United States Environmental Protection Agency Office Of Air Quality Planning And Standards Research Triangle Park, NC 27711 EPA-453/R-01-005 January 2001 FINAL REPORT Air National Emission Standards for Hazardous Air Pollutants (NESHAP) for Integrated Iron and Steel Plants - Background Information for Proposed Standards Final Report ------- EPA-453/R-01-005 National Emission Standards for Hazardous Air Pollutants (NESHAP) for Integrated Iron and Steel Plants - Background Information for Proposed Standards U.S. Environmental Protection Agency Office of Air Quality Planning and Standards Metals Group, MD-13 Research Triangle Park, NC 27711 Prepared Under Contract By: Research Triangle Institute Center for Environmental Analysis Research Triangle Park, NC 27711 January 2001 ------- This report has been reviewed by the Emission Standards Division of the Office of Air Quality Planning and Standards of the United States Environmental Protection Agency and approved for publication. Mention of trade names or commercial products is not intended to constitute endorsement or recommendation for use. Copies of this report are available through the Library Services (MD-35), U.S. Environmental Protection Agency, Research Triangle Park, NC 27711, or from the National Technical Information Services 5285 Port Royal Road, Springfield, VA 22161. ------- TABLE OF CONTENTS 1.0 INTRODUCTION 1-1 1.1 Statutory Basis 1-1 1.2 Selection of Source Category 1-2 2.0 INDUSTRY OVERVIEW 2-1 2.1 Background 2-1 2.2 Geographic Distribution 2-3 2.3 Size Distribution 2-3 2.4 Product Characterization 2-3 2.5 References 2-5 3.0 PROCESS DESCRIPTION AND BASELINE EMISSIONS 3-1 3.1 Sinter Plants 3-1 3.1.1 Emission Points 3-3 3.1.2 Factors Affecting Emissions 3-3 3.1.3 Estimates of Baseline Emissions 3-5 3.1.3.1 HAP Metal Emissions from the Windbox 3-5 3.1.3.2 PAH Emissions from the Windbox 3-7 3.1.3.3 Volatile Organic HAP Emissions from the Windbox 3-9 3.1.3.4 Emissions of D/F from the Windbox 3-9 3.1.3.5 Emissions from the Discharge End 3-9 3.1.3.6 Emissions from the Cooler 3-10 3.1.4 Uncertainties in the Emission Estimates 3-11 3.2 Blast Furnaces 3-12 3.2.1 BlastFurnaceAuxiliaries 3-14 3.2.1.1 Stoves 3-14 3.2.1.2 BlastFurnace Gas Cleaning 3-15 3.2.1.3 Pulverized Coal Injection 3-15 ------- TABLE OF CONTENTS (continued) 3.2.2 Emission Points and Factors Affecting Emissions 3-16 3.2.3 Estimates of Baseline Emissions 3-19 3.2.3.1 Casthouse PM Emissions 3-19 3.2.3.2 Miscellaneous Emission Points 3-19 3.2.3.3 Estimates of MnEmissions 3-26 3.2.3.4 Estimates of HCNEmissions 3-26 3.2.4 Uncertainties in the Emission Estimates 3-27 3.3 Basic Oxygen Process Furnace 3-33 3.3.1 Reladling, Desulfurization, and Slag Skimming 3-33 3.3.2 BOPF Shop 3-37 3.3.2.1 Bottom Blown Furnace 3-39 3.3.2.2 Combination Blowing 3-39 3.3.3 Ladle Metallurgy 3-40 3.3.4 Emission Points and Factors Affecting Emissions 3-40 3.3.5 Estimates of Baseline Emissions 3-43 3.3.5.1 BOPF Charging, Oxygen Blow, and Tapping PM Emissions 3-43 3.3.5.2 Miscellaneous Emission Points 3-43 3.3.5.3 Estimates of Mn Emissions 3-44 3.4 References 3-51 4.0 EMISSION CONTROL TECHNIQUES AND EQUIPMENT 4-1 4.1 SinterPlant 4-1 4.1.1 Windbox 4-1 4.1.1.1 Baghouses 4-3 4.1.1.2 Scrubbers 4-5 4.1.2 DischargeEnd 4-7 4.1.3 Materials Handling 4-7 4.1.4 Capture and Control System Performance 4-8 4.1.5 Pollution Prevention 4-8 4.2 Blastfurnace 4-9 ------- TABLE OF CONTENTS (continued) 4.2.1 Casthouse 4-9 4.2.2 Gas Cleaning 4-14 4.2.3 Wastewater 4-14 4.2.4 Capture and Control System Performance 4-16 4.3 BOPF Shop 4-16 4.3.1 Primary Furnace Controls 4-16 4.3.1.1 Open Hood Designs 4-17 4.3.1.2 Closed Hood Designs 4-21 4.3.2 Secondary Sources of Emissions 4-24 4.3.2.1 Furnace Controls 4-24 4.3.2.2 Ancillary Operations 4-27 4.3.3 Ladle Metallurgy Operations 4-28 4.4 References 4-36 5.0 EXISTING STATE REGULATIONS 5-1 5.1 SinterPlant 5-1 5.1.1 Windbox 5-1 5.1.2 DischargeEnd 5-1 5.1.3 SinterCooler 5-2 5.2 Blast Furnace 5-5 5.3 BOPF Shop 5-8 5.3.1 Primary Control Devices 5-8 5.3.2 BOPF Secondary Controls 5-8 5.3.3 Hot Metal Transfer, Desulfurization, Slag Skimming, and Ladle Metallurgy 5-11 5.3.4 BOPF Shop Roof Monitor 5-12 6.0 CONTROL COSTS 6-1 6.1 Approach 6-1 6.2 BOPF Primary Control Systems 6-1 6.3 Secondary Capture and Control Systems 6-2 iii ------- TABLE OF CONTENTS (continued) 6.4 Bag Leak Detection Systems 6-4 6.5 Total Nationwide Costs 6-4 6.6 References 6-5 7.0 ENVIRONMENTAL IMPACTS 7-1 7.1 Emission Reductions 7-1 7.2 Secondary Impacts 7-2 7.3 References 7-3 APPENDIX A - SUMMARY OF SINTER PLANT TESTING A-l APPENDIX B - DOCUMENTATION FOR THE MACT FLOOR B-l IV ------- FIGURES 2-1 Locations of Integrated Iron and Steel Plants 2-4 3-1 Schematic of Sinter Plant Emission Points and Typical Controls 3-4 3-2 Schematic of Blast Furnace Emission Points and Typical Controls 3-17 3-3 Schematic of BOPF Shop Emission Points and Typical Controls 3-41 v ------- TABLES 2-1 Integrated Iron and Steel Plants 2-2 3-1 Estimates of PM Emissions From Sinter Plants 3-6 3-2 Estimates of HAP Emissions From Sinter Plants 3-8 3-3 Casthouse PM Emission Estimates Submitted by the Companies 3-20 3 -4 Raw Material Handling, Slag Handling, and Furnace Slip PM Emission Estimates Submitted by the Companies 3-21 3-5 Blast Furnace Stove PM Emission Estimates Submitted by the Companies 3-23 3-6 Estimates of PM Emissions From Blast Furnace Operations 3-24 3-7 Mn Data Provided By the Companies 3-28 3-8 Estimates of Mn Emissions From Blast Furnace Operations 3-29 3-9 Estimates of HCN Emissions From Blast Furnace Wastewater Treatment 3-31 3-10 BOPF Shop Emission Control Systems-Closed Hood BOPF Shops 3-34 3-11 BOPF Shop Emission Control Systems-Open Hood BOPF Shops 3-35 3-12 Summary of Controls for Ancillary Processes 3-36 3-13 PM Emissions From the BOPF Shop Reported by the Companies 3-45 3-14 Emission Factors Used for the BOPF Shop 3-46 3-15 Estimates of PM Emissions From the BOPF Shop 3-47 3-16 Estimates of Mn Emissions From the BOPF Shop 3-49 4-1 Emissions Controls for Sinter Plant Windboxes 4-2 4-2 Sinter Discharge and Cooler Control Technologies 4-7 4-3 Casthouse Capture and Control Systems 4-10 vi ------- TABLES (continued) 4-4 Emissions Controls for Blast Furnace Casthouses 4-12 4-5 Gas Cleaning Systems for Each Furnace 4-15 4-6 Open Hood BOPF Shop Primary Control System 4-19 4-7 Operating Parameters of Closed Hood BOPF Systems - Venturi Scrubbers 4-23 4-8 Secondary Emission Control Systems in the BOPF Shop 4-29 4-9 Secondary Control Device Parameters 4-30 4-10 Ladle Metallurgy Station Control Device Parameters 4-33 5-1 Sinter Plants in the US 5-1 5-2 Controls and Emission Limits for the Discharge End 5-3 5-3 Discharge End Fugitive Emissions: Opacity Limitations 5-4 5-4 Sinter Cooler Descriptions and Limits 5-4 5-5 Casthouse Emission Controls and Opacity Limits 5-6 5-6 Emission Limits for