that have distinctive deposition characteristics. Ash accumulations occur via transport of particles to the heat-transfer surface, impaction of the surface, and sticking of the particles. The transport mechanisms important for ash deposition include small-particle mechanisms for particles less than 10 mm, which involve thermophoresis, electrophoresis, and vapor-phase and small-particle diffusion, and large-particle mechanisms for particles greater than 10 mm, which involve inertial impaction. Thermophoresis is a phenomenon that involves the transport of very small particles as a result of a thermal gradient from hot gases to cooler heattransfer surfaces. Electrophoresis is the transport of particles because of a difference in charge between the particle and the heat-transfer surface. Vapor-phase and small-particle diffusion occurs in the boundary layer next to the heat-transfer surface and results in transport of ash to the heat-transfer surface. Inertial impaction is a larger particle phenomenon where the particles are of a sufficient size and density to leave the air flow patterns around the tube and impinge on the surface of a tube or deposit. Deposits that form in the radiant section are called slag deposits. Deposits that form in the convective pass on steam tubes are called fouling deposits. Slag deposits are usually associated with a high level of liquid-phase components and are exposed to radiation from the flame. This is more of an arbitrary definition to aid in classification of deposits, since the term “slag” has been used to describe any type of deposit. Slag deposits are usually dominated by silicate liquid phases but may also contain moderate to high levels of reduced iron phases such as FeO and FeS. The liquid characteristics of the silicates are highly dependent on the quantities of Na, Mg, Ca, K, and Fe ash able to intereact with the silicates to flux them and lower the melting temperature of the generally refractive silicates. In addition, the initiating layers of slag deposits may consist of very fine particulate and can produce a reflective ash layer. This phenomenon is especially evident
154
1. Initial Coal 1800⬚–2400⬚F
5
75
mm
Minerals O−Na+ C
OH O−Ca+ + O-H
6 Ammonia Injection System
1200⬚–1800⬚F
3
C OH
4
Carbon and Inorganic Components
2. Combustion
Coal + Air
2000⬚–3000⬚F
2
1
SCR Reactor
Coal + Air
Sootblower 7
Liquids
Slag
Na, K SOX
Catalyst Layers
Solids
Vapors
Coal Flame
3. Early Combustion Products
4. Slag Deposits Formation
5. High-Temperature Fouling Deposit Formation Secondary Superheater Tube
Char and Fly Ash Rapidly Receding Char Surfaces Inorganic Droplets 0.1–5 mm Ca-Rich Fly Ash
Liquid Phase
Rebound – Solid Particle
6. Low-Temperature Fouling Deposit Formation Primary Superheater Tube
7. Blinding and Plugging of SCR Catalysts Sulfate Binding
Condensed Flame Volatiles Na or Ca Sulfate and Silicate Bonding
Sticking – Liquid Particle
FIGURE 4-2 Ash deposition phenomena in utility boilers.
Captive Surface Na-Rich Surface
Condensed Flame Volatiles Popcorn Ash Ca Sulfate Bonding
Characteristics and Behavior of Inorganic Constituents
155
when high organically associated calcium subbituminous coal is fired. These coals produce small CaO particles that usually form the initiating layers. Fouling deposits form in the convective passes of utilities and, in most cases, do not contain the high levels of liquid phases that are usually associated with slagging-type deposits. Fouling deposits contain lower levels of liquid phases as compared to slag deposits. The fouling deposit liquid phases usually consist of a combination of silicates and sulfates that bind the particles together. The formation of these deposits on heat-transfer surfaces can significantly reduce heat transfer. The heat transfer through a deposit is related to the temperature, thermal history, and physical and chemical properties of the deposited material. The heat transfer through a deposit of a given thickness is affected by thermal conductivity, emissivity, and absorptivity. In addition, if these fouling deposits are allowed to grow unabated, they can form very large and heavy deposits that will eventually shed under their own weight and may cause severe damage to boiler tubes they impact.
4.4.2 Slagging Deposits Slag deposits range in consistency from being lightly sintered to molten flowing material. The basic impact of these deposits is reduced heat transfer. The decrease in heat transfer is the result of a combination of radiative properties (emissivity) and thermal resistance (thermal conductivity) of a deposit. The physical state of the deposit has a significant effect on the heat-transfer properties. For example, a molten deposit will have higher emissivity than a sintered deposit. The molten deposits may be more difficult to remove than the sintered deposits. The factors that contribute to the formation of slag deposits include (1) gas flow patterns resulting in impacting and sticking particles, (2) low excess-air conditions causing localized reducing conditions that increase the quantity of low-melting-point phases, (3) a molten captive deposit surface forming that becomes an efficient collector of impacting particles, and (4) decreases in heat transfer causing increases in temperatures in the furnace and aggravation of slagging and fouling problems. Hatt [31] has suggested a means to describe deposits found in utility boilers. Four types of slag deposits can be described: (1) metallic slags that have a metallic luster and are usually associated with the combustion of pyrite-rich coals (reducing conditions cause the separation of the metallic portion from the other slag components); (2) amorphous, glassy slag that is relatively homogeneous with a high degree of assimilated ash particles; (3) vesicular slags that consist of amorphous slags, contain trapped bubbles, and have a spongelike appearance; and (4) sintered slag deposits that are only partially fused. The primary particle-bonding or -sintering mechanisms are dominated by aluminosilicate liquid phases that contain varying levels of Na,
156 Combustion Engineering Issues for Solid Fuel Systems
Mg, Al, K, Ca, and Fe. In addition, in high-pyrite coals, iron-rich silicates, FeO, and FeS can play a role in deposit initiation and growth. The initial layers of slag deposits are usually rich in small, lightly sintered particles along with larger particles that have impacted the surface and stuck. In some cases, heat transfer can be significantly reduced by the formation of a highly reflective ash layer in the radiant section. This reflective ash layer is a result of the transport of small particles to the heat-transfer surface and is a characteristic of coals that produce an abundance of small particles. These initial layers are composed of simple oxides, such as CaO, MgO, FeO, Fe2O3, and Fe3O4, and complex silicate phases that are stable in the radiant section of the boiler and are important for providing alkali-rich material for incorporation into future deposited ash. As ash layers build away from the metal surface, the insulating effect causes higher-outer-deposit temperatures and accelerated slag formation, resulting in a less dense inner deposit (usually up to 2.5 cm [1 in.] thick) and an outer, denser molten layer that is less thick.
4.4.3 Fouling Deposits Several types of fouling problems have been identified to occur in utility boilers. The problems have been defined by Hurley et al. [32] as high-temperature and low-temperature fouling. This definition is needed because the bonding mechanism of the deposits differ. In high-temperature fouling, the bonding of particles is because of silicate liquid phases, and in low-temperature fouling, the bonding mechanism is a result of the formation of sulfates. Condensed sulfur species principally in the form of CaSO4 are stable and are the matrix or bonding material in the low-temperature deposits. Table 4-3 illustrates the transport mechanisms for convective pass fouling. Convective pass fouling is significantly aggravated by the condensation of flame-volatilized elements. This vapor phase can contribute to the formation of fouling deposits.
TABLE 4-3 Summary of Convective Pass Deposits Type Conventional Fouling Upstream Reheater Upstream Enamel Downstream
Temperature, F ( C) Above 1,280 (694) 1,170–1,370 (632–744) 1,060–1,370 (571–744) All banks
Mechanism
Aerodynamic Diameter, mm
Inertial impaction
>10
Inertial impaction
>10
Small-particle diffusion/ thermophoresis Eddy impaction
<3 <10, >3
Characteristics and Behavior of Inorganic Constituents
157
4.4.4 High-Temperature Fouling High-temperature fouling occurs in regions of the utility’s boiler where temperatures exceed the stability of the sulfate-bearing phases. At lower temperatures, sulfates dominate, whereas at higher temperatures, silicates are more prone to produce liquids. In combustors burning coals that contain high levels of alkali and alkaline-earth elements, high-temperature fouling can be a significant problem. Figure 4-3 illustrates the deposits that are characteristic of high-temperature deposition and illustrates the liquid-phase formation that forms the glue to bond particles together. The deposits contain both silicate and sulfate phases. In most cases, the high-temperature fouling deposits are layered. The innermost layers that are next to the tube are rich in flamevolatilized species and upon initiation appear powdery in substance. The subsequent layers contain larger liquid-bearing particles that impact and stick. In addition, more flame-volatilized species condense on the surfaces of the particles acting as a fluxing agent to lower melting temperatures, causing them to sinter and develop strength. As a result of the insulating effect of the deposit layer on the tube, the outer layers of the deposit increase in temperature, eventually approaching the gas temperature. The higher temperature causes melting and interaction of the particles, resulting in the formation of a liquid phase. Once a liquid phase has formed on the outside of the deposit, it becomes an efficient collector of ash particles, regardless of the melting characteristics of the particles, and the deposit will have an exponential growth rate. The type of liquid phase that dominates the deposit processes is temperature-dependent.
FIGURE 4-3 Characteristics of a high-temperature fouling deposit.
158 Combustion Engineering Issues for Solid Fuel Systems
4.4.5 Low-Temperature Fouling Low-temperature ash deposition occurs at temperatures in the range of 920–1,150 K (920–1,150 C [1,688–2,102 F]) [29, 32]. In these systems, the sulfate phases dominate the matrix or bonding mechanism between particles. Detailed examination of deposits [32, 33] indicated that coals containing high levels of calcium readily form these deposits. The process involves the deposition and capture of calcium-rich particles along with silicate-rich particles. After deposition, the calcium-rich particles are sulfated, causing the formation of a bonding matrix. The bonding matrix is illustrated in Figure 4-4. Unlike high-temperature deposits, the silicates do not participate in the binding of ash particles. The silicate particles remain largely unreacted. The primary contributor to the formation of low-temperature deposits is the very small calcium-rich particles that are produced as a result of combusting the coal. These small particles are readily sulfated and can create a matrix that is very strong and difficult to remove using sootblowing techniques.
FIGURE 4-4 Low-temperature fouling in utility boilers.
Characteristics and Behavior of Inorganic Constituents
159
4.4.6 Ash Impacts on SCR Catalyst Ash-related impacts on SCR catalyst performance will depend on the composition of the coal, the type of firing systems, flue gas temperature, and catalyst design [34–38]. The problems currently being experienced on SCR catalysts include the following: Formation of sulfate- and phosphate-based blinding materials on
the surface of catalysts; and Carrying of deposit fragments, or popcorn ash, from other parts of
the boiler and depositing on top of the SCR catalysts. Licata et al. [34] conducted tests on a South African and German Ruhr coal and found that the German Ruhr coal significantly increased the pressure drop across the catalyst because of the accumulation of ash. They found that the German coal produced a highly adhesive ash consisting of alkali (K and Na) sulfates. In addition, they reported that the alkali elements are in a water-soluble form and highly mobile and will migrate throughout the catalyst material, reducing active sites. The water-soluble form is typical of organically associated alkali elements in coals. The German Ruhr Valley coal has about 9.5% ash and 0.9% S on an as-received basis, and the ash consists mainly of Si (38.9%), Al (23.2%), Fe (11.6%), and Ca (9.7%), with lower levels of K (1.85%) and Na (0.85%) [35]. Cichanovicz and Muzio [36] summarized the experience in Japan and Germany and indicated that the alkali elements (K and Na) reduced the acidity of the catalyst sites for total alkali content (K þ Na þ Ca þ Mg) of 8–15% of the ash in European power plants. Benson et al. [15] found that alkaline-earth elements such as calcium react with SO3 on the catalyst, resulting in plugging of pores and a decrease in the ability of NH3 to bond to catalyst sites. The levels of calcium in the coals that caused blinding ranged from 3–5% of the ash. The mechanisms for this type of low-temperature deposition have been examined and modeled in detail at the EERC in work termed Project Sodium and Project Calcium in the early 1990s [29, 32, 33, 39]. While the focus of those projects was specific to primary superheater and economizer regions of boilers and not SCR systems, the fundamental mechanisms are the same. Deposit buildup of this type can effectively blind or mask the catalyst, diminishing its reactivity for converting NO2 to N2 and water and potentially creating increased ammonia slip [34]. Arsenic and phosphates, which are not uncommon in low-rank coals, may also play a role in catalyst degeneration. Arsenic is a known catalyst poison [34] in applications such as catalytic oxidation for pollution control. Phosphates can occur in low-temperature ash deposits to create blinding effects, and they also occur with arsenic and can cause catalyst poisoning [29, 32, 33, 39]. Figure 4-5 shows pore blockage in a catalyst because of calcium sulfate growth.
160 Combustion Engineering Issues for Solid Fuel Systems
FIGURE 4-5 SCR catalyst blockage due to sulfate formation.
4.4.7 Deposit Thermal Properties The absorption of heat in a utility boiler that has deposits on the surfaces involves several processes. These processes are illustrated in Figure 4-6. The heat-transfer process involves radiative, convective, and conductive heat transfer. The first process to be considered is the radiative and convective heat transfer to the surface of the deposit, and the second is the conductive heat transfer through the deposit. Finally, convective heat transfer to the cooling fluid within the tube is considered.
FIGURE 4-6 Heat transfer through a deposit in a utility boiler.
Characteristics and Behavior of Inorganic Constituents
161
FIGURE 4-7 Thermal conductivity of deposits having different characteristics.
The ash chemical and physical characteristics will influence radiative and conductive processes. The ability of the deposit to reflect or reradiate the radiant energy from the flame is primarily due to the physical characteristics of the deposits. If the deposit is not molten, it will have a rough surface that will scatter the radiant energy, causing a high deposit emission. If the deposit is molten, the chemical composition of the deposit will influence its color and, therefore, the ability of the deposit to absorb heat. The thermal conductivity of a deposit is related to the physical structure and chemistry of the deposit. A deposit that is highly porous and composed of insulating materials will have a low thermal conductivity, whereas low-porosity materials containing iron will have higher thermal conductivities. Thermal conductivities for deposits have been measured by Wall et al. [40] and Benson et al. [7]. The thermal conductivities are illustrated in Figure 4-7. The properties of the deposit that influence heat transfer are emissivity and thermal conductivity. These properties vary with temperature, particle size, chemical composition, and density/porosity of the deposit.
4.5 Deposit Strength Development The development of strength in deposits can be explained through sintering theory. A good description of sintering theory can be found in Kingery et al. [41]. Sintering of a material is defined as a densification process as a result of heat treatment. Strength development in ash deposits is a dynamic and complex process. In most cases, strength development involves the formation of
162 Combustion Engineering Issues for Solid Fuel Systems
a liquid phase. Much of the discussion here will focus on liquid-phase sintering or viscous flow sintering (viscosity of the liquid phase is the most important factor in the development of strength in high-temperature deposits). Other sintering mechanisms such as solid-state sintering may be involved, but may not be dominant. Sintering in ash deposits involves complex physical and chemical transformations. An understanding of the mechanisms of sintering will allow for predicting strength development and identifying possible ways to weaken deposits. The ash particles that produce a deposit vary in size and composition. The particle-size and composition distribution of the deposited ash material provides an indication of the relative reactivity of the ash particles with each other and the gas-phase components. Assessment of the reactivity of deposited ash particles gives further insight into determining the deposit’s ability to form a liquid phase. The reactivity and ability to form a liquid can be ascertained by examining the base/acid (B/A) ratio distribution of the entrained ash particles. A B/A ratio distribution of the entrained ash that would indicate a high potential for liquidphase formation would have both acid- and base-rich components. Once deposited, the basic and acidic species could react, forming low-meltingpoint particles. A low potential is indicated by a dominance of a basic or acidic component. Little potential for liquid-phase formation exists if only one component is present.
4.6 Deposit Characterization Deposits are characterized to determine the materials and conditions (temperature and atmosphere) responsible for deposit formation. Several techniques are used routinely to determine the chemical and physical characteristics of the deposits. These include X-ray diffraction (XRD) and SEM analysis. XRD will provide the types of crystalline phases present in the deposit. The identification of the crystalline phases provides an indication of the materials responsible for the deposition problems and the temperatures at which the deposit was exposed. XRD can be misleading, since it only provides an identification of the crystalline phases, whereas many deposits lack crystallinity and consist mainly of amorphous materials. The amorphous materials are usually the components that cause bonding of ash particles that result in the formation of deposits. SEM analysis methods provide key information on the components that cause ash deposition and the conditions under which they formed. These methods include SEM morphological analysis and scanning electron microscopy point count (SEMPC). While automated SEM techniques provide quantified information about the chemistry, size, and number of phases and particles present, they do not provide all the information needed to fully characterize a material. Morphological analysis supplements the automated SEM analysis well
Characteristics and Behavior of Inorganic Constituents
163
FIGURE 4-8 SEM images of utility boiler deposits.
because it provides information about the physical relationships of the size, crystallinity, and juxtaposition of the phases present. For example, the morphological investigation of the deposits provides insight into the characteristics of the liquid phases responsible for growth of necks (particleto-particle bonding) between deposited particles and the formation of a sticky deposit surface capable of capturing impacting particles. Examples of SEM morphological analysis are shown in Figure 4-8. The backscatter electron images (BEIs) show the components that bond the deposited materials together. Chemical analyses of selected points on Figure 4-8 listed in Table 4-4 provide the identity of the material causing the bonding of the deposit. The SEMPC technique is used to quantitatively determine the number of phases present in ashes and deposits. The SEMPC involves microprobe analysis of a large number of randomly selected points in a polished cross-section of a sample. The SEMPC provides information on the degree of melting and interaction of the various deposited ash particles and provides quantitative information on the abundance of phases present in the ash. The insights provided by the SEMPC analysis are shown in Figure 4-9. By examining the phases present, the material responsible for the formation of the deposit can be identified. The SEMPC technique can be used to identify and quantify the number of melted phases and their viscosities. These melt phases are often responsible for ash deposition. In addition, various regions in deposits and individual entrained ash particles can be examined to determine the changes that occurred with operating conditions (temperature and oxygen levels). The results of the SEMPC analysis are listed in Table 4-5. The viscosity of the amorphous phases can be calculated as a function of temperature and plotted as illustrated in Figure 4-10 as a means to provide information on the temperature at which the deposit formed. For example, from Figure 4-10, the critical temperature where this deposit will develop
164 TABLE 4-4 Analysis of Points in Deposit (wt%, Elemental) Point No.
Description
Na
Mg
Al
Si
P
S
Cl
K
Ca
Ti
Cr
Fe
O
3, 5, 6, 8, 10, 12
Light material/ connective material Dark material/ particle centers Particle interfaces All points
0.7
3.0
7.4
20.9
0.0
2.0
1.2
0.7
29.7
1.2
0.1
3.2
30.1
0.1
2.1
6.9
33.4
0.0
1.9
0.8
0.3
4.2
0.3
0.0
1.6
48.5
0.0
2.4
2.8
1.0
0.0
18.7
0.8
0.3
23.6
0.2
0.0
2.2
48.0
0.4
2.5
5.9
19.0
0.0
4.6
1.0
0.5
19.8
0.7
0.0
9.9
35.7
2, 4, 11
1, 9 1–12
5. Physical Properties a. Density b. Strength c. Morphology
1. Liquid Phase a. Quantity b. Chemical Composition c. Variability d. Viscosity
Flow of Liquid Phase Convex Surface
Fly Ash Grains Bonded by Liquid Phase Liquid-Phase Responsible for Sintering 2. Derived Phases a. Unreacted Phases b. Partially Reacted Phases
Partially Reacted Aluminosilicates Quartz
3. Crystallization Products
Concave Surface
4. Deposit Chemistry a. Reactivity b. Base/Acid Ratio
Base/Acid Ratio
Viscosity
165
FIGURE 4-9 SEMPC technique, deposit insights.
166 Combustion Engineering Issues for Solid Fuel Systems TABLE 4-5 Phase Analysis from SEMPC Phase Name Silicate Phases Gehlenite Anorthite Albite Pyroxene Calcium Silicate Oxide Phases Quartz Iron Oxide Sulfate and Sulfide Phase Anhydrite Amorphous Phases Pure Kaolinite Kaolinite-Derived Illite-Derived Montmorillonite Unclassified Miscellaneous Phase Calcite-Derived
Nominal Formula
Percent
Ca2Al2SiO7 CaAl2Si2O8 NaAlSi3O8 (Ca,Mg,Fe)Si2O6 CaSiO3
11.5 3.8 2.4 1.0 1.0
SiO2 FeO
16.0 1.0
CaSO4
4.2
Al2Si2O5(OH)4 (Kaolinite is Al2Si2O5[OH]4) K1-1.5Al4Si7-6.5Al1-1.5O20(OH)4 (0.5Ca,Na)0.7(Al,Mg,Fe)4(Si,Al)8 20(OH)4 Miscellaneous
CaCO3(SO4)
FIGURE 4-10 Calculated viscosities from SEMPC analysis.
1.4 4.5 1.0 5.2 46.2 0.7
Characteristics and Behavior of Inorganic Constituents
167
significant strength and be difficult to remove through sootblowing is at about 983–1,038 C (1,800–1,900 F) (where about 30% of the liquid phases will be less than 5 log10 poise). When combined with morphological analysis, SEMPC can provide a good understanding of a particular deposition problem [42, 43]. This understanding, in turn, often leads to finding effective operational adjustments that can reduce or eliminate the ash-related problem.
4.7 Predicting Ash Behavior 4.7.1 Advanced Indices Advanced indices calculations provide coal ranking with respect to a coal’s potential to cause deposition in various sections of the boiler such as the waterwalls and high-temperature and low-temperature convective passes. An advanced index is based on advanced methods of analysis, detailed understanding of key processes that influence fouling, and full-scale performance data. Advanced indices are used to relate the coal characteristics as determined by CCSEM and CHF to ash behavior in a coal-fired utility boiler [14, 44]. From these advanced indices, fuel performance can be estimated in terms of slag flow behavior, abrasion and erosion wear, wall slagging, high-temperature silicate-based convective pass fouling, and low-temperature sulfate-based convective pass fouling.
4.7.2 Mechanistic Models Phenomenological or mechanistic models have been developed and are being used to predict the effects of ash-forming constituents as a function of coal composition, combustion conditions, systems geometry, and operating conditions [44]. Currently, the information needed by utilities and other industries as a function of firing conditions and coal composition includes the following: Deposition rate, which will provide information as to the frequency
of sootblowing in load shedding in order to clean up the unit; Heat-transfer recovery after cleaning by sootblower or load drop
methods; and Potential for catastrophic deposition.
4.8 References 1. Harding, N.S., L.L. Baxter, F. Wigley (eds). 2001. Power Production in the 21st Century: Impacts of Fuel Quality and Operation. Snowbird, UT: United Engineering Foundation. October 28–November 2, 2001.
168 Combustion Engineering Issues for Solid Fuel Systems 2. Mehta, A., and S. Benson (eds). 2001. Effects of Coal Quality on Power Plant Management: Ash Problems, Management, and Solutions. New York: United Engineering Foundations Inc.; and Palo Alto, CA: EPRI. 3. Baxter, L., and R. DeSollar (eds). 1996. Applications of Advanced Technology to Ash-Related Problems in Boilers. New York: Plenum Press. 4. Couch, G. 1994. Understanding Slagging and Fouling During pc Combustion. London: IEA Coal Research Report. 5. Williamson, J., and F. Wigley (eds). 1994. The Impact of Ash Deposition on Coal-Fired Plants: Proceedings of the Engineering Foundation Conference. London: Taylor & Francis. 6. Schobert, H.H. 1995. Lignites of North America. Coal Science and Technology 23. New York: Elsevier. 7. Benson, S.A., M.L. Jones, and J.N. Harb. 1993. Ash Formation and Deposition. In Fundamentals of Coal Combustion for Clean and Efficient Use (Smoot, L.D., ed.). Amsterdam, London, New York, Tokyo. Elsevier. pp. 299–373. 8. Benson, S.A. (ed.). 1992. Inorganic Transformations and Ash Deposition During Combustion. New York: American Society of Mechanical Engineers for the Engineering Foundation. 9. Bryers, R.W., and K.S. Vorres (eds). 1990. Proceedings of the Engineering Foundation Conference on Mineral Matter and Ash Deposition from Coal. Proc. of United Engineering Trustees Inc. Conference. Santa Barbara, CA. February 22–26, 1988. 10. Raask, E. 1988. Erosion Wear in Coal Utility Boilers. Washington, D.C.: Hemisphere. 11. Raask, E. 1985. Mineral Impurities in Coal Combustion. Washington, D.C.: Hemisphere. 12. Benson, S.A. (ed.). 1998. Ash Chemistry: Phase Relationships in Ashes and Slags. Fuel Process. Technol. 56(1–2): 168. 13. Laumb, M.L., S.A. Benson, and J.D. Laumb. 2001. Ash Behavior in Utility Boilers: A Decade of Fuel and Deposit Analyses. United Engineering Foundation Conference on Ash Deposition and Power Production in the 21st Century. Snowbird, UT. October 28–November 2. 14. Zygarlicke, C.J. 1999. Predicting Ash Behavior in Conventional and Advanced Power Systems: Putting Models to Work. In Impact of Mineral Impurities in Solid Fuel Combustion (Gupta, R. et al., eds.). New York: Kluwer Academic/ Plenum Publishers. pp. 709–722. 15. Benson, S.A., J.P. Hurley, C.J. Zygarlicke, E.N. Steadman, and T.A. Erickson. 1993. Predicting Ash Behavior in Utility Boilers. Energy Fuels. 7: 746–754. 16. Loehden, D., P.M. Walsh, A.N. Sayre, J.M. Bee´r, and A.F. Sarofim. 1989. Generation and Deposition of Fly Ash in the Combustion of Pulverized Coal. J. Inst. Energy. 119–127. 17. Zygarlicke, C.J., D.L. Toman, and Benson. 1990. Trends in the Evolution of Fly Ash Size During Combustion. Prepr. Pap.—Am. Chem. Soc., Div. Fuel Chem. 35(3): 621–636.
Characteristics and Behavior of Inorganic Constituents
169
18. Katrinak, K.A., and C.J. Zygarlicke. 1993. Size-Related Variations in Coal Fly Ash Composition as Determined Using Automated Scanning Electron Microscopy. Prepr. Pap.—Am. Chem. Soc., Div. of Fuel Chem. 38(4): 1203–1209. 19. Hurley, J.P., T.A. Erickson, S.A. Benson, and J.N. Brobjorg. 1991. Ash Deposition at Low Temperatures in Boilers Firing Western U.S. Coals. International Joint Power Generation Conference, San Diego, CA. 20. Benson, S.A., and D.P. Kalmanovitch. 1989. Characterization of a Full-Scale Ash-Fouling Deposit Formed from a Subbituminous Coal. Joint Power Generation Conference, Dallas, TX. October, p. 7. 21. Wall, T.F., A. Lowe, L.J. Wibberley, and I. Stewart. 1979. Mineral Matter in Coal and the Thermal Performance of Large Boilers. Prog. Energy Combust. Sci.. 5: 1–29. 22. Zygarlicke, C.J., and E.N. Steadman. 1990. Advanced SEM Techniques to Characterize Coal Minerals. Scan. Microsc. 4(3): 579–590. 23. Steadman, E.N., C.J. Zygarlicke, S.A. Benson, and M.L. Jones. 1990. A Microanalytical Approach to the Characterization of Coal, Ash, and Deposits. In Seminar on Fireside Fouling Problems. Washington, D.C.: ASME Research Committee on Corrosion and Deposits from Combustion Gases. 24. Galbreath, K.C., C.J. Zygarlicke, G. Casuccio, T. Moore, P. Gottlieb, N. Agron-Olshina, G. Huffman, A. Shah, N. Yang, J. Vleeskens, and G. Hamburg. 1996. Collaborative Study of Quantitative Coal Mineral Analysis Using Computer-Controlled Scanning Electron Microscopy. Fuel. 75(4): 424–430. 25. Benson, S.A., and P.L. Holm. 1985. Comparison of Inorganic Constituents in Three Low-Rank Coals. Ind. Eng. Chem. Prod. Res. Dev. 24: 145. 26. Benson, S.A., L. Kong, K.A. Katrinak, and K. Schumacher. 1997. September. Coal Quality Management System. Final Report for North Dakota Industrial Commission. 27. Katrinak, K.A., S.A. Benson, J.D. Laumb, R. Schwalbe, and W. Peterson. 2000, August. Matching Lignite Characteristics and Boiler Operation. Final Report for North Dakota Industrial Commission. 28. Sarofim, A.F., J.B. Howard, and A.S. Padia. 1977. The Physical Transformations of Mineral Matter in Pulverized Coal Under Simulated Combustion Conditions. Combust. Sci. Technol. 16: 187. 29. Hurley, J.P., and H.H. Schobert. 1992. Ash Formation During Pulverized Subbituminous Coal Combustion: 1. Characterization of Coals, and Inorganic Transformations During Early Stages of Burnout. Energy and Fuels. 6(1): 47–58. 30. Zygarlicke, C.J., D.L. Toman, and S.A. Benson. 1990. Trends in the Evolution of Fly Ash Size During Combustion. Prepr. Pap.—Am. Chem. Soc., Div. Fuel Chem. 35(3): 621–636. 31. Hatt, R.M. 1990. Fireside Deposits in Coal-Fired Utility Boilers. Prog. Energy Combust. Sci. 16: 235–241. 32. Hurley, J.P., T.A. Erickson, S.A. Benson, J.N. Brobjorg, E.N. Steadman, A.K. Mehta, and C.E. Schmidt. 1991. Ash Deposition at Low Temperatures in
170 Combustion Engineering Issues for Solid Fuel Systems
33.
34.
35.
36.
37.
38.
39.
40.
41. 42.
43.
44.
Boilers Firing Western U.S. Coals. In Proceedings of the International Joint Power Generation Conference. Burlington, MA: Elsevier Science. pp. 1–8. Benson, S.A., M.M. Fegley, J.P. Hurley, M.L. Jones, D.P. Kalmanovitch, B.G. Miller, S.F. Miller, E.N. Steadman, H.H. Schobert, B.J. Weber, J.R. Weinmann, and B.J. Zobeck. 1988, July. Project Sodium: A Detailed Evaluation of Sodium Effects in Low-Rank Coal Combustion Systems. Final Technical Report; EERC publication. Licata, A., H.U. Hartenstein, and H. Gutberlet. 1999. Utility Experience with SCR in Germany. 16th Annual International Pittsburgh Coal Conference, Pittsburgh, PA, October 11–15. Cichanovicz, J.E., and D.R. Broske. 1999. An Assessment of European Experience with Selective Catalytic Reduction in Germany and Denmark. EPRI–DOE–EPA Combined Utility Air Pollution Control Symposium: The MEGA Symposium. Atlanta, GA. August 16–20. Cichanovicz, J.E., and L.J. Muzio. 2001. Twenty-Five Years of SCR Evolution: Implications for U.S. Applications and Operation. EPRI–DOE–EPA Combined Utility Air Pollution Control Symposium: The MEGA Symposium, AWMA. Chicago, IL. August 20–23. Laumb, J.D., and S.A. Benson. 2001. Bench- and Pilot-Scale Studies of SCR Catalyst Blinding. In Proceedings of Power Production in the 21st Century: Impacts of Fuel Quality and Operations. Snowbird, UT: United Engineering Foundation. October. Flagan, R.C., and S.K. Friedlander. 1978. Particle Formation in Pulverized Coal Combustion—A Review. In Recent Developments in Aerosol Science (Shaw, D.T., ed.). New York: Wiley-Interscience. pp. 25–99. Hurley, J.P., and H.H. Schobert. 1993. Ash Formation During Pulverized Subbituminous Coal Combustion: 2. Inorganic Transformations During Middle and Late Stages of Burnout. Energy Fuels. 7(4): 542–553. Wall, T.F., A. Lowe, L.J. Wibberley, and I. Stewart. 1979. Mineral Matter in Coal and the Thermal Performance of Large Boilers. Prog. Energy Combust. Sci. 5: 1–29. Kingery, W.D., H.K. Bowen, and D.R. Uhlmann. 1976. Introduction to Ceramics, 2nd ed. New York: John Wiley & Sons. Erickson, T.A., E.M. O’Leary, B.C. Folkedahl, M. Ramanathan, C.J. Zygarlicke, E.N. Steadman, J.P. Hurley, and S.A. Benson. 1994. Coal Ash Behavior and Management Tools. In The Impact of Ash Deposition on Coal-Fired Plants. London: Taylor and Francis Publishing Company. pp. 271–284. Folkedahl, B.C., E.N. Steadman, D.W. Brekke, and C.J. Zygarlicke. 1994. Inorganic Phase Characterization of Coal Combustion Products Using Advanced SEM Techniques. In The Impact of Ash Deposition on Coal-Fired Plants. London: Taylor and Francis Publishing Company. pp. 399–408. Benson, S.A., J.P. Hurley, C.J. Zygarlicke, E.N. Steadman, and T.A. Erickson. 1993. Predicting Ash Behavior in Utility Boilers. Energy Fuels. 7(6): 746–754.
CHAPTER
5
Fuel Blending for Combustion Management Dao N.B. Duong Performance Engineer Foster Wheeler, NA,
David A. Tillman
Chief Engineer – Fuels and Combustion Foster Wheeler NA, and
Anthony Widenman
Director – Fuels Laboratory DTE Energy
5.1 Introduction Blending can be defined as “to combine or associate so that the separate constituents or line of demarcation can not be distinguished; to prepare by thoroughly intermingling different varieties or grades; . . . to combine into an integrated whole . . .” [1]. In the context of solid fuels, blending involves two or more combustible materials—coals, biomass fuels, petroleum cokes—to achieve a desired result. As the definition implies, blending is performed to achieve a consistent material. This requires an engineered approach to the process, rather than “a slug of this and a slug of that” and a hope for the best! Fuel blending has become increasingly common in the electric utility industry in recent years— and for a variety of reasons. Fuel blending has also become commonplace in process industries such as pulp and paper, cement production, minerals refining, and other economic entities where solid fuels are commonly burned. 171
172 Combustion Engineering Issues for Solid Fuel Systems
The basic principle of fuel blending is that, by combining two or more coals, or coals with other solid fuels, a new fuel is produced. While many parameters reflect the weighted average of the parent fuels (e.g., higher heating value), many other parameters reflect interactions between the fuels and—as a consequence—do not reflect the weighted average of the parent materials. Devolatilization patterns, total volatile evolution (% volatile matter), fuel nitrogen evolution, and ash fusion temperatures are among the parameters that reflect particle-particle interactions and do not reflect the weighted average of the parent materials. When firing blends of materials, these interactions between fuel types produce unusual consequences which are frequently deleterious but can also be advantageous.
5.1.1 Types of Fuel Blending For most industries burning solid fuels, blending involves two or more coals that may or may not be similar in nature. When coals that are similar in properties are blended, the effort is to maintain current or design conditions in the boiler. More commonly, blending of coals involves combining eastern or Interior Province bituminous coals with lower rank fuels. Such blend fuels include Powder River Basin (PRB) subbituminous coals and, more recently, off-shore coals such as coals from the “Ring of Fire,” or from Russia, South Africa, Australia, Germany, Poland, and other nations. PRB subbituminous coals currently supply approximately 20% of the all the electricity in the United States, about 400 million tons/year. As such, they have become a fuel of choice for utilities particularly in the western and midwestern United States. PRB coals, now transported by water, are also penetrating eastern markets. The “Ring of Fire” is a zone that is 40,000 km long with frequent earthquakes and volcanic eruptions that partly encircles the Pacific Ocean basin. Such coals include the Alaskan subbituminous deposits in the Beluga and Nenana fields. They also include Indonesian coals, Colombian coals, and other deposits from the nations living in this geologically active region. The properties of coals from these deposits are influenced by volcanic activity, as it has impacted the coalification process. Additional international coals of significance, originating from coalification processes that are fundamentally different from those producing coals found in the United States and Canada, come from such locations as Australia, Brazil, South Africa, Russia, China, Poland, and Germany. Blending of dissimilar coals serves both economic and environmental purposes as will be discussed subsequently. Additional blending activities are involved with dissimilar solid fuels—e.g., blending of coals with opportunity fuels (see Chapter 3). Such fuels may involve high calorific content materials such as petroleum coke and bitumen. Alternatively, such blend fuels may include various forms of
Fuel Blending for Combustion Management
173
biomass—wood and wood waste, agricultural wastes (e.g., manures), and fuel crops such as switchgrass, which contain lower heat contents than coals or petroleum cokes [2, 3]. A particular set of blending technologies has been developed for biomass firing with coal under the general heading of cofiring [4–7]. Other dissimilar fuels fired with coal include refusederived fuel produced from municipal solid waste, tire-derived fuel, and a host of industrial waste materials (e.g., byproduct aromatic carboxylic acid [BACA], which has been periodically cofired with coal at the Colbert Fossil Plant of the Tennessee Valley Authority [TVA]). The use of industrial byproducts as a supplementary fuel, blended with coal, is as diverse as engineering imaginations permit [2]. Liquid hazardous wastes and waste oils are also fired with coal and other solid fuels. Under certain circumstances, multiple opportunity fuels are fired with coal to achieve specific objectives [3, 8]. The scope and breadth of fuel blending depends on the situation at hand, the fuels available, the firing method involved (see Chapter 7), and the economic and environmental conditions being experienced by the unit or plant.
5.1.2 The Reasons for Fuel Blending Fuel blending can become attractive for a variety of economic reasons: coal supply security, coal cost, plant capacity goals, or similar plant objectives. A particular blend may be found to offer a “sweet spot” in the operations and maintenance of the plant. Environmental reasons also support the practice of fuel blending. Environmental reasons commonly revolve around reductions in greenhouse gas emissions (fossil CO2), sulfur dioxide (SO2) emissions, reductions in oxides of nitrogen (NOx) emissions, and better management of particulates. More recently, it has become apparent that blending eastern bituminous coals with PRB subbituminous coals provides for improved mercury capture. The mercury is produced, as a product of combustion, in a more readily managed oxidized compound. For those plants operating selective catalytic reduction (SCR) systems, blending can be used to help manage the arsenic content in the products of combustion and, consequently, the potential for catalyst poisoning. Local conditions also may promote fuel blending—particularly blending of coals with opportunity fuels. These conditions may include providing economic support to the customer of a utility through the purchase or acceptance of a byproduct as an opportunity fuel, thereby helping to offer financial benefits to that customer. If such benefits include job creation or job sustenance, then the public value of such activities becomes significant. Historically, one such example was the use of sawdust and sanderdust from the Andersen Windows factory outside St. Paul, Minnesota, by the 600 MWe cyclone-fired Allen S. King Generating Station of Northern States Power. The TVA’s use of BACA is another example of such synergies between
174 Combustion Engineering Issues for Solid Fuel Systems
power plants and their customers. For coal-fired cement kilns, accepting and burning hazardous wastes in controlled processes can serve the same function while providing economic benefit to both the generator of the waste and the cement kiln. The local situations also include helping to clean up areas after catastrophic events. Santee Cooper is among the utilities that have performed such a service after a major hurricane. Other utilities have performed such services after major ice storms, tornados, and other serious weather events. In short, there are many reasons to support fuel blending.
5.1.3 Issues for Fuel Blending All types of combustion systems have been used in association with fuel blending: pulverized coal systems (both wall-fired and tangentially fired) of all sizes from <40 MWe to >3000 MWe; cyclone boilers; stoker-fired boilers; fluidized-bed boilers; and process heating equipment such as process industry kilns. While the blending typically occurs in the coal yard (although it can occur up to and including at the burner tip or in the combustion system), the consequences are most commonly felt in the combustion system and in the post-combustion controls. The issues related to fuel blending typically result from the consequences of such actions with respect to both the combustion system and the post-combustion controls. These consequences can influence mills or pulverizers, burners, flame characteristics, heat transfer, formation of emissions, and related concerns. Specific issues associated with blending include (not exhaustive) the following: Equipment requirements to achieve blending, recognizing that
there is a difference between “bucket blending” (or “a slug of this and a slug of that”) and true blending where the resulting fuel characteristics are predictable and reproducible; Control requirements both for the blending system and for the combustion system as a whole; Bulk properties of the fuel (e.g., proximate analysis, ultimate analysis, ash elemental analysis, calorific value); Influences of blending on grinding properties, with particular attention to the grinding of one component of the blend in preference to the blend as a whole; Reactivity of the fuel both in terms of volatile evolution and char oxidation, as influenced by synergistic effects between components of the blend; Slagging and fouling tendencies caused not only by bulk chemistry properties (e.g., base/acid ratio) but also eutectic effects (e.g., as measured by iron oxide/calcium oxide ratios); and Operational considerations including training of personnel.
Fuel Blending for Combustion Management
175
Within these parameters, it is important to recognize that blending of two coals does not necessarily produce a product with the weighted average of the two parent fuels; rather, it produces a third and new fuel with some characteristics governed by the average of the two and with other characteristics that are unique to the blend. This has a profound influence not only on fuel characteristics but also on the equipment and its operations. To address these issues, this chapter uses Monroe Power Plant of DTE Energy as a case study. The blending facility and process used at Monroe Power Plant have been considered a pioneering program. Monroe has been well described in the literature [9–13]. The chapter then considers other alternative solutions by comparing and contrasting them with the Monroe approach. Monroe Power Plant is a 3100 MWe (net) plant firing blends of southern PRB and Central Appalachian (CA) coals. Typical PRB coals are Black Thunder, Antelope, and North Antelope. Typical CA coals are Long Fork, TECO, and other low- and mid-sulfur eastern bituminous coals. Blends are typically in the 60–65 percent PRB range (mass basis). The plant is composed of four units, each with 775–795 MWe (net) capacity. The plant consumes typically 25,000–30,000 tons/day of coal, or 2–3 unit trains of coal/ day; this is equivalent to 8.5–10 million tons of coal annually. Currently, the plant is ranked 16th in the nation for net station heat rate (Btu/kWh) with typical values on the order of 9,350–9,450 Btu/kWh [14]. It has the lowest heat rate among plants fueled primarily with PRB coal. Monroe Power Plant is fueled based on a sophisticated blending program.
5.2 Equipment and Controls Issues Associated with Fuel Blending Blending is not mixing. The process is designed to achieve a homogeneous, consistent, and reproducible blend of two or more fuels. “Bucket blending” is not really blending unless a sufficient number of transfer points are in the transport line to achieve the blend desired. Even then reproducibility is questionable. Blending is designed to provide a combustion unit with fuels of specific properties, and these properties will be unique to each blend. Further, these properties will not always be the weighted average of the parent coal properties. Blending permits changing the ratio of constituents—and the ultimate fuel properties or characteristics—by changing the blend. To accomplish this, both equipment and controls are necessary. Equipment issues include installing sufficient systems to achieve the blending objectives, installing information systems sufficient to understand what the blend is, and utilizing control systems sufficient to manage the blend.
176 Combustion Engineering Issues for Solid Fuel Systems
5.2.1 The Blending System at Monroe Power Plant The Monroe Power Plant blending system has been previously described [9–13] and is summarized here. At Monroe, the blending system begins with the rail car unloader or boat unloader. The rail car unloader—a rotary dumper system—feeds coal to an overhead trestle (see Figure 5-1). From the overhead trestle, coal is deposited onto three piles: a low-sulfur eastern bituminous coal pile, a low-sulfur western subbituminous (PRB) coal pile, and a mid-sulfur eastern bituminous coal pile. In reality, the system has duplicates of each coal pile—with one on each side of the trestle. All the piles feed an underground coal belt carrying the desired blend of coal. The underground coal belt is equipped with variable speed ploughs, belt scales, and load cells such that the blend can be controlled from the fuel supply control room. The accuracy of the blend has been shown to be 1%. However, the minimum flow of any coal, when blend accuracy is required, is nominally 15%. The blended coal is then transported to the plant on the CV04 belt; it passes through the cascade room and is distributed to the 28 silos in the plant.
FIGURE 5-1 Overview of the Monroe Power Plant. The coal transport system is prominent at the back of the plant.
Fuel Blending for Combustion Management
177
Control of the blending is of critical importance. ECG consultants of Akron, Ohio, designed and installed a digital fuel tracking system—AccuTrack—to monitor the flow of fuel from unloading to burner tip [9, 15]. This software system monitors and models the flow of coal and the blending of coal to, and within, each silo. Using the QC, Inc. X-ray fluorescence (XRF) coal analyzer, AccuTrack monitors the percentage coal blend and the essential coal parameters on the CV04 belt and at various levels within the silos [13]. Figure 5-2 is an overview of the coal handling system as depicted by AccuTrack. Table 5-1 identifies the parameters measured by the XRF analyzer. The AccuTrack system operates on a PI platform supplied by OSIsoft, Inc. As such, it is able to integrate fuel supply (ton/h), fuel quality, and operational parameters for maximum blend optimization. Figure 5-3 shows the monitoring of the coal blend. It is a screen available to all supervising operators and shift supervisors identifying and quantifying key fuel parameters in the fuel being burned; the fuel that will be burned in 1 hour, 2 hours, and 4 hours; and the fuel being put into the silos from CV04. It is color coded such that a value with green in the background is good, yellow indicates caution, red indicates trouble, and blue indicates that the value shown is better than necessary for successful operation. For example, if the fuel exhibits 10,600 Btu/lb, then the color will be blue because full load can be achieved successfully—and without deleterious side effects—at lower calorific values. The Monroe blending system, which is totally automated, is among the most extensive and sophisticated that has been constructed to date; in 1982 DTE Energy spent $400 million to install the system, and additional funds have been used to upgrade it annually [9]. The degree of sophistication is dictated by the high percentages of PRB being fired—currently between 60% and 65% depending on coal quality—and by the quantity of coal being supplied to the plant. Other utilities and power plants have taken alternative approaches depending on plant size, type, and blending objectives.
5.2.2 Alternative Blending Systems Alternative blending systems have been designed and installed based on individual plant objectives and the trade-off between capital expense and labor (O&M) expense. Virtually all blending facilities have some means to meter each coal in the blend and to control the relative percentages. Many such blending facilities are designed not only for coal/coal blends, as is the case at Monroe, but also coal-opportunity fuel blends [5]. The blending facility constructed at NiSources’ (NIPSCO) Bailly Generating Station by Foster Wheeler was designed to manage all opportunity fuels ranging from wood waste to petroleum coke [5, 7]. Since Bailly Generating Station is composed of two cyclone boilers, it exhibited significant flexibility in utilizing opportunity fuels.
178
DETROIT EDISON Monroe Power Plant Current Time: 3/31/2006 9:51:08 AM
Main Mimic TH 2
SHIP BOOM
TH9
TRIPPER ZM16
BNO1
CV07
MESB
FE11
FE09 CR01
CV06
ZM01 FE01 CV01 ZM03
ZM02
TH4
CV19
CV10
FE03
CV08
FE05
FE04
CV09
FE06
ZM06
FE07
FE02 CV02 ZM04 TH1 FE25
L
FE08
ZM07
CR02 CV20
ZM15 TRIPPER
CV03
FE10
CVR1A
FE12
ZM05
CV05
CV11
CVR1 ZMC4 ZM C4
BREAKER HOUSE
CVR2
TH11
CVC4
CVC2 R2 FG
C1A CAR C1B DUMPER C1C
C-5A CHCC C-6A
C-5 C-6
POWERPLANT
C-7A
TH3 TH12
C-8A
C-7 1
2
3
4
5
6
7
C-8 8
9
10
UNIT 2
Main
Cascade Routes Reclaim Routes
12
11
13
14
15
UNIT 1
Silos unit 1
16
17
18
19
20
21
UNIT 3
Silos Unit 2
Silos Unit 3
FIGURE 5-2 The digital fuel tracking system at Monroe Power Plant (Source: [8]).
22
23
24
25
26
27
28
UNIT 4
Silos Unit 4
Train Unloading
Ship Unloading
Fuel Blending for Combustion Management
179
TABLE 5-1 Parameters Measured by Monroe Online Analyzer Basic Fuel Parameters Higher Heating Value (Btu/lb as received) Moisture Percentage Volatile Matter Percentage Fixed Carbon Percentage Sulfur Percentage Volatile/Fixed Carbon Ratio Ultimate Analysis (by calculation from proximate analysis) Sulfur Loading (lb/106 Btu as SO2)
Ash Chemistry Parameters Ash Percentage Ash Loading (lb/106 Btu) Silica (Percentage) Alumina (Percentage) Titania (Percentage) Iron Oxide (Percentage) Magnesium Oxide (Percentage) Calcium Oxide (Percentage) Sodium Oxide (Percentage) Potassium Oxide (Percentage) Base/Acid Ratio Slagging Alkalinity Factor Calcium/Iron Ratio
The system, shown in Figure 5-4, consisted of a fuel receiving yard where one or more opportunity fuels could be received and bucket blended into a rough pile, and then a conveying system for actual blending. The conveying system consisted of a Stamler reclaimer supplying fuel to a metered air slide conveyor. The weigh-belt scale on the air slide conveyor fed quantity data to a PLC. The weigh-belt scale on the main conveyor also fed quantity data to the PLC. Using simple mathematics, the PLC allowed the plant to control the blend of opportunity fuels and coals. Subsequent to the initial program, the blending facility was used to test fire blends of PRB coals with traditional plant coals. Unlike the Monroe system, the Bailly system requires additional operators when it is used. Mobile equipment is required both to build the opportunity fuel pile and to feed the Stamler reclaimer. Price differentials between fuels can make such operations worthwhile, however. The Bailly system also does not give the operators detailed knowledge of the fuels being burned; however, most power plants lack this capability even for their base fuels. Beyond the electric utility industry, blending is very common in the process industry. Cement kilns blend tire chips and whole tires with coal for operations. Cement kilns also blend hazardous wastes into their fuel supply as a means for cost control. Pulp and paper manufacturers, operating stoker-fired power boilers, typically blend hogged wood waste with coal, oil, and other residuals in a wide variety of blends. Less attention is paid to efficiency than plant reliability and availability. Blending, then, is used in a wide array of applications.
180
Unit 1 Fuel Quality #1 11500 30 10
Unit 1 Fuel Quality Screen
Characteristics Blend
Firing
Belt TPH Unit Silo 0 C5 3 15 3 15 C5A 0
NOW 1 hrs 2 Hrs 4 Hrs 14 - 63 - 24 14-63-24 15-65-20 15-65-20
On C4 0-0-0
10500 10000 9500 15 5 4/21/2005 7:35:39 AM
Heating Value A/R (Btu/lb)
10018
10018
9906
9919
10019
Moisture (%)
20.45
20.45
21.07
21.13
20.24
Unit 1 Fuel Quality #2 0.6 100 20
Ash Loading (lb/MBtu)
6.46
6.46
6.53
6.50
6.47
0.4
Base/Acid Ratio
0.38
0.38
0.39
0.39
0.40
SO2 Rate
1.11
1.11
1.05
1.05
1.07
Silica + Alumina
59.42
59.41
58.92
59.02
57.60
Slagging Alkalinity
12.98
13.01
13.65
13.47
13.38
0.72
0.73
0.72
0.74
Fuel Volatility Ratio (VM/FC)
0.71
North Furnace Temperature
2187
South Furnace Temperature
2329
Unit 2
Unit 3
Unit 4
8.00 Hour(s)
4/21/2005 3:35:3
8.00 Hour(s)
4/21/2005 3:35:3
0.2 0 50 10 4/21/2005 7:35:39 AM Furnace Temperatures 2260 2420
U1NFURN-GA U1NFURN-GA 2187.1 FF U1SFURN-GA U1SFURN-GA 2329.1 FF
2200 2150
Limits
2100 2240 4/21/2005 7:35:39 AM
4/21/2005 3:35:39 PM
Operator Guidance 1. If Heating Value A/R is below 10,200 Btu/lb, then unit efficiency will deteriorate and mills will be stressed harder. Watch Furnace Temperature at full load. 2. If Moisture is above 20 percent, then unit efficiency will deteriorate and mills will be stressed harder. Watch Furnace Temperature at full load. 3. If Ash Loading is more than 7 lb/MBtu, then opacity excursions are expected at full load. 4. If Base/Acid Ratio is less than 0.2, then opacity excursions are expected at full load. 5. If Base/Acid Ratio is between 0.2 and 0.25, then there is a potential for opacity problems at full load. 6. If Base/Acid Ratio is more than 0.5, then slagging can be expected at full load. 7. If Base/Acid Ratio is between 0.4 and 0.5, then slagging can be caused by the fuel at full load. 8. If Fuel Volatility Ratio is below 0.25, then combustion will occur in upper furnace, watch Furnace temperature. 9. High Furnace Temperature values should provide a caution to increase excess O2 or reduce load..
FIGURE 5-3 Computer screen for monitoring the coal blend at Monroe Power Plant. Note that the blend can be viewed in terms of what is now burning, what will burn in 1–4 hours, and what is being loaded into the silos (Source: [9]).
Fuel Blending for Combustion Management
181
FIGURE 5-4 The blending system at Bailly Generating Station of NIPSCO, used to add measured amounts of petroleum coke, wood waste, mixtures of the two, or other fuels to the main coal feed to the plant. Software integrating the two feed streams controlled the blend (Source: [6]).
A wide variety of fuel blending systems has been designed between the Monroe approach and the Bailly approach. These systems exist along a continuum between capital costs and O&M costs. For some process industries, this blending occurs with less than the sophisticated information and controls more commonly found in electricity-generating stations. The blend system designs of both utilities and industries depend on real estate availability as well as plant and corporation philosophy.
5.3 Fuel and Combustion Effects of Blending Blending is performed to achieve certain effects: reduced fuel cost, reduced SO2 emissions, reduced NOx emissions, and other effects. To some extent, the requirement for mercury capture will play into fuel blending strategies, since the oxidation state of mercury impacts the ease with which it is captured. Because blend fuels do not behave as the average of the parent fuels, they must be treated as a new fuel. The new fuel will burn differently from the parent coals or the average of the parent coals. In addition, the inorganic constituents also behave differently from the averages. Knowledge of blend characteristics such as basic principles, combustibles, ash and inorganic constituents, basic properties, and behavioral aspects are necessary to better manage blending issues.
182 Combustion Engineering Issues for Solid Fuel Systems
With this knowledge, one may “design” new fuels that meet specific technical, economic, and environmental concerns. Several types of blends exist, with two types dominating: coal-on-coal blends and opportunity fuel/coal blends. Both are considered here.
5.3.1 Blending Overview As previously discussed, blend fuels of interest are typically low-rank coals such as PRB subbituminous coals and off-shore coals such as coals from the “Ring of Fire.” These fuels are typically blended with high heating value, i.e., high Btu, bituminous coals to achieve specific effects. Opportunity fuel/coal blending frequently focuses on petroleum coke, which is a high-Btu, high-sulfur, low-ash fuel. It is also a lowvolatility fuel. Further, the ash typically contains significant concentrations of vanadium and nickel. Other opportunity fuels blended with coal include the biomass fuels—largely wood and wood waste. These are high-moisture, high-volatile, low-Btu, low-ash fuels. The ash associated with these fuels is high in alkali metals and alkaline earth elements: Potassium, calcium, sodium, and magnesium are commonly found in significant concentrations.
5.3.2 The Monroe Power Plant Case Study Understanding the fuels and combustion effects of blending requires a careful assessment of the process of combustion. Only then can the subtleties associated with blending be explored. Again, the Monroe Power Plant experience is used as the basis for the discussion; Monroe provides an effective case study because it is totally dependent on the blending of fuels. 5.3.2.1 Development of Combustion Models as an Analytical Tool To perform complete combustion analyses, combustion modeling is necessary. The Monroe power plant in Monroe, Michigan, has developed extensive combustion models of its boilers. The models incorporate three major combustion mechanisms: drying, pyrolysis, and char oxidation. The models also directly address pulverizer performance including preferential grinding, sieve analysis, furnace exit gas temperature (FEGT), flame temperatures, furnace dimensions, and detailed coal characteristics. The models directly address the burnout location for coal particles of varying particle sizes (sieve analysis), and if the particles do not experience complete combustion, the model calculates percent completion. Drying, the process of driving all the moisture out of the fuel particle, is an endothermic process which is heat transfer controlled with the required
Fuel Blending for Combustion Management
183
amount of energy being largely a function of the fuel particle moisture content. Drying and heating initiate physical changes, predominantly particle shrinkage, as the moisture leaves. Equation [5-1] shows the relationship between the general conductance of the fuel particle and its drying time. ðT1 T2 Þ q ¼ kA x
q ¼ general conductance [Btu/hr]
h
k ¼ thermal conductivity
ðBtuftÞ ðhrft2 FÞ
[5-1]
i
A ¼ surface area [ft2] T1 ¼ surface temperature [ F] T2 ¼ particle center temperature [ F] Pyrolysis, the next major process of coal combustion, can be defined as the heating of fuel particles in the absence of oxygen to produce volatile matter and char—both for subsequent combustion. Pyrolysis or devolatilization can be represented by equation [5-2]: Ca Hb Oc þ heat ! H2 O þ CO2 þ H2 þ CO þ CH4 þ C2 H6 þ CH2 O þ þ tar þ char
[5-2]
The solid coal particles are composed of a series of ring structures connected as either fused rings or by ether linkages, carbon, sulfur bridges, or similar associations. Associated with the fuel backbones are functionalities such as hydroxyls, methoxyls, methyl groups, and heteroatoms like sulfur, nitrogen, and chlorine attached either by themselves or in the functional groups. Pyrolysis can be broken down into two stages: the breaking of linkages between ring structures on the backbone and then the reduction of the evolved molecules into smaller fragments by subsequent bond breaking. The second stage involves functional groups breaking away from the fuel matrix. Volatiles evolving include carbon monoxide (CO), methane, ethane, CxHx–xþ2 volatiles, OH radicals, H2O, NH3, HCN, and other products. During this process, char is produced from the solids evolving the volatile matter. Pyrolysis rates are controlled by both heat transfer and chemical reactions; they can be expressed as Arrhenius equations with reaction times derived from equation [5-3]:
E^ Re ¼ Aexp ^T R
[5-3]
184 Combustion Engineering Issues for Solid Fuel Systems
Re A E^ ^ R T
¼ reactivity ¼ pre-exponential constant [1/s] ¼ activation energy [kcal/mol] ¼ ideal gas constant [kcal/kmol K] ¼ gas temperature [K]
The final major mechanism is char oxidation. Char oxidation proceeds largely by the reaction of surface carbon to the surrounding oxygen producing CO; CO then breaks away from the char matrix and oxidizes into CO2. The following general reaction equation [5-4] describes char oxidation: C þ O2 ) CO2
[5-4]
Char oxidation involves consumption of the solid particle by heterogeneous gas-solids reactions. Char oxidation is both diffusion and chemical kinetics controlled, with diffusion of O2 into the pores rate limiting at the higher temperatures and chemical kinetics rate limiting prominent at lower temperatures. Therefore, the general equation [5-5] reflects both diffusion and chemical kinetics: dmc 2 12 ¼ pd ke ro2 16 dt
[5-5]
When equation [5-5] is integrated from initial char mass to zero, the reaction time is as follows, shown in equation [5-6]: tc ¼
tc rci di ke ro2
¼ ¼ ¼ ¼ ¼
rci di 4:5ke ro2
[5-6]
char burnout time initial char density initial diameter effective rate constant oxygen density
By incorporating parameters such as furnace dimensions, FEGT, flame temperature, preferential grinding, sieve analysis, and fuel properties including fuel reactivity, the models calculate the burnout location of varying particle sizes. Computational fluid dynamics (CFD) modeling is necessary to look at burner behavior for the purposes of burner design and particle burnout modeling. CFD modeling can be used to model the flow patterns, temperature, or mixing profiles of a fluid at the burners.
Fuel Blending for Combustion Management
185
5.3.2.2 Fuel Effects of Blending at Monroe The Monroe models provide critical insights into the effects of blending on fuel properties. Certain characteristics may be evaluated based on the average of the parent fuels. These are proximate and ultimate analysis, higher heating value, ash elemental analysis, base/acid ratio, and the percent fluxing agent in the ash. Figures 5-5 and 5-6 are representations of the relationship between the percentage of southern PRB in the blend versus higher heating value and base/acid (B/A) ratio. By taking the weighted average of the parent coals, one can derive certain characteristics. Note that B/A ratio is not totally linear; however, the polynomial expression is sufficiently close that linearity can be assumed. 5.3.2.3 Volatility and Volatile Release Patterns Volatility and volatile release patterns are governed by the interaction of fuels, not by the weighted average. Research at Penn State’s Energy Institute, conducted for Monroe Power Plant, revealed the impacts of blending on volatility of fuels. Figure 5-7 is representative of the volatile release patterns for Antelope PRB subbituminous coal; Long Fork mid-sulfur bituminous coal; a blend of 60% Antelope, 40% Long Fork; and a blend of 70% Antelope, 30% Long Fork. The release patterns for the blended fuels are faster and more complete than either parent fuels. Therefore, pyrolysis is 10700
Higher Heating Value (Btu/lb)
10600 10500 HHV (Btu/lb, a/r) = 12603 - 37(PRB) 10400 10300 10200 10100 10000 9900 PRB is percent southern PRB in blend; 100 - PRB is percent Central Appalachian coal in blend
9800 9700 50
55
60
65
70
75
80
Percent Southern PRB in Blend
FIGURE 5-5 Influence of percentage of PRB on the heat content of the blended fuel at Monroe Power Plant, assuming average coals as delivered. Note that this is a linear relationship (Source: [11]).
186 Combustion Engineering Issues for Solid Fuel Systems 0.57 0.56 B/A = 0.419 – 0.0003(PRB) + 3⫻10–5(PRB)2 Base/Acid Ratio of Blend
0.55 0.54 0.53 0.52 B/A is base/acid ratio 0.51 PRB is percent southern PRB in blend 0.5 0.49 0.48 50
55
60 65 70 Percent Southern PRB in Blend
75
80
FIGURE 5-6 Influence of percentage of PRB on the base/acid ratio of the blended fuels at Monroe Power Plant, assuming average coals as delivered. Note that this is not linear but approaches linearity (Source: [11]). 90.0% 60% Blend
Antelope
80.0% 70% Blend Total Volatile Yield (%)
70.0% Long Fork
60.0% 50.0% 40.0% 30.0% 20.0% 10.0% 0.0% 0
500
1000
1500
2000
2500
3000
3500
DTR Temperature (8F)
FIGURE 5-7 Volatile release patterns for southern PRB, Central Appalachian bituminous coals, and various blends of the two (Source: [11]).
Fuel Blending for Combustion Management
187
a faster process for the blends compared to the parent fuels. Further, devolatilization is more complete. The kinetics of devolatilization are also influenced significantly by blending. When one looks at activation energies required to initiate pyrolysis for both the parent fuels and their blends, it is again apparent that pyrolysis is more readily initiated for the blends than for the parent coals. The activation energies are less for the blends than either parent fuels. Figure 5-8 shows the Arrhenius plots for the parent coals, Figure 5-9 shows the Arrhenius plot for a blend, and Figure 5-10 compares the activation energies for the parent coals and blends. Note the lower slope of the plot for the blend compared to either parent coal. Note also the comparison of activation energies, indicating that the interaction of particles is significant. Although no mechanisms have been elucidated to explain these results, it is believed that oxygenated fragments and radicals leaving the PRB coals are interacting with bituminous coal particles and increasing the volatile release from the higher Btu fuels. Volatile release from blends also influences the ability to reduce NOx emissions by staging. NOx emissions due to fuel nitrogen can be minimized by staging the combustion and/or by knowledge of the nitrogen evolution patterns of the coal. Research by Baxter et al. [16] documents that the NOx reduction associated with staging is enhanced if the nitrogen volatiles evolve from the coal particles more quickly, rapidly, and completely. These nitrogen volatiles are most likely to evolve in a fuel-rich environment where the impacts of staging are most prominent. Tests performed by a drop tube reactor (DTR) provided information necessary to analyze the nitrogen evolution patterns for the Monroe Power Plant coals. Figure 5-11 is a representation of the nitrogen evolution pattern for the blend fuel Antelope subbituminous coal, Long Fork bituminous coal, and their blends. Note that the volatile nitrogen evolution pattern is more effective with the blended fuels than with either parent. Volatility, then, can be enhanced by blending. 5.3.2.4 Char Oxidation Char oxidation is also influenced significantly by fuel blending. Again, the area of influence is in reaction kinetics. Figure 5-12 shows the activation energies associated with the char oxidation kinetics for both the parent coals—Antelope and Long Fork—and for the 60% PRB and 70% PRB blends. Note that the PRB char oxidation kinetics govern the entire process. Once PRB is introduced into the blend, all char oxidation kinetics are essentially equal to those of the PRB coal. 5.3.2.5 Ash Chemistry Ash chemistry is heavily influenced by blending and the creation of eutectics that reduce ash fusion temperatures. These effects have been well recognized and studied over the past several years. CaO-SiO2 systems have
188 10
10
E = 4.43 kcal/mol A= 6.03 1/sec R = 1.986 kcal/kmol-K
1600C
Reactivity (1/sec)
Reactivity, R (1/sec)
E = 4.72 kcal/mol A = 6.46 1/sec R = 1.986 kcal/kmol-K 1500C 1200C
1
0.1 0.0
1000C
800C 600C
0.2
0.4
0.6
0.8
1/T X 1000 (1/K)
1.0
1.2
1.4
1600C
1500C 1000C
1
0.1 0.0
800C
0.2
0.4
0.6
0.8
600C
1.0
1.2
1.4
1/T X 1000 (1/K)
FIGURE 5-8 Arrhenius plots for devolatilization of Central Appalachian coal (left) and Southern PRB coal (right) (Source: [10]).
Fuel Blending for Combustion Management
189
10
Reactivity, R (1/sec)
E = 3.33 kcal/mol A = 4.88 1/sec R = 1.986 kcal/kmol-K 1600C 1500C 1000C
1
800C
600C
0.1 0.0
0.2
0.4
0.6 0.8 1/T X 1000 (1/K)
1.0
1.2
1.4
FIGURE 5-9 The Arrhenius plot for devolatilization of a blend of 70% PRB/30% Central Appalachian coal. Note the lower slope of the plot (Source: [10]).
Pyrolysis Activation Energy (Kcal/mol)
Pyrolysis Activation Energy vs. Percent PRB in Blend 5 4.5 4 3.5 3 2.5 2 1.5 1 0.5 0
0.00%
60.00% 70.00% Percent PRB in Blend
100.00%
FIGURE 5-10 Activation energies for a Central Appalachian coal, a Southern PRB coal, and two selected blends. Note that the activation energies for the blends are lower than for either parent coal (Source: [10]).
190 Combustion Engineering Issues for Solid Fuel Systems
Percent Nitrogen Yielded as Volatile Matter
90.00% 70% Blend 80.00%
60% Blend
70.00% 60.00%
Southern PRB Coal
50.00% 40.00% 30.00% 20.00% Central Appalachian Coal 10.00% 0.00% 0
500
1000
1500 2000 DTR Temperature (8F)
2500
3000
Char Oxidation Activation Energy (kca/mol)
FIGURE 5-11 Volatile nitrogen evolution patterns for a Central Appalachian coal, a Southern PRB coal, and two selected blends (Source: [10]).
Char Oxidation Activation Energy vs. Percent PRB in Blend 35 30 25 20 15 10 5 0
0.00%
60.00% 70.00% Percent PRB in Blend
100.00%
FIGURE 5-12 Activation energies for char oxidation of a Central Appalachian coal, a Southern PRB coal, and two selected blends. Note that the PRB dominates the process (Source: [10]).
Fuel Blending for Combustion Management
191
been studied for several years. CaO, by itself, has a liquid temperature of 2,570 F. However various combinations drive ash fusion temperatures to the 2,000–2,100 F region and, in some cases, to the 1,400–1,500 F region [11]. As increasing percentages of PRB coals—particularly southern PRB coals—are introduced into a blend, these increase the calcium component in the ash and can depress the ash fusion temperatures. Calcium is a fluxing agent, as is iron. Calcium is contained, in significant percentages, in the ash from PRB coals. Iron is more prevalent in the eastern and Interior Province bituminous coals and generally is a function of sulfur content. Higher sulfur concentrations correspond to higher iron concentrations. Together, calcium and iron significantly depress ash fusion temperatures. Calcium oxide/iron oxide ratios are of particular significance. Research cited in Singer [17] shows that, as the blends bring the ratio toward 1.0 from either side, the ash fusion temperatures are continually depressed. Research by Widenman also shows eutectic effects but not of the same characteristics as shown by Singer. The Widenman data are shown in Figures 5-13 through 5-15. These data show ash fusion temperatures as a function of flux loading (lb/106 Btu of CaO þ Fe2O3) and of Fe2O3/CaO and CaO/Fe2O3 molar ratios. Fluxing values are compared to initial-oxidizing and hemispherical-reducing ash fusion temperatures. Note that there are significant curves at both ends of the plots generated, documenting the eutectic effects of the calcium and iron oxides. Note also the comparison between Figures 5-14 and 5-15. Figure 5-15
2,700 Initial Oxidizing y = −2308.1x3 + 7872.8x2 − 9228.5x + 5960.8 R2 = 0.8765
Ash Fusion Temperature (8F)
2,600 2,500 2,400 2,300 2,200 2,100 2,000 1,900 1,800 0.6
Hemispherical Reducing y = −5625.4x3 + 17449x2 − 18039x + 8501.5 R2 = 0.7837
0.7
0.8
0.9 1 1.1 1.2 Fe2O3 + CaO (lb/106 Btu)
1.3
1.4
FIGURE 5-13 Ash fusion temperatures as a function of fluxing agent loading.
1.5
192 Combustion Engineering Issues for Solid Fuel Systems 2,700 Initial Oxidizing
Ash Fusion Temperature (8F)
2,600
y = 2469.2x0.0547 R2 = 0.8737
2,500 2,400 Hemispherical Reducing
2,300
y = 344.67x3− 803.21x2+ 672.38x + 2147.7 R2 = 0.7251
2,200 2,100 2,000 0.00
0.20
0.40
0.60 0.80 1.00 1.20 Fe2O3/CaO Molar Ratio
1.40
1.60
1.80
FIGURE 5-14 Ash fusion temperatures as a function of Fe2O3/CaO molar ratios. Note the curvature of the plots, indicating eutectic effects.
2,600 Initial Deformation, Oxidizing Conditions
Ash Fusion Temperature (8F)
2,550
y = 4.5951x2 − 80.017x + 2536.8 R2 = 0.8144
2,500 2,450 2,400 2,350 2,300 2,250 2,200 2,150 2,100 0.00
2.00
4.00 6.00 8.00 CaO/Fe2O3 Molar Ratio
10.00
12.00
FIGURE 5-15 Initial deformation ash fusion temperatures as a function of CaO/ Fe2O3 molar ratios. Note the low point at 8:1 ratio, contrasting with the data in Figure 5-14.
Fuel Blending for Combustion Management
193
inverts the ratio used for analysis. One can interpret the data to imply that, as calcium oxide is added, ash fusion temperatures will continuously decline. Alternatively, particularly with Figure 5-15, one can interpret the data to imply that, at a calcium oxide/iron oxide ratio of 8:1, the low point in ash fusion temperatures is reached. Above that ratio, ash fusion temperatures may increase again. This is consistent, conceptually, with the work cited in Singer although the numerical values are substantially different. The eutectic effects of calcium and iron are particularly significant in evaluating blends of coals for firing in a boiler. They can result in “the worst of all worlds” from an operational perspective. For the Monroe Power Plant, calcium and calcium/iron effects are particularly significant. For these reasons, they are closely monitored using the online XRF analyzer.
5.3.3 Fuel Effects for Other Locations Fuel effects for other locations are similar to those associated with the Monroe Power Plant. However, experience shows that the influence of blending on volatility or reactivity is most prominent when low-rank coals—or biomass fuels—are present. These fuels contain the most volatile and most oxygenated compounds of significance. As coals get to higher and higher ranks, the interaction between volatile matter and coal particles is of less significance, leading to the suggestion that the oxygenated compounds are responsible for the increased reactivity of the blends relative to the parent coals. Blends between subbituminous coals and lignites have also been studied, and with results comparable to those associated with bituminoussubbituminous blends. Such results include volatility kinetics, total volatiles evolved, and nitrogen volatile evolution [11, 12]. Ash behavior is similar for opportunity fuel/coal blends relative to coal/coal blends. The volatility and char oxidation concerns are typical of coal/low-rank coal blends. Of particular significance, however, is the high alkalinity of the biomass fuels and the vanadium content of the petroleum cokes. Vanadium can depress ash fusion temperatures—particularly in an oxidizing environment. It can lead to increased slag formation, particularly in the presence of such slag formers as calcium. However, because petroleum coke is low in ash, the vanadium is more of a promoter of slag from the base coal than a direct cause of slag formation. Blends of coal/biomass also have been studied with similar consequences noted [18].
5.4 Operational Issues with Fuel Blending Operational issues associated with the practice of fuel blending must be considered and addressed by combustion engineers at the plant, at the corporate office, and at various vendors. Critical operational issues such as
194 Combustion Engineering Issues for Solid Fuel Systems
FEGT management are well documented. Others, such as the impact of changing blends, are less well chronicled.
5.4.1 Managing Inorganic Constituents Operationally, the first issue is management of inorganics including furnace temperature control—focusing on FEGT. Blend fuels tend to be lower in heating value compared to design fuels; therefore, a higher throughput is necessary to maintain capacity. Higher throughputs present operational issues to plants with smaller precipitators. These issues may be opacity spikes or an increase in the opacity baseline. Through management of the fuel quality and/or an increase in the capacity of the precipitator, opacity issues can be minimized. Blend fuels such as PRB will also leave deposits of reflective ash on the boiler walls, thus reducing the heat transfer capability and increasing the temperature. Managing the inorganics, particularly as they influence FEGT, involves installing and maintaining sufficient sootblowing to remove reflective ash deposits. This may include installing water lances, water cannons, and like technologies. It is recognized that overuse of sootblowers or water cannons can have significant negative impacts on the boiler, causing tube erosion and/or tube cracking. However, blends of coal and opportunity fuel/coal with high alkalinity in the ash can cause sufficient problems to require these extensive cleaning systems. Beyond these problems, managing the slagging potential of the blend is an ongoing and ever present issue. This involves careful fuel analysis and management of the fuel constituents. As necessary, blending away from severe slagging conditions—particularly if sootblowing capability is compromised—becomes critical.
5.4.2 Managing the Fire If the blend fuel is higher in moisture compared to the design fuel, the fireball will be raised in the primary furnace. Further, changing the fuel can change the flame length and flame shape. Increased temperatures will promote slagging incidences. Furthermore, mill capacity, primary air temperatures, and flows will limit the percentage of PRB blend fuel burned. As the moisture increases, the primary air flow and temperature must also increase to maintain the thermal performance of the mill. With high percentages of low-rank fuels, the operating window for the unit (between maximum and minimum capacity) becomes increasingly narrow.
5.4.3 Managing Blend Changes For blending to be optimized, plants must have the ability to change blends to meet technical or economic conditions. Technical conditions include
Fuel Blending for Combustion Management
195
responses to in-seam variability of coal; emissions considerations including meeting New Source Review annual targets; and operational issues including opacity, slagging, and fouling. Economic issues include the relationships between fuels costs and product costs. For electricity-generating stations, the dispatch value of electricity governs what a power plant can pay for fuel, since fuel costs are about 65–70% of the total generation cost, which is plant dependent. Changing blends can be performed for short-term time frames (e.g., off-peak hours, weekends) or long-term time frames (e.g., shoulder months) depending on fuel quality and market conditions. During blend changes, fuel instability can occur. Blending continues in the silos or bunkers. Well-designed and constructed silos or bunkers typically can experience mass flow or plug flow where there is uniform flow from the inlet of the silo or bunker to the outlet. If significant ratholing and funnel flow occur in the bunkers and silos, the time that a certain blend reaches the pulverizers and burners changes significantly. The characteristic flow of the fuel determines the quality of the fuel experienced by the mills and burners. Studies at the Monroe Power Plant determined silo flow characteristics by doping the coal with a tracer agent. These studies led to understanding the performance of the silos or bunkers during blend changes. Figure 5-16 demonstrates that the silos at Monroe, like most silos and bunkers, do not exhibit mass flow or plug flow, but exhibit blend instability during blend changes.
Barium Concentration 500 450
Barium (mg/kg)
400 350 300 250 200 150 100 50 0
0
30
60
90
120 150 180 210 240 270 300 330 360 390 420 450 480 510 540
Time, minutes
FIGURE 5-16 Instability in the blend during blend changes as demonstrated by coal doping at Monroe Power Plant (Source: [9]).
196 Combustion Engineering Issues for Solid Fuel Systems
Silo and bunker performance determines the extent to which a boiler can handle blend changes without experiencing operational issues such as opacity spikes and excursions. Blend changes can be manipulated to minimize upsets and derates experienced by a boiler.
5.5 Conclusions Blending has become a prominent process for managing the supply of solid fuels at electricity-generating utilities, process industries, and institutional establishments. It requires extensive equipment in the fuel yard and in the fuel transport system. It also requires considerable attention to controls to manage the flow of fuel. When blending occurs, significant changes exist in the fuel being fed to the boiler. These changes exist in the quality of the fuel and the behavior of that fuel in a combustion setting. Operationally, these changes can influence management of the coal pile, the mills, the combustion system, and the post-combustion controls. Further, the ability to adjust blending can influence operational practices. In all, blending is increasingly practiced for both technical and economic reasons. It has increasing potential for all solid fuel users but must be accommodated in the design and operation of the combustion system—the utility boiler, the process industry boiler or kiln, or the institutional system.
5.6 References 1. Merriam Webster’s Collegiate Dictionary, 10th ed. 1994. Springfield, MA: Merriam Webster, Inc. 2. Tillman, D.A., and N.S. Harding. 2004. Fuels of Opportunity. London: Elsevier. 3. Smith, D.J. 2006. Blending of Opportunity Fuels with Coal Can Reduce Emission and Generating Costs. Coal Power. 2(2): 24–26. 4. Tillman, D.A. 2000. Biomass Cofiring: The Technology, The Experience, The Combustion Consequences. Biomass and Bioenergy. 19(6): 365–384. 5. Hus, P., and D. Tillman. 2000. Cofiring Multiple Opportunity Fuels with Coal at Bailly Generating Station. Biomass and Bioenergy. 19(6): 385–394. 6. Battista, J., E. Hughes, and D. Tillman. 2000. Biomass Cofiring at Seward Station. Biomass and Bioenergy. 19(6): 419–428. 7. Tillman, D.A. 2001. Final Report: EPRI-USDOE Cooperative Agreement: Cofiring Biomass with Coal. Clinton, NJ: Foster Wheeler. Contract No. DE-FC22-96PC96252. 8. Dobrzanski, A. 2005. Opportunity Fuels: A Plant Perspective. Proc. Electric Power Conference, Chicago, IL. March 5. 9. Peltier, R. and K. Wicker. 2003. PRB Coal Makes the Grade. Power. 147(8): 28–36.
Fuel Blending for Combustion Management
197
10. Tillman, D., A. Dobrzanski, D. Skiver, and J. Dosch. 2005. Using Real Time Coal Analysis to Improve Combustion at the Monroe Power Plant of DTE Energy. Proc. Coal Technology Association Conference. Clearwater, FL. March 20–23. 11. Tillman, D.A., et al. 2006. Fuel Blending with PRB Coals for Combustion Optimization. Proc. PRB Users Group. Atlanta, GA. April 5. 12. Tillman, D.A., and D. Duong. 2006. Fuel Blending with PRB Coals for Combustion Optimization: A Tutorial. Coal Technology Association Conference. Clearwater, FL. May 26. 13. McLenon, K. 2007. Fuel Quality: A Shift Supervisor’s Perspective. Proc. Electric Power Conference. Chicago, IL. May 1–3. 14. Anon, K. 2006. Industry Report 2005 Operating Performance Ranking. Electric Light and Power. (November-December). 15. Scavuzzo, J. 2004. The AccuTrack System. American Coal Council Annual Meeting. Kansas City, MO. July 27–29. 16. Baxter, L.L., et al. 1996. Nitrogen Release During Coal Combustion. Energy & Fuels. 10(1): 188–196. 17. Singer, J.G. (ed). 1991. Combustion: Fossil Power. Windsor, CT: Combustion Engineering, Inc. 18. Tillman, D.A. 1999. Biomass Cofiring: Field Test Results. Palo Alto, CA: EPRI. Report TR-113903.
CHAPTER
6
Fuel Preparation Donald Kawecki
Vice President – Engineering Foster Wheeler North America Power
The process of designing and operating a fuel preparation system for a power or steam generation station requires a thorough understanding of the interrelationship of the components and a clear vision of the technologies to be employed. All the components need to work as an integrated system; otherwise, operational problems will be experienced. The fuel preparation system supports the boiler by providing a metered, controlled fuel flow stream that meets the combustion system sizing and fuel quality requirements. This chapter addresses an overview of the current knowledge of fuel preparation systems for the following:
Stoker-fired boilers Biomass-fired or cofired systems Circulating fluidized-bed (CFB) boilers Pulverized coal (PC)-fired boilers
Each of these technologies has specific fuel preparation requirements to support the system needs. These requirements are driven by both the combustion technology and the specific fuel being used. The first step in the knowledge process is to know the details of the fuels being considered. Having a clear understanding of the fuels will ensure that the fuel preparation system design can grind and transport the product with a high degree of certainty and that fuels-related pitfalls can be eliminated. Once the fuels are chosen, it is important to understand the combustion system needs. Fuel preparation systems are designed to receive the 199
200 Combustion Engineering Issues for Solid Fuel Systems
expected range of fuel types and fuel quality and accurately, reliably, and safely size, dry, transport, and meter the product into the combustion chamber. Each fuel has specific preparation requirements to optimize the combustion process. It is important to know these requirements during the fuel preparation system design phase to ensure proper system operation. Fuel preparation systems are normally divided into two groups: Fuel transportation systems transport the fuel from one location to
another. These systems include belt conveyors, drag chains, screw conveyors, feeders, etc., to move and meter the flows. There is a significant body of knowledge readily available on fuel transportation systems. This chapter will cover only the fuel transportation systems that are used to meter the flow into the combustion system. Fuel preparation systems that grind, size, dry, or otherwise prepare the fuel for use in the combustion system. These systems include a wide range of sizing and grinding equipment. One of the most important items to understand when designing or analyzing a fuel preparation system is that the components are governed by volumetric design, mass flow design, or a combination of the two. Understanding the interrelationship of these concepts will produce a flexible operating system.
6.1 Know Your Fuel The most important steps in supplying and operating a fuel preparation system are to know your fuel and system requirements. This section provides an overview of the main solid fuels in use today and provides guidelines in system requirements as they relate to fuel preparation systems. Specific information on the fuels being considered is the most important factor that can be obtained prior to designing a fuel preparation system. The information contained herein is intended as a guideline for understanding the issues with the transportation and preparation of solid fuels. Before undertaking a system design, fuel-switching, or cofiring program, one should determine the specific fuel’s qualities through lab testing and consultation with the fuel supplier.
6.1.1 Fuel Types Review of the solid fuel types in use today finds that coal is by far the predominant preference. However, with the advent of higher coal prices, higher shipping costs, and the implementation of emissions regulations, there is an increased use of petroleum byproducts and biomass fuels.