Casthouse Control Devices 5-7 5-7 Emission Limits for Primary Control-Open Hood 5-9 5-8 Emission Limits for Primary Control-Closed Hood 5-10 5-9 Emission Limits for Secondary Control Devices at Closed Hood BOPF Shops 5-10 5-10 State Emission Limits for Secondary Control Devices at Open Hood BOPF Shops 5-11 5-11 State Limits for Transfer, Desulfurization, and Slag Skimming - All Baghouses 5-13 5-12 State Limits for Ladle Metallurgy Process 5-14 5-13 Summary of BOPF Roof Monitor Opacity Limits 5-15 vii ------- 6-1 Baghouse Costs 6-3 TABLES (continued) 6-2 Nationwide Cost Estimates 6-5 7-1 Estimates of Emission Reductions 7-2 vui ------- LIST OF ACRONYMS acfm Actual cubic feet per minute BOPF Basic oxygen process furnace CAA CleanAir Act cfm Cubic feet per minute CO Carbon monoxide D/F Dioxin and furan dscfm Dry standard cubic feet per minute EAF Electric arc fumace(s) ESP Electrostatic precipitator g/scm Gram(s) per standard cubic meter gr/dscf Grain(s) per dry standard cubic foot HAP Hazardous air pollutant HCN Hydrogen cyanide Ib/hr Pound per hour Ib/ton Pound per ton MACT Maximum achievable control technology Mn Managanese NESHAP National emission standard for hazardous air pollutants NSPS New source performance standard PAH Polynuclear aromatic hydrocarbon(s) Pb Lead PM Particulate matter Q-BOP Quelle basic oxygen process scfm Standard cubic feet per minute tpd Tons per day tpy Tons per year VOC Volatile organic compound(s) IX ------- 1.0 INTRODUCTION This document summarizes the basic background information used in the development of MACT standards for the integrated iron and steel manufacturing source category. All references cited in this document are available in Docket No. A-2000-44. In addition, this document is supplemented by technical memoranda to the docket to document those steps in the standards development process not covered within this compilation of background information. The balance of this chapter summarizes the statutory basis for MACT standards and the selection of the source category. Chapter 2 provides an overview of the industry. Chapter 3 discusses the processes in detail and provides estimates of baseline emissions for each process. Emission control technologies and their performance are summarized in Chapter 4. Chapter 5 presents the determination of the MACT floor. Model plants are developed in Chapter 6 (for use in estimating potential impacts), and options for emission control and monitoring are discussed. Environmental and energy impacts are estimated for the model plants and for all plants nationwide in Chapter 7. The estimated costs for emission control and monitoring are given in Chapter 8. Appendix A summarizes the emissions data and Appendix B documents the information used to develop the MACT floor. 1.1 STATUTORY BASIS Section 112 of the CAA requires the development of NESHAP for the control of HAP from both new and existing major or area sources. The statute requires the standard to reflect the maximum degree of reduction in emissions of HAP that is achievable taking into consideration the cost of achieving the emission reduction, any nonair quality health and environmental reduction, and energy requirements. This level of control is commonly referred to as MACT. Emission reductions may be accomplished through application of measures, processes, methods, systems or techniques including, but not limited to: (1) reducing the volume of, or eliminating emissions of, such pollutants through process changes, substitution of materials, or other modifications, (2) enclosing systems or processes to eliminate emissions, (3) collecting, capturing, or treating such pollutants when released from a process, stack, storage or fugitive emissions point, (4) design, equipment, work practice, or operational standards (including requirements for operator training or certification) as provided in subsection (h), or (5) a combination of the above [section 112(d)(2)]. 1-1 ------- 1.2 SELECTION OF SOURCE CATEGORY Section 112 specifically directs the EPA to develop a list of all categories of all major and area sources as appropriate emitting one or more of the HAP listed in section 112(b). The EPA published an initial list of source categories on July 16, 1992 (57 FR 31576) and may amend the list at any time. A schedule for promulgation of standards for each source category was published on December 3, 1993 (58 FR 63941). Integrated iron and steel manufacturing is one of the 174 categories of sources listed. As defined in the EPA report, "Documentation for Developing the Initial Source Category List" (EPA-450/3-91-030), the category consists of plants engaged in producing steel. The source category includes, but is not limited to, the following process units: (1) sinter production, (2) iron production, (3) iron preparation (hot metal desulfurization), (4) steel production, (5) semi-finished product preparation, (6) finished product preparation, and (7) handling and treatment of raw, intermediate, and waste materials. The iron production process includes the production of iron in blast furnaces by the reduction of iron-bearing materials with a hot gas. The steel production process includes BOPF. The listing was based on the Administrator's determination that integrated iron and steel plants may reasonably be anticipated to emit several of the listed HAP in sufficient quantity to be designated as major sources. The EPA schedule for promulgation of the section 112 emission standards requires MACT rules for the integrated iron and steel source category to be promulgated by November 15, 2000. If MACT standards for this source category are not promulgated by May 15, 2002 (18 months following the promulgation deadline), section 112(j) requires States or local agencies with approved permit programs to issue permits or revise existing permits containing either an equivalent emission limitation or an alternative emission limitation for HAP control. 1-2 ------- 2.0 INDUSTRY OVERVIEW The steel industry is composed of two distinct types of facilities: integrated plants and non-integrated plants ("minimills"). A fully integrated facility produces steel from raw materials of coal, iron ore, and scrap. Non-integrated plants do not have all of the equipment to produce steel from coal, iron ore, and scrap on-site. Instead, they purchase their raw materials in a processed form (primarily scrap). This rulemaking includes only the integrated iron and steel industry, which has sinter plants, blast furnaces, and BOPF (see Table 2-1). 2.1 BACKGROUND1 In the past 15 years, the U.S. steel industry has lost over 61 percent of its employees and 58 percent of its facilities. Slow growth in demand for steel, markets lost to other materials, increased imports, and older, less efficient production facilities are largely responsible for the industry's decline. While the integrated steel industry was contracting, minimills more than doubled their capacity in the same period and they continue to expand into new markets. Minimills use EAF to melt scrap and other materials to make steel products. In addition to fundamentally different production technologies, other differences between the integrated steel mills and minimill are also significant. Minimills have narrow product lines and often have small, non-unionized work forces that may receive higher hourly wages than a comparable unionized work force, but without union benefits. Additionally, minimills typically produce much less product per facility (less than 1 million tons of steel per year). Lower scrap prices in the 1960s and 1970s created opportunities for the minimill segment of the market to grow rapidly. Initially, the EAF technology could only be used in the production of low quality long products, such as concrete reinforcing bar. However, minimill products have improved in quality over the years and overcome technological limitations to diversify their product lines. Recently, minimills have entered new markets, such as flat-rolled products; however, more than half of the market for quality steel products still remains beyond minimill capability and is supplied by integrated producers. 2-1 ------- TABLE 2-1. INTEGRATED IRON AND STEEL PLANTS No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Company Acme Steel AK Steel AK Steel Bethlehem Steel Bethlehem Steel Geneva Steel Gulf States Steel Inland Steel LTV Steel LTV Steel National Steel National Steel Rouge Steel USX USX USX USS/Kobe Steel WCI Steel Weirton Steel Wheeling-Pittsburgh Steel City & State Riverdale, IL Ashland, KY Middletown, OH Burns Harbor, IN Sparrows Ft, MD Orem, UT Gadsden, AL East Chicago, IN Cleveland, OH East Chicago, IN Granite City, IL Ecorse, MI Dearborn, MI Braddock, PA Fairfield, AL Gary, IN Lorain, OH Warren, OH Youngstown, OH Weirton, WV Mingo Junction, OH Follansbee, WV Totals BOPF Shops Vessels 2 2 2 3 2 2 2 4 4 2 2 2 2 2 3 6 2 2 2 2 50 Shops 1 1 1 1 1 1 1 2 2 1 1 1 1 1 1 2 1 1 1 1 23 Blast Furnaces 1 1 1 2 1 3 1 3 3 2 2 3 2 2 1 4 2 1 2 2 39 Sinter Plants 1 1 1 1 1 1 1 1 1 9 2-2 ------- 2.2 GEOGRAPHIC DISTRIBUTION1 Figure 2-1 shows the locations of integrated plants that produce iron and steel. The highest geographic concentration of mills is in the Great Lakes region, where most integrated plants are based. According to the Census of Manufactures, 46 percent of steel mills are located in six Great Lakes States: New York, Pennsylvania, Ohio, Indiana, Illinois, and Michigan, with a heavy concentration of steel manufacturing in the Chicago area. Approximately 80 percent of the U.S. steelmaking capacity is in these States. The South is the next largest steel-producing region, although there are only two integrated steel plants. Steel production in the western U.S. is limited to one integrated plant and several minimills. Historically, the mill sites were selected for their proximity to water (tremendous amounts are used for cooling and processing, and for transportation) and the sources of their raw materials, iron ore and coal. Traditional steelmaking regions included the Monongahela River valley near Pittsburgh and along the Mahoning River near Youngstown, Ohio. 2.3 SIZE DISTRIBUTION1 Large, fully-integrated steel mills have declined considerably in the last 15 years, largely due to loss of market share to other materials, competition, and the high cost of pension liabilities. In comparing the 1992 Census of Manufacture data with the data from 1977, these changes are clear. While the number of establishments under SIC 3312 fell by 58 percent from 504 facilities in 1977 to 247 in 1992, the absolute number of integrated mills has always been small, and the reduction is largely due to a drop in the number of small establishments. A more relevant statistic is the reduction in employees during the same time period. The work force for these facilities was dramatically reduced as plants closed or were reorganized by bankruptcy courts. Those that remained open automated and streamlined operations resulting in a 61 percent reduction in the number of production employees over the same 15 year period. Approximately 172,000 were still employed in SIC 3312 establishments in 1992. 2.4 PRODUCT CHARACTERIZATION1 The iron and steel industry produces iron and steel mill products, such as bars, strips, and sheets, as well as formed products such as steel nails, spikes, wire, rods, pipes, and non-steel electrometallurgical products such as ferroalloys. Under SIC 3312, Blast Furnaces and Steel 2-3 ------- FIGURE 2-1. LOCATIONS OF INTEGRATED IRON AND STEEL PLANTS 2-4 ------- Mills, products also include coke and products derived from chemical recovery in the coking process, such as coal tar and distillates. Historically, the automotive and construction sectors have been the two largest steel consuming industries. Consequently, fluctuations in sales and choice of materials in these industries have a significant impact on the iron and steel industry. Over the last two decades, the structure of the steelmaking industry has changed dramatically due to new technologies, foreign competition, and loss of market share to other materials. Many of the large, fully-integrated facilities have closed, and those that are still operating have reduced their workforce, increased automation, and invested in new technologies to remain competitive. 2.5 REFERENCES 1. U.S. Environmental Protection Agency. Profile of the Iron and Steel Industry. EPA Office of Compliance Sector Notebook Project. EPA/310-R-95-005. September 1995. 2-5 ------- 3.0 PROCESS DESCRIPTION AND BASELINE EMISSIONS This chapter provides a brief description of the sintering, ironmaking, and steelmaking processes used at integrated iron and steel plants. Detailed descriptions of these processes are available in "The Making, Shaping, and Treating of Steel1." Emission points, factors affecting emissions, HAP, and the baseline level of emissions are also presented. Emission estimates are based on data submitted by individual companies, tests of sinter plants conducted by EPA, and AP-42 emission factors. 