Fuel Preparation
201
Each of these solid fuels has specific fuel preparation requirements and issues that need to be understood and incorporated into a system analysis. An overview of the fuel related issues is provided.
6.1.2 Fuel Issues The main fuel-related issues to be addressed when designing or troubleshooting a fuel preparation system are as follows: Sizing requirements of the fuel product arriving at the plant and the
fuel product sizing required for the combustion process. The difference between these two items sets the fuel preparation system design and performance requirements. Flowability of the product through the various fuel preparation stages. Understanding the ease or difficulty in moving the product is a key design criterion for a fuel preparation system. Failure to understand and address this issue can lead to system failures due to plugging, binding, component failures, maldistribution, increased erosion, and capacity constraints of components and can lead to reduced combustion performance and potential unit derates. Morphology of the product to understand how the product was formed and the condition it is being received at the plant to correctly size the equipment to prepare the product. The means of transporting, metering, and reducing the product size are directly related to the quality of the product. Grindability is an indication of the amount of energy needed to reduce the size of a product and is one of the key indicators used to size coal crushing machinery. The Hardgrove grindability index is based on comparing the tested coal to a standard so all coals can be rated against each other. The method is defined in American Society for Testing and Materials (ASTM) Standard D409. Tramp iron/refuse/excessive moisture in the received coal stream can wreak havoc on a fuel preparation system. The mining, production, and transportation of fuel product will, by its nature, include the production of tramp products such as railroad spikes, municipal metal waste, front-end loader bucket teeth, etc., and can add significant moisture due to transportation and dust suppression control. Safety must be a primary element of any system review, design, supply, or operation. All fuels have the potential for combustion given the right conditions. It is up to the system designer and operator to ensure that the systems and practices are in place to detect, prevent, and mitigate the potential. These three concepts are a critical part of any fuel preparation system design.
202 Combustion Engineering Issues for Solid Fuel Systems
6.1.3 Coal Coal was formed from vegetative matter that was built up over thousands of years (see Chapter 2 for a detailed discussion of coal characterization). Over time, these deposits were covered by sediment and aged in a location of heat and pressure to create the solid fuels we are using today. This simple explanation does not do justice to the large variations in coal composition and quality. The variations are driven by The type of vegetative materials that formed the coal deposit. The
plant cells are composed of different components such as the cell walls and the cell internals. The walls have a stronger structure than the matter inside the cells. Slower growing plants will have a greater proportion of the cell walls and thus a stronger based composition. When converted to coal, the stronger structures will produce fuels that are more difficult to size and combust and may require increased fuel preparation. The age and depth of the coal seam produce wide variations in the coal quality and the ability to size and transport it. There is a significant body of knowledge defining and categorizing coal quality. The intent is not to repeat this knowledge but to use it as a foundation to overlay the coal preparation effects. The key items that are critical to this chapter will be highlighted. ASTM has developed a coal classification system based on the fuelfixed carbon and volatile matter content. The classification system is defined in section D 388 of the ASTM standards. The coal ranking system contains four main categories that are further divided into a total of 13 separate categories. The headings in Table 6-1 note the main categories and also include peat. Peat is the youngest of the products and has not yet achieved coal status but is composed of the same constituents. Table 6-1 also notes typical fuel quality constituents important to the fuel preparation system design and operation. These values represent a relatively small sampling of the fuels available and are intended to demonstrate the difference between the coal classes. This analysis is not a comprehensive summary of available coal quality. The preceding information will be utilized throughout this chapter. General comments relating to this information are as follows: Table 6-1 represents typical values and ranges for the fuels noted.
There can be a significant variation beyond the values shown. A fuel preparation system design or analysis should be based on actual fuel analysis and fuel ranges to assure the assumptions and designs are capable of supporting the program goals. There is a significant variation in moisture levels and flowability within each category and from the “as mined” to “received at
Fuel Preparation
203
TABLE 6-1 Typical Solid Fuel Properties Item
Anthracite
Proximate Analysis (wt. %, as-received) Fixed Carbon 67–84 Volatile Matter 2–11 Moisture 2–5 Ash 10–20 HHV (Btu/lb) 14,000 Hardgrove Grindability Index (HGI) Range 25–83 Typical Values 40–50 Density lb/ft2 (vol) 45–53 Density lb/ft2 (load) 70 Flowability Good MM BtubAnalysis Weight (lb/MM Btu) Volume (ft3/MM Btu) Moisture (lb/MM Btu) Ash (lb/MM Btu)
71 1.3 2.14 5.00
Bituminous Coal 40–77 17–40 1–12 3–12 12,500 45–112 45–60 45–50 70 Good to Marginal 80 1.8 4.8 7.2
Subbituminous Coal
Lignite
Peat
32 27 37 4 7,300
13 35 50 2 5,000
45–62 50–60 35–45 70 Marginal to Poor
45–90 50–80 35–45 70 Poor
N/Aa 19–25 80 Poor
111 2.8 26.6 11
154 3.9 61.5 12.3
200 9.1 100 20
33–47 30–32 14–31 4–7 8,500
a
Not available Million Btu input Source: [1–4]
b
plant” coal analysis. Higher surface moisture levels usually translate into more issues with fuel pluggage. System designs need to take this into account. The transportation and dust control methods used can add substantial moisture to the “as mined” reported values. When utilizing a coal analysis, one must understand the provenance so corrections can be made for site-specific transportation and storage issues. The fuel density must be known to size and troubleshoot fuel preparation systems. The difficulty is that the fuel density will change depending on the sizing and surface moisture of the fuel. Fuel preparations systems contain both volumetric flow and mass flow components. Having an accurate fuel density is the bridge between these two different flow components. The densities contained in this section are noted for both a flow and load component. The flow density is normally used to calculate the volume of coal needed to meet the system requirements and storage system capacity. The load value is used to calculate the weight that the systems need to support once the volumes are set. The weight value includes margins for extremely wet or fine fuels. The information in Table 6-1 represents typical values.
204 Combustion Engineering Issues for Solid Fuel Systems The Btu input analysis section of the table is the key to designing
fuel preparation systems. This section defines the quantities per million Btu (as-received). This section is very useful for setting up a system plan or accounting for fuel variations in a fuel-switching or cofiring program. For example, if one is considering switching from bituminous coal to subbituminous Powder River Basin (PRB) coal, the following could be determined by dividing the subbituminous coal values by the bituminous coal values. As noted in Table 6-2, this brief analysis points to the significant changes experienced when switching between coal types.
6.1.4 Petroleum-Based Products Petroleum-based products, commonly known as petcoke (i.e., petroleum coke), include a wide range of constituents. Table 6-3 notes the main types and the key items relating to fuel preparation. TABLE 6-2 Solid Fuel Btu Input Analysis Summary Bituminous Coal
Subbituminous Coal
% Change
Potential Issues
80
111
þ 40%
1.8
2.8
þ 56%
4.8
26.6
þ 554%
7.2
11
þ 53%
Feeders capacity, mill capability Conveyor capacity, silo volume Primary air drying capability, air heater performance Increased boiler erosion, slagging, fouling
Item
Delayed
Shot
Fluid
Flexi
Proximate Analysis (wt.%, as-received) Fixed Carbon Volatile Matter Moisture Ash HHV (Btu/lb) Hardgrove Grindability Index (HGI) Density lb/ft2 (vol) (dry) Density lb/ft2 (load) Flowability
80.2 4.5 7.6 0.7 14,300 54 23–31 55 Marginal
89.6 3.1 6.3 1.0 14,360 39 23–31 55 Marginal
91.5 5.0 2.2 1.3 14,020 35 23–31 55 Marginal
94.9 1.3 2.9 1.0 13,970 55 23–31 55 Marginal
MM Btu Analysis Weight (lb/MM Btu) Volume (ft3/ MM Btu) Moisture (lb/MM Btu) Ash (lb/ MM Btu)
TABLE 6-3 Typical Properties of Petroleum Coke
Source: [4, 5]
Fuel Preparation
205
There is much less range in the petcoke constituents than in coal. This simplifies the designs somewhat. However, the wide range of moistures noted in Table 6-3 and the wider range of moistures that can be received at the plant level will create flowability issues that will need to be accounted for in the system design.
6.1.5 Biomass A wide range of biomass is being increasingly used as an alternative source of fuel in power and steam-generating plants. The nature of the biomass makes fuel preparation extremely problematic. Care must be taken to assure the reliable transportation and preparation of these fuels. Biomass is broken into two main categories as follows: Woody biomass is primarily residue from the wood processing
industry, forestry products, or urban wood waste. Table 6-4 notes some of the key properties when considering woody biomass. Herbaceous biomass is primarily agricultural and agribusiness wastes. Their fuel-handling and preparation properties and requirements are significantly different from woody biomass and must be treated differently. Table 6-5 lists some of the key properties when considering the use of herbaceous biomass. Use of either biomass type in a modern power plant presents significant challenges due to the nature of the fuels. It is rare when these fuels are used as the primary fuel; they are usually used to augment the fuel supply in percentages not exceeding 15% of the thermal heat input.
TABLE 6-4 Typical Properties of Woody Biomass Item
Pine Chips
Mixed Sawdust
Proximate Analysis (wt.%, as-received) Fixed Carbon 8.4 Volatile Matter 46.6 Moisture 45.0 Ash <1 HHV (Btu/lb) 4,676 Density lb/ft2 (vol) 16–25 Density lb/ft2 (load) 40 Flowability Poor Source: [6, 7]
11.4 48.0 40.0 <1 5,040 16–25 40 Poor
Urban Wood Waste
12.5 52.6 30.8 4.1 5,788 16–25 40 Poor
206 Combustion Engineering Issues for Solid Fuel Systems TABLE 6-5 Typical Properties of Herbaceous Biomass Fresh Switchgrass
Item
Proximate Analysis (wt.%, Fixed Carbon Volatile Matter Moisture Ash HHV (Btu/lb) Density lb/ft2 (vol) Density lb/ft2 (load) Flowability
as-received) 13.7 64.8 15.0 6.6 7,750 5–10 20 Poor
Weathered Switchgrass
Reed Canary Grass
Mulch Hay
12.6 69.5 15.0 2.9 8,150 5–10 20 Poor
6.9 26.5 65.2 1.4 7,103 5–10 20 Poor
13.8 62.5 19.5 4.3 8,058 5–10 20 Poor
Source: [8, 9]
6.2 Fuel Storage Silo The function of a fuel storage silo is to maintain an adequate fuel capacity to minimize fuel interruptions and station manpower requirements. These goals can only be met by a silo that has been sized and designed correctly. This section provides guidelines to use in designing a new system or troubleshooting an existing system.
6.2.1 Storage Capacity Fuel silo storage capacity is normally based on three items: 1. Fuel usage rate; 2. Plant capability to support the silo refill; and 3. Frequency of fuel supply delivery to site. Fuel usage rate is determined by the downstream process and is usually set up for the maximum expected value. It is important to note that combustion and steam generation processes are based on a heat-liberation rate and not a mass flow or volumetric flow rate. The fuel supply system supports the downstream process, so it must be sized to meet the process needs. This presents a design conflict because the silo is a volumetric storage device and the downstream process is a heat liberation-based requirement. The steps to develop the required flow rate are 1. Determine the required furnace heat input. 2. Convert the heat input to a mass flow. 3. Convert the mass flow to a volumetric flow.
Fuel Preparation
207
Furnace heat input values are obtained from the furnace manufacturer or the process engineer and are the design basis for setting the silo design requirements. The required values may be included in a summary performance sheet or on a thermal or mass balance summary page. Furnace heat liberation is normally expressed as MM Btu/h or millions of Btu per hour. Care must be taken to ensure that this is the heat into the furnace and not the heat absorbed by the boiler, the difference being the efficiency of the boiler, i.e.: Furnace Heat Input ¼ Heat Absorbed by the BoilerðBoiler DutyÞ= Boiler Efficiency
[6-1]
The fuel mass flow is obtained by the following equation: Fuel Mass Flow ¼ Furnace Heat Input=Fuel HHVðBtu=lbÞ
[6-2]
The volumetric fuel flow is as follows: 3
Fuel Volumetric Flow ¼ Fuel Mass Flowðlb=hÞ Fuel Densityðft =lbÞ [6-3] The preceding equations are used to determine the volumetric flow rate required from the silo. Plant capability to support the silo fill is based on the available staffing plan. It is desirable to design a system that is supported by a one-shift filling operation. This means that the fuel can be received and off-loaded into the silos during a normal day shift without overtime or second shift operation. Second and third shift operation is always available for backup fill time or system maintenance. For one-shift operation, the silo will need to store sufficient fuel to get through the two off-shifts and a portion of the day shift. The expected operation of the plant during the off-shifts will also need to be taken into account. If the plant will be shut down or run at lowered loads during the off hours, the silo sizing can be adjusted accordingly. Once the base silo capability is set, a sensitivity study can be performed for different fuel densities, boiler operation, silos in service, and offloading scenarios to minimize the risk of running out of fuel during normal and abnormal operation, as well as to set the maximum load requirements for the silo support structure. The importance of understanding the basic concept noted above comes into play when considering fuel switching and multiple use of fuels. The changes in fuel density and heating value associated with fuel switching can have a significant impact on the silo fill time analysis noted previously. Just ratioing the heating values and fuel densities for the fuel switch
208 Combustion Engineering Issues for Solid Fuel Systems
will give an indication of the relative fill time cycling and the manpower effects. In many cases, switching to a lower grade coal on a PC plant will require the use of an additional shift for silo filling. These calculations can be performed using the HHV and fuel densities noted in Section 6.1.
6.2.2 Silo/Bunker Design Considerations Once the required fuel silo active volume is set, the key design criteria for fuel silos are as follows: 1. Having a continuous, controlled flow to the furnace; 2. Minimizing the potential for developing hazardous conditions; and 3. Fitting the silo into the plant arrangement. Having a continuous, controlled flow to the furnace is the most critical requirement for designing a fuel silo. The fuel silo must be designed with an understanding of the fuel flowability characteristics and safety requirements; otherwise, there will be a greater risk for interrupted fuel supply and system shutdowns due to upset conditions. To have a continuous, controlled fuel flow leaving the silo, it is recommended that a mass flow design be used and a funnel flow design be avoided [10]. In funnel flow, an active flow channel forms in the center of the silo above the outlet with nonflowing material around the silo periphery. The result is a last-in/first-out flow sequence that produces stagnant areas where the fuel can accumulate, harden, combust, and/or segregate depending on the fuels used. The nature of the funnel flow can also produce surging out of the silo that is known as ratholing and has the potential for the fuel to bridge in the silo. Both of these nondesirable effects will reduce the system controllability. In mass flow, all the materials in the silo are in motion concurrently, and the silo empties on a first-in/ first-out basis. Mass flow minimizes the amount of time that the fuel is in the silo, so it also minimizes the potential for plugging, bridging, sintering, and storage silo fires. The areas to consider when designing a solid fuel silo for mass flow operation are as follows [11]: 1. Fuel flowability characteristics that are determined by the fuel type, fuel sizing, and moisture content. Fuel flowability is the greatest variable in the silo design process. 2. Pressure differential across the fuel silo outlet. A higher pressure differential will produce more of an upward force that will push against the fuel. The higher the differential, the higher the chance of the solid fuel bridging across the silo outlet section. 3. Coal silo diameter and hopper design. 4. Coal silo lining materials.
Fuel Preparation
209
Achieving mass flow in a silo requires the correct blend of the following items: 1. Understanding the silo storage and usage scenarios. It is important to understand how long the fuel can remain in a silo. Fuels that sit idle in a silo will compact over time, potentially leading to bridging and binding of the product, and can result in funnel flow, bridging, and ratholing. If the plant process requires longer idle times, special considerations should be made to assure flowability on the subsequent start-up. 2. An outlet section large enough to prevent material bridging. The outlet section has to be large enough to prevent arching and bridging while achieving the required fuel output. The required flow area is set by knowing the volumetric flow requirement leaving the silo and the design of the flow control device at the silo outlet. Use of a round outlet will require a larger width profile outlet than the use of a rectangular outlet section. Round outlets also require steeper slopes than rectangular silo outlet sections. The general flowability relationship is that a circular outlet diameter is equivalent to a rectangular outlet diagonal dimension. 3. Hopper walls steep and smooth enough to keep the material flowing. In a new design, the hopper angle and smoothness requirements are based on a flowability study of the fuel product. The information required to determine the required hopper slopes are the fuel type, product sizing, moisture levels, expected storage, and expected usage. These values set the flowability requirement for the silo hopper section. The variables adjusted to ensure flowability are the hopper slope and the hopper wall smoothness. General guidelines for hopper slopes as measured from the vertical by solid fuel type are noted in Table 6-6. Due to the complex geometries in most silo outlet sections, these are considered general guidelines. These angles should be confirmed and adjusted based on a product flowability study. In general, a steeper hopper angle will have more capability to offset the binding and plugging action of the fuel and will have a greater probability of producing a mass flow fuel output. TABLE 6-6 Minimum Silo Wall Angles Fuel Type Bituminous coal Subbituminous coal Petroleum coke
Round Silo Sections
Rectangular Silo Sections
>75 >80 >80
>70 >75 >75
210 Combustion Engineering Issues for Solid Fuel Systems
4. Use of a smoother silo outlet section liner to promote flowability. The silo outlets are subject to erosion, so the choice of the outlet material must look at both flowability and maintainability. Refractory lining is normally used for highly erosive products. Refractory linings are not recommended for solid fuel silos because of their poor flowability characteristics. The main silo outlet types are as follows: Carbon steel is a common silo outlet section material and is used as the base material for almost all silos. The material is formable, relatively inexpensive, and readily repairable. Silos can be made of carbon steel and lined with carbon steel for reparability. However, carbon steel has a relatively rough surface that will promote bridging. Stainless steel has better flowability characteristics than carbon steel but is much more expensive. Stainless steel can also be supplied with different surface finishes to match the fuel flowability characteristics. When stainless steel is used, it is usually installed as a liner in the silo outlet section and not as part of the silo structure. Stainless steel liners are the most common liners in use for solid fuels on new units. They provide better flowability characteristics than the carbon steel liners and have a robust wear capability. Advanced materials such as Teflon coating are also used in more extreme cases. These composite or chemical-based products provide the best flowability characteristics but are more expensive than the metal-based products and have a lower allowable temperature range than the other products. These advanced materials are seeing increased retrofit usage for coal-switching programs on existing units where the original design fuel had better flowability characteristics than the fuel being considered. Installation of this product in the existing silo can overcome the reduced flowability characteristics of the fuel being considered. In general, these products have a maximum short-term temperature range of up to 300 F. Above this temperature the materials will fail, resulting in loss of the flowability characteristics. Replacement of the failed sections is required to regain the flowability characteristics. If these products need to be used on highly reactive fuels that exhibit a high spontaneous combustion tendency, such as PRB coals and lignite, great care must be taken to ensure that the silos are equipped with adequate fire detection, suppression, and remediation equipment and that the fuels do not sit idle in the silos for any extended periods of time. Flanging the silo hopper area for ease of replacement should also be considered. 5. A silo outlet that is completely active with no ledges or dead zones. This is a very important item for both a new silo design and for repair and maintenance on existing silos. Ledges, plate
Fuel Preparation
211
overlaps, weld buildup, etc., all provide a location for the fuel to hang up and potentially bridge. A well-designed silo will minimize, if not eliminate, these areas. Items to consider are as follows: i. Silo liner plates installed with vertical weld lines instead of a patchwork of smaller plates field welded; ii. Flow plates installed over horizontal weld lines to promote flowability; iii. Hopper outlet shut-off dampers that are completely removed from the fuel stream when open. The damper frame should be designed so it does not project into the flow area; and iv. Use of air hammers, hammer plates, active bottoms, air purges, etc., installed in key areas of the silo outside to promote flowability in critical silo areas. 6. An installation design and quality control procedure that assures that the silo, hopper, and linings are installed per the design requirements. The best-designed system will not work if it is not installed and maintained correctly.
6.2.3 Safety Considerations Solid fuels are, to varying degrees, prone to self-combustion, and coal dust is considered a highly explosive product. Care must be taken to eliminate this potential during the loading, storage, and use of the fuel. Safety cannot be ignored. Following is a brief listing of detection, prevention and remediation techniques and equipment for consideration on the coal transport and storage system areas: Detection: CO detection, thermocouples, O2 monitors, methane detectors. Prevention: Proactive cleaning, evacuation fans and baghouse systems, silo inerting. Remediation: Sprinklers, CO2 system, foam and other sprays, water deluge system, fire-retardant injection lances. The details and recommendations should come from a qualified system supplier who has a high degree of familiarity with the current codes, standards, and applications and experience with the solid fuels under consideration.
6.3 Solid Fuel Flow Control Supplying a controlled, metered, reliable fuel flow is one of the most important requirements of any fuel preparation system. Fuel flow variations can easily upset the combustion process and cause a wide range of issues from emissions violations, furnace component slagging and fouling incidents, and tube corrosion to unit trips and outages.
212 Combustion Engineering Issues for Solid Fuel Systems
The coal flow control feeders are normally located just upstream of the furnace fuel injection point to gain the maximum advantage of the metering process. The further the flow metering is from the combustion process, the more variability can be expected in the fuel delivery. Design of an accurate, reliable fuel flow control system requires an understanding of the fuel properties and an understanding of the process requirements. Fuel issues include
Moisture levels; Required fuel flow; Product sizing (maximum and minimum); Product density and density variability; Fuel silo design considerations; and Furnace input control limitations.
Not accounting for these properties in the design and operation of the fuel feed system can contribute to a poorly operating furnace and can lead to long-term problems. It is also important for the fuel feeder to fit within the fuel preparation and delivery system. The feeder sizing and draw rates must match the fuel silo design conditions to minimize plugging potential and ensure mass flow conditions are maintained. The two main types of fuel flow control devices are volumetric and gravimetric. Volumetric feeders provide a controlled volume of fuel to the downstream components. Volumetric feeders cover the widest range of application from simple screw feeders to utility-grade volumetric control devices. Volumetric feeders are best used when there is little variation in fuel density and/or the downstream system does not rely on an extremely accurate heat input. Volumetric feeders are usually used on stoker-fired, biomass-fired, and industrial-sized CFB boilers. Volumetric feeders can be belt, screw, rotary valve, or vibrating type. Gravimetric feeders utilize a load cell system or a lever/balance system to determine the weight of the fuel passing through it. The use of a gravimetric feeder system will provide a more accurate feed rate calculation than the volumetric feeders because it will compensate for density variation in the fuel supply. These feeders are usually used when control of the heat input (MM Btu/h) to the boiler is required to optimize the combustion, boiler efficiency, and emissions. Almost, if not all, modernday pulverized coal-fired and CFB utility-generating stations utilize gravimetric feeder systems. Gravimetric feeders are weigh-belt type devices. Belt feeders are best used with elongated hopper outlets and when handling a wide range of fuel products. The exception is the metering of extremely fine products. There is a large body of knowledge available on the specific design methods and components of all feeder types, so the details will not be
Fuel Preparation
213
reiterated herein. The important items to consider when choosing, optimizing, and troubleshooting a feeder system are the fuel-related issues associated with operation on one fuel versus multiple fuels. It is important to understand the usage of the different types of feeders for the different fuel types and applications. Tables 6-7 through 6-9 present general guidelines for fuel feeder usage within the Boiler House (i.e., from the fuel silos to the boiler). If the applications being considered are outside the guidelines presented in these tables, then a higher degree of due diligence is required, and specific flowability and usage analysis should be performed by flow analysis specialists. Table 6-10 notes the current technology of solid fuel feeder types used for the indicated plant designs. The feeder types used have changed through the years, the plant complexity and size have increased, and a greater range of coals is being utilized. If the application being considered is not represented in this table, then a higher degree of due diligence is required to assure that the system requirements can be met with the necessary degree of accuracy and reliability. TABLE 6-7 Utility Plant Feeder Selection Based on Fuel Characteristics Variable Maximum practical particle size Particle degrades easily Material is dry powder Material sensitive to over-pressure HHV control of a large-density variation
Screw
Belt
Rotary Valve
Up to 2 inches Good Good Good Poor
2-4 inches Good Good Good Good
3/4 inch Good Excellent Good Good
Source: [12]
TABLE 6-8 Feeder Selection Based on Utility Plant Application Consideration Variable Ability to tolerate direct impact Hopper outlet configuration Gravmetric or volumetric operation Ability to seal against gas pressure Amount of return spillage Ability to control dust Ease of cleanout Tolerance to tramp metal Source: [12]
Screw
Belt
Rotary Valve
Good
Poor
Poor
Square, round, or rectangular Vol
Square, round, or rectangular Either
Square or round Vol
Poor
N/A
Good
None Good Poor Poor
Controllable Controllable Good Good
None Good Good Poor
214 Combustion Engineering Issues for Solid Fuel Systems TABLE 6-9 Feeder Design Limitations for Utility Plant Application Variable
Screw
Maximum temperature Maximum tons/h Turndown
750 F 15 10:1
Belt
Rotary Valve
300 F 50 10:1
750 F 15 10:1
Source: [12]
TABLE 6-10 Feeder Selection Based on Fuel Type Variable Stoker Herbaceous biomass Urban waste biomass Industrial CFB Utility CFB Industrial PC Utility PC
Screw
Belt
Rotary Valve
Yes Yes Yes Yes
Yes Yes Yes Yes Yes Yes Yes
Yes
Yes Yes
Source: [12]
6.4 Fuel Sizing Equipment Four basic crushing methods are used to size solid fuels. All fuel sizing equipment uses one or more of these methods to reduce the size of the fuel product traveling to the combustion chamber. The four methods and use guidelines are as follows [13]: 1. Impaction is the sharp, instantaneous impingement of one moving object against another. One or both of the objects can be moving. Within this category are two major subgroups of gravity impact or the dropping of a fuel particle or object to crush the fuel and dynamic impact, or impaction driven by an outside power source. Dynamic impact is usually specified for fuel preparation when the finished product must be well graded and must meet intermediate top and bottom size requirements. 2. Attrition is the sizing of fuels by rubbing between two hard surfaces. The material is reduced in size by forcing it through a given clearance. Attrition is a dynamic method of sizing. Attrition crushing is usually used for fuel preparation when the material is friable
Fuel Preparation
215
or mildly abrasive and when a closed circuit system is not desirable to control top size (i.e., classifier system to reject oversized for regrinding). 3. Shear is a cleaving or cutting action and is usually combined with other methods. In this case, the fuels are forced against a shear bar and sized accordingly. Shear is a dynamic method of sizing. Shear crushing is normally used when material is somewhat friable, for primary crushing with no more than a 6:1 reduction ratio, and when a relatively coarse product is required. 4. Compression is the grinding of the fuel between two hard surfaces. Compression crushing is normally used when a material is hard and abrasive and when a relatively coarse product is required. Fuel sizing equipment must be chosen taking into account the grinding characteristics of the fuel, the fuel size into the equipment, and the required fuel size output from the equipment. The first step in developing a fuel preparation plan or troubleshooting an existing system is to set the fuel size requirements. This includes understanding the size range and conditions being delivered to the site and the size and conditions that need to be output to the combustion process. Table 6-11 notes typical process requirements for the combustion systems. The fuel preparation system design needs to take the “as-received” fuel sample and size it to the process conditions. This information is used to TABLE 6-11 Combustion Process Fuel Sizing Requirements
Fuel Type
Anthracite Low volatile bituminous coal High vol bituminous coal Subbituminous coal Lignite Peat Herbaceous biomass Woody biomass
AsReceived Coal Size
PC
CFB
Stoker
Cyclone
<2" <2"
Process Distribution (<200/ <50 mesh) 80%/99% 75%/99%
Process Top Size 1/4" 3/8"
Process Top Size 5/16"–1 1/4" 3/4"–1 1/2"
Process Top Size 1/4" 1/4"
<2"
75%/99%
1/2"
3/4"–1 1/2"
1/4"
<2"
65%/98%
1/2"
3/4"–1 1/2"
1/4"
<2" <2" Varies
65%/98% N/A 1/4"
1" 1" 2"
3/4"–1 1/2" 1" 3/4"–1 1/2"
1/4" N/A <3/4"
Varies
1/4"
2"
3/4"–1 1/2"
1/2"–1 1/2"
216 Combustion Engineering Issues for Solid Fuel Systems
gain an understanding of the typical size requirements. Actual values through the supply chain should be used for system designs and troubleshooting. It should be noted that all fuels are not recommended for all the combustion systems presented. Table 6-11 presents the probable fuel size requirements when the other fuel constituents are applicable for the combustion system. Specific fuel details should be reviewed with the original equipment manufacturer (OEM), the combustion system supplier, or other industry experts to define the plant-specific applicability and sizing requirements. One key element of the fuel sizing that is not included in Table 6-11 is the size distribution requirements. Different combustion technologies require different fuel size distributions to support the combustion process. Understanding the fuel flow and size distribution requirements will define the fuel sizing equipment. The four combustion systems noted in Table 6-11 have different operating characteristics that produce different fuel size requirements. Following is a brief summary of the combustion system characteristics that define the fuel size requirements. PC boilers utilize a suspension-fired combustion system with a limited residence time. Finer fuel product is preferred with the fuel top size being the controlling factor. PC boilers need to complete the combustion process before the furnace outlet to minimize the potential for tube overheating and severe slagging and fouling. The larger coal particles will take longer to burn and are usually the main cause of upper furnace and convection pass issues. The fuel size requirements for PC firing are driven to minimize the larger particle size as measured by the percentage passing a 50-mesh screen. PC-fired units also require that the fuel be preheated and dried to augment the combustion process. The coal grinding equipment must meet the needs of the firing systems for the boiler to operate efficiently. The coal/air temperatures entering the burners are generally in the range of 135 F to 200 F depending on the fuel type and moisture level. The temperature is chosen based on balancing the improved combustion against the safety of the system. As the coal volatile matter increases, the potential for coal fires in the fuel preparation system increases. All operating procedure and system control requirements should meet the equipment manufacturer recommendations and safety code requirements. CFB boilers utilize a mixture of fuel bed and suspension firing. The CFB has a significant fuel residence time in the combustion window, and oversized product is rejected back to the furnace for further combustion. The fuel particles are sized to support this process, so the quantity and quality of both the top size and the fines are important. If the top size is too large, the product will not circulate, and if there is a high percentage of fuel fines, the fuel will bypass the recirculation process and leave through the convection pass before the combustion is complete. The fuel sizing requirements for CFB firing are to provide a minimum of oversized
Fuel Preparation
217
and fine particles to maximize the fuel recirculation. CFB boilers do not require the fuel to be dried and heated to support the combustion process. Stoker-fired boilers have more flexibility for larger sized fuel particles, but the fuel preparation systems should minimize the finer particles. Finer particles can block the grate flow areas and can entrain in the furnace. Stoker-fired boilers do not require the fuel to be dried and heated to support the combustion process. Cyclone-fired boilers can handle only a limited amount of fine particles but can handle a wide range of fuels. The fuel needs to be sized so there is limited oversized product. Cyclone-fired boilers do not require the fuel to be dried and heated to support the combustion process. However, the fuel must have enough ash to develop and maintain a liquid slag layer in the cyclone. Table 6-12 contains a summary of the fuel preparation technologies commonly used to prepare fuel for combustion. Table 6-12 notes the technology and the principal usage guidelines. Following are brief descriptions of the crushers noted in Table 6-12. TABLE 6-12 Fuel Sizing Equipment Usage Summary Technology
Grinding Type
Use
Output
Bradford Breakers
Gravity
Primary crushing
Relatively coarse, Minimum fines, 100% sized
Cage Mills
Impact
Fine to coarse
Granulators
Impact, Compression
Primary crushing, Final sizing Primary crushing, Final sizing
Hammermills
Impact, Attrition, Shear
Primary crushing, Final sizing
Impactors
Impact
Final sizing
Double Roll Crushers
Compression
Final Sizing
Ball Mills
Gravity Impact
Final Sizing
Vertical Spindle Mills
Compression
Final Sizing
Medium to coarse, Minimum fines Controlled top size, Medium fineness, High fines Controlled top size, Minimum fines Controlled top size, Minimum fines Controlled top size, Maximum fines Controlled top size, Maximum fines
System Mine or plant primary sizing and tramp material cleaning CFB, Stoker, Cyclone CFB, Stoker, Cyclone
CFB, Stoker, Cyclone
CFB, Stoker
CFB, Stoker
PC Fired, Arch Fired PC Fired, Arch Fired
218 Combustion Engineering Issues for Solid Fuel Systems
Bradford Breakers are normally used for crushing, sizing, and cleaning run of mine coals at the mine or plant site. They produce a relatively coarse product with minimum fines that are 100% sized. The Bradford Breaker, which is shown in Figure 6-1, is a large cylinder made of a perforated plate fitted with internal shelves. As the cylinder rotates, the shelves lift the feed, and the feed slides off the shelves and drops onto the screen plates below, where it shatters along its natural cleavage lines. The size of the screen plate perforations determines the product size. Sized product flows through the screen, but oversized pieces are relifted and redropped until they pass through. Tramp iron, lumber, or other debris is transported to the discharge end of the cylinder where it is scooped out for disposal by a refuse plow [13]. Cage mills are used for sizing of friable, dry materials. A cage mill, shown in Figure 6-2, is composed of a series of concentric sleeves that are driven. The feed enters the center of the mill and is ground, through impaction, as it is struck by the first row of sleeves. After impact, the shattered coal is passed to the second row of sleeves, where the sizing process continues. Size reduction continues through the subsequent rows until the fuel is thrown against an impact plate and then exits the mill. Varying the spacing between the sleeves controls product size by controlling the speed of the rotating cage [13]. Granulators crush by a combination of impact and rolling compression, utilizing rows of ring hammers that crush with a slow, positive rolling action; this action produces a granular product with minimum fines.
FIGURE 6-1 Pennsylvania Crusher Bradford Breaker. (Courtesy of Pennsylvania Crusher Corp., 2006)
Fuel Preparation
219
CAGE MILL
Cage Mill View
FIGURE 6-2 Pennsylvania Crusher cage mill (Source: [14]).
Product size is determined by screen openings and is adjusted by changing the clearance between the cage and the path of the ring hammers [13]. A granulator is shown in Figure 6-3. Hammermills reduce the material sizing in a two-step process. The material is first reduced by dynamic impact from the rotating hammers and continues with attrition and shear in the small clearance between the hammers and screen bars. Hammermill crushing produces great product uniformity with a minimum of oversized product and will produce high throughput capacity. Modern-day systems use a reversible hammer system to increase the time between maintenance shutdowns and maximize the hammer material use [13]. An example of a hammermill is shown in Figure 6-4. Impactors are used to produce an optimum percentage of products below 1/8" size and with a minimum amount of fine particles. The modern
FIGURE 6-3 Pennsylvania Crusher granulator. (Courtesy of Pennsylvania Crusher Corp., 2006)
220 Combustion Engineering Issues for Solid Fuel Systems
FIGURE 6-4 Pennsylvania Crusher Hammermill. (Courtesy of Pennsylvania Crusher Corp., 2006)
impactors can maintain their throughput with wet coals and are able to pass through noncrushables. Their rotors can usually be run in either direction to provide equal wear on both hammer faces [13]. An impactor is shown in Figure 6-5. Double roll crushers compression grind the fuel between the two counter-rotating rolls. As the two rolls rotate toward each other, the material is pulled down into the crushing zone, where it is grabbed and compressed by the rolls. Product size is determined by the size of the gap between the rolls [13]. An example of a double roll crusher is shown in Figure 6-6. Ball mill pulverizers use an air sweeping action for impaction and gravity grinding of the fuel in a large rotating drum filled with hardened steel balls. The main components are a large drum lined with a wear-resistant material and with hollow trunnions at each end and a classifier that sizes the final product. The drum is mounted horizontally and supported by trunnion bearings. As the ball mill rotates, the profiled liners lift and
FIGURE 6-5 Pennsylvania Crusher impactor and Coalpactor mills. (Courtesy of Pennsylvania Crusher Corp., 2006)
Fuel Preparation
221
FIGURE 6-6 Pennsylvania Crusher double roll crushing mill. (Courtesy of Pennsylvania Crusher Corp., 2006)
tumble the fuel and steel balls at a low speed, pulverizing the fuel. The ground product is sized by an integral or externally mounted classifier that passes the correct size fuel and rejects the oversized fuel back into the mill for further grinding. An example of a ball mill is shown in Figure 6-7. FOSTER WHEELER POWER GROUP
BALL MILL COAL PULVERIZER COAL CONDUIT PRIMARY AIR INLET DUCT
DISTRIBUTOR RAW COAL INLET CLASSIFIER PULVERIZED COAL DISCHARGE CONDUIT
SOUND PROOF HOUSING REINFORCED RIBBON CONVEYOR
CONVEYOR SPOKES TRUNNION TUBE MILL MOTOR RING BULL GEAR PINION PINION BEARING
AIR INLET SCREEN CONVEYOR ROLLER BEARING / AIR SEAL ASSEMBLY AIR INLET ELBOW
END LINER DOUBLE WAVE, DOUBLE SIZE ACCESS DOOR BALL DISCHARGE RECOVERY TRAY DRUM DOUBLE WAVE DRUM LINER TRUNNION BEARING
FIGURE 6-7 Foster Wheeler ball mill pulverizer. (Courtesy of Foster Wheeler North American Power, 2006)
222 Combustion Engineering Issues for Solid Fuel Systems
Hot air is used to transport and dry the fuel during the grinding process. The inlet air temperature is varied to maintain a set air/coal temperature leaving the mill [14]. Ball mill pulverizers produce a high fineness at the cost of high power usage. They have historically been used for anthracite and petcoke grinding. The lower moisture levels and hard-to-burn combustion characteristics of both of these fuels lend themselves to use in the ball mill. Vertical spindle mills use an air sweeping action for attrition and compression grinding of the fuel between a rotating table and large grinding rollers. The rollers use an external loading mechanism to increase the grinding capability on the system. The mill operates by maintaining a bed of fuel between the grinding tables and the grinding rollers. The rollers impart pressure on the fuel bed, causing the coal particles to rub against each other and break apart. Hot air is used to transport and dry the fuel. The inlet air temperature is varied to maintain the set air/coal temperature leaving the mill. Figure 6-8 notes the key components of a Foster Wheeler MBF Pulverizer and shows the coal flow path through the system. FW MBF COAL MILL Pulverized Fuel Adjustable Classifier
Grinding Roller Ass’y.
Replaceable Grinding Segments
Raw Coal
Coarse Particle Return
Rotating Primary Air Port
Tramp Iron Outlet Chute
Speed Reducer
Primary Air
FIGURE 6-8 Foster Wheeler MBF coal mill. (Courtesy of Foster Wheeler North American Power, 2006)
Fuel Preparation
223
Vertical spindle pulverizers are the most common grinding mill on pulverized coal-fired utility steam generators. They provide a reasonable fuel fineness capability with a fast fuel change response, low horsepower utilization, and a relatively long time between worn part replacements. The coal mills have undergone continual improvements by all the manufacturers as new materials are being used and more stringent combustion and emissions requirements have come into play. Use of computational tools and extensive field experience on the thousands of mills in operation have produced a range of retrofit options to improve existing mill performance. These options include the following: 1. Advanced static and dynamic classifiers that can improve the fuel fineness and distribution leaving the mill while maintaining or improving the mill throughput capability. The advanced static classifier optimizes the separation efficiency by optimizing the coal flow direction and velocity through the classifier system. The dynamic classifier increases both the fineness levels and coal mill throughput capability by use of a rotating cage installed inside a ring of stationary classifier blades. This rotating cage will knock down the oversized fuel particles and send them to the mill for regrinding while preferentially passing the finer coal particles. The classifiers are controlled with a variable speed drive so the overall mill performance can be optimized for the coal quality and combustion system requirements. An example of a dynamic classifier is shown in Figure 6-9.
FW Dynamic Classifier
Raw Coal Feed Pipe
Drive Belt
Bearing Assembly Variable Speed Drive Motor Stationary Vane Ring
Rotating Cage (Rotor) Rejects Cone
Classifier Housing
Primary Separating Zone Rejects Hopper With Flaps
FIGURE 6-9 Foster Wheeler Dynamic Classifier. (Courtesy of Foster Wheeler North American Power, 2006)
224 Combustion Engineering Issues for Solid Fuel Systems
FW MBF Rotating Airport Wear Ring
Upper Ring
Lower Ring Bolted Connection to Table
FIGURE 6-10 Foster Wheeler rotating airports. (Courtesy of Foster Wheeler North American Power, 2006)
2. Rotating airports (shown in Figure 6-10) are attached to the grinding table. These airports will distribute the primary air flow more evenly throughout the coal mill body. This increases the drying capability of the system and minimizes the erosion of the mill internals. 3. Hydraulic tensioning systems adjust the grinding force based on the coal flow and fuel quality. This will reduce the system power consumption and wear rates over the life of the unit. 4. Optimized roller and grinding table profiles grind with higher efficiency. 5. Advanced materials increase the time between required maintenance outages. All the preceding potential modifications should be considered for new system installation, fuel-switching and blending programs, and for troubleshooting remedies for overstressed systems. A wide variety of equipment and designs is available on the market for both original equipment manufacturers and independent engineering companies. These products will produce a wide range of performance improvements and enhancements for both new and retrofit programs. However, caution must be used when determining if the potential modifications will enhance the fuel preparation and combustion system performance. An analysis of the overall effects from the fuel-handling system through the emissions and boiler performance must be performed to assure that the potential solution fits within the plant capabilities.
Fuel Preparation
225
6.5 Pulverized Coal System Analysis Guidelines The majority of utility-sized solid fuel-fired power stations use air-swept pulverizers to supply fuel to the combustion systems and the majority of the systems are vertical spindle coal mills. To safely, reliably, and efficiently operate the power station, the coal mill must be fully integrated into the boiler island and must meet the combustion system requirements. The following sections provide an overview of how to integrate the coal mill into the boiler system and provide guidelines for troubleshooting the systems. The following key areas need to be understood when designing, optimizing, or troubleshooting a coal mill system:
Coal mill throughput capability; Coal mill fineness capability; Coal mill thermal capability; Coal pipe velocities; Coal pipe distribution; Firing system safety issues; Primary air flow and temperature requirements; Air heater leakage changes; and Primary air fan capability.
A coal mill performs the following tasks: 1. Grinding the raw coal to the required fineness levels; 2. Drying the surface moisture from the ground coal so it flows freely and burns easily; 3. Separating the ground coal traveling to the burners into the correct size fractions to support combustion; and 4. Transporting the coal through the pulverizer system to the burners. All four tasks are performed concurrently and must be correctly balanced to assure the desired coal mill performance is maintained. They are performed by the use of these three systems: 1. Primary air system to supply the required transport airflow at the temperature needed to dry the coal; 2. Grinding rollers, grinding ring, and tensioning system that supply the energy to grind the coal; and 3. Classifier system to separate the ground coal into a stream that meets the fineness requirements (sent to the burners) and a stream that needs further grinding (sent back to the grinding section).
226 Combustion Engineering Issues for Solid Fuel Systems
6.5.1 Mill Sizing and Standard Ratings Boiler manufacturers utilize a standard series of coal mills with overlapping throughput capability that are adapted to the particular job requirements. The number of mills in the series has been minimized and standardized to keep the supply costs down. Making a new mill size requires thousands of engineering hours and large capital costs for forging and casting patterns. By having standard-size mills, these development costs are paid only once per mill size and not on every job. During the design phase of a project, the manufacturer will review the customer requirements to determine the mill size that meets all the requirements. Items such as range of fuel analysis, spare mill requirements, combustion system fineness needs, etc., will all be evaluated and balanced to determine the final mill sizing and configuration. When fuel switching is being considered for existing units, the existing capability of the coal mills system is one of the key analysis areas. The current market trend is to switch to lower grade fuels that cut into the mill system margins. In many fuel-switching scenarios, the coal mill system is the limiting factor in determining the percentage of lower grade fuels that may be used. This points to the importance of understanding the coal mill system performance drivers and how they integrate into the overall plant system. The criteria to choose a mill size are as follows:
Maximum coal throughput—lb/h; Raw coal surface moisture—%; Raw coal sizing to the coal mill—inches; Raw coal Hardgrove grindability index—HGI; Required coal fineness levels to support the combustion goals— % passing 200 and 50 mesh; and Required coal/air mixture flows and temperatures. Each of these criteria has an effect on the coal mill sizing and performance. Coal mill manufacturers have developed a series of correction curves that define how the mill will perform for each of these criteria. Figures 6-11 through 6-13 represent typical correction curves for vertical spindle mills. These curves address the fuel quality effects on mill capability and will be used to present the interrelationship of the various fuel and performance criteria. Each manufacturer and each mill type has different designs and operating constraints that slightly modify the curves. These curves are presented to demonstrate the concepts. Actual system analysis should be undertaken with the manufacturer’s support to assure accurate results. Review of the graphs provides the understanding shown in Table 6-13 of the basic coal quality relationships for the vertical spindle pulverizers. Understanding these concepts will aid in designing, reviewing, and troubleshooting coal mill systems. A fuel switch calculation procedure is
Fuel Preparation
227
Typical Fineness Correction Curve
Fineness Correction Factor
1.40 1.30 1.20 1.10 1.00 0.90 0.80 50
60
70
80
Pulverizer Fineness - % passing 200 mesh
FIGURE 6-11 Coal fineness correction curve. Typical Hardgove Correction Curve
Hardgrove Correction Factor
1.20 1.10 1.00 0.90 0.80 40
45
50
55
60
Coal Hardgrove Value- Hg
FIGURE 6-12 Coal Hardgrove grindability index correction curve.
presented later to demonstrate the method of analyzing switching from one fuel to another. Once the coal mill is sized to support the required system performance, the operating characteristics are determined. The first system requirement is to develop the relationship of air flow versus coal flow. This is developed via an air/coal ratio analysis. Figure 6-14 represents a typical air/coal curve for a vertical spindle pulverizer. As seen in Figure 6-14, the coal and air follow a linear relationship down to a minimum value. The minimum value is set to assure sufficient airflow to dry the coal and transport it through the mill and pipes to the furnace. If the airflow is too low, the coal will drop out of suspension and cause coal mill and coal pipe plugging.
228 Combustion Engineering Issues for Solid Fuel Systems Typical Coal Moisture Correction Curve Moisture Correction Factor
1.10 1.05 1.00 0.95 0.90 0.85 0.80 5
10
15
20
25
30
35
Coal Moisture - As Received
FIGURE 6-13 Coal total “as-received” moisture correction curve.
TABLE 6-13 Coal Mill Coal Quality Effects Item
If
Coal Moisture Raw Coal Size Hardgrove Grindability Index Coal Fineness—% through mesh
Increased Increased Decreased Increased
Will Result In Decreased Decreased Decreased Decreased
mill mill mill mill
capability capability capability capability
Typical Air/Coal Curve
Primary Air Flow - Klb/Hr
250 225 200 175 150 125 100 0
25
50 75 100 Pulverizer Coal Flow To Mill - Klb/Hr
FIGURE 6-14 Typical air/coal ratio curve.
125
150
Fuel Preparation
229
Once the air/coal ratio relationship is developed, the system pressure drop is determined. The coal mill pressure drop is one of the largest and most variable components of the primary air fan design requirements. The primary air system has to be sized to supply the correct airflow at the needed pressure to dry and transport the fuel from the mill to the furnace. Limitations in the primary air system will directly impact the capacity of the coal mill system. Table 6-14 summarizes the relationship of coal mill variable criteria to mill pressure drop (DP) over the normal mill operating range.
6.5.2 Coal Mill Capacity and Capability Analysis To demonstrate the relationships presented in the preceding sections, a theoretical performance analysis on a pulverizer system utilizing an eastern bituminous coal to set the mill design criteria has been prepared. The analysis determined the eastern bituminous (Ebit) coal flow requirements to meet a given boiler’s steam flow demands and dictated the coal mill capability based on the mill size, setup, and coal quality. An analysis was then performed on the system to demonstrate the effect of a PRB coal conversion on the coal mill system performance. Table 6-15 notes the design basis for the coal mill study based on a typical coal mill performance that has been sized to meet the system requirements. A second column notes the coal mill requirements for a potential PRB fuel change from the original design. This section will show how the original design components are interrelated and will show the analysis method and potential results from switching to a different fuel. The review covered the three critical coal mill performance areas as follows: Coal throughput capability; Primary air capability; and Thermal capability.
For a coal mill to operate at a required coal flow, all three of the performance areas noted here must be operating in acceptable ranges. If one of these three areas is short, it will limit the coal throughput. TABLE 6-14 Coal Mill Pressure Drop (DP) Effects Item Coal Flow Primary Air Flow Classifier Opening Hardgrove Grindability Index
If Increased Increased Decreased Decreased
Will Result in Increased Increased Increased Increased
Mill Mill Mill Mill
DP DP DP DP
230 Combustion Engineering Issues for Solid Fuel Systems TABLE 6-15 PRB Conversion Fuel Supply System Analysis Summary Item
Units
Coal type HHV Total coal flow @ boiler design point Coal mills—total Coal mills operating—design Coal mill capability Per mill coal flow Per mill primary air flow Air to coal ratio Total primary air flow Coal analysis Density Moisture—total Moisture—surface Grindability Raw coal sizing Required fineness 50 Mesh 200 Mesh Mill exit temperature
Btu/lb lb/h Qty Qty lb/h lb/h lb/h lb/h
Design Coal
Conversion Coal
Eastern Bituminous 12,000 300,000
Powder River Basin 8,000 450,000
4 3 120,000 100,000 170,000 1.70 510,000
4 4 113,000 112,500 180,000 1.60 720,000
45 6
55 30
ft3/lb wt.% asreceived wt.% asreceived HGI inches
3
14
45 3/4
50 3/4
99þ% 75% 160
99þ% 65% 135
F
6.5.2.1 Coal Throughput Capability Coal mills have a mass flow throughput capability that cannot be exceeded. This capability is set by the amount of coal inventory required in the mill. If the inventory exceeds the maximum amount, the mill will plug. The inventory is set by the degree of grinding and drying required to meet the coal throughput requirements. For this analysis, it is assumed that the existing unit coal mill system was sized to achieve full load coal flow with three mills each operating at 100,000 lb/h coal flow, with the fourth mill as a spare. It was also assumed that there was an additional 20% capability in the existing mills that would allow normal system operation at 120,000 lb/h coal flow from each mill. The first step in the analysis is to determine what the revised mill coal flow capability is for the new fuel. This value is approximated by prorating the known capacity by the capacity correction factors, which are summarized in Table 6-16. The factor summary is developed by the following: Xtotal ¼ Xf Xhgi Xm
[6-4]
Fuel Preparation
231
TABLE 6-16 PRB Conversion Coal Mill Capability Analysis Summary
Item
Factor
Fineness Hardgrove Total Moisture Factor Summary
Xf Xhgi Xm X total
Ebit Correction Factor
Ebit Value 70% 45 6%
1.00 0.90 1.10 0.99
PRB Value
PRB Correction factor
65% 50 30%
1.10 1.00 0.85 0.94
For the Ebit coal; the factor Summary Xtotal ¼ 1:00 0:90 1:10 ¼ 0:99 The revised mill capability is determined by the following: Revised mill capability ¼ design capability ðrevised factor summary= design factor summaryÞ
[6-5]
Thus, the revised coal mill capability is Capability ¼ 120; 000 lb=h ð0:94=0:99Þ ¼ 113; 000 lb=h
[6-6]
This analysis has shown there is approximately a 6% coal mill throughput capability derating (113,000/120,000) lb/h due to the differences in the eastern bituminous coal and PRB coal constituents. The second step in the throughput analysis is to determine whether the coal mills with the revised maximum capability are capable of supplying the required coal flows. This is accomplished by comparing the required coal flow to the revised mill capability times the number of coal mills available. It is clear from the review that the coal mill system will not be able to supply the required PRB coal flow with three-mill operation. Therefore, the system will require four-mill operation at the full boiler load. Four-mill operation results in the following: Revised coal mill system capability ¼ 113; 000 lb=h per mill 4 mills ¼ 452; 000 lb=h
½6-7
Where the required coal flow ¼ 450,000 lb/h per Table 6-15. System Capability ¼ 452; 000 lb=h capability=450; 000 lb=h required capacity ¼ 1:0
½6-8
232 Combustion Engineering Issues for Solid Fuel Systems
Therefore, the coal mills have just enough grinding capacity to support full boiler load operation with all four mills in service. Additional factors to be taken into account to determine if this theoretical capacity is acceptable for day-to-day operation are as follows: The requirement for unit derates during coal mill maintenance and
coal mill maintenance rebuild outages; The effect of coal variability on the coal mill capability; and The ability of the coal offloading, transport, and storage silo system
to support the increased coal flow requirements. 6.5.2.2 Primary Air Capability The second area of analysis is to determine whether the existing primary air system has enough capacity to support the PRB coal conversion’s increased coal flow requirements. This is accomplished by determining the primary air flow required by the coal mills and adding in the primary air leakage through the air heater to develop the primary air fan requirements. The primary air flow required by the coal mills is easily developed with the coal mill air/coal relationship mill curve presented previously. The total primary air flow is found using the following equation: Total primary air flow ¼ no mills in service primary air flow per mill from the typical air=coal curve in Figure 6-14 [6-9] In the example case, this calculates to Total revised primary air flow ¼ 4 mills 180; 000 lb primary air per mill ðfound from the air=coal curve at 112; 500 lb=h coal flowÞ ¼ 720; 000 lb=h
½6-10
This value compares to the Ebit design value of 510,000. This means the coal mills grinding the PRB coal will require 41% more primary air than the current Ebit operation. 6.5.2.3 Air Heater Leakage Air heater leakage can be a significant contributor to the primary air fan flow requirements depending on the air heater configuration. There are two typical systems for supplying heated air to a utility coal mill: Tubular air heaters that fully separate the air and flue gas streams. In
this system, one of the gases flows inside a tube or between plates
Fuel Preparation
233
while the other flows outside the tube or over the plates. The heat transfer is across the metal boundary with no direct contact between the gases, and the air leakage is minimal. Tubular air heaters are usually used on industrial and smaller and older utility stations. Regenerative air heaters rotate a basket of packed metal through the hot flue gas, to heat up the metal baskets, into the cold primary air stream that absorbs the heat from the metal. Metal plates are used to seal between the low pressure flue gas and the high pressure primary air. As these seal plates wear between maintenance outages, the amount of leakage will increase. The vast majority of the power stations use regenerative air heaters. There are two types of regenerative air heater systems applied to the utility industry. The first has individual air heaters for the primary air stream and for the secondary air stream. The second has one air heater with separate compartments for both the primary and secondary air streams. This type is referred to as a trisector because of the three streams (primary air, secondary air, and flue gas). For this analysis, it is assumed that a trisector regenerative air heater was supplied. Air heater leakage is defined as the amount of air that “leaks from the high pressure air side of the rotary air heater to the low pressure flue gas side.” Air heater leakage is calculated by dividing the amount of primary air that leaks into the flue gas stream by the amount of flue gas entering the air heater. For a trisector air heater, it is difficult to calculate the individual components of the leakage so it is normally reported on a total air heater basis. This makes it difficult to calculate the primary air fan mass flow change for use in a fan analysis. Primary air fans generally operate in the 45–60 inches water column (w.c.) range, while the flue gas on a balance draft system will be at negative pressure. This can create a 50–80 inches w.c. differential pressure between the primary air side and the flue gas side. This high differential pressure combined with the large seal area between the primary air and the flue gas can produce large flow changes for relatively small seal plate wear. Primary air heater leakage rates of 25–50% of the primary air flow required by the coal mills are relatively common. When performing a fuel switch that significantly increases the primary air flow to the coal mills, the primary air heater leakage will trend to and often exceed the higher side of this range. If a fuel switch is being considered, the air heater supplier or the boiler manufacturer should be consulted to sort through the complexities of this issue to ensure that the analysis produces accurate results. It is important to note that the amount of air heater leakage exhibits a significant variation over time. Following a maintenance outage, the total air heater leakage can be measured in the 5–15% range. Over time as the air heater seals wear, this leakage value can double. It is extremely important to understand the primary air leakage changes over the maintenance life
234 Combustion Engineering Issues for Solid Fuel Systems TABLE 6-17 PRB Conversion Primary Air Flow Analysis Summary Item
Factor
Primary Air Flow to Mills Primary Air Heater Leakage Primary Air Heater Leakage Flow Total Primary Air Fans Flow
lb/h % lb/h lb/h
Ebit
PRB
510,000 25% 127,500 637,500
720,000 35% 252,000 972,000
of the equipment to ensure that the primary air supply system is capable of supporting the required coal flow. Testing the air heater leakage before and after a maintenance outage is the best way to develop this understanding. To simplify this analysis, it is assumed that the primary air heater leakage rate will increase linearly with the coal mill air flow change and the initial Ebit leakage rate was 25%, based on these assumptions Table 6-17 was developed. This result indicates the primary air system will need to produce a 50þ% higher flow to support full boiler load PRB coal operation. The higher flow rate will also produce a higher system pressure drop that will further reduce the fan capability. One should note that most primary air fan systems are supplied with significant margins to allow for the high air heater leakage rates. However, extensive experience has found that primary air systems can rarely supply the required airflows for a 100% Ebit coal to PRB coal conversion when they were originally sized for only Ebit coals. To support this PRB coal conversion, the primary air fan system will most likely need significant modification, additions, or even replacements with higher capability equipment. 6.5.2.4 Thermal Requirements To pulverize and transport the coal from the mill to the burners and to support the combustion process, the mill needs to dry the coal moisture to acceptable levels. This is accomplished by adjusting the coal mill inlet hot air temperature and flow to maintain the required mill outlet coal/air temperature. Coal drying can be improved by increasing the hot air flow and/or increasing the hot air temperature. The required air inlet temperature can be developed by performing a thermal balance around the coal mill. This balance will look at the following:
Primary air flow and temperature into the mill; Seal air flow and temperature into the mill; Coal flow, temperature, and moisture levels into the mill; Coal/air temperature and humidity leaving the mill; Remaining coal moisture in coal leaving the mill; and Mill losses and additions as required for the specific mill under analysis.
Fuel Preparation
235
One of the largest contributors for determining the inlet air temperature is the amount of moisture dried in the coal. The amount of moisture dried is a function of
Total moisture level in the coal; Coal rank; Surface moisture level in the coal; Coal fineness level leaving the mill; and Air/coal mixture temperature leaving the coal mill.
It is important to note that the drying characteristics of coals vary according to their rank and classification. Eastern bituminous coals tend to have low total moisture levels (5–10%) and very low inherent moisture levels (1–2%), whereas Powder River Basin coals have very high total moisture levels (up to 35%) and inherent moisture levels (up to approximately 18%). The PRB coals will require significantly higher inlet air temperatures than the eastern bituminous coals due to their significantly higher moisture levels. Figures 6-15 and 6-16 were developed based on typical drying requirements for the eastern bituminous and PRB coals. These graphs note the required coal mill inlet air temperature for a range of total coal moisture levels and air/coal ratio values.
Eastern Bituminous Coal 800 20%
Inlet Air Temp (⬚F)
700 15%
600
500
10%
400 5% 300 0% 200 1.00
1.25
1.50
1.75
2.00
2.25
Air to Coal Ratio
FIGURE 6-15 Eastern bituminous coal mill inlet temperature requirements.
2.50
236 Combustion Engineering Issues for Solid Fuel Systems Powder River Basin Coal 900
Inlet Air Temp (⬚F)
800
700
35% 30%
600
25% 500
20% 15%
400
300 1.00
1.25
1.50
1.75 Air to Coal Ratio
2.00
2.25
2.50
FIGURE 6-16 Powder River Basin coal mill inlet temperature requirements.
The Eastern bituminous coal development is based on the following general assumptions: 1. 2. 3. 4. 5.
80 F ambient air temperature and coal inlet temperature; Coal dried to 2% remaining moisture; 10% coal mill radiation loss; 150 F coal/air exit temperature; and Moisture lines based on as-received total coal moisture.
The actual requirements for developing a site-specific curve will vary slightly from the preceding assumptions. This curve is provided for general reference. Before undertaking a final system review or design, one should develop and incorporate the site-specific fuel and system requirements into the analysis. The PRB coal development is based on the following general assumptions: 1. 80 F ambient air temperature and coal inlet temp; 2. Assumed 50% of the total coal moisture is dried during the grinding process; 3. 10% coal mill radiation loss; 4. 135 F coal/air exit temperature; and 5. Moisture lines based on as-received total coal moisture.
Fuel Preparation
237
A comparison of the preceding curves for the analysis finds the following. The eastern bituminous coal case with a 1.7 air/coal ratio and 6% total coal moisture will require a coal mill inlet air temperature of approximately 400 F. The PRB coal conversion case with a 1.6 air/coal ratio and 30% total coal moisture will require a coal mill inlet temperature of approximately 750 F. This is a significant increase in primary air system and air heater performance that is most likely beyond the existing system capabilities. This implies that the existing system will need to be updated with a system that can provide the required primary air flow and temperature. The current trend in the industry is to utilize in duct heaters that use natural gas, steam, or electricity to provide the required primary air heating or to utilize low blend percentages of the PRB coals until the system limitations are reached. 6.5.2.5 Analysis Summary A review of the analysis performed finds that there are significant differences in the system requirements for the eastern bituminous and PRB coals. The significant heating value and moisture differences make it very difficult to switch between the two fuels. The key information from this analysis is that the coal mills have the capability to process the additional PRB coal flow due to the original design margins in the system, but it is very unlikely that the remaining components of the fuel preparation system will be able to support the coal mills. This is a theme repeated regularly when one reviews fuel switch options. All the fuel preparation system components must be reviewed concurrently to assure the system is capable of meeting the plant requirements. The preceding analysis is a foundation that can also be utilized for determining the effects of alternative coals on plant performance. Looking at additional coal sourcing within the same fuel classification for economics or supply flexibility should include the analysis points noted previously to ensure that the system requirements are met. The analysis methods presented can be used for screening the fuels for general applicability. If system issues arise during the analysis, they should be addressed with the equipment suppliers to determine the best path forward.
6.5.3 Coal Mill Capability Test Plan A well-defined and comprehensive coal mill system capability test plan is required to accurately benchmark the existing operation of a coal mill system prior to undertaking a comprehensive fuel-switching evaluation. Once the system is benchmarked, reasonable engineering evaluations can be performed on system upgrades and fuel-switching and blending options. Without a benchmarked system, the probability of falling short of the program goals is very high.
238 Combustion Engineering Issues for Solid Fuel Systems
The intent of a comprehensive coal mill capability test plan is to Determine the maximum reliable operating system coal flows; Determine the fineness versus coal load characteristics for the
plant fuels; Determine the mill system pressure balance versus coal flow char-
acteristics up to maximum coal mill coal flow; and Provide maximum coal flow operating data for the primary air fan,
forced draft fan, and combustion system analysis. Following are the major components of the test plan. Details of the test procedures are available in numerous ASTM standards and other available documentation. 1. Clean air testing to calibrate the coal mill airflow elements. The clean air testing is performed by traversing the coal pipes with clean air flow through the mill. The data are reduced to a calculated air flow that is compared to the coal mill air flow measuring device. 2. Coal mill fineness testing is performed by extracting isokinetic coal samples from the coal pipes during steady state coal mill operation. The samples are then sieved through 50 mesh, 100 mesh, and 200 mesh sieves (or as defined by test requirements) per standard protocols to understand the size fraction ranges of the coal particles. This information is used to compare the operating characteristics with subsequent testing to characterize the coal mill performance over the expected operating ranges of coals and coal loading. Having the coal mill benchmarked during normal operating conditions is also useful in diagnosing mill problems and off-design operation. 3. Coal mill pressure balance testing is performed to characterize the operating pressure in the critical areas of the coal mill. The data are taken for the high and low sides of the coal mill pressure transmitters. The coal mill pressure balance data are useful in understanding the internal working of the mill. The coal mill pressure balance will change based on the condition of the grinding components, the classifier setting, and the fuel quality. Having a benchmark understanding of the fuel preparation system performance will help in diagnosing mill problems or off-design operation. 4. A raw coal sample is taken during each coal mill test and analyzed for a) Proximate analysis; b) Ultimate analysis; c) Ash mineral analysis; d) Raw coal size; and e) Hardgrove grindability index.
Fuel Preparation
239
Benchmarking the coal being ground during the coal mill testing is one of the most important steps in setting up a coal mill test program: 1. Primary air duct pressure: Taken from the high side of the primary air to windbox differential transmitter. This information is useful for analyzing the primary air supply system capability. 2. Coal pipe distribution: To be performed using a dirty air pitot test procedure to determine the coal pipe to coal pipe coal flow splits. This information is very useful in diagnosing and controlling the boiler carbon monoxide emissions and unburned carbon levels. The more even the distribution between the coal pipes the better the coal burners will work. 3. Air heater leakage and pressure balance testing: Testing is performed by measuring gas constituents and temperatures around the air heater to determine the efficiency heating the primary airflows, the amount of air bypassing the air heater, and the leakage from the primary air side to the flue gas. 4. Primary air fan performance: To determine the fan system operating characteristics and available margins.
6.6 References 1. Stultz, S.C., and J.B. Kitto (eds). 1992. Steam, Its Generation and Use, 40th ed. Barberton, Ohio: Babcock & Wilcox. p. 8–6, Table 5. 2. Biomass Energy Foundation website, www.biomassenergyfoundation.org/: Biomass Energy Foundation: Proximate/Ultimate Analysis, accessed December 29, 2007. 3. Babcock & Wilcox Company. 1960. Steam, Its Generation and Use. New York: McKibbin & Son. 4. The Engineering Tool Box, Fuels—Densities and Specific Volumes, www. engineeringtoolbox.com, accessed 2006. 5. Tillman, D.A., and N.S. Harding. 2004. Fuels of Opportunity: Characteristics and Uses in Combustion Systems. Oxford, UK: Elsevier Ltd. p. 34, Table 2.2. 6. Tillman, D.A., and N.S. Harding. 2004. Fuels of Opportunity: Characteristics and Uses in Combustion Systems. Oxford, UK: Elsevier Ltd. p. 138, Table 4.3. 7. Wood Energy—Wood Fuel Characteristics, Biomass Fuel: Wood Energy. http://www.woodenergy.ie/iopen24/defaultarticle.php?cArticlePath=5. 8. Tillman, D.A., and N.S. Harding. 2004. Fuels of Opportunity: Characteristics and Uses in Combustion Systems. Oxford, UK: Elsevier Ltd. p. 194, Table 5.2. 9. Tillman, D.A. 2000. Biomass Cofiring: The Technology, The Experience, The Combustion Consequences. Biomass & Bioenergy. 19: 365–384. 10. Purutyan, H., B.H. Pittenger, and J.W. Carson. 1999. Six Steps to Designing a Storage Vessel That Really Works. Powder and Bulk Engineering, November 1999, Vol. 13, No. 11, pp. 56–68. 13(11).
240 Combustion Engineering Issues for Solid Fuel Systems 11. Royal, T.A., and J.W. Carson. 1991. Fine Powder Flow Phenomena in Bins, Hoppers, and Processing Vessels. Proc. of Bulk 2000: Bulk Material Handling Towards the Year 2000, London. 29–31 October. 12. Carson, J.W., and G. Petro. How to Design Efficient and Reliable Feeders for Bulk Solids. Original basis for paper printed in various forms, www.jenike. com. 13. Pennsylvania Crusher Corp. 2003. Bulletin 4050-D 00-3-03R-1M: Handbook of Crushing. 14. Kukoski, A.E. 1992. Ball Mill Pulverizer Design. Proc. of Low Volatile Fuels, Double Arch Firing Technology Applications Symposium, Beijing, PRC June 5–6.
CHAPTER
7
Conventional Firing Systems Peter Marx
President Advanced Combustion Technology, Inc. and
Jeffrey Morin Staff Engineer Advanced Combustion Technology, Inc.
7.1 Overview When one is designing any type of system, it is most often thought that the main objective is to create one that serves its purpose through an efficient, cost-effective, and law-abiding process. For the combustion of solid fuels for power generation or industrial usage, the science behind the types of fuels available, prevailing economics, and environmental awareness at the time of construction dictates which type of fuel burning system is advantageous for the given situation. Throughout the history of power generation and industrial application of fuels in boilers, process heaters, and kilns, a myriad of combustion systems has been designed to meet the requirements of societal and economic conditions; however, key system designs have proven to bridge the wants and needs of society and the economic entities therein. Such designs include the traditional stoker, pulverized, and cyclone firing systems that may be seen in power plants and industrial applications across the globe. This chapter provides an overview of the traditional methods of solid fuel firing focusing on fuels, general design parameters, and applications. The interplay of these parameters determines the selection of the appropriate technology for burning the various types of solid fuel: various ranks of coal, biomass, petroleum coke, and waste-based energy sources. 241
242 Combustion Engineering Issues for Solid Fuel Systems
7.2 Types of Traditional Combustion Systems Traditional combustion systems include stokers, pulverized firing systems, and cyclonic firing systems. Each approach has been developed to meet specific needs and applications; each approach is still used today although pulverized firing systems tend to dominate the field.
7.2.1 Stoker Firing Systems Stoker firing systems involve the burning of material on a grate in a furnace. The type of material burned can range from coal to biomass. Since the material combustion takes place on and above a grate within a furnace, the combustion is often much more stable than other traditional combustion systems and can also have a higher combustion capacity [1]. Often, the furnace is simply used to generate a combustible gas for which the combustion is later completed within a boiler. Such systems are often classified as closecoupled gasifier-combustor systems. Figure 7-1 shows an example of a typical stoker-firing system. Stoker firing systems are classified by the manner in which the fuel is loaded onto the grate. Two of the main types of stoker systems used today are chain grate stokers and spreader-stokers [2].
7.2.2 Pulverized Firing Systems Pulverized firing systems are the most common methods for burning coal and are widely used around the world for power generation, specifically wall-fired or tangentially fired boilers employing pulverized coal-firing [3]. Arch-fired systems and roof-fired systems also exist, directing the flow of combustion products initially downward. As the name implies, pulverized
FUEL INLETS
TO THE BOILER
FUEL PILE
GRATE
FIGURE 7-1 Flat grate stoker-fired system (Source: [2]).
Conventional Firing Systems
243
firing systems employ the use of pulverizers to grind fuel to a small particle size (e.g., >70% passing 200 mesh [or 74 mm]). Chapter 6 provided a detailed discussion of pulverizers for fuel preparation. The pulverized fuel is pneumatically transported into the furnace, where it is combusted. A description of the combustion process is provided in Chapter 1. Pulverized firing systems are also used in process industry settings such as rotary kilns.
7.2.3 Cyclone Firing Systems Cyclone firing systems are based on using centrifugal forces to facilitate burning of crushed (e.g., 3/8" 0") fuel particles [3]. Since spiraling air drags the particles outwardly, they build up against the outer wall of the chamber, resulting in very high temperatures at the walls. Cyclone-fired systems operate at extremely high temperatures (typically >3,300 F), thus they often create high NOx emissions (e.g., 1.2–1.9 lb/106 Btu). Cyclone firing systems were originally designed to capitalize on the slagging properties of certain coals—Illinois Basin and Western Kentucky coals, Northern Appalachian coals—and to remove 70% of the inorganic matter in these coals as tapped slag or bottom ash.
7.2.4 Fluidized-Bed Systems Fluidized-bed combustion technology is a relatively new technology, with most of its coal-fired development starting in the 1950s and 1960s, with major commercialization occurring in the 1980s. It is a leading technology for the combustion of a range of fuels because of its inherent advantages over conventional combustion systems, including fuel flexibility, low NOx emissions, in situ control of SO2 emissions, excellent heat transfer, high combustion efficiency, and good system availability. A detailed discussion of fluidized-bed systems and their applications is provided in Chapter 8.
7.3 Applications and Uses of Conventional Firing Systems 7.3.1 Electricity Generation Coal-fired systems are the dominant sources of energy in the power industry; pulverized coal systems dominate, followed by cyclone-fired systems. (Circulating fluidized-bed systems, used increasingly for power generations, are discussed separately in Chapter 8.) A few installations employ stoker-firing, and many of the smaller independent power plants also use stoker technology. In the United States, coal generates roughly 50% of our electricity [4]. The next closest source of electricity is nuclear power (19.3%) followed by natural gas (18.7%). Data describing this distribution of electricity generation are presented in Table 7-1.
244 Combustion Engineering Issues for Solid Fuel Systems TABLE 7-1 Distribution of Electricity Generation by Fuel, 2005 Fuel/Energy Source Coal Petroleum Natural Gas Other Gases Nuclear Hydroelectric Other Renewables Hydro-Pumped Storage Other Total
Net Thousand Megawatt-hours Generated
Percentage of Total
2,013,179 122,522 757,974 16,317 781,986 269,587 94,932 6558 4,749 4,054,688
49.7 3.0 18.7 0.4 19.3 6.6 2.3 0.2 0.1 100.0
Source: [4]
The overall process for electricity generation starts with the combustion of the fuel. The heat that is generated during the combustion process raises the temperature of a series of tubes that are contained within a boiler. The tubes contain water, which eventually changes to highpressure/high-temperature steam due to the high temperatures present within the boiler. The high-pressure steam from the boiler passes through the blades of a turbine, thus spinning the turbine shaft. The rotation of the turbine shaft is then used within a generator that creates the final product of electrical current. The history of power generation in coal-fired boilers is one of increasing the pressure and temperature of the steam, and of adding reheat steam cycles, for improved boiler thermal efficiency and reduced fuel consumption per unit of energy generated. This phenomenon is the reduction in net unit heat rate, expressed typically as Btu/kWh. Figure 7-2 illustrates the increase in pressure of utility boilers during the period 1945–1996. Steam temperatures rose from an average design temperature of 900 F in 1945 to 1,000 F in the period from 1960–1996. There were two units built with design temperatures of 1,100 F and 1,200 F during the periods of 1960–1965 and 1956–1960, respectively. The metallurgy available at that time could not sustain such temperatures, and values in the 1,000–1,025 F range were commonly accepted as state of the art [5]. Today, cycles designed for 1,050 F are common, and there is constant pressure to raise the main steam temperature to 1,100 F and beyond. Reheat cycles also became common—and virtually always employed—by 1960 in coal-fired utility boilers (see Figure 7-3). As shown in Figure 7-3, a few units were designed with double reheat cycles; however, single reheat became the standard. The combination of more severe steam conditions and reheat cycles permitted utility boilers to continuously increase in capacity, as is shown in Figure 7-4. While the very large
Conventional Firing Systems
245
Increase in Main Steam Pressure over Time Main Steam Pressure (psig)
5500 Maximum Main Steam Pressure
5000 4500 4000 3500 3000 2500 2000 1500 1000
Average Main Steam Pressure
500
6 99
0 –1
19
91
99
5 19
86
–1
98
0 19
81
–1
98
5 19
76
–1
97
0 19
71
–1
97
5 19
66
–1
96
0 19
61
–1
96
5 19
56
–1
95
0 –1
95 51 19
–1 46 19
19
40
–1
94
5
0
Time Period (5-year Increments)
FIGURE 7-2 Increase in main steam pressure among coal-fired utility boilers (data from [5]).
Implementation of Reheat Cycles in Boilers 200
Total Number of Boilers
180 Number of Boilers
160 140 120 100 80 60
Boilers with Reheat Boilers with Double Reheat
40 20
19 40 –1 94 19 5 46 –1 95 19 0 51 –1 95 19 5 56 –1 96 19 0 61 –1 96 19 5 66 –1 97 19 0 71 –1 97 19 5 76 –1 98 19 0 81 –1 98 19 5 86 –1 99 19 0 91 –1 99 6
0
Time Period (5-year Increments)
FIGURE 7-3 Implementation of reheat cycles to increase coal-fired boiler efficiency (data from [5]).
246 Combustion Engineering Issues for Solid Fuel Systems
6 99
0 –1
19
91
99
5 –1 86 19
–1
98
0 19
81
98
5 –1 76 19
–1
97
0 19
71
97
5 –1 66 19
–1
96
0 19
61
96
5 –1
95 56 19
–1
95 51 19
–1
94 46
–1 19
40 19
0
750 700 650 600 550 500 450 400 350 300 250 200 150 100 50 0
5
Megawatts (Nameplate)
Average MW per Boiler
Time Period (5-year Increments)
FIGURE 7-4 Average capacity of coal-fired utility boilers over time (data from [5]).
units such as Cumberland Fossil Plant of the Tennessee Valley Authority (TVA)—10 106 lb/hr or 1,300 MWe per boiler—are rarely on the drawing boards, 600–800 MWe supercritical boilers are still proposed and built using pulverized coal technology. Within the United States, some 50% of the electricity is generated by use of coal—largely by pulverized firing systems. Cyclone boilers account for only about 15% of the coal-fired capacity among utilities, and stoker firing accounts for much less. Because of the vast resources and reserves of coal in the United States (see Chapter 2), coal is the natural choice of fuel due to the fact that it is so readily available within the country. Coal and the other solid fuels—petroleum coke, various forms of biomass, wastes— remain the backbone of the U.S. energy supply. Further, utility boilers are at the heart of the solid fuels combustion community as they consume over 90% of the coal produced in the United States.
7.3.2 Industrial Boilers, Kilns, and Process Heaters Solid fuel combustion systems are used across the world to provide high amounts of intense heat for various practices. Many of the combustion systems are used to simply provide heat for boilers to create steam. The steam can then be used in various processes such as heating, pulp and paper processing, textile formation, refineries, food processing, and
Conventional Firing Systems
247
pharmaceutical purposes. Such systems include, but are not limited to, the following: Process steam boilers for industry (e.g., steam boilers supplying
energy to the dry kilns of sawmills); Power boilers for the pulp and paper industry (commonly cogenera-
tion boilers producing steam at 950 psig/1,250 F for backpressure or automatic expansion cogeneration turbines, and exhausting steam at 150 or 50 psig for process applications); Process rotary kilns such as cement kilns, lime kilns for pulp and paper manufacture, and ore processing kilns for the mining industry; and Process heaters of other designs for a myriad of manufacturing industries. These do not include the use of coal, coke, and other solid fuels in the smelting and refining of metals (e.g., coal and coke for blast furnaces, wood chips for nickel refining, which has been practiced periodically). Of these, industrial boilers are most prominent. There are three main types of industrial-sized boilers that are used which include watertube, firetube, and cast iron. Watertube boilers are the conventional style used in the utility industry and in the process industry; and they are designed in such a way that water passes through tubes that are in direct contact with the heat created by the fuel combustion. Watertube boilers are mainly used for steam generation. Further, for the large cogeneration boilers, the watertube design is the only appropriate technology. Firetube boilers are designed so that the hot combustion gases flow through the tubes, which are then surrounded by the intended heating water contained in a large open vessel. Firetube boilers are limited in both capacity and steam pressure. Typically, these are small (e.g., 50,000 lb/h or less), low pressure (e.g., <50 psig) units. Cast iron boilers designed in a similar fashion to firetube boilers; however, they are made out of cast iron rather than steel. Firetube and cast iron boilers are generally used for space heating systems, and as portable power boilers [6].
7.4 Basic Issues 7.4.1 Fuel Selection When one is dealing with solid fuel combustion systems, it is crucial to understand the economic workings behind the process just like any capital investment. Many different factors can go into fuel selection, most of which are based on availability and final usage costs. For example, with recent “threats” for a carbon dioxide limit, natural gas has become an increasingly
248 Combustion Engineering Issues for Solid Fuel Systems
attractive fuel because natural gas combustion creates less carbon dioxide than does burning coal. The main advantage that can be seen with natural gas as well as oil is that there is little to no ash formation in the combustion process. Latent heat loss for natural gas and oil systems ranges from 10% to 6%, respectively. It is found that the most rapid heat release upon combustion is associated with natural gas; however, the heat flux for natural gas is not as great as it is for oil and coal [6]. While natural gas is a much “cleaner” fuel in terms of CO2 output and has a much easier start-up than coal, it is not necessarily the best choice of fuel. Currently, the world’s natural gas wells and oil refineries are at their maximum capacity. If every power plant decided to switch over to a natural gas system, gas availability would be inadequate, gas prices would skyrocket, and a crippling domino of price increases would occur for all industrial production. When coal is used as the fuel of choice for a combustion system, a great deal of research is needed to select the optimal approach for burning. While coal has many advantages in the United States, such as its relatively low cost, high availability, low latent heat (about 4%), and low flue gas moisture, coal can exhibit varying characteristics in the combustion process and system. It is a natural material that may or may not be washed or otherwise treated in a coal preparation plant; however, it is not refined into an engineered product. Because of this, the type of coal used can depend on emissions regulation [7]. Sulfur and SO2 regulations originally caused this fuel selection phenomenon. The sulfur regulations were a significant impetus promoting the use of Powder River Basin coals (see Chapter 2). Nitrogen oxides are often unwanted byproducts that have higher occurrences with coal firing because of the reactivity and the concentration of fuel-bound nitrogen typically expressed in lb N/106 Btu. Coal firing also creates a great deal of fly ash that can be problematic, especially when additional systems such as electrostatic precipitators are needed for particulate removal. Fuel selection also influences the use of opportunity fuels (see Chapter 3). In energy utilization systems where the choice of fuel is an opportunity fuel—biomass, for example—fuel selection is of critical importance. Gratefired systems designed for wood waste, or wood waste and coal, may experience unacceptable levels of slagging and fouling if hays, straws, or other biomass fuels are employed. The sodium, potassium, and chlorine contents of such fuels must be considered in system design and operation. Systems that can handle some opportunity fuels as supplements may limit their use (e.g., most petroleum coke firing systems limit the use of this fuel to 20% for flame stability reasons). Further, some coals are more suitable to petroleum coke as a supplement than other coals. High volatility coals are more suitable to blending with petroleum coke than medium or low volatility coals. Flame stability becomes the issue in this case. Other opportunity fuels fired include tire-derived fuels, pitches, and a wide array of wastes. Firing of opportunity fuels in conventional power systems requires significant attention to the selection of both the coal and the opportunity fuel [8, 9].
Conventional Firing Systems
249
7.4.2 Operational Considerations Operational issues associated with solid fuel combustion include management of temperatures, heat transfer, and deposition—slagging and fouling. The design and operation of firing systems must consider all these. Temperatures are directly impacted by the design and operation of burner systems. Cyclones, for example, generate temperatures of 3,300– 3,550 F. As such, the heat liberation is on the order of 350,000–500,000 Btu/ft3-h in the cyclone barrel [9, 10]. This accounts for the high levels of NOx generation. Cell burners—placing two or three relatively small wall-fired pulverized coal burners in a confined area—also liberate a significant amount of energy and generate high temperatures. Heat transfer issues are related to temperature issues. They are also related to slagging and fouling. Slagging and fouling may be a direct consequence of the firing system configuration. Immediate contact of the flame with separated overfire air— particularly in a close-coupled arrangement—can quench combustion reactions and can also cause condensation of inorganic matter. “Eyebrows” and “smiles” can be caused by such phenomena. Managing slagging and fouling of other types also is required of the firing systems. Figure 7-5 depicts slag formed by a pulverized coal firing system with intense heat release in the burner zone.