3.1 SINTER PLANTS Sintering is a process that recovers the raw material value of many waste materials generated at iron and steel plants that would otherwise be landfilled or stockpiled. An important function of the sinter plant is to return waste iron-bearing materials to the blast furnace to produce iron. Another function is to provide part or all of the flux material (e.g., limestone, dolomite) for the ironmaking process.1' 2 Feed material to the sintering process includes ore fines, reverts (including blast furnace dust, mill scale, and other byproducts of steelmaking), recycled hot and cold fines from the sintering process, and trim materials (calcite fines, and other supplemental materials needed to produce a sinter product with prescribed chemistry and tonnage). The materials are proportioned and mixed to prepare a chemically uniform feed to the sinter strand, so that the sinter will have qualities desired for satisfactory operation of the blast furnace. The chemical quality of the sinter is often assessed in terms of its basicity, which is the percent total basic oxides divided by the percent total acid oxides ((CaO+MgO)/(SiO2+Al2O3)); sinter basicity is generally 1.0 to 3.0. The relative amounts of each material are determined based on the desired basicity, the rate of consumption of material at the sinter strand, the amount of sinter fines that must be recycled, and the total carbon content needed for proper ignition of the feed material.2 The sintering machine accepts feed material and conveys it down the length of the moving strand. Near the feed end of the grate, the bed is ignited on the surface by gas burners and, as the mixture moves along on the traveling grate, air is pulled down through the mixture to bum the fuel by downdraft combustion; either coke oven gas or natural gas may be used for fuel to ignite the undersize coke or coal in the feed. As the grates move continuously over a series of windboxes toward the discharge end of the strand, the combustion front in the bed moves progressively downward. This creates sufficient heat and temperature to agglomerates the fine 3-1 ------- particles, forming a cake of porous clinker, and providing the strength and other properties needed for use in the blast furnace. The sinter machine strand is composed of pallets which ride on rails over the windboxes. Each pallet has a grated bottom, open ends where the cars come together, and sideboards of maximum height for the sinter bed. The windboxes provide for a controlled distribution of combustion air as it is drawn through the sinter bed. Air is drawn down through the burden, into the windboxes and through an initial separator to a large fan. Very coarse particles are recovered in the windboxes. Other somewhat less coarse particles are removed by the separator. After the fan, the gases are further cleaned before discharge to the atmosphere.2 Each sinter strand generally has 12 to 22 windboxes. The height of the sinter bed varies between 9 and 24 inches. The cake of porous clinker is discharged from the sinter strand to a breaker which reduces the sinter to smaller pieces, generally less than 6 inches in diameter. The crushed product is screened before and/or after cooling; in older plants one or both steps of screening may be absent. Fines and other pieces similar for use as a hearth layer are returned to the feed system.2 The sinter is cooled to below 300* F so that it can be handled on conveyor belts. The sinter product is then transferred to feed areas for the blast furnace. Sinter coolers are often used in conjunction with a water quench and circular or straight line moving beds with forced or induced draft, or they may be quiescent. A portion of the cooling air may be fed to the windbox system to utilize its heat content. The finished product is then ready to be used in the blast furnace feed (burden), along with iron ore pellets, coke, and fluxing agents.1'2 The amount of return fines may fluctuate if the quality of the sinter changes or if the efficiency of screening changes. Some facilities may use a hearth layer, although some older plants do not have the necessary equipment for creating the hearth layer. The amount of flux material varies depending on the percentage of sinter used in the blast furnace burden, the flux requirement of the blast furnace, and other production factors in the ironmaking process. Economics generally favor a high, or super flux sinter.2 There are currently nine sinter plants in operation in the United States. Four of the plants are located in Indiana, with two in Ohio, and one each in Utah, Maryland, and West Virginia. The plants range in capacity from 0.5 to 4.4 million tpy with a total nationwide capacity of 17.6 million tpy. 3.1.1 Emission Points 3-2 ------- The emission points associated with the sinter plant are shown in Figure 3-1. The figure also indicates the typical emission control devices, if any, that have been installed for each emission point. The most significant source of emissions is the windbox, which is controlled either by a baghouse or wet scrubber at each of the nine plants. This emission point is a potential source of organic HAP as well as metal HAP because oil and other organics may be present in the sinter feed material. The other emission points shown in the figure are primarily sources of PM emissions. Emissions from the discharge end of the sintering operation are also controlled at each of the plants (the discharge end emissions points include discharge, crusher, hot screen, cold screen, and the cooler at some plants). Emissions from material storage and handling, mixing, and sinter storage are generally uncontrolled. 3.1.2 Factors Affecting Emissions Several factors can affect the PM emissions, and consequently, the emissions of HAP metals in the PM. For example, PM emissions from the windbox are affected by the amount of fines (e.g., pollution control dust from the steelmaking process) and their particle size distribution; an increase in fines can result in a larger quantity of PM being emitted as well as lower particle sizes in the emissions. The composition of the feed material, such as the amount of manganese and lead, also affects the quantity of these HAP that comprise the PM. Operating parameters, such as the bed air flow rate, bed depth, proper proportioning and mixing of the feed materials, and condition of the grate and machine seals affect the generation of PM from the windboxes.2 3-3 ------- 1. Emissions from raw material storage, handling 4 None 2. Emissions 3. Windbox 4. Discharge fr 9m mixing f None 1 ) emissions 4 Baghouse or scrubber V J emissions 4 Baghouse V. J 5. Crusher emissions Baghouse Raw material storage and handling ^oke A Fines -lux T Slag Ore | Sludge cale Sinter to blast furnace ^^_^^_^ Mixing drum A Air and fuel i Water Sinter ^^^^^ storage "* i None V J \ i 9. Sinter storage emissions Cold screen Sintering L Baghouse V J i Discharge Crusher Fines 1 8. Cold screen emissions l_ Cooler Baghouse V J I 7. Cooler emissions ^ ' Hot screen Baghouse V J \ i 6. Hot screen emissions FIGURE 3-1. SCHEMATIC OF SINTER PLANT EMISSION POINTS AND TYPICAL CONTROLS 3-4 ------- Emissions of hydrocarbons, pyrolysis products, and products of incomplete combustion are also affected by the feed composition, especially the amount of oily material in the feed, as well as by the combustion conditions. Hydrocarbon vapors, identified by a bluish plume, originate from oil in the feed when it is vaporized on the sinter strand ahead of the flame front and is evaporated or pyrolized. The oil in the feed originates from oily mill scale, blast furnace sludge, and coke breeze, which may contain tarry material and oil absorbed from the sump in which it is recovered.2 Emissions from the discharge end, including screening and crushing, are primarily PM and are affected by the amount of fines generated and their composition (i.e., the amount of metal HAP that comprise the PM) and by the ventilation rate that is used. The capture efficiency of the hoods used on the discharge end is a major factor affecting the fugitive emissions from the process. Emissions from the sinter cooler are affected by the quantity of fines in the sinter product being cooled and the type of cooler, whether quiescent, circular, or straight line moving beds, and whether they use forced or induced draft.2 3.1.3 Estimates of Baseline Emissions The major emission points of interest for the sinter plant and those for which data are available are the windbox stack, the discharge end (includes the discharge, crushing, screening, and transfer points, which are usually ducted to a common control device), and the cooler stack. At a few plants, emissions from the cooler are also ducted to the control device used to control emissions from the discharge end. 3.1.3.1 HAP Metal Emissions from the Windbox. Emission test data were obtained from several plants to characterize typical PM emissions from the control device that treats the combustion air and offgases from the sinter plant windboxes. The PM data, when combined with dust analyses for HAP metals, provide one means to estimate potential HAP metal emissions. The PM data summarized in Table 3-1 for the windbox were taken from References 3 through 18. Most of the data were obtained from responses to a survey of the industry (section 114 information collection request) and from test reports provided by individual companies. In addition, EPA conducted tests at two sinter plants in 1997 and measured HAP metal emissions from a plant with a baghouse and one with a scrubber.17'18 3-5 ------- TABLE 3-1. ESTIMATES OF PM EMISSIONS FROM SINTER PLANTS3 Plant AK Steel, Mddletown, OH3' 4 Bethlehem Steel, Bums Harbor, IN5' 6 Bethlehem Steel, Sparrows Point, MD7 Geneva Steel, Provo, UT8' 9 Inland Steel, East Chicago, IN10'11 LTV Steel, East Chicago, IN USX Gary, IN14 WCI Steel, Warren, OH15 Wheeling-Pittsburgh, Follansbee, WV16 Totals Capacity (million tpy) 0.9 2.9 4.0 0.8 1.4 1.9 4.4 0.8 0.5 17.6 Control device Windbox Scrubber Scrubber Scrubber Baghouse Baghouse Scrubber Baghouse Baghouse Scrubber - Discharge Baghouse Baghouse Baghouse Cyclone Baghouse Scrubber Baghouse Baghouse Baghouse - Cooler Baghouse None Cyclone No cooler None None None Baghouse None - PM Emissions (tpy) Windbox 148 247 507 22 60 14217 200e 5.418 116 1,447 Discharge 172 87b 196 8.0 41 7012,13 132b 1.818 8.8 717 Cooler c l,450d 245 -- 700d 950d 2,200d c 250d 5,795 Total 320 1,784 948 30 801 716,380 2,532 7 375 7,959 a Emission estimates without footnotes are as reported in the reference under "Plant." b Based on an emission factor of 0.06 Ib/ton (see text). 0 Included with discharge emissions (common control device). d Based on an emission factor of 1 IbAon (see text). e No emissions data because the plant recently upgraded control to a baghouse; used emission factor of 0.09 Ib/ton based on average factor from Geneva and Inland Steel. 3-6 ------- HAP metals that have been reported in the PM include antimony, arsenic, beryllium, cadmium, chromium, cobalt, lead, manganese, mercury, nickel, and selenium.17> 18 However, manganese (Mn) and lead (Pb) have been the most prevalent by far of the metal HAP; all other metal HAP combined represent less than 1 percent of the quantity of Pb and Mn. Consequently, the focus of the baseline emission estimates will be on Mn and Pb as the HAP metals of interest. The emission estimates for lead and manganese in Table 3-2 that are referenced are the values that were measured and reported by the companies and also include the results of two tests conducted by EPA in 1997. These data were used to develop estimates of Mn and Pb as a percent of PM that could be applied to the other plants to estimate emissions. For example, Bethlehem Steel reported manganese as 0.3 percent of PM7, Inland reported it as 0.8 percent10' n, and the two EPA tests showed a wide range of 0.05 to 3.5 percent.17'18 To estimate Mn emissions from other plants, an average value of 1.2 percent of PM was used. For Pb, Bethlehem reported a value of 2 percent of the PM,5'6 Inland reported 0.3 percent,10' n AK Steel reported 1.3 percent3'4, and EPA measured 2.2 percent at WCI.18 For the estimates in Table 3-2, an average value of 1.5 percent of PM was used to estimate Pb emissions. 3.1.3.2 PAH Emissions from the Windbox. In the two sinter plant tests conducted by EPA in 1997, PAH known as the "7-PAH" and "16-PAH" were analyzed. The 7 PAH include benzo(a)anthracene, benzo(a)pyrene, benzo(b)fluoranthene, benzo(k)fluoranthene, chrysene, dibenzo(a,h)anthracene, indeno(l,2,3- cd)pyrene. The 16-PAH add to this list naphthalene, phenanthrene, pyrene. acenaphthene, acenapthylene, anthracene, benzo(ghi)perylene, fluoranthene, and fluorene. Based on the test results, the emission factors given below were developed for plants with baghouses and scrubbers:17'18 Control device 7-PAH (Ib/ton) 16-PAH (Ib/ton) Baghouse 6.5 xlO'4 1.0 xlO'2 Scrubber 2.4 xlO'5 9.7 xlO'4 3-7 ------- TABLE 3-2. ESTIMATES OF HAP EMISSIONS FROM SINTER PLANTS Plant AK Steel, Middletown, OH Bethlehem Steel, Burns Harbor, IN Bethlehem Steel, Sparrows Point, MD Geneva Steel, Provo, UT Inland Steel, East Chicago, IN LTV Steel, East Chicago, IN USX, Gary, IN WCI Steel, Warren, OH Wheeling-Pittsburgh, Follansbee, WV Totals HAP emissions (tpy) Windbox Mna 0.133-4 3.0 1.57 0.03s 0.510'11 0.0617 2.4 O.I18 1.4 9.12 Pbb 2.6s'4 495,6 7.6 0.03s 0.210-11 1417 3.0 0.0818 1.7 34.1 Volatiles0 9.9 32 44 8.8 15 21 48 8.8 5.5 194 7-PAH" 0.01 0.03 0.05 0.3 0.5 0.0117 1.4 0.3 0.006 2.6 16-PAH* 0.4 1.4 1.9 4.0 7.0 0.617 22 4.218 0.2 42 Discharge Mnf 0.533-4 0.7 0.67 0.00s 0.510-11 0.712- 13 1.0 0.0718 0.07 4.2 Coole r Mnf g 11 0.77 h 5.3 7.1 16.5 g 1.9 42 Total 14 53 56 13 29 43 94 14 11 327 ' Based on 1.2 percent of PM emissions in Table 3-1 unless specific reference is given. b Based on 1.5 percent of PM emissions in Table 3-1 unless specific reference is given. 