FIGURE 7-5 A photograph of slag formed by intense heat release in the burner zone of a pulverized coal boiler. Note the glassy appearance of the slag as formed. (Photo courtesy of A. Widenman)
250 Combustion Engineering Issues for Solid Fuel Systems
7.4.3 Airborne Emissions Airborne emissions are the final consideration reviewed. Because of its significance, the major pollutants—particulates, SO2, NOx, CO2, and hazardous air pollutants—are reviewed separately. 7.4.3.1 Particulates Particulates are naturally occurring solids or liquids that are formed from the reactions of various substances. Particulates can originate from a variety of chemical interactions and may be emitted into the air surrounding their formation. Examples of particulates formed during a reaction include solid ash particles, unburned carbon, and condensable volatile organic compounds (VOCs). Depending on the combustion application, various restrictions exist regarding particulate formation. The size, quantity, and characteristics of the particles formed during a reaction process determine how operating conditions should be modified to minimize their creation. The main issues associated with particulate emission include concerns for public and worker health—particularly issues associated with breathing fine particles [11]. Chapter 9 deals specifically with this and other pollution-control issues. 7.4.3.2 SO2 Sulfur oxides are one of the most costly products of solid combustion due to the amount created during certain types of combustion and how it is not easily removed from the process. The main methods that have been used to reduce sulfur oxides in the past years have been to use coals or other solid fuels (e.g., biomass) that have a lower sulfur content than many of the traditional coals; alternatively, the options include using wet, semidry, or dry scrubbers (see Chapter 9). Overall, SO2 formation accounts for more than 90% of the sulfur contained within coals. SO3 is another byproduct that can be formed in the combustion process (as well as with the conversion of SO2 to SO3 and then H2SO4 in the atmosphere). H2SO4 has been identified as a health hazard and is also a component for acid rain formation. 7.4.3.3 NOx NOx is a pollutant that most combustion plants keep at a minimum, in accordance with environmental legislation dating back to the Clean Air Act of 1970. Nitrogen oxides most often come in the form of NO but can also be produced as NO2. NO can oxidize in the atmosphere to NO2. NO2 can then dissociate in the atmosphere to NO and atomic oxygen. Ozone (O3) may be formed from the atomic oxygen and is typically regarded as one of the main components of urban smog. NO2 can produce other pollutants such as peroxyacetyl nitrate (PAN) if it combines with hydrocarbon
Conventional Firing Systems
251
radicals. If NOx is oxidized to nitric acid or nitrates and is coupled with wet or dry decomposition, it can be removed from the atmosphere. NOx is formed either by combustion reactions associated with fuel nitrogen, or by fixation of atmospheric nitrogen at very high temperatures (thermal NOx). The former reactions occur when nitrogen in the fuel—particularly volatile nitrogen—reacts with an excess of oxygen in the combustion oxidant stream. Thermal NOx is formed typically by the extended Zeldovich mechanism and by the “Prompt NO” mechanism as defined by Fenimore (see Chapter 1 for details). The production of NOx is directly correlated with the combustion temperature and the availability of oxygen, which is supplied by the firing system. 7.4.3.4 CO2 Until recently, production of carbon dioxide was a major objective, not a major concern for combustion systems. With current global warming concerns, government agencies have been looking to lessen the effects that power generation has on the environment. CO2 is cited as one of the main greenhouse gases that could be changing the overall global temperature. According to the Environmental Integrity Project [11], upwards of 40% of all carbon dioxide emissions come from power plants. This proves to be a major issue for the power industry as nearly all combustion processes involve CO2 as a product of the burned fuel. CO2 emissions are directly a function of the carbon content of the fuel. Alternative forms of carbon evolving from combustion processes include unburned carbon in the fly ash, carbon monoxide (CO), and hydrocarbon emissions from the combustion process. All such outcomes involve less efficient burning of the fuel and “throwing away” of otherwise useful energy. A new industry dealing with “Carbon Sequestration” has evolved from present policies and possible future regulations. One suggestion has been to consider storing the carbon dioxide produced from the plants on the ocean floor in a liquid form. Other suggestions include using algae to adsorb the CO2 followed by converting the algae to use as a biofuel, using the carbon dioxide in geological sites such as unminable coal seams, or even trapping the carbon in mineral formations known as carbonate salts. The current combustion response is to burn the fuel in an oxycombustion system; pure (or relatively pure) oxygen is blended with flue gas for thermodynamic and heat transfer reasons; this oxidant consisting of O2, CO2, and H2O is then substituted for combustion air. The result is a flue gas stream with a high concentration of CO2—sufficient for capture and subsequent sequestration [12, 13, 14]. 7.4.3.5 Other Emissions (Hazardous Air Pollutants) Other emissions resulting from combustion systems include carbon monoxide (CO), hydrocarbons (regulated as total hydrocarbons, nonmethane
252 Combustion Engineering Issues for Solid Fuel Systems
hydrocarbons, or volatile organic carbon), and selected trace metals such as mercury. Hazardous air pollutants (HAPs) include a suite of metals, hydrogen chloride, fluorides, and an array of other compounds.
7.5 Firing Systems and Combustion Issues Conventional firing systems all approach combustion issues—temperature, heat release, operational considerations, and emissions management—in unique ways. The following sections detail firing systems applications to these issues. Because stoker firing is the most traditional approach, it is discussed first, followed by pulverized coal and cyclone firing.
7.5.1 Stoker Firing 7.5.1.1 Basic Description and Identification of Types Several grate designs are employed, depending on the material being burned, including the following: Traveling “pinhole” grates where the underfire air serves both as
primary combustion air and as the cooling medium for the grate bars; Chain grates, which are one variation on the traveling grate; Stationary flat grates with a dumping section at the end, again using combustion air for grate cooling as well as combustion; ash is removed either by air jets or steam jets, or may be hand raked in small systems; Hydro-grates, where the grate is cooled by water from the feedwater circuit, and where the combustion function is separated from the grate bar cooling function; Sloping grates either with all stationary parts or with some moving parts; many sloping grates used for biomass or municipal solid waste combustion are designed to agitate the bed of fuel either by reverse upward thrusts (e.g., the Martin Grate) or by other mechanical actions; and Roller grates which are set on a slope where roller action tumbles and agitates the bed material.
There are also several types of fuel-feeding mechanisms. For chain grate stokers and sloping grate systems, fuel can be fed into a feed chute and then dumped or dragged onto the grate in the furnace. Wood-fired stokers, such as the large power boilers in pulp mills or in small power plants, typically use “wind swept spouts” where a portion of the combustion air injects the fuel above the bed. It then falls to the bed. Coal-fired stokers commonly use “paddle wheel” fuel feeders to throw the fuel onto the grate.
Conventional Firing Systems
253
Detroit Stoker has designed and offers a combination air injector-paddle wheel for combined firing of wood and coal in large pulp mill boilers. In the United States, where the forest products industry along with other process industries dominate the use of stoker firing, the main type of stoker used is the spreader style. Both traveling “pinhole” grates and hydrogrates are employed. With the spreader style, feeders spread the fuel onto grates below. The bed on the grate is relatively thin and uniformly distributed. This contrasts with chain grates and sloping grates where the fuel bed is relatively deep. Once on the grates, the fuel dries, ignites, and finally burns. The volatile matter released during the initial stages of combustion burns in the furnace space above the bed while the char burns on the grate. The purpose of the grates is to collect the ash from the burned fuel in an ash pit where it is then discarded. Typically, 80% of the ash exits from the stoker boiler as bottom ash or grate ash, and only about 20% exits as fly ash. Advantages for spreader stokers, relative to other types of stokers, include their high combustion capacity, their ability to respond to changes in power demands relatively quickly, and the fact that the system can be shut down in a relatively short amount of time [2]. In the United States, the main type of stoker used is the spreader style for prepared fuels such as hogged wood waste. Chain grates, vibrating grates, and similar designs are more common when firing mixed municipal waste in incinerators. With the spreader style, feeders spread the fuel onto grates below. Once on the grates, the fuel dries, ignites, and finally burns. The purpose of the grates is to collect the ash from the burned fuel below in an ash pit where it is then discarded. Advantages for spreader stokers include their high combustion capacity, their ability to quickly respond to changes in power demands, and the fact that the system can be shut down in a relatively short amount of time. A common issue among spreaders is the fact that the mechanical grates often require a great deal of maintenance. Air-cooled pinhole grate stokers, which are used in the largest units, are designed to have a continually moving grate that moves at a constant velocity. The solid fuel is loaded on one end of the grate where its thickness is regulated to maintain optimal burning levels. Primary air is sent through the moving grate to ensure proper combustion inside the furnace. The fuel moves down with the grate and is eventually ignited once it enters the furnace. As the fuel burns on the grate, its thickness decreases. At the end of the furnace, the remaining ash from the combustion process is dumped into some type of pit where the grate travels around a roller. 7.5.1.2 Fuel Selection for Stokers Stoker systems are fairly versatile when it comes to fuel selection choices. Stoker systems can run on basically any type of coal as long as
254 Combustion Engineering Issues for Solid Fuel Systems
environmental precautions are met post-firing. Stokers also have their place in the renewable energy field with their ability to burn biomass fuels such as wood wastes and straws. As a renewable energy source, the carbon dioxide created by burning such fuels is thought to be negligible because it is reabsorbed into the biomaterial from the atmosphere as it is regrown. Also, very small amounts of metal and sulfur are released during biomaterial combustion. With stoker systems, unwanted human wastes such as chipped old tires, mixed municipal waste, and many industrial byproducts can be used in the combustion process. One main advantage to stoker systems is that fuel can be easily mixed depending on what is available at the time of combustion. 7.5.1.3 Fuel Preparation When fuel is prepared for stoker firing systems, it is crucial to understand the burning attributes of the fuel. Depending on the type of fuel, the height at which the fuel is loaded (or spread) onto the bed can make a significant difference with final ash products. For typical chain grate stokers, top size is usually limited to 0.75 to 1.5 inches. For spreader stokers, the fuel should be only about 1 inch on the bed. The density of the fuel on the grate also makes a difference for complete combustion purposes. It should be noted that when the fuel is ground for combustion, fine particles should be kept to a minimum so that they do not fall through the grate. For chain grate stokers, fine particles should be minimized, certainly kept to less than 50% passing through a 6-mm round-hole screen [7]. With spreader stokers, coals should have 95% less than 1.25 inches, with 2 inches maximum size. Fuel loaded into a stoker firing system may need to be dried depending on the fuel used. For biomass and other waste fuels, a moisture level less than 55%–60% is optimal for firing; however, fuel moisture mainly depends on the type of firing system used [7, 15, 16]. 7.5.1.4 Design Parameters Spreader stokers can be used in a wide variety of boiler sizes. Depending on the fuel type, inputs can range from 50 106 Btu/h to 530 106 Btu/h for coal and 900 106 Btu/h for some waste fuels [7]. The use of 900 106 Btu/h is very rare, but 600–750 106 Btu/h is common for the large stoker-fired units. These would include the McNeill Generating Station in Burlington, Vermont, fired with wood waste; the power boilers at the Longview, Washington, Columbus, Mississippi, Plymouth, North Carolina, New Bern, North Carolina, and Valliant, Oklahoma, pulp and paper mills of Weyerhaeuser Company; and many other examples of small power plants and pulp mills of Georgia Pacific, International Paper Company, and other forest industry companies. Because the fuel particles are relatively large (e.g., 1–2" 0"), rates of heat release are somewhat lower than those associated with pulverized
Conventional Firing Systems
255
firing or cyclones. For coal, plan area heat release rates are commonly specified by potential purchasers at 500,000–750,000 Btu/ft2-h. For biomass such as hogged wood waste, plan area heat release rates are more commonly 750,000 – 1 106 Btu/ft2-h. Volumetric heat release rates for coal can be on the order of 20,000–25,000 Btu/ft3-h, while volumetric heat release rates for biomass fuels are best specified at 15,000 Btu/ft3-h. In a few cases, 20,000 Btu/ft3-h is used although this can stress the furnace. Excess air, measured as excess O2, is somewhat higher in stoker firing than in pulverized coal firing. While some advanced units can employ excess air at 3% O2, more common experience is in the 4–5% O2 range. Excess air depends heavily on the moisture in the fuel. Stoker firing is capable of handling fuels with up to 55–60% moisture while functioning properly. For this reason, stoker firing is commonly used for burning bark and wood waste in pulp and paper mills. Combustion temperatures for stoker firing depend on the fuel being fired. Municipal waste combustion typically involves temperatures from 1,800 F to 2,200 F. Biomass firing is typically in the 2,000–2,500 F combustion temperature range. Coal firing is comparable to biomass firing. Temperature is heavily dependent on the condition of the fuel, particularly the moisture content. 7.5.1.5 Functioning of Grates As identified previously, several different types of grates are used for the fuel to burn on, each with their own unique function. The main purpose of the grates is to minimize ash and fuel falling through while allowing for efficient fuel burning. Grate cleaning types can be separated into two categories: stationary and dumping. Stationary cleaning grates are rarely used because of the dangers that are imposed on the operator for reasons of direct exposure to the furnace and high opacity levels. For dumping grates, there is a separate section for each fuel feeder. To remove the ashes, the fuel feed is stopped at one of the grate sections and the air is cut off. The fuel is then dumped and the feeder is restarted. Reciprocating grates remove ash by moving back and forth with alternating stationary grates. Ash is pushed forward and eventually falls into the hopper by the stepped grates, which rest on one another. Vibrating grates rest on flexing plates that are vibrated by an eccentric drive. The timed vibrating intervals bring the ash to the front so that it may be dropped into the ash pit. The most popular type of ash removal system is the traveling-grate spreader stoker. Airflow is supplied through the grate as it moves forward and eventually dumps the remaining ash at the front of the loader. Water-cooled grates may also be used in certain situations and employ a 6 grade down toward the end with the ash pit. Vibrations are often coupled with water-cooled grate systems. Figure 7-6 shows an example of a spreader system, and in particular, one that is water-cooled and vibrating.
256 Combustion Engineering Issues for Solid Fuel Systems
Over Fire Air Nozzles
Air Swept Refuse Distributor Spout
Over Fire Air Nozzles Eccentric Grate Drive
FIGURE 7-6 Configuration of a water-cooled stoker system (Source: [7]).
7.5.2 Pulverized Firing 7.5.2.1 Applications As previously stated, pulverized firing systems are used in a variety of applications. Pulverized firing is the current system of choice for the power industry as well as for other industrial expenditures. Many industrial boilers use the pulverized firing system in heat treatment plants, heat generation equipment, as well as chemical and food processing. There have been special designs of pulverized firing systems such as blast furnaces that are often used bloomeries for iron, blowing houses for tin, and smelt mills for lead. Cement kilns also make use of the pulverized firing system to heat raw materials such as limestone, clay, and sand to temperatures around 2,700 F to form the main cement ingredient called clinker. As stated earlier, coal is the primary fuel used in pulverized firing systems. When coal is fired, ash removal is often a major issue that requires the employment of multiple systems to accomplish the task. In Figure 7-7, the schematic of a fairly basic pulverized coal plant is shown where ash and particulate removal takes place. When coal is fired, the ash created is mostly fly ash; however, 15–35% can be bottom ash or slag.
Conventional Firing Systems
Furnace Coal Handling
Coal Bunker Burners Coal Feeder Pulverizer
Stack
Convective Pass Air Preheater Particulate Collection Device
FD Fan
Ash
257
SO2 Control Device
ID Fan
Bottom Ash/ Slag Removal
FIGURE 7-7 Example of a pulverized coal-firing plant with ash removal (Source: [7]).
7.5.2.2 Basic Description and Identification of Types With pulverized coal units, three basic designs are used to achieve maximum contact between the reactants: wall-fired, pulverized-coal firing; tangentially fired pulverized coal units; and vertical- or arch-fired pulverized coal boilers. One type of design involves the division of the fuel and air into multiple streams that are each treated separately, creating numerous flame envelopes. This style of combustion limits mechanical mixing of the particulates at the beginning, and often it can be difficult to achieve an optimal air-fuel ratio. The other design involves a conjoined chamber that the air and fuel are brought through together. Mixing occurs from the burning throat, and the continual turbulence makes it so that air and fuel distribution is less of a problem. 7.5.2.3 Wall-Fired Pulverized Coal Boilers and Firing Systems The wall-fired boiler setup is typical for pulverized firing systems. The burners may be arranged all on one wall of a furnace or on opposing walls of a furnace. Special cases exist in which some furnaces have burners on all four walls; however, this is rare. The fuel and air are brought into the nozzle tangential to the opening, allowing for high rotational interaction. Secondary air comes in from the windbox and may be directed by inlet vanes or specific swirler designs that can be used to manipulate the shape of the flame. With wall-fired systems, typical temperatures range from 2,500–3,200 F. Since many of the power plants that are currently in operation were constructed prior to the start of the strict environmental regulations that plants must follow today, many wall-fired systems are being retrofitted with new burners and overfire air to help lower NOx emissions and the formation of other harmful pollutants. Wall-fired burners are also the common design for process industry rotary kilns (e.g., cement kilns) and other process heating systems.
258 Combustion Engineering Issues for Solid Fuel Systems
In today’s regulatory climate, all wall-fired burners are low-NOx burners. The typical design, originated by the International Flame Research Foundation (IFRF) some 40 years ago, is a swirl-stabilized burner [17, 18]. Alternatively, coal can be distributed and the flame can be shaped by use of impellers. In swirl-stabilized burners, primary air carries the fuel to the furnace. Typically, the burner uses a primary air-fuel ratio of 1.7–2:1 (lb air/lb fuel). Secondary air, and at times tertiary air, swirls around the outside of the flame, thereby shaping the flame and stabilizing the activity. Figures 7-8 and 7-9 show, schematically, the performance of a typical wallfired swirl-stabilized burner. Flame shaping can be used either to feature a shorter, “bushy” flame or a longer, more laminar flame. In general, the heat input to the burner, measured in 106 Btu/h, divided by 5–7, indicates the approximate length. Higher swirl numbers will result in shorter, bushier flames, and lower swirl numbers will result in longer, “lazier” flames. As a practical matter, coal transported to the wall-fired burner forms ropes as a function of coal pipe elbows. Burners are commonly equipped with “rope breaker” designs ranging from distribution discs to pumpkin teeth. Figure 7-10 shows the installation of an Advanced Combustion Technology (ACT) wall-fired burner at Detroit Edison’s Monroe Power Plant. Note the distribution disc at the entrance to the burner pipe. This disc ensures that the coal rope will be broken up and that the coal will be evenly distributed in the burner pipe and in the flame. The shaping of the flame can then be depicted either by computational fluid dynamics modeling (see Figure 7-11) or by photography (see
FIGURE 7-8 Functioning of a wall-fired swirl-stabilized burner. Note the secondary air shaping the flame and creating the recirculation zones in the center of the flame. (Courtesy of Advanced Combustion Technology)
Conventional Firing Systems
259
FIGURE 7-9 Operation of a swirl-stabilized burner. Note the flame shaping effect of the secondary air. (Courtesy of Advanced Combustion Technology)
Figure 7-12). Figure 7-11 depicts a flame shape as a function of temperature. Note that the darkened edges of the flame in the picture are the hottest. Note, also, the lines indicating flow. These lines show the recirculation zone at the center of the flame. The figure is shown only as a representation of CFD modeling and the consequent shape of the flame. Figure 7-12 is a photograph of a single swirl-stabilized flame from the side view. The swirl creates the wide trajectory of the flame, and the swirl can be adjusted by adjusting the spin vanes for the secondary air. Note that, with reasonable swirl, the flame can have a wide trajectory and a relatively short distance, depending on the heat input and heat release of the burner. Under most circumstances, each burner will fire a maximum of 200–250 106 Btu/h of fuel, supporting 25 MWe of generation. Larger burners have been deployed from time to time, increasing the flame and heat release intensity for each burner. Such larger installations increase the potential for slagging in the boiler, depending on the fuel being fired. There are numerous vendors of wall-fired burners including the primary boiler manufacturers—Foster Wheeler, Alstom, Babcock & Wilcox,
260 Combustion Engineering Issues for Solid Fuel Systems
FIGURE 7-10 Installation of a swirl-stabilized wall-fired burner at Monroe Power Plant of DTE. Note the distribution disc at the back of the burner in the upper right quadrant, and the swirlers employed for the secondary air to shape the flame. (Courtesy of Advanced Combustion Technology) 3.70E+03 3.60E+03 3.49E+03 3.39E+03 3.29E+03 3.18E+03 3.08E+03 2.98E+03 2.87E+03 2.77E+03 2.67E+03 2.56E+03 2.46E+03 2.36E+03 2.25E+03 2.15E+03 2.05E+03 1.94E+03 1.84E+03 1.74E+03 1.63E+03 1.53E+03 1.43E+03 1.32E+03 1.22E+03 1.12E+03 1.01E+03 9.10E+02 8.07E+02 7.03E+02 6.00E+02
FIGURE 7-11 Computational fluid dynamics depiction of the temperature in a wallfired burner flame. (Courtesy of Advanced Combustion Technology)
Conventional Firing Systems
261
FIGURE 7-12 Photograph of a single wall-fired burner swirl-stabilized flame. (Courtesy of Advanced Combustion Technology)
Babcock Power—and numerous specialty manufacturers—ACT, GE-EER, Advanced Burner Technology owned by Siemans-Westinghouse, and many more. Each burner manufacturer provides unique characteristics for the product. Figure 7-13 shows a typical Foster Wheeler burner, which is one of many such products in the marketplace. In recent years, experiments have been conducted to modify wall-fired burners either for oxygen-enhanced combustion or for biomass cofiring [9, 20]. These modifications are designed for further NOx reductions and for the generation of green—CO2 neutral—power. Praxair, Inc. developed the oxygen-enhanced pulverized coal technology and demonstrated it at the Mt. Tom Station. Foster Wheeler, GPU Genco, and Electric Power Research Institute (EPRI), working with the U.S. Department of Energy (DOE), developed the use of wall-fired burners for cofiring sawdust with coal. Wall-fired burners have been commonly used to fire blends of petroleum coke with coal, holding the percentage of petroleum coke to a maximum of 20% for flame stability purposes. Designs now exist to fire
262 Combustion Engineering Issues for Solid Fuel Systems
Oil Guide Tube Perforated Plate/ Air Inlet
Coal Inlet
Burner Throat Refractory/Quarl
Flame Scanner Inner Air Zone Cone Damper
Adjustable Inner Sleeve Split Flame Tip Axial Swirler
FIGURE 7-13 A Foster Wheeler wall-fired burner, representing the many products in the marketplace (Source: [19]). (Photo courtesy of Foster Wheeler North American Power)
pulverized coal boilers with coal in a stream of oxygen and recycled flue gas using wall-fired technology. Babcock & Wilcox, teamed with Air Liquide, is developing a 50 MWe project to use this approach for generation of a highly concentrated stream of CO2, which can then be captured and sequestered in the oil fields of Canada [12]. 7.5.2.4 Tangentially Fired Pulverized Coal Boilers With tangentially fired burners, each corner of a square or rectangular furnace houses a series of burners located in the corners of the furnace; these burners are directed toward the center of the furnace. The burners are aimed in such a manner that they fire tangentially toward a center flame circle. Typical values are 4–6 for square furnaces. The boiler functions much like a single burner with multiple fuel injection points. As the burners are stacked above one another in the furnace, the vortex created by the flames becomes greater and greater. In comparison to wall-fired units, mixing at the throat of the burner is often decreased due to the decreased turbulent zone along with the fact that air and fuel are usually separated and sent into the furnace with straight paths. Mixing does occur, however, at the center vortex of the furnace where the separate streams react with one another to create turbulence. Temperature changes affected by variation in load may be counteracted with movable nozzles that can be raised or lowered to adjust heat absorption in the furnace. The design and construction of tangentially fired boilers
Conventional Firing Systems
263
have long been the province of Alstom (formerly Combustion Engineering) although Mitsubishi Boiler and many other vendors supply tangentially fired systems. Each corner, then, contains multiple burner or coal injection penetrations. Further, the corners also contain air injection locations. Given the need for low NOx firing, each corner also contains injection points for separated overfire air (SOFA). Figure 7-14 depicts a tangentially fired system developed by Foster Wheeler, for application to existing tangentially fired boilers. The angle of tangentially fired coal injectors and secondary air injectors can be varied. The ability to tilt burners up and down, or in yaw, has long been a feature of these boilers. Depending on the angle at which the burners are placed and the tilt at which they are set, the inner furnace aerodynamics may be manipulated such that optimal combustion is achieved. Typically, the intense rotational speed and low pressure region at the center of the “fireball” make it so that combustion particles and gases are attracted to the center of the furnace. Tangentially fired boilers have been very successful in low-NOx firing configurations. With the application of low-NOx firing systems and firing either lignite or Powder River Basin coal, tangentially fired boilers at Gibbons Creek and at Limestone Station have achieved <0.15 lb NOx/106 Btu.
Separated OFA w / Yaw & Tilt
Separated OFA w / Yaw & Tilt
OFA
OFA
Aux Boundary Air
Aux Boundary Air
Aux Boundary Air
Coal
Coal
LFSC
Coal
LFSC TLN 1
LFSC TLN 2
TLN 3
FIGURE 7-14 The tangentially fired system developed by Foster Wheeler for low-NOx boilers (Source: [19]). (Photo courtesy of Foster Wheeler North American Power)
264 Combustion Engineering Issues for Solid Fuel Systems
Tangentially fired boilers have also been successfully retrofitted to fire various biomass fuels in the pursuit of CO2-neutral power generation. Demonstrations with woody biomass have been performed at Albright Generating Station of Allegheny Power. Demonstrations with herbaceous biomass—switchgrass and a variety of other grasses and hays—have been successful at the Gadsden Generating Station of Southern Company and the Ottumwa Generating Station of Alliant Energy. The Albright demonstration also showed the capability of biomass to reduce NOx emissions during tangentially fired boiler operations.
7.5.2.5 Vertically Fired (Arch-Fired) Boilers Vertically fired—down-fired—systems are most often used for low volatile solid fuels that are difficult to ignite. With vertically fired systems, the high moisture or ash-free volatile matter is forced through nozzles that are suspended from arched walls in the furnace. High-pressure air is used to ensure that the fuel achieves optimal circulation within the furnace. Tertiary air is introduced along the sides of the furnace, which helps direct the flow pattern within the furnace. A “U-shaped” flame is formed at the bottom of the furnace where the volatile gases travel back up the center of the furnace. The high temperatures traveling back up the furnace help to ignite the incoming fuel. A main advantage of vertically fired systems is that the direction of the flow makes it so that the largest fuel particles are contained within the furnace for the longest amount of time, allowing for complete combustion. Figure 7-15 shows the vertically fired system. Figure 7-16 shows the typical arch-fired burner. In these systems, primary and secondary air enter through the burner with the tertiary air entering at the sides of the boiler.
7.5.2.6 Pulverized Coal Burner Systems Over the past 40 years, burner designs have changed in many aspects, specifically to meet the strict government standards regarding NOx emissions. Prior to more stringent emissions regulations, burner systems were designed to provide intense turbulence to achieve maximum combustion and high-energy release from reactants. Newer burner systems are designed with the knowledge that fuel-bound nitrogen can be converted to unwanted NOx formation in the presence of excess oxygen and extremely high temperatures. With tangentially fired systems, lower NOx conditions are initially experienced due to the fact that the initial incoming fuel and secondary air come parallel to one another, minimizing mixing until the fuel reaches the center fireball. Once the fuel reaches the center fireball, centrifugal forces drag the largest particulate matter to the outside of the flame. As the particles ignite and lose mass during their combustion, they move toward the center of the fireball experiencing relatively low NOx
Conventional Firing Systems
265
FIGURE 7-15 A Foster Wheeler double arch-fired boiler (Source: [21, 22]). (Photo courtesy of Foster Wheeler North American Power)
formations. Another effort that may be employed to reduce NOx conditions is to introduce the air at a later point in the furnace rather than directly at the burner mouth. When overfire air ports are used, around 20% of the necessary combustion air may be diverted to a later point in the furnace to minimize the effects of excess oxygen on the fuel.
7.5.2.7 Typical and Maximum Conditions In pulverized coal systems, the coal ignites around 730 F and starts to decompose. As the coal combusts, temperatures rapidly rise throughout the furnace, peaking around 3,000 F. Once completely broken down, the coal material that was unable to be exploited in the initial combustion process is later burned in the furnace, and the unburned material collects in the form of ash.
7.5.2.8 Fuel Preparation For a pulverized fuel firing system, it is crucial that the particles of the fuel meet a certain size requirement for maximum performance efficiency. When one is dealing with coal, fuel should be ground to a size where
266 Combustion Engineering Issues for Solid Fuel Systems MAIN FLAME SCANNER VENT VALVE
AIR/COAL INLET PIPE
BURNER VENT PIPE
RIFFLE DISTRIBUTOR
ADJUSTABLE ROD AND VANE
SIGHT PORT
DOUBLE CYCLONE BURNER
OIL BURNER (OPTIONAL)
BURNER NOZZLE FLAME SCANNER AND IGNITOR VENT AIR DAMPER VENT PIPE
ADJUSTABLE ROD AND VANE DOUBLE CYCLONE BURNER
SECONDARY AIR PORTS
CONTROL DAMPERS
ENG39
FIGURE 7-16 An arch-fired or vertically fired burner system (Source: [21]). (Photo courtesy of Foster Wheeler North American Power)
at least 70% passes through a 200-mesh screen (70 mm) and less than 1% is retained on a 52-mesh screen (300 mm). The criticality of pulverizer performance cannot be overstated (see also Chapter 6). Numerous experiments have been performed firing finer and finer materials in conventional pulverized firing systems—leading to
Conventional Firing Systems
267
experiments with micronized coal. Those experiments, while periodically successful, also demonstrated that coal can be ground “too fine” under certain circumstances.
7.5.2.9 Effect of Moisture The performance for pulverized fuel firing systems weighs heavily on the moisture that the coal experiences. Any moisture greater than 10% decreases the performance of the system. Often, drying air is used from hot gases within the furnace to preheat the fuel, especially when dealing with wet coals. Depending on the total moisture contained within the coal, coal moisture can be associated with a decrease in boiler efficiency. According to Lawn [3], there is approximately 0.07% efficiency loss per 1% moisture content.
7.5.2.10 Swirling Flow Airflow around a burner is the main way in which combustion flames are controlled in large-scale applications. When the air entering the boiler at the combustion zones is swirled, radial and axial pressure gradients may be formed that can change the pre-existing flow field. A spiral flow can be formed by bringing air into the burner at a tangential direction. Equilibrium is then formed between the centrifugal forces acting on the firing particles (from the swirl) and the forces exerted on the tube walls. A torroidal vortex forms (internal reverse flow zone, or IRZ), which causes the air to expand rapidly at the entry boundaries of the burner exit so that the flow expands rapidly. The internal reverse flow ensures that the coal is completely combusted under low-oxygenated air conditions. The reduction of oxygen at the combustion zone is accomplished by the IRZ, which pulls the secondary air back to the burner mouth. The high temperature of the recirculating air is what also ignites the coal as it comes out of the burner with the primary air. When the coal is burned under low oxygenated conditions, a lower NOx production level can be achieved. Depending on the degree of swirl, the angle at which the outgoing fuel and air is released can be affected. The swirl number can be computed by determining the ratios of the mean axial and tangential velocities with respect to the swirl generator.
7.5.2.11 Overfire Air Systems as Burner-Based Emissions Control It is known that under normal combustion conditions in the presence of excess O2 there is a greater chance for NOx formation from the fuelbound nitrogen. The problem with this fact is that without the proper amount of oxygen in the coal burning process, incomplete combustion can occur. One method of ensuring complete combustion of the fuel
268 Combustion Engineering Issues for Solid Fuel Systems
without supplying surplus oxygen that would create NOx is to introduce the oxygen later in the furnace. The design of introducing air at a later point in a furnace after the burners is known as overfire air. With overfire air systems, the surplus O2 may be introduced after peak flame temperatures to ensure complete combustion with low NOx emissions. The extension of the overfire air principle for NOx management led to the development of systems using reburn technology. In reburn technology, the main body of fuel is combusted at a stoichiometric ratio of 1.0. Additional fuel—typically natural gas but also pulverized coal—is introduced above the main fireball, reducing the stoichiometric ratio to 0.8–0.9. Then a final supply of combustion air is introduced, raising the stoichiometric ratio to 1.15–1.2. This completes the final burnout.
7.5.3 Cyclone Firing 7.5.3.1 Basic Description and Identification of Types For cyclone firing systems, basically three main setups are used; they are known as front-fired, opposed-fired, and tangentially fired systems. These three designs are fundamentally the same as the pulverized wall-fired and tangentially fired designs; however, the intrinsic difference is that the combustion region is contained within the cyclone combustor. Centrifugal burners work to create an intense vortex within the burner so that the fuel particles are suspended for maximum combustion (see Figure 7-17). For cyclone combustors, a majority of the slag is retained within the combustor where it exits through the tap hole. The cyclone firing system developed by Babcock & Wilcox is somewhat different from the figures shown previously. Crushed coal (typically 1/4" to 3/8" 0") enters the system through one of three types of burners:
Primary air and fuel
Exhaust and secondary burnout region
Secondary Tangential air inlets-velocities 60-140 m/s
Tap hole for slag removal
FIGURE 7-17 Vortex formation in a cyclone burner (Source: [3]).
Conventional Firing Systems
269
scroll, vortex, or radial. Scroll and radial feeders enter at a tangent to the burner or scroll section of the cyclone. With primary air, the coal is picked up and moved from the wear blocks in the burner section to the cyclone barrel itself. There, the swirling action forces the coal particles to the wall of the cyclone barrel. Secondary air sweeps the wall, causing intense combustion of the coal and subsequent slag formation. The gaseous combustion products exit through the re-entrant throat and enter the primary furnace. The slag is tapped through a slot in the cyclone exit; it flows down the side of the boiler and to the slag tap in the center of the primary furnace. The slag enters a tank of water where it is quenched and broken into fine pieces. The vortex feeder varies from the scroll or radial feeder in that the coal is deposited directly onto the wear blocks, without swirl being imparted. Primary air enters the burner section and moves the coal to the cyclone barrel. 7.5.3.2 Typical and Maximum Conditions There are two main ways in which the temperature controls are chosen for cyclone combustors, which are dependant on whether or not the combustion device is to be slagging or nonslagging. For slagging devices, the temperature must remain at or above 2,750 F to ensure that the liquid ash remains mobile for removal through the taps. For nonslagging devices, wall temperatures are kept constant at about 2,200 F so that the ash does not reach its fusion temperature. Temperatures are kept to a minimum for the nonslagging burners by introducing cool primary air into the combustion chamber. In the Babcock & Wilcox cyclone, temperatures inside the cyclone barrel are typically 3,300–3,550 F. 7.5.3.3 NOx Formation and Cyclones Due to the extremely high burning temperatures, there are often problems associated with NOx formation with cyclone combustors. The high heat allows for the formation of the NOx through thermal interaction with atmospheric nitrogen as well as the formation of NOx from the fuel-bound nitrogen. Typical NOx emissions for cyclone combustors range from 1,000–1,200 ppm or 1–2 lb NOx/106 Btu. These can be reduced dramatically by using separated overfire air systems. 7.5.3.4 Design and Operating Parameters Typically speaking, there has been a dramatic increase in boiler size over the past 50 years. The maximum boiler size has increased from 300 tons/h of steam to over 3,000 tons/h. An increase in boiler size has corresponded with an increase in maximum power generation. Average-sized units range in the 150–700 MWe range for electrical output; however, units with an output of over 1,000 MWe exist in the United States and Japan [3].
270 Combustion Engineering Issues for Solid Fuel Systems
The largest single unit is Paradise Fossil Plant Unit #3, a 1,156 MWe supercritical boiler fired with 23 large cyclone barrels. Steam pressures within the boiler are usually found to be between 2,350 and 2,500 psia, where final steam temperatures range from 1,000–1,050 F. Each cyclone is capable of supporting 40–50 MWe of generation. While boilers are often a common use for cyclone burners, dryers and kilns often embrace the cyclonic technology for its lack of ash in the flue gas as well as having clean air at the burner throat. For cyclones that work without slagging, there have been relatively few recorded issues with regards to maintenance and operation. For cyclone burners that do have slag, a common problem is the overheating of refractory caused by molten slag that has the ability to wear away the inner walls of the burner relatively quickly. Operationally, cyclone boilers are capable of firing a wide variety of fuels—typically high-volatile coals but also including blends with tire-derived fuel (TDF), wood waste, petroleum coke, and an array of opportunity fuels. All have been tested, and some—typically TDF and pet coke—are commonly fed when available at acceptable prices.
FIGURE 7-18 An illustration of good cyclone combustion. Note that the slag in the cyclone barrel is fluid and flowing through the tap into the furnace (and subsequently into the slag tank). Note also the clear combustion in the primary furnace (Source: [23]). (Photo courtesy of Kirby Letheby)
Conventional Firing Systems
271
FIGURE 7-19 An example of poor cyclone combustion. Note the pooling of the slag in the cyclone barrel with the possibility that cyclonic flow of gases is disrupted, and the combustion process associated with the slag (Source: [23]). (Photo courtesy of Kirby Letheby)
Despite the fuel flexibility of cyclone boilers, operations are not always forgiving. Figure 7-18 shows what good cyclone combustion should look like. Figure 7-19 shows one form of poor combustion in cyclone boilers, and Figure 7-20 shows excessive deposition on superheater tubes caused by poor cyclone operations [23].
7.6 Concluding Statements There are numerous traditional firing systems, each capable of addressing specific requirements and conditions. These requirements may be the capacity of the boiler, the type of fuel available, or the application to process industry. Each type of system has unique strengths and limitations. Choice and application of each system must be carefully considered. The application and use of each system must then be approached with care.
272 Combustion Engineering Issues for Solid Fuel Systems
FIGURE 7-20 Excessive deposition on superheater tubes caused by poor cyclone combustion (Source: [23]). (Photo courtesy of Kirby Letheby)
7.7 References 1. FAO. 2007. Power and Heat Plants. FAO Corporate Document Repository. Forestry Department. http://www.fao.org/docrep/p2396E/p2396e03.htm, accessed June 2, 2007. 2. Chain Grate Stoker. 2005. The Aarkay Group. http://www.aarkay.net/eng/ cgs.html, accessed June 3, 2007. 3. Lawn, C.J. 1987. Principles of Combustion Engineering for Boilers. Orlando FL: Academic Press. pp. 4–31, 118–119, 200–215, 251–260, 316–326, 453–504. 4. U.S. Energy Information Administration. 2006. Electric Power Annual with Data for 2005. Washington, D.C.: U.S. Department of Energy. 5. Utility Data Institute (UDI). 1996. Power Plant Equipment Directory, 2nd Ed. Washington, D.C.: McGraw-Hill.
Conventional Firing Systems
273
6. Baukal, C.E., and J. Zink. 2001. The John Zink Combustion Handbook. Danvers MA: CRC P LLC. pp. 15–28, 510–675. 7. Elliott, T.C. 1989. Standard Handbook of Powerplant Engineering. New York: McGraw-Hill. pp. 1.158–1.2069, 3.108–3.114. 8. Tillman, D. 2001. Final Report: EPRI-USDOE Cooperative Agreement: Cofiring Biomass with Coal. Contract No. DE-FC22-96PC96252. 9. Payette, K., and D. Tillman. 2004. Designing an Opportunity Fuel with Biomass and Tire-Derived Fuel for Cofiring at Willow Island Generating Station and Cofiring Sawdust with Coal at Albright Generating Station: Final Report to USDOE. Contract DE-FC26-00NT40894. 10. Stultz, S.C., and J.B. Kitto (eds). 2005. Steam, 41st ed. Barberton, OH: Babcock & Wilcox. 11. 50 Dirtiest Power Plants. 2007. Environmental Integrity Project. July 27. http://www.environmentalintegrity.org/pub385.cfm, accessed June 3, 2007. 12. Farzan, H., S. Vecci, D. MacDonald, K. McCauley, P. Pranda, R. Varagani, F. Gautier, and N. Perrin. 2007. State of the Art of Oxy-Coal Combustion Technology for CO2 Control from Coal-Fired Boilers—Are We Ready for Commercial Installation? Proc. 32nd International Technical Conference on Coal Utilization and Fuel Systems. Coal Technology Assn. Clearwater, FL. June 10–15. 13. Ochs, T., D. Oryshchyn, J. Ciferno, and C. Summers. 2007. Ranking of Enabling Technologies for Oxy-Fuel Based Carbon Capture. Proc. 32nd International Technical Conference on Coal Utilization and Fuel Systems. Coal Technology Assn. Clearwater, FL. June 10–15. 14. Combustion. Proc. 32nd International Technical Conference on Coal Utilization and Fuel Systems. Coal Technology Assn. Clearwater, FL. June 10–15. 15. Junge, D.C. 1975. Boilers Fired with Wood and Bark Residues. Res. Bulletin 17. Corvallis, OR: Forest Research Laboratory, Oregon State University. 16. Junge, D.C. 1979. Design Guideline Handbook for Industrial Spreader Stoker Boilers Fired with Wood and Bark Residue Fuels. Corvallis, OR: Oregon State University Press. 17. Beer, J.M., and N.A. Chigier. 1972. Combustion Aerodynamics. New York: John Wiley and Sons. pp. 86–144. 18. Orfanoudakis, A.G., A. Hatziapostolou, K. Krallis, E. Mastorakos, K. Sardi, D.G. Pavlou, and N. Vlachakis. 2005. Design, Evaluation Measurements and Modelling of a Small Swirl Stabilised Laboratory Burner. IFRF Combustion Journal (December): 1–29. 19. Grusha, J., S. Laux, and S. Slingerland. 2002. Operational and Performance Update on New Ultra Low NOx Combustion Systems with Fuel/Air Monitoring and Control Technology. Proc. 26th International Technical Conference on Coal Utilization and Fuel Systems. Coal Technology Assn. Clearwater, FL. March 4–7. 20. Battista, J., E. Hughes, and D. Tillman. 2000. Biomass Cofiring at Seward Station. Biomass and Bioenergy. 19(6): 419–428.
274 Combustion Engineering Issues for Solid Fuel Systems 21. Eberle, John S., J. Antonio Garcia-Mallol, and Robert N. Simmerman. 2002. Advanced FW Arch Firing: NOx Reduction in Central Power Station. Proc. Pittsburgh Coal Conference, Pittsburgh, PA. September. 22. Garcia-Mallol, J.A., T. Steitz, C.Y. Chu, and P. Jiang. 2005. Ultra-Low NOx Advanced FW Arch Firing: Central Power Station Applications. 2nd U.S. China NOx and SO2 Control Workshop. Dalian, Liaoning, Peoples Republic of China. August. 23. Letheby, Kirby. 2002. Utility Perspectives on Opportunity Fuels. Proc. 26th International Technical Conference on Coal Utilization and Fuel Systems. Coal Technology Assn. Clearwater, FL. March 4–7.
CHAPTER
8
Fluidized-Bed Firing Systems Bruce G. Miller
Associate Director, Energy Institute The Pennsylvania State University and
Sharon Falcone Miller Research Associate, Energy Institute The Pennsylvania State University
8.1 Introduction Fluidized-bed combustion (FBC) is a leading technology for the combustion of a range of fuels, fossil and others, because of several inherent advantages it has over conventional combustion systems including fuel flexibility, low NOx emissions, in situ control of SO2 emissions, excellent heat transfer, high combustion efficiency, and good system availability. The development of the fluidized-bed concept started in 1922 with the Winkler patent for gasification of lignite [1]. The first major implementation of the concept was around 1940 in the chemical industry to promote catalytic reactions. In the 1950s, the pioneering work on coal-fired fluidized-bed combustion started in Great Britain, specifically by the National Coal Board and the Central Electricity Generating Board [2, 3]. The U.S. Department of Interior’s Office of Coal Research, one of the predecessors of the current Department of Energy (DOE), began studying the FBC concept in the early 1960s (and it still continues sponsoring research into advanced FBC systems as discussed later in the chapter) because it recognized that the fluidized-bed boiler represented a potentially lower cost, more effective, and cleaner method to burn coal [4]. In addition to Great Britain and the United States, 275
276 Combustion Engineering Issues for Solid Fuel Systems
other countries like Finland, Germany, and China started programs to develop fluidized-bed combustion, as they wanted to establish a technology that was able to utilize low-grade fuels with low emissions [1]. Around 1990, atmospheric FBC crossed the commercial threshold with every major U.S. boiler manufacturer and many international boiler manufacturers currently offering fluidized-bed boilers as a standard package. In this chapter, the two main FBC groups, atmospheric FBC and pressurized FBC, and the two minor subgroups, bubbling and circulating fluidized-bed, will be discussed. A brief technology overview will be presented focusing on fluidization characteristics, heat transfer, fuel flexibility, emissions formation and control, and combustion efficiency. Ash chemistry and agglomeration issues will be discussed, as will the role of FBC boilers in clean coal technology development in the United States and worldwide. The chapter will conclude with a discussion of unique opportunities for FBC technology, specifically a potential role in providing security to the food industry.
8.2 Fluidized-Bed Combustion Systems In a typical FBC system, the fuel (or fuels), an inert material such as sand or ash (referred to as bed material), and limestone (when added for sulfur control) are kept suspended through the action of combustion air distributed below the combustor floor. The function of the inert material is to disperse the incoming fuel particles throughout the bed, heat the fuel particles quickly to the ignition temperature, act as a thermal flywheel for the combustion process by storing a large amount of thermal energy, and provide sufficient residence time for combustion. The FBC concept is attractive because it increases turbulence, contains a substantial amount of thermal energy, and permits lower combustion temperatures. Turbulence is promoted by fluidization, making the entire mass of solids behave much as a liquid. Improved mixing permits the generation of heat at a substantially lower and more uniformly distributed temperature than occurs in conventional systems such as stoker-fired boilers or pulverized coal-fired boilers. The turbulent nature of the bed increases the rate of heat transfer on tubes immersed in the bed. The high thermal energy in the bed allows a variety of fuels to be utilized, including low-quality fuels with high mineral matter or moisture contents, as well as a multiple fuels in the same unit. The bed temperature in an FBC boiler is typically 1,450–1,650 F (780–900 C). This temperature is below that at which most inorganic components melt and form slag and agglomerates and is well below that at which thermally induced NOx production occurs. At this temperature range, the reactions of SO2 with a suitable sorbent, commonly limestone, are thermodynamically and kinetically balanced [5]. Sulfur capture at a given sorbent rate decreases significantly outside the 1,450–1,650 F range. Nitrogen oxides can be reduced to nitrogen by use of ammonia-based compounds at typical FBC temperatures.
Fluidized-Bed Firing Systems
277
Although FBC offers a number of significant advantages, there are also some disadvantages with the technology. This includes the necessity for handling large amounts of solids—fuel, limestone, and bed material, higher carbon-in-ash levels than those from pulverized coal-fired systems, and increased N2O formation due to the lower combustion temperatures [6]. The principle of FBC systems is illustrated in Figures 8-1 [7, 8] and 8-2 [9]. The fundamental distinguishing feature of all FBC units is the velocity of air through the unit, as illustrated in Figure 8-1. Bubbling beds have lower fluidization velocities, and the concept is to prevent solids from elutriating (i.e., removed by the gas stream) from the bed into the convective pass. Circulating fluidized-bed units apply even higher velocities to promote solids elutriation. These will be discussed further in the following sections, where the four different variants of FBC are presented. FBC can be divided into two main groups based on the fluidizing gas velocity: bubbling fluidized-bed combustion and circulating fluidized-bed combustion. In addition, FBCs that are operated under pressure are classified as pressurized bubbling fluidized-bed combustors and pressurized circulating fluidized-bed combustors. Each of the combustion systems is briefly described in the following sections.
ENTRAINED-FLOW REACTORS SEMI-ENTRAINED FULLY ENTRAINED
Fuel
Fuel
Fuel Solids
Air
FIXED BED 1-4 ft/sec
BUBBLING BED 4-12 ft/sec
Velocity
Air
CIRCULATING BED 12-30 ft/sec
Air TRANSPORT REACTOR above 30 ft/sec
y
locit
s ve
n ga
Mea
Air
Slip velocity
Increasing solids throughput
Mean solid velocity
Increasing expansion
FIGURE 8-1 Fluidizing velocity of air for various bed systems (Adapted from [7] and [8]).
278 Combustion Engineering Issues for Solid Fuel Systems Suspension Firing
Stoker Firing (Fixed Bed)
(Entrained Bed)
CFB
GAS
GAS
AIR
AIR FUEL
AIR ASH
BFB GAS
GAS
FUEL
Fluidized-Bed Firing
FUEL & SORBENT
FUEL & SORBENT
ASH
AIR ASH
AIR ASH
Velocity
1-4 ft/sec
15-30 ft/sec
4-12 ft/sec
12-30 ft/sec
Fuel Feed Size
<2 inches
<300 µm
<0.25 inches
<0.25 inches
Furnace Temperature
2000-2500⬚F
2200-2800⬚F
1500-1700⬚F
1500-1700⬚F
FIGURE 8-2 Comparison of combustion processes (Modified from [9] and [10]).
8.2.1 Bubbling Fluidized-Bed Combustion (BFBC) A fluidized bed consists of a bed of particles above a grid so that a fluid can be passed upward through the bed. The grid supports the bed, provides a uniform flow of air across the whole base of the bed, and prevents the bed material from dropping into the plenum chamber. The bed material commonly includes the fuel, often a sorbent for sulfur control, fuel ash derived from the mineral matter in the fuel, and sometimes an inert additive such as sand if there is insufficient bed material to maintain the design mass of the bed. As the gas velocity (which is combustion air in a fluidizedbed combustor but can also consist of recirculated flue gas or oxygen in the case of enhanced oxygen combustion) increases, the bed undergoes three stages of fluidization—fixed bed, bubbling fluidized-bed, and circulating fluidized-bed. As the gas velocity is increased through a fixed bed, there is a point at which the gas-particle drag force overcomes the force of gravity on the sand. At this point, the inter-particle distances increase, the bed expands, and the particles appear to be suspended in the gas [6]. This is the onset of fluidization, and the gas velocity at this point is referred to as the minimum fluidizing velocity. In practice, BFBC boilers are operated at gas velocities that are several times greater than the minimum fluidizing velocity. Bubbles passing through the bed occupy 20–50% of the bed volume and give intensive agitation and mixing of the bed particles [6]. The particles still remain in relatively close contact with minimal bed loss and a welldefined upper surface.
Fluidized-Bed Firing Systems
279
LIMESTONE
FUEL
FLUE GAS
SOLIDS
PARTICULATE RECYCLE STANDPIPE AIR INLET
PLENUM CHAMBER DISTRIBUTOR PLATE
WASTE BED DRAIN
FIGURE 8-3 Generalized schematic diagram of a bubbling fluidized-bed boiler (Modified from [1] and [11]).
A generalized schematic diagram of a BFBC boiler is shown in Figure 8-3 [11]. BFBC boilers typically operate at temperatures of 1,450– 1,650 F (790–900 C), with a mean bed particle size between 1,000 and 1,200 mm, and fluidizing velocities between the minimum fluidizing velocity (which is a function of particle size) and the entraining velocity of the fluidized solid particles (i.e., 4 to 12 ft/sec). Under these conditions, a well-defined bed surface separates the high-solids loaded bed and the lowsolids loaded freeboard regions. Most bubbling-bed units, however, utilize reinjection of the solids escaping the bed to increase combustion efficiency. Flue gases leave the bed and pass through the freeboard, where coarse particles entrained in the flue gas fall back to the bed by gravity. After exiting the combustor, the flue gases pass into a convective section where heat is further recovered. The flue gases then pass through a particulate control device or devices, which is typically a cyclone followed by either a baghouse or electrostatic precipitator. (A discussion of particulate removal devices can be found in Chapter 9 and elsewhere [10].) The cleaned flue gases are then discharged into the atmosphere through a stack. BFBC technology is mature for cogeneration and industrial-sized applications. While units as large as 120–140 MWe have been installed, the mean capacity from 1976 to present day is approximately 30 MWe [1]. The trend for scaling the BFBC boiler toward utility size stopped around
280 Combustion Engineering Issues for Solid Fuel Systems
1990 due to increased competition from CFBC boilers, which at that time surpassed the scale of BFB boilers. Future technology development is likely to be limited to ensuring fuel flexibility on existing designs for the increasing use of biomass and/or waste as feedstock with, or instead of, coal [1]. Further efforts are set to improve plant construction and operation, increase efficiency, and reduce capital and operating costs.
8.2.2 Circulating Fluidized-Bed Combustion (CFBC) When the gas velocity through the bed is increased to a point exceeding the particle free-fall velocity, the bed particles become entrained and are removed from the combustor. At this point, the gas velocity is referred to as the particle terminal velocity [6]. The entrained solids are separated from the gas stream using a cyclone and recycled to the bed. A generalized schematic diagram of a CFBC boiler is shown in Figure 8-4.
GAS CONVECTION PASS
FUEL
HOT CYCLONE
LIMESTONE
COMBUSTOR
SOLIDS
FLUE GAS RECYCLE
SECONDARY AIR
HEAT EXCHANGER (OPTIONAL)
PRIMARY AIR
WASTE BED DRAIN
FIGURE 8-4 Generalized diagram of a circulating fluidized-bed boiler (Modified from [1] and [11]).
Fluidized-Bed Firing Systems
281
Similarly as in BFBC boilers, combustion air is continuously blown through the bed from the bottom of the combustor, and combustion occurs at temperatures around 1,450–1,650 F (780–900 C). However, CFBC boilers operate with a mean bed particle size between 100 and 300 mm and fluidizing velocities up to about 30 ft/sec. Because CFBC boilers promote elutriation and the solids are entrained at a high rate by the gas, bed inventory can be maintained only by recirculation of solids separated by the off-gas by a high efficiency cyclone. This allows for effective combustion of the carbon and maximum sulfur capture of the sorbent (when added). Notwithstanding the high gas velocity, the mean solids velocity in the combustor is lowered due to the aggregate behavior of the solids. Clusters of solids are continuously formed, flow downward against the gas stream, are dispersed, are re-entrained, and form clusters again. The solids thus flow upward in the combustor at a much lower mean velocity than the gas. The slip velocity between gas and solids is very high with corresponding high heat and mass transfer. After leaving the cyclone, the flue gases enter the convective pass, where useful heat is further removed. The flue gases then pass through a baghouse and discharge through a stack. CFBC technology has a number of advantages over BFBC technology, including [6] the following: Ability to design and build larger, higher capacity units; Improved combustion efficiency and sulfur retention due to the use
of finer particles, turbulent gas-particle mixing, and a high recycle rate; Smaller bed area due to the use of high fluidizing velocities; Reduced number of fuel feed points due to the smaller combustor size and turbulent mixing; Reduced erosion and corrosion of heat transfer tubes because tubes immersed in the fluidized-bed cooler are subjected to significantly lower gas and particle velocities than in a BFBC and the cooler is subjected to oxidizing conditions, whereas reducing conditions occur near the fuel feed points in the BFBC; and Increased convective heat transfer coefficients.
While CFBC technology has several advantages over BFBC technology, there are also some areas of concern or disadvantages: Increased height of the CFBC boiler compared to the BFBC boiler; Additional height and size of the fluidized-bed ash cooler; Greater pressure drop across the CFBC boiler resulting in increased
fan power requirements; Need for high cyclone efficiencies for bed material recovery for
solids recycle; and Higher erosion in the combustor, cyclone, and associated ducting
due to the high gas velocities.
282 Combustion Engineering Issues for Solid Fuel Systems
Similar to BFBC technology, CFBC technology is commercially available. CFB boilers have evolved into the utility boiler size range with a number of units as large as 300þ MWe in operation, and are poised to enter into the realm of larger once-through, supercritical units [12]. However, several opportunities remain to support further development of supercritical CFB boilers and the pressurized version. Further development in efficiency improvement, fuel flexibility, effective scale-up, and reductions in capital cost are underway [1].
8.2.3 Pressurized Fluidized-Bed Combustion (PFBC) As in atmospheric designs, two types of pressurized fluid beds have been applied to power generation: bubbling and circulating. Bubbling bed designs (PBFBC) have been developed and a few are in operation, whereas circulating designs (PCFBC) are mainly in the pilot and conceptual design stages. An example of a PBFBC boiler is shown in Figure 8-5; this is the Tidd PFBC demonstration, the United States’ first PFBC combined-cycle demonstration [13]. Combustion occurs within a large pressure vessel at pressures of 10–15 atmospheres and temperatures similar to atmospheric FBC boilers. Because the system is operated under pressure, provisions must be made to feed the fuel and sorbent to the combustor and remove the ash from the combustor across pressure boundaries. Also, primary particulate removal is performed under pressure using high-efficiency cyclones or high-temperature
COMBUSTOR VESSEL
TRANSFORMER
HOT GAS CLEANING UNITS BOILER TRANSFORMER
GENERATOR 54MW
CONDENSER
STEAM TURBINE
AIR
FLUIDIZED BED COAL DOLOMITE
GENERATOR 16MW GAS TURBINE
STACK
CYCLONE ASH BED ASH
ECONOMIZER PRECIPITATOR
DEAERATOR L.P. HEATERS CONDENSATE PUMPS
H.P. HEATER TANK PUMPS
FEED PUMPS
FIGURE 8-5 Tidd PFBC combined-cycle process (Source: [13]).
FLY ASH
Fluidized-Bed Firing Systems
283
filters. Advantages of PFBC include improved cycle efficiency, reduced emissions, improved combustion, reduced boiler size, and reduced tube erosion [14]. Elevated pressures, and in some cases temperatures, produce a highpressure gas stream that can drive a gas turbine, and steam generated from the heat in the fluidized-bed is sent to a steam turbine, creating a highly efficient combined cycle steam. About 80% of the electricity is generated in a conventional steam turbine-generator set, and the balance of the electricity is generated in a gas turbine [1]. Cycle efficiencies of 40% are currently achieved in PBFBC boilers but efficiencies greater than 50% are targeted in second-generation PCFBC boilers. The increased pressure of PFBC systems and corresponding air/gas density allows much lower superficial fluidizing air velocities. For a PBFBC boiler, this is approximately 3 ft/sec compared to 10 ft/sec in a BFBC boiler. This reduces erosion of submersed boiler tubes and also permits the use of much deeper beds. The combined effect of lower velocity and deeper bed results in an increased in-bed gas residence time and smoother fluidization, resulting in better gas-solids contact and ultimately better SO2 capture and improved combustion efficiency. Another advantage of PFBC is reduced boiler size. The high gas density results in a smaller required bed plan area. The lower velocity reduces the total height required for the bed and freeboard [15].
8.3 Heat Transfer In fluidized-beds, good mixing is usually achieved, which gives good heat distribution and a uniform temperature distribution. This, in turn, results in effective gas-solids contact giving a high rate of heat transfer from burning fuel to the waterwalls or immersed tubes. In conventional furnaces (stokerfired or pulverized coal-fired), the solids loading in the gas stream are low (i.e., approximately 10 lb per 1,000 lb of gas), and heat transfer from the gas to the waterwalls is primarily by radiation with a lesser contribution from convective heat transfer [14]. In contrast, in a CFB boiler, the gas leaving the furnace contains a high solids concentration, which can be greater than 5000 lb per 1000 lb of gas, and hence, convective heat transfer dominates over radiative heat transfer. For equal temperatures, the heat transfer coefficients in an FBC boiler are considerably higher than those in a conventional furnace [14]. However, because the temperatures are lower in an FBC boiler, the overall heat fluxes between the two systems are similar. What differs between the two systems are the types of heat transfer surfaces employed between the combustion systems. In a conventional system, heat is transferred to the waterwalls (i.e., containment structure) and tubes in the convective pass. In an FBC boiler, three zones of heat transfer must be considered: in-bed, splash zone (interface between the bed and freeboard), and above-bed. In addition, heat transfer areas outside the furnace must also be considered.
284 Combustion Engineering Issues for Solid Fuel Systems
In a BFB boiler, heat transfer surfaces include tube banks in the dense bed, waterwalls in the dense bed, and tubes in the convective pass. CFB boilers do not incorporate tube bank surface in the bed and rely on heat absorption of the containment walls, internal partitions such as division walls and wingwalls, external ash coolers, and tubes in the convective pass.
8.4 Combustion Efficiency Combustion efficiency, defined as the ratio of heat released by the fuel to the heat input by the fuel, is generally high in FBC systems. The combustion efficiency is typically higher than stoker-fired systems and is comparable to pulverized coal-fired systems. It is generally higher in a CFB boiler than in a BFB boiler because of the use of finer particles, more turbulent environment, and a high solids recycle rate [14]. Similarly, PCFBC boilers achieve higher efficiencies due to smaller and more frequent bubbles, which result in better gas-solid contact. Combustion efficiency is affected by fuel type, bed temperature, gas velocity, and excess air levels. Combustion efficiency increases with fuel volatile matter content and bed temperature. Combustion efficiency decreases with increasing superficial gas velocity. Combustion efficiency initially increases with increasing excess air level and then decreases. This is believed to be due to an increase in CO and hydrocarbon emissions as the excess air level increases to higher levels [14].
8.5 Fuel Flexibility The flexibility in fuel utilization, mainly the ability to use low-quality fuels, makes FBC boilers an attractive technology. These fuels can be fired solely, in combination with other low-grade fuels, or cofired with coal. The lower combustion temperatures permit burning high-fouling and -slagging fuels at temperatures below their ash fusion temperature. This greatly reduces operating problems associated with these fuels; however, care is still required because these fuels are those containing significant concentrations of alkali and alkaline earth elements, as discussed in more detail in Section 8.7. In addition, fuels with low heating values, due to high moisture and/or ash contents, or low volatile matter contents can be successfully burned using an FBC boiler because of the large mass of hot bed material and long residence time that the fuel spends in the bed. Examples of these include fuels such as brown coal, peat, and sludge with moisture contents up to 60%; waste coals with ash contents up to 76% and higher heating values as low as 2,600 Btu/lb; and petroleum coke with volatile matter content less than 10%. The fact that many of these low-grade fuels are difficult to reduce to fine size, due to high ash contents or fibrous structures (as in the case of biomass), makes them candidates for FBC technology, since the fuel does not need to be pulverized but can be processed to sizes of 0.25 inches 0.
Fluidized-Bed Firing Systems
285
Although a main advantage of FBC boilers is that they can be designed to burn a wide variety of low-grade fuels, once an FBC boiler has been designed, there are limitations in deviating from the design values so as not to exceed design limits [14]. As of 2005, biomass is the main fuel type used in BFB boilers with more than 2,000 MWe of installed capacity [1]. This is followed by peat, various ranks of coal (bituminous coal, subbituminous coal, and lignite), coal wastes, and other wastes, each contributing <500 MWe in installed capacity. In contrast, the main fuel type used in CFBC boilers is bituminous coal (10,000 MWe) followed by lignite (4000 MWe), with smaller contributions from coal wastes, petroleum coke, biomass, other ranks of coal (anthracite, subbituminous coal, brown coal), peat, wastes, and RDF pellets, where each category contributes 2,000 MWe or less [1]. Figure 8-6 illustrates the wide range of fuels that have been used/ tested in FBCs; however, the range of fuels is endless. Candidate fuels include, not inclusive, the various ranks of coal (anthracite, bituminous coal, subbituminous coal, lignite, brown coal), waste coal from coal-cleaning operations, petroleum coke, oil shale, refinery bottoms, peat, woody biomass, herbaceous biomass, manure and litter, animal-tissue biomass, tires, paper mill sludge, sewage sludge, refuse-derived fuel, pellet-derived fuel, plastics, industrial wastes, and more. Table 8-1 lists various characteristics of a diverse range of fuels that are used in FBC boilers or have been tested in pilot-scale units.
Higher Heating Value (Btu/Ib)
15,000
CONSUMER PDF MIXED PLASTICS COLORED CONSUMER PDF OR PRINTED WOOD AND MIXED PLASTICS PLASTICS RF PELLETS
10,000 A N I M A L
5,000
T I S S U E
PVC
RDF
POLYOLEFIN PLASTICS BITUMINOUS COALS
ANTHRACITE
ANIMAL PRODUCTS
CHIPBOARD DEMOLITION WOOD
PETROLEUM COKE
COLORED OR PRINTED PLASTICS, CLEAN
PLYWOOD AGRICULTURAL RESIDUE
PDF COMMERCIAL
PDF INDUSTRIAL
WASTE COALS BROWN COALS, LIGNITE
WOOD BIOMASS
FIBER RESIDUE
PEAT BARK
MSW
CONSUMER PDF PAPER AND WOOD
2,000 MULTIPLE CHALLENGES
SOME CHALLENGES
NO CHALLENGE
STANDARD DESIGN
FIGURE 8-6 Illustration of applicable fuels for FBC technology (Source: Modified from [16]).
286 TABLE 8-1 Examples of the Range of Characteristics of Feedstocks Utilized in FBCs or Pilot-Scale Units Bituminous Coala Prox. Anal. (%, as-rec.) Moisture 2.0 Volatile Matter 31.1 Ash 61.7 Fixed Carbon 5.2 Ult. Anal. (%, as-rec.) Carbon 85.5 Hydrogen 5.2 Nitrogen 1.0 Sulfur 0.7 Oxygen 0.4 Ash 5.2 Chlorine (ppm) 1,875 Heating Value 13,650 (Btu/lb as-rec.) Density (g/cc) 0.70 Inorganic Elements (% ash basis) CaO 1.84 Na2O 0.37 K2O 2.30 Cr (ppm) 200 Inorganic Elements (ppb fuel basis) Hg 50
Waste Coalb,c
Pine Shavingsd,e
Red Oak Shavingsd,e
Plywoodf
Panelingf
Sewage Sludgeb,c
Switchgrassa
Reed Canary Grassd,e
21.3 24.2 31.5 23.0
45.0 46.6 0.1 8.8
28.8 56.6 1.1 13.5
17.5 67.3 0.7 14.5
19.0 64.7 0.9 15.4
86.8 6.4 0.6 6.2
9.6 76.2 11.2 3.1
65.2 26.5 1.4 6.9
54.5 4.9 1.0 3.5 6.8 29.3 600 7,434
27.0 3.5 0.1 <0.1 69.2 0.1 – 4,685
36.7 4.1 0.4 <0.1 57.6 1.1 – 5,745
43.0 5.2 0.2 <0.1 50.8 0.7 – 6,980
40.9 4.7 2.4 <0.1 51.0 0.9 – 6,500
29.2 4.6 3.3 1.4 14.2 47.3 1,250 736
49.6 6.5 0.3 0.1 30.8 3.1 1,300 7,322
15.9 2.1 0.3 0.1 80.2 1.4 – 2,520
–
0.10
0.10
–
–
–
0.72
0.05
16.1 0.65 1.60 28
8.75 1.38 4.94 –
45.7 1.39 6.10 –
– – – –
– – – –
13.0 0.95 1.80 170
10.7 0.62 2.40 50
9.57 2.34 18.1 –
90
–
–
–
–
130
50
–
Hayd,e Prox. Anal. (%, as-rec.) Moisture 19.5 Volatile Matter 62.5 Ash 13.7 Fixed Carbon 4.3 Ult. Anal. (%, as-rec.) Carbon 37.4 Hydrogen 4.6 Nitrogen 1.4 Sulfur 0.2 Oxygen 52.1 Ash 4.3 Chlorine (ppm) – Heating Value (Btu/lb as-rec.) 6,487 Density (g/cc) 0.04 Inorganic Elements (% ash basis) CaO 12.9 Na2O 0.6 K2O 40.5 Cr (ppm) – Inorganic Elements (ppb fuel basis) Hg – a
b
c
d
e
f
g
References: [17], [18], [19], [20], [21], [22], [23]
Rice Husksf
Poultry Litterd,e
Sheep Manured,e
Petroleum Cokef
7.0 59.1 14.7 19.2
20.0 44.2 13.6 22.2
47.8 34.0 10.9 7.3
7.6 4.5 0.7 80.2
35.6 4.1 0.7 0.1 44.8 14.7 – 5,950 –
30.5 4.5 2.8 0.5 48.1 13.6 – 5,120 –
21.2 2.7 1.1 0.3 63.8 10.9 – 3,600 0.37
– – – –
12.7 3.6 4.0 –
–
–
Tire-Derived Fuelf
Meat & Bone Meala
Cow Carcassesa,g
1.2 63.5 6.2 29.1
3.0 57.5 5.9 33.6
31.6 55.8 11.1 1.5
81.1 3.6 2.6 4.4 0.0 0.7 750 13,211 –
81.7 7.2 0.6 1.6 2.7 6.2 – 15,500 –
35.3 5.2 7.4 0.3 15.2 33.6 1,100 6,412 0.72
28.6 7.7 0.8 <0.1 51.7 11.1 1,200 3,850 0.74
4.64 12.8 23.4 –
2.2 1.8 0.3 –
3.6 0.8 0.7 –
49.1 2.16 0.37 <10
– – – –
–
–
–
<20
–
287
288 Combustion Engineering Issues for Solid Fuel Systems D
1.06 D
PLAN AREA PLAN AREA W 1.08W = 1.00 = 1.15
1.08 D PLAN AREA 1.16W = 1.25
1.24 D PLAN AREA 1.26W = 1.56
1.52h h
1.15h
1.26 D PLAN AREA 1.29W = 1.63
2.1h
1.17h
H
1.05 H
Medium Volatile Bituminous
High Volatile Bituminous or Subbituminous
1.07 . H
Low Sodium Lignite
1.30 H
Medium Sodium Lignite
1.45 H
High Sodium Lignite
FIGURE 8-7 Impact of coal type on pulverized coal-fired furnace design (Source: [24]).
The fuel characteristics have a minor impact on a CFB utility boiler design compared to that of a pulverized coal-fired boiler. Parameters that must be considered when arriving at a final pulverized coal-fired design include the heat release rate, fuel properties (e.g., ash fusion temperatures, volatile matter content, ash content), percentage of excess air, production of emissions, boiler efficiency, and steam temperature [8], with the most important item to consider being the fuel burned. Furnaces for burning coal are more liberally sized than those for gas or fuel oil firing. This is necessary to complete combustion within the furnace and to prevent formation of fouling or slagging deposits. This is depicted in Figure 8-7, where the impact of coal type on furnace size is shown [24]. By contrast, Figure 8-8 shows the impact of coal type on a CFBC design, which is small [24].
8.6 Pollutant Formation and Control Fluidized-bed coal combustors have been called the “commercial success story of the last decade in the power generation business” and are perhaps the most significant advance in coal-fired boiler technology in half a century. Originally, development of the technology was focused on manufacturing a compact, package boiler that could be pre-assembled at the factory and shipped to a plant site, thereby providing a lower cost alternative to onsite assembly of conventional boilers. By the mid-1960s, however, it became apparent that the fluidized-bed boiler not only
289
Fluidized-Bed Firing Systems D
1.08 D
1.06 D
PLAN AREA PLAN AREA 1.05W W = 1.00 = 1.05
H
Medium Volatile Bituminous
1.04 H
High Volatile Bituminous or Subbituminous
PLAN AREA 1.12W = 1.12
1.06 H
Low Sodium Lignite
1.24 D
1.26 D
PLAN AREA PLAN AREA 1.19W 1.20W = 1.19 = 1.20
1.08 H
Medium Sodium Lignite
1.08 H
High Sodium Lignite
FIGURE 8-8 Impact of coal type on circulating fluidized-bed furnace design (Source: [24]).
represented a potentially lower cost, more efficient manner to burn coal, but also generated less emissions than conventional boilers. Because the technology can control sulfur dioxide and nitrogen oxides at lower cost than conventional boilers, FBC technology has been developed from the “package boiler” concept to the utility boiler concept. This section discusses pollutant formation and control with an emphasis on sulfur and nitrogen oxides.
8.6.1 Sulfur Dioxide Sulfur in the fuel is oxidized to SO2 during the combustion process. In an FBC boiler, the SO2 is captured in situ by adding a sorbent material, which most commonly is limestone but sometimes can be dolomite. Sulfur retention can be greater than 95% in an FBC boiler, but sorbent utilization levels are relatively low (e.g., typically only about 40% of the calcium is utilized in the capture of sulfur). A discussion of the effect of operating parameters and sorbent characteristics on sulfur capture will be discussed in the following sections. The vast majority of the SO2 is removed by the in-bed reactions discussed in subsequent sections; however, in recent years polishing scrubbers have also been employed to meet increasingly stringent SO2 regulations. 8.6.1.1 Transformation of Sorbents in the FBC Process In an FBC system, limestone and dolomite will undergo thermal decomposition, a process commonly known as calcination. The decomposition of limestone proceeds according to the following equation: CaCO3 ! CaO þ CO2
[8-1]
290 Combustion Engineering Issues for Solid Fuel Systems
Calcination of limestone is an endothermic reaction, which occurs when limestone is heated above 1,400 F (760 C). Calcination is necessary before the limestone can absorb and react with gaseous sulfur dioxide. Calcined limestone is porous due to the loss of carbon dioxide. Capture of the gaseous sulfur dioxide is accomplished via the following equation to produce a solid product, calcium sulfate: CaO þ SO2 þ 1=2 O2 ! CaSO4
[8-2]
The reaction of porous calcium oxide with sulfur dioxide involves a continuous variation in the physical structure of the reacting solid with conversion. One of the major factors responsible for these changes is the formation of calcium sulfate, which has a higher molar volume than calcium oxide (52 cm3/mole for CaSO4 compared to 17 cm3/mole for CaO). Because of this expansion in the solid volume from the reactant to the product, the pore network within the reactant will be progressively blocked as conversion increases. As a result, the reaction rate will decay rapidly as soon as a shell of the product layer is formed on the outside of the reacting solid. For pure CaO prepared by the calcination of reagent grade CaCO3, the theoretical maximum conversion of CaCO3 to CaSO4 has been calculated to be 57%. At this conversion, the void spaces within the solid are completely filled by the CaSO4 formed. In practice, the actual conversion obtained using natural limestones is much lower. Calcium utilization as low as 15–20% has been reported in some cases [25] and is typically 30–40%. Low utilization efficiency can be explained by early pore blockage making it more difficult for SO2 to reach the reactive CaO in the particle interior. Sulfur dioxide can usually penetrate only a small distance (50–100 mm) into a particle until pore plugging inhibits further sulfation [26]. Therefore, fine particles can be sulfated more completely than coarse particles, but they also tend to be elutriated from the system before they have had time to become fully sulfated. Physically, the limestone undergoes structural changes such as porosity, pore volume, and surface area development; changes in pore size distribution; and sintering. The particles also undergo degradation, commonly referred to as attrition, with the resultant small particles removed from the system with the offgas. The effect of each of these physical properties on sulfur capture will be discussed later. The decomposition of dolomite, MgCa(CO3)2, which is a double carbonate of magnesium and calcium, is similar in many respects to the decomposition of limestone. The reaction proceeds either in a single step or in stages, depending on the operating conditions and the chemical composition of the sample. For a single-step decomposition, the overall process can be described by this equation: MgCaðCO3 Þ2 ! MgO þ CaO þ CO2
[8-3]
Fluidized-Bed Firing Systems
291
When decomposition occurs in stages, MgCO3 will decompose first. The following reactions will take place in sequence: MgCaðCO3 Þ2 ! MgO þ CaO3 þ CO2
[8-4]
CaCO3 ! CaO þ CO2
[8-1]
MgO will not react with sulfur dioxide in the reaction gas at temperatures above 1,400 F; therefore, the sulfation reaction of dolomite is basically the reaction of sulfur dioxide with calcium oxide (Equation [8-2]). In PFBC systems, however, the partial pressure of CO2 (1,400 F; 760 C) is so high that calcination of limestone does not proceed because of thermodynamic restrictions. For example, at 1,560 F (850 C), calcium carbonate does not calcine if the CO2 partial pressure exceeds 0.5 atmosphere. In these conditions, the sulfation reaction for limestone is as follows: CaCO3 þ SO2 þ 1=2 O2 ! CaSO4 þ CO2
[8-5]
Increasing the pressure from 1 to 5 atmospheres significantly increases the sulfation rate and calcium utilization [27, 28]. Under pressure, dolomite may partially calcine, and sulfation is represented by the following reaction: CaCO3
MgCO3 þ SO2 þ 1=2 O2 ! CaSO4 MgO þ 2CO2
[8-6]
8.6.1.2 Bed Temperature The effect of bed temperature on sulfur capture is well known. As previously discussed, the peak sulfur retention in an atmospheric FBC boiler is between 1,450 F and 1,650 F (790 C and 900 C), which coincides with increasing extent of calcinations. Plants with insufficient heat transfer area have exhibited high bed temperature and poor limestone utilization. In addition to the level of the bed temperature, the uniformity of the bed temperature is important because hot and cold locations in the bed must be avoided. An optimum temperature exists which is sensitive to sorbent type, particle size, calcination conditions, and pressure. This temperature effect is due to an irreversible effect, the generation and loss of pores by sintering [29, 30], and decomposition of CaSO4 formed during oxidizing conditions [31]. For PFBC boilers, however, there is no pronounced maximum for sulfur retention as a function of temperature. Within the normal operation range, sulfur capture is found to increase slightly with temperature [6]. When dolomite is used as a sorbent, the increase in sulfur retention with increase in bed temperature can continue up to as high as 1,750 F (950 C) [6].
292 Combustion Engineering Issues for Solid Fuel Systems
8.6.1.3 Particle Residence Time The residence time of the sorbent particles in an FBC boiler is a function of the particle size (and the fluidizing gas velocity). The particle size distribution is a continuum; however, various particle sizes are found in the system. Sorbent particle size (and hence residence time) is important because the rate of sulfation is proportional to the particle surface area, which is inversely proportional to the particle diameter. In other words, the smaller the particle, the faster the rate of sulfation. Unfortunately, the smaller particles, which may have a fast sulfation rate, have the shortest residence time in the combustor (2 seconds, similar to that of the gas residence time). The larger particles may have slower rates of sulfur capture, but they are contained in the system for a significantly longer period of time. 8.6.1.4 Bed Quality Bed quality, which includes limestone distribution, mixing, and fluidization, is important in capturing sulfur in the bed. Obviously, the better the contact between the limestone, sulfur dioxide, and oxygen, the greater the quantity of sulfur that can be captured. Also, increasing the fluidizing gas velocity decreases the residence time of the gas in the bed. This, in turn, reduces the effective contact between SO2 and the sorbent and, consequently, reduces sulfur capture. 8.6.1.5 Gaseous Environment Another important parameter is the gaseous atmosphere in the combustor. Oxygen is necessary for the sulfation reaction. FBC boilers normally operate with about 4% vol O2 in the bed, which is far in excess of the SO2 concentration. Under this condition, it is commonly assumed that the sulfation reaction has a zero dependency on O2 concentration. Oxygen can indirectly reduce the rate of CaO sintering, however. Sintering is a complex process which results in a reduction in the total surface area and hence a change in the pore size. Carbon dioxide enhances sintering, which can be detrimental to sulfur capture by causing CaO particle shrinkage and densification (which occurs during the later stages of sintering). Excess O2 will reduce the CO2 partial pressure and hence impede sintering. The initial rate of sulfation is approximately proportional to the SO2 concentration, but the reaction is terminated by pore blockage more rapidly as the SO2 concentration is raised. In addition, it appears that the ultimate extent of sulfation attainable is slightly dependent on the amount of SO2 present in the bulk gas. 8.6.1.6 Combustor Pressure Combustor pressure also affects sulfur capture in an FBC boiler. No peak in the sulfur retention versus temperature curve is observed at high pressure
Fluidized-Bed Firing Systems
293
(12 atmospheres). Limestone performance is reduced, but there is an improvement in dolomite performance. Dolomite is the preferred sorbent in pressurized fluidized-bed combustors. 8.6.1.7 Chemical Composition Sorbent performance in FBC boilers has been well documented in the literature. It is generally accepted that sorbent performance has no relationship to chemical composition [6, 25, 26, 32, 33]. A high-purity limestone (>95% CaCO3) will not necessarily perform better than a low-purity limestone (<70% CaCO3). 8.6.1.8 Porosity The total porosity, a dimensionless quantity, is defined as the ratio of the sum of the pore volume to the total volume. It is an important physical property because it provides insight into whether the solid has sufficient interior vacancy in which the sulfation reaction can occur. Porosities of raw stones generally vary between 0.3% and 12.0% and that of calcines from 20% to 50%, which is less than the theoretical porosity of 54% for CaO derived from carbonates [34]. The porosity of a stone increases upon heating because of the release of CO2. The rate of porosity increase is dependent on the calcination conditions. Since this porosity increase is so condition specific, it is impossible to predict a priori. However, porosity increases for different stones of between a factor of 5 and 180 for the same calcination conditions have been reported [35]. This increase in porosity can be quickly negated by four processes: sulfation; thermal sintering (>1,470 F [>950 C]); moisture-activated sintering, which is believed to be most prevalent at temperatures around 1,110 F (600 C); and CO2-activated sintering, which is most intense at approximately 1,650 F (900 C) [36]. The importance of using an initially porous stone can be seen from the following equation [37]: xs ¼ eo =ðz 1Þð1 eo Þ
[8-7]
where xs ¼ the maximum conversion at the surface of CaO to CaSO4 before pore closure occurs; eo ¼ the initial porosity of the calcine; and z ¼ molar volume of CaSO4/molar volume of CaO. Using 0.53 for the initial porosity of CaO, xs is maximized at 56% utilization. However, stone utilizations typically range between 30% and 40%. The reason for the 16–26% difference is partly that the porosity of
294 Combustion Engineering Issues for Solid Fuel Systems
the stone is quickly reduced by the formation of CaSO4, and only a 30–40 mol% fraction of the outermost region of the stone reacts with the gaseous SO2. This phenomenon is often referred to as pore mouth plugging and is the dominant cause for only partial sulfation of sorbent particles in the size range between 18 and 140 mesh (1,000 and 105 mm) [36]. There are primarily four methods to increase stone utilization via optimizing calcine structure. They are (1) calcining with high CO2 partial pressures to produce a wide-mouth pore size distribution, (2) using slow calcination rates, (3) calcining at temperatures between 1,380 F and 1,470 F (750 C and 950 C), and (4) using impure stones. Particle porosity is believed to be the controlling parameter for larger particles. Porosity accounts for larger product molar volumes, which in turn can result in larger sorbent utilizations at long residence times (i.e., when pore plugging becomes rate limiting) [38]. The pore size distribution plays only a minor role in controlling sulfation, whereas increasing the sorbent porosity appears to be the single most promising method to enhance sulfation, especially if the stone is initially 15% porous [39]. However, it has been shown that large-mouth pores greatly enhance stone utilizations and identified pore volume distribution and grain size as the determining factors for stone utilization [40]. 8.6.1.9 Surface Area Surface area is related to porosity. As with porosity, the surface area of the resulting calcine increases substantially upon heating. The surface area development is controlled by the particle size and calcining conditions—temperature, heating rate, and environment. Surface areas as large as 90 m2/g have been obtained by using low temperatures (1,110– 1,650oF [600–900oC]), high heating rates, small particles, and CO 2-free environments. As with porosity, internal surface area is the limiting parameter for smaller particles, smaller product molar volumes, lower conversions, and short residence times (i.e., when kinetics may be rate determining) [38]. Similarly to chemical composition and porosity, there is no experimental correlation between surface area and stone performance. However, for an irreversible first-order gas-solid reaction under chemical control, the time required to reach a given conversion is inversely proportional to the active surface area and inversely proportional to the square of the surface under product layer diffusion control. For fluidized-bed applications, the latter relation is of most interest [41]. 8.6.1.10 Particle Size Conversion of CaO to CaSO4 increases as particle size decreases over the entire particle size range of 1–1,000 mm. From a fundamental viewpoint, reducing the particle size is beneficial because less CaO will be inaccessible
Fluidized-Bed Firing Systems
295
(on a weight basis). The ideal particle size distribution for an FBC power plant is site specific. That is, it depends on the manufacturer and specifications of the combustor and cyclone, the economics of producing and feeding fines, the ash content of the fuel, and the propensity for the stone to attrit. These are a few of the many considerations that must be included in the evaluation process. Laboratory results indicate that reducing the particle size will be beneficial in capturing more sulfur per unit mass of stone. However, in the field, operating conditions and economics will dictate the specifications of the particle size distribution.
8.6.2 Nitrogen Oxides The nitrogen oxides (mainly NO, NO2, and N2O but referred to as NOx) that are formed in an FBC derive from two sources—oxidation of fuel nitrogen, which is referred to as fuel NOx, and reactions of oxygen and nitrogen in the air, which is referred to as thermal NOx. Since thermal NOx is a product of a high-temperature process, mainly above 2,700 F (1,480 C), it makes only a minor contribution to the overall NOx emissions, and fuel NOx is the primary contributor. Typically, more than 90% of the NOx is in the form of NO, approximately 20–300 ppm as N2O, with the balance NO2 [6]. A discussion of NOx formation mechanisms, the effect of fuel characteristics and operating parameters on NOx formation, and NOx reduction techniques will be presented in the following sections. 8.6.2.1 NOx Formation The formation of NOx compounds is complex. Upon heating the fuel, the organically bound nitrogen volatilizes forming volatile-nitrogen and charnitrogen. The volatile-nitrogen compounds are primarily released as HCN and NH3. In the char, the nitrogen is bound in aromatic structures [6]. HCN and NH3 undergo the following reactions during combustion [6]: HCN þ 5=4 O2 ! NO þ CO þ 1=2 H2 O
[8-8]
HCN þ 3=2 O2 þ NO ! N2 O þ CO þ 1=2 H2 O
[8-9]
HCN þ 3=4 O2 ! 1=2 N2 þ CO2 þ 1=2 H2 O
[8-10]
NH3 þ 5=4 O2 ! NO þ 3=2 H2 O
[8-11]
NH3 þ 3=4 O2 ! 1=2 N2 þ 3=2 H2 O
[8-12]
During char oxidation, the nitrogen is mainly oxidized to NO and N2O. These are, in turn, partially reduced to N2. NO is reduced through
296 Combustion Engineering Issues for Solid Fuel Systems
reactions with NH3, carbon in the char, and CO. N2O can be reduced by temperature effects or reactions with char and CO. NO and N2O reduction reactions are as follows [6]: NO þ NH3 þ 1=4 O2 ! N2 þ 3=2 H2 O
[8-13]
NO þ 2=3 NH3 ! 5=6 N2 þ H2 O
[8-14]
NO þ CðcharÞ ! 1=2 N2 þ CO
[8-15]
N2 O ! N2 þ 1=2 O2
[8-16]
N2 O þ CðcharÞ ! N2 þ CO
[8-17]
N2 O þ CO ! N2 þ CO2
[8-18]
8.6.2.2 Fuel Nitrogen and Volatile Matter Content: Fuel Rank In general, NO and N2O emissions increase with increasing nitrogen content in the fuel, whereas with increasing fuel volatile matter content, NO emissions usually increase but N2O emissions decrease. N2O emissions are primarily dependent on the type of fuel (as well as combustion temperature, as discussed later), with low-rank fuels generating lower amounts of N2O than higher rank fuels, i.e., N2O emissions from lowest to highest—biomass, peat, oil shale, brown coal/lignite, bituminous coal [42, 43]. This is attributed to the amount of NH3 released from the fuel as lower rank fuels (i.e., higher volatile matter fuels) release more NH3 and their NH3 to HCN ratio is greater than for higher rank fuels. Oxidation of NH3 results in NO formation, while HCN produces both NO and N2O. 8.6.2.3 Combustion Temperature Combustion temperature has a significant effect on both NO and N2O emissions [42, 43]. NO emissions increase, but N2O emissions decrease. Higher temperatures promote the oxidation of nitrogen radicals to NO and reduce the char and CO concentrations, which in turn decrease the reduction of NO to N2 on the char surface. The reduction reactions of N2O with hydrogen radicals (H, OH) are significantly enhanced via the reactions [6]: N2 O þ H ! N2 þ OH
[8-19]
N2 O þ OH ! N2 þ HO2
[8-20]
Fluidized-Bed Firing Systems
297
8.6.2.4 Excess Air An increase in excess air leads to increases in NO and N2O emissions. As excess air is increased, the combustion rate increases. However, excess air has a minor effect on N2O emissions compared to fuel type and combustion temperature [43]. This is especially true in pressurized units [6]. 8.6.2.5 Gas Velocity/Residence Time The superficial gas velocity, i.e., residence time in the bed, has an effect, although minor, on N2O emissions but not NO emissions. With increasing gas velocity, the contact time between the gas and particles decreases, thereby resulting in higher N2O emissions because there is less time for the carbon-N2O reduction reaction to occur [43]. 8.6.2.6 Limestone Effects The effect of limestone addition on NOx emissions is complex. In a BFBC boiler, the limestone is primarily in the dense phase of the bed where CO concentration is high and O2 concentration is low and the catalyzed reduction of NO by CO may dominate over oxidation of NH3, thereby leading to lower NO emissions [6]. In a CFBC boiler, however, the limestone is distributed through the entire combustor, and oxidation of volatiles to NO may occur in the upper zone where CO concentrations are lower, thereby increasing NO emissions. Limestone addition is usually found to decrease N2O emissions although the effect is minor [6, 43]. The decrease is attributed to the limestone catalyzing the decomposition of N2O. 8.6.2.7 NOx Reduction Techniques The FBC process inherently produces lower NOx emissions due to its lower operating temperature. However, where necessary, additional combustion modifications or flue gas treatment for NOx control can also be employed. Techniques currently used for FBC systems include reducing the peak temperature by flue gas recirculation, natural gas reburning, overfire air/air staging, fuel reburning, low excess air, and reduced air preheat [10]. Postcombustion control is also used, including selective noncatalytic reduction (SNCR) and selective catalytic reduction (SCR), which achieve 35–90% NOx reductions. Air staging, SNCR, and SCR will be briefly discussed here but are presented in more detail in Chapter 9 and elsewhere [10]. Air staging can reduce both NO and N2O emissions from FBC boilers. This is accomplished by putting less than the theoretical amount of combustion air through the distributor plate and adding the remainder of the combustion air above the dense bed or, in the case of CFBC boilers, injecting some of the secondary air into the cyclone. As a result, some of the fuel nitrogen compounds decompose into molecular nitrogen rather
298 Combustion Engineering Issues for Solid Fuel Systems
than form NOx due to the reducing atmosphere in the bed. Char and CO concentrations in the bed increase, thereby enhancing the rates of NO and N2O reduction on the char. SNCR is a technology that involves injecting nitrogen-containing chemicals into the FBC boiler within a specific temperature window without the expensive use of catalysts. The chemicals, with the two most common being ammonia and urea, selectively react with NO in the presence of oxygen to form molecular nitrogen and water. The main reactions when using ammonia or urea are, respectively, as follows: 4NO þ 4NH3 þ O2 ! 4N2 þ 6H2 O
[8-21]
4NO þ 2COðNH2 Þ2 þ O2 ! 4N2 þ 2CO2 þ 4H2 O
[8-22]
The optimum temperature window for ammonia is 1,560–1,920 F (850–1,050 C) and for urea is 1,830–2,100 F (1,000–1,150 C). While both have been used, ammonia is favored due to its optimum temperature window and the fact that ammonia produces less N2O than urea [43]. SNCR operation is very effective for reducing NO, but it does increase N2O emissions. Fortunately, the overall concentration of N2O is much less than NO. Where it is necessary to reduce N2O emissions even further, a combination SNCR/SCR can be employed for overall NOx control [43].
8.6.3 Particulate Matter Particulate matter is generated from two sources: the fuel and sorbent. The mineral matter in the fuel is released during combustion, and this residue, along with calcined sorbent and reacted sorbent, is referred to as ash. In addition to the ash, there can be unburned carbon in the residue although this amount should be minimal, provided the system is properly operating. Some of the ash remains in the fluidized-bed and is discharged by the bed material drain system. This ash is normally larger than 105 mm and is relatively easy to handle and transport [14]. The remaining ash is fine and leaves the boiler in the flue gas. This material is typically less than 44 mm and requires a high-efficiency collection device. Normally, this device is a fabric filter but can also be an electrostatic precipitator (see Chapter 9 for a discussion of particle-removal devices). Ash from an FBC system contains a significant amount of CaSO4, CaO, and CaCO3 and is more alkaline than that from conventional combustion systems.
8.6.4 Carbon Monoxide/Hydrocarbons Carbon monoxide is the product of incomplete combustion of carbon. It is formed when the oxygen supplied is less than the amount required for
Fluidized-Bed Firing Systems
299
stoichiometric combustion of carbon to CO2, when there is inadequate fuel/air mixing, or insufficient residence time for combustion. Fuel reactivity and bed temperature also influence CO emissions. Hydrocarbon emissions are produced under similar conditions. Typical flue gas concentrations are less than 200 ppm for CO and 20 ppm for hydrocarbons in a CFB boiler burning coal [14]. Overfire air, i.e., air staging, is a common technique used for controlling CO and hydrocarbon emissions.
8.6.5 Trace Elements All solid fuels contain small concentrations of trace elements, usually measured in parts per million (ppm). Trace elements enter the atmosphere through natural processes, and sources of trace elements include soil, seawater, and volcanic eruptions. Human activities such as power generation and combustion of fuels in the industrial and commercial sectors also lead to emissions of some elements. Title III of the U.S. Clean Air Act Amendments of 1990 designates 188 hazardous air pollutants. Included in the list are 11 trace elements: antimony (Sb), arsenic (As), beryllium (Be), cadmium (Cd), chromium (Cr), cobalt (Co), lead (Pb), manganese (Mn), mercury (Hg), nickel (Ni), and selenium (Se). In addition, barium (Ba) is regulated by the Resources Conservation and Recovery Act, and boron (B) and molybdenum (Mo) are regulated Irrigation Water Standards. Vanadium (V) is regulated based on its oxidation state, and vanadium pentoxide (V2O5) is a highly toxic regulated compound. Other elements, such as fluorine (F) and chlorine (Cl), which produce acid gases (i.e., HF and HCl) upon combustion, and radionuclides such as radon (Rn), thorium (Th), and uranium (U) are also of interest. The distribution of trace elements in the bottom ash, ash collected in the particulate control devices, and fly ash and gaseous constituents emitted into the atmosphere depends on many factors, including the volatility of the elements, temperature profiles across the system, pollution control devices, and operating conditions [44, 45]. Numerous studies have shown that trace elements can be classified into three broad categories based on their partitioning during coal combustion. A summary of these studies is presented by Clarke and Sloss [45], and Figure 8-9 illustrates the classification scheme for selected elements. Class I elements are the least volatile and are concentrated in the coarse residues (i.e., bottom ash) or are equally divided between coarse residues and finer particles (i.e., fly ash). Class II elements will volatilize in the boiler but condense downstream and are concentrated in the finer sized particles. Class III elements are the most volatile and exist entirely in the vapor phase. Overlap between the classifications exists and is a function of fuel, combustion system design, and operating conditions, especially temperature [45].
300 Combustion Engineering Issues for Solid Fuel Systems
Increasing Volatility
Hg Cl F
Rn
Class III Volatilized and Emitted Fully in the Vapor PhaseNot Enriched in the Fly Ash
B Se As Cd K Na Pb Sb Ti
Class II Enriched in the Fly Ash and Depleted in the Bottom Ash
Ba Be Ca Co Cr Cu Mg Mo Ni Sr U V Al Fe Mn Si Th
Class I Equally Distributed Between the Bottom Ash and the Fly Ash
FIGURE 8-9 Classification scheme for selected trace elements relative to their volatility and partitioning in power plants (Adapted from Clarke and Sloss [45] and Miller et al. [46]).
The operating temperature in an FBC boiler is lower than that from a conventional system (i.e., pulverized coal or stoker-fired) which may lead to reduced volatilization of some elements. However, this reduction may be offset to some degree by the longer residence times at a relatively high temperature in the FBC boiler, allowing more volatilization to occur [45]. Mercury, which has become an element of much interest due to its toxicity and recent regulation for coal-fired boilers [47], has been shown to be captured by the limestone that is used for sulfur control in CFBC boilers. The mercury is captured by the fine limestone particles and removed by the particulate control device. According to Hughes and Littlejohn [48], most trace elements associated with the particulates and emissions of the trace elements from the system depend primarily on the efficiency of the particulate control device. Conventional baghouse filters have an overall removal efficiency of greater than 99%. Once a dust cake is formed, the efficiency for removal of even the smaller particles (down to 0.1 mm) can approach 100% [45, 49]. The following discussion is evidence of the low mercury emissions rate from CFBC boilers where the mercury is associated with the limestone particles captured by bagfilters. In 1999, the U.S. Environmental Protection Agency (EPA) approved an Information Collection Request (ICR) to study mercury emissions from power plants to provide some framework for comparison as a basis for implementing mercury control legislation. Part III of the ICR was to determine speciated mercury emissions from stationary sources. This included
Fluidized-Bed Firing Systems
301
power plants of various sizes, systems, and firing configurations throughout the United States. The details of the ICR can be found at the U.S. EPA website [50]. Three CFBC units were included in the ICR: Stockton CoGen Plant, Stockton, California; PG&E Scrubgrass Generating Station, Unit 1, Kennerdell, Pennsylvania; and Tractebel Power Inc., Kline Township Cogeneration Plant Unit 1, New York. The Stockton plant is a CFB boiler that fired bituminous coal and fluid coke, with mercury concentrations of 26–29 ppb and 12–45 ppb, respectively, at 55,000 lb/h during the test. Ammonia and limestone were also fed into the system for emissions control. The emissions tests showed no detection of mercury at the baghouse outlet. Detection limits were equivalent to 0.1 mg/dscm (milligram per dry, standard cubic meter), which translates to 0.00005 to 0.00008 lb/h. Mercury was detected in the particulate samples at the baghouse inlet; however, no oxidized or elemental mercury (gas phase) was detected. The Scrubgrass plant fired waste bituminous coal in a CFB boiler with limestone injection. The fuel was fed at a rate of 75,000 lb/h (3.95 lb Hg/h based on 527 ppb Hg in the waste coal). Emissions tests showed that 99.8% of the mercury was captured with the ash in the baghouse. This corresponds to emissions of 1.39 lb Hg/year. Capture efficiency across the baghouse resulted in a reduction of 99.99%, 85.88%, and 19.00%, respectively, in particle-bound mercury, oxidized mercury, and elemental mercury. Similarly, Tractebel’s plant fired anthracite cleaning wastes (i.e., culm) in a CFB boiler. Test results showed that 99.80% of the mercury was captured with the ash in the baghouse. This amounts to approximately 0.53 lb Hg release per year. The average mercury in the fuel during the test was 330 ppb; however, the plant reported that the mercury concentration is normally about 170 ppb, which correlates to emissions of 0.25 to 0.30 lb per year. The baghouse removal efficiency of the particulate-bound mercury was 99.995%. Oxidized and elemental mercury removal through the baghouse was 43.69% and 91.16%, respectively. Overall, certain trace elements are captured by the bed material, which is dependent on the type of element, bed temperature, and fuel characteristics. The remaining elements will exit the combustor by the flue gases and, as the temperature decreases, condense onto cooler surfaces and fly ash particles, which will be collected by the particulate-collection devices. The highly volatile elements will remain in the flue gas or condense on the submicron particles that escape the particulate-removal device.
8.7 Ash Chemistry and Agglomeration Issues An important issue that must be assessed in an FBC system is the behavior of the inorganic elements toward bed agglomeration. As previously discussed, the lower bed temperatures are advantageous in that the temperatures can be kept below the ash softening temperatures, especially when
302 Combustion Engineering Issues for Solid Fuel Systems
firing coals. However, because an FBC boiler can fire a wide range of fuels with varying ash chemistries, the interactions of the ash with the bed material can lead to agglomeration, which has a detrimental effect on fluidization. This is especially true with biomass materials where care must be taken in the design of the system (e.g., incorporation of kaolin clay injection systems for agglomeration control) or selection of the feedstocks. This is illustrated in this section, in which the occurrence of inorganic elements in various biomass materials and their effect on combustion behavior in a CFBC system were investigated. This was part of a design for a CFB boiler to be located at Penn State University in which various biomass materials were considered for cofiring with coal [20, 21, 51–55]. It has long been recognized that the mode of occurrence of inorganic elements in fossil fuels has a direct bearing on their behavior during combustion [56–59]. The occurrence of inorganic elements in bio-based fuels is also important. Inorganic species are incorporated in biomass in several ways due to the chemical makeup of the biomass, its origin and the manner in which it is collected for utilization as a fuel. The fuel may be of plant or animal base or a mixture of both due to farming practices (i.e., mixture of manure and bedding). Inorganic species can occur as ion-exchangeable cations, as coordination complexes, and as discrete minerals. In the case of firing a single fuel, such as coal, it is possible to predict ash behavior to avoid system problems. However, it becomes more complex to predict ash behavior in the case of firing multiple fuels in proportions that vary with time, e.g., seasonal changes, and are extremely heterogeneous. Like low-rank coals, biomass materials often contain significant amounts of alkali metals, e.g., potassium and sodium, and alkaline earth elements, e.g., calcium and magnesium, which are rapidly released into the gas phase and interact with other elements, resulting in problems with fouling, slagging, and corrosion. In general, potassium and sodium that are associated with the organic structure of the fuel tend to be problematic in that they can contribute to the formation of inorganic phases that have lower melting points. Studies conducted on ash formation during coal combustion show that the incorporation of moderate amounts of alkali and alkaline earth elements into silicates enhances the coalescence and agglomeration of inorganics due to formation of “sticky” molten phases [57, 58, 60, 61]. The presence of low-melting point phases in a fluidizedbed combustor results in the formation of clinkers that can compromise the bed fluidity. It is also important to recognize that the blending of biomass feedstocks and coal does not necessarily result in simply an additive effect of problematic elements. Changes in the feed blend may or may not have devastating effects on system operation. Predicting these effects is based on understanding the manner in which the inorganics in fuels interact during combustion and their effect on the chemical and physical properties of the ash and gas phases in the system.
Fluidized-Bed Firing Systems
303
This section presents a review of chemical fractionation data of various biomass materials and utilization of models to predict sintering potential and viscosity behavior of ash produced during combustion, with an emphasis on fluidized-bed combustion. As previously mentioned, this was part of a fuels evaluation stage in the design of the Penn State FBC system.
8.7.1 Chemical Fractionation of Biomass The chemical fractionation procedure is based on an element’s varying solubility as a result of its occurrence in a fuel. A procedure used to fractionate low-rank coals at the University of North Dakota Energy and Environmental Research Center [62] later modified by Baxter [63] was further modified by Penn State [20] to better address handling issues particular to biomass fuels. A schematic representation of the method is shown in Figure 8-10. Each step results in a liquid and solid residue sample, which are both analyzed for the following major and minor elements: Al, Ba, Ca, Fe, K, Mn, Mg, Na, P, Si, Sr, S, and Ti, using decoupled argon plasma (DCP) spectroscopy. Analysis of both the liquid and solid residue was conducted so that a material balance could be performed. In the first step of leaching, water was used to remove elements that are in a water-soluble form. This consists primarily of water-soluble salts, e.g., alkali sulfates, carbonates, and chlorides. The remainder of the residue from Step 1 was subjected to a second leaching step using ammonium acetate to remove elements that are bound loosely to organic matter, e.g., ionexchangeable elements such as potassium, calcium, sulfur, and sodium. Again, the leachate and a portion of the residue from this step were analyzed for major and minor elements.
Deionized Water Leaches raw fuel
Step 1
Water-Soluble Salts
Ammonium Acetate (1M) Leaches residue from Step 1
Step 2
Ion-exchangeable Material
Hydrochloric Acid (1M) Leaches residue from Step 2
Step 3
Acid-Soluble Salts Carbonates, Sulfates, and Mono-Sulfides
FIGURE 8-10 Schematic representation of the chemical fractionation method.
304 Combustion Engineering Issues for Solid Fuel Systems
The final leaching step used hydrochloric acid to remove elementbearing minerals that exist as acid-soluble salts such as carbonates, sulfates, mono-sulfide minerals, and simple oxides. Again, the leachate and a portion of the residue from this step were analyzed by DCP spectroscopy. The insoluble portion of the fuel is generally made up of silicates and other acid-insoluble mineral phases. Certain biomass fuels are inherently difficult to work with given the chemical fractionation procedure. It is often difficult to obtain a representative sample given the heterogeneous nature of the fuels. As an example, the manure samples evaluated consisted of a mixture of manure, straw, and sand taken from the floor of the two dairy barns, and chicken litter was a combination of the chicken manure and the wood shavings that are used as a bed material in the chicken barns. This heterogeneity was noted as a major problem in a round-robin study conducted by von Puttkamer et al. [64]. It is also extremely difficult to grind such heterogeneous samples given the different grindability of straw, sand, and dried manure. It is also difficult to work with samples that contain material that have inherently different wettabilities and densities, e.g., sand versus straw. Often, only the liquid portion of the sample is analyzed due to time and cost considerations. This is not always appropriate, as it was observed that material balances for individual elements between the sum of the solid and liquid samples and the original parent were not as good as generally obtained in coal samples. Zevenhoven-Onderwater et al. [65] also reported such difficulties in obtaining good material balances between analysis of solid and liquid samples. In short, biomass fuels required special consideration when applying the chemical fractionation procedure. Therefore, modifications to the preparation, e.g., cutting and grinding the sample, and filtering steps, e.g., addition of a centrifuge step, were made to accommodate the physical characteristics of the biomass fuels [20]. 8.7.1.1 Results of the Chemical Fractionation Study Proximate, ultimate, and ash analyses of the fuels are given in Table 8-2. Figures 8-11 through 8-19 show the weight percent of selected elements that occur as water soluble and/or ion-exchangeable, acid soluble, or insoluble in the different biomass fuels studied. For the purpose of discussion, the water-soluble and ion-exchangeable portions are combined because they are both indicative of species that are highly reactive during combustion, i.e., organically bound or water-soluble mineral phases such as carbonates. The combined water-soluble and ionexchangeable portions are referred to as water-soluble/ion-exchangeable. Acid-soluble elements are usually derived from acid-soluble mineral phases, e.g., pyrite and some clays. Insoluble phases are generally minerals such as quartz and aluminosilicates. Many of the insoluble and some of the acid soluble portions are indicative of the presence of dirt and other contaminates that make up the fuel sample and must be considered as part of the
TABLE 8-2 Proximate, Ultimate and Ash Analysis of Cofire Coal and Biomass Fuels Cofire Coal
305
Moisture Proximate analysis (wt.%, db) Volatile matter Ash Fixed carbon Ultimate analysis (wt.%, db) Carbon Hydrogen Nitrogen Sulfur Oxygen HHV (Btu/lb, db) HHV (kJ/kg, db) Bulk density (lb/ft3) Bulk density (g/cc) Ash Analysis (wt.%) Al2O3 BaO CaO Fe2O3 K2O MgO MnO Na2O P2O5 SiO2 SO3 SrO TiO2
Pine Shavings
Reed Canary Grass
Sheep Manure
Dairy Free-Stall Manure
Dairy Tie-Stall Manure
Misc. Manure
Poultry Litter
5.0
45.0
65.2
47.8
70.3
69.8
50.5
20.0
24.2 14.7 61.1
84.7 0.1 15.2
76.1 4.1 19.8
65.2 20.9 14.0
30.6 62.3 7.1
30.1 62.5 7.4
21.8 73.5 4.8
55.3 17.0 7.7
72.8 3.9 1.5 2.3 4.8 13,118 30,493 – –
49.1 6.4 0.2 0.2 44.0 8373 19,455 11.9 0.10
45.8 6.1 1.0 0.1 42.9 7239 16,828 3.12 0.05
40.6 5.1 2.1 0.6 30.7 6895 16,021 23.1 0.37
22.1 2.9 1.1 0.1 11.5 3799 8832 50.5 0.81
22.6 2.9 1.1 0.1 10.8 8203 19,070 50.5 0.40
19.6 2.5 1.0 0.1 3.3 3114 7238 43.7 0.7
38.1 5.6 3.5 0.6 30.9 6399 14,874 – –
25.34 – 2.28 18.34 2.22 0.82 – 0.25 0.4 48.2 0.67 – –
13.4 0.15 8.75 5.94 4.94 3.35 0.49 1.38 1.44 57.2 0.05 0.80 1.16
1.66 0.05 9.57 1.47 18.1 5.29 0.11 2.34 13.8 43.0 0.02 0.11 4.99
3.08 0.05 12.8 1.95 23.4 5.74 0.17 4.64 9.21 29.3 5.52 0.03 0.20
0.96 0.02 6.38 1.29 6.75 2.65 0.17 1.32 2.90 74.98 0.04 0.10 2.06
2.26 0.02 23.3 1.37 10.7 8.91 0.14 7.04 14.7 26.0 0.14 0.11 5.08
1.34 0.01 3.44 0.93 1.77 1.06 0.03 0.88 2.54 84.82 0.01 0.14 1.20
9.14 0.05 12.7 4.04 9.94 4.01 0.36 3.60 14.0 39.4 2.58 0.03 0.51
306 Combustion Engineering Issues for Solid Fuel Systems 100 90 80 70 Water soluble + ion exchangeable Acid soluble
Weight %
60 50
Insoluble
40 30 20 10 0 Sheep Manure
Chicken Litter
Dairy Dairy Tie-Stall Free-Stall Manure Manure
Misc. Manure
Pine Reed Canary Shavings Grass
FIGURE 8-11 Occurrence of potassium in biomass. 100 90 80
Weight %
70 60
Water soluble + ion exchangeable Acid soluble Insoluble
50 40 30 20 10 0 Sheep Manure
Chicken Dairy Litter Tie-Stall Manure
Dairy Misc. Pine Reed Canary Free-Stall Manure Shavings Grass Manure
FIGURE 8-12 Occurrence of sodium in biomass.
total fuel analysis. It should be mentioned that sample reproducibility is also difficult due to variability of the fuels. Potassium occurs predominantly in water-soluble/ion-exchangeable forms (Figure 8-11). In all four manures and the Reed Canary grass, 95% of the total potassium is water-soluble/ion-exchangeable. The pine shavings and chicken litter contained a moderate amount of water-soluble/ ion-exchangeable potassium, with the balance being in the insoluble form.
Fluidized-Bed Firing Systems
307
100 90 80
Weight %
70 Water soluble + ion exchangeable Acid soluble Insoluble
60 50 40 30 20 10 0 Sheep Manure
Chicken Dairy Litter Tie-Stall Manure
Dairy Misc. Free-Stall Manure Manure
Reed Canary Pine Shavings Grass
FIGURE 8-13 Occurrence of calcium in biomass. 100 90 80
Weight %
70 60
Water soluble + ion exchangeable Acid soluble Insoluble
50 40 30 20 10 0 Sheep Manure
Chicken Dairy Dairy Misc. Pine Reed Canary Tie-Stall Free-Stall Manure Shavings Grass Litter Manure Manure
FIGURE 8-14 Occurrence of magnesium in biomass.
The insoluble potassium in the chicken litter is attributed to the significant amount of wood chips that make up the litter. Sodium is also present predominantly (90%) in a water-soluble/ionexchangeable form in all the biomass fuels except for the pine shavings (76%) (Figure 8-12). The remaining sodium is present mostly in an insoluble form. Virtually all the calcium in the fuels is either present as watersoluble/ion-exchangeable or acid-soluble (Figure 8-13). Less than 1.6% of
308 Combustion Engineering Issues for Solid Fuel Systems 100 90 80
Weight %
70 60
Water soluble + ion exchangeable Acid soluble Insoluble
50 40 30 20 10 0 Sheep Manure
Chicken Dairy Litter Tie-Stall Manure
Dairy Free-Stall Manure
Misc. Pine Reed Canary Manure Shavings Grass
FIGURE 8-15 Occurrence of aluminum in biomass. 100 90 80
Weight %
70 60
Water soluble + ion exchangeable Acid soluble Insoluble
50 40 30 20 10 0 Sheep Manure
Chicken Dairy Dairy Misc. Pine Reed Canary Litter Tie-Stall Free-Stall Manure Shavings Grass Manure Manure
FIGURE 8-16 Occurrence of silicon in biomass.
the calcium remained in the insoluble portion of the fuel. Unlike potassium and sodium, there was a significant portion of acid-soluble calcium ranging from 5.6–68%. The plant fuels tended to have significantly less acid-soluble calcium (5–6%) than the manure samples (17–68%). Magnesium followed a similar trend as calcium with slightly lower amounts of acid-soluble magnesium and some insoluble magnesium present (Figure 8-14). Interestingly,
Fluidized-Bed Firing Systems
309
100 90 80
Weight %
70 60
Water soluble + ion exchangeable Acid soluble Insoluble
50 40 30 20 10 0 Misc. Pine Reed Canary Sheep Chicken Dairy Dairy Tie-Stall Free-Stall Manure Shavings Grass Manure Litter Manure Manure
FIGURE 8-17 Occurrence of sulfur in biomass. 100 90 80
Weight %
70 60 50
Water soluble + ion exchangeable Acid soluble Insoluble
40 30 20 10 0 Sheep Manure
Chicken Dairy Litter Tie-Stall Manure
Dairy Misc. Free-Stall Manure Manure
Pine Reed Canary Shavings Grass
FIGURE 8-18 Occurrence of iron in biomass.
calcium is not involved in the formation of melt phases predicted by the thermodynamic modeling discussed in the next section. Aluminum and silicon are concentrated in the insoluble portion of the fuels (Figures 8-15 and 8-16). This is expected given that many of the manure samples also included dirt, i.e., quartz, clay minerals, from the stall
310 Combustion Engineering Issues for Solid Fuel Systems 100 90 80
Weight %
70 60
Water soluble + ion exchangeable Acid soluble Insoluble
50 40 30 20 10 0 Sheep Manure
Chicken Dairy Tie-Stall Litter Manure
Misc. Dairy Free-Stall Manure Manure
Pine Reed Canary Shavings Grass
FIGURE 8-19 Occurrence of phosphorus in biomass.
as well as hay/straw and sand. There was some water-soluble/ion-exchangeable and acid-soluble aluminum present in some of the samples (Figure 8-15). The source of this aluminum was not determined. Water-soluble/ionexchangeable silicon was measured in the pine shavings and Reed Canary grass. Silicon is not typically found in ion-exchangeable form, so no explanation is given. Material balance of silicon was not very good. This is attributed to the varied contamination of sand/dirt in many of the samples. Sulfur occurred predominantly in the water-soluble/ion-exchangeable portion (23–96%) (Figure 8-17). The percent of sulfur that was acid-soluble ranged from 2–46%. Iron was the only element present predominantly in the form of acid-soluble species (26–71%) (Figure 8-18). Phosphorous was present predominantly in a water-soluble/ionexchangeable form ( 60%) (Figure 8-19). The difference was found mostly in the acid-soluble form ranging from 0.3% in the Reed Canary grass to 34.8% in the miscellaneous manure sample. In general, a significant portion of the alkali and alkaline earth elements occur in the water-soluble/ion-exchangeable portion of the biofuels. Zevenhoven-Onderwater et al. reported similar results for forest residue, Salix (low Si) and Salix (high Si) [65]. The high percentage of alkali and alkaline earth elements in the water-soluble/ion-exchangeable form is cause for concern given their potential for forming molten phases in the bed during CFB combustion. Extraneous quartz is fairly inert within the gas stream in the absence of volatilized alkali and alkaline earth elements. Volatilized alkali and alkaline earth elements can migrate into the silicate structure,
Fluidized-Bed Firing Systems
311
forming phases that have lower melting points. It is important not only to look at the elemental concentration on a fuel basis, but also to consider the interaction of elements at the temperature regime for a given system to better assess potential fuel blends for a particular combustion system.
8.7.2 Thermodynamic Modeling to Predict Inorganic Phases A series of fuel blends was used as input into a Gibbs free energy minimization program called FactSage developed at the Facility for the Analysis of Chemical Thermodynamics (FACT), Centre for Research in Computational Thermochemistry (CRCT), E´cole Polytechnique de Montre´al, Canada, and GTT Technologies [66]. The program calculates equilibrium composition for a given system at a set of defined temperature and/or pressure conditions. An average biomass and coal fuel blend was identified as a feasible fuel blend to be fired at Penn State University (Baseline Blend, Table 8-3), as part of the study to evaluate cofiring agricultural and opportunity fuels with coal. (Details of the study are found elsewhere [20].) The coal identified is a medium volatile bituminous coal. The inorganic composition of potential fuel blends is given in Table 8-4. The average fuel blend composition was used as input into the FactSage Thermodynamic modeling program to determine the state of the inorganic phases present in the bed. An average temperature of 1,171 K (898 C, 1,650 F), to represent an average anticipated bed temperature, and a firing rate of 58.6 MWt (200 million [MM] Btu/h) were used. The equilibrium phases predicted by FactSage are given in Table 8-5. TABLE 8-3 Percent Thermal Input of Proposed and Theoretical Fuel Blends Based on a Firing Rate of 58.6 MWt (200 MMBtu/h) % Thermal Input Fuel Coal Sewage Sludge Sheep Manure Chicken Litter Dairy Tie-Stall Manure Dairy Free-Stall Manure Misc. Manure Red Oak Shavings Pine Shavings Reed Canary Grass
Baseline Blend 83.8 0.4 0.1 0.0 0.4 0.0 0.3 8.4 6.5 0.2
Chicken Litter
Manure Blend 1
Manure Blend 2
ManureCoal Cofire 84.9
59.0
25
3.9
21.5 8.1 11.7
25 25 25
4.0 3.4 3.9
100
312 Combustion Engineering Issues for Solid Fuel Systems TABLE 8-4 Inorganic Analysis of Fuel Blends (Fuel Basis, As-Fired) Weight % Oxide
Baseline Fuel Blend
Chicken Litter
Manure Blend 1
Manure Blend 2
Manure-Coal Cofire
Al2O3 BaO CaO Fe2O3 K2O MgO MnO Na2O P2O5 SiO2 SO3 SrO Ash %
2.29 0.00 0.43 1.65 0.23 0.11 0.00 0.03 0.07 4.87 0.07 0.00 9.78
2.53 0.01 3.50 1.12 2.75 1.11 0.10 0.99 3.86 10.89 0.71 0.01 27.72
0.24 0.00 0.98 0.16 1.36 0.40 0.01 0.32 0.68 6.38 0.27 0.01 10.93
0.17 0.00 0.64 0.12 0.63 0.24 0.01 0.18 0.43 8.00 0.08 0.01 10.67
1.96 0.00 1.10 1.35 0.60 0.30 0.01 0.20 0.63 8.66 0.13 0.01 0.11
The most basic scenario was to input the chemical analysis of the fuel blend in the oxide form. It is acknowledged that the elements may or may not be present as oxides. At 1,171 K (1,650 F), the phases present in equilibrium are given in Table 8-5. In some cases, mineral names are assigned to chemical formulas. This does not necessary imply any information regarding the crystallinity of the phase but only a match with regard to chemical composition. In the Baseline Blend fuel, there are no liquid phases present at 1,171 K. All the alkali earth elements are tied up in aluminosilicates that have melting points higher than 1,171 K. The coal provides a significant source of aluminum to favor the formation of aluminosilicates versus silicates that have lower melting points. At equilibrium at 1,171 K, the chicken litter fuel contains the liquid phase Na2SO4 (3.1 wt.%) (Table 8-5). The chicken litter contains significant amounts of sodium as compared to the other fuels. The remaining alkali earth elements are divided among other silicates. Previous work conducted at Penn State involved combustion studies of chicken litter in a fluidizedbed combustor during which significant clinkering occurred in the bed. As is common practice, kaolin clay was added to the fuel feed to reduce the occurrence of clinkering in the bed [67, 68]. Kaolinite [Al2Si2O5(OH)4] is the main constituent of kaolin clay. The net effect of the clay is to increase the aluminum in the ash that shifts the equilibrium composition away from the formation of phases having lower melting points. In addition, the kaolin also dilutes the concentration of alkali and alkaline earth
Fluidized-Bed Firing Systems
313
TABLE 8-5 Inorganic Phases Predicted at Equilibrium at 1171 K. All Phases are Solid Unless Followed by (l), Indicating a Liquid Phase. Liquid Phases are also Indicated in Bold Typeface. Weight % Phase SiO2/tridymite CaAl2Si2O8/anorthite Fe2O3/hematite Al6Si2O13/mullite KAlSi2O6/leucite Mg2Al4Si5O18/cordierite NaAlSi3O8 CaSO4/anhydrite Ca3Fe2Si3O12/andradite MgOCa2O2Si2O4/ akermanite Na2Ca2Si3O9 Mg2SiO4/forsterite K3Na(SO4)2 Na2SO4(l) CaOMgOSiO2/ monticellite K2Si4O9(l) Na2Ca3Si6O16 MgOCaOSi2O4/diopside Na2Mg2Si6O15 K2SO4
Baseline Blend 25.7 19.4 17.1 14.8 11.1 8.3 2.7 1.2
Chicken Litter
7.8
Manure Blend 1
Manure Blend 2
ManureCoal Cofire
11.0
50.0
1.7
1.2
27.0 18.7 9.1
10.8
7.3
19.4 11.8 1.5 1.1
25.7 13.9 29.4 8.2 7.0 3.1 4.9 31.0 22.2 15.4 8.4 6.2
13.3 13.2 10.0 3.3 1.7
11.3
elements. The net effect is to shift the reaction in favor of forming aluminosilicates having higher melting points. Interestingly, calcium is not involved in the formation of melt phases. As mentioned earlier, calcium occurs predominantly in an acid soluble form in the chicken litter. Hald [69] studied the addition of limestone on the formation of liquid phases during combustion of coal and straw and suggested that CaO was only a minor contributor to the formation of melt phases. The extent to which organically bound alkali and alkaline earth elements volatilize depends on the combustion temperature, as suggested by the work by Helble et al. [70, 71]. However, the volatility of sodium or presence of sodium volatiles in the gas stream decreases with temperature. The reason for this is that at higher temperatures the organically bound sodium will react with silicate particles in the char and will not be released into the gas stream [72, 73]. At combustion temperatures less than 1900 K (2,960 F), sodium chloride and sodium cations are vaporized from the char.
314 Combustion Engineering Issues for Solid Fuel Systems
At temperatures greater than 1900 K, inherent quartz begins to soften, allowing diffusion of sodium into the silicate structure. This reaction of sodium with inherent silicate particles at high temperatures usually results in the formation of molten silicate particles which ultimately coalesce. The coalescence or agglomeration of silicate particles is greatly enhanced due to the incorporation of alkali and alkaline earth elements. Two manure blends utilizing no coal support were run using FactSage. It is recognized that this may not necessarily represent a real-life scenario but serves to evaluate the unique nature of biomass. Each blend consisted of sheep, dairy tie- and dairy free-stall, and miscellaneous manure. Manure Blend 1 was based on similar feed rates for the dairy and miscellaneous manures (6,820 kg/h) and 13,545 kg/h feed rate for the sheep manure. The sheep manure has significantly higher levels of potassium, calcium, and sodium than the other manures. Manure Blend 2 is based on equal thermal input by the different manures. Manure Blend 1 had significant amounts of liquid phase (33 wt.%) K2Si4O9(l) present at equilibrium. Manure Blend 1 contained approximately twice as much K2O as Manure Blend 2. Manure Blend 1 had potassium contained within three species: K2Si4O9(l) contained 60% of the potassium; KAlSi2O6 contained 16% of the total potassium, and K2SO4 contained 24% of the potassium. Manure Blend 2 had potassium contained within the same three species as Manure Blend 1 with 60% of the potassium in the liquid phase (K2Si4O9), which accounted for 13.3 wt.% of the total inorganic material. The high percentage of liquid phase is attributed to the low concentration of Al2O3 present in the fuel. Potassium aluminosilicates tend to have higher melting points than potassium silicates. Zevenhoven-Onderwater et al. defined a T15 (critical temperature) as the temperature at which 15 wt.% of the ash is present in a molten phase, thereby enabling fly ash deposition in the flue gas pass or formation of sticky bottom ash and possible bed sintering and agglomeration [65]. The T15 for forest residue was exceeded between 873 K (1,110 F) and 1,133 K (1,600 F). The T15 for the Salix (low Si) was reached between 1,113 K (1,600 F) and 1,273 K (1,830 F). Salix (high Si) was predicted at temperatures greater than 1,303 K (1,890 F). The importance of alkali and alkaline earth elements in fuels can be demonstrated by the SiO2-K2O system (Figure 8-20). SiO2 (quartz) has a melting point of 1,883 K (2,930 F). However, the introduction of a minor amount of K2O, e.g., 0.02 mass fraction, into the system results in the formation of K2Si4O9(l) (9 wt.%) at 1,171 K (1,650 F). K2Si4O9(l) is in equilibrium with SiO2(s4) (tridymite) up to 1,732 K (2,660 F). An increase in the mass fraction of K2O to 0.2 increases the mass fraction of the liquid phase to 68% and ultimately leads to the formation of additional potassium silicate melt phases at lower temperatures with tridymite being consumed. Baxter and Jenkins [74] noted the impact of potassium in depressing the melting point of silicon. Baxter and Jenkins studied straw ash deposits and found that the molten region had a silicon
Fluidized-Bed Firing Systems
315
1227.0
Temperature (K)
1210.4
.
1193.8
K2O(liq) + K2SiO3(s)
9% liquid phase 0.02 mass fraction K2O
K2Si2O5(s3) + K2SiO3(s)
inc. liquid phase
1177.2
inc. K2O mass fraction 10x
K2Si2O5(s3) + K2Si4O9(liq)
1160.6
68% liquid phase 0.20 mass fraction K2O
K2Si4O9(liq)+ SiO2(s4) 1144.0 0
.2
.4 .6 mass SiO2/(SiO2+K2O)
.8
1.0
FIGURE 8-20 SiO2-K2O binary system at equilibrium.
to potassium ratio of less than 4:1 and a ratio over 25:1 in the granular region of the deposit. It should be noted that the introduction of Al2O3 into a system results in a reduction or absence of the K2Si4O9(l) phase. In Figure 8-21 the SiO2Al2O3 system is shown in which K2O makes up 0.1 mass fraction of the total system. At 0.02 mass fraction Al2O3, the liquid phase accounts for 33% of the total mass at equilibrium. Increasing the mass fraction of Al2O3 10 times to 0.20 reduces the mass fraction of liquid phase to 3%. Increasing the mass fraction of Al2O3 to 0.22 eliminates the K2Si4O9(l) phase. Mullite (KAlSi2O6) and leucite (Al6Si2O13) solid phases are in equilibrium with tridymite. The presence of Al2O3 in the system favors the formation of potassium aluminosilicates that have higher melting points compared to potassium silicates. The decrease in liquid phases with increasing aluminum content is precisely what is reflected in the reported inorganic phases in the Baseline Blend with a normalized mass fraction of Al2O3 of 31% as compared to the Manure 1 Blend at 3%. A manure cofire blend was also run in which coal provides 85% of the thermal input. The inorganic composition of the cofire blend is given in Table 8-4 and the equilibrium composition in Table 8-5. No liquid phases were calculated to be present at the target temperature of 1,171 K. In fact, no liquid phases were predicted at temperatures up to 1,866 K (2,900 F).
316 Combustion Engineering Issues for Solid Fuel Systems 1227.0
Temperature (K)
1210.4
.
KAlSi2O6(s2) + SiO2(s4) + K2Si4O9(liq)
1193.8 Al2O3(s4) + KAlSi2O6(s2) +
KAlSi2O6(s2) + Al6Si2O13(s) + SiO2(s4)
33% liquid phase at 0.02 mass fraction Al2O3
Al6Si2O13(s)
1177.2 Al2O3(s4) + KAl9O14(s) + KAlSiO4(s2)
1160.6
dec. liquid phase
Al2O3(s4) + KAlSi2O6(s2) + KAlSiO4(s2)
inc. Al2O3 mass fraction 10x
3% liquid phase at 0.20 mass fraction Al2O3
1144.0 0
.2
.4
.6
.8
1.0
mass SiO2/(SiO2+K2O+Al2O3)
FIGURE 8-21 SiO2-Al2O3 binary system at equilibrium and 0.1 mass fraction K2O.
The cofire of coal with the manure blend provides adequate aluminum and silicon to favor the formation of phases that incorporate the alkali earth elements that have higher melting points compared to phases formed in Manure Blend 1 and 2. A manure cofire in which coal provides 50% of the thermal input was run on FactSage. Interestingly, the equilibrium composition contained a trace (0.6 wt.%) Na2SO4(l) at 1,171 K.
8.7.3 Viscosity of Inorganic Melt Phases A series of coal-biomass blends was studied to determine the effect of cofiring biofuels on the formation of liquid phases. Typical temperature ranges for an FBC system were used as well as those for a pulverized coal-fired (dry bottom) boiler and a cyclone-fired boiler for comparison. A composite ash composition was calculated for each of the fuel blends and is given in Table 8-6. The average fuel blend composition and chemical fraction data were used as input into the FactSage program. Equilibrium compositions at temperatures ranging from 1,400–4,000 F were calculated for the nine fuel blends. Typical temperatures for a fluidized-bed boiler, pulverized coal-fired (dry bottom boiler), and cyclone-fired boiler range from 1,450–1,750 F, 2,500–3,000 F, and 3,300–3,600 F, respectively. The output of the FactSage model was then used as input into a viscosity program developed at Penn
TABLE 8-6 Ash Composition (Weight %) of Coal-Biomass Fuel Blends (200 MMBtu/h) Fuel
% Thermal input
Al2O3
CaO
Fe2O3
K2O
MgO
Na2O
P2O5
SiO2
SO3
1 2
100% Coal 90% Coal 10% Pine Wood 80% Coal 20% Pine Wood 80% Coal 20% Manure 90% Coal 10% Switchgrass 80% Coal 20% Switchgrass 70% Coal 30% Switchgrass 60% Coal 40% Switchgrass 80% Coal 10% Pine Wood 10% Switchgrass
25.32 25.48
2.28 2.44
18.34 18.37
2.22 2.31
0.82 0.88
0.25 0.28
– 0.43
48.20 49.10
0.67 0.67
25.19
2.60
18.07
2.37
0.94
0.30
0.45
49.30
0.65
11.74
4.66
8.48
4.34
1.70
1.13
2.63
63.67
0.74
24.26
2.69
17.44
2.63
1.02
0.31
0.59
50.24
0.76
22.73
3.09
16.21
3.02
1.21
0.38
0.79
51.62
0.89
19.70
3.30
13.90
5.00
1.50
0.40
1.60
54.00
0.60
17.60
3.60
12.20
6.00
1.70
0.40
2.00
55.80
0.60
23.88
2.86
17.08
2.72
1.08
0.34
0.63
50.54
0.78
3 4 5 6 7 8 9
317
318 Combustion Engineering Issues for Solid Fuel Systems 100 90 80 FUELS 70
1 4
Weight %
60
5 6
50
7 8
40 100% Coal 30 Increasing % biofuel cofire 20 10 0 1,200
1,400
1,600
1,800
2,000
2,200
2,400
2,600
2,800
3,000
3,200
3,400
3,600
Temperature, F
FIGURE 8-22 Weight percent liquid phase for coal and coal-switchgrass/manure blends.
State by Folkedahl [75]. The model predicts viscosity as a function of temperature. Theoretically, the percentage of liquid/slag phase and its composition predicted by the FactSage program as a function of temperature would relate to the viscosity curves generated by the viscosity model. As a first approximation, the percent liquid phase present at a given temperature was calculated as a function of temperature. Selected fuels are shown in Figure 8-22 to demonstrate the trend in increasing weight percent of liquid phase with increasing percent of biomass cofire. In general, the primary solid phases include SiO2 (tridymite), Al6Si2O13 (mullite), Fe2O3 (hematite), CaAl2Si2O8 (anorthite), KAlSi2O6 (leucite), Mg2Al4Si5O18 (cordierite,) and Ca2SO4 (anhydrite). The 80% coal/20% manure blend (Fuel 4) had the greatest weight percent of liquid phase present at all temperatures. All the fuels except for Fuel 4 showed no liquid phase present at temperatures typical of FBC systems (1,450–1,750 F). Fuel 4 has a greater percentage of SiO2 than the other fuels but less Al2O3 and Fe2O3. However, Fuel 4 has the highest percentage of K2O, Na2O, and CaO (see Table 8-6). Fuel blends 5–8 represent successive additions of 10% thermal input (10–40%) of biomass to coal. Below 2,500 F, the coal (Fuel 1) has the lowest percent liquid phase. With increasing addition of switchgrass, the percent of liquid phase increases (Fuels 5–8). However, above 2,733 F the coal is calculated to have the highest percent of liquid phase, and the addition of
Fluidized-Bed Firing Systems
319
switchgrass results in a reduction in the percent liquid phase. At 3,400 F, all the fuels are predicted to be approximately 100% liquid. 8.7.3.1 Viscosity Results Selected viscosity curves generated by the GMDH model are shown in Figure 8-23. Table 8-7 contains the T250 taken from the calculated viscosity curves. The predicted T250 values for all the fuel blends fall below the recommended 2,600 F cutoff except for the coal T250, which exceeded the cutoff by only 47 F. Fuel 1 (100% coal) had the highest viscosity values predicted by the model at all temperatures. The coal/wood and coal/manure blends had similar viscosity curves. The similarity in the viscosity curves is supported by the similar percent liquid phase predicted by the FactSage model, with the exception of the coal/manure blend (Table 8-7). One would assume that the increased percentage of liquid phase in the coal/manure blend would result in a reduction in viscosity, but this is not the case. Fuel blends 5 and 6, in Figure 8-23, represent successive additions of 10% thermal input (10–20%) of biomass to coal (Fuel 1). The addition of 10% switchgrass to the coal results in a significant reduction in the viscosity at any given temperature, as well as its T250 (Fuels 5–8, Table 8-7). However,
10000 Fuel 1 (0) Fuel 5 (10) Fuel 6 (20) Fuel 4 (20 Manure)
Viscosity, poise
1000
100% Coal
100 Coal -Biomass Blends
10
1 1200 1400 1600 1800 2000 2200 2400 2600 2800 3000 3200 3400 3600 Temp F
FIGURE 8-23 Viscosity curves generated by the GMDH model for selected fuel blends (coalþswitchgrass, coalþmanure). The dashed line represents 250 poise. Numbers in parentheses represent percent of thermal input by biomass.
320 Combustion Engineering Issues for Solid Fuel Systems TABLE 8-7 Calculated T250 Values and Weight Percent Liquid at T250 Fuel 1 2 3 4 5 6 7 8 9
T250 ( F)
Wt% at T250
2,647 2,263 2,265 2,280 2,269 2,276 2,330 2,346 2,271
80.2 25.1 41.9 87.4 26.0 38.3 51.9 59.0 25.2
further addition of switchgrass results in a slight increase in T250 and the percent liquid phase. As mentioned previously, at temperatures greater than 2,733 F, the percent liquid phase is inversely related to the amount of switchgrass added to the coal. If the viscosity is directly related to the percent liquid phase, then viscosity should be inversely related to the percent switchgrass in a blend. Interestingly, the theoretical successive additions of switchgrass to the coal do not result in continued reduction in viscosity but rather a slight increase approaching a theoretical 100% switchgrass curve. The position of the coal, 100% switchgrass, and curves of Fuels 5–8 supports the fact that the physical characteristic of a slag is not the sum of its fuel components. One possibility would be the influence of the composition variation of the slag. As previously mentioned, highly siliceous melts generally have higher viscosity, which it reduces with the introduction of modifying alkalis into the melt. This relationship is not always linear.
8.7.4 Conclusions The biomass fuels presented demonstrate the impact that certain elements have on potential clinkering or fouling problems. The FactSage equilibrium calculations suggest that a cofire of biomass fuels with an appropriate nonfouling coal should not pose any problems in a CFB system given that the coal makes up a majority of the thermal input. Chicken litter was successfully fired in a fluidized-bed combustor at Penn State University only after the addition of kaolin clay reduced the presence of low melting phases in the bed. FactSage consistently predicted K2Si4O9(l) to be present at 1,171 K with biomass fuels having low aluminum levels and significant concentration of alkali and alkaline earth elements. Only 10% (normalized with respect to SiO2 and Al2O3) of K2O present in a system was enough to result in the formation of K2Si4O9(l) at equilibrium that could compromise a CFB system. Thermodynamically, it appears that the baseline cofire blend evaluated for the CFB boiler for cofiring
Fluidized-Bed Firing Systems
321
biomass and other wastes along with coal-based fuels is feasible and that there is flexibility in the biomass blends that can be handled. In general, the biomass fuels studied have a higher concentration of alkali metals and alkaline earth elements associated with water-soluble minerals and/or ion-exchangeable sites than most coals. Cofiring of biofuels having higher concentration and occurrence of these elements with coal can result in ash that has a higher propensity to form liquid phases as lower temperatures. This increases the risk of ash deposition and slagging in combustion systems. The melt phases also tend to have lower viscosity compared to highly siliceous compounds. However, it should be noted that the blending of fuels does not have a linear additive effect on melting points and viscosity. If the biomass fuels are cofired at <20% thermal input, then there does not seem to be an increased risk of ash deposition or slagging that would compromise a combustion system. For most of the blends at this level of cofire, the total aluminum and silicon content of the coal is not sufficiently diluted to suppress the melting point (i.e., eutectic) to a point that would be unacceptable in a fluidized-bed boiler.
8.8 FBC Boilers and Their Role in Clean Coal Technology Development Coal is the most abundant fossil energy resource worldwide and is found in every major region of the world except for the Middle East. Coal consumption is concentrated in the electricity generation sector with about 65% of the coal consumed worldwide used for producing electricity, with the industrial sector accounting for most of the remainder [76]. Coal-fired electricity generation accounts for 41% of the world electric supply. Power generation accounts for almost all the projected growth in coal consumption worldwide, which is projected to increase from 114.5 quadrillion Btu in 2004 to 199.1 quadrillion Btu in 2030. Although coal-fired power generation has remained well established in many regions, especially the United States, Canada, Australia, India, and China, and major growth in coal consumption is predicted, its future worldwide requires continuing improvements in its environmental performance, thermal efficiency, and economics [77, 78]. Consequently, there is much activity worldwide in developing clean coal technologies, one of which is FBC technology. In this section, a brief review of FBC technology’s role in clean coal development in the United States and worldwide is provided. A summary of FBC-related projects, funded in part by the U.S. Department of Energy (DOE), is included in Section 8.8.1 to illustrate the strategic program initiated by the United States to further development of FBC boilers. These activities, as well as those implemented in other countries, have been instrumental in developing next-generation FBC boilers. A summary of FBC boilers’ role in, and development needs for, clean coal technology worldwide is presented in Section 8.8.2. A summary of developments needed for further deployment of FBC
322 Combustion Engineering Issues for Solid Fuel Systems
systems as a clean coal technology is provided in Section 8.8.3. This section concludes with a discussion of a unique, niche opportunity for FBC boilers that has received increasing interest from food industry and government agencies in the United States.
8.8.1 United States Coal is recognized as an essential component in providing the United States with energy and economic stability and security to its citizens. Coal, which accounts for over 94% of the proven fossil energy reserves in the United States, supplies over 50% of the electricity [79] to the nation. To support continued domestic growth, demand for electricity is expected to increase by nearly 41% by 2030. Coal is expected to provide much of that generation, for reasons of energy security and economic stability, increasing to approximately 57%. The expanded use of coal is dependent on developing technological capabilities that address reduced emissions, improved efficiency, and favorable economics. This generation of technologies has been designated “clean coal technologies.” 8.8.1.1 Clean Coal Technology Development Program (CCTDP) The United States government, through the Department of Energy (DOE), recognized that moving clean coal technologies into the marketplace required a reduction in financial and technological risk to industry. To overcome these risks, the DOE initiated the Clean Coal Technology Demonstration Program (CCTDP) in 1985, which is a cost-shared initiative between the government and industry where industry must provide a minimum of 50% cost share [79]. The CCTDP is nearly concluded, with 38 projects funded for a total combined commitment by the federal government and private sector of $5.2 billion (all dollars are U.S. dollars unless stated otherwise). One aspect of the CCTDP was advanced power generation systems, which focused on enhancing power generation efficiency, producing nearly zero pollutant emissions, and providing hydrogen separation and carbon dioxide capture and sequestration. A comprehensive review of the program can be found elsewhere [10,79]. The emphasis of this program category included technologies that could effectively repower aging power plants faced with the need to both control emissions and respond to growing power demands. Repowering is an important option because existing power generation sites have significant value and warrant investment since the infrastructure is in place and siting powerplants represents a major undertaking. These advanced systems offer reductions in greenhouse gas emissions, SO2, NOx, and particulate emissions far below New Source Performance Standards, and salable solid and liquid byproducts. Of 11 projects funded, five included fluidized-bed combustion technologies. Following is a brief discussion of each.
Fluidized-Bed Firing Systems
323
City of Lakeland (Florida), Lakeland Electric was selected by the DOE for two CCTDP projects in 1989 and 1993; however, in 2003 these projects were terminated due to economic issues [80, 81]. The first project, a PFBC project, was to demonstrate Foster Wheeler Corporation’s PCFB technology coupled with Siemens Westinghouse’s ceramic candle-type, hot-gas cleanup system and power generation technologies, which were to represent a cost-effective, high-efficiency, low-emissions means of adding generating capacity at Greenfield sites or in repowering applications [82]. The second project, to be performed on the same boiler, was to demonstrate topped PCFB technology in a fully commercial power generation setting, thereby advancing the technology for future plants that will operate at higher gas turbine inlet temperatures and will be expected to achieve cycle efficiencies in excess of 45%. JEA (formerly Jacksonville [Florida] Electric Authority) is demonstrating CFB (atmospheric) combustion at a scale larger than previously operated. The objective of the project is to demonstrate CFB combustion, using a Foster Wheeler CFB boiler, rated at 297.5 MWe, which represented a scale larger than previously operated (when constructed in 1990). The objective was to verify expectations of the technology’s economic, environmental, and technical performance; to provide potential users with the data necessary for evaluating a large-scale CFB boiler as a commercial alternative; to accomplish greater than 90% SO2 removal; and to reduce NOx emissions by 60% when compared to conventional technology. The project was successful, it met its objectives, and the final report was prepared in June 2005. The Ohio Power Company performed a PFBC demonstration to verify expectations of PFBC economic, environmental, and technical performance in a combined-cycle repowering application at utility scale; to accomplish greater than 90% SO2 removal; and to achieve an NOx emission level of 0.3 lb/MM Btu at full load [83]. The demonstration was performed at the Ohio Power Company 70 MWe Tidd Plant (Brilliant, Ohio), Unit No. 1, and was the first large-scale operational PFBC system in the United States. It was successfully completed in March 1995 and, operationally, the PFBC boiler, provided by the Babcock & Wilcox Company, demonstrated commercial readiness and met its emissions performance targets by achieving SO2 removal efficiencies of 90–95% at full load with calcium-to-sulfur (Ca/S) molar ratios of 1:1.4 and 1:5, respectively; NOx emissions were 0.15 to 0.33 lb/ MM Btu; CO emissions were less than 0.1 lb/MM Btu; and particulate emissions were less than 0.02 lb/MM Btu. Tri-State Generation and Transmission Association, Inc., demonstrated the feasibility of CFBC technology at the utility scale and evaluated the economic, environmental, and operational performance at that scale [84]. Three small, coal-fired stoker boilers at the Nucla Station (Nucla, Colorado) were replaced with a new 110 MWe Foster Wheeler Energy Corporation CFBC boiler. Environmentally, SO2 capture efficiencies of 70% and 95% were
324 Combustion Engineering Issues for Solid Fuel Systems
achieved at Ca/S molar ratios of 1.5 and 4.0, respectively; NOx emissions averaged 0.18 lb/MM Btu; CO emissions ranged from 70–140 ppm; particulate emissions ranged from 0.0072–0.0125 lb/MM Btu (or 99.9% removal efficiency); and solid waste was essentially benign and showed potential as an agricultural solid amendment, soil/roadbed stabilizer, or landfill cap [82]. 8.8.1.2 Clean Coal Power Initiative A follow-up program to the CCTDP is the Clean Coal Power Initiative (CCPI), which was initiated in 2002 by President Bush to implement his National Energy Policy, and is a technology demonstration program that fosters more efficient clean coal technologies for use in existing and new power generation facilities in the United States. Like the CCTDP, it is a cost-shared partnership between government and industry aimed at accelerating commercial deployment of advanced technologies for clean, reliable, and affordable electricity. The 10-year initiative will be funded at a federal cost share estimated at $2 billion, with a matching industry cost share of at least 50%. The CCPI will be conducted over four solicitations (Rounds 1 through 4). As of March 31, 2005, five projects that were selected were awarded cooperative agreements, and one project was in negotiation. Total value of the six projects, one which involves CFBC technology, is $940 million, of which the DOE is cost-sharing $259 million. In October 2004, four projects were selected for demonstration; however, none involves FBC technology. The one FBC project, which is currently in its design phase, is being performed by Western Greenbrier Co-Generation, LLC (a public service entity formed to serve the interests of three municipalities in West Virginia—Rainelle, Rupert, and Quinwood). The powerplant will be an innovative Alstom Power, Inc., CFB boiler system utilizing bituminous coal waste and integrated with an advanced multipollutant control system for SOx, NOx, particulate matter, and mercury. An integrated coproduction facility will manufacture structural bricks using ash from the boiler and wood waste from an adjacent industrial process.
8.8.2 Worldwide Numerous countries worldwide, in addition to the United States, have developed clean coal programs and roadmaps because of the increased interest in and concern regarding future energy supplies and the potential environmental effects [85]. In each of these countries, the importance of coal as a major energy source is recognized, and steps are being taken to utilize this fuel source in an environmentally acceptable manner. The major countries with strategic programs in place include Australia, Canada, European Union (with Germany and the United Kingdom leading supporters), and Japan. This section will not discuss each in detail; rather, it will present a summary of FBC’s role in, and development needs for, clean coal technology.
Fluidized-Bed Firing Systems
325
In essentially all the programs, CFBC is a near-term commercial clean coal technology that will be further developed [85]. This technology is expected to focus on using low-cost and opportunity fuels, increasing unit size, and transitioning from supercritical to advanced supercritical and ultra-supercritical steam cycles. In Australia, activities include fluidized-bed development for clean power from lignite, and power systems evaluations for a combined gasification-PFBC system. FBC’s role in Canada is focusing on utilizing low-cost and opportunity fuels and transitioning from CFBC (today) to PFBC units (in 2020). In Germany, projects recommended for the medium term include PFBC systems with partial gasification and externally fired combined cycle. Japan is developing PFBC and advanced PFBC systems to use waste, biomass, and other solid fuels simultaneously with the goal of developing high-efficiency units for reducing CO2 emissions.
8.8.3 Further Developments Needed for Conventional Clean Coal Technologies CFBC technology has proven itself as a viable candidate for clean coal technology. However, in the short-to-medium term, reducing capital costs and improving efficiency and environmental performance will aid further deployment of established technologies including CFBC [85]. PFBC technology is not yet well established, and continuing evolution is not likely to occur without a significant installed base of reference operating plants. PFBC boilers no longer appear to offer sufficient gains in efficiency to justify their greater complexity compared with supercritical pulverized coal combustion with flue gas desulfurization. This, along with the lack of direct supply from the original developers of PFBC, is also a constraint on its further large-scale development and deployment. PFBC appears to be a niche technology for small plants burning low-grade coals or as the basis for some hybrid cycles [85].
8.9 Unique Opportunities for FBCs As discussed in the previous section, FBC technology is a key contributor in the implementation of clean coal strategies. As part of these strategies, PFBC has been identified as playing a role in providing niche opportunities, most likely over the medium to long term. There are near-term opportunities, though, for FBC technology to support niche markets that are not currently being addressed. One such potential application is the ability for FBC technology to provide security to a nation’s food supply, which is being explored in the United States but is relevant worldwide [10]. The unique opportunity is discussed in this section and is of interest to the food industry mainly because of an FBC boiler’s ability to accommodate low-grade fuels. The following discussion is a summary of work that was performed at Penn State, at the request
326 Combustion Engineering Issues for Solid Fuel Systems
of and with support from the food industry (i.e., National Cattlemen’s Beef Association [NCBA] and Cargill Meats Solutions) [23, 86–91], State government (i.e., Pennsylvania Energy Development Authority [PEDA]) [17], and federal government (i.e., U.S. Department of Energy [DOE]) [92].
8.9.1 Background of Opportunity/Food Industry Issue The United States agricultural sector, of which animal agriculture comprises a substantial portion, accounts for about 13% of the United States’ gross domestic product and nearly 17% of United States jobs [93]. The value of U.S. livestock commodities is enormous, amounting to $105 billion during 2003 with United States animal agriculture contributing approximately 26, 19, and 35 billion pounds of beef, pork, and poultry, respectively, to the food supply [93]. With the large animal agriculture industry come agricultural security problems, specifically carcass disposal. Typically, animal-production mortalities and natural disasters in the United States result in about four billion pounds of carcasses annually, of which approximately 2.5–2.7 billion pounds are cattle, nearly one million are swine, and the remainder sheep and poultry [93–95]. In the event of an accidental or intentional (through terroristic action) introduction of animal diseases, these quantities could escalate substantially. This is of special concern with concentrated animal production operations and mobile food-animal populations. In addition to carcass disposal, future Food and Drug Administration (FDA) regulations may require the removal of deadstock, nonambulatory cattle, and specified risk materials (SRMs) from being utilized in the production of rendered animal feed ingredients because of the concern of catastrophic outbreaks of bovine spongiform encephalopathy (BSE), commonly called “mad cow disease.” BSE is a prion disease in cattle that leads to neuron damage in the brain, resulting in brain degradation and death. BSE is caused by a conformation change in the PrP prion protein to the PrPsc prion protein. BSE infection occurs when a molecule of PrPsc enters a cow’s body and comes into contact with molecules of PrP. PrP is found throughout the cow’s body but occurs mainly in nerve cells and immune cell membranes, materials that are referred to as SRMs [96]. Therefore, infectivity is known to occur through ingestion or inhalation of PrPsc, peripheral exposure to a place such as damaged skin, or hereditary inheritance. Skin contact with an infected object is not believed to be a cause of infection [96]. Historically, byproducts from the meatpacking and rendering industries were processed by rendering into cattle feed, producing a protein (i.e., meat and bone meal [MBM]) or fat (i.e., tallow). In Europe, this practice has been halted because outbreaks of BSE have infected cattle populations around the world. Now, in Europe, most of these byproducts are incinerated, providing no additional value to the facility [97]. In the United States, impending legislation by the FDA may likely create a similar situation.
Fluidized-Bed Firing Systems
327
Research shows that although PrPsc is resistant to common decontamination practices including autoclaving, chemical solvents, and biological proteases, incineration is successful in destroying the infective protein in MBM [98]. No prior research has been conducted in studying the combustion of animal tissue that is potentially infected with BSE. If possible, it may be more efficient to burn tissue instead of first processing the carcass or SRMs into MBM or tallow prior to combustion. Also, animal tissue combustion may be useful if a future catastrophic BSE occurrence calls for the need for large-scale carcass disposal [99]. In the case of such an event, rapid decontamination is necessary to prevent spread of the infection because BSE transmission from carcasses has also been documented. In the United States, the estimated annual supply of fed-cattle (i.e., cattle less than 30 months in age) SRMs (which consist of the tonsils and a portion of the small intestine called the distal ileum; see Figure 8-24), cow carcasses from meatpacking facilities, cow carcasses from on-farm mortalities, and cull-cow (i.e., cattle older than 30 months in age) SRMs (which consist of the skull, brain, trigeminal ganglia, eyes, tonsils, spinal cord, spinal column, dorsal root ganglia, and distal ileum) is 850, 75, 2,500, and 650 million lb, respectively [97, 100–102]. With heating values ranging from 2,800–6,200 Btu/lb as fired, approximately 15 trillion Btu of energy are available for use as a potential FBC boiler fuel [93, 94, 103]. The heating value of carcasses and SRMs is illustrated in Figure 8-25 and
FIGURE 8-24 Specified risk material (Source: Modified from [100]).
328 Combustion Engineering Issues for Solid Fuel Systems 25,000
Natural Gas Fuel Oils Animal Fats/ Greases
Btu/lb (as-fired)
20,000
Coals
15,000
Animal Proteins 10,000
Coal Refuse
Woody Resides
Herbaceous Plants/ Crops Animal Tissue Manure/ Biomass Litter
5,000
R
N
at
ur
N o. al G 2 N F as es o. 6 uel C ta ho u Fu Oil ic ran el e W t G Oil hi re te as G e re as e La r Ta d Po llo ul w tr Bi y F a tu m t Su i bb A nou n itu s t m hra in ou cite s C Fe L oal at ign he ite r Bl Me oo al Po d M ea ul t l ry M ea P M t & or ea l k C oa Bon Me l R e al M C efu ea oa s l l R e-h ef igh us R e ai lro low W oo a d F d C ibe Tie hi p rb s Fe s/S oa d- ha rd C vi a n C ttle gs ul S l C -Co RM ow w C SR ar M D Po cas ai ry ultr ses /B y ee Li f M tte an r ur e R ee Sw Cor n itc d C an hgr ar as s y G ra ss
-
FIGURE 8-25 Comparison of as-fired energy densities of selected fuels evaluated at Penn State.
compared to that of various boiler feedstocks, both fossil fuels and biomass fuels, which have been tested at Penn State. Feedstocks with energy densities as low as 4,000 Btu/lb (e.g., poultry litter) are fired in boilers as sole fuels, and fuels with even lower energy densities are successfully cofired with coal [102].
8.9.2 Disposal Options The National Agricultural Biosecurity Center Consortium USDA APHIS (Animal and Plant Health Inspection Service) Cooperative Agreement Project, Carcass Disposal Working Group reviewed eight disposal options for carcasses/SRMs [93]. These included burial, incineration, composting, rendering, lactic acid fermentation, alkaline hydrolysis, anaerobic digestions, and nontraditional and novel technologies. Nontraditional and novel technologies included thermal depolymerization, plasma arc process, refeeding (e.g., to alligators), napalm, ocean disposal, nontraditional rendering, and novel pyrolysis technology. The review included discussions on the principles of operation of the disposal technologies, disease agent considerations, implications to the environment, advantages and disadvantages of the technologies, and cross-cutting and policy issues. The study did not review combustion in boilers, which was definitely an oversight. The U.S. Environmental Protection Agency (EPA) has subsequently noted that
Fluidized-Bed Firing Systems
329
there are several design options for industrial and utility boilers available offering significant capacity for utilizing carcasses and SRMs; boilers have good control over combustion processes; and they provide particulate matter and acid control [104]. Although the EPA identified disadvantages (in their opinion) of this technique, stating that “fluidized-bed boilers are not permitted to burn wastes, they lack suitable feeding equipment for animal-tissue biomass (ATB), and there will be permit issues dealing with biocontaminants,” no technical issues appear to be insurmountable. Handling the carcasses/SRMs is not considered problematic. Fluidized-bed combustors have historically utilized low-grade fuels such as paper mill sludge, coal-water mixtures/pastes, waste coal, and others. A variety of handling equipment is available for these fuels. Similarly, the rendering industry processes and handles carcasses and other animal-tissue biomass; consequently, the handling, grinding, and delivery of the ATB into a boiler are not considered difficult technical issues. Also, fluidized-bed boilers have been used to combust hazardous wastes, plastics, and waste oils with complete combustion achieved.
8.9.3 Cofiring ATB in Coal-Fired Boilers for Carcass Disposal The concept of cofiring ATB with coal in an FBC boiler was explored at Penn State. In July 2004, Penn State, Cargill Beef Solutions, and the McDonald’s Corporation hosted a workshop at Penn State to discuss and develop strategies to utilize ATB as a fuel on both a routine and large-scale emergency basis, and stimulate public-private collaboration [87]. Representatives included the food industry, BSE experts, the rendering industry, the boiler manufacturing industry, federal government agencies (energy, agriculture, and environment), state government agencies (agriculture), the cogeneration industry, the cement industry, equipment vendors, banking, and academia. A major highlight of the meeting was that the boiler vendors informed the attendees that the concept of cofiring ATB with coal was technically sound. One of the key points developed during the workshop was the need to perform pilot-scale testing to demonstrate to regulatory agencies, the U.S. Department of Agriculture, the FDA, and industry the technical viability of this option. Consequently, three test programs were conducted for NCBA/Cargill Food Solutions, PEDA/Cargill Food Solutions, and the DOE.
8.9.4 Summary of ATB/Coal Cofiring in a Pilot-Scale Fluidized-Bed Combustor Tests were performed in a pilot-scale bubbling fluidized-bed combustor to evaluate the technical feasibility of cofiring ATB with coal. A total of 58 tests were performed in the three projects, which are briefly discussed in the following sections, in which the following parameters were investigated:
330 Combustion Engineering Issues for Solid Fuel Systems
Three types of ATB (discussed previously); With and without flue gas recirculation; Varying ratios of ATB-to-coal based on thermal input; Three feed locations—one in-bed, two locations above-bed; Varying oxygen concentration; With and without the use of overfire air; Two coal particle size distributions; and Oxygen injection into the bed and at two locations in the freeboard.
8.9.4.1 NCBA/Cargill Food Solutions Tests The analysis of the ATB used in the testing is provided in Table 8-8. A summary of average operating conditions and emissions for selected tests is provided in Table 8-9. The testing, which was composed of 35 tests investigating the effect of fuel type, feed location, and flue gas recirculation (FGR; with and without) on combustion efficiency, successfully demonstrated that carcasses and SRMs can be cofired with coal in a BFB combustor. Feeding ATB into the fluidized-bed combustor did, however, present several challenges. Specifically, handling/feeding issues resulting from the small scale of the equipment and the extremely heterogeneous nature of the ATB were encountered during the testing. Through statistical analysis, it was shown that the ATB feed location had a greater effect on CO emissions (which, along with hydrocarbon emissions, were used as an indication of the extent of combustion efficiency), which were used as an indication of combustion performance, than the fuel type due to feeding difficulties. Baseline coal tests,
TABLE 8-8 Proximate, Ultimate, and Heating Value Analyses of ATB Samples and Coal Used in the Pilot-Scale Testing Volatile Fixed Moisture (as Matter Carbon Ash (dry (dry (dry determined weight %) weight %) weight %) weight %) Cull-Cow Carcasses (ATB1) Cull-Cattle SRMs (ATB2) Fed-Cattle SRMs (ATB3) Coal
31.6 44.8 57.3 1.2 C (dry weight %)
Cull-Cow Carcasses (ATB1) Cull-Cattle SRMs (ATB2) Fed-Cattle SRMs (ATB3) Coal
40.8 33.3 30.3 81.3
81.6 80.3 96.9 32.5
2.2 1.6 1.3 62.5
16.2 18.1 1.8 5.0
HV (as-fired Btu/lb) 3,841 2,425 3,893 13,897
H (dry N (dry S (dry O (dry weight %) weight %) weight %) weight %) 12.8 13.1 7.9 5.9
0.3 1.4 1.8 1.4
0.1 0.1 0.1 0.7
26.4 32.9 58.1 5.8
TABLE 8-9 Summary of Average Operating Conditions and Emissions for Selected Tests
Fuel Firing Rate: Btu/h coal Btu/h ATB ATB Feed Location Main Bed Velocity (ft/s): Temperatures (oF): T2: Upper Bed T4: Middle Freeboard T5: Upper Freeboard Emissions: O2 (%) CO (ppm) @ 3% O2 CO2 (%) @ 3% O2 SO2 (ppm) @ 3% O2 NOx (ppm) @ 3% O2 HCc (ppm) @ 3% O2 a
Flue gas recirculation N/A - not applicable c HC: Total Hydrocarbons b
100% Coal
100% Coal 18% FGRa
100% Coal 30% FGR
87% Coal 13% ATB1
87% Coal 13% ATB1 27% FGR
83% Coal 17% ATB1 23% FGR
417,000 N/Ab N/A
333,600 N/A N/A
250,290 N/A N/A
229,396 31,147 In-bed
258,535 38,152 Above-bed
246,779 51,882 Above-bed
6.3
6.3
6.1
6.5
6.6
1,597 1,269 1,201
1,681 1,366 1,290
1,633 1,367 1,295
1,690 1,314 1,233
10.3 1,689 15.9 537 439 50
8.4 1,344 16.6 507 411 35
7.8 1,415 16.3 492 362 49
10.1 1,364 15.8 471 493 74
85% Coal 15% ATB2
90% Coal 10% ATB2 25% FGR
230,154 24,710 Above-bed
6.6
222,563 37,990 Abovebed 6.6
1,663 1,600 1,532
1,716 1,598 1,503
3.5 1,656 15.3 482 311 160
4.0 2,570 14.9 420 342 207
78% Coal 22% ATB3
82% Coal 18% ATB3 24% FGR
242,789 54,899 Above-bed
6.5
196,849 56,744 Abovebed 6.7
1,678 1,439 1,350
1,702 1,551 1,472
1,688 1,521 1,415
1,696 1,627 1,532
9.7 1,952 15.4 475 508 184
4.6 2,098 15.3 327 353 221
8.7 2,912 15.1 545 466 281
2.8 2,607 14.9 465 307 338
6.6
331
332 Combustion Engineering Issues for Solid Fuel Systems
using a Middle Kittanning seam high-volatile bituminous coal, and tests cofiring ATB when injected into the bed were statistically indistinguishable. Fuel feeding issues would not be expected at the full scale since full-scale units routinely handle low-quality fuels. 8.9.4.2 PEDA/Cargill Food Solutions Tests In the second series of testing, which was composed of six tests, the three ATB samples were fed into the combustor at an optimized overbed feed location (compared to the first series of testing), with FGR and overfire air, and using a different coal particle size distribution (coarser than the first series of testing). The ATB feed location was moved from about 2 feet above the bed (in a combustor with a total height of 10 feet) to just above the dense bed. The increase in residence time along with the injection of overfire was used to reduce CO emissions. Bench-scale fluidized-bed combustion testing firing coal with MBM has shown that CO emissions increased as the rate of MBM input increased [105]. This was attributed to the volatile MBM-derived hydrocarbon concentration (from the fats) that suppresses the further oxidation of CO that normally occurs with oxygenated free radicals (i.e., OH and HO2). These radicals react easier with hydrocarbons than with CO; therefore, CO oxidation is restricted when volatile hydrocarbons are emitted. The bench-scale testing did not use overfire air (which is similar to the NCBA/Cargill tests). The use of overfire air was expected to minimize this effect and allow further oxidation of the CO. An improvement in combustion was observed in the second series of tests, based on CO and hydrocarbon emissions, which is illustrated in Table 8-10. 8.9.4.3 DOE Oxygen-Enhanced Combustion Testing In the final series of testing, which was composed of 17 tests, the effect of two ATB types (fed-cow SRMs and cow carcasses) and the use of oxygenTABLE 8-10 Comparison of Results Between the NCBA/Cargill Tests and the PEDA/Cargill Tests NCBA/Cargill
Coal Cull-Cow carcasses Coal Cull-Cow SRMs Coal Fed-Cattle SRMs a
a
PEDA/Cargill
CO
HCs
CO
HCsb
755 2570 798 2098 830 2607
56 207 28 221 41 338
632 1167 547 1618c 580 933
5 2 <2 532 <2 <2
Online hydrocarbon analyzer Bag samples c Greater quantity of larger hydrocarbon molecules; more ATB feed problems were encountered b
Fluidized-Bed Firing Systems
333
enhanced combustion on combustion efficiency were evaluated. Oxygen was injected in the bed (with the main combustion air/recirculated flue gas mixture) or into the middle of the freeboard. Approximately 10% of the air was replaced by oxygen in the tests. No improvement in combustion efficiency was noted during these tests.
8.9.5 Closing Statements Because of the advantages that FBC technology affords, niche opportunities are likely to appear in the near and medium term. One of these is potentially the cofiring of animal-tissue biomass with coal as a method to provide food security worldwide. Currently, this concept is being explored in the United States but is applicable in any meat-producing country. Activities to date have demonstrated that it is technically feasible to successfully cofire ATB with coal in a BFBC system. Testing to date has not investigated the deactivation of the PrPsc prion protein because this type of research can be conducted only in a few select laboratories. The next steps in the implementation of this technology are to demonstrate the cofiring concept at the demonstration scale (using noncontaminated ATB) and determine if the PrPsc prion protein is deactivated under FBC conditions (i.e., temperature and residence time). Tests performed to date in the laboratory have shown that FBC temperatures appear to be adequate (as the PrPsc prion protein has been shown to be deactivated at temperatures lower than FBC temperatures); however, tests to date have been conducted with residence times of minutes (i.e., 15 minutes) when residence times in FBC boilers can be seconds for small particles and minutes for larger particles.
8.10 References 1. Koornmeef, J., M. Junginger, and A. Faaij. 2007. Development of Fluidized Bed Combustion—An Overview of Trends, Performance, and Cost. Progress in Energy and Combustion Science. 33: 19–55. 2. Elliot, M.A. (ed). 1981. Chemistry of Coal Utilization, secondary suppl. New York: John Wiley & Sons. 3. Schobert, H.H. 1987. Coal: The Energy Source of the Past and Future. Washington, D.C.: American Chemical Society. 4. Department of Energy. 2007. Fluidized-Bed Combustion. Washington, D.C.: U.S. Department of Energy. www.netl.doe.gov/technologies/coalpower/ advresearch/combustion/FBC/fbc-overview.html (accessed July 16, 2007). 5. Tang, J.T., and F. Engstrom. 1987. Technical Assessment on the Ahlstrom Pyroflow Circulating and Conventional Bubbling Fluidized-Bed Combustion Systems. Proc. of the International Conference on Fluidized-Bed Combustion. p. 37. May 3–7.