0 Based on an emission factor of 0.022 Ib/t (see text). d Based on emission factors of 6.5 x 10"4 and 2.4 x 10"5 Ib/t for baghouses and scrubbers, respectively. e Based on emission factors of 1.0 x 10"2 and 9.7 x 10"4 Ib/t for baghouses and scrubbers, respectively. f Based on 0.75 percent of PM emissions in Table 3-1 unless specific reference is given. B Combined with discharge emissions (common control device). h No cooler. ------- 3.1.3.3 Volatile Organic HAP Emissions from the Windbox. Several plants reported emissions of VOC from the windbox, some of which are HAP. For example, a test report provided by Inland Steel reported 0.39 Ib/ton of non-methane hydrocarbons (expressed as propane)20, and another test report provided by LTV Steel reported 167 Ib/hr of VOC (expressed as carbon).21 For a typical production rate of 416 ton/hr of sinter12'13, the emission factor would be 0.40 Ib/ton. Based on an emission factor of 0.39 to 0.40 Ib VOC/ton, VOC emissions from the windbox for the individual plants would range from 100 to 880 tpy with a nationwide total of 3,800 tpy. Speciated data for volatile HAP were provided by Bethlehem Steel22: Compound Ib/yr Benzene 25,283 Carbon disulfide 21,507 Toluene 10,015 Xylene 4.186 Total 64,971 Chloromethane, ethyl benzene, and styrene were also reported at much lower levels, about 10 percent of the quantity of toluene that was measured. For an annual production rate of 2.92 million tons of sinter in 19925'6, these HAP compounds were emitted at a rate of 0.022 Ib/ton. This emission factor was applied to the other plants to estimate volatile HAP emissions. 3.1.3.4 Emissions of D/F from the Windbox. Testing was conducted for D/F by EPA at two sinter plants.17'18 The results expressed as TEQ (total equivalent to 2,3,7,8-TCDD) were 0.7 g/yr (5.5 x 10'7 g/ton of sinter) for the plant with a wet scrubber and 2.8 g/yr (3.4 x 10"6 g/ton of sinter) for the plant with a baghouse. Based on a nationwide capacity of 10.2 million tpy for plants with scrubbers and 7 million tpy for plants with baghouses, the nationwide estimate of TEQ from sinter plants is 29 g/yr. 3.1.3.5 Emissions from the Discharge End. The only significant HAP reported by the companies in emissions from the discharge end were metals, and Mn was by far the most prevalent. Test data for PM were obtained from several plants (those not marked with a footnote in Table 3-1). The data from the reporting plants were used to derive a PM emission factor for the other plants: 5-9 ------- Plant Control PM (Ib/ton) LTV12'13 scrubber 0.074 Weirton19 scrubber 0.041 WCI18 baghouse 0.0044 Inland10'n baghouse 0.082 Bethlehem7 baghouse 0.098 Wheeling-Pittsburgh16 baghouse 0.035 Average 0.06 Other than the very low results for WCI, there are no obvious differences in the emission factors for scrubbers and baghouses; consequently, the average value of 0.06 Ib/ton was used to estimate PM emissions from the discharge end for the other plants. Dust analyses provided by Bethlehem Steel showed Mn to be 0.3 percent of PM for discharge emissions,7 and similar data from Inland revealed a value of 1.2 percent.10' n A midrange value of 0.75 percent of PM was used to estimate Mn emissions for the other plants. 3.1.3.6 Emissions from the Cooler. Test data were available for 2 tests conducted at USS Gary Works for an uncontrolled cooler in October 1979 and December 1987. The results showed a concentration of 0.033 gr/dscf, 147 Ib/hr, and 518,700 dscfm. The resulting emission factor is about 1 Ib/ton. Test data were also available from Bethlehem's Sparrows Point plant for a cooler controlled by a cyclone with a rated efficiency of 90 percent. The cyclone achieves an outlet concentration of 0.02 gr/dscf with a flow of 640,000 acfm at 600* F with a resulting emission factor of about 0.12 Ib/ton.7 Assuming 90 percent control, the uncontrolled emission factor would be 1.2 Ib/ton. For the estimates presented in this section, an uncontrolled emission factor of 1 Ib/ton is used. This emission factor likely represents coolers with very high flow rates of air through the bed of hot sinter. If other plants use lower flow rates or quiescent coolers, the uncontrolled emissions may be much lower than 1 Ib/ton. This factor was coupled with the concentration derived for Mn from the discharge end (0.75 percent of PM) because the composition of the sinter dust from the discharge end and cooler should be about the same. Consequently, the estimates of Mn emissions from the cooler for the plants without controls are based on PM emissions of 1 Ib/ton and 0.75 percent Mn in the PM. 3-10 ------- 3.1.4 Uncertainties in the Emission Estimates A major uncertainty in the emission estimates is the quantity of emissions that are not captured and escape as fugitive emissions with the ventilation air. The plants reported measured emissions from point sources, which were the stacks from which emissions from the control device were discharged. However, the capture efficiency of hoods used on several emission points associated with the discharge end was reported as about 95 percent,7 which means that the quantity that was not captured was far more than the quantity emitted from control devices that were generally rated as 99 to 99.9 percent efficient in the control of PM. Some of the larger particles may settle out in the building, and other PM that escapes capture is emitted with the ventilation air to the atmosphere. Uncertainty is also introduced by differences in the composition of the feed materials used by the plants. The percent of Pb and Mn in the dust may be directly related to the amount of these metals in the feed materials. In addition, some of the more volatile metal compounds may be more concentrated in fine particles (i.e., the concentration of HAP metals may vary as a function of particle size). The quantity and type of organics in the feed material (such as oily scale), may also affect the type and quantity of organic compounds that are emitted. 3-11 ------- 3.2 BLASTFURNACES The blast furnace converts iron oxide into molten iron for subsequent refining in the BOPF shop to produce steel. A typical burden (feed) may consist of iron ore, pellets, sinter, limestone, coke, mill scale, BOPF slag, and other iron bearing materials. The burden material is charged into the top of the furnace and slowly descends through the furnace. The coke provides the thermal energy required for the process and provides carbon to reduce the iron oxide and to remove oxygen in the form of CO. The blast furnace is a vertical shaft furnace. Raw materials are charged into the top of the furnace and fall to the top of the burden of raw materials already in the furnace. As they descend in the furnace, they are heated by a countercurrent flow of gas. Heated air is injected through the tuyeres, located near the bottom of the furnace just above the hearth. The air moves countercurrent to the burden, consuming the coke (carbon). Raw materials are introduced at the top of the blast furnace; the hottest temperature zone in the furnace is at the hearth level, where the burden is molten. The furnace filling is controlled by the level of burden in the furnace. When the level is below a preset point, the stockhouse