334 Combustion Engineering Issues for Solid Fuel Systems 6. Wu, Z. 2003. Understanding Fluidized-Bed Combustion. London: IEA Coal Research. 7. Anon. 1988. Boilers and Auxiliary Equipment. Power, Special Edition. (June). 8. Elliot, T.C. (ed). 1989. Standard Handbook of Powerplant Engineering. New York: McGraw-Hill. 9. Raskin, N. 2004. Overview of Circulating Fluidized-Bed Boiler Technology. Workshop on the Utilization of Animal Tissue Biomass in Boilers and Other Industrial Processes. University Park, PA. July 22. 10. Miller, B.G. 2005. Coal Energy Systems. United Kingdom: Elsevier. 11. Gaglia, B.N., and A. Hall. 1987. Comparison of Bubbling and Circulating Fluidized-Bed Industrial Steam Generation. Proc. of the International Conference on Fluidized-Bed Combustion. May 3–7. 12. Goidich, S.J., S. Wu, Z. Fan, and A.C. Bose. 2005. Design Aspects of the UltraSupercritical CFB Boiler. Proc. of the 22nd International Pittsburgh Coal Conference. University of Pittsburgh. September 12–15. 13. U.S. Department of Energy and American Electric Power Service Corporation. 1990. Tidd: The Nation’s First PFBC Combined-Cycle Demonstration, Clean Coal Technology Topical Report Number 1. (March). 14. Stultz, S.C., and J.B. Kitto (eds). 1992. Steam—Its Generation and Use. Barberton, OH: Babcock & Wilcox. 15. Robertson, A., Z. Fan, D. Horaazak, R. Newby, H. Goldstein, and A.C. Bose, 2005. 2nd Generation PFB Plant with Supercritical Pressure Steam Turbine. Proc. of the 22nd International Pittsburgh Coal Conference. University of Pittsburgh. September 12–15. 16. Foster Wheeler. 2006. Biomass Circulating Fluid Bed Boilers, Promotional Presentation Module 38 Biomass CFB 071106. 17. Miller, B.G., S. Falcone Miller, R.T. Wincek, and R.S. Wasco. 2006. Fuel Flexibility in Boilers for Fuel Cost Reduction and Enhanced Food Supply Security Final Report. Prepared for Pennsylvania Energy Development Authority, Agreement No. PG050020. June 30. 509 pp. 18. Rostam-Abadi, M., R. Harvey, Y. Lu, B.G. Miller, S. Falcone Miller, and R.T. Wincek. 2005. Combustion of Sewage Sludge and Coal Fines in a Pilot-Scale Circulating Fluidized-Bed Combustor. Proc. of Electric Power 2005, 7th Annual Conference and Exhibit. Chicago, IL. April 5–7. 19. Miller, B.G., S. Falcone Miller, and R.T. Wincek. 2004. Combustion of Sewage Sludge and Waste Coal in a Pilot-Scale Circulating Fluidized-Bed Combustor. Prepared for Illinois State Geological Survey. March 9. 40 pp. 20. Miller, B.G., S. Falcone Miller, R. Cooper, J. Gaudlip, M. Lapinsky, N. Raskin, T. Steitz, and J.J. Battista. 2003. Feasibility Analysis for Installing a Circulating Fluidized Bed Boiler for Cofiring Multiple Biofuels and Other Wastes with Coal at Penn State University Final Report. Prepared for U.S. Department of Energy, National Energy Technology Laboratory, Pittsburgh, PA. March 24. DE-FG26-00NT40809, 181 pp. 21. Miller, B.G., S. Falcone Miller, R.E. Cooper, Jr., N. Raskin, and J.J. Battista. 2002. A Feasibility Study for Cofiring Agricultural and Other Wastes with
Fluidized-Bed Firing Systems
22. 23.
24. 25. 26. 27.
28.
29.
30.
31.
32.
33.
34.
335
Coal at Penn State University. Proc. of the Nineteenth Annual International Pittsburgh Coal Conference, Coal—Energy and the Environment. Pittsburgh, PA. September 23–27. Tillman, D.A., and N.S. Harding. 2004. Fuels of Opportunity: Characteristics and Uses in Combustion Systems. United Kingdom: Elsevier. Miller, B.G., S. Falcone Miller, E.M. Fedorowicz, D.W. Harlan, L.A. Detwiler, and M.L. Rossman. 2006. Pilot-Scale Fluidized-Bed Combustor Testing Cofiring Animal-Tissue Biomass with Coal as a Carcass Disposal Option. Energy & Fuels, 20: 1828–1835. Foster Wheeler. 2007. Utility CFB Steam Generators, Promotional Presentation Module 31 Utility CFB Detail 042707. Haji-Sulaiman, M.Z. 1988. Sorbent Performance in Fluidized Bed Combustion, Ph.D. Thesis. The Pennsylvania State University. May. Rickman, W.S. 1986. Modeling Sulfur Capture by Recycle Fines in FluidizedBed Combustion, EPRI CS-4442, Final Report. (March), pp. 2–3. Miller, B.G., D.E. Romans, and A.W. Scaroni. 1990. Characterization of Limestones for FBC Systems. Proc. of The National Stone Association SO2 Emission Control Conference. Pittsburgh, PA. September 9–11. Morrison, J.L., B.G. Miller, D.E. Romans, S.V. Pisupati, and A.W. Scaroni. 1993. Fluidized-Bed Boilers—SO2 Capture Aspects. Proc. of SO2 Capture Seminar “Sorbent Options and Considerations.” Cincinnati, OH. September 19–21. Christman, P.G., and T.G. Edgar. 1982. The Effect of Temperature on the Sulfation of Limestones. Proc. of 1982 American Institute of Chemical Engineers Meeting. Los Angeles, CA. Dennis, J.S., and A.N. Hayhurst. 1984. The Effect of Temperature on the Kinetics and Extent of SO2 Uptake by Calcarious Materials During FBC of Coal. Proc. of the 20th Symp. (Int.) on Combustion, The Combustion Institute. pp. 1347–1305. Hansen, P.F.G., K.D. Johansen, L.H. Bank, and K. Ostergaard. 1991. Sulfur Retention on Limestone Under Fluidized-Bed Combustion Conditions: An Experimental Study. Proc. of the 11th International Conference on FBC. Montreal, Ontario, Canada. April 21–24. pp. 73–81. Pisupati, S.V., J.L. Morrison, D.E. Romans, B.G. Miller, and A.W. Scaroni. 1993. Importance of Calcium Carbonate Content on the Sulfur Capture Performance of Naturally-Occurring Sorbens in a 30 MWe Circulation Fluidized-Bed Plant. Proc. of the 12th International FBC Conference. La Jolla, CA. May 8–13. Romans, D.E., A.W. Scaroni, and B.G. Miller. 1991. Evaluation of Sorbents for Capturing SO2 in Fluidized-Bed Combustion Systems. Proc. of the 14th Annual Energy-Sources Technology Conference & Exhibition, American Society of Mechanical Engineers. Dallas, TX. January 22. Boynton, R.S. 1980. Chemistry and Technology of Lime and Limestone. New York: John Wiley & Sons Publishing.
336 Combustion Engineering Issues for Solid Fuel Systems 35. Fee, D.C., et al. 1982. Sulfur Control in Fluidized-Bed Combustors: Methodology for Predicting the Performance of Limestone and Dolomite Sorbents, ANL/FE-80-10. 36. Newton, G.H., S.L. Chen, and J.C. Kramlich. 1989. Role of Porosity Loss in Limiting SO2 Capture by Calcium Based Sorbents. AIChE Journal. 35(6): 988–994. 37. Bhatia, S.K., and D.D. Perlmutter. 1981. The Effect of Pore Structure on FluidSolid Reactions: Applications to the SO2-Lime Reaction. AIChE Journal. 27(2): 226–234. 38. Simons, G.A., and A.R. Garmen. 1986. Small Pore Closure and the Deactivation of the Limestone Sulfation Reaction. AIChE Journal. 32(9): 1491–1499. 39. Simons, G.A. 1988. Parameters Limiting Sulfation By CaO. AIChE Journal. 34(1): 167–170. 40. Ulerich, M.E., E.P. O’Neill, and D.L. Keairns. 1978. Thermogravimetric Study of the Effect of Pore Volume-Pore Size Distribution on the Sulfation of Calcined Limestone. Thermochimica Acta. 26: 269–282. 41. Borgwardt, R.H., N.F. Roache, and K.R. Bruce. 1989. Surface Area of Calcium Oxide and Kinetics of Calcium Sulfide Formation. Environmental Progress. 3(2): 129–135. 42. Hiltunen, M., et al. 1991. N2O Emissions from CFB Boilers: Experimental Results and Chemical Interpretation. Proc. 11th Conference on FluidizedBed Combustion. Montreal, Ontario, Canada. April 21–24. 43. Hiltunen, M. 2006. Nitrous Oxide (N2O) Emissions in Fluidized-Bed Boilers. ECCP I Review: Non-CO2 Gases. Brussells, Belgium. January 30. 44. Davis, W.T. (ed). 2000. Air Pollution Engineering Manual, 2nd ed. New York: John Wiley & Sons, p. 9. 45. Clarke, L.E., and L.E. Sloss. 1992. Trace Elements: Emissions from Coal Combustion and Gasification. London: IEA Coal Research. 46. Miller, S.J., S.R. Ness, G.F. Weber, T.A. Erickson, D.J. Hassett, S.B. Hawthorne, K.A. Katrinak, and P.K.K. Louie. 1996. A Comprehensive Assessment of Toxic Emissions from Coal-Fired Power Plants: Phase I Results from the U.S. Department of Energy Study Final Report. Contract No. DE-FC2193MC30097 (Subtask 2.3.3). 47. Standards of Performance for New and Existing Stationary Sources: Electric Utility Steam Generating Unites, amended. 2005. Code of Federal Regulations, Part 60, 63, 72, and 75, Title 40. 48. Hughes, I.S.C., and R.F. Littlejohn. 1987. Emissions of Trace Elements Emissions from AFBC. Proc. of the International Conference on Fluidized-Bed Combustion: FBC Comes of Age, American Society of Mechanical Engineers. New York May 3. 49. Miller, B.G., S. Falcone Miller, R.T. Wincek, and A.W. Scaroni. 1999. A Demonstration of Fine Particulate and Mercury Removal in a Coal-Fired Industrial Boiler Using Ceramic Membrane Filters and Conventional Fabric Filters. Proc. EPRI-DOE-EPA Combined Utility Air Pollutant Symposium. Atlanta, GA. August 16–20.
Fluidized-Bed Firing Systems
337
50. www.epa.gov/ttn/atw/combust/utiltox/mercury.html 51. Miller, B.G., S. Falcone Miller, R.E. Cooper, Jr., N. Raskin, and J.J. Battista. 2002. Biomass Cofiring: A Feasibility Study for Cofiring Agricultural and Other Wastes with Coal at Penn State University. Proc. of the 27th International Technical Conference on Coal Utilization & Fuel Systems. Clearwater, FL. March 4–7. 52. Falcone Miller, S., B.G. Miller, and D. Tillman. 2002. The Propensity of Liquid Phases Forming During Coal-Opportunity Fuel (Biomass) Cofiring as a Function of Ash Chemistry and Temperature. Proc. of the 27th International Technical Conference on Coal Utilization & Fuel Systems. Clearwater, FL. March 4–7. 53. Falcone Miller, S., B.G. Miller, and C.M. Jawdy. 2002. The Occurrence of Inorganic Elements in Various Biofuels and Its Effect on the Formation of Melt Phases During Combustion. Proc. of the 2002 International Joint Power Generation Conference. Phoenix, AZ. June 24–26. 54. Falcone Miller, S., and B.G. Miller. 2004. The Effect of Cofiring Coal and Biomass on Utilization of Coal Combustion Products: The U.S. Perspective. Proc. of the 11th International Conference on Ashes from Power Generation. Zakopane, Poland. October 13–16. 55. Falcone Miller, S., and B.G. Miller. 2007. The Occurrence of Inorganic Elements in Various Biofuels and Its Effect on Ash Chemistry and Behavior and Use in Combustion Products. Fuel Processing Technology. 88(11): 1155–1164. 56. Watt, J.D. 1969. The Physical and Chemical Behaviour of the Mineral Matter in Coal Under the Conditions Met in Combustion Plant, Part II, BCUR, Leatherhead. 57. Raask, E. 1985. Mineral Impurities in Coal Combustion. Washington, D.C.: Hemisphere Publishing. 58. Falcone, S.K., and H.H. Schobert. 1986. Mineral Transformations During Ashing of Selected Low-Rank Coals. In Mineral Matter and Ash in Coal, (K. S. Vorres, ed.) ACS Symposium Series No. 301. Washington, D.C.: American Chemical Society. pp. 114–127. 59. Schobert, H.H. 1995. Lignites of North America. Coal Science and Technology, vol. 23, chapter 5. Amsterdam: Elsevier. 60. Vorres, K.S., S. Greenburg, and R.B. Poeppel. 1985. Viscosity of Synthetic Coal Ash Slags, Argonne National Laboratory Report ANL/FE-85-10. 61. Jung, B. 1990. Viscous Sintering of Coal Ashes in Combustion Systems. Ph.D. Thesis in Fuel Science. The Pennsylvania State University. 62. Benson, S.A., and P. Holms. 1985. Comparison of Inorganic Constituents in Three Low-Rank Coals. Ind Chem Eng Prod Res Dev. 24:145–149. 63. Baxter, L.L. 1994. Pollutant Emission and Deposit Formation During Combustion of Biomass Fuel. Livermore, California: Sandia National Laboratories. Quarterly Report to National Renewable Energly Laboratory.
338 Combustion Engineering Issues for Solid Fuel Systems 64. von Puttkamer, T., S. Unterber, and K.R.G. Hein. 2000. Round Robin on Biomass Fuels. ACS National Meeting, Division of Fuel Chemistry. 45(3): Washington, D.C. 65. Zevenhoven-Onderwater, M., J.P. Blomquist, B.J. Skrifvars, R. Backman, and M. Hupa. 2000. The Prediction of Behavior of Ashes from Five Different Solid Fuels in Fluidized-Bed Combustion. Fuel. 79: 1353–1361. 66. FACTSage 5.0, Developed at the Facility for the Analysis of Chemical Thermodynamics (FACT), Centre for Research in Computational Thermochemistry (CRCT), E´cole Polytechnique de Montre´al, Canada, and GTT Technologies, Herzogenrath, Germany. Released March 2001. 67. Jawdy, C.M., S. Falcone Miller, and B.G. Miller. 2000. Chicken Litter Combustion Analysis. Unpublished internal report. 68. Virr, M.J. 2001. The Development of a Modular System to Burn Farm Animal Waste to Generate Heat and Power. Proc. of the 16th International Conference on Fluidized Bed Combustion, Reno, NV. May 13–16. 69. Hald, P. 1995. The Behavior of Alkali Metals in Biomass Conversion Systems. ACS National Meeting, Division of Fuel Chemistry. 40(3). 70. Helble, J.J., S. Srinivasachar, A.A. Boni, L.E. Bool, N.B. Gallagher, T.W. Peterson, J.O.L. Wendt, F.E. Huggins, N. Shah, G.P. Huffman, K.A. Graham, A.F. Sarofim, and J.M. Beer. 1991. Mechanisms of Ash Evolution—A Fundamental Study Part II: Bituminous Coals and the Role of Iron and Potassium. Proc. of the Engineering Foundation Conference on Inorganic Transformations and Ash Deposition During Combustion. Palm Coast, FL. March 10–15. 71. Helble, J.J., S. Srinivasachar, A.A. Boni, S.G. Kang, K.A. Graham, A.F. Sarofin, J.M. Beer, N.B. Gallagher, L.E. Bool, T.W. Peterson, J.O.L. Wendt, N. Shah, F.E. Huggins, and G.P. Huffman. 1991. Mechanisms of Ash Evolution—A Fundamental Study Part I: Low-Rank Coals and the Role of Calcium. Proc. of the Engineering Foundation Conference on Inorganic Transformations and Ash Deposition During Combustion. Palm Coast, FL. March 10–15. 72. Neville, M., and A.F. Sarofim. 1985. The Fate of Sodium During Pulverized Coal Combustion. Fuel. 64: 384–390. 73. Gallagher, N.B., T.W. Peterson, and J.O.L. Wendt. 1991. Alkali/Silicate Interactions During Pulverized Coal Combustion. ACS Division of Fuel Chemistry Preprints. 36(1): 181–190. 74. Baxter, L.L., and B.M. Jenkins. 1995. Laboratory Illustrations of the Transformations and Deposition of Inorganic Material in Biomass Boilers. 210th ACS National Meeting, Fuel Science Division. Chicago, IL. March 20–24. 75. Folkedahl, B.C. 1997. A Study of the Viscosity of Coal Ash and Slag. Ph.D. Thesis, Penn State University. 76. EIA. 2007. International Energy Outlook 2007. Washington, D.C.: U.S. Department of Energy, Energy Information Administration. May. 77. EIA. 2007. Annual Energy Outlook 2007. Washington, D.C.: U.S. Department of Energy, Energy Information Administration. February. 78. Henderson, C. 2003. Clean Coal Technologies. London: IEA Coal Research.
Fluidized-Bed Firing Systems
339
79. Department of Energy. 2006. Clean Coal Technology Programs: Program Update 2006. Washington, D.C.: Office of Fossil Energy, U.S. Department of Energy. September. 80. Department of Energy. 2003.DOE/FE-0215P-54 Clean Coal Today. Washington, D.C.: Office of Fossil Energy, U.S. Department of Energy. 54, Summer. 81. Raskin, N. 2004. Foster Wheeler Power Group, Inc., personal communications, January 8. 82. Department of Energy. 2002. Clean Coal Technology Demonstration Program: Program Update 2001 Including Power Plant Improvement Initiative Projects. Washington, D.C.: Office of Fossil Energy, U.S. Department of Energy. July. 83. Ohio Power Company. 1995. Tidd PFBC Demonstration Project: Final Report, DE-FC21-87MC24132, August. 84. Colorado-Ute Electric Association, Inc. 1991. NUCLA Circulating Atmospheric Fluidized-Bed Demonstration Project: Final Report, DE-FC21-89MC25137. October. 85. Henderson, C. 2003. Clean Coal Technologies Roadmaps. London: IEA Coal Research. 86. Miller, B.G., S. Falcone Miller, and D.W. Harlan. 2004. Utilizing AnimalTissue Biomass in Coal-Fired Boilers: An Option for Providing Biosecurity. 29th International Technical Conference on Coal Utilization & Fuel Systems. Clearwater, FL. April 18–22. 87. Miller, B.G., D.W. Harlan, L.A. Detwiler, and S. Falcone Miller. 2005. Utilizing Animal Tissue Biomass in Coal-Fired Boilers: A Step Closer Towards Implementation? 30th International Technical Conference on Coal Utilization & Fuel Systems. Clearwater, FL. April 17–21. 88. Miller, B.G., S. Falcone Miller, R.S. Wasco, R.T. Wincek, and D.J. Clifford. 2005. Demonstrate the Technical Feasibility of Cofiring Animal-Tissue Biomass (SRMs and Carcasses) with Coal in a Pilot-Scale Bubbling FluidizedBed Combustor. Prepared for the National Cattlemen’s Beef Association. May 31. pp. 216. 89. Miller, B.G., L.A. Detwiler, S. Falcone Miller, D.W. Harlan, and M.L. Rossman. 2006. An Overview of Activities at Penn State Evaluating the Concept of Utilizing Coal-Fired Boilers for Carcass Disposal. Council of Industrial Boiler Owners Fluidized-Bed XIX Conference. Hershey, PA. May 8–10. 90. Miller, B.G., S. Falcone Miller, R.T. Wincek, R.S. Wasco, D.W. Harlan, L.A. Detwiler, and M.L. Rossman. 2006. Cofiring Animal-Tissue Biomass in Coal-Fired Boilers to Dispose of Specified Risk Materials and Carcasses. The 31st International Technical Conference on Coal Utilization & Fuel Systems. Clearwater, FL. May 21–25. 91. Miller, B.G., S. Falcone Miller, D.W. Harlan, L.A. Detwiler, and M.L. Rossman. 2006. Cofiring Animal-Tissue Biomass in Coal-Fired Boilers to Dispose of Specified Risk Materials and Carcasses: An Overview of an University/Industry Collaboration. National Carcass Disposal Symposium 2006. Beltsville, MD. December 4–7.
340 Combustion Engineering Issues for Solid Fuel Systems 92. Miller, B.G., S. Falcone Miller, R.S. Wasco, and R.T. Wincek. 2007. OxygenEnhanced Combustion of Animal-Tissue Biomass Cofired in a Pilot-Scale Coal-Fired Fluidized-Bed Combustor. Prepared for the U.S. Department of Energy, National Energy Technology Laboratory, Agreement No. DE-AP2605NT53791. April 30. 93. National Agricultural Biosecurity Center Consortium, Carcass Disposal Working Group. 2004. Carcass Disposal: A Comprehensive Review. National Agricultural Biosecurity Center, Kansas State University. August. 94. Hamilton, R. 2004. Review of On-Farm Mortalities. Presented at the Workshop on the Utilization of Animal Tissue Biomass in Boilers and Other Industrial Processes. University Park, PA. July 22. 95. Informa Economics. 2004. An Economic and Environmental Assessment of Eliminating Specified Risk Materials and Cattle Mortalities from Existing Markets. Prepared for the National Renderers Association. August. 96. Novakofski, J., et al. 2005. Prion Biology Relevant to Bovine Spongiform Encephalopathy. J. Animal Science. 83(6): 1455–1776. 97. Nottrodt, -Ing. A. 2001. Technical Requirements and General Recommendations for the Disposal of Meat and Bone Meal and Tallow. Report for the German Federal Ministry for Environment, Nature Protection, and Reactor Safety. February 23. 98. Brown, P., et al. 2000. New Studies on the Heat Resistance of HamsterAdapted Scrapie Agent: Threshold Survival after Ashing at 600 C Suggests an Inorganic Template of Replication. Proc. National Academy of Science. 97(7): 3418–3421. 99. Detwiler, L. 2004. Overview of Animal Tissue Disposal Issues. Presented at the Workshop on the Utilization of Animal Tissue Biomass in Boilers and Other Industrial Processes. University Park, PA. July 22. 100. Alberta Agriculture. 2005. Specified Risk Material. Alberta Agriculture, Food and Rural Development, Regulatory Services Branch. 101. United States Department of Agriculture. FSIS. http://www.fsis.usda.gov/ OPPDE/rdad/FRPPubs/03-025IF.htm. 102. Miller, B.G., and S. Falcone Miller. 2003. Utilizing Biomass in Industrial Boilers: The Role of Biomass and Industrial Boilers in Providing Energy/ National Security. The First CIBO Industrial Renewable Energy & Biomass Conference. Minneapolis, MN. April 7–9. 103. Harlan, D.H. 2004. Review of Packer Issues. Presented at the Workshop on the Utilization of Animal Tissue Biomass in Boilers and Other Industrial Processes. University Park, PA. July 22. 104. Kremer, F., and P. Lemieux. Treatment/Disposal of TSE Contaminated Wastes. Presented at the Workshop on the Utilization of Animal Tissue Biomass in Boilers and Other Industrial Processes. University Park, PA. July 22. 105. Fryda, L., K. Panopoulos, P. Vourliotis, E. Pavlidou, and E. Kakara. 2006. Experimental Investigation of Fluidized-Bed Co-Combustion of Meat and Bone Meal with Coals and Olive Bagasse. Fuel. 85: 1685–1699.
CHAPTER
9
Post-Combustion Emissions Control David Nordstrand Subject Matter Expert – Emissions Control DTE Energy,
Dao N.B. Duong
Performance Engineer Foster Wheeler, NA, and
Bruce G. Miller Associate Director, Energy Institute The Pennsylvania State University
9.1 Introduction This chapter presents post-combustion emissions control strategies. This includes technologies for controlling particulate matter and acid gas control with an emphasis on sulfur dioxide and nitrogen oxides. Options for controlling mercury, the newest pollutant to be regulated, are also presented. The chapter concludes with a general discussion of carbon dioxide capture from flue gas streams, which is anticipated to be regulated in the future.
9.2 Particulate Capture 9.2.1 Introduction Particulate matter emissions from coal-fired electric utility boilers in the United States have decreased significantly since implementation of the 341
342 Combustion Engineering Issues for Solid Fuel Systems
1970 Clean Air Act Amendments. Several particulate control technologies are available for coal-fired power plants, including electrostatic precipitators (ESPs), fabric filters (baghouses), wet particulate scrubbers, mechanical collectors (cyclones), and hot-gas particulate filters. Of these, ESPs and fabric filters are currently the technologies of choice because they can meet current and pending legislation for particulate matter levels. While cleaning large volumes of flue gas, they achieve very high collection efficiencies and can remove fine particles. When operating properly, ESPs and baghouses can achieve overall collection efficiencies of 99.9% of primary particulates (over 99% control of PM10, i.e., particulate matter with an aerodynamic diameter less than 10 mm, and 95% control of PM2.5, i.e., particulate matter with an aerodynamic diameter less than 2.5 mm), thereby achieving the 1978 New Source Performance Standards required limit of 0.03 lb particulate matter per million Btu [1]. The primary particulate matter collection devices used in the power generation industry—ESPs and fabric filters—are discussed in this section.
9.2.2 Electrostatic Precipitation 9.2.2.1 Introduction The electrostatic precipitation section begins with a brief discussion of theory. A more detailed discussion of theory can be found elsewhere [1]. This discussion focuses on issues pertaining to equipment arrangement, process control, operation, and diagnostics. It concludes with a discussion of flue gas conditioning. Environmental controls for coal-fired boilers coincide with the increase in using pulverized coal for power production. The reduction of the coal particle’s size to increase its surface area advanced the field of combustion by achieving better combustion efficiency; however, it did result in causing pollution problems of the inorganic constituents of the coal. Prior to the use of pulverized coal technology, the ash primarily remained in the burning bed (i.e., stoker-fired boilers) and was removed as cinders. Finely grinding the coal resulted in the ash that is generated in the combustion process to remain in the gas stream and exit through the stack. The ash could then “fly” for miles, impacting a large area downwind; hence, the term “fly ash” was born. Utilities as well as the customer had mixed emotions to the problem in that the sight of the ash leaving the stack became the symbol of the modern technology that had harnessed the power. Most early corporate photographs showed the stations in full production with exhaust flowing from the stacks; they even lit the stacks for a more dramatic view at night. It was the downwind neighbors’ complaints that led to the control of the discharge. For the power industry, electrostatic precipitation (ESP) began with the installation of a commercial unit at The Detroit Edison Company Trenton Station in 1923. The Detroit Edison Company engaged F.G. Cottrell and his electrostatic precipitation technology after the company’s new
Post-Combustion Emissions Control
343
station (which was constructed without particulate control) started up in the rural community of Trenton, and the company became inundated with complaints. It was Cottrell who advanced the ESP technology by incorporating synchronous mechanical rectifiers with high-voltage transformers to convert high power levels of alternating current (AC) volts to direct current (DC) volts. The synchronous rectifier was the standard until the introduction of the silicone rectifier in the 1950s. It is through the control of the process that advances have been made, while the basic theory of operation has remained unchanged from its conception by the original inventors. 9.2.2.2 Theory The ESP process relies on a chargeable particle in the gas stream that can then be attracted to the grounded plate for collection within an electrical field. The chemistry of the particle defines the electrical properties that allow an electrical field to collect the ash. Charging occurs through the use of an electrode and grounded plate. Hence, the ESP is a chemical, electrical, and mechanical device. The three defined functions of the process are charging, collection, and removal. Charging A high-voltage negative charge is applied to an electrode to develop an electric field between the electrode and a grounded plate. If a particle passes through the field and accepts an additional electron, its charge changes to the negative state and is attracted to a positively grounded plate. This process, ionization, occurs to all elements but the noble gases; however, the effect is minimal except for a few of the compounds found in fly ash. It is the distribution of the strong elements that are receptive to ionization that supports the electron transfer. Ultimately, this is related to the particle size where the larger particles have more variability in composition. Increasing the available electrons results in more ionization occurring. This is Coulomb’s law. The law states that the force is proportional to the sum of and intensity of the charge applied. Figure 9-1 shows the relationship between particle size and the electric field required to attract the particles. As can be seen, larger particles require less energy for charging. The size distribution found in fly ash ranges from 0.5 mm to over 50 mm, with the larger sizes typically resulting from the use of eastern U.S. coals, whereas the smaller sizes are more typical from using western U.S. coals. Collection A charged particle migrates to the plate under the force of the electrical field by Coulomb’s force, but additional forces that the particle experiences impact the projected path. Charged particles reach the plate only when the electrical force is greater than the sum of the mechanical force, the gravitational force, the inertial force, and the flue gas viscous force. This can be represented as follows:
344 Combustion Engineering Issues for Solid Fuel Systems 105
Electrical Field, volts/meter
104
103
102
101 0.1
1
10
100
Particle Size, microns
FIGURE 9-1 Electric field required to attract particles as a function of particle size.
Charges on the ash þ Electrical Field > Mass Velocity
[9-1]
The inertial and viscous forces are the retarding forces that act on the particle to change its path based on the electrical force being applied. The movement is referred to as the migration velocity. When Stokes’ law on viscosity is applied, it can be seen in Figure 9-2 that increasing the particle size, charge, or the electric field will increase the migration velocity. Combining the charging and collection process into one graph defines both the time and length required to collect a given size particle. Again, eastern U.S. coals are favored over western U.S. coals because of their larger particle size. Particle collection, by theory, is manageable only when the gas and particles are limited to linear flow. In the real-world precipitator, the linear model does not apply and can only be modeled using turbulent gas flow equations that are exponential equations. This can be seen in Figure 9-2 where the migration velocity asymptotically approaches a finite value. This graph can be used to define the efficiency of the precipitator by the time and distance required to move the smallest particle to the grounded plate. The performance or collection efficiency of an ESP is defined as the mass of particulate collected divided by the mass of particulate entering the ESP over a given time period.
Post-Combustion Emissions Control
345
0.400
Migration Velocity, ft/sec
0.350
0.300
0.250
0.200
0.150 0.1
1 Particle Size, microns
10
FIGURE 9-2 Relationship between particle size and velocity for collection.
Ash Removal The last function of a precipitator is to remove the ash from the plates and the electric field. The difficulty arises when the electric field, the ash layer on the plates, the mass gas flow, and the striking energy are not matched and the ash becomes re-entrained into the gas flow. The preferred process is that the electric field holding the ash layer that is directly adjacent to the plate be of such strength that the strike energy just breaks this bond and gravity dislodges the particulate matter, which is an agglomerated ash layer, into the hopper. This has always been an issue with precipitator performance. Deficiencies in either the mechanical condition of the precipitator or in the ash characteristics will result in ash re-entrainment. 9.2.2.3 Equipment Arrangement Knowing the basic terms and location of the precipitator components helps to understand how the machine is configured and the overall process. Basic terms are as follows: Bus Section: The smallest division within the machine that contains assembled plates and electrodes that cannot be either mechanically or electrically subdivided. A bus section can be fed directly from the Transformer rectifier (TRSet) or from a high-voltage switch that ties many bus sections together to feed from a single TRSet; Field: Bus sections running with the gas flow;
346 Combustion Engineering Issues for Solid Fuel Systems
Cell: Bus sections running perpendicular to gas flow; Chamber: A group of bus sections isolated from other bus sections by a divider wall; and Hopper: Ash collection point under the bus sections. Fly ash collection will never reach 100%; however, when multiple precipitators or bus sections are integrated, the process can approach 100%. By rule of thumb, the inlet bus section will collect 80% of the ash delivered at its face. All fields after the inlet will collect 70% of the ash. The efficiencies of multiple field precipitators are illustrated in Table 9-1. The typical precipitator designed to collect the ash from high-sulfur American coals was originally 3 fields deep by 4–8 cells wide. Changing air quality regulations, specifically the need to reduce SO2, have a negative impact on the performance of the 3-field design. To counter the fuel change, more fields were added to improve the efficiency. The tendency to design 6-field precipitators is influenced by European designs that can collect the ash from lower-sulfur coals, a requirement of many governments in Europe. A secondary benefit of the European design is that the larger-sized chamber allows the gas velocity to be about half that of an American design and results in more collection because the particles are in the electric field for a longer time. A typical upgrade to the older 3-field design revolves around the repowering at the bus section level. The best improvements are for units with switch-controlled bus sections in the field direction. The result can be a 5-field precipitator that will have efficiency near the best-in-class 6-field units. 9.2.2.4 Resistivity Resistivity is the most important parameter in EPS technology and is often the source of most malfunctions. Values are offered as guides to good performances that center around 5 109 ohms-cm, this being the current density developed because of the ash chemistry and the electric field.
TABLE 9-1 Precipitator Efficiency by the Number of Fields Number of Fields
Amount Collected (%)
Amount Bypassed (%)
Overall Efficiency (%)
1 2 3 4 5 6
80.00 14.00 4.20 1.26 0.38 0.11
20.00 6.00 1.80 0.54 0.16 0.05
80.00 94.00 98.20 99.46 99.84 99.95
Post-Combustion Emissions Control
347
Resistivity, ohms-cm
1.00E+11 0.5% S
1.00E+10
1.0 % S 1.5% S
1.00E+09 1.00E+08
88 0
82 0
76 0
70 0
64 0
58 0
52 0
46 0
40 0
34 0
28 0
22 0
1.00E+07 Temperature, ⬚F
FIGURE 9-3 The effect of temperature on resistivity based on sulfur content of the coal.
When developing a method for course correction to control the variations in collection, one must rely on the data at hand and the fundamentals of collection theory. Figure 9-3 is a curve showing the relationship between temperature and resistivity defined by the sulfur content for a family of coals. Two very important relationships are displayed in this graph. The first is the impact that chemistry has on collection. Notice that the higher the percentage of sulfur, the lower the resistivity becomes. This is also true for other elements like iron and sodium. The second impact is the shape of the curves as temperature changes. A high-resistivity coal may be collectable in a different temperature range, and what would be considered a good coal for collection can cause problems at a higher temperature. Relying on the unit’s temperature indicator of the flue gas entering the precipitator without knowledge to the full temperature distribution across the full face of the precipitator can limit the analysis of resistivity and lead to an incorrect solution. Typically, the gas leaving a regenerative rotating air preheater is an average of a group of thermocouples placed on equal divisions of the duct area or height. That means that some part of the lowest and highest one-quarter of the temperature spread is not accounted for, and the collection process on the outer bounds of a good fuel is malfunctioning due to too high or too low of a resistivity. 9.2.2.5 Process Control The first two operations of the precipitator, charging and collection, are performed by a processor that is normally referred to as automatic voltage control (AVC). The processor also provides protection for the TRSet and hardware by limiting the electric field to preset parameters that match the design of the equipment. The basic components are shown in Figure 9-4.
348 Combustion Engineering Issues for Solid Fuel Systems TRANSFORMER-RECTIFIER SET TRANSFORMER POWER SOURCE LINEAR REACTOR
ESP DIODE BRIDGE
SECONDARY AMP METER
PRIMARY AMP METER SCRs & AVC
PRIMARY VOLTAGE METER
SECONDARY K-VOLT METER
FEEDBACK LOP THROUGH BUILDING STEEL
FIGURE 9-4 Power and protection circuit for precipitator control.
Digital technology has advanced the precipitator AVC into the computer age, which has allowed for more robust monitoring and finite control of the process. However, with few exceptions, the logic is still the same as that developed for analog circuitry. For proper precipitator function, a quality DC voltage signal must be sent into the section to develop the electric field. Along with good input, feedback for equipment protection is required. With those two functions in mind, one can define the components. The incoming power is from an AC single-phase source, typically 480 volts and protected for 200–300 amps. The AVC provides voltage trim by turning on the SCR (shown in Figure 9-4) for a predetermined time. The SCR rectifies the signal so it is all positive going to the transformer. The gate control circuit does the trimming, and the triggering circuit turns on the SCR. One cycle of an AC wave has 180 for flow available or at full power into the unit. This is then divided with 90 positive and 90 negative flow. The lower the angle selected, the less power is sent to the precipitator. Good practice limits the full power to an angle that ensures that power from only one full wave is sent per trigger. Selecting an angle too near the full cycle can result in more than one cycle being sent per trigger and activating the protection circuit that will turn off the precipitator. The linear reactor refines the signal going to the transformer. The precipitator is a DC device that must continuously maintain a flow of electrons from electrode to plate. The transformer in the TRSet provides a step-up in the voltage from the power plant bus voltage to 40,000 volts. The conservation of energy applies, and the resulting current drops to the milliamp range. The diode bridge then filters out much of the AC ripple left in the signal before it enters the precipitator proper.
Post-Combustion Emissions Control
349
Spark Limit
Primary voltage limit
Lower Resistivity Lower Load
Primary voltage under limit
Current
T/R secondary current limit
Secondary voltage limit
T/R primary current limit
Voltage
FIGURE 9-5 Control limits for the controller feedback signal.
The protection of the equipment is also the function of the AVC. Using the four feedback signals—primary volts, current and secondary kilovolts, and current—the AVC monitors and then corrects for deficiencies when ordering the triggering of the SCR. All the feedback signals have control limits that define where the precipitator can operate. The relationship between them is shown in Figure 9-5. Added to the graph is the impact of the spark limit as the diagonal line impacting the performance at higher voltages and higher resistivities. Spark rate will control the process by lowering load to maintain opacity. Lower resistivity or loads require less voltage but more current to supply the electric field. After a spark has been drawn, the controller provides equipment protection. The control is done after the controller senses a rapid rise in current from the last cycle and/or with a drop in secondary voltage. Controllers accomplish this in the first cycle of the spark. The voltage is quenched for a user-defined time, typically 2–3 cycles, and then the controller re-introduces the voltage in a fast ramp of 3–4 cycles to a setback point. This point is set by the controller based on the voltage at which the spark occurred. A value of 10–15% works well for an inlet section and 5–10% for the other sections as well as the outlet. A slow ramp then takes over for about 10 cycles until the voltage level has been returned to its original level or the section sparks again. The sequence is shown in Figure 9-6. All these spark control parameters are user-definable and should be experimented with under normal operation so the correct response can be
Current
350 Combustion Engineering Issues for Solid Fuel Systems
Set Back
Slow Ramp Fast Ramp
Time Quench Time
FIGURE 9-6 History of a spark and the controller’s reaction.
understood. Making changes during a system malfunction is not advised. An oscilloscope can be used to determine the response, and its use is the best way to set up the spark rate parameters. A voltage and current trace of a spark and the protection that is occurring are shown in Figure 9-7. Controllers have advanced to where the oscilloscope traces can be displayed on a handheld or computer interface. This type of data cannot be replaced with the unit-wide process data systems due to the slow 1-second time response typical of the network. The third function of the precipitator is to remove the ash. Controllers for this job have advanced from the rolling cam timers to PLC logic to digital processors. Each step forward has improved the cycling of the rapping equipment, but it is the digital processor that has allowed for signals from the boiler and precipitator indicators to be fed back into the sequencing of the rappers. WF1 DE MONROE 8/17/04 SI and SV on LBC-S2 Shielded Cables
1> 2>
1) Ref A 200 mVolt 10 ms 2) Ref B 200 mVolt 10 ms
FIGURE 9-7 Oscilloscope trace of a spark and control.
Post-Combustion Emissions Control
351
Removing the ash from the precipitator requires breaking the chemical–electric field bond that is holding the ash to the plate. The performed force is to strike the plate and shake loose the ash. With gravity and the electric field weakening due to ash buildup, the ash will drop into a receiving hopper. Devices for striking are of three types: Dropping weight: A steel weight is supported within a nonconductive sleeve with a magnetic coil surrounding it. Applying voltage to the coil will lift the weight, and removing the power will drop the weight to generate the striking force; Hammer: This is similar to the dropping weight rapper, but the hammer is on either a rolling shaft or a lifting rod. The drive is then a electric motor gearbox controlled by applying power to the motor; and Vibrator: The vibrator applies a force by shaking the hardware with an agitator-style motor. It is controlled by switching the power on and off. All the rapper styles have drawbacks that have resulted in differing preferences by the manufacturers and suppliers. In operation, all function if the chemical, electrical, and mechanical parameters of the precipitator are maintained. A fuel or boiler hardware change can impact the equilibrium of the precipitator. Modification to the rappers may be necessary to regain the equilibrium. Timing of these modifications is important in that waiting too long to see how the precipitator reacts may result in some adverse conditions that cannot be corrected without an outage of the unit. In general, if the modifications are expected to change the ash entering the precipitator, then the rapper program should be rewritten to maintain equilibrium. If, for example, the gas flow will increase by 1%, then the rappers will need to remove ash in a corresponding shortened time and with more energy to provide a larger electric field to collect the faster moving flue gas flow and ash. In operation, the precipitator follows theory, with minor deviations that can be explained by the process parameters. The time delay between good precipitator operation and a derate caused by process changes is anywhere from 12–48 hours, more than enough time to recognize the problem and positively react to the change. 9.2.2.6 Operating an Electrostatic Precipitator As with any operating unit, there are procedures to follow. Remembering that resistivity is the parameter to maintain and that the equipment and fuels will vary is important. The first sign of one of them changing is an increase in opacity. Noticing this increase is the first step in correcting the collection problem. There is nothing wrong with first considering that the collection problem is due to a malfunction of the precipitator; however, one should not just limit the investigation to this equipment. The electrical components of the
352 Combustion Engineering Issues for Solid Fuel Systems 2
Kvolts 40
1.5
30
mamps
20
1
Sparks
mamps
Kvolts & Spark Rate
50
0.5
10
0
0 Inlet Field
Center Field
Outlet Field
Fields into Flow
FIGURE 9-8 Expected performance trends of a properly operating precipitator.
Opacity
precipitator can be quickly diagnosed if the precipitator is at fault. Figure 9-8 shows expected response within a gas pass through a precipitator. The logic of this response is from the collection theory where multiple fields are required and less ash is available in each field as it passes through the precipitator. The electric field is developed between the negative electrode and the grounded plate based on the number of ash particles present that will accept a charge. Also important is the particle size distribution and the thermal gradients across the face of the precipitator. In an inlet field with a large volume of ash present, an electric field capable of charging the particles does so at a lower current level. Moving further into each field, less ash should be in the stream, and the electric field expands until the available particles take a charge. One should return to the first indicator noticed, opacity, and check on the trend over time, looking for resistivity being out of bounds. A lower than acceptable resistivity will display as a very active, broadband baseline with a standard deviation of 2–3% (see Figure 9-9). Additional spiking will
Tim e
FIGURE 9-9 Low-resistivity response in opacity.
Post-Combustion Emissions Control
353
be present over normal rapper response because of the odd resistivity value that the unit is reacting to. In the figure, it can be seen that no spiking pattern is distinguishable, and very little warning that a spike is going to occur is available. A higher resistivity will display very tight baseline opacity with quick, high rapper spikes, as shown in Figure 9-10. Ideally, the process should operate between the two extremes with spikes less than 30% (on a marginal precipitator) and a baseline trend that can be definable. Applying corrections to the parameters of the process can direct the precipitator toward proper collection. The boiler parameters that have the greatest impact on precipitator operation are excess air and stack gas temperature. Both of these parameters were part of the design basis for the unit and are a function of the design coal the unit was intended to burn. Very few units can operate their lifetime on design coal because even a mine mouth plant will face variations in the coal seam. As the coal source changes, the margin in the design base is eroded to the point that some equipment will limit the unit’s capability to maintain rated load. Typical equipment to be limited by capacity are the coal mills, fans, and the precipitator. All these have a first-order impact on opacity.
Opacity
Excess Air The air available to properly burn the coal is basic to the combustion process. Over-supplying air increases the velocity through the ESP, and out the stack. This can be observed in the summer when a unit exceeds induced draft fan capacity because the air volume to the forced draft fan has increased due to higher ambient temperatures. Typically, reducing O2 by 0.1% can sufficiently drop the gas volume to regain collection in the precipitator. What occurs in the precipitator has been a mechanical process balance change caused by the increased volume of gas. This additional gas is then compressed to fit within the defined volume of the precipitator case. The result is an increase in velocity, which then reduces the time the
Time
FIGURE 9-10 High-resistivity response in opacity.
354 Combustion Engineering Issues for Solid Fuel Systems
ash particles are in the precipitator. The design velocity is referred to as the retention time and ranges from 3–6 ft/sec depending on the original design of the process. Reducing excess air strictly for precipitator performance is not advisable without performing the correct combustion testing or having the right process indicators in place. A CO monitoring grid would be the recommended instrumentation to control excess air. It should also be noted that combustion tuning that does not consider the rest of the process is just as dangerous as reducing O2 for opacity control and producing a few additional megawatts. Testing for CO across the economizer outlet combined with flame stability adjustment will support the true excess air requirements for the unit. The original equipment manufacturer (OEM) has a defined value for CO that should not be exceeded but should be used as the control for adjusting O2. Once the value is established, the testing needs to be redone periodically so that any drift is controlled. A secondary benefit to periodic testing is to define air in-leakage from failed expansion joints or duct corrosion that occurs prior to the precipitator, which puts an additional flow load on the precipitator. The addition of low NOx burners to an existing or retrofitted unit will raise the combustion air requirements. Typically, at startup, few mass gas flow problems are encountered due to the relatively tight conditions of the new equipment. Understanding that equipment deteriorates over time is important, and those setpoints that define the burner operation must be recorded so that the process drift can be corrected later. The O2 operating curve is normally modified to handle the additional air requirements. If the unit is opacity limited at loads well below the unit rating, then the combustion testing needs to be performed at lower loads to validate that the curve is correct. Most retrofitted curves have been drawn with straight lines based on too little test data. Stack Gas Temperature Resistivity is a function of the gas temperature in the precipitator; this is the most important operating variable and has the highest impact on opacity. Stack gas temperature is usually monitored by an array of three to four thermocouples mounted in the outlet duct of the air heater. The outputs are summed and the average is displayed in the control room; however, two sets of data have been lost with this arrangement. The first is the true thermal profile of the air heater, and the second is the real data that exist from the thermocouple array. The thermal profile can be developed by manually testing the duct after the air heater. Again, the data must be obtained for a properly running unit so that any deviations can be detected. Air heater heat transfer is basically linear and repeatable as a function of the wheel’s travel through a full revolution when determining the temperature spread across the face of the precipitator. Having these data for a properly running precipitator will define the thermal range that the process will operate within. If possible, one should vary the stack gas temperature
Post-Combustion Emissions Control
355
to determine how much latitude the system has. Most mid-1960 designs will have a 100 F air heater spread with a variability of 15 F on top of that value for daily operating control. If no tools are available to vary the stack gas temperature, then one should use the seasonal temperature range to determine the spread. Some units will have a consistent load limit when the ambient temperature reaches a high or low level that may appear to be a fan limit but could really be a resistivity limit. Coal Chemistry Coal chemistry is not usually considered as an operating parameter mainly because very little can be done to change coal’s composition once it is on-site. It is important to document the impact of changing coal composition with process data so that in future the same coal can be used intelligently. The indicators of the fuel quality are not new because most units have had equipment problems when certain values stray from the design parameters. What is important is recognizing that the localized equipment problems are impacting the process as a whole. Decreasing the heating value of the coal (which results in higher coal and combustion air feed rates to maintain the same thermal input) has impacts such as increasing coal mill throughput and increased fan requirements. What is not normally considered is the effect of the additional air volume on the mass gas flow through the precipitator. This can be measured by testing but does require a rather tight boiler casing to quantify the change. Knowing there is a change is more important than having a definitive number. An increase in the percentage of ash can cause the reduction in heating value; plus, it will add to the ash loading on the precipitator. The design coal ash content and heating value will provide a value for the precipitator’s collection capabilities. The precipitator will bypass any additional ash as opacity and a load limit. An increase in moisture delivered with the coal or added after delivery affects the precipitator. Under controlled conditions, water can be a very good conditioner of the ash and will enhance collection. But, as the quantity of water increases, the resistivity begins to decrease due to chemistry changes and thermal losses due to increased moisture in the flue gas. An increase in the spark rate with a drop in power is expected from a moisture unbalance. Adding large quantities of water, like in an economizer tube leak, will result in a total collapse of the electric field and a dramatic increase in opacity. The sulfur content of the coal, because of environmental regulations, is now a controlled variable in the coal. Even when the coal has reduced sulfur levels, older precipitators will still collect particles provided either that the process is controlled to required parameters for temperature and flow or a conditioning agent is used. Using continuous emissions monitoring (CEM) for SO2 will provide a very good indicator as to the available sulfur for collection. SO2 is an indicator of the available sulfur that has been converted to SO3 by the boiler process. Knowing how the precipitator
356 Combustion Engineering Issues for Solid Fuel Systems
functions with a given percentage of sulfur from the coal helps to maintain the equilibrium of the process. Once the good range is defined, it is easy to adjust the boiler O2 and stack temperature to follow the drift in sulfur. If there is less sulfur, one may lower the temperature. If there is more sulfur, one may increase the temperature. O2 is used as a trim after the best opacity level has been achieved with temperature adjustment. 9.2.2.7 Diagnostics Being diligent on maintaining process balance will minimize problems but will not eliminate them entirely. Often the problems arise when the unit is operating near the full load rating or in the attempt to exceed the unit rating. Defining the conditions and moment when the problem first occurs simplifies the search for the parameter that changed because many of the boiler parameters for the high load point are known, can be found, or can be calculated. Here, it is important to have a good sense as to how the boiler parameters should look to detect the outliers. Also, it is easy to repeat the process as a test to determine which parameters are out of bounds. The second way that problems occur is that no one is paying attention to boiler operation or conditions are such that no one expects the precipitator to be a problem. A good example of this is during off-peak times when load is reduced, opacity has dropped, and the stack gas temperature is allowed to drift below the proper resistivity range. The outcome is a change in the ash chemistry that leaves a very tenacious ash on the plates which is difficult to be rapped off. Then, when the unit must achieve full load and stack gas temperature returns to normal, the precipitator will not respond and opacity will increase, thereby derating the unit. This same condition can occur in units that blend coals, for example, using a higher blend of a poor-quality coal, in an attempt to reduce the fuel cost. The third way that problems arise is that equipment fails. Although equipment failure is the cause, it is not the primary reason this occurs. Usually, it is a past missed opportunity to correct a problem that results in the equipment failure. Despite the cause, the malfunction must be dealt with; therefore, one must look into the operation of the precipitator to find the source of the problem even if the source is elsewhere. Methods for interrogating the precipitator control data for clues are included in the suppliers’ operating manuals. Following their procedures will determine what the problem is, if it is precipitator based. Force-fitting the data by assuming the precipitator controls are reacting very close to a described condition may lead to an incorrect conclusion. The full data set, including the boiler and fuel, needs to be analyzed to resolve the problem. It is imperative to have a very good set of test data on the unit that was obtained when the unit was in good repair and performing in an efficient state. Subsequent data gathering has to be done regularly and include
Post-Combustion Emissions Control
357
the coal and ash analysis, plus the precipitator readings. Data sets of this nature require that the unit be properly instrumented and that the devices are in calibration. Even a marginally monitored boiler can provide meaningful information through observation. With the precipitator control readings in hand, what do the data tell? Following are examples of typical malfunctions. One section malfunctioning is an indication of a mechanical problem within the equipment. A mechanical problem is defined as one needing local repair so that the section can return to normal operation. Unit derate history will provide information as to which and how many of the sections can be out of service and the unit still be able make load. It is best to place a process warning on the need to do maintenance at one failure short of the derate number so that load loss does not occur. A complete gas path malfunctioning is an indication of a boiler or burner issue. Here, the source will be a change in equipment condition that will allow for a localized disturbance in the system balance. The temperature of the flue gas is the first parameter to check, followed by the quality of the combustion process with regards to carbon carryover into the precipitator. Large boilers generally have two or more forced draft and induced draft fans plus air preheaters that develop distinct flow paths for the flue gas traveling through the process. The accepted balancing monitoring parameter is fan motor amps with the assumption that equal work by the motors means the flow is balanced side-to-side for the most efficient operation. In reality, the flow balance in this manner is not as important to the precipitator as is balanced temperature. A small adjustment in the fan bias control will normally bring the temperature out of the air preheaters in line with each other. Remember that balancing the temperature into the ESP minimizes the overall temperature spread the precipitator is reacting to. The average temperature value for resistivity should also be corrected at this time. Carbon carryover, also known as loss-on-ignition (LOI) or carbon in the ash depending on how it is formed, requires laboratory analyses to get a finite value. However, with practice and knowledge of the color of the normal ash produced by the unit, one can see a general conditional change that will suggest the presence of a problem. Finding the source will require a review of the burner settings and the fineness of the grind. Total precipitator performance degradation is an indication of a fuel issue. The symptoms vary depending on how the fuel changed. A review of the fuel analysis that came with the shipment should be the first step. In order of importance, one should check for changes in sulfur, ash, moisture, and heating value against the unit standard fuel analysis that gives full load. It would also be valuable to compare the new fuel to the design fuel, particularly if the present standard fuel is known to have absorbed most of the design margin. Determining the impact of each change is then
358 Combustion Engineering Issues for Solid Fuel Systems
based on how other equipment in the system reacts to the fuel change. The impacts of coal changes were discussed previously and should be applied when determining the problem. Adding ash mineral analysis to the review opens up analysis to the chemical conditioning of the process. Again, the first requirement is to compare the ash analysis to the design standard fuel specification for changes in individual compounds. When reviewing the ash mineral analysis, one needs to remember that the values are a percentage of the total sample volume. The more sampling that is done, the better the representation will be. Here, the focus is on those compounds that enhance and impede collection. Sulfur in the ash mineral analysis reports all the sulfur compounds produced by the combustion of the coal in the presence of the ash. Of all the sulfur compounds, only SO3 has the additional electron that supports collection. The better indicator is the CEM SO2 value with an SO2 calculation from the fuel analysis: SO2 =MMBtu ¼ ð%SCoal Þ 2000 =HHVCoal
[9-2]
where SO2/MMBtu ¼ lb SO2 per million Btu heat input; %SCoal ¼ percent sulfur in the coal, dry basis; and HHVCoal ¼ higher heating value of the coal, Btu/lb, dry basis. This can be an indicator of how much of the sulfur in the coal can be converted to SO3 (typically about 1% of the SO2 is converted to SO3) and how much has passed through the precipitator versus remained in the boiler or ended up as bottom ash. Other elements in the ash, like iron, will tie up the sulfur as slag before the conversion to SO3 can be accomplished. The CEM provides data on the minimum SO2 value that the process will maintain collection within the opacity limit and the consistency of the sulfur content of the coal. This value should be independent of the fuel origin. SO3, then, is a byproduct of the combustion process and generated in the high-temperature region of the boiler. In a gaseous form, it is not a stable compound and will convert back to SO2 in the presence of other sulfur compounds, oxygen, and lower temperature. The hope is to have the gas condense in the presence of the ash and then mix and agglomerate, thereby allowing the new, larger particle to have the spare electron. If a chance encounter does not occur, then the SO3 can change into a liquid or a mist. As a liquid, SO3 will not mix with the ash, not act as a collection agent, and will pass through the precipitator uncollected. Unfortunately, liquid SO3 will reflect light and will be seen by the opacity meter, causing an increase in opacity. This is of concern for units that have selective catalytic reactors. Counter measures to ensure maximum SO3 include all the actions
Post-Combustion Emissions Control
359
that constitute good combustion, such as proper air/fuel mixture, operating the boiler with the fireball in the furnace and not about the bullnose, controlling furnace temperature, and keeping the backpasses clean. Iron, unlike sulfur, will enhance collection because of its natural reaction to being affected by a magnetic field. The downside to iron is its capability to react with sulfur in the absence of oxygen and then attack waterwall tubes. The preference between the two should be for boiler protection. If there is a choice of the coals to be burned, lower iron should be preferred; however, this results in less sulfur available to assist with particulate collection. However, when iron is minimized, which also reduces sulfur (since sulfur and iron are often associated together because they are components of pyrite), other less acceptable compounds increase in concentration. Lithium has a strong attraction to the negative plate when in the electric field, but typically there is not enough available in the ash to have a significant effect. Most laboratories cannot test for the presence of lithium, and the result is that the analysis typically reports values less than 1%. Sodium can assist with the collection process even at the lower operating temperatures of cold-side precipitators. Sodium does enhance the presence of slag in the boiler if the percentages are high. Sodium contents as low as 2% in the ash can be considered high for an eastern U.S. coal-designed boiler. Alumina and silica are often combined because they are the bulk of the composition and can increase the resistivity of the ash. If the combined concentration of alumina and silica exceeds 80%, resistivity will increase and collection is impacted in a small precipitator. If the unit is burning a blend of eastern and western U.S. fuels, the important number is the percentage of the western fuel in the blend. When alumina and silica are heated, a glassy material can be produced. Glass has properties that make it a good electrical insulator and semiconductor. This causes the heavy alumina and silica particles to retain most of the charge when it becomes attached to the plate. The charge bond between the particle and the plate will be greater than the rapping force used to clean the plates. As the heavy alumna and silica particles build up, the resistivity of the plate increases, causing a drop in the effective power being used for collection. High alumina and silica content in the ash requires high power applied in the rappers. Base/acid ratio, an indictor of how the ash will behave in the boiler with regards to slagging, can be also be used as a measure of how well the ash will collect in the precipitator. These are the data behind the rule of thumb that coals that are good for the boiler are bad for the precipitator and vice versa. Base=Acid ¼
Fe2 O3 þ CaO þ MgO þ Na2 O þ K2 O SiO2 þ Al2 O3 þ TiO2
[9-3]
360 Combustion Engineering Issues for Solid Fuel Systems
Note that alumina and silica are in the denominator of the equation, and iron is in the numerator. As has already been discussed, increasing iron will support precipitator collection, as will reducing the amount of alumna and silica. By trending base/acid ratio over all fuels burned and comparing it to the visual on boiler deposits and opacity, one can determine a very good indictor of a coal’s performance from the ash mineral analysis. 9.2.2.8 Resistivity Conditioning Modifying the resistivity with additives is a popular method for resolving collection problems caused by high-resistivity ash. The process revolves around spraying a product into the gas steam to come into contact with the ash particles. Because collection is a function of the electric field and a particle having a spare electron, all these additives affect the particle’s electron count. Physically, the additives have to either coat the particle or cause the particle to stick to another particle to accomplish the task. The thermal condition at which the additive is applied is product sensitive and may require the chemical plant that supplies the product to the duct to include a heat source. All the products require sufficient retention time and proper flows into the precipitator to ensure that mixing takes place. A short duct distance from the air heater to the precipitator with a high flow profile will complicate the conditioning process. The most widely used process is to inject SO3 into the duct. The system to produce SO3 is simple and straightforward in mimicking the natural generation of SO3 in the boiler. SO3 is an example of an additive that coats the ash. The system can use elemental sulfur in either a molten or dry form, plus commercially available SO2 as a feedstock. The elemental sulfur has to be converted into SO2, which is done by burning the material in the presence of oxygen. The heat source is the process air elevated to temperatures higher than the combustion point of sulfur. The resulting SO2 gas is then forced through a catalyst of vanadium pentoxide to generate SO3. The SO3 is then blown into the duct with enough velocity and heat to condense on the fly ash beyond the nozzle of the injection probe. This electron-rich particle will be collected when it enters the electrical field of the precipitator. The use of SO2 as a feedstock requires additional safeguards and energy to maintain the high temperature for conversion to SO3. Sulfur can be over-injected, causing the inlet fields to become current limited. Under these conditions, two undesirable actions occur: First, the inlets will set up an electric field that favors the very low resistivity particles, thereby minimizing collection; and, second, it allows more ash to be bypassed into the latter fields of the precipitator, increasing the spark rate in the outlet fields, which also limits collection. Sulfur’s capability to modify resistivity has a relationship to the temperature at which the precipitator is operating. Raising or lowering the temperature will assist sulfur in controlling resistivity. The reality of adjusting sulfur and temperature includes factoring in temperature changes
Post-Combustion Emissions Control
361
by the rotating air heater. The air heater impact can be eliminated but requires a graduated injection delivery system developed solely for the specific unit. Anhydrous ammonia has also been widely been used for conditioning because of its ability to agglomerate particles together. The ammonia is stored under pressure and piped to a control valve for metering. After the valve, process air is added to transport the ammonia into the duct from its probes. At this point, the distribution of the ash in the presence of the ammonia dictates the success of attaching particles together. Once in the electric field, the new larger particle will be collected if an available free electron is part of the composition. Eastern U.S. fuels high in alumina and silica can be more collectable when using anhydrous ammonia as a collection agent. Ammonia should not be over-injected because it will cause the collapse of the electric field. Combining sulfur with ammonia is a possibility and can be successful at enhancing collection. Adjustment of the two agents should consider sulfur first because of its dependence on temperature, followed by trimming with the ammonia. The expected result is reducing the opacity baseline with the sulfur and then limiting the spiking with the ammonia. Proprietary additives are available to improve collection; however, they have to either coat or agglomerate the particles to improve collection. These additives are evaluated by performance testing because the chemical compound will not be divulged. A test of this type will require the plant owner to expend funds and labor to install and operate a temporary system for the demonstration. The unit parameters, i.e., gas flow and pressure, need to be monitored to ensure that the additive is not adversely impacting the process. The test duration needs to be long, i.e., more than a week, which is a typical test length, and it is better if it lasts over a month. Timing of the test should be during an off-production season and preferably before a periodic outage so the internals of the ESP can be inspected. Part of the testing should include varying unit parameters to simulate adverse operating conditions. Possible actions would be to elevate the air temperature entering the forced draft fans to model summer operation or to increase O2 to increase velocity, as would be seen for a plugged backpass. Most additives work at the basic modular level and, as such, may require favorable ash chemistry to be effective. Varying the coal as part of the test will provide an indication of the system limitations if fuel quality were to degrade in the future.
9.2.3 Baghouse/Fabric Filters 9.2.3.1 Overview Up to about 1970, the development and use of fabric filters or baghouses was limited due to two crucial factors: material availability and bag chemical resistance. The availability of the materials limited installations to
362 Combustion Engineering Issues for Solid Fuel Systems
temperatures below 250 F, and the chemical resistance characteristics of the bags reduced fabric filtration. As advancements were made, the interest in baghouses increased as a result of successful installations on large coalfired boilers that proved to have good operation and high efficiencies in particulate removal. A fabric filter or baghouse will collect dry particulate matter from the flue gas as it passes through the filter material. The baghouse will typically be composed of multiple compartments that utilize long, vertically supported, fabric filters or bags that separate particulate matter from the flue gas. 9.2.3.2 Basic Principles As the flue gas exits the boiler and flows through the baghouse, particulate matter from the gas forms a “cake” on the fabric filter. This cake serves to increase both the cloth efficiency and gas flow resistance. As a result of this cake formation, periodic cleaning is required during continuous operation. Three different baghouse designs are available, each with its own cleaning mechanism; these are the shaker, reverse-air, and pulse-jet designs. Some of the key design criteria in baghouses are air-to-cloth ratio, particle size, fabric material, support equipment, and drag or pressure drop. Air-to-cloth (A/C) ratio is the ratio of the flue gas flow through the apparatus (in actual cubic feet per minute—acfm) to the area of available fabric (in square feet). For the reverse-air and pulse-jet designs, A/C ratios for coal-fired plants will range from 1.5–2.3 ft/min (0.45–0.7 m/min) and 3.0–4.0 ft/min (0.9–1.2 m/min), respectively. For all three systems, particle size is a key factor in the final design. As the quantity of fine particulates increases, the filter cake opening decreases, and a finer bag weave is required. Fabric materials will be selected based on ash composition, particle size distribution, flue gas temperature, moisture quantity, flue gas dew point, and the collected material’s hydroscopic behavior. Support equipment will include isolation dampers, hoppers, insulation, mechanical supports, controls, and several others. Drag or pressure drop will include the drop across the bags, cake deposit, and the attachment of the tubesheet to the bags. A general equation for drag or pressure drop is [2] as follows: DP ¼ a V þ b c t V 2 where DP a V b c t
¼ pressure drop, inches water gauge (wg). (kPa); ¼ constant, inches wg.-min/ft (kPa-min/m); ¼ filtration velocity, ft/min (m/min); ¼ cake coefficient, inches wg.-min-ft/lb (kPa-min-m/kg); ¼ inlet dust concentration, lb/ft3 (kg/m3); and ¼ time, min.
[9-4]
Post-Combustion Emissions Control
363
In equation [9-4], the constant a is determined through actual operating data; the value increases as the fabric filter ages. Cake coefficients will range anywhere from 0.1–700. For coal fly ash, this range will be from approximately 5–20. When special fabrics and finishes are utilized, the achievable outlet emissions level is lower for fabric filters than electrostatic precipitators. In addition, because the hopper material for a fabric filter system will be drier, the ash handling and disposal process is easier. 9.2.3.3 Specific Designs Shaker Design The shaker design is the oldest design of the three. As the name implies, this design uses a shaking action to fracture the filter cake. The cleaning process begins by isolating a compartment of the baghouse. Mechanical energy in the form of a sinusoidal acceleration is applied to the top of the bag. As the sinusoidal acceleration travels down along the bag, it fractures the cake. This fractured cake then collects in the hoppers. Shaker designs have been applied to both inside collectors where the particulates are collected on the inside of the bag filters and outside collectors where the particulates are collected on the outside of the bag filters. Figure 9-11 is a general schematic for an inside collector, showing both the normal operation and cleaning cycle. Application temperatures will range from ambient temperature for dust collection/control to elevated gas temperatures observed in a large utility boiler system. As a result, limitations to the system are cloth material and temperature, which depend on the material’s capability to withstand the shaking action without incurring excessive damage. In addition, proper weave construction, finish, and weight of the fabric all play critical roles in this design. Shaking Drive Clean Gas Outlet
Damper Closed During Cleaning
Filter Bag
Dirty Gas Inlet
Tubesheet
Under Normal Operation
Under Cleaning
FIGURE 9-11 Schematic diagram of a shaker design under normal operation and under cleaning.
364 Combustion Engineering Issues for Solid Fuel Systems
Pulse-Jet Design The pulse-jet design operates mainly by the use of outside collection. Deposit removal for this system introduces a pulse of compressed air that is injected into the open end of an isolated bag filter. The pulse then will create a shock wave that travels through the bag length causing a momentary expansion of the bag, fracturing the cake deposit. Wire or mesh frame cages are used in this design to prevent the filters from collapsing during cleaning operation. Depending on the gas composition, customer preference, and component reuse, varying wire or mesh frame cage materials are available. Cleaning can be accomplished with either all compartments in service or with one compartment out of service. Figure 9-12 shows a generic schematic of the pulse-jet design. In this particular design, rows of filter bags are pulsed sequentially. By cleaning in this manner, the operation of an entire compartment is not compromised. The particular compartment does not need to be isolated during the cleaning cycle. Compressed air that is utilized can vary from low pressure and high volume to high pressure and low volume. The compressed air is discharged from what are known as pulse valves, which are rapid opening and closing valves. The air then travels along a supply pipe to nozzles or holes to the top of the bag, where it is released. Reverse-Air Design In the reverse-air design, clean flue gas from the outlet duct is drawn back in and used to blow in the direction counter to normal flow. This action relaxes the filters in the isolated compartment and fractures the cake. The bags are then re-inflated before more “dirty” flue gas is allowed to travel through the compartment. To achieve this cleaning mechanism, the design must utilize a reverse-gas fan and dampers to supply a controlled flow of clean flue gas to the bags. Figure 9-13 is a generic schematic of the reverse-air design.
Clean Gas Outlet
Fabric Filter with With Cage
Dirty Gas Inlet
Dirty Gas Outlet
High Pressure Cleaning Air
Clean Gas Inlet Under Normal Operation
Under Cleaning Single Filter Bag Isolation
FIGURE 9-12 Schematic diagram for pulse-jet design under normal operation and under cleaning.
Post-Combustion Emissions Control
365
Reverse Gas In Damper Closed
Clean Gas Outlet
Damper Open
Damper Closed During Cleaning
Filter Bag Anti-Collapse Rings Dirty Gas Inlet
Tubesheet
Under Normal Operation
Gas Out
Under Cleaning
FIGURE 9-13 Schematic diagram of the reverse-air design under normal operation and under cleaning.
To prevent a full collapse of the filter bag during the cleaning process, multiple rings are sewn throughout the bag length. Application temperatures will range anywhere from ambient to over 500 F. Glass cloths are often used because they can withstand the wide range of temperature without incurring excessive physical damage. Reverse-air design has the gentlest cleaning mechanism of the three designs. However, because of this passive cleaning mechanism, in certain applications the operating pressure drop becomes unacceptable. To overcome this issue, some units will utilize sonic air horns to assist with the cleaning. In addition, cleaning frequencies will depend on the filter cake resistivity and inlet grain loading. In general, as the inlet grain loading increases, the cleaning frequency also increases. This results in longer cleaning times, thereby not permitting the full use of all bag filter compartments. 9.2.3.4 Collection Efficiency Baghouse performance and efficiency depends on several factors: the air-tocloth ratio, the permeability of the fabric, the nature of the particulate, and even the cleaning frequency. With correct design and component selection, the efficiency of baghouses can be as high as 99.9% or 0.005 grains per dry standard cubic feet as the outlet loading. Some of the operating issues that may lead to fabric filter downtime or efficiency reduction are bag blinding, bag erosion, and filter bag deterioration. Bag blinding will typically be caused by frequent or prolonged periods of operation at or below the dew point. When the filter is operating under such conditions, high pressure drops can occur, which then become a fan or process limitation. Furthermore, bag blinding can lead to faster bag replacement. Bag erosion is typically caused by high velocity flue gas
366 Combustion Engineering Issues for Solid Fuel Systems
impacting the filter material. To minimize erosion effects, one typically limits flue gas velocities to less than 300 ft/min at the bag neck. Deteriorations in the filter bags can either be caused by poor material or finish selection or by operating abnormally outside the design range of temperature, volumetric flow, and flue gas chemistry. 9.2.3.5 Conclusions Most particulate control systems typically require either a pressure drop or collection area to properly perform. In addition, changes to the gas volume or particulate loading will typically cause the efficiencies to decrease or the emissions output to increase. Baghouses, on the other hand, do not materially change the final emissions output, since the collected material becomes the capturing medium. When western U.S. fuels are burned, baghouses may be more suitable as the particulate emissions control than electrostatic precipitators. The overall system utilizes fabric materials and finishes most suitable for the application. Considerations such as weave, thread count, weight, and finish for the filter bag are important. In addition, bag inlet velocities, air-tocloth ratio, bag-hanging hardware, and damper design are among some of the other factors that determine the stability and performance of the baghouse. From an economic standpoint, the shaker type has the lowest capital cost; however, it also has the highest maintenance cost. Therefore, the pulse-jet and reverse-air types are used more often than the shaker type. Baghouses or fabric filters vary in applications, ranging from coal-fired utilities, circulating fluidized-bed boilers, industrial boilers, stoker-fired units, refuse-derived fuel units, and furnaces and boilers in the cement and steel industry.
9.3 Acid Gas Control 9.3.1 Acid Gases of Importance: SO2 , HCl Acid gases such as sulfur dioxide (SO2), sulfur trioxide (SO3), hydrogen chloride (HCl), and hydrogen fluoride (HF) cause widespread damage when they precipitate as acid rain or when they form weak acidic solutions with moisture at ground level. These effects will include damage to soil productivity, damage to vegetation and aquatic life, and damage to manmade structures. As a result, emission regulations have become more stringent over the years, pushing the industry to limit the emissions on these acid gases, primarily SO2. SO2 emissions are mainly generated by two principal industrial sources: fossil fuel combustion and metallurgical ore refining. In practice, the reduction of SO2 has been achieved by utilizing lower sulfur fuels, fuel desulfurization methods, or by installing a flue gas desulfurization (FGD) system. In some cases, when possible, combinations of the different methods are employed to meet the emissions level required.
Post-Combustion Emissions Control
367
Many utilities are currently utilizing lower sulfur fuels; this is done by either blending the lower sulfur fuel with the parent coal or completely switching to lower sulfur fuels. In coal washing, pyritic sulfur is removed, with sulfur reductions up to 50% achieved. Depending on the application, cost, and efficiencies required, FGD systems available for acid gas control are wet scrubbers, spray dryer absorbers (or semi-dry systems), and dry injection systems.
9.3.2 Array of Technologies Depending on Application In general, scrubber systems rely on chemical reactions with a sorbent to remove acid gases and small quantities of heavy metals in the flue gas. In a wet scrubber system, the liquid sorbent is sprayed into the flue gas; the acid gases will then get diffused or scrubbed into the liquid. With additional treatment, the mixture of acid gas and liquid can be used to produce synthetic gypsum, which may be sold. In the spray dryer absorber system, a fine mist of lime slurry is injected into the flue gas as it enters the spray dryer absorber, where the acid gas and lime reaction takes place. In a dry scrubbing system, without the use of a slurry, the reagent is injected directly into the duct in a dry, powdery sodium compound. In both the spray dryer absorber and dry injection designs, the systems will typically be coupled with baghouses to get further acid gas removal. The baghouse acts as a secondary acid gas removal stage, with removal efficiencies sometimes as high as 15–20%.
9.3.3 Wet Scrubber Technology 9.3.3.1 Basic Principles Wet scrubber systems utilize a liquid sorbent that is sprayed into the flue gas. The available reagents for wet scrubbers are lime (CaO), limestone (CaCO3), magnesium-enhanced lime (MgO and CaO), ammonia (NH3), and sodium carbonate (Na2CO3). However, the reagents utilized most often are lime or limestone. Liquid lime/limestone will be sprayed into the flue gas in an absorber vessel. The acid gas comes in contact with the lime/ limestone and reacts with the liquid. With additional treatment, the mixture of acid gas and liquid can produce synthetic gypsum that may be sold to various vendors. 9.3.3.2 Typical Designs/Scale of Operations In general, there are three main arrangements for wet scrubbers: electrostatic precipitators (ESPs) with single-stage wet scrubbing, baghouse with single-stage wet scrubbing, and two-stage scrubbing. Single-stage wet scrubbing is the most common arrangement; an electrostatic precipitator (ESP) precedes the wet scrubber in this arrangement. If the required removal efficiency is 85% or less, a gas bypass is used to pipe
368 Combustion Engineering Issues for Solid Fuel Systems
a fraction of the unreacted flue gas. This fraction is then mixed with the treated flue gas to reheat the flue gas before it exits the stack. On the other hand, if a higher efficiency is required, entrainment separators are used to remove slurry droplets. All the flue gas is then reheated 25–50 F above the dew point before it exits the stack. When an operation is burning certain types of fuels, particularly western U.S. fuels, the required size of an electrostatic precipitator may be very large to be able to meet the emissions standard. As a result, fabric filters need to be used prior to the wet scrubber. Two-stage scrubbing involves SO2 absorption first in a venturi scrubber and then finally in a second-stage absorber. Wet scrubbing in this manner is particularly useful when the fly ash contains higher quantities of calcium, magnesium, and sodium. Figure 9-14 is a generic schematic showing the major components and layout utilized for a wet scrubber system. These are the limestone storage, pulverizer, feed slurry storage tank, absorber reaction tank and absorber tower, and the synthetic gypsum production system. The reaction tank in this system has to be sized to provide sufficient time for sulfur components to precipitate and for the dissolution of additives to occur. In the design and operation of wet scrubbers, the two major problems that are encountered are corrosion and wet-dry interface scaling; these problems can be minimized by use of special construction materials and limited recycling of scrubber solutions. As the level of dissolved solids in the lime/limestone slurry increases, scaling potential also increases. If excess scaling is experienced, large pressure drops of the gas can cause a shutdown. When one is designing an absorber tower, the relationship between the overall mass transfer coefficient and factors such as gas velocity, tower height, number of stages, stage spacing, nozzle characteristics, and liquid-to-gas (L/G) ratios must be considered. In addition, wet scrubber systems also require a water treatment facility.
Flue Gas To Stack Wash Water
Recycle Water From Synthetic Gypsum System
Limestone Storage
Pulverizer
Feed Slurry Storage Tank
Limestone Feed Slurry
Absorber Tower
Absorber Reaction Tank
Absorber Spray Headers Synthetic Gypsum Production System
Oxidation Air
FIGURE 9-14 Schematic diagram of generic wet scrubber system components and layout.
Post-Combustion Emissions Control
369
Wet scrubbers require a relatively small amount of space; have low capital cost; and can process high-temperature, high-acidity, and high-humidity flue gas streams. Consequently, wet scrubbers are commonly used in large utility boilers. 9.3.3.3 Efficiencies Geometric shape, injection locations, gas residence time, gas velocities, and several other factors determine the performance and efficiency of wet scrubber systems. New systems will typically achieve SO2 removal efficiencies of 95% and sometimes even better. Additionally, according to some sources, HCl removal can be 95% or better, and HF removal can be in excess of 33%. The rate of sulfate ion oxidation also affects the performance and reliability of the scrubbing process. Typically, high sulfate-to-sulfite ratios are easier to dewater, resulting in more stable disposal products. There are two dominant oxidation processes; these are natural and forced oxidation. With natural oxidation, a reaction between the soluble sulfite in the slurry and oxygen in the flue gas results in the conversion of sulfite to sulfate. Forced oxidation uses compressed air that is injected into the process, typically in the reaction tank, promoting oxidation with efficiencies ranging from 90% to 99% plus. Factors influencing the oxidation rate are oxygen concentration in the flue gas, ionic strength, slurry pH, and the L/G ratio.
9.3.4 Spray Dryer Absorbers 9.3.4.1 Basic Principles In the spray dryer absorber system, a fine mist of calcium-based slurry is injected into the flue gas as it enters the spray dryer absorber. This mist of slurry is atomized by rotary cup spray atomizers or dual-fluid nozzles. As the water in the spray droplets evaporate, the flue gas is cooled from higher temperatures down to the range of 160–180 F; this temperature range is dependent on the relationship between approach to saturation and removal efficiency. In the absorber, SO2 is absorbed by the droplets, which then enable the reaction between the lime and SO2. The desulfurized flue gas, reaction products, unreacted lime, and fly ash then exit the absorber and enter the baghouse, where further acid gas removal takes place. If the dry material is recycled from the baghouse or sometimes ESP and mixed with fresh slurry, synthetic gypsum can be produced if the mixture is oxidized. Spray dryer absorber systems are typically used for low-sulfur coals. 9.3.4.2 Typical Designs/Scale of Operation Some of the major components of spray dryer absorber systems are absorber reactor, sorbent tank, reagent tank, baghouse, I.D. fan, and the fly ash handling system, as shown in Figure 9-15. The most critical of these components is the spray dryer absorber, with designs varying from single inlets
370 Combustion Engineering Issues for Solid Fuel Systems
FLUE GAS
BAGHOUSE REACTOR
SORBENT SLURRY TANK
I.D. FAN
FLY ASH HANDLING SYSTEM
H2O REAGENT
FIGURE 9-15 Typical spray dryer absorber and particulate control system components and arrangement.
and single rotary atomizers to three inlets and atomizers. The amount of flue gas entering the scrubber dictates the spray dryer absorber design utilized. Baghouses are used more often than precipitators in the spray dryer absorber design, since they facilitate further acid gas removal. In this system, the reacted calcium chloride, calcium sulfate, and calcium carbonate will exit at the bottom of the reactor. A cyclonic section of the reactor or a separate cyclone will separate the flue gas and reacted solids. The flue gas will continue to the baghouse where further removal of the acid gas and particulate take place. 9.3.4.3 Efficiencies Variables such as the type and quality of the additive used, the degree of dryness achieved by the spray dryer, the heat available for drying, and the amount of solids product recycled to the atomizer all affect the efficiency that a spray dryer absorber can achieve. The type and quality of the additive used are affected essentially by the rate of dissolution and the pH of the dissolved additive. Spray dryer absorbers typically cannot operate at saturation, since the droplets cannot dry sufficiently; drying efficiencies are directly related to how closely the outlet conditions approach saturation. An important indicator of how much evaporative heat is available from the flue gas is the spray-down temperature. Spray-down temperature is the difference between the spray dryer absorber inlet and outlet temperatures which corresponds to the amount of water supplied to the atomizing system. When some of the recycle solids product from either the fabric filter or ESP are mixed with fresh lime slurry, lime utilization has been shown to decrease [3]. Spray dryers can typically achieve SO2 removal efficiencies ranging anywhere from 70–95% [4].
Post-Combustion Emissions Control
371
9.3.4.4 Waste Streams Waste streams in spray dryer absorber systems will typically contain calcium sulfite [CaSO3], calcium sulfate [CaSO4], calcium hydroxide [Ca(OH2)], and ash. Calcium sulfite particles will typically be present in the form of thin platelets, whereas calcium sulfate particles will be in thick and rod-shaped crystal form. The actual morphology present will have significant influence on processes such as dewatering of the wet sludge. If the structure is crystalline, the solids will appear dry and stable; however, under vibration, the material may re-liquefy and flow. A rod-shaped crystal will typically settle more rapidly and retain less water than the crystalline structure. The majority of the waste from this system is collected in the baghouse with a portion stored in a recycle storage silo. Wastes in storage silos eventually get mixed with the lime slurry to increase reagent utilization. In addition, before the waste is loaded onto trucks for disposal or sale, pug mills must be used to treat the waste products. If oxidized, synthetic gypsum can be produced and sold to certain vendors. Other potential uses of the waste are in agricultural soil conditioning and in the preparation of bricks or aggregates when mixed with other waste products like fly ash.
9.3.5 Dry Injection Systems 9.3.5.1 Basic Principles In a dry scrubbing system, without the production of a slurry, the reagent is injected directly into the duct as a dry, powdery compound and can be calcium or sodium based. Again, a baghouse is typically used in this design for further acid gas removal. In sodium-based systems, the sodium compounds will contain ores such as nahcolite and trona which appear to be most cost effective among the available ores. These ores will commonly be composed of sodium bicarbonate (NaHCO3), sodium carbonate (Na2CO3), and sodium sesquicarbonate (Na2CO3 NaHCO3 2H2O). The following represents the major chemical reactions of SO2 with the sodium compounds: 2NaHCO3 þ SO2 þ 1=2 O2 ) NaSO4 þ 2CO þ H2 O 2ðNa2 CO3 NaHCO3 2H2 OÞ þ 3SO2 þ 3=2 O2 ) 3Na2 SO4 þ 4CO2 þ 5H2 O
[9-5] [9-6]
To achieve sufficient removal efficiencies, injection temperatures need to be approximately 275 F (135 C). If flue gas temperatures are too high, sometimes the gas stream is quenched to achieve the required temperature range of 270–280 F (132–138 C). When the gas temperature is lowered, condensation of volatile vapors of metal chlorides, arsenic, mercury, dioxins, and other organics can take place on the sorbent and fabric filter.
372 Combustion Engineering Issues for Solid Fuel Systems
9.3.5.2 Typical Designs/Scale of Operations Dry injection systems are utilized more commonly on a smaller scale such as industrial boilers and municipal solid waste incinerators. Some of the major components that must be considered in the design of dry injection systems are reagent pulverizer, injection fan, reagent storage bin, and baghouse, as shown in Figure 9-16. 9.3.5.3 Efficiencies Dry injection systems are capable of efficiencies greater than 90% as long as the stoichiometric ratios, i.e., molar ratio of alkali to sulfur, are in the range of 1.0 to 2.0. Typically, dry injection systems will have removal efficiencies ranging from 50–70%. In addition, as much as 23% NOx reduction can be achieved.
9.3.6 Reactions For scrubber systems which utilize lime or limestone, the overall reactions for lime and limestone are shown in reactions [9-7] and [9-8], respectively. CaO þ SO2 ) CaSO3 CaSO3 þ 1=2 O2 ) CaSO4 CaCO3 þ SO2 ) CaSO3 þ CO2 CaSO3 þ 1=2 O2 ) CaSO4
[9-7]
[9-8]
FLUE GAS
Reagent Storage System Blower
REAGENT PULVERIZER
BAGHOUSE
I.D. FAN
FLUE GAS & REAGENT
FLY ASH HANDLING SYSTEM
FIGURE 9-16 Typical dry injection system and particulate control system component and arrangement.
Post-Combustion Emissions Control
373
Wet scrubber systems using limestone to produce synthetic gypsum will undergo several chemical reactions. These are absorption, neutralization, regeneration, oxidation, and precipitation to produce synthetic gypsum, shown in reactions [9-9] through [9-17] [3]. Absorption: þ SO2 þ H2 O ) H2 SO3 ðHSO 3 þH Þ
[9-9]
Neutralization: H2 SO3 þ SO2 3 ) 2HSO3 H2 SO3 þ HCO 3 ) HSO3 þ H2 CO3
[9-10] [9-11]
Regeneration: þ2 þ 2SO2 CaðOHÞ2 þ 2HSO 3 ) Ca 3 þ 2H2 O
[9-12]
CaðOHÞ2 þ 2H2 CO3 ) Caþ2 þ 2HCO 3 þ 2H2 O
[9-13]
Oxidation: 2 þ HSO 3 þ 1=2 O2 ) SO4 þ H 2 SO2 3 þ 1=2 O2 ) SO4
[9-14] [9-15]
Precipitation: 2 ðm þ nÞCaþ2 þ mSO2 3 þ nSO4 þ xH2 O ) CaðmþnÞ ðSO3 Þm ðSO4 Þn xH2 O
[9-16] Caþ2 þ SO2 4 þ 2H2 O ) CaSO4 2H2 O
[9-17]
For dry scrubber systems, particularly the spray dryer absorber design, several reactions must occur to remove SO2 from the flue gas. These are dissolving of gaseous SO2, dissolving of lime, hydrolysis of SO2, neutralization, and precipitation, as shown in reactions [9-18] through [9-24] [2]. Dissolving of Gaseous SO2: SO2 ðgÞ , SO2 ðaqÞ
[9-18]
374 Combustion Engineering Issues for Solid Fuel Systems
Dissolution of Lime: CaðOHÞ2 ðsÞ ) Caþ2 þ 2OH
[9-19]
Hydrolysis of SO2: þ SO2 ðaqÞ þ H2 O , HSO 3 þH
[9-20]
SO2 ðaqÞ þ OH , HSO 3
[9-21]
Neutralization:
OH þ Hþ , H2 O 2 HSO 3 þ OH , SO3 þ H2 O
[9-22] [9-23]
Precipitation: Caþ2 þ SO2 3 þ 1=2 H2 O , CaSO3 1=2 H2 OðsÞ
[9-24]
Each of these reactions typically occurs as heat is transferred from the flue gas to the slurry droplet and then evaporates the water. By having a high pH in the droplet environment (10 to 12.5), a low concentration of acid is maintained in the liquid phase, thereby enhancing SO2 absorption from the flue gas. In general, for both the wet and dry scrubber systems utilizing either lime or limestone, as the pH level increases, the SO2 removal also increases. The pH level will primarily affect the sulfite (SO32) or bisulfite (HSO3) and bicarbonate (HCO3) or carbonic-acid (H2CO3) equilibrium reactions shown previously. Other factors that affect the SO2 removal rates are the magnesium and chlorine concentrations. Increased magnesium levels will typically improve the SO2 removals by an increase in the soluble alkalinity level. On the other hand, in the absence of magnesium, increasing the chlorine content reduces the SO2 removal efficiency by lowering the pH level. In wet scrubbers, however, if the pH level is too high, considerable carbon dioxide absorption will occur. Carbon dioxide will react with the lime to form calcium carbonate, which is insoluble at high pH levels, and thus decreases the efficiency level. Therefore, an optimal pH level for wet scrubbers is typically around 8.5. 9.3.6.1 Kinetics and Thermodynamics In limestone scrubber systems, the condition of limestone calcine or “burn” can have a significant impact on the lime reactivity. If the temperature at which limestone is calcined is too high, the lime can be deadburned and become unreactive.
Post-Combustion Emissions Control
375
With wet scrubber technology, the rate-limiting reaction is believed to be the dissolution of limestone; this reaction rate can be expressed by equation [9-25] [3]: r ¼ KA½Hþ
[9-25]
where r ¼ rate of limestone dissolution, mol/l-s; K ¼ experimental constant, s/cm2; A ¼ surface area of suspended limestone, cm2; and Hþ ¼ hydrogen ion concentration (10-pH), mol/l. From the relationship of equation [9-25], to increase the dissolution rate, either the surface area or the hydrogen ion concentration must be increased. Typically, limestone is ground to have 80–90% passing through 325 mesh (45 mm) to achieve a reasonable surface area. In the spray dryer absorber design, the rate of SO2 mass transfer, the gas phase heat transfer to the droplet surface, and the mass transfer of the water molecules away from the surface all play critical roles. The SO2 mass transfer rate or diffusion rate can be described by equation [9-26] [3]: dC ¼ ADab ðCbulk C Þ dt
[9-26]
where dC/dt A Dab Cbulk C*
¼ ¼ ¼ ¼ ¼
rate of SO2 mass transfer; droplet surface area; diffusivity of SO2; concentration of SO2 in bulk gas phase; and equilibrium concentration of SO2 that would exist in the gas at the droplet surface.
For the mass transfer of SO2 to occur, the droplet must remain wet for a certain period of time. The rate of mass and heat transfer for a particular water molecule can be understood by equation [9-27] [3]: dM ¼ KAðT Ts Þ dt
[9-27]
where dM/dt ¼ rate of mass transfer; K ¼ combined heat and mass transfer coefficient, experimentally derived; A ¼ droplet surface area;
376 Combustion Engineering Issues for Solid Fuel Systems
T Ts
¼ gas-phase bulk temperature; and ¼ adiabatic saturation temperature of gas/liquid system, typically assumed as surface temperature of droplet.
By taking the integral of equation [9-27] from initial to final moisture content, one can determine the time required for the wet-phase SO2 mass transfer to occur. Calculations have shown that more than 90% of the mass transfer is complete within the first second of contact between the gas and slurry. Furthermore, the larger the droplet size, the longer the drying time.
9.4 NOx Control 9.4.1 Introduction The formation of NOx is generated primarily by three forms: fuel NOx, thermal NOx, and prompt NOx. Fuel NOx is produced by the oxidation of nitrogen inherent in the fuel source. Thus, fuels with high nitrogen content such as coal will produce greater quantities of NOx. When the combustion temperature in the burner zone exceeds approximately 2,650 F (1,950 C), thermal NOx is formed from the oxidation of the nitrogen in the air. Prompt NOx, which produces the least amount of NOx, is formed from the oxidation of hydrocarbon radicals near the combustion flame. In a combustion unit, typically 95% of the NOx is in the form of nitric oxide (NO) with the remaining 5% existing in the form of nitrogen dioxide (NO2). Nitrogen dioxide is an unstable molecule at high temperatures. As a result, most of the NOx emissions from stacks exist in the form of NO. However, when the NO exits the stack, it then converts to NO2. Several removal systems are available; the application of each type will depend but is not limited to factors such as boiler size, temperature, other post-combustion equipment and their corresponding size, monetary budget, and real estate or space available. Conventionally, low NOx burners, staged combustion, and gas recirculation or reburn are used to reduce the NOx levels leaving the boiler. However, with increasingly stringent emissions control, these techniques mainly complement the reduction of NOx rather than act as the primary means of removal. Selective catalytic reduction (SCR) and selective noncatalytic reduction (SNCR) are the predominant methods of NOx reduction, with more emphasis on SCR as the emissions levels become more stringent. The two systems have traditionally been used separately; however, hybrid NOx control systems have recently been introduced. The hybrid systems incorporate both a redesigned SNCR system along with a compact induct SCR system (typically, single-layer catalyst). The SNCR system acts as the initial stage of NOx removal, and the SCR system provides the secondary and final NOx removal stage. When the two traditional systems
Post-Combustion Emissions Control
377
are incorporated, a lower capital cost is realized for a system that can still achieve removal efficiencies of approximately 80%.
9.4.2 Post-Combustion Technologies of Significance 9.4.2.1 Selective Noncatalytic Reduction (SNCR) SNCR technology injects a reagent, typically ammonia or urea, entrained in a carrier fluid such as water, steam, or air directly into the combustion source. SNCR injection is based on the homogeneous gas-phase reaction between the flue gas nitric oxide (NO) and ammonia (NH3) producing products of reaction—diatomic nitrogen (N2) and water (H2O). Typically, these systems will achieve 30–50% NOx reductions with proper design. If an ammonia-based system is used, the ammonia will react with the NOx through reaction [9-28]: 2NO þ 4NH3 þ 2O2 þ 2H2 O ) 3N2 þ 8H2 O
[9-28]
In some cases, the permitting process and storage facility required for an ammonia system are difficult. As a result, diluted urea systems must then be utilized. Reaction [9-29] depicts the relationship between urea and NOx. Notice that N2 and H2O are not the only products of the reaction, as CO2 is also produced: COðNH2 Þ2 þ 2NO þ 1=2 O2 ) 2N2 þ CO2 þ 2H2 O
[9-29]
Both reactions [9-28] and [9-29] consider only the more dominant NOx form. For both systems, the optimal temperature range is 1,600–2,000 F (870–1,095 C). If the NO and ammonia or urea reaction occurs above 2,200 F (1,205 C), more NOx is produced by reaction [9-30]. On the other hand, if the reaction occurs below 1,600 F (870 C), ammonia may pass through unreacted. Unreacted ammonia, known as ammonia slip or breakthrough, if in large enough quantities will react with sulfur trioxide to form ammonia sulfate [(NH4)2SO4] or ammonium bisulfate [NH4HSO4]. Ammonia sulfate will contribute to plume formation, whereas ammonium bisulfate will cause fouling and corrosion problems to equipment downstream such as the air heaters. 2NH3 þ 5=4 O2 þ 1=2 H2 O ) NO þ 2H2 O
[9-30]
Temperature control is a very crucial factor to the success of this system. Sometimes this temperature window cannot be achieved by the combustion unit. The addition of hydrogen gas to aid in the reaction effectively lowers this temperature window to 1,300 F (700 C), allowing for more flexibility with respect to injection locations.
378 Combustion Engineering Issues for Solid Fuel Systems
9.4.2.2 Selective Catalytic Reduction (SCR) SCR utilizes either ammonia or urea in a carrier fluid such as steam or air and injects the mixture through an injection grid into the flue gas. The mixture of flue gas and ammonia or urea will then travel to a catalytic reactor, where essentially only N2 and H2O are formed. Collection efficiencies will range from 30% to over 90% NOx removal depending on such factors as catalyst volume, flue gas sodium and potassium concentrations, and age of the catalyst. The estimated catalyst life will range from 2–5 years, depending on each application. SCR systems will tend to be much higher in capital cost than SNCR systems; however, their achievable efficiencies will either double or sometimes triple that of the SNCR system. As emission regulations become more stringent, installing an SCR system may be the only means for meeting them. The reaction mechanisms for NO and NO2 are as follows: 4NO þ 4NH3 þ O2 ) 4N2 þ 6H2 O
[9-31]
2NO2 þ 4NH3 þ O2 ) 3N2 þ 6H2 O
[9-32]
Again, because NO is the dominant NOx form, reaction [9-31] has the greatest impact on the system. Kinetics and Thermodynamics The most optimal temperature range for an SCR system is between 600 F and 800 F (315 C and 425 C). If the temperature is below this range, significant ammonium sulfate and ammonium bisulfate may be formed and consequently corrode the catalyst. On the other hand, if the temperature range is exceeded, catalyst sintering and damage may occur. In addition, the injection distance is critical to maximize the amount of mixing before it reaches the catalyst bed. Space velocity is another important factor for determining the NOx removal level. Space velocity can be defined as the total flue gas volume per hour divided by the bulk volume of the catalyst [measured in hour1]. Lastly, in order to achieve efficiency in excess of 80%, the molar ratio of NH3 to NO has to be in excess of 1. System Designs SCR systems can be placed in several different locations downstream of the combustion unit. When the SCR reactor is located at the economizer outlet, this is referred to as the hot side or high dust location, since the flue gas will be higher in temperature and particulate matter content. On the other hand, if the reactor is located downstream of the air heater, particulate control system, and flue gas desulfurization system (if installed), this is located on the cold side or low dust region. Less damage will be incurred by the catalyst bed; however, reheating of the flue gas may be necessary for the reaction to occur. This ultimately increases the operating cost along with the capital cost.
Post-Combustion Emissions Control
379
Typically, SCR design is equipped with an economizer bypass which can be used for low-load operation. By diverting part of the hot flue gas from the economizer section through the bypass duct and mixing it with the cooler flue gas, one can accomplish an increase in temperature. This ensures SCR operation within the temperature window that is optimum for NOx reduction. As a consequence, some boiler efficiency reduction takes place because more fuel is required to meet the steam output temperature. Other factors such as residence time, degree of mixing, and stoichiometric ratio also affect removal efficiency. As with most reactions, the higher the residence time, the higher the NOx removal rate. Residence time is typically expressed as space velocity, which is defined as the inverse of residence time and can be experimentally determined through measurement. To have sufficient contact between the reactants, the degree of mixing between the flue gas and reagent becomes very important. The reagent will be injected in a pressurized gas-phase and may also be injected with a carrier fluid such as steam or air to increase the penetration. Theoretically, a stoichiometric ratio of 1 should be sufficient for high NOx removal. However, this linear assumption is sufficient only up to about 85% removal. To achieve higher removal efficiencies, a higher stoichiometric ratio is required. Catalyst Types SCR catalysts consist of active metals or ceramics that incorporate porous structures with activated sites. These sites will have acid groups located on the compound structure end where reduction reactions can take place. As the hot flue gas and reagent diffuse through the catalyst, NOx is chemically reduced at the activated sites to nitrogen and water. Once the reaction is completed, the site where the reduction occurred is reactivated through oxidation. When a catalyst is utilized, the required activation energy is reduced; thus, the reaction rate is increased. Several catalyst types are used, and each can vary in material type and structure. Catalyst types and materials are typically distinguished by three categories: base metal, zeolite, and precious metal. The most common materials are titanium dioxide with small amounts of vanadium, molybdenum, tungsten, or the combination of other reagents. Typically, two structures are used for the catalyst bed: ceramic honeycomb and pleated metal flat plate. These structures provide greater surface area-to-volume ratio while limiting plugging issues. Catalyst selection is based on several criteria: resistance to toxic materials, resistance to abrasion, resistance to thermal cycling, resistance to oxidation of SO2 to SO3, and resistance to plugging. If considerations for fly ash erosion in coal-fired boilers are not properly addressed, catalyst service life will be less than expected. As discussed previously, the optimal temperature range for SCR operation ranges from 600–800 F (315–425 C). Operating below this temperature window will increase the potential for ammonia slip and consequent ammonium sulfate and ammonium bisulfate formation. However, depending on the flue gas SO3 concentration, the
380 Combustion Engineering Issues for Solid Fuel Systems
temperature at which ammonium bisulfate forms will vary. As SO3 and unreacted ammonia concentrations increase, the minimum operating gas temperature also increases. Catalyst Variability Catalyst formation will vary from single component, multicomponent, to active phase with a support structure. The types, materials, and formations used will vary depending on the application. To properly design the catalyst for maximum reduction in surface area and prevention or minimization of catalyst plugging, one must consider pitch size. “Pitch” is a term typically associated with honeycomb and metal plate designs. It affects the flue gas velocity experienced in the interstitial spaces and can be defined as the width of the catalyst cell plus the cell wall thickness. In general, a wider pitch will result in lower gas velocities. In a high-temperature or high-dust SCR, the typical catalyst pitch for a honeycomb structure ranges from 7–9 mm for a coal-fired application. If the environment is cold temperature or low dust, the catalyst size is reduced to the range of 4–7 mm.
9.5 Mercury Control 9.5.1 Mercury Emissions from Existing Control Technologies from Coal-Fired Power Plants Estimates for mercury emissions from coal-fired power plants with various control technologies, based on the EPA’s 1999 Information Collection Request (ICR) data [1], are given in Table 9-2 [5]. These data show that mercury emissions were estimated at 49 short tons in 1999. This estimate was based on 84 units tested in the third phase of the ICR (out of more than 1,100 units in the United States) and, since there are questions on bias based on the number of samples from eastern versus western coal-fired boilers, various estimates of mercury emissions range from 40–52 short tons/year [5]. The ICR data indicate that the speciation of mercury exiting the stack of the boilers is primarily gas-phase oxidized (i.e., 43%) or elemental (i.e., 54%) mercury with some particulate-bound (i.e., 3%) mercury present [6]. Table 9-2 provides information on the influence of various existing air pollution control devices (APCDs) on mercury removal; however, mercury capture across the APCDs can vary significantly based on coal properties; fly ash properties, including unburned carbon; specific APCD configurations; and other factors [6]. Mercury removals across cold-side ESPs averaged 27%, compared to 4% for hot-side ESPs [5]. Removals for fabric filters were higher, averaging 58%, owing to additional gas-solid contact time for oxidation. Both wet and dry FGD systems removed 80–90% of the gaseous oxidized mercury, but elemental mercury was not affected. High mercury removals (i.e., 86%) in fluidized-bed combustors with fabric filters were attributed to mercury capture on high carbon content fly ash.
381
Post-Combustion Emissions Control TABLE 9-2 Estimated Mercury Removal by Various Control Technologies
Control Technology ESP cold ESP cold þ FGD wet ESP hot Fabric filter Venturi particulate scrubber Spray dryer þ fabric filter ESP hot þ FGD wet Fabric filter þ FGD wet Spray dryer þ ESP cold FBC þ fabric filter Integrated GasificationCombined Cycle FBC þ ESP cold Totals
Short Tons of Mercury Entering
No. of U.S. Power Plants
Number of ICR Part III Test Sites
Estimated Mercury Removals (%)a
Hg Emission Calculation, EPRI ICR (short tons)
39.4 16.8 5.5 2.9 2.2
674 117 120 58 32
18 11 9 9 9
27 49 4 58 18
28.8 8.6 5.3 1.2 1.8
1.6
47
10
38
1
1.6 1.5 0.3 3.4 0.07
20 14 5 39 2
6 2 3 5 2
26 88 18 86 4
1.2 0.2 0.2 0.5 0.1
0.02 75.3
1 1,128
1 84
— —
0.1 48.8
a Removals as percentage of mercury in coal calculated by EPRI Note: EPRI, Electric Power Research Institute; ICRs, Information Collection Requests; ESP, electrostatic precipitator; FGD, flue gas desulfurization; FBC, fluidized-bed combustion Source: Modified from [5]
Differences in mercury emissions as a function of coal rank are among the most significant findings of the ICR and subsequent DOE testing, as can be observed in Table 9-3. Table 9-3 lists the ranges (note that values less than zero in the ICR data, due to mercury measurement limitations, have been changed to zero for averaging purposes) and average co-benefit mercury capture by coal rank and APCD configuration [7]. Specifically, it is the fact that units burning subbituminous coal and lignite frequently demonstrate worse mercury capture than similarly equipped bituminous coal-fired plants. An exception to this is the case where a subbituminous coal is fired in a boiler containing an SCR and cold-side ESP. The SCR oxidizes the elemental mercury, which enhances the ESP’s effectiveness in capturing the mercury. These data also demonstrate the improved mercury-capture effectiveness of wet FGD systems when an upstream SCR system is in service. For example, average mercury capture for bituminous coal-fired plants equipped with a cold-side ESP and wet FGD increased from 69–86% with the addition of an SCR. FGD systems are more efficient at mercury removal with upstream SCR systems; however, FGD systems also remove significant mercury quantities without an SCR
382 Combustion Engineering Issues for Solid Fuel Systems TABLE 9-3 Average Co-Benefit Mercury Capture by Coal Rank and APCD Configurations APCD Configuration CS-ESP CS-ESPþWet FGD HS-ESP HS-ESPþWet FGD FF FFþWet FGD SDAþFF SDAþCS-ESP PS PSþWet FGD
Average Percentage Mercury Capture (Range of Mercury Capture) Bituminous Coal w/o SCR 28 (0–92) 69 (41–91) 15 (0–43) 49 (38–59) 90 (84–93) 98 (97–99) 98 (97–99) N/A N/A 32 (7–58)
w/ SCR 8 (0–18) 85 (70–97) N/A N/A N/A N/A 95 (89–99) N/A N/A 91 (88–93)
Subbituminous Coal w/o SCR 13 (0–61) 29 (2–60) 7 (0–27) 29 (0–49) 72 (53–87) N/A 19 (0–47) 38 (0–63) 9 (5–14) 10 (0–74)
w/ SCR 69 (58–79) N/A N/A N/A N/A N/A N/A N/A N/A N/A
Lignite w/o SCR 8 (0–18) 44 (21–56) N/A N/A N/A N/A 4 (0–8) N/A N/A 33 (9–51)
Note: CS-ESP, cold-side ESP; HS-ESP, hot-side ESP; PS, particulate scrubber; SDA, spray dryer absorber; FF, fabric filter; SCR, selective catalytic reactor; FGD, flue gas desulfurization Source: [7]
system. The data indicate that for pulverized coal-fired units, the greatest co-benefit for mercury control is obtained for bituminous coal-fired units equipped with a fabric filter for particulate matter control and either a WFGD or spray dryer absorber for sulfur dioxide control. In these cases, average mercury capture of 98% was observed for both cases. The worst performing pulverized bituminous coal-fired units were those equipped only with a hot-side ESP [6]. The rank-dependency on mercury removal is due to the speciation of the mercury in the flue gas, which can vary significantly between power plants depending on coal properties. Power plants that burn bituminous coal typically have higher levels of oxidized mercury than power plants that burn subbituminous coal or lignite, which is attributed to the higher chlorine and sulfur content of the bituminous coal. The oxidized mercury, as well as the particulate mercury, can be effectively captured in some conventional control devices such as an ESP, fabric filter, or FGD system, whereas elemental mercury is not as readily captured. The oxidized mercury can be more readily adsorbed onto fly ash particles and collected with the ash in either an ESP or fabric filter. Also, because the most likely form of oxidized mercury present in the flue gas, mercuric chloride (HgCl2), is water soluble, it is more readily absorbed in the scrubbing slurry of plants equipped with wet FGD systems compared to elemental mercury, which is not water soluble [6]. The use of SCRs to significantly increase oxidation and improve removal of mercury is actively being pursued.
Post-Combustion Emissions Control
383
9.5.2 Mercury Legislation On May 18, 2005, the U.S. Environmental Protection Agency (EPA) issued a final regulation for the control of mercury emissions from coal-fired power plants [8]. The Clean Air Mercury Rule (CAMR) establishes a nationwide cap-and-trade program that will be implemented in two phases and applies to both new and existing plants. Based on 1999 power plant emissions estimates, the cap-and-trade provision of the rule would reduce mercury emissions by almost 70% from 48 tons/y to 38 tons/y by 2010 and to 15 tons/y in 2018 [9]. As a result, CAMR requires an overall average reduction in mercury emissions of approximately 69% to meet the Phase II emissions cap. Meanwhile, several states have adopted, or are considering, legislation that will impose more stringent regulations on mercury emissions from coal-fired boilers than those included in CAMR [10]. EPA has projected that co-benefit mercury reductions achieved through further SO2 and NOx emissions controls required under the Clean Air Interstate Rule (CAIR) will likely enable industry compliance with the Phase I cap. However, both improvements in co-benefit mercury capture and development of new mercury control technologies will be needed to achieve the level of control necessary to meet the Phase II cap [7].
9.5.3 Technologies for Mercury Control Many research organizations, federal agencies, technology vendors, and utilities are actively in the process of identifying, developing, and demonstrating cost-effective mercury control technologies for the electric utility industry. The goal is to have technologies ready for commercial demonstration by 2007 that can reduce uncontrolled mercury emissions by 50–70% and greater than 90% by 2010, while reducing cost by 25–50% compared to baseline cost estimates of $60,000/lb mercury removed (1999 dollars). Approaches for controlling mercury include coal treatment/combustion modifications, sorbent injection, and FGD enhancement/oxidation [11]. To date, use of activated carbon injection (ACI) has shown the most promise as a near-term mercury control strategy. The DOE and others are conducting field tests of a number of alternative approaches to enhance ACI mercury-capture performance for both bituminous and low-rank coal applications, including the use of chemically treated powdered activated carbons (PACs) that compensate for low chlorine concentrations in the combustion flue gas. Other mercury control technologies are being tested to enhance mercury capture for plants equipped with wet flue gas desulfurization (FGD) systems. These FGD-related technologies include (1) coal and flue gas chemical additives and fixed-bed catalysts to increase levels of oxidized mercury in the flue gas; and (2) wet FGD chemical additives to promote mercury capture and prevent re-emission of previously captured mercury from the FGD absorber vessel [9]. A summary of PAC injection and wet FGD technologies for mercury control is provided in the following sections.
384 Combustion Engineering Issues for Solid Fuel Systems
9.5.3.1 Sorbent Injection Although existing APCD can capture some mercury, innovative control technologies will be needed to comply with the CAMR Phase II mercury emissions cap. To date, ACI has shown the most promise as a near-term mercury control technology, although continuous long-term operation is required to determine the effect on plant operations. In a typical configuration, PAC is injected downstream of the power plant’s air heater and upstream of the particulate control device—either an ESP or a fabric filter. The PAC adsorbs the mercury from the combustion flue gas and is subsequently captured along with the fly ash in the particulate control device. A variation of this concept is the TOXECON TM process in which a separate baghouse is installed after the primary particulate collector (especially when it is a hot-side ESP) and air heater, and PAC is injected prior to the TOXECON TM unit i.e., TOXECON I TM. This concept allows for separate treatment or disposal of fly ash collected in the primary particulate control device. A variation of the process is the injection of PAC into a downstream ESP collection field to eliminate the requirement of a retrofit fabric filter and allow for potential sorbent recycling i.e., TOXECON II TM configuration. Overview of Powdered Activated Carbon Injection for Mercury Control The performance of PAC in capturing mercury is influenced by the flue gas characteristics, which are determined by factors such as coal type, APCD configuration, and additions to the flue gas including SO3 for flue gas conditioning [12]. Research has shown that HCl and sulfur species (i.e., SO2 and SO3) in the flue gas significantly impact the adsorption capacity of fly ash and activated carbon for mercury. Specifically, it has been found that [11] HCl and H2SO4 accumulate on the surface on the carbon; HCl increases the mercury removal effectiveness of activated car-
bon and fly ash for mercury, particularly as the flue gas concentration increases from 1 to 10 ppm. The relative enhancement in mercury removal performance is not as great above 10 ppm HCl. Other strong Brnsted acids such as the hydrogen halides—HCl, HBr, or HI—should have a similar effect. Halogens such as Cl2 and Br2 should also be effective at enhancing mercury removal effectiveness, but this may be the result of the halogens reacting directly with mercury rather than the halides, thereby promoting the effectiveness of the activated carbon; and SO2 and SO3 reduce the equilibrium capacity of activated carbon and fly ash for mercury. Activated carbon catalyzes SO2 to H2SO4 in the flue gas. Because the concentration of SO2 is much higher than mercury in the flue gas, the overall adsorption capacity of mercury is likely dependent on the SO2 and SO3 concentrations in the gas because they form H2SO4 on the surface of the carbon.
Post-Combustion Emissions Control
385
100
Mercury Removal (%)
80
60
Brayton Point: Bit-ESP (DARCO Hg)
40
Yates 1: Bit-ESP (Super HOK) Gaston: Bit-FF (DARCO Hg) Pleasant Prairie: Subbit-ESP (DARCO Hg) Holcomb: Subbit-SDA/FF (DARCO Hg)
20
Monroe: 60:40 Subbit-Bit blend-ESP (DARCO Hg) Leland Olds: Lignite-ESP (DARCO Hg) Stanton 10: Lignite-SDA/FF (DARCO Hg )
0 0
5
10 15 20 25 ACl Concentration (lb/MMacf)
30
35
FIGURE 9-17 Mercury removal as a function of activated carbon injection concentration (Source: [11]). Note: Legend lists power plant name: coal rank-APCD configuration (PAC type).
Figure 9-17 illustrates some results from conventional PAC injection tests performed through numerous DOE test programs [11]. Conventional PAC injection was the focus of initial field testing (which was in 2001–2002) and serves as the benchmark for all field PAC injection tests. This work showed that a maximum of approximately 65% mercury capture could be achieved when firing a subbituminous coal in a power plant using an ESP. Also, when using conventional PAC costs of 50¢/lb, it was found that the cost to remove 70% mercury from a bituminous coal-fired plant with an ESP would cost $70,000/lb of mercury removed, which is higher than the targets listed earlier [13]. The conventional PAC testing was followed by work with chemically treated PACs that were developed for low-rank coal applications due to the low mercury-capture results in the initial testing. Some results from the chemically treated PAC testing are shown in Figure 9-18 [11]. With PAC injection at 1 lb/MMacf (million actual cubic feet) of flue gas, mercury removal ranges from 70–95% for low-rank coals. For 90% mercury removal, costs are estimated at $6,000/lb mercury removed for a subbituminous coal/SDA-FF combination to <$20,000/lb mercury removed for a subbituminous coal/ESP combination assuming chemically treated PAC costs of $0.75 to $1.00/lb [11]. Results from testing using the TOXECON TM technology are given in Table 9-4 (fabric filter configuration) and Figure 9-19 (ESP configuration) [12]. In the fabric filter configuration, mercury removals of 70–90% have
386 Combustion Engineering Issues for Solid Fuel Systems 100
Mercury Removal (%)
80 Low-rank coal; Total Hg removal 60 Bituminous coal; ACl Hg removal 40
Meramec: Subbit-ESP (DARCO Hg-LH) Stanton 1: Subbit-ESP (B-PAC) Dave Johnston: Subbit-ESP (MerClean 8) Holcomb: Subbit-SDA/FF (DARCO Hg-LH) St. Clair: 85:15 Subbit/Bit blend-ESP (B-PAC) Stanton 10: Lignite-SDA/FF (DARCO Hg-LH) Lee: Bit-ESP (B-PAC )
20
0 0
1
2 3 4 5 6 ACl Concentration (lb/MMacf)
7
8
FIGURE 9-18 Mercury removal as a function of chemically treated activated carbon injection concentration (Source: [11]). Note: Legend lists power plant name: coal rank-APCD configuration (PAC type). TABLE 9-4 Demonstrated Mercury Removal at 2 or 5 lb/MMacf ESP (5 lb/ MMacf)
ESP with SO3a (5 lb/MMacf)
Low S, very low Cl (PRB or North Dakota lignite) Low S, >50 ppm Cl (Some Texas lignites)
78–95% with brominated PAC 78–95%
40–91% with brominated PAC 40–91%
70–90%
Low S, bituminous coal Low S, bituminous coal with SCR Low S, bituminous coal
55–75% 15–70%
40–91% NAc
NDb ND
90–95% with brominated PAC 90–95% with brominated PAC ND ND
<15%
NA
NA
NA
TOXECON TM (2 lb/MMacf) 70–90%
SDAþFF (2 lb/MMacf)
a
SO3 from SO3 injection Not available c Not applicable or configuration unlikely Source: Modified from [12] b
been achieved using low-rank coals, while removal efficiencies of 50–90% are expected when using low-sulfur bituminous coals [12]. From Figure 9-19, it can be seen that 50–80% mercury reduction was achieved when injecting chemically treated PAC at 4–5 lb/MMacf into the next-to-last ESP field (i.e., F5) and last ESP field (i.e., F7) [11].
Post-Combustion Emissions Control
387
100
Mercury Removal (%)
80
60
40
DARCO Hg, F5 DARCO Hg, F7 DARCO Hg-LH, F5 DARCO Hg-LH, F7 DARCO Hg-LH, F5 & F7
20
0 0
2
4 6 8 ACl Concentration (lb/MMacf)
10
FIGURE 9-19 Mercury removal as a function of activated carbon injection concentration using the TOXECON II TM configuration at the Independence Station (Source: [11]). Note: DARCO Hg is a conventional PAC, whereas DARCO Hg-LH is a brominated PAC.
Balance-of-Plant Issues Long-term tests are necessary to evaluate the effect of PAC injection on plant performance, and projects addressing this issue have recently started. To date through short-term field tests with plants using ESPs, no impact of PAC injection upstream of the air preheater has been observed. In general, no balance-of-plant issues, including increases in particulate emissions and changes in ESP operation, have been identified as a result of PAC injection in the ductwork upstream of the APCDs. One exception was an increase in the arc rate at low boiler load conditions as compared to baseline arcing, during PAC injection [12]. Particulate measurements collected at the outlet of the ESP when injecting PAC in the TOXECON II configuration indicated that the PAC injection did not impact particulate emissions [12]. However, stack opacity and ESP outlet particulate spikes associated with PAC injection following fourth field raps were observed. Two significant balance-of-plant issues were discovered during TOXECON I TM testing at Presque Isle [12]. The first was that PAC can self-ignite in the hoppers of the baghouse when it is allowed to accumulate and when exposed to external heating from hopper heaters. The second was that the ash/PAC mixture becomes stickier than ash alone and harder to remove from the hoppers and transport with the ash removal system when it is heated. The first issue is being addressed by frequently evacuating the hoppers and controlling the maximum temperature of the heating
388 Combustion Engineering Issues for Solid Fuel Systems
elements to less than 300 F. The second issue can be addressed through equipment design. No balance-of-plant problems have been noted at sites using spray dryer absorbers during PAC injection. This includes opacity or changes in spray dryer absorber or fabric filter operation. 9.5.3.2 Wet Flue Gas Desulfurization Although PAC injection has shown the most promise as a near-term mercury control technology, testing is underway to enhance mercury capture for plants equipped with wet FGD systems. These FGD-related technologies include (1) coal and flue gas chemical additives with fixed-bed catalysts to increase levels of oxidized mercury in the flue gas; and (2) wet FGD chemical additives to promote mercury capture and prevent re-emissions of previously captured mercury from the FGD absorber vessel. There is much interest in these activities because the use of FGD systems at coalfired power plants will increase significantly over the next 15 years due to the implementation of CAIR, which establishes a market-based allowance cap-and-trade program to permanently cap emissions of SO2 and NOx in 28 eastern U.S. states and the District of Columbia. The SO2 emission caps were based on percent reductions from the total number of Title V, Phase II allowances currently allocated to sources in the affected states—a 50% reduction for 2010 and 65% reduction for 2015. When fully implemented, CAIR will reduce SO2 emissions by more than 70% from 2003 levels. Table 9-5 provides a summary of the power generation capacity equipped with FGD controls in 2004, categorized by coal rank and FGD type [7]. Total FGD capacity is projected to increase from 100 GW in 2004 to 150 GW in 2010 to 180 GW in 2015 to 231 GW in 2020 [14]. It is anticipated that most of this will be wet-based systems with dry FGD increasing from 10 GW in 2004 to 21 GW by 2015. Wet FGD systems, especially those associated with bituminous coalfired power plants equipped with SCR systems, appear to be good candidates for capturing mercury. With the projected increase in wet FGD systems for bituminous coal-fired power plants, the co-benefit of capturing of mercury with SO2 can be realized. Research is currently underway to
TABLE 9-5 2004 U.S. Coal-Fired Generation Capacity with FGD Controls, GW
Wet FGD Dry FGD Total FGD Total Coal-Fired Source: [7]
Bituminous Coal
Subbituminous Coal
Lignite
Total
57 4 61 213
24 5 29 96
9 1 10 15
90 10 100 324
Post-Combustion Emissions Control
389
evaluate technologies that facilitate mercury oxidation and to ensure that captured mercury is not re-emitted from FGD systems. Research, to date, is encouraging.
9.6 Carbon Dioxide Capture 9.6.1 Introduction The increased CO2 concentration in the atmosphere from fossil fuel combustion is causing concerns for global warming. The capture and sequestration of CO2 from stationary combustion sources is considered an important option for the control of CO2 emissions because CO2 is likely to be regulated in the future as part of a carbon management program. Currently, however, no cost-effective technologies for coal-fired power plants are available. This section will briefly summarize the leading candidate technology for CO2 capture from coal-fired power plants.
9.6.2 Approaches for Capturing Carbon Dioxide from Coal-Fired Power Plants There are three key technologies for capturing CO2 from coal-fired power plants: oxyfuel combustion, pre-combustion (integrated gasificationcombined cycle [IGCC], which is discussed in Chapter 11), and postcombustion CO2 scrubbing. Each technology must be integrated with CO2 dehydration, compression, transport, and storage, though, in order to sequester the CO2 so that it is not emitted into the atmosphere. Each of these three coal-based technologies is important because the International Energy Agency projects each of the three to reduce CO2 emissions by 1.3 Giga tons/y in 2050 [15]. Here, post-combustion CO2 capture will be discussed.
9.6.3 Post-Combustion Carbon Dioxide Scrubbing Post-combustion CO2 capture from a coal-fired power plant is the most difficult method for removing CO2 from a gas stream of the three technologies presented in the preceding section. In oxyfuel combustion and IGCC operation, a concentrated CO2 stream is produced, from which it is relatively easy to remove the CO2 as compared to that from a conventional combustion flue gas, where the CO2 concentration is only about 12–18% by volume. In general, CO2 can be separated, recovered, and purified from concentrated CO2 sources by chemical and physical methods such as absorption, adsorption, or membrane separation. These separation and purification steps can produce pure CO2 from power plant flue gas, but they add considerable cost to the CO2 conversion or sequestration system. Industrially, separation of CO2 is usually performed utilizing the amine absorption
390 Combustion Engineering Issues for Solid Fuel Systems
process with monoethanol amine (MEA) [16]. The main reaction responsible for CO2 chemical interaction with amine (i.e., chemical absorption) is believed to be the carbamate formation: CO2 þ 2R2 NH $ R2 NHþ 2 þ R2 NCOO
[9-33]
where R is an alkyl group. The Fluor Danial Econamine FG CO2 Recovery Process that was developed by DOW Chemical is a widely used commercial process that uses an amine solution, containing a proprietary additive, to remove CO2 economically from low-pressure, oxygen-containing flue gas streams similar to flue gas [17]. Some large-scale designs have been developed for CO2 recovery from flue gas for use in CO2-enhanced oil recovery [18]. Activated carbons and molecular sieves are readily available commercially, and many studies have been conducted on CO2 adsorption using such materials as well as other adsorbents as zeolites, pillard clays, and metal oxides [19]. In summary, while there are technologies available for capturing CO2 from flue gas, much work needs to be done to make them more economical. Currently, amine scrubbing with CO2 compression to 1,200 psig cost approximately $2,000/kW and reduced the net power plant output by 12.5% [1]. Post-combustion capture is an area that needs further research development, especially with impending legislation on carbon management.
9.7 References 1. Miller, B.G. 2005. Coal Energy Systems. Oxford, UK: Elsevier. 2. Kitto, J.B., and S.C. Stultz. 2005. Steam: Its Generation and Use. Barberton, OH: Babcock & Wilcox. 3. Singer, G.J. 1991. Combustion: Fossil Power Systems, 4th ed. Windsor, CT: Combustion Engineering Inc. 4. Institute of Clean Air Companies. (n.d.). Acid Gas/SO2 Controls. http://www. icac.com/i4a/pages/index.cfm?pageid=3401, accessed December 30, 2007. 5. Pavlish, J.J., E.A. Sondreal, M.D. Mann, E.S. Olson, K.C. Galbreath, D.L. Laudal, and S.A. Benson. 2003. Status Review of Mercury Control Options for CoalFired Power Plants. Fuel Processing Technology. 82: 89–165. 6. Feeley, T.J., J. Murphy, J. Hoffman, and S.A. Renninger. 2003. A Review of DOE/NETL’s Mercury Control Technology R&D Program for Coal-Fired Power Plants. DOE/NETL Hg R&D Review. April www.netl.doe.gov/. 7. Miller, C.E., T.J. Feeley III, W.J. Aljoe, B.W. Lani, K.T. Schroeder, C. Kairies, A.T. McNemar, A.P. Jones, and J.T. Murphy. 2006. Mercury Capture and Fate Using Wet FGD at Coal-Fired Power Plants. DOE NETL Mercury and Wet FGD R&D. August. http://www.netl.doe.gov/technologies/coalpower/ewr/ index.html. 8. The Clean Air Mercury Rule. May 18, 2005. Code of Federal Regulations.
Post-Combustion Emissions Control
391
9. Kilgroe, J.D., C.B. Sedman, R.K. Srivastava, J.V. Ryan, C.W. Lee, and S.A. Thorneloe. 2002. Control of Mercury Emissions from Coal-Fired Electric Utility Boilers Interim Report Including Errata Dated 3–12–02, EPA-600/ R-01-109. Washington, D.C.: U.S. Environmental Protection Agency, U.S. Government Printing Office. 10. State Mercury Program for Utilities. 2007. Washington, D.C.: National Association of Clean Air Agencies. http://www.4cleanair.org/Documents/State Table.pdf. 11. Feeley, T.J. 2006. U.S. DOE’s Hg Control Technology RD&D Program— Significant Progress, But More Work to Be Done! Mercury 2006—Conference on Mercury as a Global Pollutant, Madison, WI, August 6–11. http://www. netl.doe.gov/technologies/coalpower/ewr/index.html. 12. Sjostrom, S., T. Campbell, J. Bustard, and R. Stewart. 2007. Activated Carbon Injection for Mercury Control: Overview. 32nd International Technical Conference on Coal Utilization & Fuel Systems. Clearwater, FL. June 10–16. 13. Jones, A.P., J.W. Hoffmann, D.N. Smith, T.J. Feeley III, and J.T. Murphy. 2007. DOE/NETl’s Phase II Mercury Control Technology Field Testing Program, Updated Economic Analysis of Activated Carbon Injection. May. http://www.netl.doe.gov/technologies/coalpower/ewr/index.html. 14. Multi-Pollutant Regulatory Analysis: CAIR/CAMR/CAVR. 2005. U.S. Environmental Protection Agency. October http://www.epa.gov/airmarkets/mp/ cair_camr_cavr.pdf. 15. Princiotta, F. 2007. Mitigating Global Climate Change through Power-Generation Technology. Chemical Engineering Progress. 103(11), pp. 24–32. 16. Chakam, A. 1997. CO2, Capture Process: Opportunities for Improved Energy Efficiencies, Energy Conversion and Management. 38: S51–S56. 17. Sander, M.T., and C.L. Mariz. 1995. The Fluor Daniel Econamine FG Process: Past Experience and Present-Day Focus. Energy Conversion and Management. 33: 818. 18. Mariz, C.L. 1998. Carbon Dioxide Recovery: Large Scale Design Trends. Journal of Canadian Petroleum Technology. 37(7): 42–47. 19. Miller, B.G., A.L. Boehman, P. Hatcher, H. Knicker, and A. Krishnana, et al. 2000. The Development of Coal-Based Technologies for Department of Defense Facilities: Phase II Final Report, DE-FC22-92PC92162, prepared for the U.S. Department of Energy Federal Energy Technology Center, Pittsburgh, PA, July 31.
CHAPTER
10
Some Computer Applications for Combustion Engineering with Solid Fuels Michael Santucci President ECG Consultants,
James Scavuzzo Vice President ECG Consultants, and
Joe Hoffman
Staff Engineer ECG Consultants
10.1 Introduction The electrical generation industry is a vital element in the economic drive train. Until the last decades of the 20th century, the industry was marginally on the radar screen, as it was functioning as a regulated monopoly and with energy costs to customers that were usually tolerable, and controlled by Public Utility Commissions or Public Service Commissions. Several major developments impacted the status quo, and the effects are still far from settled. The more significant events contributing to the industry flux were the following: Environmental limitations of combustion emissions (particulates,
SO2, NOx and—more recently—mercury with CO2 somewhere near the horizon); 393
394 Combustion Engineering Issues for Solid Fuel Systems Three Mile Island Unit #1 incident—and later the Chernobyl
disaster—marking the curtailment of capital investment in the nuclear fuel option for the United States; The NO GO (in early 1970s) then GO with plentiful, inexpensive natural gas—but now not so plentiful and definitely costly; The escalation of oil prices and trickle down effects to other fuels; and Deregulation of the electric power industry. The industry situation today shows coal-fired generation as the backbone of the electric supply—the coal being an essential resource as a reasonable cost, plentiful, domestically available fuel. The use of coal is also supplemented by other solid fuels (see Chapter 3): petroleum coke, wood waste, agricultural biomass, and a host of industrial products used in niche applications. If full advantage of coal’s availability and optimum use are to be realized in existing plants, the complexity of the total picture requires the integration of all the coal issues related to the supply, transportation, storage, costs, constituent analysis, combustion properties, and contingencies, as well as the understanding of the end product—the electric market with all its nuances. This requires optimum use of computers.
10.1.1 Computer Applications in Combustion Engineering Computers have long been used in combustion engineering—particularly in the areas of combustion analysis and combustion control. Combustion analysis has involved extensive modeling of chemical reactions. Initial models were often simple heat and material balance models with or without temperature calculations. The National Aeronautics and Space Administration (NASA), using the JANAF (Joint Army, Navy, and Air Force) thermodynamic data, produced one of many Gibbs Free Energy Minimization models to calculate products of combustion and temperatures at equilibrium. These models ultimately made their way from mainframe computers to desktop and laptop personal computers, and gained wide acceptance in the analysis of combustion of solid fuels. 10.1.1.1 Analytical Modeling Computer modeling with thermodynamic data extended beyond bulk combustion problems to the management of trace metals and the fate of inorganic matter. Models such as FACTSAGE provide key insights into slagging and fouling, and the viscosity of inorganic matter released from solid fuels during combustion processes (see Chapters 4 and 8). The most advanced computer modeling technique is generally considered to be the use of computational fluid dynamics (CFD) to evaluate combustion within the primary furnace. Its applications extend well beyond
Some Computer Applications for Combustion Engineering with Solid Fuels
395
combustion, but within the combustion arena, it has distinct applications. These include the following: Temperature mapping, including temperature gradient mapping; Flame shaping, including the use of the models to provide guidance
on setting of burners for optimal firing practices; and Evaluation of emissions formation and control, including setting
burners to minimize the formation of NOx and other pollutants. Unlike the Gibbs codes, the CFD models rely more on kinetics of the combustion reactions to drive the results. CFD modeling has achieved wide acceptance in the combustion arena. It is used not only to analyze the phenomena occurring in the furnace and boiler, but also to analyze the flow of gaseous combustion products through post-combustion systems. The extension of CFD modeling now being developed is the use of three-dimensional models such that the practitioner or analyst can enter “inside” the furnace or boiler and experience what is going on. Several universities and commercial companies are developing this approach. The capabilities of the computer have enabled a much greater understanding of the combustion process and boiler design. Prior engineering and design was largely based on the extension of empirical data to the development of new, larger equipment and boilers and, all too often, an expensive trial-anderror solution process to problems. The tremendous data storage capacity and the rapid computational abilities of the computer has brought with it the development of techniques such as complex finite element (FE) models, threedimensional/multimechanism heat transfer problems, computational fluid dynamics, neural network analysis, and data historian systems (e.g., OSI PIW). The OSI PIW system was originally developed to track the complex flows and reactions in the refinery and petrochemical industry. With the system’s inherent ability to efficiently store large volumes of data and the evolution of accurate, reliable instrumentation, the PI and similar systems’ application to boiler and combustion and the allowed extensive parametric studies and their relative influence. Many ancillary programs have been developed to provide functionality to the OSI PI data. “E-notification” is an example of a program that provides an interest criteria communications link with initiation triggered either on a program-to-time basis or width-critical indices of process values. This technique provides operations and management with real-time advice for corrective action and/or management decisions. Finite element analysis with high accuracy of both stresses and strains in complex geometries of mechanical parts has been possible only with the capability of computers to make many calculations in nano-time. This procedure has enabled precise engineering of mechanical parts to provide functional pieces with allowable stresses and minimal material inventory.
396 Combustion Engineering Issues for Solid Fuel Systems
Advanced heat transfer studies are possible much in the same way as the finite element analysis. Simultaneous, multiple heat transfer mechanisms can be analyzed and, together with CFD, have enabled the development of large, high efficiency boilers. CFD programs are used to study the variables of combustion with the numerous criteria including NOx production exothermic gas flows and the mixing process with overfire air. Boiler configuration changes and revised gas flows can be studied with computer runs to verify the results of contemplated changes. 10.1.1.2 Computer Applications for Process Control Computers have been employed for process control of boilers, kilns, and other combustion devices for several decades. These control systems evolved into the distributed control system (DCS) that has replaced the bench board and hard station approach in virtually all large power plants. The DCS approach has not been static. The recent evolution being employed in some generating stations—and proposed to more—is the closed loop “neural network” control system. This system relies on the computer “learning” the consequences of actions taken in the control of a boiler and then incorporating those learned lessons into the actual control of the boiler. This system is empirically driven at the present time and relies on the computer’s capability to learn the consequences of actions; it relies to very little extent on first principles. Neural networks are data processes enabling parametric studies of the many factors that are influential but difficult to discern as to their contribution to an activity and interaction with the multiplicity of other parameters. These programs analyze large volumes of process data generated over time but do require variation of parameters—often intentional disturbances—to distinguish trends of influence. 10.1.1.3 Computer Applications for Fuel Control A recent development in the application of computers for combustion control is the use of computers to monitor and influence the quality of the fuel being fed to the boiler. This approach has become more widely applied as utilities and industries blend fuels (see also Chapter 5 for a discussion of the use of this approach). One such system, AccuTrack, is detailed in this chapter. This use of computers integrates information systems, modeling, and information management for the benefit of combustion engineers and, more significantly, plant operators.
10.2 Background The leaders in the competitive, deregulated field of electric generation must utilize all available techniques to maximize the many aspects of fuel opportunities, and this is especially applicable to coal as the prevailing
Some Computer Applications for Combustion Engineering with Solid Fuels
397
base-load fuel. For instance, the incremental price of a MWh in the electric market swings drastically with higher usage during the day to the low demand in the middle of the night. The summer peaks from air conditioning loads command premium prices that, to a well-managed plant, mean sales opportunities and corporate profits. But the realization of the success requires thorough planning; effective facilities; installation/utilization of appropriate data sources; data processing with meaningful, actionable output; and participative management to see the process as a vital, continuous practice by plant and support personnel. Attentive, knowledgeable management can get it done, but sufficient, accurate data are the key: “One cannot manage what is not measured.” W. Edwards Deming, father of Total Quality Management
In the original planning, the plant siting considerations strongly factored both the coal supply and quality. This coal availability either used local sources such as mine-mouth or readily available transportation facilities such as barge or railroad. Since the plant startup, there may have been situations that required or facilitated alternative coal supplies. The new emissions regulations might have necessitated compliance fuels, often taking advantage of lower sulfur fuels from the northern plain states or selected eastern bituminous coals. Rail and barge conveyance has expanded with greater efficiencies and larger payloads giving transportation alternatives. But the gradual degradation of coal quality from design has been prevalent with the exhaustion of the prime coal seams, changes in mining methods, and economic opportunities for high-value coal; i.e., relatively lower $/MMBtu (million Btu). The last item invariably comes with penalties that might include lower heating value (i.e., low Btu/lb), lower grindability, and higher slagging or fouling tendencies, any of which can cause a boiler derate from full-load capability but can provide substantial benefit in the fuel cost component ($/MWh) during off-peak loads.
10.3 Process for Fuels Opportunity Realization No goal is achieved without research and identification of objectives, methodical design, planning, validation and, finally, a step-by-step execution through fulfillment. In the process for realization of the rewards in fuel opportunities, the following steps are enumerated for accomplishment of objectives: identify current fuels opportunities, validate objectives, develop effective designs, install and implement processes, and report effectively (see Figure 10-1).
10.3.1 Identify Current Fuels Opportunities Too frequently, generation plant practices have accepted a passive role in the fuel selectivity and use for boiler operations. With the increasing fuel
398 Combustion Engineering Issues for Solid Fuel Systems
Fuels Opportunity - Coal Challenges Variable Coal Prices Market Volatility - Fuel and Electricity Emission Regulations Yard/Plant Constraints
Objective to Achieve Success Enable Fuel Flexibility Adapt Fuel for Emission Compliance Provide/Enable Key Information Pursue Operational ExcellenceMaximize Plant Output value Make Data Output Actionable Use Fuel Opportunities while Minimizing Risk thru “Timed Delivery” to Boiler
FIGURE 10-1 Coal management opportunities face challenges that must be met with the attainment of multiple objectives if success is to be realized.
costs and low excess generation margins, the optimum use of fuel for cost containment and generation availability has become increasingly important. The AccuTrack system was developed to address the specific problems and to make the best practical and economic choice to assure the generation that most efficient cost and/or availability during peak demands. The current market of fuel includes high-quality, more costly fuels as well as lesser cost or “high value” (lower $/MMBtu) that, if selected, must be burned under controlled conditions with realization of both the benefits and consequences. A typical situation arising from the random handling of coal is shown in Figure 10-2. Without due consideration of the fuel and its implications to the plant operations, the fuel quality and plant hardware limitations resulted in a sizable, sustained unit derate and a substantial loss of generation sales opportunity during the summer peak. Had there been a thorough understanding of the coal/generation hardware/electric market, provisions for fuel could have been made to capture all or a large portion of the sales opportunity or—more directly—the realization of corporate profits. Although there might be fewer or more categories of available fuels, the example of Figure 10-3 shows the possible grouping of three categories of coal quality. The design fuel is shown with a heating value of 13,500 Btu/lb coal and a Hardgrove Grindability Index of 60 that allowed fullload capability of the boiler with spare mill capability (83% of rated). The “rocket fuel” is a coal with high heating value, and grindability coal would allow full-load operation even with significantly lower mill capacity. Validation of the existing fuel monitoring, coal-handling, storage, and reclaim equipment is required to determine its adequacy to provide the coal
Some Computer Applications for Combustion Engineering with Solid Fuels
399
Full Load NMW 3130 MKB 90% Pulv Cap Factor 10000 Btu/NKW-Hr Heatrate
MCR
450 400 350
NMW
300 250 200 150 100
August Lost Sales Opportunity ~$1.2M
Unit/Pulverizer Limitation w/loaded coal
50
2/28/2007
2/14/2007
1/31/2007
1/3/2007
1/17/2007
12/20/2006
12/6/2006
11/8/2006
11/22/2006
10/25/2006
9/27/2006
10/11/2006
9/13/2006
8/30/2006
8/2/2006
8/16/2006
7/5/2006
7/19/2006
6/7/2006
6/21/2006
5/24/2006
5/10/2006
4/26/2006
4/12/2006
3/29/2006
0
FIGURE 10-2 Unit output shortfalls (derates) and lost sales opportunities resulting from coal quality impacts to pulverizer and/or boiler operations.
selectivity, handling capability, coal analysis, and data to provide “the right coal at the right time.” New equipment and upgrades can take advantage of state-of-the-art advances in hardware design, materials, digital accuracy, and automated features. Economic evaluations of cost and budget justification for an AccuTrack system must take into account both the peak load scenario with the premium coal requirement as well as the off-peak sales potential, the latter through the prudent use of “high-value fuels.” The process development must bring all involved personnel on-board, those from the corporate fuels department, systems sales/generation dispatch, yard personnel, plant/boiler operations, technical support, and management. The process must be well understood and the respective contributions made clear to all participants. Their clearly defined responsibilities must be executed in a timely manner and actionable information made available for appropriate execution.
10.3.2 Validate Objectives and Develop Effective Design With the opportunities defined, the economically justified objectives must be accomplished by acquisition and processing of the critical data necessary
400 Combustion Engineering Issues for Solid Fuel Systems Full Load Fuels 3130 MKB 70% Fineness 14000 Design Fuel
13750
Rocket 13500
BTU
13250
13000
Hi value Mids
12750
12500 100% Mill Capacity
80% Mill Capacity
90% Mill Capacity
12250
12000 45
50
55
60
65 Hardness HGI
70
75
80
85
FIGURE 10-3 Example of fuel sources “A” through “L” with grouping by quality and suitability for full-load capability under various pulverizer capacity availabilities.
to point in the proper direction. This software is built to accomplish a stepwise process with the following elements: Track incoming coal fuel specifications and tonnage from mine-to-
yard-to-bunker-to-burner; Based on numerous criteria including fuel quality, cost, and avail-
ability, recommend coal blend specifications to the bunkers to meet projected generation requirements and economic objectives within the constraints of equipment limitations and regulated emissions; Create a transaction record for each fueling event; and Provide the operator with tools to know current and projected nearterm coal characteristics to optimize the combustion process while avoiding boiler upsets and maintaining emissions consistent with regulatory limitations. The accounting necessity to track coal receipt inventory, use, and value is evident. In addition, operator advisories and economic opportunities will be realized only by the appropriate use of available quality and value of fuel.
Some Computer Applications for Combustion Engineering with Solid Fuels
401
The projected hour-to-hour blended coal quality will be needed to proactively comply with the available electrical sales opportunity, emission limitations, or capacity requirements. Each transaction is a unique fuel record of quantity, quality, and value that will be integrated for composite records. These individual records can consist of a few tons represented by an analyzer sample to as much as a shipload tonnage represented by a characteristic coal analysis. In the process development, all portions of the coal handling and fuel impacts to the boiler operations must be investigated and evaluated against a standard justified by the fuels opportunity study. The items in Figure 10-4 are specific to a particular plant but illustrate the types of equipment deficiencies that must be addressed to implement the earnings potential of the computer-controlled process. During these investigations, deficiencies are commonly found that are the result of inattention to equipment installed as part of the original design but disregarded by both management and the workforce due to the insensitivity of their respective contribution to the overall impact to the plant operations, generation cost optimization, and electric sales/profits potential. For example, all plants were designed with the technology of available scales and coal-sampling equipment. Scale repairs and periodic calibrations come due but might well be relegated to a lesser priority because they are not directly impacting availability, and the same applies to coal-sampling systems. Customary practice degenerates to the point that vendor weights and analysis are accepted without question, and inaccurate weights of fuel input to the boiler are left to be rectified in inventory surveys. These types of coal accounting shortcomings occur despite the dominance of the fuel impact to both the generation costs and the sensitivity to opportunity sales.
Equipment Deficiencies Observations Scales are outdated and not functioning properly Sampler Systems are inoperable. Annunciator Alarm Panel is not functioning. Automation of Yard requires additional equipment No Boiler FEGT Monitors
Recommendations Replace/Install 5 Scales Replace As-Received and As-Fired Samplers with Modular Systems Digitize Alarm Panel in PI Install Necessary Equipment w/remote capability Install FEGT Monitors
FIGURE 10-4 Typical listing of “as-found” conditions in coal-fired plant with recommended corrective actions.
402 Combustion Engineering Issues for Solid Fuel Systems
Undertaking these initiatives will maximize the return on efforts to achieve key elements of any corporate strategy: Maximizing Generation—In a market where Reserve Margin Fore-
casts are dropping from 11% to 6% over the next 3 years, the need to meet full capability during peak demand will be essential. The peak demand growth will be accompanied by base-load growth, the greater portion being coal-fired generation; Fuel Flexibility—Burn the “right fuel at the right time.” During high-demand periods, this axiom will require the selective use of premium fuels and, during low-demand situations, the prudent use of the high-value fuels to take advantage of opportunity sales—the latter with full understanding of all the implications (e.g., slagging potential, capacity limitations, emissions impact, etc.); Environmental Compliance—Understand fuel impacts to the boiler and air quality control systems in complying with more stringent environmental limitations enacted by the governmental agencies; and Accurate Reporting of Inventory—Asset reporting accuracy required by Sarbanes-Oxley legislation makes high demands of data accuracy. Responsibilities of and consequences to management signatories with regard to accurate, timely asset reporting are significant. The flow chart of Figure 10-5 shows the inputs, calculations/tabulations and postings for the AccuTrack system. It is evident that a large
Inputs
Real-time Calculations and Processing
Yard and Plant RealTime Data
Tracking Sile Fills
Tracking Ship ment Unloads
Manual Inputs
On-the-Belt Blending
Shipment Status
Corporate Fuel Data Systems
Loading Destinations
Unload Quantities
Fuel Type & Feeder Quantities
Unload Destinations
Vendor/ Internal Lab Data On-Line Analyzer Data Day Ahead Dispatch
Fill Coal Specs Silo/Bunker Flow Model
Shipment Coal Specs
Tracking yard Inventory On-Hand Inventory Quantities & Specs Audit Inventories with Corporate
Shipment Variance Results DFTS Server
Results MS SQL Server
Website Interface
OSI PI Server
Process Book Client
Dedicated server with MS SQL and ECG’s server-side tracking software
FIGURE 10-5 AccuTrack software data inputs, processing and result postings.
Some Computer Applications for Combustion Engineering with Solid Fuels
403
number of entries are involved, and the responsible parties must make accurate, conscientious entries. Information transfer is frequently described as the lifeblood of large corporations. Large utilities house information in a variety of places; i.e., corporate fuel purchasing accounting systems are found to frequently contain actionable data for the plants via the OSI PI system. This system is custom designed to collect data from multiple inputs and make the data actionable at the plant level.
10.4 Successfully Applying Computer Technology to Fuels Control As with any project, the successful installation of such computer software must address all the following considerations: People—Change management, coordination, and mutual under-
standing are critical to acceptance at the plant, dispatching, and corporate fuels groups. It could be that these groups have conflicting goals. The fuels group could be rewarded for buying less expensive coal, while the plant emphasis is founded in greater output or lesser overtime expense. A cheaper fuel is not a good value unless it is part of a “right fuel at the right time” strategy with purchase and use to optimize the benefits. The understanding of the elements of this method and goals by all involved personnel is the best way to realize the desired result from properly managed coal use. Technology—Over the past 25 years or so, there have been many advances in generating plant technology. But very few of the advances have been applied to the coal-handling portions of the plant; the available upgrades were probably dismissed with the existing installations being deemed “adequate.” With the closer look and realization of the fuel opportunities available through such software, the justification of remote controls, accurate sensing elements, digital information networks, and OSI PI data records are given in a new light. Equipment—Subject to a design review, a plant coal-handling installation might require significant upgrades or new installations to provide the fuel flexibility required through such software application. Besides the technology aspects, the system upgrade might well include additional scales, variable speed belts with blending capabilities, coal samplers, and analyzers. System upgraded costs, though significant, may be readily justified with the projected savings, given the ever-increasing fuel prices, declining generation reserves, and growing sales opportunities, the best solution for an immediate payback through lowered fuel costs, fewer forced
404 Combustion Engineering Issues for Solid Fuel Systems
outages, and significantly improved availability. Ancillary services such as ramp rates now figure into the value of generation at a coalfired power plant. Plant personnel armed with the foreknowledge of coal quality impact can better meet their commitments. Figures 10-6 and 10-7 depict the Detroit Edison–Monroe Power Plant, the former an aerial view of the overall yard layout and the latter a schematic of the primary belts, unloading to storage, reclaim feeders, belt scales, and transfer points. Note that the pile is segregated into three separate coal classification storage areas. The yard area shows the two primary receiving stations consisting of a ship unloader and a unit train rotary dumper, with all received coal normally delivered to the appropriate storage area. The desired coal blend for the programmed boiler feed is facilitated via the selectively controlled underfed rotary plow feeders beneath the storage pile. The online analyzer at the belt feed to the bunkers verifies the intended coal quality or implies the coal blend tuning necessary to meet the operational objectives. Delivery to each of the boiler-mill storage silos provides sufficient coal for about 8 hours at full-load operation of the associated pulverizer. The data summary of Figure 10-8 includes both graphical and tabular displays providing a real-time status of the current Monroe Plant fueling status. An operations supervisor and a fuel yard manager typically use this
Ship Unloading system Two 72 inch belts that support~5000 TPH Unloads a 32,000 ton ship in ~8 hours Train Unloading System 72 inch belt that supports ~5000 TPH Unloads a 14,000 ton train in ~6 hours Reclaim 6 rotary plow feeders support ~2200 TPH Specifically designed for blending 3 types of coal An emergency reclaim is available at ~ 1400 TPH 28 silos (7 per unit) that hold 380 tons each (10,640 total) Consumption 1350 TPH at full load (~5.5 hr capacity) Ouerall 13 belt scales ~60 gates and over 160 field inputs used in tracking system X-Ray type coal analyzer
Rotary Rail Unloader
Ship Unloader
Overhead unload tripper
Low Sulfur Southern
Low Sulfur Western
Sampling building with analyzer off C4
Med Sulfur Eastern
FIGURE 10-6 DTE–Monroe Power Plant: Aerial view showing major elements of the fuel unloading, handling, and storage facilities. Note that the primary fuel reclaim is via underfed rotary plow feeders under each of the three segregated coal types.
Some Computer Applications for Combustion Engineering with Solid Fuels
LSS (#1)
LSW (#2)
MSE (#3)
405
CV-07
<== North 1425 ft CV-01 S Ship CV-02 S
1975 ft
FE-03
S
FE-05
S
FE-07
S
FE-04
S
FE-06
S
FE-08
S
CV-06 CV-08 CV-09
FE-25
13 belt scales
CV-03
6 Rotary plow feeders
Calibrated Reference Scale
Analyzer OLCA
S CVC4
R1A C-5
~60 gate indicators
S
S R1
ZM05
ZM05
C5-A C-6A
S
CV05 Train Unloading
S
R2 C-6
28 silos
To Yard
CVC2
C-7A
C-8A
C-7 1 2
3
4 5
Unit #2
6
7
C-8 8 9 10 11 12 13 14
15 16 17 18 19 20 21
22 23 24 25 26 27 28
Unit #1
Unit #3
Unit #4
FIGURE 10-7 DTE–Monroe Power Plant coal yard schematic.
System health
Inventory results On-thebelt coal specs
Analyzer results Silo desintations
Vendor supplied specs
Tripper locations Blending results
FIGURE 10-8 DTE–Monroe Power Plant: Real-time fuel tracking overview graphics.
406 Combustion Engineering Issues for Solid Fuel Systems
AccuTrack screen to review the various fuel qualities by type available and being loaded to the units in real time. An illustration of volume predictions of fuel over the near term is shown in Figure 10-9. As a part of the software methodology, the stochastic model will be a flow-blended fuel according to the bunker loading history and present inventory. The outlined fuel packet of 2–4 hours is correlated with a projected coal burn rate, and the characteristics are based on the history of bunker loading and the associated prorated analysis. The program will present to the operator an associated rate-dependent fuel projection summary that will be a guide to the forecast combustion properties. As a permanent record, the “as-fired” fuel analysis will provide a means for correlation of the fuel with associated data such as emissions, scrubber performance, and others—both measured and calculated—that are available from the PI historian. Figure 10-10 is a customized screen for a particular plant with its blended fuel options. The screen provides the boiler operator with an advisory display of both existing and projected coal parameters and an index guide as to the coal’s probable consequence to boiler operations. The matrix display of numerical values shows coal parameters for the present as well as those projected for next 1-hour, 2-hour, and 4-hour intervals to allow the operator to anticipate conditions and take appropriate action as might be required. Within the program, the calculation engine compares the numerical values of the present/projected fuel with preprogrammed indices or standards and, if triggered, will illuminate the screen with the appropriate warning. Particles Removed after 8 hrs Particles Removed between 6 to 8 hrs Particles Removed between 4 to 6 hrs Particles Removed between 2 to 4 hrs Particles Removed in 2 hrs
60
50
40
30
20
Material That will be removed 2-4 Hours from Now
10
0 0
10
20
30
40
50
60
FIGURE 10-9 Fuel quality/bunker flow near-term projection model.
70
80
Some Computer Applications for Combustion Engineering with Solid Fuels
407
Low Alarm: Opacity potential for the next 2 hours
High Alarm: Slagging potential in 4 hours
FIGURE 10-10 Fuel quality summary for boiler operations advisory.
Yellow means CAUTION—degrading conditions could occur—and RED indicates that a higher potential or likely prospect for degrading conditions. These warnings can apply to physical limitations of the installed equipment or to regulatory restraints such as emissions. Precautionary measures taken by the boiler operator might be to reduce the load or to activate sootblowers, initiate air heater cleaning, or to make pulverizer adjustments. In summary, the Figure 10-10 information displays provide the boiler operator with the following: Real-time and projected data from PI-ProcessBook; Simple table of forecasted data that allow the operator to advise to
adjust for changes in coal through Adjust excess air, Bias mills according to present fuel or desired firing configuration, Adjust load and/or Modify soot blowing schedule; and Color-coded variables which provide visual prompts with alarming codes: Green: “Okay,” Yellow: Low Alarm or CAUTION—POSSIBLE DERATE CONDITIONS, Red: High Alarm or PROBABLE DERATE CONDITIONS. The fuel information available through the coal analysis and fuel tracking/accounting provides valuable data for the near-term operator advice of
408 Combustion Engineering Issues for Solid Fuel Systems
combustion-related consequences as well as major components of real-time cost of generation. Those relevant properties are tracked from the scales, coal analyzer, and/or other resources, through the bunker and to the combustion process. The following list provides those items with significance to operator considerations and/or accounting and can be adjusted according to the customer wishes:
Btu (Btu/lb)—Basic measure of fuel quality Cost ($/MMBtu)—Basic measure of fuel value SO2 (lb/MMBtu)—Regulatory emission concern Ash Loading (lb/MMBtu)—Slagging potential, opacity/emissions concern Base-to-Acid ratio—Slagging potential, opacity/emissions concern Moisture (%)—Possible pulverizer limitation, heat rate effects Iron (%)—Slagging influence, ash properties Calcium (%)—Slagging influence, ash properties SiþAl (%)—Slagging influence, ash properties Alkalinity—Slagging influence, ash properties Volatility—Flame stability, FEGT effects
Additional properties can be derived or entered as might be applicable to a specific plant requirement.
10.5 AccuTrack Situation Challenges and Response The best practical design of an AccuTrack system cannot address all situations. Many elements and events that can occur will require the best judgment of operators and the technical staff to make the proper decision regarding the capabilities of the system and the immediate objectives. The following is a summary of challenges and considerations that might be addressed as the need arises: Delay from One Scale to Downstream Scale—At every plant location,
the length and capacity of the belts (i.e., fuel inventory on the belts) and locations of scales and analyzers become significant considerations requiring compensation. Although fuel blending changes can be made promptly, an inventory on the belt and in the bunker—often a fairly large quantity of coal but precisely traced through AccuTrack compensation—will require time to be burned. For instance, should there be an unanticipated change in operating conditions such as an abrupt boiler derate or sales opportunity, there will be a revised forecast of the fuel blend delivered to the burner. The fuel revision will be immediately displayed to the boiler operator for appropriate action. Update Frequency of Analyzer—The analyzer output frequency is a function of two elements: first, the interval for the sampler to obtain a representative fuel sample for analysis. The second interval
Some Computer Applications for Combustion Engineering with Solid Fuels
409
is a time for the analyzer to scan the captured sample and post the analysis result. This delay, though relatively small, is adjusted in the program. Calibration of Analyzer—The calibration of the analysis instrumentation is a periodic event and requires technical expertise and timely attention. If scheduling of calibration during fuel handling downtime cannot be arranged, coal quality entries must use vendorsupplied or prior analysis. Invalid Inputs—In every real-world installation, a fuel accounting support group must recognize and then reconcile input errors. These errors might arise due to human entry mistakes or sensor unavailability, but all inputs must be scanned for suspicious data and corrected accordingly. Mixing in Open Bunkers—Despite field verification of the bunker flow model, irregularities in the bunker can distort the predicted coal flow. The existing program may be upset by various bunker conditions such as a “slip” of a major coal hang-up and the nontypical coal such as high moisture, frozen coal, or excessive fines. (Note that several compensation factors are used to adjust the bunker flow model, including coal size, bunker geometry, moisture, and fuel type.) Projecting Day-Ahead Load—In most plants the bunker storage contains a significant amount of coal, often 8–24 hours’ supply at full load. In such an installation, the coal bunker inventory is based on a projected operational scenario with on-peak or off-peak generation arrangements. Situations can arise such that the forecast has been markedly changed, and there will be an obvious lag before the existing bunker inventory has been exhausted and a boiler coal change can be realized. Handling Unit Derates—Various conditions arise such that the operator will have to change unit output based on the plant equipment capabilities. They might include equipment derates, maintenance, operational difficulties such as slagging, emissions criteria, or external limitations. The coal blend to the bunker can be changed promptly and the AccuTrack bunker forecast revised immediately, but there might be a lag time before the optimized fuel is available at the burner.
In the computer system, installation of a coal analyzer and a thorough understanding of the coal-quality variables and the benefits/consequences allow the projection of various factors that affect optimal boiler operations. Meaningful presentation of this combustion information makes the data actionable, both in the present and the near term; is a valuable asset to the boiler operator; and can assure adequate warning of the degrading conditions and allow proactive measures by the operator to assure the
410 Combustion Engineering Issues for Solid Fuel Systems
optimum tuning of combustion variables for the rated unit output, emissions compliance, and/or the most effective combustion. To achieve the objective, the operator must track the coal movement from the pile to the bunker. The coal flows on long- and high-capacity belts from the pile to the burner. The flow on the belts and the numerous surge bins in between are mostly a stream flow. The surge bins act as a temporary buffer in between to regulate flow. A coal analyzer is sufficient to correctly predict the type of coal reaching the bunker from different coal piles given the belt movements.
10.6 Modeling the Flow of Coal in Bunkers and Silos The coal flow through bunkers and silos is extremely complicated. This is due to the fact that the bunkers have a large storage capacity and the discharge of bunkers is throttled and flow rate controlled by a feeder to a pulverizer. This causes significant and complex pattern mixing to occur in the bunker as the coal flows down from the top of the bunker to its exit at the bottom. Further complexities arise when a mill is shut off for a certain duration of time while the other mills are still running on a shared bunker. Understanding these flow patterns is important for knowing when the bunker will be discharging a particular type of coal and, in turn, in maximizing the fuel efficiency. Several different models are available for predicting coal flow in a bunker. The main types considered for implementation in AccuTrack software are as follows:
Plug flow; Discrete element modeling; Void model; and Stochastic model.
10.6.1 Plug Flow Models The plug flow model is the simplest case and uses the classic first in, first out (FIFO) assumption, a flow condition without mixing. This model has been the general standard for accounting convenience but is of little value to the boiler operator in providing predictable fuel performance. The plug flow model might be justified under two restrictive conditions: first, when the fuel loadings to the boiler are uniform (rarely a good assumption) and, second, when the bunker capacity is small or the height-to-planer area is high (as in a tall, slender silo).
10.6.2 Discrete Element Modeling (DEM) The discrete element modeling (DEM) method is the most sophisticated method of modeling granular flow in a bunker. A DEM model will generally
Some Computer Applications for Combustion Engineering with Solid Fuels
411
consist of a group of spheres which have the forces acting on them due to the following:
Friction; Collision; Cohesion; Damping; Gravity; and Smaller or molecular forces, which can also be included; e.g., van der Waals or Coulomb forces.
With a well-defined set of input parameters, the DEM simulations yield the most accurate results for granular flow in a bunker application. However, the DEM modeling is complex and extremely processor intensive. A single computer simulation can last for several days even on high-powered multiprocessor systems. This drawback makes DEM impractical for calculations applicable to real-time fuel blending optimization and operator advisories.
10.6.3 Void Model The void model is based on void creation because the material is discharged from the bottom of the bin or bunker. With time, and further discharged from the bunker, the further void creation and spread are calculated based on a probability model. This model is relatively simple, but it does not consider the interactive physical forces between the particles. Although fast enough for real-time simulation of granular flow, the model does not accurately capture the flow pattern occurring in the bunker, evidently forcing too much mixing at the particle level than what occurs in both scale models and field tests of granular flow. It is generally observed that granular materials like coal maintain the inter-particle packing over a large distance when flowing under the influence of gravity. The void model excessive mixing error tends to increase with time.
10.6.4 Stochastic Model The stochastic model is an event-driven, probability-based model that closely captures the flow pattern of the material in a bunker without being computation intensive. The current software implements a simplified version of the stochastic model and has been validated in both scale model and field validation tests. The AccuTrack model very closely predicts the flow patterns in a bunker and is stable over time. In summary, the stochastic model chosen has the following attributes: It is fast and can be run on a readily available processor. Compared
to the excellent credentials of the DEM model, the results closely
412 Combustion Engineering Issues for Solid Fuel Systems
match the DEM results but require only a small fraction of the run time, an essential requirement if the system objective is a real-time advisory to the operator and management; It does not produce the more extensive particle mixing as in the void model. This is a feature that has been validated as more representative of actual flow conditions; and The model is relatively straightforward yet produces results with a high degree of accuracy as verified with field tests. Examples of FIFO flows and stochastic flow are shown in Figures 10-11 through 10-13.
10.6.5 Bunker Geometry Coal bunkers generally are designed to store a reasonably large volume of fuel, usually about 8–24 hours’ fuel equivalent of full-load capability. This capacity is a configuration of utilitarian construction to fit the overall plant design, and that will promote the continuous flow of fuels without dead areas. This is accomplished by making the walls of the bunker with the greatest reasonable slope and selective use of lower friction materials such as stainless steel so that a granular material can slide easily in the lower 100
300
400
500
50 45 40 35 30 25 20 15 10 5 0
Percentage out of Total Flow
200
0
10 20 30 40 50 60 70 80
110 100 90 80 70 60 50 40 30 20 10
0
Layer 1 Layer 2
0
1
2 3 4 5 6 7 Hours of Bunker Flow
8
600 100
200
300
400
500
600
700
800
900
1000
FIGURE 10-11 At the left, the initial state of the bunker at time T ¼ 0 hrs. The graph on the right illustrates the proportional contribution of the two different fuels, one red and the other yellow, to the flow rate from the bunker outlet. The differentiated band of flowable, particulate solid fuel is assumed to have distinct properties worth tracking. The following illustrated sequence “A” is based on the FIFO (first-in, first-out) or plug flow assumption, and the “B” sequence is the AccuTrack stochastic model with both models identical at time T ¼ 0.
Some Computer Applications for Combustion Engineering with Solid Fuels 50
120 Percentage out of Total Flow
45 40 35 30 25 20 15 10 5 0
413
0
20
40
60
Layer 1 Layer 2
100 80 60 40 20 0
80
0
2 4 6 Hours of Bunker Flow
8
FIGURE 10-12A FIFO assumed flow with fuels distribution (left) and constituent contribution to bunker outlet flow at time T ¼ 2.5 hrs.
200
300
400
500
50 45 40 35 30 25 20 15 10 5 0 0
Percentage out of Total Flow
100
10
20
30
40
50
60
70
80
110 100 90 80 70 60 50 40 30 20 10 0 0
Layer 1 Layer 2
1
2 3 4 5 6 Hours of Bunker Flow
7
8
600
100
200
300
400
500
600
700
800
900
1000
FIGURE 10-12B AccuTrack stochastic flow with fuels distribution and constituent contribution to bunker outlet flow at time T ¼ 2.5 hrs.
areas of the hopper. Similarly, bridging is inhibited by making the hopper outlet area as large as possible. Development of a bunker flow model from the multidimensional scale model apparatus must be verified with real-world data. The illustration of Figure 10-14 shows the validation test setup and results using RFID transponder chips in a field test of the model. The RFID chips were placed in the bunker at a specific time, at predetermined locations, and atop the coal with known elevations and surface profile (inventory). These consumable RFID chips were of such size and character as to be identical with a
414 Combustion Engineering Issues for Solid Fuel Systems 50
120 Percentage out of Total Flow
45 40 35 30 25 20 15 10 5 0
0
20
40
60
80 60 40 20 0
80
Layer 1 Layer 2
100
0
2
4 6 Hours of Bunker Flow
8
FIGURE 10-13A FIFO assumed flow with fuels distribution and constituent contribution to bunker outlet flow at time T ¼ 7.5 hrs.
100 50 45
Percentage out of Total Flow
200
40 35
300
30 25 20
400
15 10 5
500
0 0
10
20
30
40
50
60
70
80
110 100 90 80 70 60 50 40 30 20 10 0 0
Layer 1 Layer 2
1
2 3 4 5 6 Hours of Bunker Flow
7
8
600 100
200
300
400
500
600
700
800
900
1000
FIGURE 10-13B AccuTrack stochastic flow with fuels distribution within the bunker and constituent contribution to bunker outlet flow at time T ¼ 7.5 hrs. Note that the probabilistic flow technique will give similar but slightly different results each time the program is run.
comparable coal particle at the same location and with every expectation that it would follow a similar flow path. The individually identified RFID chips were recorded with a time stamp as they passed through the belt of the respective feeders. Over 95% of the chips projected to arrive within 8 hours were recorded by the scanners, a recording rate that was thought to be very good. Those chips whose projection of arrival over 8 hours might have been in “dead zones” in close proximity to the bunker walls.
Some Computer Applications for Combustion Engineering with Solid Fuels
415
SCALE CONVE
RFID Chips
YOR T-
1B
Tracer Testing Method ELEPHANT SNOUT
600 500 400 Predicted Time Actual Time
300 200
BUN KE R
100 0 1
2
Tag Tracking ID Description 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
3
4
5
10311 10312 10313 10314 10315 10316 10317 10318 10319 10320 10321 10322 10323 10324 10325
6
7
8
Initial Location X Y 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75
9
10
11
12
Z 22 22 22 22 22 22 22 22 22 22 22 22 22 22 22
50 53 55 57 60 65 65 65 65 65 60 56 54 52 50
13
14
15
Predicted Time Actual Time 360 340 360 345 400 380 450 475 520 530 520 540 520 500 520 485 520 525 520 500 450 500 400 425 360 400 360 340 360 345
B
GAMMETRIC FEEDERS
RFID Reader
PULVERIZING MILLS
FIGURE 10-14 Illustration showing field test validation method and results of bunker fuel flow model.
The test results of Figure 10-14, comparing the expected arrival with the actual time stamp, are shown in both tabular and graphical format. The calculated chip arrivals as calculated by the program take into account the subsequent fuel loading and the flow rate to the feeders. The very close agreement of the site-specific computer output with the actual time stamp can be noted, and the small differences might be attributed to the randomness of the particle flow path within the bunker.