functions continuously, filling the skips with predetermined weights of materials in the ordered sequence. The top of the blast furnace is enclosed so that blast furnace gas can be drawn off above the stock level and a bell and hopper arrangement can be used for charging the furnace. Most installations use a combination of two bells so that a gas tight space can be provided between the two bells to prevent gas from escaping while the lower bell is opened. Raw materials are taken to the furnace top by a skip hoist or a conveyor belt and dropped into the upper hopper. With the large bell closed, the small bell is lowered and the charge material is dropped into the large-bell hopper. When the large-bell hopper is full, the small bell is held closed, the large bell is lowered, and the material is dumped into the blast furnace without allowing any of the gas to escape. A more recent innovation, used on several blast furnaces in the industry, is the Paul Wurth bell-less top, in which the charge materials are deposited into hoppers located at the top of the furnace. The hoppers can be depressurized for loading and repressurized for discharging the material into the furnace. There are at least two hoppers so that while one is being loaded, the other can be discharged into the furnace. As the charge material enters the furnace, it is directed by a rotating chute to various locations on top of the stockline.1 3-12 ------- With this design, the furnace burns fuel more efficiently, leaks less, and can hold pressure. There is also not a problem with wearing a hole in the bell or sealing bell rods.1 In the blast furnace process, the heated raw materials react chemically with one another. The principal set of reactions are the complex ones between coke, air, and iron ore. Part of the coke is consumed by the oxygen in the air to produce heat for the process. Another part of the coke combines with the oxygen in the iron ore and releases free iron, which melts, drips to the bottom of the furnace, and collects in the hearth. A final portion of the carbon dissolves in the iron. The heat of the blast furnace serves to calcine the limestone. The resulting calcium oxide reacts with the impurities in the ore, principally sulfur, and, in molten form, descends to the hearth. The slag, being about one-third the density of the iron, floats in a separate layer on the iron bath. Ironmaking is a continuous process within the blast furnace; however, it is a semi-continuous process with respect to periodic charging of materials into the top of the furnace and periodic tapping of molten iron and slag from the bottom of the furnace. Periodically, the hearth becomes full of molten iron and slag. Because there is a limit to the amount that can be tolerated before it interferes with the furnace operation, they must be removed from the furnace at regular intervals. The iron notch, which is used for tapping the hot metal, is located just above the floor of the hearth; each furnace has one or more iron notches When the furnace is in operation, the iron notch is completely filled with a refractory material, called taphole clay. To cast the hot metal from the furnace, a tapping hole is drilled through this material. The hot metal flows through this hole and is discharged into a trough, which is a long narrow basin typically 3 to 5 feet wide and 26 to 40 feet long; the trough generally has a slightly sloping bottom away from the furnace. At the far end of the trough, there is a dam to hold back the hot metal until the depth of the metal in the trough is sufficient to contact the bottom of a refractory skimmer block. The skimmer holds back the slag and diverts it into the slag runners. The hot metal flows over the dam and down the iron runner, where it is directed in sequence to a train of ladles positioned under stationary spouts along the runner. At several large blast furnaces, a tilting spout is used, positioned between two hot metal tracks. The spout is first tilted to fill the ladle on one track and then to fill the one on the other track. While the second ladle is being filled, the first one can be replaced with an empty one so that the cast can be continued uninterrupted while several ladles are filled. 3-13 ------- After the flows of iron and slag cease, the tap hole is plugged with fresh clay by a device called a "mud gun", and the ironmaking process resumes. The hot metal is transported from the blast furnace to the BOPF shop in refractory-lined ladles that have a course of insulating material between the lining and the steel shell.23 Blast furnace gas (primarily CO) is collected from offtakes at the top of the furnace; this gas is cleaned of PM and is used to fire the blast furnace stoves that heat the furnace air. Excess blast furnace gas is used as a fuel in other processes at the plant. There are currently a total of 39 blast furnaces at 20 plants that are owned by 14 companies in the U.S. The plants are located in 10 different States, with the largest number in Ohio and Indiana. Each furnace has the capacity to produce 700,000 to 3,440,000 tpy of hot metal. 3.2.1 Blast Furnace Auxiliaries 3.2.1.1 Stoves. About 30 percent of the blast furnace gas is utilized to heat the hot air blast by means of the blast furnace stoves; there are generally 3 to 4 stoves per blast furnace. The remainder is used for other heating purposes throughout the facility. Before the blast air is delivered to the blast furnace tuyeres, it is preheated by passing it through regenerative stoves that are heated primarily by combustion of the blast furnace off-gas. In this way, some of the energy of the off-gas that would otherwise have been lost from the process is returned to the process. The additional thermal energy returned to the blast furnace as heat decreases the amount of fuel that has to be burned for each unit of hot metal and thus improves the efficiency of the process. In many furnaces, the off-gas is enriched by the addition of a fuel with much higher calorific value, such as natural gas or coke oven gas, to obtain even higher hot blast temperatures. This decreases the fuel requirement and increases the hot metal production rate to a greater extent than is possible when burning off gas alone to heat the stoves. 3.2.1.2 Blast Furnace Gas Cleaning. As the blast furnace gas leaves the top of the furnace, it contains dust particles varying in size from about 6 millimeters to a few microns. The dust that is carried out of the top, referred to as flue dust, is made up of fine particles of coke and burden material and extremely fine particles of chemical compounds formed from reactions within the blast furnace. Before the blast furnace gas can be burned in either the hot blast stoves or the boiler house, it must be cleaned to remove most of the flue 3-14 ------- dust and prevent plugging and damaging of the checkers or burners and to keep the dust from being discharged into the atmosphere with the products of combustion. The gas normally passes through a dry dustcatcher, where the coarser particles are removed, and then through a wet-cleaning system, where the very fine particles are scrubbed from the gas with water. 