10.6.6 Validation of Bunker Modeling The following methods are used to validate the bunker model: 1. A scale model proportional to an actual bunker design is used with a material of closely matching coal characteristics for the mock simulation. A representative model of bunkers at the eastern power plant is shown in Figure 10-15. Several materials closely matching the coal characteristics were used in the scaled test model. For the material tracking of the element packets, two different color materials were used in the bunker, as shown in Figure 10-16. The dynamics were traced using time-indexed photo frames. 2. To validate the actual flow characteristics of coal through the bunker/hoppers under a variety of bunker configurations and equipment operations, radio frequency identification devices (RFIDs)
416 Combustion Engineering Issues for Solid Fuel Systems
FIGURE 10-15 Bunker size (1/54th) physical model.
FIGURE 10-16 The bunker filled with two different types of material. The dynamics were traced using time-indexed photo frames.
Some Computer Applications for Combustion Engineering with Solid Fuels
417
testing was done in a power plant bunker. Examples of RFID tags are shown in Figure 10-17. RFID antennas were mounted in the stock feeders and cabled back to a single reader/computer for acquisition. RFID tags (100) were then placed at predetermined positions within the bunker. The tags were then allowed to flow with the coal through the bunker to the feeder. As the tags flow past the reader, the times and unique ID numbers are recorded. The tags are consumed in the pulverizers. Once the data were analyzed, the bunker model was tuned to the specific bunker. RFID tags were used for the purpose of identification or tracking using radio waves. In bunker modeling, the RFID tags can be used to determine how long it takes coal to travel through a bunker. This is done by recording the time difference from when the tags are inserted to when they exit the feeder where the RFID antennas will sense their passing. By better understanding flow in the different areas of the bunker, the model can be appropriately adapted to simulate real-time flow. The RFID tags themselves are 11/8" 11/8" squares that are 1/8" thick and in no way affect the performance of the coal or require any permitting, as in the case of coal doping. Field tests for coal flow model verification were carried out at the Conemaugh Station using RFID tags to monitor flow patterns and flow rates within the bunkers. Results from this testing are shown in Figures 10-18 and 10-19. The results of the test and its comparison to the model timings are shown in Figure 10-20. The information available for bunkered coal is as follows: 1. The bunker particles are projected on a grid, with each grid point having information of which pile and feeder it came from; 2. The bunkered coal also has about 10 properties associated with it that are tracked and/or adjusted to provide guidance to operations with regard to combustion characteristics, boiler capability, emission concerns, and real-time generation costs;
FIGURE 10-17 RFID tags that are placed in the bunker for tracing solid fuels flow.
418 Combustion Engineering Issues for Solid Fuel Systems
FIGURE 10-18 (A) Feeder belt discharge (empty) shown with door open; (B) distributed coal flow from belt running to provide optimum probability for tag detection by the scanner; (C) feeder door, location to attach transponder antenna. Observation ports on door to run antenna wire to reader.
3. The flow with known coal quality bunkering data will provide realtime as fired fuel data with 5 minute updates; and 4. The AccuTrack model has a fuel forecast mode in which it predicts the coal characteristics to the boiler based on bunker loadings. This advisory to operations allows for proactive actions for timing of its boilers for specific information and to address potential problems like slagging, emission limitations, capacity limitations, etc. The bunker model is supplied with inputs including the following: a. Bunker-by-bunker geometry; b. Fuel properties of the coal filling the top of the bunker; c. Bunker fuel top profile is based on ultrasonic level detector data above each discharge to the feeders. This upper profile data
Some Computer Applications for Combustion Engineering with Solid Fuels
419
FIGURE 10-19 Placement of RFID tags in the bunker.
Average Time Comparison 7.0000
Time in Hours
6.0000 5.0000 4.0000
Model Actual
3.0000 2.0000 1.0000 0.0000 9.25
37
46.25
58.05
59.15
61.05
64.75
69.25
69.85
72
72.35
73.5
X coordinate
FIGURE 10-20 Comparison of the model results to RFID test results.
reference is a means to refine the upper surface computational complexities during the simultaneous bunker outlet fuel flow and fuel replenishment from above; and d. Fuel fill rate into the bunker and the respective drawdown from each feeder discharge sets an instantaneous bunker coal quality profile. Projections of coal quality over time will change with revisions of either or both of the fill and drawdown schedules.
420 Combustion Engineering Issues for Solid Fuel Systems
Additional inputs can be provided depending on the relation of the coal flow with regards to that property. The multidimensional model assumes that the bunker is approximately symmetric on the third dimension; that is, both the foreground and background geometries are a reflection about the centerplane. A revised model is currently under development for implementation in the AccuTrack fuel flow model for tackling problems associated with nonsymmetric bunkers.
10.7 Conclusions Regarding the AccuTrack Approach to Computer Management of Fuel Properties Plant-specific editions of the AccuTrack process have been installed and have been implemented as a standard operating practice. Besides the fundamental accounting contribution, the process has been particularly helpful to overall plant operations with key elements being increasingly evident with longer term usage. As an example, the following are instances of particular emphasis at the DTE/Monroe Plant: Boiler operations during higher demand were particularly sensitive
and the information most valuable to the operator in the attention afforded to control stack opacity and slagging/fouling in the boiler; Backup data using vendor analyses and as-received sample analyses for coal from the stockpiles were most valuable for entering default data during analyzer and/or sampler outages; Coal blends from the segregated stockpiles were based largely on vendor-supplied analysis and the proportionality projected for an hour-by-hour coal-quality target. Results from the as-burned analyzers just prior to the bunker showed very good agreement with the targeted blend coal quality, an indication that the storage segregation of coal type was effective and that the blending proportionality was accurate; With the installation of a well-designed system, reliance on the automated data output is emphasized. Operator intervention for manual entries does occur, and the manual input requires controlled permissives for the necessary data entry; and Despite the high reliability of a well-designed system, maintenance and calibration are required to assure accuracy and availability of the system and the equipment. Periodic calibration, system checks, and maintenance requirements must be in place and apply to both the PI and to other IT systems.
Some Computer Applications for Combustion Engineering with Solid Fuels
421
10.8 Summary With the dominance of fuel costs in the generation accounting equation, the necessity to control, wisely burn, and accurately track fuel inventories is self-evident. Systems such as AccuTrack are developed for that purpose and can effectively implement the necessary elements to accomplish the fuel-associated cost control. Corporate and plant personnel participate in the following tasks to fulfill the mission: Inventory Tracking is essential for accounting and generation cost
control. This is accomplished by accurate, reliable measurement of the receipt, storage, transport, and consumption/disposition of available fuels through state-of-the-art equipment to provide actionable data to all involved parties; e.g., corporate fuel, yard, operations, dispatch, etc.; Fuel Flexibility is created through management initiatives specifying real-time integration of Accounting Fuel Management Systems with Plant Performance objectives to enable strategic use of various fuel supplies to maximize generation (sales) opportunities while operating within the capabilities of the boiler and in compliance with emission regulations; and Fulfillment of Operational Excellence is accomplished by optimized planning and operations with precision by burning the “right fuel at the right time” to support the plant mission while complying with environmental requirements.
CHAPTER
11
Gasification Christopher Higman Chief Consultant Syngas Consultants Ltd.
11.1 Introduction to Gasification In a drive for more efficient use of coal—reducing net station heat rates—and in a drive for reduced airborne emissions, gasification has emerged as an alternative to combustion. The idea of coupling a coal-based synthesis gas generator to a combustion turbine is not new; it was proposed by Wilhelm Gumz in 1950 [1], but only with the development of large, highly efficient turbines has it emerged as an economic possibility. The inherent possibility of pre-combustion CO2 removal at modest additional cost—as practiced on a large scale in the ammonia industry—has increased the interest in gasification for the generation of electric power [2]. Gasification can be described as the conversion of any carbonaceous feedstock into a gaseous product with a useful chemical heating value. Early processes for the production of gas from coal emphasized devolatilization and pyrolysis reactions creating a gas with significant hydrocarbon content for lighting purposes. These were the town gas processes of the 19th century. With the introduction of the incandescent mantle and the development of chemical syntheses at the beginning of the 20th century, the emphasis shifted to water gas and partial oxidation processes generating synthesis gas (or syngas), in which carbon monoxide and hydrogen are the principal components. Syngas manufactured from coal, petroleum coke, and various other feedstocks such as refinery residues is processed into ammonia (for fertilizer production) and methanol. In both cases, gasification-based applications 423
424 Combustion Engineering Issues for Solid Fuel Systems
amount to about 10% of global production. Other chemical applications include, for example, the production of pure hydrogen (typically in oil refineries) or carbon monoxide for acetic acid manufacture and synthesis gas for oxo-alcohol synthesis. In all these cases, the raw synthesis gas from the gasifier must be purified. Typically, this requires desulfurization (often down to the parts per billion level) and in many cases adjustment of the hydrogen-to-carbon monoxide ratio. Synthetic fuels have been produced from coal gasification using Fischer-Tropsch technology on a commercial scale in South Africa since the 1950s. Integrated gasification-combined cycle (IGCC), in which syngas generated from coal or petroleum residues is used to fuel a combustion turbine in a combined cycle power plant, is the subject of increasing interest on account of both its high efficiency and the extremely low rates of atmospheric emissions.
11.2 Gasification Theory The principal reactions, which take place during the gasification of pure carbon, are those involving carbon, carbon monoxide, carbon dioxide, hydrogen, water (or steam), and methane. The most important are Partial oxidation :
C þ 1=2 O2 ! CO
and the water gas reaction :
C þ H2 O ⇆ CO þ H2
111 MJ=kmol [11-1] þ131 MJ=kmol [11-2]
In gasification processes, the reactions with free oxygen are all essentially complete. The carbon conversion is in general also largely complete. Most processes operate at sufficiently high temperatures that equilibrium is reached and the final gas composition is determined by the CO shift reaction :
CO þ H2 O ⇆ CO2 þ H2
and the steam methane reforming reaction :
CH4 þ H2 O ⇆ CO þ 3 H2
41 MJ=kmol [11-3] þ 206 MJ=kmol [11-4]
Gasification temperatures are generally sufficiently high that, thermodynamically as well as in practice, no hydrocarbons other than methane can be present in any appreciable quantity. Depending on the reactor arrangement, it is, however, possible that some pyrolysis products survive and are contained in the synthesis gas. It is instructive to see how the gas composition changes with pressure and temperature. This is shown in Figures 11-1 and 11-2 with the calculations all performed at 1,832 F (1,000 C). The increase in methane and CO2 content in the synthesis gas with increasing pressure can be seen clearly in Figure 11-1. When looking at yields, as in Figure 11-2, one needs to distinguish between a syngas (H2 þ CO) basis, which is relevant for
Gasification
Content [mol%]
70%
425
CO2
60%
CO
50%
H2 CH4
40%
H2O
30% 20% 10% 0% 0
20
40
60 80 Pressure [bar]
100
120
FIGURE 11-1 Variation of syngas compositions with pressure (at 1,832 F).
10
1.0 (H2+CO)/Feed
0.9
8
0.8
7
0.7
6
0.6
5
0.5
O2/ Heat content dry
Heat content dry/Feed
4
0.4
3
0.3
2
0.2
1
0.1
O2/(H2+CO)
0 0
20
40
60 Pressure [bar]
80
[mol/mol] or [mol/MJ]
[mol/mol] or [MJ/mol]
9
100
0.0 120
FIGURE 11-2 Variations of yields with pressure (at 1,832 F).
chemicals applications, and heating value basis, which is applicable for power (IGCC) applications. The syngas yield of synthesis gas drops with increasing pressure, whereas the heating value yield increases, reflecting the higher methane content. Similarly, the variation in oxygen demand goes in opposite directions depending on whether it is expressed per unit of syngas or per unit of heating value. As most modern gasification processes operate at pressures of 400 psig or higher, temperatures of above 2,400 F are required to produce a synthesis gas with a low methane content. At these temperatures, the same trends exist as at 1,832 F (1,000 C) although they are less pronounced. Note that while an increased methane content is beneficial for power generation in terms of heat supply to a combustion turbine, large quantities of methane
426 Combustion Engineering Issues for Solid Fuel Systems
generated by processes optimized for synthetic natural gas (SNG) production, for example, would be counterproductive in a CO2 capture scenario. Nonetheless, even low-temperature fluid-bed processes operating around 1,800 F generate sufficiently little methane that 90% carbon emissions reduction can be achieved with conventional technology. A comparison of gas compositions and yields with temperature is shown in Figures 11-3 and 11-4. Here, one can see that, with increasing temperature, the gas becomes increasingly CO rich. In Figure 11-4, the increased oxygen demand at higher temperature operation is apparent. The syngas yield goes through a mild maximum between 2,200 F (1,204 C) and 2,400 F (1,315 C).
Content [mol %]
70%
CO2
60%
CO
50%
H2
40%
CH4 H2O
30% 20% 10% 0% 1000
1100
1200 1300 1400 Temperature [°C]
1500
1600
10
1.0
9
0.9
8
0.8 O2/ Heat content dry
7 6 5 4
(H2+CO)/Feed Heat content dry/Feed
0.7 0.6
O2/(H2+CO)
0.5 0.4
3
0.3
2
0.2
1
0.1
0 1000
1100
1200 1300 Temperature [°C]
1400
FIGURE 11-4 Variations of yields with temperature (at 450 psig).
0.0 1500
[mol/mol] or [mol]/MJ
[mol/mol] or [MJ/mol]
FIGURE 11-3 Variation of syngas compositions with temperature (at 450 psig).
Gasification
427
11.3 Features of Gasification Systems In the practical realization of gasification processes, a broad range of reactor types has been and continues to be used. Although the usual classification is by bed type, there are other features to be considered, and many of these can be selected independently of the bed type. The most important of these will be discussed in the following sections.
11.3.1 Bed Type For most purposes, reactor types can be grouped into one of three categories: moving-bed gasifiers, fluid-bed gasifiers, and entrained-flow gasifiers. The gasifiers in each of these three categories share certain characteristics, which differentiate them from gasifiers in other categories. Some of these characteristics are summarized in Table 11-1. Moving-bed gasifiers (sometimes called fixed-bed gasifiers, since although the feedstock is moving through the bed, the location of the bed itself is fixed in space) are characterized by a bed, in which the coal moves slowly downward under gravity as it is gasified, generally, but not universally, by a counter-current blast. In such a counter-current arrangement, the hot synthesis gas from the gasification zone is used to preheat and pyrolyze the downward flowing coal. With this arrangement, the oxygen consumption is very low, but pyrolysis products as well as moisture brought into the reactor with the coal are present in the product synthesis gas. The outlet temperature of the synthesis gas is generally low, even if high slagging temperatures are reached in the heart of the bed. Moving bed processes operate on lump coal. An excessive amount of fine particles, particularly if the coal has strong caking properties, can block the passage of the upflowing syngas. Fluid-bed gasifiers offer extremely good mixing between feed and oxidant, which promotes both heat and mass transfer. This ensures an even distribution of material in the bed, and hence a certain amount of only partially reacted fuel is inevitably removed with the ash. This places a limitation on the carbon conversion of fluid-bed processes. The operation of fluid-bed gasifiers is generally restricted to temperatures below the softening point of the ash, since agglomeration of soft ash particles will disturb the fluidization of the bed. Some attempts have been made to operate into the ash softening zone to promote a limited and controlled agglomeration of ash with the aim of increasing carbon conversion, but this mode of operation has so far not been successfully translated into commercial-scale plants. Sizing of the particles in the feed is critical; material that is too fine will tend to become entrained in the syngas and leave the bed overhead. This is usually partially captured in a cyclone and returned to the bed. The lower temperature operation of fluid-bed processes means that they are better placed to handle reactive feedstocks such as low-rank coals and biomass.
428 TABLE 11-1 Characteristics of Different Categories of Gasification Processes Category
Moving Bed
Fluid Bed
Entrained Flow
Ash conditions Typical processes
Dry ash Lurgi
Slagging BGL
Dry ash Winkler, HTW, CFB, KRB Transport Gasifier
Agglomerating KRW, U-Gas
Slagging Shell, GEE, E-Gas, Siemens, KT
Feed characteristics Size Acceptability of fines
1/4"–2" limited
1/4"–1/2" good
1/4"–1/2" better
<200 mm unlimited
yes (with stirrer)
1/4"–2" Better than dry ash yes
possibly
yes
yes
any
high
low
any
any (dry feed) high (slurry feed)
low (800–1,200 F)
moderate (1,650–1,900 F)
high (2,300–2,900 F)
low low hydrocarbons in gas
moderate moderate lower carbon conversion
moderate (1,650– 1,900 F) moderate moderate lower carbon conversion
Acceptability of caking coal Preferred coal rank
Operating characteristics Outlet gas temperature low (800–1,200 F) Oxidant demand Steam demand Other characteristics
low high hydrocarbons in gas
high low pure gas, high carbon conversion
Gasification
429
The history and development of coal gasification and fluid-bed technology have been intimately linked since development of the Winkler process in the early 1920s. Winkler’s process operated in a fluidization regime, where a clear distinction exists between the dense phase or bed and the freeboard where the solid particles disengage from the gas. This regime is the classic or stationary-fluidized bed (bubbling-fluidized bed). With increasing gas velocity, a point is reached where all the solid particles are carried with the gas and full pneumatic transport is achieved. At intermediate gas velocities, the differential velocity between gas and solids reaches a maximum, and this regime of high, so-called “slip velocity” is known as the circulating fluidized bed. Over the years, gasification processes have been developed using all three regimes, each process exploiting the particular characteristics of a regime to the application targeted by the process development. Entrained-flow gasifiers operate with feed and blast in co-current flow. The residence time in these processes is short (a few seconds). The feed is ground to a size of 200 mm or less to promote mass transfer and allow transport in the gas. Given the short residence time, high temperatures are required to ensure a good conversion, and therefore all entrained-flow gasifiers operate in the slagging range. The high-temperature operation creates a high oxygen demand for this type of process. An advantage of entrainedflow gasifiers is that they do not have any specific technical limitations on the type of coal used, although coals with a high moisture or ash content will drive the oxygen consumption to levels where alternative processes may have an economic advantage. Additionally, the ash is produced in the form of an inert slag or frit. This is achieved with the penalty of additional effort in coal preparation as well as a high oxygen consumption, especially in the case of coal-water slurries and/or coals with a high moisture or ash content. Nonetheless, even if entrained flow has been selected as the means of contacting fuel and gasification agent, this still leaves a considerable variety of alternatives open in the design approach, as can be judged from the characteristics of some important entrained-flow processes shown in Table 11-2. The majority of successful coal gasification processes that have been developed since 1950 are entrained-flow, slagging gasifiers operating at pressures of 275–1,000 psig and at high temperatures of at least 2,400 F. Entrained-flow gasifiers have become the preferred gasifiers for hard coals and have been selected for the majority of commercial-sized IGCC applications. Other Configurations and Processes Other configurations have been developed, particularly for biomass and waste applications. A number of processes include a first pyrolysis stage at moderate temperatures so as to separate out the high volatile content from the char. In a second, high-temperature stage, the volatiles are gasified at high
430 Combustion Engineering Issues for Solid Fuel Systems TABLE 11-2 Features of Various Entrained-Flow Processes Process
Stages
Feed
Flow
Reactor Wall
Koppers-Totzek Shell SCGP
1 1
dry dry
up up
jacket membrane
Prenflo (now Shell) Siemens (formerly GSP) GEE (formerly Texaco) E-Gas
1
dry
up
membrane
1
dry
down
1
slurry
down
Membrane (or refractory) refractory
2
slurry
up
refractory
CCP (Japan)
2
dry
up
membrane
Primary Syngas Cooling
Oxidant
syngas cooler gas quench and syngas cooler gas quench and syngas cooler water quench and/ or syngas cooler
oxygen oxygen
water quench or syngas cooler two stage gasification two stage gasification
oxygen
oxygen oxygen
oxygen air
temperature to overcome the problem of high tar content in the syngas normally associated with biomass gasification. Detailed integration and execution vary considerably among the different processes adopting this strategy. Further, one or two processes do not fit into any of these three main categories. This includes in situ gasification of coal in the underground seam as well as molten bath processes.
11.3.2 Flow Direction Flow direction must be examined from two points of view. The flow direction of the reactants in relation to one another must be considered. In an entrained-bed process, the reactants flow, by definition, is co-current, but in a moving-bed reactor one has a choice. The most well-known movingbed reactor, the Lurgi-Sasol dry bottom gasifier, operates in countercurrent, but a number of smaller gasifier designs use co-current flow to reduce the tar make. Co-current moving bed gasifiers are, however, limited to small-scale operations. For entrained-flow gasifiers, the choice is different, namely whether to go for an up-flow design, such as with the Shell SCGP and the ConocoPhillips E-Gas processes, or a down-flow design, such as GE Energy (GEE) or Siemens.
11.3.3 Feed Preparation The feed system is an important choice for entrained-flow gasifiers. GEE and E-Gas use a coal-water slurry, whereas Shell and Siemens use dry-feed systems.
Gasification
431
In a dry-feed system, the coal is ground and dried in a roller mill with a hot-gas drying circuit, similar to those used in conventional pulverized coal units. The pulverized coal is then fed through a lock hopper system into the pressurized feed vessel. The coal is then transported to the burners from the feed vessel by pneumatic conveying in the dense phase. The carrier gas is typically pure nitrogen from the air separation unit (ASU), but for some chemical applications where nitrogen is undesirable, CO2 can be used. The lock hopper system, which relies on gravity flow to move the coal from the uppermost, atmospheric bunker through the lock hopper into the feed vessel, requires a support structure every bit as tall as the gasifier itself, so there is great interest in the development of a “solids pump,” which could reduce the cost of the feed system. For wet-feed systems, the slurry is made in a rod mill into which precrushed coal and water are fed. The coal is ground in a wet milling process to a size of about 100 microns. The product slurry is sieved to remove oversized material, which can be discarded or recycled according to need. The slurry is pumped to the reactor pressure typically with a membrane piston pump. In chemical applications, slurry systems benefit by being able to operate at high pressures (up to 1,200 psig); however, bringing liquid water into the gasifier causes considerable reduction of efficiency compared with a dry-feed gasifier. In the case of dry-feed gasifiers, the amount of carrier gas required for the pneumatic transport of the coal into the gasifier increases with pressure. The economic limit for dry feed systems is generally considered to be about 550 psig. The optimum feed system is intimately connected to the type of coal to be processed. While for bituminous coals there is not much economic difference between the two systems, for subbituminous coals or lignites with a high inherent moisture content, these differences can become considerable. The inherent moisture does not contribute to the transport properties of a slurry, so it must be added to the about 35% free water content of the slurry. The total water content entering the gasifier can become so large that economic operation with a slurry feed is impossible. The situation is different with a dry feed. The coal must be dried to a level that leaves it dry enough to allow unhindered pneumatic transport, which in the case of lignites might still be as high as 10–15%. The efficiency penalty in the gasifier itself is therefore relatively small. Modern recuperative drying processes, which recover the latent heat of the steam driven off from the raw coal, help to reduce the energy cost of the drying process.
11.3.4 Operating Temperature Operation temperature is another fundamental choice, whether to go for a slagging or nonslagging operation. While a decision may be connected with the bed type—all entrained-flow processes operate in the slagging
432 Combustion Engineering Issues for Solid Fuel Systems
zone—the moving bed offers a choice between, e.g., the Lurgi-Sasol dry bottom gasifier and the BGL slagging gasifier. In all cases, it is necessary to ensure that the temperature is either high enough that there is an adequate safety margin above the ash fluid temperature (and its TCV critical viscosity temperature) that the slag flows easily or that the operating temperature is sufficiently lower than the ash softening temperature that ash particle agglomeration does not interfere with the operation of the bed (whether fluid or moving). In between these two temperatures is an effective “no-go zone” in which sticky ash will create operating problems in any system.
11.3.5 Oxidant The choice of oxygen or air is another issue which gives rise to discussion. Historically, the high-temperature entrained-flow processes used oxygen, and there are good reasons for this. Partly, this has been dictated by the fact that in the period between 1935 and 1985 most gasifiers were built for chemical applications where the presence of large quantities of nitrogen in the syngas was detrimental to the downstream synthesis process. (Note that this also applies to ammonia, where only about 25–30% of the nitrogen associated with the oxygen used in the gasifier is required for the synthesis.) But there are other technical reasons why oxygen is preferred over air in this configuration, even where the presence of nitrogen is not a fundamental problem. As can be seen from Figure 11-5, while the cold gas efficiency does not vary much over the range 85–90% oxygen, it falls off ever more rapidly the closer one approaches the 21% of the 5
100 Cold Gas Efficiency
Cold gas Efficiency [%]
90
4
80 70
3
60 50 kmoles O2 per 100 kg maf coal
40
2
30 kmoles O2 per kmole syngas
20
1
10 0 0
20
40 60 Oxygen content [mol%]
80
0 100
FIGURE 11-5 Variations of cold gas efficiency with air enrichment (at 2,732 F).
Gasification
433
atmosphere. Essentially, this represents the penalty of having to heat the nitrogen to the gasification temperature, which was chosen as 2,732 F (1,500 C) in the example. The oxygen demand is increased, and the syngas quantities to be cooled and treated are approximately double. These disadvantages are more than enough to offset the capital and operating costs of an air-separation plant. The situation is different, however, with a fluid-bed reactor operating at, say, 1,800 F. The efficiency penalty is slightly less because the nitrogen must not be heated to as high a temperature as in the case of an entrainedflow gasifier. More important is the fact that the fluid-bed gasifier operating with oxygen must inject substantial quantities of steam to achieve a high carbon conversion, while maintaining the bed at a temperature lower than the ash softening point. This steam is lost to the condensation part of the combined cycle, and the net effect is to favor air over oxygen in this case [3].
11.3.6 Reactor Containment The reactor containment is another choice that requires consideration. Some means of protecting the pressure shell from the reaction temperature is required. The alternatives are (a) refractory lining; (b) a water-cooled membrane wall between the reaction space and the pressure shell; or (c) a water jacket integral with the pressure shell. Although most processes have settled for one system or another, Siemens offers a choice between all three according to application. Refractory lining is the cheapest first-cost solution, but for slagging gasifiers it requires regular maintenance. The hot face may require replacement every 2 years depending on gasifier operating temperatures and coal (ash) quality, with possibly some minor maintenance in between. A full, hot face change-out can take up to 4 weeks, so plants with a highavailability requirement tend to install a spare gasifier, which negates the cost advantage. The membrane wall solution relies on a layer of solidified slag between the water-cooled wall and the hot gas space to protect it. The liquid slag then flows down the wall of solid slag, as shown in Figure 11-6. This solution requires considerably more in terms of investment than refractory, but once installed, it is relatively maintenance free. Membrane walls have proven successful, trouble-free operation of 10 or more years, and they have a predicted life of about 20 years.
11.3.7 Primary Syngas Cooling The final choice to consider is that of primary syngas cooling. The purpose of the primary cooling in a slagging gasifier is to bridge the “no-go zone” between free-flowing liquid slag and dry solid ash. The method of
434 Combustion Engineering Issues for Solid Fuel Systems
FIGURE 11-6 Gasifier containment solutions.
cooling must be chosen so that no sticky ash adheres to heat exchangers or other surfaces while it is in the intermediate temperature range. Here, one can select between water quench, gas quench, and radiant cooling together with or without a convective cooler. GEE, for instance, offers a choice between a full-water quench and (for coal gasification) a radiant cooler. Shell offers a gas quench. Staged gasification is a feature that can increase efficiency (see Figure 11-7). It is used by E-Gas and two processes under development in Japan: CCP and Eagle. Part of the feed is injected to a first stage, where all the oxygen is consumed. This stage operates hot, under slagging conditions, and the slag is drawn off at the bottom. The remainder of the feed is added at the second stage, where it reacts with the hot synthesis gas from the first stage. In the second stage, the additional coal is dried and devolatilized, and part of the fixed carbon is gasified. The gas leaving the second stage is cooled to below the ash fusion temperature by the endothermic reactions. The ash is therefore dry at this point, and the gas can be cooled in a convective heat exchanger. The remaining ungasified char and the ash are recovered from the gas in a cyclone or filter and recycled to the first stage. This ensures a high, overall carbon conversion despite the low reactor exit temperature. It also has the effect that the ash from the second stage feed is also slagged and discharged from the first stage.
Gasification
435
FIGURE 11-7 Two-stage gasification.
11.3.8 Primary Gas Cleaning Primary gas cleaning, i.e., particulate removal as well as removal of chlorides and bulk ammonia from the gas, is usually considered as part of the complete gasification process and as such is supplied by all the gasification technology vendors. Particulate removal may be performed wet, in a scrubber (e.g., GEE or Siemens) or dry, with a candle filter (e.g., E-Gas or Shell). The wet systems produce a “black water” from the scrubber bottoms that must be cooled and cleaned. The bulk of the chlorides and ammonia is removed in the scrubber, although the process condensate from the low-temperature gas cooling will still contain significant quantities of ammonia. While the water handling in the dry systems is much simpler, the candle filter itself is a source of maintenance cost. Different materials are used for the candles. Shell uses ceramic candles; E-Gas uses sintered metal. In both cases, a subsequent water wash is required to remove the ammonium chloride. The wash water is, however, largely free of particulate matter.
11.3.9 Fuel Issues Properties of fuels are described in Chapters 2 and 3 and, for most issues related to fuel properties, the reader is referred there. There are, however, a few issues specific to gasification which will be discussed here. Ash fusion properties are generally measured for both oxidizing and reducing atmospheres. For gasification, the values measured under reducing conditions are the relevant ones, and it is necessary always to ensure that these are in fact the values quoted when listing out fuel properties. This also applies when measuring slag viscosities.
436 Combustion Engineering Issues for Solid Fuel Systems
Gasifiers are generally very flexible in their fuel requirements. The Wabash demonstration unit was designed for coal feed but has been operated for several years on 100% petroleum coke, simply because this was a cheaper fuel. The only limit on sulfur content is the size of the sulfur recovery unit. Nuon has successfully added up to 20% biomass to the feed of its IGCC in Buggenum, although this unit was also designed only for coal. Limitations to increasing the biomass rate are in the feed system (the lower energy density of the biomass lowers the fuel input) and the low heating value of the syngas (caused by the high water content of the biomass). One should note, however, that building a plant that can process 100% low-sulfur, high-moisture, high-ash lignite and also 100% petroleum coke has its costs because each section must be designed for the extreme case of each fuel.
11.4 Commercial Gasification Systems An extremely wide variety of gasifiers has been implemented, particularly in the field of small-scale biomass and waste gasification. The number of processes operating or under construction at sizes over 100 MWth is limited, however, and only these are discussed here.
11.4.1 GE Energy (formerly Texaco) The GEE coal gasifier is a down-flow, entrained-flow, slagging design. It uses a coal-water slurry feed, which makes for a simpler, cheaper design than a dry-feed operation, but at the cost of a slight efficiency penalty. The reactor containment uses a refractory lining. Depending on application, cooling may be by water quench or radiant cooling (see Figures 11-8 and 11-9). The first Texaco coal gasifier was demonstrated in Oberhausen, Germany, in 1978. In 1984, commercial units for Eastman in Kingsport, Tennessee, and for Ube in Japan were placed into service. In the same year, the Cool Water 100 MWe demonstration IGCC was started up. In 1996, a 250 MWe power plant for Tampa Electric went into operation in Polk County, Florida. Tampa Electric has announced its intention to build additional 630 MWe capacity based on the same technology. GE Energy acquired the former Texaco gasification business in 2004. There are currently a total of 24 plants contracted, more than half of which are in operation.
11.4.2 Shell The coal gasifier (SCGP) is an up-flow, entrained-flow gasifier operating on a dry feed and at slagging temperatures. The up-flow arrangement allows separation of syngas and slag largely within the reactor itself. Typically, there are four side-mounted burners located in the lower part of the reactor.
Gasification
FIGURE 11-8 GEE gasifier with radiant cooling.
FIGURE 11-9 GEE gasifier with quench cooling.
437
438 Combustion Engineering Issues for Solid Fuel Systems
FIGURE 11-10 Shell coal gasifier (With permission from Shell).
The vessel containment uses a membrane wall, which has a demonstrated lifetime of over 10 years. Syngas cooling is via a cooled gas recycle quench to bring any ash particles in the syngas through the ash solidification temperature range and a convective steam-raising syngas cooler. A schematic of the gasifier is shown in Figure 11-10. When low-rank coals are gasified, it is important that the pulverized coal is absolutely free of surface water, which could cause the particles to agglomerate and disturb the pneumatic conveying function. For such applications, Shell uses RWE’s Mechanical Vapor Compression process, in which the steam driven off from the coal in a fluid-bed dryer is compressed and used as the fluid-bed heating medium. Thus, at the cost of the compression energy, the latent heat of the steam is recovered, which substantially reduces the heat rate penalty when using low-rank coals [4]. The Shell process was first demonstrated in a pilot plant in Hamburg in 1978 built by a joint-venture between Shell and Koppers. It is currently operated by Nuon in its 250 MWe power plant in Buggenum, The Netherlands. Nuon has recently announced a project to build three more such gasifiers at a 1200 MWe multi-fuel power site in Eemshaven [5]. A 350 MWe plant, designed by Koppers, based on the results of the original joint venture operates in Puertollano, Spain. About 15 plants, mostly of the 2,000 short tons/day (t/d) plus class, are in various stages of planning, construction, and operation in China, mostly for chemical operations.
Gasification
439
FIGURE 11-11 E-Gas two-stage gasifier (With permission from ConocoPhillips).
11.4.3 E-Gas (ConocoPhillips) The E-Gas process, shown in Figure 11-11, uses a coal-water slurry feed into an up-flow, entrained-flow gasifier. A key identifying feature is the use of two-stage gasification. The second stage uses the heat in the syngas product from the first stage to devolatize and gasify the second-stage feed. This allows the syngas outlet to be cooler than the ash melting temperature, although the first stage is operating at slagging temperatures. This contributes to improved cold gas efficiency for the process. Further cooling is by means of a convective cooler. The reactor is refractory lined. The E-Gas technology, now owned by ConocoPhillips, was originally developed by Dow, which built a 550 t/d pilot plant in Plaquemine, Louisiana, in 1983. This was followed by a 1600 t/d 165 MWe IGCC production unit on the same site, which operated on subbituminous coal between 1987 and 1995. These plants provided the basis for the Wabash River 250 MWe IGCC, which went on-stream in 1996 [6]. E-Gas technology is being applied in a number of 630 MWe power plants currently in planning in the United States.
11.4.4 Siemens (formerly Future Energy GSP) The Siemens fuel gasifier (SFG) is a down-flow, entrained-flow gasifier offering a dry-feed capability for high efficiency and a membrane wall reactor containment, which requires minimum maintenance. The process, shown in Figure 11-12, was developed in the early 1980s specifically to gasify high-sodium lignite from the central German fields. The very high
440 Combustion Engineering Issues for Solid Fuel Systems FUEL
GAS TO PILOT BURNER OXYGEN
BURNER
PRESSUR. WATER OUTLET
COOLING SCREEN
PRESSUR. WATER INLET QUENCH WATER COOLING JACKET
GAS OUTLET
WATER OVEREFLOW
GRANULATED SLAG
FIGURE 11-12 Siemens (formerly GSP) gasifier (Source: [8]).
inherent moisture content of this fuel dictated the use of a dry feed. The high sodium content led the developers to use a membrane wall and a water quench. The first commercial plant with a 200 MWth feed capacity went into service in 1984 in Schwarze Pumpe, Germany, where it continued to operate on lignite until 1990, when on German reunification the
Gasification
441
operator changed his business model and reconfigured the plant to use waste liquid feeds. Six gasifiers are currently in various stages of planning and construction in China, including two 500 MWth units [7].
11.4.5 KBR Transport Gasifier The KBR Transport Gasifier, shown in Figure 11-13, is a fluid-bed gasifier operating in a high-velocity regime, which has been under development since the early 1990s. The objective of this development was to demonstrate higher circulation rates, velocities, and riser densities than conventional circulating beds, resulting in higher throughput and better mixing and heat transfer rates. The first commercial demonstration unit will be in a 280 MWe under construction in Orlando, Florida [9]. The fuel (and, if required, limestone sorbent for bulk sulfur removal) is fed to the reactor through separate lock hoppers. Fuel and, when required, sorbent are mixed in the mixing zone with oxidant and steam and recirculated solids from the standpipe. The gas with entrained solids moves upward from the mixing zone into the riser. The riser outlet makes two turns before entering the disengager, where larger particles are TO PRIMARY GAS COOLER
DISENGAGER
RISER
CYCLONE
MIXING ZONE
LOOPSEAL
COAL SORBENT AIR STEAM J-LEG STARTUP BURNER (PROPANE)
STANDPIPE
SOLIDS AIR OXYGEN STEAM
FIGURE 11-13 KBR transport gasifier (With permission from KBR).
442 Combustion Engineering Issues for Solid Fuel Systems
removed by gravity separation. Smaller particles are largely removed from the gas in the cyclone. The solids collected by the disengager and cyclone are recycled to the mixing zone via the standpipe and J-leg. In the pilot plant, the sorbent added to the fuel reacts with the sulfur present to form CaS. This, together with a char-ash mixture, leaves the reactor from the standpipe via a screw cooler. These solids and the fines from the candle filter are combusted in an atmospheric fluid-bed combustor. The 72 t/d pilot plant reactor was operated in combustion mode from 1997 to 1999. Since then it has been operated in gasification mode, using both air and oxygen as oxidant. Average carbon conversion rates are about 95%, and values of up to 98% have been achieved. Gasification takes place at 1,650–1,800 F; and pressures in the pilot plant, between 140 and 250 psig.
11.4.6 Lurgi The Lurgi dry-bottom gasifier, shown in Figure 11-14, is still the process with the largest production of syngas worldwide, based on Fischer-Tropsch applications in South Africa and an SNG application in Beulah, North COAL
COAL LOCK
TAR RECYCLE JACKET STEAM
COAL DISTRIBUTOR
WASH COOLER
GRATE WATER JACKET STEAM & OXYGEN
ASH LOCK
ASH
FIGURE 11-14 Sasol-Lurgi dry-bottom gasifier (With permission from Lurgi AG).
Gasification
443
Dakota. It is a fixed-bed process. The counter-current flow provides a high methane content in the syngas, which gives the process a high cold-gas efficiency. The syngas also contains tars devolatilized from the incoming feed, and these must be removed from the syngas for most applications. In the facilities mentioned here, the tars are further processed into high-value chemicals. Modifications of the process have been developed. The BGL is a slagging version without the lower grate, which is able to recycle the tars to the lower part of the gasifier. Another modification is the MPG liquids gasifier, which was developed to process the tars into syngas in a separate gasifier.
11.4.7 Raw Gas Analyses Typical raw gas analyses from the various gasifiers are shown in Table 11-3. Note that values will vary according to coal quality, oxygen purity, operating pressures, and other parameters.
11.5 Trace Components in Gasifier Syngas Apart from the main components of synthesis gas, H2, CO, H2O, CO2, and methane, many other components can have an influence on downstream gas processing. Typically, some of these components can act as catalyst poisons; require special treatment to avoid environmental problems; or can create corrosion, fouling, or solvent degradation in the gas-processing scheme. Thus, an integrated plant design needs to address the origin and fate of these components.
11.5.1 Sulfur Compounds In high-temperature processes, all sulfur components in the feed are converted to H2S or COS. Other compounds such as SOx or CS2 are essentially TABLE 11-3 Typical Raw Gas Analyses Process CO2 CO H2 CH4 N2 þ Ar
mol% mol% mol% mol% mol%
GEE (quench)
Shell
E-Gas
Siemens
KBR (air)
KBR (O2)
Lurgi
20.7 41.7 37.1 0.1 0.4
1.7 62.4 31.0 <0.1 4.9
12.8 48.7 35.9 1.3 1.3
4.4 57.3 29.7 <0.1 8.6
8.4 23.6 12.0 2.3 53.8
22.7 38.2 34.6 3.1 1.3
31.5 15.7 42.6 9.5 0.7
Note: All analyses are given on dry, sulfur-free basis. Values will vary according to coal quality, oxygen purity, operating pressure, etc. Lurgi value for methane includes higher hydrocarbons.
444 Combustion Engineering Issues for Solid Fuel Systems
absent. This is not the case in low-temperature processes such as the Lurgi process, where tars and other species have not been completely cracked, and mercaptans, for example, may be present in the gas. The H2S:COS ratio of a raw gas is determined by the hydrogenation and hydrolysis reactions. Under typical gasification conditions, H2S is the dominant species and approximately 93–96% of the sulfur is in this form, the rest being COS.
11.5.2 Nitrogen Compounds Nitrogen enters the gasifier in two forms, either as molecular nitrogen, generally in the gasification agent (but also as a component in gaseous feeds) or as organic nitrogen in the fuel. While the bulk of the nitrogen in the synthesis gas is present as molecular nitrogen, most gasifiers produce small amounts of HCN and NH3.
11.5.3 Chlorine Compounds Chlorine compounds are present in most coals. They will react with ammonia in the raw gas to form ammonium chloride (NH4Cl). At high temperatures, ammonium chloride is in the vapor phase, but below 480–540 F it becomes solid and presents a fouling risk to the gas cooling train. At lower temperatures still, below the water dew point of the gas, it goes into solution and is highly corrosive. These aspects have to be considered in the design of the cooling train. Metals in the feedstock will also from chlorides (e.g., sodium chloride). Many of these have melting points in the range 650–1,450 F and represent a fouling risk in heat exchangers. For this reason, water quench cooling is usually selected when designing to process a high-sodium feed.
11.5.4 Unsaturated Hydrocarbons The existence of unsaturated hydrocarbons in the raw synthesis gas varies very widely. In low-temperature, counter-current processes such as the Lurgi process, there will, in general, be significant quantities of aromatics and other unsaturates in the volatiles and tars, though the exact amount will also depend heavily on the feedstock. For high-temperature entrained-flow processes, the presence of any hydrocarbon other than methane, whether saturated or unsaturated, is minimal.
11.5.5 Oxygen In high-temperature gasification processes, whether of coal or oil, the oxygen is completely consumed in the reaction, and no oxygen is contained in the synthesis gas. One should, however, be aware of the potential of
Gasification
445
introducing small quantities of oxygen into the gas in the subsequent processing, either in low-pressure processes or via the wash water, and take appropriate precautions when oxygen-sensitive catalysts are to be used in the downstream processing.
11.5.6 Formic Acid At higher partial pressures, carbon monoxide will react with water to form formic acid. The thermodynamics of the reaction favor formic acid formation at lower temperatures so that this is particularly noticeable in the gas condensate. At the lower end of the operating temperature range for COS hydrolysis (300 F), formic acid can be formed on the catalyst.
11.5.7 Carbon In coal gasification there is always a certain amount of the initial carbon feedstock which is carried over unconverted in the form of char as particulate matter into the gas. Typically, in an entrained-flow gasifier, this is less than 0.5%. This carbon is encased in particles of ash and thus inaccessible to further conversion. It can be extracted from the gas in a particulate filter (Shell, E-Gas) or a wet scrubber (GEE, Siemens). Plants operating at lower carbon conversion can separate most of it from the slag and recycle it. This mode of operation does, however, carry an oxygen consumption penalty.
11.5.8 Metal Carbonyls The formation of nickel and iron carbonyls can take place in the presence of gaseous carbon monoxide in contact with metallic nickel or iron or their sulfides. The thermodynamics favor this reaction at low temperatures, so the formation takes place during gas cooling. This is primarily an issue which needs to be addressed when processing fuels derived from oil refining such as petroleum coke or other refinery residues. The consequences of any metal carbonyl slip into the gas treatment units depend very much on the treatment scheme, since any subsequent reheating, e.g., for a CO shift, will tend to decompose the carbonyls back to the metals or their sulfides. Carbonyls are not removed from the gas by amine washes but are removed by physical solvents such as Rectisol or Selexol.
11.5.9 Mercury Mercury can be present in coal although the quantities vary widely from source to source. Mercury presents a potential hazard both for the integrity of the plant and as a toxic emission for the environment. When coal is gasified, mercury from the feed will appear at least in part in the synthesis gas, and so for these feeds, it is necessary to address this feed contaminant.
446 Combustion Engineering Issues for Solid Fuel Systems
Considerable research is being conducted into developing technologies for removal of mercury from flue gas, and they are all extremely expensive. The situation for gasification technologies is different. Proven and economic methods for mercury removal are available and are standard practice in the LNG industry to prevent mercury present in natural gas entering the cryogenic plant and causing serious damage to the equipment. In these applications, mercury is adsorbed onto sulfur-impregnated carbon, which can achieve recovery rates of the order of magnitude of 95%. Experience of this technology in a syngas application has been demonstrated for over 20 years in the Eastman methanol plant in Kingsport, Tennessee.
11.5.10 Arsenic One of the problems associated with coal utilization is that in coal many of the elements of the periodic table can be found in minor concentrations. An element of emerging concern is arsenic, which may be present in concentrations in the order of 1–10 ppm in coal. Toxic elements are of no concern when they end up bound in the slag or in stable chemical compounds. The problem with arsenic is that under reducing conditions it forms the volatile compound AsH3. It is a known poison for ammonia catalysts, but recorded instances of this occurring in commercial plants have not been found.
11.6 Gas Treating 11.6.1 Introduction Primary gas treating, which includes particulate removal and a water wash to extract ammonia and chlorides from the gas, is generally considered to be an integral part of the gasification process. The measures taken vary slightly from process to process and were discussed in Section 11.3. Additional treatment always includes desulfurization, for which a number of different solutions are available. Other gas treatment requirements will be project specific depending on application (power or chemicals) and regulatory requirements. These might include mercury removal, and if carbon capture is required, CO Shift (converting CO and steam to hydrogen and CO2), and CO2 removal. Where hydrogen is required, then membranes and/or a pressure swing adsorption (PSA) unit might also be used. Synthesis gas made from coal does not contain any oxides of nitrogen, but of course these are generated in the combustors of a gas turbine. The levels produced are lower than that from a conventional combustion-based power plant, but typically higher than for a natural gas-fired turbine, so that in some regulatory regimes, a selective catalytic reactor (SCR) may be required. Further details are discussed in Section 11.8.2.
Gasification
447
11.6.2 Desulfurization The desulfurization technologies applied to synthesis gas have a long commercial-scale history in the refining and natural gas industries. The primary technologies are based on the chemical or physical absorption of H2S in a suitable solvent. About 5% of the sulfur in the raw gas is present as COS, so that where deep desulfurization is required, attention must be given to its removal as well. Physical solvents have at least a partial capability of removing COS, whereas chemical solvents do not. In the latter case the COS is therefore hydrolyzed to H2S before the main desulfurization step. The H2S extracted is then processed to a useable/saleable product such as elemental sulfur (in a Claus plant) or sulfuric acid. The most common desulfurization processes can be grouped under four main headings: Chemical solvents, mainly amines, such as MEA, DEA, MDEA,
DIPA, and others. This group also includes potassium carbonatebased systems such as the Benfield wash. The most widely employed systems today are based on MDEA, which will be described in more detail later. Physical solvents, the most applicable being Selexol (mostly for power applications) and Rectisol (mostly for chemical applications). Physical/chemical solvents, of which Sulfinol is the most prominent example. Liquid redox systems, which include the special feature that elemental sulfur is produced as part of the solvent regeneration step without an additional processing unit. Absorption processes are characterized by washing the synthesis gas with a liquid solvent which selectively removes the acid components (mainly H2S and CO2) from the gas. The laden solvent is regenerated, releasing the acid components and recirculated to the absorber. The washing or absorption process takes place in a column, which is usually fitted with packing (structured or unstructured) or trays. The loading capacity of a physical wash depends primarily on Henry’s law and is therefore practically proportional to the partial pressure of the component to be removed (see Figure 11-15). This leads to the fact that the solution rate for any particular operating pressure is approximately proportional to the volume of raw gas to be processed. In contrast, the loading capacity of a chemical wash is limited by the quantity of the active component of the solution. Once a saturation level is reached, only a minor additional loading can be achieved by physical absorption in the solution. The solution rate is approximately proportional to the volume of acid gas removed.
Partial pressure, [bar]
448 Combustion Engineering Issues for Solid Fuel Systems
Physical solvent Chemical solvent
ΔIch ΔIph Loading capacity, [kmol/m3 solvent]
FIGURE 11-15 Loading capacities of physical and chemical solvents.
Generally, solvent regeneration is achieved by one of or a combination of flashing, stripping, and reboiling. Both flashing and stripping reduce the partial pressure of the acid component. Reboiling raises the temperature to break the chemical bond and, in the case of physical solvents, also uses the solvent vapor to strip out the acid gas components. In such systems, the acid components are released in the same chemical form in which they were absorbed.
11.6.3 Chemical Solvent Processes 11.6.3.1 Amine Processes Figure 11-16 shows the flow sheet of a typical MDEA wash, although this flow sheet is representative of many other chemical washing processes. The raw syngas is contacted in a wash column with lean MDEA solution, which absorbs the H2S and some of the CO2. MDEA is to some extent selective in that the bonding of the amine with H2S takes places faster than with CO2 and advantage can be taken of this in the design. The rich solution is preheated by heat exchange with the lean solution and enters the regenerator. Reboiling breaks the chemical bind, and the acid gas components discharged at the top of the regenerator are cooled to condense out the water, which is recycled.
11.6.4 Physical Solvent Processes 11.6.4.1 Physical Washes The important characteristics for any successful physical solvent are as follows: Good solubility for CO2, H2S, and COS in the operating range,
preferably with significantly better absorption for H2S and COS
Gasification
449
FIGURE 11-16 Typical MDEA flow sheet.
compared with CO2 if selectivity is an important issue for the application of interest. Low viscosity at the lower end of the operating temperature range. Although lowering the operating temperature increases the solubility, the viscosity governs, in effect, the practical limit to lowering the operating temperature. A high boiling point reduces vapor losses when operating at ambient or near ambient temperatures. 11.6.4.2 Selexol The Selexol process was originally developed by Allied Chemical Corporation and is now owned by UOP. It uses dimethyl ethers of polyethylene glycol (DMPEG). The physical properties of DMPEG are listed in Table 11-4. The typical operating temperature range is 15–100 F. The ability to operate in this temperature range offers substantially reduced costs by eliminating or minimizing refrigeration duty. On the other hand, for a chemical application such as ammonia, the residual sulfur in the treated gas may be 1 ppm H2S and COS each after the CO2 wash [10]. This is, however, not an issue in power applications where the sulfur slip is less critical. Selexol has a number of references for such plants, including the original Cool Water demonstration unit and most recently the 550 MW Sarlux IGCC facility in Italy. The ratio of absorption coefficients for H2S, COS, and CO2 is about 1:4:9 in descending order of solubility [11]. A plant designed for, say, 1 ppm COS in the clean gas would require about four times circulation rate of a plant for 1 ppm H2S together with all the associated capital and
450 Combustion Engineering Issues for Solid Fuel Systems TABLE 11-4 Properties of Physical Solvents Process Solvent Formula Mol. weight Boiling point at 760 Torr Melting point Viscosity
Specific weight Selectivity at working temperature
lb/lb mol F F cP cP cP kg/m3 (H2S:CO2)
Selexol
Rectisol
DMPEG CH3O(C2H4O)xCH3 178 to 442 415 to 870 –4 to –20 4.7 at 86 F 5.8 at 77 F 8.3 at 59 F 1.031 1:9
Methanol CH3OH 32 147 –137 0.85 at 5 F 1.4 at –22 F 2.4 at –58 F 790 1:9.5
operating costs. In a gasification environment, it is therefore preferable to convert as much COS as possible to H2S upstream of a Selexol wash. In a plant using raw gas shift for hydrogen or ammonia, this will take place simultaneously on the catalyst with the carbon monoxide shift. Where no CO shift is desired, then COS hydrolysis upstream of the Selexol unit provides a cost-effective solution to the COS issue. Other characteristics favorable for gasification applications include high solubilities for HCN and NH3 as well as for nickel and iron carbonyls. The Selexol flow sheet in Figure 11-17 exhibits the typical characteristics of most physical absorption systems. The intermediate flash allows co-absorbed syngas components (H2 and CO) to be recovered and recompressed back into the main stream. For other applications such as H2S concentration in the acid gas or separate CO2 recovery using staged flashing, techniques not shown here may be applied. 11.6.4.3 Rectisol The Rectisol process, which uses cold methanol as solvent, was originally developed to provide a treatment for gas from the Lurgi moving-bed gasifier, which, in addition to H2S and CO2, contains hydrocarbons, ammonia, hydrogen cyanide, and other impurities. Figure 11-18 contains a flow sheet of the Rectisol process. In the typical operating range of –20 to –75 F, the Henry’s law absorption coefficients of methanol are extremely high, and the process can achieve gas purities unmatched by other processes. This has made it a standard solution in chemical applications such as ammonia, methanol, or methanation, where the synthesis catalysts require sulfur removal to less than 0.1 ppm. This performance has, however, a price in that the refrigeration duty required for operation at these temperatures involves considerable capital and operating expense.
Gasification TREATED GAS
451
ACID GAS TO SRU
ACID GAS CONDENSER MECHANICAL FILTER
LEAN SOLUTION COOLER
MAKE UP WATER LEAN SOLUTION PUMP
REFLUX PUMP STRIPPER
ABSORBER RECYCLE COMPRESSOR
RAW SYNGAS
STEAM
REBOILER HYDRAULIC TURBINE FLASH DRUM RICH/LEAN EXCHANGER
FIGURE 11-17 Selexol flow sheet for selective H2S removal (Source: [11]).
METHANOL SYNTHESIS RAW GAS
CO SHIFT MP FUEL GAS CO2
REFRIGERANT
ACID GAS
STEAM H2S ABSORBER
H2S FLASH COLUMN
HOT REGENERATOR
H2S REMOVAL UNIT
FIGURE 11-18 Flow sheet of selective Rectisol process.
CO2 ABSORBER
CO2 FLASH TOWER
CO2 REMOVAL UNIT
452 Combustion Engineering Issues for Solid Fuel Systems
Methanol as a solvent exhibits considerable selectivity, as can be seen in Table 11-4. This allows substantial flexibility in the flow sheeting of the Rectisol process, and both standard (nonselective) and selective variants of the process are regularly applied according to circumstances. As a physical wash, which uses at least in part flash regeneration, part of the CO2 can be recovered under an intermediate pressure. Typically, with a raw gas pressure of 710 psig, about 60–75% of the CO2 would be recoverable at 40–60 psig. Where CO2 recovery is desired, whether for urea production in an ammonia application or for sequestration, this can provide significant compression savings. Figure 11-18 shows the selective Rectisol variant as applied to methanol production. The incoming raw gas is cooled down to about –20 F, the operating temperature of the H2S absorber. Both H2S and COS are washed out with the cold methanol to a residual total sulfur content of less than 100 ppb. The desulfurized gas is then shifted outside the Rectisol unit, the degree of shift being dependent on the final product. Carbon dioxide is then removed from the shifted gas in the CO2 absorber. This column is divided into two sections: a bulk CO2 removal section using flashregenerated methanol and a fine CO2 removal section in which hotregenerated methanol is used. The CO2 removal section operates at lower temperatures, typically about –75 F. The permissible CO2 slip is dependent on the application. For methanol synthesis gas, 1 mol% residual CO2 is quite adequate. For hydrogen production based on methanation, typically 100 ppm would be appropriate. For ammonia, where the gas is subsequently treated in a cryogenic nitrogen wash, 10 ppm would be typical. Following the solvent circuit is an intermediate H2S flash from which co-absorbed hydrogen and carbon monoxide are recovered and recompressed back into the raw gas. The flashed methanol is then reheated before entering the hot regenerator. Here, the acid gas is driven out of the methanol by reboiling, and a Claus gas with an H2S content of 25–30% (depending on the sulfur content of the feedstock) is recovered. Minor adaptations are possible to increase the H2S content if desired. The hot-regenerated methanol, which is the purest methanol in the circuit, is used for the fine CO2 removal. The methanol from the CO2 removal is subjected to flash regeneration in a multistage flash tower. The configuration shown is typical for the methanol application with only atmospheric flash regeneration. For a hydrogen or ammonia application where better absorption is required, the final flash stage may be under vacuum, or it may use stripping nitrogen from the air separation plant. Finally, the loop is closed with the flash-regenerated methanol returning to the H2S absorber. Water entering the Rectisol unit with the syngas must be removed; an additional small water-methanol distillation column is included in the process to cope with this. Typically, the refrigerant is supplied at between –20 F and –40 F. Depending on application, different refrigerants can be
Gasification
453
used. In an ammonia plant, ammonia is used and the refrigeration system is integrated with that of the synthesis. In a refinery environment, propane or propylene may be the medium of choice. The Rectisol technology is capable of removing not only conventional acid gas components but also, for example, HCN and hydrocarbons. Supp [12] describes a typical hydrocarbon prewash system. Mercury capture using Rectisol as a cold trap to condense out metallic mercury is also documented [13]. 11.6.4.4 Liquid Redox Processes Liquid redox processes can selectively remove H2S in the presence of high CO2 concentrations. They remove H2S only and will not remove CO2, COS, CS2, or mercaptans. The active agent oxidizes the H2S to sulfur in an absorber column. The solution is circulated to an oxidizer for regeneration. The sulfur is recovered from the solution and processed typically by filtration and melting (to remove moisture) to make a marketable product grade. Typical of liquid redox processes is the older Stretford process, which uses a vanadium-based agent. Modern processes such as Lo-Cat or Sulferox use iron, which is held in dilute solution (high circulation rates) by the common chelating agent ethylene diamine tetra acetic acid (EDTA). The net effect of a liquid redox plant is comparable to an amine plant plus a Claus sulfur recovery plant. Usually, the amine/Claus combination is favored over liquid redox for large plants. A liquid redox plant comes into its own in smaller plants with sulfur capacities under 15 t/d and where the H2S:CO2 ratio in the raw gas is low.
11.6.5 Membranes Permeable gas separation membranes utilize differences in solubility and diffusion of different gas components in polymer materials. In recent years, membrane-based technologies have gained increasing importance in gas processing. They can be used to skim off hydrogen out of the raw syngas. The degree of selectivity between different gas components depends on the membrane used, and if pure hydrogen is required, an additional PSA step is usually applied. Membrane separation units are typically supplied skid mounted.
11.6.6 COS Hydrolysis In all synthesis gases produced by gasification, sulfur is present not only as H2S, but also as COS. Typically, a syngas from the gasification of a refinery residue with 4% sulfur may contain about 0.9 mol% H2S and 0.05 mol% COS. While some washes such as Rectisol can remove the COS along with the H2S, others, particularly amine washes, require the COS to be
454 Combustion Engineering Issues for Solid Fuel Systems
converted selectively to H2S if the sulfur is to be substantially removed. This is best achieved by catalytic COS hydrolysis, according to the reaction COS þ H2 O ⇆ H2 S þ CO2
30 MJ=kmol
[11-5]
Commercially, this reaction takes place over a catalyst at a temperature in the range of 300–500 F. Various catalysts are available, including a promoted chromium oxide-alumina, pure activated alumina, or titanium oxide. Lower temperatures favor the hydrolysis equilibrium. Typically, the optimum operating temperature is in the range 300–400 F. Depending on process conditions, the residual COS can be reduced to the range of 5–30 ml/Nm3. Some catalysts also promote the hydrolysis of HCN. The catalyst operates in the sulfided state and is not poisoned by heavy metals or arsenic. Halogens in the gas will, however, reduce activity, selectivity, and lifetime—a fact that needs to be addressed carefully in coal gasification applications. The gasification of refinery residues may form nickel and iron carbonyls that can decompose depositing nickel or iron sulfide on the catalyst bed, thus creating an increased pressure drop over the system. Typically, a guard bed, which can be taken offline during normal operation, is installed upstream of the COS hydrolysis bed to catch these deposits. For any application, care must be taken to avoid catalyst degradation by liquid water. Catalyst manufacturers recommend that the bed temperature be maintained at least 50 F above the water dew point under all circumstances.
11.6.7 CO Shift Besides having an important influence on the composition of the raw syngas from the gasifier itself, the CO shift reaction can be and is operated as an additional and separate process from the gasifier at much lower temperatures in order to modify the H2/CO ratio of the syngas or maximize the total hydrogen production from the unit. As can be seen, one mole of hydrogen can be produced from every mole of CO. The reaction is largely independent of pressure. The equilibrium for hydrogen production is favored by low temperature. CO þ H2 O ⇆ CO2 þ H2
41 MJ=kmol
[11-6]
The CO shift reaction will operate with a variety of catalysts between 400 F and 930 F. The types of catalyst are distinguished by their temperature range of operation and the quality (sulfur content) of the syngas to be treated.
Gasification
455
11.6.7.1 Clean Gas Shift High-Temperature (HT) Shift Conventional (high-temperature) shift uses an iron oxide-based catalyst promoted typically with chromium and more recently with copper. The operating range of these catalysts is between 570 F and 930 F. Much above 930 F, sintering of the catalyst sets in and it is deactivated. An HT shift catalyst is tolerant of sulfur up to a practical limit of about 100 ppm, but is likely to lose mechanical strength, particularly if subjected to changing amounts of sulfur. An important aspect in the design of CO shift in the gasification environment, where inlet CO contents of 45% (petroleum residue fed) up to 65% (coal) are common, is the handling of the heat of the reaction, particularly under end-of-run conditions, where an inlet temperature of 660 F or more may be necessary. On the one hand, the reaction must be performed in several stages to avoid excessive catalyst temperatures and to have an advantageous equilibrium. On the other hand, optimum use must be made of the heat. One such arrangement is shown in Figure 11-19. Desulfurized syngas containing about 45 mol% CO, which leaves the acid gas removal (AGR) system at about 770 psig, and ambient temperature is heated and water saturated at a temperature of about 420 F by water which has been preheated with hot reactor effluent gas. The saturated gas is further preheated to the catalyst inlet temperature of between 570 F and 680 F. The steam loading from the saturator is such that only the stoichiometric steam demand for the reaction is required to be added from external sources. In the first stage, the CO is reduced to a level of about 7–8 mol% at an outlet temperature of about 900 F. The outlet gas is cooled to a temperature of about 720 F in the gas and water preheaters before entering the second catalyst bed. Here, the residual CO is reduced to about 3.2 mol%. The gas is then cooled in a direct contact desaturator tower. There are a number of different designs, particularly for the first reactor, which incorporate the SATURATOR
REACTOR I
REACTOR II
DESATURATOR
PROCESS STEAM
SHIFT GAS
HEAT RECOVERY DESULFURIZED GAS
FIGURE 11-19 CO Shift with saturator-desaturator circuit.
456 Combustion Engineering Issues for Solid Fuel Systems
gas-gas heat exchanger as an internal. In such reactors, the exchanger is arranged centrally inside an annular catalyst bed with an axial (Lurgi) or axial-radial (Casale) gas flow pattern. Alternative methods of controlling the catalyst outlet temperature include interbed condensate injection. The use of an isothermal steam-raising reactor has been proposed, and although such a solution has been employed in a steam reformer plant, none is recorded at the high CO inlet concentrations involved in a gasification plant. Typical catalyst lifetime for the first bed in a gasification situation is 2–3 years, which is considerably shorter than for a steam-reforming situation. This is generally attributed to the high operating temperatures associated with high CO concentrations in the inlet gas. On a molesconverted basis over the lifetime of the catalyst, the performance in the gasification context is comparable with that of steam reforming. Low-Temperature (LT) Shift Low-temperature shift operates in the temperature range 300–520 F and uses a copper-zinc-aluminum catalyst. It is used in most steam reforming-based ammonia plants to reduce residual CO to about 0.3 mol%, but has generally not been applied in gasificationbased units. On the one hand, it is highly sulfur sensitive, and even with 0.1 ppm H2S in the inlet, gas will over time become poisoned. A second reason for its lack of use in, particularly, oil gasification plants is the effect of the higher pressure on the water dew point in the gas. Operation near the dew point will cause capillary condensation and consequent damage to the catalyst. With a dew point of about 420 F and a temperature rise of about 50 F, there is not much margin for error below the upper temperature limit of 520 F when recrystalization of the copper catalyst begins. The first application of low-temperature shift at high pressure was in Shell’s Pernis gasification facility, which has now performed successfully for several years [14]. 11.6.7.2 Raw Gas Shift For applications in which it is desired to perform CO shift on raw syngas, a cobalt-molybdenum catalyst, variously described as a sour shift or dirty shift catalyst, can be used. In some parts of the literature, this catalyst is described as sulfur tolerant. This is actually a misnomer, since the catalyst requires sulfur in the feed gas to maintain it in the active sulfided state. It is generally applied after a water quench of the raw syngas, which typically will provide a gas at about 480 F saturated with sufficient water to conduct the shift reaction without any further steam addition. For an ammonia application, the raw gas shift is typically configured as two or three adiabatic beds with intermediate cooling, resulting in a residual CO of about 1.6 or 0.8 mol%, respectively. An important side effect of the raw gas shift catalyst is its ability to handle a number of other impurities characteristic of gasification. COS
Gasification
457
and other organic sulfur compounds are largely converted to H2S, which eases the task of the downstream AGR. HCN and any unsaturated hydrocarbons are hydrogenated. Carbonyls are decomposed and deposited as sulfides, which increases pressure drop over the bed. Selective removal of arsenic in the feed is also claimed [15].
11.6.8 Mercury Removal Mercury removal has in the past not been a requirement in gasification systems except in the specific chemical application of Eastman in Kingsport, Tennessee. Simple activated carbon beds filled with a sulfur-impregnated carbon are sufficient to remove 95% of the mercury present in the gas. The beds have a typical lifetime of about 2 years between change-outs. It is anticipated that in the future all IGCC applications will include mercury removal.
11.7 Complete Systems Complete systems, be it for power production or chemical applications, require considerable skill in finding an optimum integration. Some contaminants in the gas must be removed early in the overall gas treatment, so as to prevent disturbances such catalyst deactivation or corrosion in downstream facilities. The optimum temperature for each treatment stage must be considered in setting up a flow scheme. And, while respecting both of these conditions, one needs to find an optimal energy balance. There are, of course, a number of different solutions, and some typical results of such integration are discussed in the following sections.
11.7.1 Integrated Gasification-Combined Cycle (IGCC) The basic structure of the IGCC is shown in Figure 11-20. Within this framework, however, there is considerable scope for variation, as is demonstrated by four existing coal-based IGCCs of the 250–350 MWe class in Polk County, Florida; Wabash, Indiana; Buggenum, The Netherlands; and Puertollano, Spain. Some reference is also made to liquid-feed IGCCs in the size range up to 550 MWe. The purpose of this section is to review the choices available within the basic structure. Note that the main systems in the IGCC have been grouped into four major blocks, which are further discussed in the following paragraphs. The first variation to be considered is the extent of air-side integration, which can range from 0%, as in Polk or Wabash, to 100%, as in Buggenum or Puertollano, where the degree of integration is defined as the percentage of air supplied to the air separation unit (ASU) by extraction from the gas turbine. This difference in the first generation (1990s) of IGCCs can largely be attributed to the gas turbines available at the time.
458 Combustion Engineering Issues for Solid Fuel Systems CCU block
BALANCE OF PLANT
ASU block
COMBINED CYCLE UNIT
N2 ASU O2
FEEDSTOCK PREPARATION
Gasification block GASIFICATION & HT GAS COOLING
PARTICULATE REMOVAL
PRETREATMENT & LT GAS COOLING
ACID GAS REMOVAL
H2S SULFUR RECOVERY
Gas treatment block
FIGURE 11-20 Basic structure of an IGCC.
The 100% integration is clearly disadvantageous in today’s economic environment, since a long start-up period using an expensive backup fuel is required. On the other hand, zero air extraction from the gas turbine air compressor does not allow optimum use of the machine over a full range of ambient conditions. The optimum degree of integration is dependant on many factors, including the ambient temperature range and the turbine selection. A typical figure today might be around 30%. The quality of the oxygen is another variable. Over the range 85% (Puertollano) to 95% (most other plants), the optimum curve for energy consumption is fairly flat. Additional energy is required in the ASU to raise the quality further. A purity of 99.5% O2 is typically not attractive for a straight IGCC application, but for chemical applications it is generally the standard. Where production of power with one or more chemicals (often known as polygeneration) is to be considered, one would need to review the oxygen purity specification on an individual case. If the co-product were to be ammonia, for instance, 95% O2 would be perfectly satisfactory. Other choices to be made include the distillation pressure in the ASU (making use of the higher pressure available from the extraction air) and the choice of oxygen compression system (gas phase or liquid phase). These choices are best left to the ASU vendor. Finally, a decision needs to be made on the provision of storage for liquid oxygen and liquid nitrogen. Both storage facilities are interconnected with the plant availability. Essentially, oxygen storage can provide
Gasification
459
an opportunity for improving availability. Conversely, the poorer the availability, the more liquid nitrogen storage is desirable. Air blown gasification, i.e., without any ASU at all, is not considered here, although it should be noted that one such plant (Southern Company, Orlando, Florida; 280 MWe) using the KBR Transport Gasifier is currently in the planning stages. 11.7.1.2 Gasification Block Variations in the gasification block are almost entirely dependent on the choice of technology supplier. Important differences are discussed in this section. In feedstock preparation, rod mills are used for slurry preparation by GEE and ConocoPhillips and roller mills are used by Shell and Siemens. In both cases, the particle size is of the order of magnitude of <100 microns. In the case of fluid-bed processes such as the Transport Gasifier mentioned previously, the particle size is much larger (6 mm). The feedstock pressurization is directly connected to the feedstock preparation method. Slurry feed units use slurry pumps. Dry feed units need to employ lock hoppers and pneumatic conveying. For an entrained-flow gasifier, the flow direction can be down-flow (GEE or Siemens) or up-flow (ConocoPhillips or Shell). The temperature containment can be with refractory (GEE or ConocoPhillips) or using a membrane wall (Shell or GSP). Only ConocoPhillips uses a two-stage gasifier. The other technology suppliers use single-stage gasification. Syngas cooling is available in a number of variations. Water quench (GEE or Siemens) is not currently used in the coal-based IGCC configuration, primarily because of the associated efficiency penalty. It is used, however, in chemical applications, particularly where CO shift for hydrogen manufacture is involved (Kingsport, Tennessee; Coffeyville, Kansas). It is also used in a number of refinery-based IGCC units. Should CO2 capture have to be implemented from the beginning of a project (as opposed to being retrofitted later), then this cooling technique would probably be favored also for coal-based IGCC applications. Of the different steam-raising configurations, radiant cooling is offered only by GEE (e.g., Polk). ConocoPhillips uses a firetube convection cooler after the second stage of its E-Gas gasifier. GEE has also used a firetube convection cooler as a second cooling stage in Polk, but it has been deleted in current designs. Shell uses a gas quench and watertube syngas cooling. A certain amount of syngas pretreatment is generally included in the scope of the gasification technology supplier. This includes removal of particulate matter and a number of trace components in the gas, particularly ammonia and chlorides. Shell and ConocoPhillips remove the particulates and the water-soluble gases in separate stages, using a candle filter (sinter metal for ConocoPhillips and ceramic for Shell) for particulate removal
460 Combustion Engineering Issues for Solid Fuel Systems
and a water wash for ammonia and chlorides. GEE combines these steps in a single scrubber. Slag removal from the pressurized gasifier is achieved using a lock hopper arrangement in most processes. Only ConocoPhillips has a proprietary continuous let-down system. 11.7.1.3 Gas Treatment and Sulfur Recovery Although sulfur species (primarily H2S) are the principal targets of the gas treatment system, it is necessary to consider the full range of potential contaminants, which include COS (a minor sulfur species) and mercury. Depending on the selection of desulfurization technology, COS will probably need to be hydrolyzed to H2S to achieve the required level of sulfur removal. Typical temperatures for COS hydrolysis are between 320 F and 390 F (160 C and 200 C). Mercury removal is best performed at ambient temperatures upstream of the acid gas removal, so some of the gas treatment will need to be integrated with the low-temperature gas cooling. Mercury removal from syngas has been practiced industrially only at Kingsport although it is a regular feature of natural gas pretreatment in LNG plants. There is an extremely wide variety of AGR systems on the market. They can be classified as chemical washes (which include all amines such as MDEA or ADIP) and physical washes such as Selexol or Rectisol. In addition, it is possible to have a mixed characteristic solvent such as Sulfinol. All these named processes have been used in IGCC or chemical plant gasification operations. Selection is based on requirements for high purity (Rectisol) versus low cost (MDEA), with Selexol and Sulfinol lying in between on both counts. All these processes have a long track record in industrial practice, all with high availability records. The chemical washes are generally not capable of absorbing COS, which must be converted to H2S in a COS hydrolysis step upstream of the wash. Physical washes can absorb COS. In the case of Selexol, this capability is not very strong, and economics usually dictate the use of a COS hydrolysis as well (but not after a CO shift, as in Coffeyville). Rectisol does not require any upstream COS hydrolysis. Sulfur recovery is generally achieved using Claus technology, although Polk is an exception in that it manufactures sulfuric acid rather than elemental sulfur. Differences in the Claus technology itself are generally only of a detailed nature. Considerable variety is shown in the handling of the tailgas from the Claus plant, which in addition to H2S also contains small quantities of SO2, COS, CS2, and elemental sulfur. In all plants, these are hydrogenated back to H2S over a catalyst. In some plants, this gas is then treated separately in another washing stage and then incinerated and discharged to the atmosphere. In others, it is recycled to a point and mixed with the raw gas upstream of the main AGR so that this remaining gas is
Gasification
461
treated there. The point at which the recycle is fed into the main gas stream varies. In all plants, syngas dilution is used to reduce the NOx emissions from the gas turbine. The dilution medium may be nitrogen only (e.g., Polk initially), steam only (e.g., Wabash), or a combination of the two (e.g., Buggenum). Steam is generally added by saturation using low-level heat to provide the necessary hot water. In some cases it is added by direct injection (e.g., Pernis). 11.7.1.4 Combined Cycle Power Plant The combined cycle unit (CCU) block, as described in Figure 11-21, covers the typical scope of an NGCC complete with balance of plant. The principal difference lies in the use of syngas as a fuel. Note that to date little experience is available on the use of selective catalytic NOx reduction (SCR) in IGCCs, where the residual sulfur content in the syngas could impact on the availability of the HRSG. Two units in Italy are equipped with SCRs, but the required NOx emission levels are not comparable with the values required of an NGCC plant. In one case, the SCR is used only when the gas turbine operates on the backup distillate fuel. The only significant experience with SCR in IGCC is at the Negishi plant of Nippon Oil. Values of <2.6 ppm NOx and <2.0 ppm SOx have been reported [16]. It should be noted that the “Balance of Plant” in an IGCC will include a flare and a process wastewater system.
BALANCE OF PLANT
COMBINED CYCLE UNIT
N2 ASU O2
FEEDSTOCK PREPARATION
H2
GASIFICATION & HT GAS COOLING
PARTICULATE REMOVAL
SELECTIVE ACID GAS REMOVAL
CO SHIFT & LT GAS COOLING
CO2 H2S SULFUR RECOVERY
FIGURE 11-21 IGCC with carbon capture.
462 Combustion Engineering Issues for Solid Fuel Systems
11.7.2 IGCC with Carbon Capture Integration of carbon capture into an IGCC is relatively simple using techniques common in, for example, the fertilizer industry, where CO2 is captured in the ammonia plant and compressed to typically 2,200–2,900 psig for urea manufacture. The integration of carbon capture into the IGCC flow sheet is shown in Figure 11-21. The principal changes to the flow sheet of the noncarbon capture case are the introduction of a CO shift reactor after the particulate removal and the modification to the acid gas removal to draw off the CO2 stream separately from the H2S. The configuration up to the outlet of the AGR is a regular feature of coal-based ammonia plants. The CO shift uses a sulfided chrome-molybdenum catalyst. The selective AGR is generally a physical wash. This has the added advantage that much of the CO2 is available under pressure, which reduces compression costs. One well-documented example is the 1000 t/d (100 MWe equivalent) ammonia plant in Coffeyville, Kansas, which feeds petroleum coke into a GEE Quench Gasifier and uses Selexol for acid gas removal. It achieves on-steam times of 98% between annual turnarounds [17]. There are a number of basically similar plants in China using the Shell gasifier with Rectisol. The main area where industrial experience is limited is on the use of hydrogen as fuel in a gas turbine, although there is more experience available than generally appreciated, much of it in industrial applications. One example described by GE as the “H2 Fleet Leader” is a frame 6B unit operating regularly on 85–97% hydrogen at “an availability of 96.5%þ running in uninterrupted operation, 24 hours a day over the year” since 1997 [18].
11.7.3 Methanol The following description of an integrated methanol plant serves to illustrate a chemical application. Many of the considerations involved would apply equally to ammonia, Fischer-Tropsch, or SNG production. The flow sheet shown in Figure 11-22 is a simplified version of that used by Eastman in its Kingsport, Tennessee, plant, the main simplification being that no separate pure carbon monoxide stream has been shown. Also not shown are the two activated carbon beds for mercury removal downstream the low-temperature gas cooling. The main considerations to be applied in developing a synthesis gas production scheme for methanol manufacture include the selection of gasification pressure, of syngas cooling arrangement, and of an acid gas removal system. The optimization of oxidant quality is not a consideration, since any inerts in the syngas lower the conversion in the synthesis. The oxygen should simply be as pure as reasonably possible, which in effect means 99.5% purity.
Gasification METHANOL PLANT
463
METHANOL
CO/H2 99.5% O2
SLURRY PREP.
CO2
LT COOLING
GASIFIER
PART. REMOVAL
CO SHIFT
LT COOLING
RECTISOL AGR
H2S COAL CLAUS/ TGT
SULFUR
FIGURE 11-22 Methanol plant.
The selection of the exact pressure to run the methanol synthesis loop will depend on an OPEX/CAPEX optimization. For medium-size units of, for instance, 600 mt/d, the loop would operate at about 700 psig, i.e., without any intermediate syngas compression. For a large unit, of say 2,000 mt/d, the pressure would be somewhat higher. For a smaller unit, the clean gas from the CO2 removal unit is fed to the suction side of the loop gas circulator, while the CO shift bypass gas has sufficient pressure to enter the loop on the discharge side. For the larger plant, the gases are mixed together at the suction of the booster compressor. Rectisol is the most advantageous acid gas removal system in this case. It is the only wash that will achieve the desired degree of desulfurization (0.1 ppm total sulfur). Alternative systems would require an additional stage of COS hydrogenation and subsequent zinc oxide bed for final cleanup. In the case of a methanol plant, Rectisol has the added advantage that the wash liquor is the plant product itself, thus enabling some saving of infrastructure. It is worth pointing out that the use of an optimized synthesis gas quality also influences the choice of synthesis technology. Methanol formation from CO has a significantly higher heat of reaction than that from CO2. The higher proportion of methanol produced from CO, when using syngas from coal or oil gasification, means that in such a plant more attention must be paid to the issue of heat removal than, for instance, in a steam reformer-based plant. It is necessary not only to remove the larger quantity of heat compared with a natural gas-based unit, but also to perform this in a manner that prevents the slightest local overheating to avoid byproduct formation, since the production of impurities from side reactions increases
464 Combustion Engineering Issues for Solid Fuel Systems
with increasing temperature. The intense and intimate cooling provided by the boiling water in an isothermal reactor has, therefore, made it the preferred reactor system for gasifier-based methanol plants. Over 90% of gasification-based methanol production operates with isothermal reactors. This includes all the large-capacity units.
11.8 Benefits and Limits of Gasification 11.8.1 Efficiency One of the motivations for development of the IGCC power plant was to harness the high efficiency of combined cycle gas turbines for use with coal. When one looks at heat rates or efficiencies for IGCCs, it is always important to understand the basis for the numbers quoted. Generally, they are based on higher heating value (HHV) of gasifier feed and net station output (after deducting all parasitic loads). European numbers may, however, be based on the lower heating value (LHV). On this basis, efficiency numbers appear higher and heat rates lower. The prototype 100 MWe Cool Water IGCC run between 1984 and 1988 operated with a heat rate of 10,950 HHV Btu/kWh net [19]. Wabash, representative of the 250 MWe class of IGCC built in the mid-1990s and based on the GE 7FA gas turbine, had a heat rate of 8,900 HHV Btu/kWh net [6]. The 630 MWe class currently in planning is based two larger GE 7FB or Siemens SGT6-5000F gas turbines and has a heat rate of about 8,500 HHV Btu/kWh net on bituminous coals. Dry-feed gasifiers achieve a similar heat rate, when operating on high-moisture subbituminous coals such as those from the Powder River Basin (PRB). The heat rate of a slurry-feed gasifier operating with PRB coals increases to 9,000 HHV Btu/kWh net or more.
11.8.2 Environmental Impact Another motivation for development of the IGCC power plant was its potential for extremely low environmental emission rates compared with other coal-based technologies. The first generation IGCC plants built during the 1990s have achieved the goals set at the time, which are discussed under the individual pollutants in the following sections. The technology to reduce these rates further is available, though its use will depend on a balance between cost and regulatory requirements. At the time of writing, there is still a need to clarify some of the permitting issues—particularly in respect to NOx simply because of the lack of precedents. On the one hand, there is the view that an IGCC is a coal-fired plant and should be permitted as such. On the other hand, others see the main emissions source as being the gas turbine and seek to apply the same regulations as natural gas, irrespective of the difference in fuel characteristics.
Gasification
465
11.8.2.1 Sulfur Emissions An IGCC can readily reduce sulfur emissions to about 4 ppm SO2 (wet basis, referred to 1% O2) in the turbine exhaust. This is equivalent to about 30 ppm total sulfur (H2S þ COS) in the dry, undiluted gas leaving the acid gas removal, a value achievable with an MDEA or Selexol system. If an SCR is required for Denox, then this would typically need to be reduced to about half this value, which would also be possible with a rather more elaborate version of Selexol. A further two orders of magnitude reduction would be possible using Rectisol as currently used for chemical or synthetic fuel applications. This is, however, considered to be unnecessarily expensive. 11.8.2.2 NOx Emissions The high hydrogen content of syngas (which would be even higher in the carbon capture scenario) prohibits the use of current dry low NOx burners as developed for natural gas. Instead, diffusion burners are used. Typically, one could achieve about 15 ppm (dry basis referred to 15% O2) in the exhaust of a gas turbine without SCR. In fact, some IGCCs such as the plant in Buggenum regularly achieve values under 10 ppmv without SCR. Some areas may, however, require SCR to achieve lower NOx emission rates of 3 ppmv (dry basis referred to 15% O2) comparable with those for natural gas-fired turbines. In such a case, it is necessary to desulfurize the fuel gas further to about 15 ppmv so as to avoid formation of ammonium bisulfate from ammonia slip in the SCR, which can deposit on heat exchange surfaces downstream the SCR. The SCR in the oil-fired IGCC at Negishi in Japan is reported as meeting its permit level of 2.6 ppmv (dry basis referred to 15% O2) [16]. 11.8.2.3 Mercury Mercury can be removed from the fuel gas with a fixed bed of activated carbon. Approximately 95% of the mercury leaving the gasifier in the fuel gas is captured. 11.8.2.4 Other Emissions Typical values for other pollutants referred to the gross heat input to the gasifier, based on a heat rate of 8,500 Btu (HHV)/kW (net), are as follows:
Particulate matter: 0.0145 lb/million Btu (including condensables) CO: 10–25 ppm (dry basis referred to 15% O2) Unburned hydrocarbons: 7 ppmv (wet basis) VOC: 1.4 ppmv (wet basis)
11.8.2.5 Start-up Emissions During start-up of an IGCC, there are short-term emissions, caused by flaring un- or partly desulfurized gas until the acid gas removal and sulfur
466 Combustion Engineering Issues for Solid Fuel Systems
recovery units are operating stably or operating the gas turbine on start-up fuel without SCR. These can add considerably to the annual emissions, depending on the number of start-ups. Nonetheless, the level of continuous emissions is so low that even when adding these start-up emissions, the totalized annual emissions are still an order of magnitude lower than those using conventional technology, particularly when using SCR.
11.8.3 Availability It is generally acknowledged that while early IGCC plants met their efficiency and environmental goals, the availability results were not good. This was surprising also to those within the industry accustomed to high availabilities such as those achieved by, say, Eastman at its Kingsport methanol plant, where 98% is regularly reported. An analysis of the causes of outage revealed some unexpected results [20]. Much of the lack of availability was due to fleet issues on early models of the gas turbines involved (in some cases up to 25% loss of annual availability), which had no relation with their utilization in an IGCC environment. This is contrasted with three refinery-based IGCC units built in Italy about 5 years later, which after a 2–3 year ramp-up period are reporting availabilities (and on-stream times) of 90–95%. One of these plants achieved over 90% availability in its second year of operation.
11.8.4 Capital Requirements A discussion of capital costs of IGCC with numerical values is difficult in a period of rapid inflation in the capital plant industry. It is, however, generally accepted that current IGCCs require between 10% and 20% additional investment when compared with conventional technology. Existing IGCCs are all tailor-made units with all the associated high costs of engineering, procurement, and construction. Vendors of IGCC systems have recognized the necessity of reducing costs through modularization and standardization and are all preparing so-called reference designs for two-train 630 MW net output plants. One vendor has estimated that current efforts could reduce the cost premium to about 10% [21]. Introduction of syngas operation to the next generation of gas turbines (H-class), which have now started operation with natural gas, will also reduce the specific costs per kW installed capacity by extracting a higher power output from the same gas production facility. It should also be noted that CO2 capture from the high-pressure fuel gas is much less costly from flue gas and that, in the event of carbon capture becoming a necessity, the overall cost of electricity is expected to be lower with IGCC than with conventional technologies and post-combustion capture.
Gasification
467
11.9 References 1. Gumz, W., 1950. Gas Producers and Blast Furnaces. New York: John Wiley & Sons. 2. Higman, C., and M. van der Burgt. 2003. Gasification. Amsterdam: Gulf Professional Publishers. 3. Rogers, L.H., et al. 2005. Power from PRB—Four Conceptual IGCC Plant Designs Using the Transport Gasifier. 22nd Annual International Pittsburgh Coal Conference. September 13. 4. Zuideveld, P.L. 2005. Shell Coal Gasification Process Using Low Rank Coal. Gasification Technologies Conference, San Francisco. October 9–12. 5. De Kler, R. 2007. “Nuon’s Magnum Plant: A Step Towards Sustainability.” Modern Power Systems. April. pp. 24–28. 6. Wabash River Energy Ltd. 2000. Wabash River Coal Gasification Repowering Project: Final Technical Report. Wabash, IN: Author. 7. Klemmer, K.D. 2006. The Siemens Gasification Process and Its Application in the Chinese Market. Gasification Technologies Conference, Washington. October 1–4. 8. Schingnitz, M. 2003. “Gasification: An Opportunity to Design Environmentally Compatible Processes in the Chemical and Pulp & Paper Industry.” Chem. Eng. Technol. 26:9, Weinheim: Wiley-VCH Verlag. 9. Pinkston, T., and F. Morton. 2006. Orlando Gasification Project: Demonstration of a 285 MW Coal-Based Transport Gasifier. International Technical Conference on Coal Utilization & Fuel Systems, Clearwater, FL. May 21–25. 10. Sharp, C.R., et al. 2002. Recent Selexol and Membrane/PSA Operating Experiences with Gasification for Power and Hydrogen. Gasification Technologies Conference, San Francisco. October. 11. Kubek, D.J., E. Polla, and F.P. Wilcher. 1997. Purification and Recovery Options for Gasification. IChemE Conference “Gasification Technology in Practice.” Milan. February. 12. Supp, E. 1990. How to Produce Methanol from Coal. Berlin: Springer. 13. Koss, U., M. Meyer, and H. Schlichting. 2002. Zero Emissions IGCC with RectisolW Technology. Gasification Technologies Conference, San Francisco. October. 14. de Graaf, J.D., P.L. Zuiderveld, F.G. van Dongen, and H. Ho¨lscher. 2000. Shell Pernis Netherlands Refinery Residue Gasification Project. IChemE Conference “Gasification for the Future.” Noordwijk. April. 15. BASF Technical leaflet, BASF Catalyst K 8-11 (undated). 16. Yamaguchi, M. 2004. First Year of Operational Experience with the Negishi IGCC. Gasification Technologies Conference, Washington. October 3–6. 17. Barkley, N.E. 2006. Petroleum Coke Gasification Based Ammonia Plant. AIChE Ammonia Safety Conference, Vancouver. September 10–14. 18. Jones, R.M., et al. 2006. Gas Turbine Requirements for a Carbon Constrained Environment. IChemE European Gasification Conference, Barcelona. April 25–27.
468 Combustion Engineering Issues for Solid Fuel Systems 19. Electric Power Research Institute (EPRI). 1990. Cool Water Coal Gasification Program: Final Report. Palo Alto, CA: EPRI. 20. Electric Power Research Institute (EPRI). 2007. Integrated Gasification Combined Cycle (IGCC) Design Considerations for High Availability, Vol. 1: Lessons from Existing Operations. Product No. 1012226. Palo Alto, CA: EPRI. 21. Lowe, E., 2006. “IGCC Reference Plant Inked.” World-Generation, May/June. pp. 1, 9, 16.
CHAPTER
12
Policy Considerations for Combustion Engineering Melanie McCoy General Manager Wyandotte Municipal Utilities
12.1 Introduction Engineers do not make policies, but they implement them. Policies created by others drive many engineering decisions. Policies that affect combustion engineers come from a variety of agencies and organizations: federal, state, and local governmental laws, ordinances, and regulations; corporate policies driven either by stockholders or by senior executives; and policies of clients. Policies are influenced by a wide variety of organizations within and outside government. In addition to the policy-making organizations themselves, policies are influenced by such groups as the Sierra Club, Natural Resources Defense Council, other environmental groups, U.S. Chamber of Commerce, Edison Electric Institute, labor unions, think tanks (e.g., the Heritage Foundation), research organizations including the National Academy of Sciences and the Electric Power Research Institute, and many others. Policy is made by addressing the conflicts between these interest groups—sometimes achieving full compromise and sometimes favoring the politically stronger interest groups. Most governmental policies are initiated by elected officials, and then the governmental agencies are charged with developing the rules and enforcing them. Elected officials typically provide broad perspectives and directives in legislation; the administrative agencies translate these into specific actions to take, or not to take, and into specific regulatory limits. Policy is then ratified, or modified, through the judicial system. 469
470 Combustion Engineering Issues for Solid Fuel Systems
Most commonly, policy is thought of in terms of environmental policy that dominates the news and many of the engineering design and operations issues. However, policies influenced by governmental directives, let alone by ownership or client programs, influence such issues as fuel selection; worker safety in design, construction, and operations; system efficiency (heat rate); product marketing; ownership patterns; and a host of other issues. Table 12-1 highlights some of the federal, state, and local policy actions in recent years that have influenced solid fuel utilization and system design. A detailed review of all these examples is beyond the scope of this chapter. Focus will be given to environmental issues with some comment on other policy issues. Further, the chapter will focus on coal issues, since coal is the dominant solid fuel; some comment will be given to biomass and the renewable fuels, along with petroleum coke and waste fuels.
TABLE 12-1 Examples of Governmental Actions Influencing Power Plant Design and Solid Fuel Systems Area of Consideration
Federal Policies and Actions
State and Local Policies
Fuel Selection
Permits for DM&E to enter
Renewable portfolio standards
a
Worker Health and Safety Worker Hiring
Product Marketing Environmental Protection
Ownership a
the Powder River Basin Research programs for coal and fuel utilization Research programs for improved coal mining Deregulation of interstate natural gas OSHA Mine Safety (MSA) 1964 Civil Rights Act and Successive Acts Davis-Bacon Act—Public Projects Public Utilities Regulatory Policies Act Electricity Deregulation Clean Air Act and Amendments Clean Water Act Resource Conservation and Recovery Act Coal Mine Land Reclamation Acts Anti-Trust Laws
Regional train purchased by Canadian Pacific.
promoting green power (biomass use)
Subsidies to biomass use
Comparable state programs
State Deregulation of Utilities State Implementation Acts and Programs
Land Use (e.g., zoning) acts Permitting programs for new power plants, transmission lines, etc.
Policy Considerations for Combustion Engineering 471
12.1.1 Combustion Engineers Do Not Make Policy Combustion engineers do not make policy. Rather, they respond to policies of governmental agencies, owners (public or private), and clients. Their response takes the form of system designs to conform to all laws, ordinances, and regulations of federal, state and local governments. Their responses also take the form of conformance to client specifications and to their own company standards. With respect to the latter, redundancy for reliability becomes a concern. For some engineers, this becomes “a pair and a spare” or “belt and suspenders” thinking; for others, responding to investment and financial priorities, different approaches to reliability may be employed. Because of their role as inevitable instruments implementing policy, engineers must work within the given system. Within the energy arena, engineers work within the constraints of the flow of energy in the U.S. society, reflected in Figures 12-1 through 12-3 [1]. As these figures indicate, large quantities of coal are utilized in the energy sector, and hence, solid fuel systems are designed such that coal can be transported and handled in bulk—using unit trains, barges, etc. The coal system is no longer readily capable of delivering single railroad cars to users—except in special cases. Unit trains are the norm. Engineers must respond within that system, a system created as a consequence of previous public and private policies.
Petroleum 2.79 Coal 23.79
Exports 4.93 Other 2.14
Natural Gas 19.02
Fossil Fuels 56.03
Crude Oil 10.87
Coal 22.51 Domestic Production 71.03
NGPL 2.35
Natural Gas 22.43 Supply 104.80
Commercial 18.00 Fossil Fuels 84.76
Nuclear Electric Power 8.21
Consumption 99.87
Industrial 32.43
Petroleum 39.76
Renewable Energy 6.79
m leu tro 3 Pe 29.0
l entia Resid 5 21.0
Imports 34.49
Nuclear Electric Power 8.21 Renewable Energy 6.84
Transportation 28.40
Adjustments 0.72
Other 5.476
FIGURE 12-1 The flow of fuels in the U.S. economy (quadrillion Btu), 2006 (Source: [1]).
472 Combustion Engineering Issues for Solid Fuel Systems
Coal 20.66
Fossil Fuels 28.60
Natural Gas 7.07
Energy Consumed To Generate Electricity 41.27
Conversion Losses 26.71
Plant Use 0.73 T & D Losses 1.31
Petroleum 0.69 Other Gasses 0.18
Gross Generation of Electricity 14.56
Nuclear Electric Power 8.21
Net Generation of Electricity 13.83
Renewable Energy 4.28
Other 0.18
End Use 13.03
Residential 4.62
Retail Sales Com mercial 4. 44 12.51
Indus trial 3 .42
Unaccounted for 0.45
Net Imports of Electricity 0.06
Direct Use 0.53
Transportation 0.03
FIGURE 12-2 The flow of fuels to electricity in the U.S. economy (quadrillion Btu), 2006 (Source: [1]).
Exports 49.6
Surface 803.4
Bituminous Coal 564.6
Production 1,161.4
Residential 0.4
Consumption 1,114.2
Commercial 3.8
Industrial 83.5
Electric Power 1,026.5
Subbituminous Underground Coal 511.1 358.1 Lignite 84.2 Anthracite 1.5 Waste Imports Coal 36.2 Supplied 13.6
Losses Stock and Change 39.9 Unaccounted for 7.6
FIGURE 12-3 The flow of coal in the U.S. economy (million tons), 2006 (Source: [1]).
12.1.2 Combustion Engineers Respond to Policy Combustion engineers, working within their discipline, must keep their installations—and designs—within the constraints of policies that have been translated into laws, ordinances, and regulations. At the same time, they must meet the corporate and client specifications and standards. Meeting such constraints can have far-reaching effects. As an example, the National Environmental Policy Act (NEPA) was passed in 1970; this
Policy Considerations for Combustion Engineering 473
was followed by the creation of the Environmental Protection Agency in December of that year. NEPA called for a significant reduction in emissions of sulfur dioxide (SO2). Combustion engineers responded by adapting existing power plants—and designing new power plants—to burn lowsulfur subbituminous coals largely from the Powder River Basin (PRB). The consequences of responding to this policy included an increasing use of PRB coal—to the point that about as much of this fuel is burned as bituminous coal, when measured on a tonnage basis (see Figure 12-4). Engineers did not develop the policy, but their response caused a significant shift in the energy business.
12.2 Environmental Policy and the Engineering Response The Environmental Protection Agency (EPA) is responsible for regulating and controlling pollutants. The EPA will set the policy and administer, while individual states will be responsible to develop their own programs to comply with the federal policy. In the absence of federal action, states may, on occasion, take the lead and develop their own programs. Depending on the state agenda, these programs may be more onerous than the federal requirements, but they cannot be less. Occasionally, bordering states may influence other states’ policies. For instance, the eastern states have 1400
Coal Production (million tons/y)
1200
Bituminous coal Subbituminous coal Lignite Anthracite Total US Coal Production
1000
800
600
400
200
0 1948
1958
1968
1978 Year
1988
1998
2008
FIGURE 12-4 U.S. coal production by type. Note the rise in subbituminous coal production since 1970 (Source: [1]).
474 Combustion Engineering Issues for Solid Fuel Systems
filed suits against the Midwest for air pollution that is carried eastward from the Midwest’s coal-fired plants. Another example of different state rules would be the ozone reduction plan that was designed to reduce NOx in the eastern states.
12.2.1 A Historical Perspective The history of environmental events and legislation, and the engineering response to that legislation in the solid fuels combustion community, is well documented. Although water pollution was first addressed by federal legislation—dating back to 1948—air pollution and solid waste legislation are the best examples of policy legislation and the engineering response. The year 1970 was the watershed year with the passage of the National Environmental Policy Act (NEPA), the passage of the Clean Air Act, and the creation of the Environmental Protection Agency. From this watershed year, under President Richard Nixon, regulation of particulates, sulfur dioxide (SO2), oxides of nitrogen (NOx), and other emissions became a first order of priority for combustion engineers. Subsequent actions governing the combustion of alternative and waste fuels included the creation of the Boiler and Industrial Furnace (BIF) regulations applied to the use of hazardous wastes in the power and industrial communities (e.g., cement kilns, mineral processing kilns, pulp mills, and other industries). The 1970 legislation was strengthened over subsequent years. It led the combustion community to develop advanced post-combustion control technologies beyond the initial electrostatic precipitators (ESPs) and baghouses designed and installed in the early and middle 20th century installations. The consequence of this increasingly stringent environmental policy has been a continuous reduction in power plant emissions despite increasing population, electricity consumption, coal and solid fuel consumption to fuel electricity generation, and overall economic activity. This growth of coal consumption is shown in Figure 12-4.
12.2.2 Environmental Policy and Legislation Since 1990 Environmental regulations have been developing for decades, but a major impact was the Clean Air Act Amendments of 1990, which established the first “Cap and Trade” program. Title IV of the Clean Air Act Amendments set a goal of reducing annual SO2 emissions by 10 million tons below 1980 levels. To be able to comply with this, a facility would need to install the equipment or technology to reduce the emissions, switch fuel to comply, or purchase allowances. The cap and trade program allows compliance flexibility. Allowances are allocated based on historic fuel consumptions and specific emission rates prior to the start of the program, and each unit holds allowances for 30 years. The program also allows units to bank allowances with early reductions and then use them for later years.
Policy Considerations for Combustion Engineering
475
The purpose of the cap and trade program was to allow units that were more economical to control or over-control and trade their allowance to units that were more technically challenging and costly. Since most air pollutants are a global or regional issue, this is effective in reducing the overall impact on the environment. The Clean Air Act Amendments also called for reducing NOx emissions 2 million tons by 2000. To comply, the EPA finalized the NOx State Implementation Plan in 1998. Compliance with the regulation requires both fuel and control equipment modifications. This allows the most economical solution to be developed depending on the unit configuration, site location, equipment limitations, and economic impact. This is the exciting challenge where a combustion engineer must develop the best solution. To make sure the best economic solution is put in place, a company compares installation costs, together with the operating costs, to the allowance purchase costs. Planning the future, based on free markets with emission allowances, will need to be a risk debated in the company. The regulations may require absolute compliance with a set limit, or compliance through the cap and trade system, or a combination of both. Each unit has a Renewable Operating Permit (ROP), which defines the set limits for each pollutant. If a new unit is under construction, the New Source Review Regulations (NSR) will require the best available control technology (BACT) to be installed. In addition, if a unit undergoes a major modification, the NSR requirements may apply. The NSR was developed to allow older plants to continue to operate, or be “grandfathered,” and not require that the new lower emission limits apply. This would be an economical hardship, and many plants would be decommissioned, requiring massive new plant construction. However, if the plan is modified above a certain level of investment, the EPA may require BACT. As with most of the environmental laws, the impact on the plant is not predictable. Each law leaves room for interpretation and, depending on the company’s appetite for risk, will determine the plan. As with NSR today, the cases are still in court, and each one is being ruled on differently. An example of mixed directives is the NSR suit that First Energy received where new low-NOx burners were installed and listed as one item that triggered the NSR suit. Another example may be when a superheater needs to be replaced, and a better design is available. The better design may allow more efficient operation of the boiler and lower emissions, yet it could be interpreted to trigger NSR. In addition to a company’s appetite for risk, there is also the corporate reputation and branding that can impact the plant. For a company to remain a good corporate citizen and community representative, voluntary initiatives like “Clean Corporate Citizen” and “Energy Start Certified” are available. Each of the initiatives has an impact on operation and maintenance. The culture of the plant can be changed and improved with these, if they are
476 Combustion Engineering Issues for Solid Fuel Systems
instituted in the correct frame of mind. For instance, the housekeeping required in an Environmental Management System can have exponential benefits on employee morale if it is not presented in a punitive way. The impact of the 1990 legislation can readily be seen in Figure 12-5, showing the decrease in SO2 and NOx emissions since 1988. Engineers have responded with fuel switching, more efficient generation, and with post-combustion controls such as scrubbers for SO2 control and selective catalytic reactors (SCR) for NOx control. Particulate emissions also have been reduced dramatically. This legislation spurred the growth in use of subbituminous coal, largely from the Powder River Basin and other low-sulfur western coal deposits, as shown in Figure 12-4. Today PRB subbituminous coal is used almost as extensively as the various deposits of bituminous coals. In addition to fuel switching, this legislation has played a significant role in the development and deployment of fluidized-bed combustion technology. This technology, described in Chapter 8, features both fuel flexibility and inherent control of both SO2 and NOx. It also employs mechanisms that control other emissions including hydrogen chloride (HCl) and related products of combustion from heteroatoms. Beyond the support for fluidized-bed technology, this body of legislation spurred the development of other NOx control techniques: low-NOx 18 SO2 emissions from coal Airborne Emissions (million tons/y)
16
NOx emissions from coal
14 12 10 8 6 4 2 0 1988
1990
1992
1994
1996 1998 Year
2000
2002
2004
2006
FIGURE 12-5 Annual airborne emissions from coal combustion: 1988–2005 (Source: [1]).
Policy Considerations for Combustion Engineering 477
burners, reburn technology, selective noncatalytic NOx reduction, selective catalytic NOx reduction, and others. Technologies developed to enhance acid gas control included both wet and dry scrubbers, spray dryer-absorber technologies, and the use of alternative reagents such as trona. All these have been discussed in previous chapters; all are the consequence of environmental policy. Environmental policy will continue to evolve, seeking more stringent controls of more compounds now treated as emissions. These include trace metals and other hazardous air pollutants (HAPs) including (not exhaustive) mercury, beryllium, lead, HCl, hydrogen fluoride (HF), and condensable organics considered fine particulate (e.g., PM2.5). Due to considerations of the potential for global warming, carbon dioxide (CO2), methane (CH4), and other “greenhouse gas” compounds are now being pursued if not totally regulated.
12.2.3 Mechanisms of Engineering Response to Environmental Policy To remain in compliance with the evolving policies, plants will continue to modify operations. New equipment may be required to comply with EPA rules. Physical modifications to the plant may require permit changes. To remain current on the changes, many companies work collaboratively with each other to deal with the rulemaking and compliance. In addition there are organizations to support specific aspects such as pollution control equipment, IT technology, or safety updates. The U.S. Department of Energy (DOE) is one organization that will offer solicitations to provide co-funding of projects. This allows one plant to install a technology and other companies to contribute and share in the learning of technology. In addition, DOE funds are available to assist in projects that are considered developmental. One example of this cooperation is the mercury control technology of activated carbon injection. This technology was installed as the TOXECON Project, at the Presque Isle Power Plant of WE Energies, a plant in Northern Michigan. The project was installed with support from DOE and participation by the Electric Power Research Institute (EPRI) [3]. As the plant experienced success and failures, the information was shared with the industry. The instrumentation to monitor the mercury emissions out the stack was also new to the industry. The measurement of such small amounts required new instruments to be developed. As several companies tried to get the equipment to market, each one seemed to experience some startup challenges. In addition, the need for the instruments on 1,300 units in a short time created a large logistical problem.
478 Combustion Engineering Issues for Solid Fuel Systems
Collaboration and governmental support will need to only increase in the coming years. To support the federal actions, the DOE will continue to initiate research and development programs to develop or improve technologies. The results of this research will allow the EPA rulemaking and assist in quantifying the environmental impacts or benefits. Environmental regulations will continue to evolve and reduce emissions. As each new administration sets emissions limits, a ratchet effect will require each regulated pollutant to achieve lower emissions. These limits may be legislated or regulated. On some occasions the bureaucracy cannot agree on the limits, and no law is passed. The rule is then issued, plants must comply, but there is no guarantee the limits will remain when a new administration takes control. These increased regulations will proceed even as the production and consumption of coal, petroleum coke, biomass, and the array of solid fuels increase. Electricity production and consumption projections shown in Figure 12-6 and coal demand projections shown in Figure 12-7 document the confluence of environmental regulations and energy needs that will require continuing collaboration and continuing engineering response to governmental and private policies in the energy arena. The environmental issue of global warming—and increased concentrations of CO2 in the atmosphere—also presents challenges for combustion engineers. McKibben [4] documents the facts: Before the industrial revolution CO2 concentrations in the atmosphere were at 280 ppm. By the 1950s, that concentration reached 315 ppm. Today, it is 380 ppm and increasing by some 2 ppm/y. There have been numerous debates concerning “Is it real? Is it significant?” There have been additional debates concerning the best way to approach the management of the issue.
2,500
Projections
History
Commercial Residential
Quadrillion Btu
2,000
1,500 Industrial 1,000 500 0 1980
1995
2005 Year
2015
2030
FIGURE 12-6 U.S. electricity growth projections (Source: [2]).
Policy Considerations for Combustion Engineering 479 40
History
Projections Total
Quadrillion Btu
30
20
West
10
Appalachia Interior
0 1970
1980
1990
2000 Year
2010
2020
2030
FIGURE 12-7 Projected growth in U.S. coal utilization (Source: [2]).
These debates have led to challenges to combustion engineers including the use of integrated gasification-combined cycle plants to improve generation efficiency and produce a concentrated stream of CO2, and also the use of oxygen-based combustion to produce a concentrated stream of CO2 for sequestration. These approaches are the bases of additional cooperation between industry and government. One approach to greenhouse gas that has been promoted is additional use of “green energy”—renewable and sustainable energy including biomass, wind, solar, and a variety of other energy forms. Given the policy approaches to this arena, however, renewable energy is sometimes considered as green energy—the new pork. As the consensus shifts to the need for clean coal technology, the economics of various fuels is impacted by the emissions in addition to actual fuel cost. The technology required to make coal “clean” needs to address the NOx, SO2, Hg, and CO2. Renewable fuels are becoming more economic, and with government subsidies to support the technology, some will emerge as true opportunity fuels (see Chapter 3). The cofiring of fuels to obtain the renewable credits will be a challenge for the fuel-feed systems in addition to the ash and combustion. A fluidized bed-boiler has the capability to cofire tire-derived fuel, petroleum coke, wood waste, and sewage sludge. With additional provisions, cofiring of selected biomass fuels can also be accomplished in cyclone and pulverized coal boilers. If the permitting authority in the state is agreeable, and the opportunity fuels in the area can be delivered at the right price, it may be a great time to modify the boiler to make it happen. Such modifications were pioneered by EPRI and DOE in the latter years of the 20th century and beginning of the 21st century, with demonstrations at
480 Combustion Engineering Issues for Solid Fuel Systems
FIGURE 12-8 The installation for cofiring sawdust with coal at the Albright generating station. Sawdust is separately injected into a tangentially fired pulverized coal boiler. [Photo by David Tillman]
(not exhaustive) Plant Gadsden (Southern Company), Albright and Willow Island Generating Stations (Allegheny Energy Supply Co., LLC), Seward Generating Station (GPU Genco—now Reliant Energy), Bailly and Michigan City Generating Stations of NiSources, and numerous generating stations of the Tennessee Valley Authority. Again, this was an example of private industry–government cooperation. Figure 12-8 depicts the cofiring installation at Albright Generating Station, a cofiring demonstration still in operation.
12.3 Energy Policy and Combustion Engineering Energy policy has long been debated, and established, at the federal level. Starting with the Project Energy Independence of the Nixon administration and continuing to the present, each administration attempts in earnest to improve the industry, but the unintended consequences can have lasting impacts—both good and bad. Energy policy is established both by explicit acts and by budgetary allocations to DOE.
Policy Considerations for Combustion Engineering
481
12.3.1 Energy Policy and Fuel Selection National energy policy can drive the fuel selection or switching. The Carter administration’s Fuel Use Act, which has since been repealed, caused baseload plants to switch back to coal. That act prohibited the use of oil or natural gas in generating stations for more than 1500 h/y except in specific circumstances. The engineering response was a significant push for new coal-fired generating capacity that was installed in the period 1976–1983. More recently, federal energy policy has focused on using market forces to determine fuel selection. This approach, favored by all presidents from Ronald Reagan to the present, also has significant consequences. Under this approach, the economy can drive the fuel selection, as oil has been priced out of any consideration in today’s market. The Public Utilities Regulatory Policies Act (PURPA)—a key element in President Carter’s energy policy—allowed the formation of independent power producers (IPPs) to enter the industry. As nonutility generators, they could produce power, and the regional utility would be required to purchase the energy.
12.3.2 Deregulation and Its Precursors Some states encouraged the building boom of these smaller size plants by legislating a favorable rate for the sale of electricity, such as New York, and the 6 cent rule. The IPPs in New York were guaranteed 6 cents per kilowatt for all the power they generated. California established a similarly favorable climate with its Standard Offer #4, leveling the price of electricity and favoring rapid payback of investments. Other states created a more hostile environment. These solid fuel IPPs were ideal for fuel flexibility because most were smaller in size. They could also be built near the mines to take advantage of fuel price. Fuels burned in these IPP plants included anthracite culm, bituminous gob, wood waste, and a host of other opportunity fuels as well as conventional fossil fuels. The disadvantage was that the maintenance, operation, and engineering required to get these smaller, sometimes unique plants up and operating was a challenge for small plant staffs. Deregulation of electricity generation and sale followed the spirit of the PURPA legislation, and was encouraged by both Democrats and Republicans. It came into being through President Clinton’s administration. It involves both federal actions and state actions; it has been applied inconsistently across the United States. In preparation for deregulation, many plants installed new equipment in order to be ready when competition arrived. Other plants were limited in resources, and attention was given to the wave of combustion turbines and combined cycle plants. After the price of gas escalated and the combustion turbine bubble burst, coal plants once again came into favor as the most economic approach.
482 Combustion Engineering Issues for Solid Fuel Systems
12.3.3 Energy Efficiency and Energy Policy Energy efficiency, or the efficiency of generating electricity, has long been the province of private industry energy policy. The use of increasing steam pressures and temperatures and increasing boiler capacities to achieve efficiencies was discussed in Chapter 1. These gains have been encouraged by investment policy and market forces. Some federal actions encouraging efficiency have been voluntary programs. The energy savings through efficiency projects in Energy Star programs can inspire pride of ownership and even a competitive spirit to improve. In the past, many power plants did not see a need to reduce their internal auxiliary power. In many cases, the fuel costs are passed on to the consumers. The efficiency gained would not benefit the organization. However, as more states implement an efficiency program regulation, such as the Michigan 21st Century Energy plan, the energy efficiency is required. As Michigan determined, the future energy needs will be met by a combination of energy efficiency, renewable energy, and new base load plants. Energy efficiency has also been encouraged by federal investments in cooperative research and development programs. Such programs have led to the deployment of IGCC technology used to increase the efficiency of electricity generation while reducing emissions and, potentially, providing a concentrated stream of CO2 for capture and sequestration. The Clean Coal Technology program of DOE, along with associated programs, has led the way in promoting this technology. Polk County Generating Station, a 250 MW IGCC shown in Figure 12-9, is an example of the DOE working with industry to advance energy efficiency while reducing airborne emissions. Where the DOE budget allocates funds is a strong policy push that industry responds to.
12.4 Other Federal, State, Local, and Private Policies Impacting Combustion Engineers In addition to the environmental and energy policies, health and safety regulations have been developed to protect humans from harmful impacts of pollutants or equipment. These are administered and enforced by the Occupational Safety and Health Administration (OSHA). Examples of laws that changed the life of all power plant personnel are the Confined Space rules. These rules required either the re-engineering of many areas that require access by plants on a routine basis, or new safety practices and policies. Labor laws are developed as a result of evolving legal rulings and can affect the operations of the plants. State laws set limits on the number of hours a person can work. With the limits, a maintenance outage or equipment startup schedule may need to be changed to accommodate a staff schedule. This is especially difficult if the plant staff does not have the depth and multiple persons trained as backup.
Policy Considerations for Combustion Engineering
483
FIGURE 12-9 The Polk County IGCC. This installation was one of the early IGCC projects jointly funded by industry and the federal government to enhance electricity generation efficiency and reduce the release of airborne emissions. [Photo by David Tillman]
Insurance requirements at the plant may require new equipment or modifications in addition to inspections. As the insurance industry responds to claims, their policies adapt and change. An example is vibration monitoring or oil analysis. In the past, the preventive maintenance activities would be considered good practice. Now, these techniques are required, and without them, insurance may be denied. Insurance companies provided safety provisions and policies long before the creation of OSHA. Boiler safety inspections were regularly scheduled as a condition of being insured. Similarly, all worksite conditions were inspected to ensure that certain standards and practices were met. Local organizations or influence groups may change policies based on sentiment or societal issues. Organizations continue to change and processes within the companies may influence operations, budgeting, or hiring and staffing policies. A regional environmental group may influence the air or water effluents, even above the laws. The increasing awareness of diversity has changed staffing and hiring practices. The need for engineers and
484 Combustion Engineering Issues for Solid Fuel Systems
the importance of retention of trained engineers for an organization have given rise to the mentoring programs.
12.5 Conclusions Numerous organizations from the federal government to private industry make a vast array of policies that impact the practice of combustion engineering for solid fuels. These impacts range from meeting environmental requirements to improving boiler and kiln efficiency to health and safety concerns, labor practices, and a host of other issues. Energy engineers respond to these policies by developing programs in cooperation with the federal government and private organizations (e.g., EPRI), by adjusting designs, and by implementing technologies consistent with the ever-changing policies. Policy uncertainty is one primary problem for combustion engineers and their corporations. Such uncertainty leads to difficult design and operations decisions—and to delays and cost increases in projects. Combustion engineers do not make policy. However, through these mechanisms, they have responded, and will continue to respond, to policy pressures from all forces.
12.6 References 1. U.S. Energy Information Administration. 2007. Annual Energy Review 2006. Washington, D.C: U.S. Department of Energy. 2. U.S. Energy Information Administration. 2007. Annual Energy Outlook 2007 with Projections to 2030. Washington, D.C: U.S. Department of Energy. 3. WE Energies. 2006. TOXECON Project: Wisconsin Energy Presque Isle Power Plant. Presentation May 18, 2006. 4. McKibben, B. 2007. Carbon’s New Math. National Geographic. 212(4): 32–36.
Index A Absorber reaction tank, 368 tower, 368 AccuTrack, 177, 396, 398–399, 402, 406, 408–414, 418, 420–421 Acid gas, 299, 341, 366–367, 369–371, 447–448, 450–453, 455, 458, 460–463, 465, 477 rain, 250, 366 soluble, 122, 138, 303–304, 306–310, 313 Activation energies, 20, 75–76, 187, 189–190 Adiabatic flame temperature, 25 Advanced burner technology, 261 Advanced combustion technology (ACT), 258–261 Advanced materials, 210, 224 AES, 106 Agglomeration, 95–96, 121, 276, 301–302, 314, 427, 432 Agricultural security, 326 Air heater leakage, 225, 232–234, 239 liquide, 262 staging, 297, 299 Airborne emissions, 21, 29, 94–95, 107, 109, 250–252, 423, 476, 482–483 Air-to-cloth ratio (A/C), 362 Air pollution control devices (APCD), 380–382, 384–387 Air separation unit (ASU), 431, 457–459, 461 Albright Generating Station, 106, 264, 480 Algae, 251 Alkali chlorides, 105 metals, 6, 105, 123, 182, 302, 321 sulfates, 303 Alkaline earth, 6, 137–138, 148–150, 157, 159, 182, 284, 302, 310, 312–314, 320–321
Allegheny Energy Supply Co., LLC, 106–107, 480 Allen Fossil Plant, 107 Allen S. King Generating Station, 173 Alliant Energy, 117, 264 Alstom, 259, 263, 324 Aluminosilicates, 136–137, 149, 165, 304, 312–315 Aluminum, 22, 87, 138–143, 149, 308–310, 312, 315–316, 320–321, 456 Amazon, 4 American Society for Testing and Materials (ASTM), 41–42, 61, 68–73, 93, 201–202, 238 Amine absorption, 389 processes, 448 Ammonia, 154, 159, 276, 298, 301, 361, 367, 377–380, 423, 432, 435, 444, 446, 449–450, 452–453, 456, 458–460, 462, 465 Ammonium acetate, 121–122, 138, 303 Ammonium bisulfate (NH4HSO4), 377–380, 465 Ammonium sulfate ((NH4)2SO4), 377–379 Amorphous, 88, 155, 162–163, 166 Angiosperms, 99 Anhydrous ammonia, 361 Animal agriculture industry, 326 fats, 83, 328 proteins, 328 Animal-tissue biomass (ATB), 285, 329–333 Antelope, 67, 175, 185–187 Anthracite, 5, 18, 34, 36, 39, 42, 45–47, 49, 51–53, 55, 57–58, 61–62, 86, 136, 146–148, 203, 215, 222, 285, 301, 328, 472–473, 481 Appalachian Region, 55, 58
485
486 Index Arch-fired, 242, 257, 264–266 Aromaticity, 17, 38, 74, 76–77, 111, 113 Aromatic structures, 7, 15, 35–36, 74, 295 Arsenic, 22–25, 29, 67, 72, 104, 123, 159, 173, 299, 371, 446, 454, 457 Ash behavior, 135, 167, 193, 302 chemistry, 90, 95, 121, 179, 187–193, 276, 301, 346, 356, 361 deposition, 121, 133–134, 136, 153–154, 158, 162–163, 314, 321 deposits, 159, 161–162, 194, 314 elemental analysis, 6, 88, 102–103, 174, 185 formation, 133–134, 136, 151–152, 248, 302 fusion temperature, 41, 53, 72, 91, 102–103, 172, 187, 191–193, 284, 288, 434 mineral analysis, 134, 238, 358, 360 removal, 255–257, 345, 387 softening temperature, 301, 432 Atmospheric FBC, 276, 282, 291 Atomic absorption spectroscopy (AAS), 123 Atomic ratios, 5, 7, 96, 113 Attrition, 214–215, 217, 219, 222, 290 Australia, 3, 35, 46–47, 51–52, 56, 150–151, 172, 321, 324–325 Automatic voltage control (AVC), 347–349
B Babcock power, 261 Babcock & Wilcox, 259, 262, 268–269, 323 Backpressure turbine, 98, 102, 108, 247 Baghouse, 26, 211, 279, 281, 300–301, 342, 361–363, 365–367, 369–372, 384, 387, 474 Bailly Generating Station, 89, 101, 104, 177, 181 Balance-of-plant, 387–388 Ball mill pulverizers, 220–222 Ball-and-race mills, 28 B/A ratio, 23, 162, 185 Barium, 22, 25, 123, 134, 136, 195, 299 Base acid ratio, 22, 165, 174, 179–180, 185–186, 359–360 Bed material, 252, 276–278, 281, 284, 298, 301–302, 304 temperature, 276, 284, 291, 299, 301, 311, 454 volume, 278 Belt conveyors, 200 Beluga field, 172 Beryllium, 23–25, 115, 123, 299, 477 Best available control technology (BACT), 475 Beulah lignite, 6, 16, 18, 21, 67, 77–78 BGL, 428, 432, 443 Biofuel, 251, 318
Biomass, 3, 5–8, 15, 17–18, 21–23, 25–29, 35, 37, 75–76, 85, 96–124, 171, 173, 182, 193, 196, 199–200, 205–206, 212, 214–215, 241–242, 246, 248, 250, 252, 254–255, 261, 264, 280, 284–285, 296, 302–311, 314, 316–321, 325, 328–329, 333, 394, 427, 429–430, 436, 470, 478–479 Bitumen, 172 Bituminous coals, 5, 7–8, 25, 34, 36–37, 41–42, 45–49, 51–52, 55–61, 64–66, 74, 76, 78–79, 112, 141–143, 145–146, 148–151, 172–173, 175–176, 182, 185–187, 191, 203–204, 209, 215, 229–231, 235–237, 285–286, 288–289, 296, 301, 311, 324, 328, 332, 381–383, 385–386, 397, 431, 464, 472–473, 476 Black thunder coal, 67, 75–76, 78, 175 Blast furnace gas, 83, 86 Blending, 27, 30, 107, 111, 171–177, 179, 181–182, 185, 187, 193–196, 224, 237, 248, 302, 321, 367, 402–405, 408, 411, 420 Blue haze, 21 Boat unloader, 176 Boiler efficiency, 94, 96, 107, 207, 212, 245, 267, 288, 379 Boiler and industrial furnace (BIF) regulation, 474 Bottom ash, 23–24, 70, 243, 253, 256–257, 299–300, 314, 358 Bovine spongiform encephalopathy (BSE), 326–327, 329 Bowl mills, 28 Bradford breakers, 217–218 Brazil, 4, 46, 172 Brown coal, 5, 41–42, 51, 54, 284–285, 296 Bubbling fluidized-bed, 94, 277–279, 329, 333 Bulk density, 7, 99, 105, 118, 123, 305 Bunker, 108, 195–196, 208, 257, 400, 402, 404, 406, 408–420, 431 Bus section, 345–346 Byproduct aromatic carboxylic acid (BACA), 173–174
C Cadmium, 22–25, 67, 115, 123, 299 Cage mills, 217–219 Cake, 300, 362–365 Calcinations, 289–291, 293–294 Calcined limestone, 94, 290–291, 374 Calcium, 12, 22–23, 25, 70, 96, 122, 134, 136, 138–143, 150–151, 155, 158–159, 179, 182, 191, 193, 289–291, 302–303, 307–309, 313–314, 323, 368–369, 371, 408
Index Calcium sulfate (CaSO4), 94–95, 149, 154, 156, 159, 166, 290–291, 293–294, 298, 313, 370–373 Calcium sulfite (CaSO3), 371–372, 374 Calorific value, 43, 68, 70, 73, 88–89, 93, 96, 110–112, 115–116, 174, 177 Canada, 3, 35, 46, 61, 92, 172, 262, 311, 321, 324–325 Candle filter, 435, 442, 459 CaO/Fe2O3 ratio, 23, 191–192 Capital costs, 181, 226, 282, 325, 366, 369, 377–378, 466 Cap-and-trade, 383, 388, 474–475 Carbon black, 110–111, 113 capture, 446, 461–462, 465 sequestration, 251, 322, 389, 452, 479, 482 steel, 210 Carbonates, 122, 134, 136–138, 151, 251, 290–291, 293, 303–304, 367, 370–371, 374, 447 Carbon dioxide, 36, 137, 247–248, 251, 254, 290, 292, 322, 341, 374, 389, 424, 452, 477 detection, 211 system, 211 Carboniferous period, 34 Carbon 13 Nuclear Magnetic Resonance (13C NMR), 4, 74 Carcass disposal, 326–329 Cargill meats solutions, 326 Cast iron boiler, 247 Catalysts, 108, 123–124, 134, 154, 159–160, 173, 298, 360, 376, 378–380, 383, 388, 443, 445–446, 450, 454–457, 460, 462 CCP, 430, 434 Cell, 40, 202, 212, 249, 326, 346, 380 Cell burners, 349 Cellulose, 15, 35, 99–100 Cement industries, 11, 87, 329 Central Appalachian coals, 9, 185, 188–190 Central Electricity Generating Board, 275 CFB. See Circulating fluidized-bed CFD. See Computational fluid dynamics modeling Chain grate stokers, 242, 252, 254 Chalcophiles, 23 Chamber, 200, 214, 243, 257, 269, 278–279, 346, 469 Charging, 343–344, 347, 352 Char oxidation, 7, 19–21, 74, 76, 96, 174, 182, 184, 187, 190, 193, 295 Chemical characteristics, 7, 99, 151 fractionation, 23, 77, 122–123, 135, 138, 148–149, 303–304
487
solvents, 327, 447–448 structure, 7, 118 Chevron, 84 China, 3, 35, 45–47, 51–53, 55–56, 116, 123, 150–151, 172, 276, 321, 438, 441, 462 Chlorides, 105, 252, 303, 313, 366, 370–371, 382, 435, 444, 446, 459–460, 476 Chlorine, 11, 24–25, 37, 63–67, 71, 105, 116, 119, 183, 248, 286–287, 299, 374, 382–383, 444 Chromium, 22–25, 63–67, 72, 98, 104, 115, 123, 299, 454–455 Circulating fluidized-bed, 4, 11, 29, 85–87, 94–96, 103, 105, 107, 199, 212, 214–217, 243, 276–278, 280, 282–284, 288, 299, 301–302, 310, 320, 323–324, 366, 428 City of Lakeland, 323 Clay minerals, 134, 137, 149, 153, 309 Clays, 134, 137–138, 149, 151, 153, 256, 302, 304, 309, 312, 320, 390 Clean Air Act, 250, 474 Clean Air Act Amendments, 299, 342, 470, 474–475 Clean Air Interstate Rule (CAIR), 383, 388 Clean Air Mercury Rule (CAMR), 383–384 Clean air testing, 238 Clean coal power initiative, 324 Clean coal technology, 276, 321–322, 324–325, 479, 482 Clean corporate citizen, 475 Clean gas shift, 455–456 CO2. See Carbon dioxide Coal chemistry, 355 classification, 37, 41–42, 68, 202, 404 forming periods, 34 grade, 41 injectors, 263 mill capacity, 229 fineness testing, 238 pressure balance testing, 238 nitrogen, 33, 187, 376 petrography, 33 rank, 7, 23, 36–38, 70, 73, 136, 148, 167, 202, 235, 381–382, 385–386, 388 reactivity, 74, 76–77 reflectance, 73 seam, 35, 39, 54, 202, 251, 353, 397 Coal reserves United States, 34, 47–50 world, 34, 44–47 Coal-water slurry, 430, 436, 439
488 Index Cofiring, 27, 75, 85–86, 89, 93–94, 96, 98, 106–110, 114–118, 123–124, 173, 200, 204, 261, 302, 311, 316, 320–321, 329, 332–333, 479–480 Cogeneration, 108, 247, 279, 301, 329 Coke granular, 88 needle, 88 shot, 90–91 sponge, 88 oven gas, 83, 86 Colbert fossil plants, 106, 173 Collection, 1, 70, 136, 257, 298, 300–301, 342–347, 351–353, 355, 358–361, 363–366, 378, 380–381, 384 Collection efficiency, 344, 365 Colombia, 3, 150–151, 172 Combined cycle combustion turbines (CCCT), 85, 106 Combustion air, 11, 25, 70, 251–252, 265, 268, 276, 278, 281, 297, 333, 354–355 efficiency, 243, 275–276, 279, 281, 283–284, 330, 333, 342 engineering, 1–2, 5, 26, 28–30, 394, 480, 484 models, 182 systems, 12, 22, 24, 26, 28, 30, 93, 108, 122, 133–134, 136, 151–152, 174, 215–216, 225, 241–243, 246–247, 251, 275–277, 283, 298, 321, 395 Commercial, 2, 4, 41–42, 83, 96, 98, 105, 107–109, 111, 121, 123, 150, 243, 276, 282, 288, 299, 323–325, 342, 360, 383, 390, 395, 424, 436, 440–441, 446, 454 Compression, 215, 218, 220, 222, 389–390, 438, 452, 458, 462–463 Computational fluid dynamics modeling, 4, 184, 258–259, 395 Computer-controlled scanning electron microscopy (CCSEM), 122, 135, 137–138, 144–147, 167 Confined space rules, 482 ConocoPhillips, 439, 459–460 Continuous emissions monitoring (CEM), 355, 358 Controllers, 349–350 Coordination complexes, 302 Corrosion, 71, 119, 133, 211, 281, 302, 354, 368, 377, 443, 457 COS, 443–445, 447–450, 452–454, 456, 460, 463, 465 CO shift, 424, 445–446, 450, 454–456, 459–460, 462–463 CO shift reaction, 424, 454 COS hydrolysis, 445, 450, 453–454, 460
Cottrell, F. G., 342 Coulomb’s law, 343 Crushing, 27, 73, 201, 214–215, 218–221 Crystallinity, 162–163, 312 CS2, 443, 453, 460 Cull-cow, 327 Cyclone boiler, 12–13, 27, 85–86, 92–94, 96, 107–109, 114–115, 174, 177, 246, 270–271 firing systems, 28, 241, 243, 268
D Data historian systems, 395 DCP spectroscopy, 303–304 Deadstock, 326 Deep-set hammermills, 28 Delayed coking, 88 Denmark, 116–117 Deposit characterization, 162 strength, 161 Deposition, 22, 34–35, 61, 71–72, 105, 108, 121, 133–136, 153–154, 157–159, 162–163, 167, 249, 271–272, 314 Deregulation, 27, 85, 394, 481 Design parameters, 241, 254, 355 Detroit Stoker, 253 Devolatilization, 15–18, 74–77, 96, 152, 172, 183, 187–189, 423 Devolatilize, 12, 434, 443 Diagnostics, 342, 356–360 Diffusion, 19–20, 135, 153, 184, 294, 314, 375, 453, 465 Diode bridge, 348 Discrete element model (DEM), 410–412 Discrete minerals, 33, 38, 136, 302 Distilled water, 121–122 Distributed control system (DCS), 396 Dolomite, 138, 289–291, 293 Drag, 243, 253, 264, 278, 362 Drag chains, 200 Dropping weight, 351 Drop-tube reactor (DTR), 17, 74, 186–187, 190 Dry-feed systems, 430 Dry injection, 367, 371–372 DTE energy, 9, 27, 175, 177 Dynamic classifiers, 223
E Eagle, 434 Eastman Kodak, 92 Economic Commission for Europe Codification, 41 Edison Electric Institute, 469 E-Gas, 430, 434–435, 439, 445, 459
Index Electrical field, 343–344, 360 Electricity, 1–5, 26–29, 83–85, 87, 93, 97–98, 105–106, 108–109, 172, 181, 195–196, 237, 243–244, 246, 275, 283, 321–322, 324, 466, 472, 474, 478, 481–483 Electricity generation, 2, 5, 26–29, 84–85, 97, 243–244, 321, 474, 481–483 Electric Power Research Institute (EPRI), 261, 381, 469, 477, 479, 484 Electrification, 2 Electrophoresis, 135, 153 Electrostatic precipitation, 342 Electrostatic precipitators (ESP), 21, 26, 248, 342, 363, 366–367, 474 Elemental mercury, 301, 380–382 Energy policy, 480–482 Energy Start Certified, 475 Entrained-flow gasifier, 427, 429–430, 436, 439, 445, 459 Entraining velocity, 279 Environmental impact, 108, 464, 478 Environmental Integrity Project, 251 Erosion, 133–134, 167, 194, 201, 210, 224, 281, 283, 365–366, 379 Estonia, 3 Exinite, 39 Extractives, 35, 99–100
F Fabric filter, 26, 298, 342, 361–366, 368, 370–371, 380, 382, 384–385, 388 FactSage, 311, 314, 316, 318–320, 394 FBC. See Fluidized-bed combustion Fed-cattle, 327 Fiberboard, 97–98 Field, 94, 117–118, 138, 211, 223, 242, 254, 267, 295, 342–349, 351–352, 355, 359–361, 383–387, 396, 409, 411–413, 415, 417, 436 Fine particulates, 133, 153, 362, 477 Finite element models (FE models), 395 Finland, 94, 97, 107, 276 Fireball, 28, 194, 263–264, 268, 359 Fire retardant injection lances, 211 Firetube boiler, 247 First Coal Age, 34 Fischer-Tropsch, 424, 442, 462 Fixed bed, 278, 383, 388, 443, 465 Fixed bed gasifier, 427 Flame stability, 248, 261, 354, 408 temperature, 25, 28, 182, 268 Flexicoke, 88–89 Flowability, 201–202, 205, 208–211, 213 Flue gas desulfurization (FGD), 366–367, 380–383, 388–389
489
Fluid-bed gasifier, 433, 441 Fluid coking, 88 Fluidization, 276–278, 283, 292, 302, 427, 429 Fluidized-bed combustion, 4, 25, 95, 243, 275–278, 280, 282–285, 289, 291–293, 295, 297–298, 300–303, 316, 318, 321–322, 324–325, 327, 329, 332–333, 476 Fluidized-bed cooler, 281 Fluidized-bed firing, 29, 275–333 Fluidizing velocity, 277–279 Fluorides, 252 Fluxing agents, 23 Fly ash, 23–24, 28–29, 70, 92–94, 135, 150, 152, 248, 251, 253, 256, 299, 301, 314, 342–343, 346, 360, 363, 368–369, 371, 379–380, 382, 384 Foam, 211 Food industry, 276, 322, 325–326, 329 processing, 246, 256 Food and drug administration (FDA), 326, 329 Formic acid, 445 Fort Union lignites, 148 region, 148 Foster Wheeler, 177, 221–224, 259, 261–263, 265–266, 323 Fouling, 6–7, 22–23, 27, 29, 71, 90, 95, 105, 108, 117, 121–122, 124, 148–149, 152, 155–158, 167, 174, 195, 211, 216, 248–249, 284, 288, 302, 320, 377, 394, 397, 420, 443–444 Fouling deposits, 11, 28–29, 149, 153, 155–157 Freeboard, 279, 283, 330, 333, 429 Free radicals, 19, 332 Free swelling index, 72 Front-fired, 268 Fuel blending, 27, 30, 171–175, 181, 187, 193, 408, 411 bound nitrogen, 248, 264, 269 flexibility, 29, 93, 243, 271, 275–276, 280, 282, 284, 402–403, 421, 476, 481 matrix, 15, 18, 23, 35, 183 nitrogen evolution, 172 NOx, 21, 295, 376 preparation system, 199–202, 211, 215–216, 237–238 transportation systems, 200 Fuel Use Act, 481 Functional groups, 7–8, 15, 36, 38, 74–76, 183 Furnace exit gas temperature (FEGT), 109, 148, 182, 184, 194, 401, 408 Future Energy GSP, 439
490 Index G Gannon Station, 107 Gas particle drag, 278 separation membranes, 453 treatment, 297, 445–446, 457, 460 velocity, 153, 277–278, 280–281, 284, 292, 297, 346, 368, 380, 429 Gasification, 4, 20, 26, 29–30, 87, 107, 117, 275, 325, 423–425, 427–430, 433–436, 439, 442, 445–446, 450, 453–457, 459–460, 462–464, 479 GasGE-EER, 261 GE Energy (GEE), 430, 434–437, 445, 459–460, 462 General Motors, 94 Geochemical, 34 Georgia Pacific, 254 German Ruhr, 159 Germany, 3, 45, 52, 96, 159, 172, 276, 324–325, 436, 440 Gibbons Creek, 263 Gibbs free energy, 25, 311, 394 Global warming, 85, 251, 389, 477–478 GMDH model, 319 GPU Genco, 106, 261, 480 Granulators, 217–219 Graphite furnace atomic absorption spectroscopy, 72 Grate-fired boiler, 27, 103 Gravimetric feeders, 212 Gravity, 7, 29, 99, 137, 214, 220, 278–279, 345, 351, 411, 427, 431, 442 Great Britain, 275 Greenidge Station, 106 Green power, 85 Grindability, 7, 27, 71, 88, 201, 226–227, 238, 304, 397–398 Grinding, 27, 73, 174, 182, 184, 200, 215–216, 220–225, 230, 232, 236, 238, 304, 329, 342 Grounded plate, 343–344, 352 Gulf Coast lignites, 148 region, 148 Gulf Oil, 94 Gumz, Wilhelm, 423 Gymnosperms, 99 Gypsum, 30, 138, 367–369, 371, 373
H Halogens, 384, 454 Hammer, 211, 219–220, 351 Hardgrove grindability index (HGI), 27, 71, 88, 226 Hazardous air pollutants, 250–252, 299, 477
Hazardous wastes, 86, 173–174, 179, 329, 474 HCN, 17, 183, 295–296, 444, 450, 453–454, 457 H/C ratio. See Hydrogen/carbon ratio Heat release, 24–25, 29, 77, 108, 248–249, 252, 254–255, 259, 288 transfer, 14, 25, 94–95, 133–135, 150–151, 153, 155–156, 160–161, 167, 174, 182–183, 194, 233, 243, 249, 251, 275–276, 281, 283–284, 291, 354, 375, 395–396, 441 Heating value, 34, 37–38, 41, 68–71, 138, 194, 207, 237, 327, 330, 355, 357, 398, 423, 425 Hemicellulose, 35, 99–100 Henry’s law, 447, 450 Heritage Foundation, 469 Heteroatoms, 7–8, 16–17, 37, 74, 183, 476 Higher heating value, 8, 25, 119, 172, 185, 284, 358, 464–465 High-temperature fouling, 156–157 shift, 455 Hogged wood waste, 3, 83, 103, 179, 253, 255 Hopper, 208–212, 255, 345–346, 351, 362–363, 387, 413, 415, 431, 441, 459–460 Horizontal ball mills, 28 Hot gas cleanup, 323 H2S, 18, 443–444, 447–454, 456–457, 460, 462, 465 H2SO4, 250, 387 Humic substances, 35, 40 Hydroelectric resources, 2 Hydrogen, 11, 15, 18, 33, 36, 40, 70, 322, 375, 377, 423–424, 446, 450, 452–454, 459, 462, 465 Hydrogen/carbon ratio, 5, 112 Hydrogen chloride (HCl), 138, 299, 366, 369, 384, 476–477 Hydrogen fluoride (HF), 299, 366, 369, 477 Hydro-grates, 252
I IEA, 32, 80, 168, 334, 336, 338–339 Ignition temperature, 276 Illinois, 59, 76, 243 Impaction, 135, 153, 214, 218, 220 Incinerator, 253, 372 Independent power producers (IPP), 86, 97, 119, 481 India, 45, 321 Indiana, 59, 107, 457 Indonesia, 150 Industrial boilers, 11, 79, 84, 96, 246–247, 256, 366, 372 power plants, 1
Index Inertial impaction, 135, 153 Inertinite, 39–40 Inert material, 276 Information collection request (ICR), 300–301, 380–381 Inorganic elements, 33, 38, 134, 136–138, 301–302 Integrated gasification-combined cycle (IGCC), 4, 87, 96, 107, 389, 424–425, 429, 436, 439, 449, 457–462, 464–466, 482–483 Interior region, 58 Internal reverse flow zone (IRZ), 267 International Flame Research Foundation, 258 International Organization for Standarization (ISO), 41, 61, 68 International Paper Company, 254 Ion-exchangeable cations, 302 Ireland, 3 Iron, 12, 22–23, 96, 111, 114, 135, 149, 151, 153, 156, 161, 174, 191, 193, 201, 256, 309–310, 347, 358–360, 408, 445, 453
J Jacksonville Electric Authority, 87, 107, 116, 323 Japan, 150, 159, 269, 324–325, 434, 436, 465 JEA. See Jacksonville Electric Authority
K Kaolin clay, 302, 312, 320 KBR, 441–443, 459 Kentucky, 58–59, 243 Kilns, 1, 11, 24, 26–28, 84, 86–87, 98, 114, 174, 179, 241, 243, 246–247, 256–257, 270, 396, 474 Kinetic parameter, 7 Kinetics of devolatilization, 17, 96, 187 Kingston Fossil Plant, 106 Korea, 94, 150
L Landfill gas, 85 Lead, 12, 22–24, 74, 119, 148, 193, 201, 212, 256, 299–300, 302, 347, 356, 365, 473, 477 Lignin, 35, 99–100 Lignite, 2, 8, 16, 34–37, 45, 48, 51–52, 54, 59, 61, 78, 134, 138, 148–149, 210, 263, 275, 285, 296, 325, 381–382, 436, 439–440 Limestone, 94–95, 150, 256, 263, 276–277, 289–293, 297, 300–301, 313, 367–368, 372–375, 441 dissolution, 375 station, 263 Linear reactor, 348 Liptinite, 39–40
491
Liquid phase, 29, 135, 153, 157, 162, 312–315, 318–320, 374, 458 redox, 447, 453 Liquid-to-gas ratio (L/G ratio), 368–369 Lithium, 359 Lithophiles, 23, 92 Litter, 285, 304, 306–307, 312–313, 328 Long Fork, 175, 185, 187 Loss-on-ignition (LOI), 109, 357 Low-NOx burners, 134, 258, 475 Low sulfur coal, 48, 116, 369 Low-temperature fouling, 156, 158 shift, 456 Lurgi, 442, 444, 450, 456 Lurgi-Sasol, 430, 432
M Macerals, 33, 37–40 Mad cow disease, 326 Magnesium, 22, 134, 136, 138–143, 149–151, 179, 182, 290, 302, 307–308, 367–368, 374 Manure, 173, 285, 287, 302, 304–319, 328 Mass transfer, 20, 281, 368, 375–376, 427, 429 Maximum volatile yield, 16, 18, 74–75, 96 McNeill Generating Station, 254 Meat and bone meal (MBM), 326–327, 332 Mercuric chloride (HgCl2), 382 Mercury, 22–25, 29, 63–67, 72, 90, 104, 107, 109, 115, 123, 173, 181, 252, 299–301, 324, 341, 371, 380–389, 393, 445–446, 453, 457, 460, 462, 465, 477 Mercury emissions, 107, 109, 123, 300, 380–381, 383–384, 477 Metal carbonyls, 445 Metallic slag, 155 Metallurgy, 9, 11, 244 Methane, 36, 70, 183, 211, 424–426, 443–444, 477 Methane detectors, 211 Methanol, 96, 423, 446, 450–452, 462–464, 466 Michigan City Generating Station, 107, 480 Microbiological degradation, 35 Middle East, 45–46, 321 Mineral(s), 22–23, 33, 38, 40, 71, 134, 136–138, 144–147, 149–154, 171, 302, 304, 309, 321 formation, 251 grains, 40, 135–138, 152 matter, 36, 39, 41–42, 68–70, 136, 276, 278, 298 Mitsubishi Boiler, 263 Mixed municipal wastes, 5, 253–254
492 Index Moisture, 6, 8, 11–16, 21, 25, 27, 36–38, 41–42, 53–56, 62, 64–70, 89, 99–101, 103, 105, 113, 116, 119–120, 139–143, 148, 179–180, 182–183, 194, 201–206, 208–209, 212, 216, 222, 225–226, 228, 230–231, 234–237, 248, 254–255, 264, 267, 276, 284, 286–287, 293, 305, 330, 355, 357, 362, 366, 376, 408–409, 427, 429, 431, 436, 440, 453, 464 Monoethanol amine (MEA), 390, 447 Monroe power plant, 9, 27, 175–176, 178, 180, 182, 185–187, 193, 195, 258, 260, 404–405 Morphology, 153, 165, 201, 371 Moving-bed gasifier, 427, 430, 450 Mt. Tom Station, 261 Municipal waste, 5, 253–255
N Nahcolite, 371 National Academy of Sciences, 469 National Cattlemen’s Beef Association (NCBA), 326, 329–330, 332 National Coal Board, 275 National Environmental Policy Act (NEPA), 472–474 Natural Resources Defense Council, 469 Nenana field, 172 Neural network, 395–396 New Source Performance Standards, 322, 342 New Source Review, 195, 475 New York State Electric and Gas Co., 106 NH3, 14, 159, 183, 295–297, 367, 377–378, 444, 450 Nickel, 22–25, 63–67, 88, 90, 92, 95, 104, 115, 123, 182, 247, 299, 445, 450, 454 NiSources, 87, 177, 480 Nitric acid, 251 Nitrogen, 4, 11, 15–18, 21–22, 33, 37, 53–56, 62, 64–67, 70, 74, 77–79, 89–90, 100–101, 109, 113, 116, 119–121, 139–143, 172–173, 183, 187, 190, 193, 248, 250–251, 264, 267, 269, 276, 286–287, 289, 295–298, 305, 341, 376–377, 379, 431–433, 444, 446, 452, 458–459, 461, 474 N2O, 21, 277, 295–298 NO2, 21, 159, 250, 295, 376, 378 NOx, 4, 11, 21–22, 25–26, 28–30, 70, 78–79, 85, 92–94, 96, 100, 107, 109–110, 116, 134, 173, 181, 183, 187, 239, 243, 249–251, 257–258, 261, 263–265, 267–269, 275–276, 295, 297–298, 323–324, 331, 354, 372, 376–379, 383, 388, 393, 395–396, 423–424, 445, 450, 452, 461–462, 464–465, 474–477, 479 Noncondensing extraction turbine, 102
North Antelope, 175 Northern Indiana Public Service Co. (NIPSCO), 87, 95, 107, 177, 181 Northern States Power, 87, 107, 173 Nucla Station, 323 Nuclear fuels, 2 power, 243 Nuclear magnetic resonance, 4, 33, 74 Nuon, 436, 438
O O3, 250 Occupational Safety and Health Administration (OSHA), 470, 482–483 O/C ratio. See Oxygen\carbon ratio Office of Coal Research, 275 Ohio Power Company, 323 Oil reserves-world, 34 Oil Shale, 3, 5, 285, 296 O2 monitors, 211 Opacity, 116, 180, 194–196, 255, 349, 351–356, 358, 360, 387–388, 407–408, 420 Operating costs, 280, 433, 450, 475 Opportunity fuels, 5, 15, 75, 83–86, 94, 98, 103, 121, 124, 172–173, 177, 179, 182, 248, 270, 311, 325, 479, 481 Opposed fired, 268 Organically-associated cations, 134, 136, 152 Oriental Chemical Industries, 94 OSI PI, 395, 402–403 OSIsoft, Inc., 177 Ottumwa Generating Station, 264 Overfire air, 26, 28, 93, 249, 257, 263, 265, 267–269, 297, 299, 330, 332, 396 Oxidant, 11, 251, 427–428, 430, 432, 441–442, 462 Oxidized mercury, 301, 380, 382–383, 388 Oxycombustion, 251 Oxyfuel, 389 Oxygen, 4–5, 7, 11, 15, 18–23, 25, 33, 36, 38, 40, 53–56, 62, 64–67, 70, 74–76, 78, 89, 101, 112–113, 120, 134, 136–137, 139–143, 163, 183–184, 187, 193, 250–251, 261–262, 264–265, 267–268, 278, 286–287, 292, 295, 298, 305, 330, 332–333, 358–360, 369, 390, 424–427, 429–430, 432–434, 440–445, 458, 462, 479 Oxygen\carbon ratio, 5, 90, 96, 112–113 Oxygen-enhanced combustion, 4, 11, 78, 261, 332 Ozone, 250, 474
P Package boiler, 288–289 Paradise Fossil Plant, 270 Parr formulas, 42, 69
Index Particleboard, 98, 100–101 Particles, 7, 12–14, 73, 94, 112, 116, 135, 137–138, 148–153, 155–158, 162–163, 182–183, 187, 193, 216–217, 219, 222–223, 238, 243, 250, 254, 263–265, 267–269, 276, 278–281, 284, 290, 292, 294, 297, 299–301, 313–314, 333, 342–344, 346, 352, 354–355, 359–361, 371, 382, 406, 411, 417, 427, 429, 438, 441–442, 445 Particulate control device (PCD), 279, 299–300, 384 Particulate matter, 264, 298, 324, 329, 341–342, 345, 362, 378, 382, 435, 445, 459, 465 Peat, 3, 34–37, 40, 202–203, 215, 284–285, 296 Pellet derived fuel, 285 Penn State, 21, 75, 185, 302–303, 311–312, 320, 325, 328–329 Pennsylvania, 49–50, 58, 61, 71, 86–87, 97, 121, 218–221, 301, 326 Pennsylvania Energy Development Authority (PEDA), 326, 329, 332 Pennsylvania State University, 71, 121 Peroxyacetyl nitrate (PAN), 250 Petroleum coke, 2–6, 8, 15–18, 21, 26–28, 75, 77, 83–96, 114, 172, 177, 181–182, 193, 204, 209, 241, 246, 248, 261, 270, 284–285, 287, 394, 423, 436, 445, 462, 470, 478–479 Petrox, 94 PG&E Scrubgrass Generating Station, 301 pH, 369–370, 374–375 Phosphorus, 22, 139–143, 148–149, 310 Physical/chemical solvents, 327, 445, 447–448, 450 Physical properties, 7, 27, 165, 290, 302, 449 Physical solvents, 445, 447–448, 450 Physical washes, 448–449, 460 Pine shavings, 286, 305–311 Pittsburgh seam bituminous coal, 76, 78 Planer shavings, 83, 98, 100 Plant Hammond, 106 Plant Kraft, 106 PLC logic, 350 Plug flow, 195, 410, 412 Plywood, 97–101, 285–286 Poland, 3, 35, 46–47, 51–52, 54–55, 150, 172 Polk County IGCC, 87, 483 Pore blockage, 159, 290 Pore size distribution, 290, 294 Pore volume, 290, 293–294 Porosity, 7, 38, 99, 119, 161, 290, 293–294 Porphyrin complexes, 92 Portfolio Standards, 85, 470 Potassium, 22, 101, 119, 122, 134, 136, 138–143, 148–149, 179, 192, 248, 302–303, 306–308, 314–315, 378, 447
493
Powdered activated carbon (PAC), 383–388 Powder River Basin, 9, 49, 76, 83, 85, 149, 172, 204, 230, 235–236, 248, 263, 464, 470, 473, 476 Powder River Basin coals, 83, 235, 248 Praxair, Inc., 261 Precipitator control, 348, 356–357 President Carter, 481 President Clinton, 481 President Richard Nixon, 474 President Ronald Reagan, 481 Presque Isle Power Plant, 477 Pressure drop, 159, 229, 234, 281, 362, 365–366, 368, 454, 457 Pressurized bubbling fluidized-bed (PBFBC), 282–283 Pressurized circulating fluidized-bed (PCFB), 277, 282–284, 323 Primary air, 28–29, 194, 204, 221–222, 224–225, 228–230, 232–234, 237–239, 253, 258, 267–269, 280 Primary gas cleaning, 435 Primary syngas cooling, 430, 433–434 Prion, 326, 333 Process control, 342, 347–351, 396 Process heater, 13, 241, 346–347 Process industry furnace, 1 Process industry kiln, 1, 28, 174, 179, 196, 243, 257 Project Energy Independence, 480 Prompt NO, 22, 251, 376 Proximate analysis, 5–6, 16, 39, 53–56, 68–70, 74–76, 89–90, 100–101, 112–113, 118–120, 137–138, 140–143, 174, 179, 185, 203–206, 238, 304–305, 330 Public Service Commissions, 393 Public Utilities Regulatory Policies Act (PURPA), 26, 86, 481 Public Utility Commissions, 393 Pulp and paper industries, 4, 11, 24, 102, 108, 247 Pulp mill sludge, 86 Pulse jet, 362, 364, 366 Pulverized coal boiler, 28, 93, 115, 121, 249, 256, 262, 479–480 Pulverized coal-fired boilers, 276 Pulverizers, 26–28, 71, 109, 115, 174, 182, 195, 220–223, 225–229, 243, 257, 266, 368, 372, 399–400, 404, 407–408, 410, 417 Pyridine, 37 Pyrite, 22–23, 137–138, 144–146, 149, 151, 153, 155–156, 304, 359 Pyrolysis, 7, 12–18, 22, 74–75, 77, 182–183, 185, 187, 189, 328, 423–424, 427, 429 Pyrrole, 37
494 Index Q Quartz, 22, 134, 138, 144–146, 151, 153, 165–166, 304, 309–310, 314
R Radial feeders, 269 Radicals, 12, 16, 18–20, 22, 183, 187, 251, 296, 332, 376 Rail car unloader, 176 Rapper, 350–351, 353, 359 Raw gas shift, 450, 456–457 Reburn(ing), 26, 93, 108, 268, 297, 376, 477 Reciprocating grates, 255 Recirculation zone, 258–259 Rectisol, 445, 447, 450–453, 460, 462–463, 465 Recycled flue gas, 262 Reed Canary grass, 117, 120–122, 206, 286, 305–311, 328 Refineries, 86, 90, 94, 246, 248, 285, 395, 423–424, 445, 453–454, 459, 466 Refinery off-gases, 83 Refractory lining, 210, 433, 436 Refuse-derived fuel, 85, 103, 285, 366 Regenerative air heater, 233 Reheat boilers, 9, 11, 95, 102, 108–109, 245 Reheat steam, 29, 94, 109, 244 Reliant Energy, 106, 480 Rendering industry, 329 Renewable energy, 3, 254, 471–472, 479, 482 Reserves, 3, 33–34, 44–50, 246, 322, 402–403 Residence time, 94, 105, 216, 276, 283–284, 292, 294, 297, 299–300, 332–333, 369, 379, 429 Residential, 2, 57, 471–472, 478 Resistivity, 346–347, 349, 351–357, 359–360, 365 Resistivity conditioning, 360–361 Resources, 2–4, 33–34, 44, 47–48, 52, 83, 85, 110, 116, 246, 299, 321, 394, 408, 469–470, 481 Reverse air, 362, 364–366 RFID chips, 413–415 Rice hulls, 83, 117–121 Ring of Fire, 37, 172, 182 Roll crushers, 217, 220 Roller grates, 252 Roll-wheel pulverizers, 28 Rotary dumper, 176, 404 Russia, 3, 45–47, 51–55, 150–151, 172
S Safety, 201, 208, 211, 216, 225, 432, 470, 477, 482–484 Salix, 310, 314 Sandia National Laboratories, 121 Santee Cooper, 174
Sawdust, 6, 16, 18, 21, 77, 83, 97–103, 105–109, 173, 205, 261, 480 Scaling, 279, 368 Scanning electron microscopy (SEM), 122, 135, 137, 162–163 SCR. See Selective catalytic reduction Screw conveyors, 200 Scroll feeders, 269 Scrubbers, 30, 250, 289, 342, 367–369, 374, 476–477 Scrubber technologies, 30 Secondary air, 28–29, 233, 257–260, 263–264, 266–267, 269, 280, 297 Sedimentary rock, 33 Selective catalytic reduction, 26, 29–30, 93–94, 108, 123–124, 134, 154, 159–160, 173, 297–298, 348–349, 376, 378–382, 386, 388, 446, 461, 465–466, 476 Selexol, 445, 447, 449–451, 460, 462, 465 Separated overfire air (SOFA), 26, 28, 93, 109, 249, 263, 269 Sewage sludge, 85, 285–286, 311, 479 Seward Generating Station, 106–107, 480 Shaft power, 1, 105 Shaker, 362–363, 366 Shawville Generating Station, 106 Shear, 215, 217, 219 Shell, 118, 290, 428, 430, 433–436, 438, 443, 445, 456, 459, 462 Siberia, 4 Siemans, 261 Siemans-Westinghouse, 261 Sierra Club, 469 Sieve analysis, 182, 184 Silica, 12, 15, 22, 101, 148–149, 151, 179–180, 359–361 Silicon, 22, 139–143, 308–310, 314, 316, 321, 343 Silos, 176–178, 180, 195, 207–208, 210, 213, 371, 404–405, 410 Sintered slag, 155 Sintering, 149, 155, 161–162, 165, 208, 290–293, 303, 314, 378, 455 Sizing requirements, 201, 215–216 Slag, 11, 23, 28–29, 71, 93, 114, 133–135, 148, 153–156, 167, 193, 217, 243, 249, 256–257, 268–271, 276, 318, 320, 358–359, 429, 432–436, 440, 445–446, 460 Slag deposits, 135, 153–156 Slagging, 6–7, 22–23, 27–29, 71–72, 90, 93, 105, 108, 117, 121–122, 124, 148–149, 151–152, 155, 167, 174, 179–180, 194–195, 204, 211, 216, 243, 248–249, 259, 269–270, 284, 288, 302, 321, 359, 394, 397, 402, 407–409, 418, 420, 427–429, 431–434, 436, 439, 443
Index deposits, 155, 288 fuels, 29, 284 gasifiers, 429, 433 Slip velocity, 277, 281, 429 Sloping grates, 252–253 SNCR, 26, 297–298, 376–378 Sodium, 22, 98, 101, 122, 134, 136, 138–143, 148–149, 159, 182, 248, 302–303, 306–308, 312–314, 347, 359, 367–368, 371, 378, 440, 444 Sodium bicarbonate (NaHCO3), 371 Sodium carbonate (NaCO3), 367, 371 Sodium cations, 313 Sodium chloride, 313, 444 Sodium sesquicarbonate (Na2 CO3*NaHCO3*2H2O), 371 Solid wastes, 2, 4, 103, 173, 252, 324, 372, 474 Sootblowers, 26, 148, 194, 407 Sootblowing, 28, 158, 167, 194 Sorbent, 276, 278, 281–282, 289, 291–294, 298, 367, 369–371, 383–384, 390, 441–442 injection, 383–384 utilization, 289, 294 South Africa, 3, 35, 46–47, 51–52, 150–151, 159, 172, 424, 442 South African, 150–151, 159 Southern Company, 118, 264, 459, 480 Space velocity, 378–379 Specific gravity, 7, 99 Specified risk materials (SRMs), 326–330, 332 Spent pulping liquors, 4–5, 86 Spray dryer, 367, 369–371, 373, 375, 381–382, 388, 477 Spreader stoker, 103, 105, 242, 253–255 Staged gasification, 434–435 Stainless steel, 210, 412 Stamler reclaimer, 179 Static classifiers, 223 Stationary grates, 255 Steam, 1, 4, 9–11, 14, 24, 26, 29, 71, 83, 86, 88, 94–96, 98, 102, 105–106, 108–109, 150, 153, 199, 205–206, 223, 229, 237, 239, 244–247, 252, 269–270, 282–283, 288, 325, 360, 377–379, 424, 428, 431, 433, 438, 441–442, 446, 451, 455–456, 459, 461–463, 482 Steam methane reforming reaction, 424 Steel industries, 2, 366 Stochastic Model, 406, 410–412 Stockton CoGen Plant, 301 Stoker, 12, 27, 29, 105, 115, 121, 214–215, 217, 241–243, 252–254 Stoker-fired boilers, 174, 199, 217, 276, 342 Stoker firing, 13, 23, 29, 242–243, 246, 252–255, 278
495
Straw, 83, 117–119, 121–123, 248, 254, 304, 310, 313–314 Strontium, 22, 63–67, 134, 136 Subbituminous coal, 5–8, 16–17, 34, 36, 45, 47, 49, 51–52, 57, 61, 76, 78, 113, 138, 148, 155, 172–173, 182, 185, 187, 193, 203–204, 209, 215, 285, 328, 381–382, 385, 388, 431, 439, 464, 473, 476 Sulfation, 94, 290–294 Sulfides, 24, 92, 122, 134–136, 303, 445, 457 Sulfur, 15–16, 18, 21, 23–24, 33, 37, 41–43 48 forms, 71 recovery, 436, 453, 458, 460–461 regulations, 248 Sulfur dioxide (SO2), 21, 93, 173, 289–292, 341, 366, 382, 473–474 Sulfur trioxide (SO3), 21, 53–56, 63–67, 91, 102–103, 122, 139–143, 159, 250, 305, 312, 317, 355, 358, 360, 366, 371–374, 377, 379–380, 384, 386 Supercritical boilers, 4, 11, 87, 106, 246 Surface area, 14, 38, 149, 183, 290, 292, 294, 342, 375, 379–380 Surface-mined coal, 54 Suspension-fired, 26, 216 Suspension firing, 25, 27–28, 115, 216, 278 Swirl-stabilized burner, 28, 258–259 Switchgrass, 6, 8, 16, 18, 21, 27, 77, 116–122, 173, 206, 264, 286, 317–320, 328 Syngas, 423–427, 430, 432–433, 436, 438–439, 442–443, 446, 448, 450–456, 459–463, 465–466 Syngas pretreatment, 459 Synthesis gas, 423–425, 427, 434, 443–447, 452–453, 462–463 Synthetic natural gas (SNG), 426, 442, 462 System availability, 243, 275
T Taiwan, 150 Tampa Electric Company, 87, 107 Tangentially-fired boiler, 28, 85, 93, 174, 242, 257, 262–264, 480 Teflon coating, 210 Tennessee Valley Authority (TVA), 106–107, 173, 246, 480 Tertiary air, 258, 264 Texas, 50, 58–59, 61, 386 Textile formation, 246 Thermal efficiency, 244, 321 Thermal flywheel, 276 Thermal NOx, 21–22, 251, 295, 376 Thermogravimetric analysis (TGA), 74, 77 Thermophoresis, 135, 153, 156 Tidd, 282, 323
496 Index Tire, 2, 4–5, 23, 26–27, 85–86, 103, 110–116, 173, 179, 248, 254, 270, 285, 287, 479 Tire derived fuel (TDF), 2, 4–5, 26–27, 85–86, 103, 110–116, 173, 248, 270, 287, 479 TOXECON, 384–387, 477 Trace metals, 12, 22–24, 29, 85, 88, 90–91, 96, 101–102, 104, 123–124, 252, 394, 477 Tractebel Power Inc., 301 Tramp iron, 201, 218, 222 Transformer, 282, 343, 345, 348 Transformer rectifier (TRSet), 345, 347–348 Trenton Station, 342 Tri-State Generation and Transmission Association, Inc., 323 Trona, 371, 477 Tubular air heater, 232–233 Turbines, 1, 29, 85, 87, 94, 98, 102, 106, 108, 244, 247, 282–283, 323, 423–425, 446, 451, 457–458, 461–462, 464–466, 481 Turbulence, 12, 257, 262, 264, 276
U Ukraine, 46–47, 51–52 Ultimate analysis, 6, 53–56, 70, 75, 89–90, 100–101, 112–113, 118–121, 139–143, 174, 179, 185, 238, 305 Ultra supercritical boilers, 4 Unburned carbon, 92–93, 239, 250–251, 298, 380 United States, 2–3, 33–35, 37, 41, 44–52, 54–59, 61–62, 64–67, 84–85, 87, 92–93, 96–98, 106, 110–111, 115–117, 119, 121, 124, 138, 148, 150, 172, 243, 246, 248, 253, 269, 275–276, 282, 301, 321–327, 333, 341, 380, 394, 439, 473, 481 University of North Dakota Energy and Environmental Research Center, 303 Unsaturated hydrocarbons, 444, 457 Upper Carboniferous Period, 34 Urban wood waste, 98, 100–101, 104, 106–107, 205 Urea, 100, 298, 377–378, 452, 462 U.S. Chamber of Commerce, 469Utility boiler, 10–11, 84–85, 98, 106, 109, 134–135, 138, 153–156, 158, 160, 163, 167, 196, 244–246, 282, 288–289, 329, 341, 363, 369 U.S. Department of Energy (DOE), 109–110, 261, 272, 275, 321–324, 326, 329, 332–333, 381, 383, 385, 477–480, 482 U.S. Department of Interior, 275 U.S. Environmental Protection Agency (EPA), 72, 114, 300–301, 328–329, 383, 473, 475, 477–478 Utility power plants, 1, 213–214
V Vanadium, 6, 22–24, 63–67, 88, 90, 92, 95–96, 104, 123, 182, 193, 299, 360, 379, 453 Veneer, 97–98 Vertical spindle mills, 217, 222, 226 Vesicular slag, 155 Vibrator, 351 Vietnam, 51–52 Viscosity, 23, 72, 162–163, 165, 303, 316–321, 344, 394, 432, 449–450 Vitrinite, 37–40, 43, 62, 64–66, 73 Void model, 410–412 Volatile evolution, 17, 73, 78–79, 172, 174, 193 Volatile matter, 15–18, 36–39, 41–43, 53–55, 62, 64–70, 73–79, 89, 93, 113, 139–143, 172, 179, 183, 190, 193, 202–206, 216, 253, 264, 284, 286–288, 296, 305, 330 Volatile matter evolution, 77–79 Volatile organic compounds, 250, 465 Volatile release patterns, 185–187 Volumetric feeders, 212
W Wall-fired boiler, 93, 257 Waste oils, 173, 329 Wastewater treatment gas, 85 Water deluge system, 211 Water gas shift reaction, 424 Water jacket, 433, 442 Water-cooled membrane wall, 433 Watertube boiler, 247 WE Energies, 477 West Virginia, 50, 58, 61, 324 Western Greenbrier Co-Generation, LLC, 106–107, 324, 480 Wet-feed systems, 431 Weyerhaeuser Company, 254 Willow Island Generating Station, 107, 480 Wind power, 2 Winkler, 275, 428–429 Wood chips, 247, 307, 328 Wood waste, 2–3, 5, 27, 83–86, 98, 100–104, 106–107, 173, 177, 179, 181–182, 205, 248, 253–255, 270, 285, 324, 394, 479, 481
X X-ray fluorescence (XRF), 122, 139–143, 177, 193
Z Zeldovich mechanism, 22, 251 Zinc, 22–25, 111, 113–116, 123, 456, 463