3.2.1.3 Pulverized Coal Injection. At least six facilities in the industry have installed pulverized coal injection systems to replace some of the coke required for the blast furnace. Coal injection systems are much less costly than building new coke batteries and have fewer environmental problems, as temperatures are not high enough to liberate any problem elements in the coal. As much as 40 percent of the furnace coke can be replaced on a one-for-one basis with coal. The quantity of coal that can be used is affected by quality of the coke and is also limited by the amount of oxygen available at the tuyeres. In preparing coal for injection, the first step is a grinding or pulverizing operation; most systems take the coal down to 80 percent-200 mesh. The coal is stored under a controlled atmosphere, brought up to furnace pressure in feed tanks, and pneumatically conveyed in a single pipe to the blast furnace area. The coal-air mixture is then divided in a static distributor for delivery to each pipe by way of individual pipes.24 Uniform distribution to the furnace tuyeres is critical. At the tuyeres, fine coal meets the hot blast at around 2,000 • F. The obj ect of coal inj ection is to get the particles broken down to atoms of carbon and combusted with oxygen before the end of the raceway, the combustion zone in front of the tuyeres. Coal injection has a positive effect on blast furnace operations. The flame temperature can be more effectively controlled and there is an indication that slips occur less frequently.25 3.2.2 Emission Points and Factors Affecting Emissions A schematic of the emission points is given in Figure 3-2 and described in this section. The major emissions of interest occur from the casthouse during tapping when molten iron and slag are removed from the furnace. Emissions occur at the taphole, from the trough, from the runners that transport the iron and slag, and from the ladle that receives the molten iron. These emissions include flakes of graphite (carbon) called "kish" that is released as the metal cools (because the solubility of carbon in the metal decreases as it cools) and metal oxides that form when the reduced metal (e.g., iron, manganese) reacts with oxygen in the air.23 Factors 3-15 ------- affecting these emissions include the duration of tapping, exposed surface area of metal and slag, length of runners, and the presence/absence of runner covers and flame suppression, which reduce contact with air. Gaseous and particulate emissions occur from slag handling as the slag is discharged and allowed to cool. Particulate emissions also occur when the solidified slag is later broken up and removed. These emissions are generally uncontrolled. Emissions from raw material handling occur from the storage, sizing, screening, mixing, and transport of the feed materials that comprise the blast furnace burden. These raw materials that generate dust include iron ore, pellets, sinter, coke, and flux materials such as limestone and silica23 Emissions are affected by the extent to which fine particles are generated, use of enclosures and extent of exposure to the atmosphere, use of water sprays or other materials for suppression, etc. The gas leaving the blast furnace is primarily CO and nitrogen and is heavily laden with PM. The gas is cleaned and is used as fuel in the blast furnace stoves and other operations at the plant. Emissions occur from the stove stack when this gas is burned. The quantity and composition of these emissions are affected by the amount and type of particles remaining after cleaning and the combustion conditions when the fuel is burned. 3-16 ------- 1. Emissions matei , i from raw 2. Emissions from •ials furnace slips i A ( None j ( None i Raw material storage and handling ^ ^ Ore, pellets Coke Flux EOF slag Sinter ^ ' Slag transfer, disposal C None J \ Feed ^ Blast furnace ei TT Coal Slag Hot 1 metal i ' Casting I operations BFgas F-» 1 1 r— -1 i Coal drier ^ 1 i i Hot metal to ( Baghouse steelmaking f Suppression; ^ covered runners; \^ baghouse ^/ Dust t . Clean BFgas r ~^ • ~ • • Wastewater 1 1 ^ ' Blast furnace -^^ — ' Wastewater 1 1 i Hot air I 1 | . | Coal j Sludge | i i J ( None J ( None J I I | | | ! ! ! j 5. Emissions from 4. Stove stack BF gas to 3. Emissions from ' coal drier emissions other operations wastewater treatment 7. Emissions from slag transfer, disposal 6. Casting emissions FIGURE 3-2. SCHEMATIC OF BLAST FURNACE EMISSION POINTS AND TYPICAL CONTROLS 3-17 ------- Emissions also occur from furnace "slips." A slip occurs when the burden material hangs or bridges in the furnace rather than continuing its downward movement. When this happens, the solid material below the "hang" continues to move downward and form a void below the hang that is filled with hot gas at very high pressure. When the hang finally collapses, the sudden downward thrust of the burden material forces the hot gas upward with the force of an explosion. To prevent damage to the furnace, the pressure is relieved through bleeder stacks on top of the furnace that discharge the particle-laden gas directly to the atmosphere.1 Factors that are believed to contribute to blast furnace slips include re-solidification of previously fused slag and molten iron, an excessive quantity of fines in the coke, alkalis such as oxides of sodium and potassium, and overblowing of the furnace (excess air).23 One plant reported that slips were very infrequent now because they used pellets rather than iron ore.19 Older blast furnaces are reported to experience more slips than are newer furnaces.23 The quantity of emissions from slips is related to the duration of the slips, their frequency, how fast the pressure rises, and how quickly it is relieved. Emissions are also discharged uncontrolled to the atmosphere during a practice known as "back drafting." Back drafting occurs when it is necessary to take the furnace out of blast for a short period of time (generally less than 2 hours) to perform maintenance. The blast air is stopped, the bleeders are opened to pull some of the furnace gas out of the top, and gas is also drawn back through the tuyeres to a hot stove where it is burned and discharged through the stove stack. Some plants use a back-draft stack to discharge the gas rather than drawing the gas back through a stove.1'23 Only one plant reported their level of emissions from back drafting (200 tpy of PM).8 No other information was available on the frequency of back drafting or the level of emissions. Emissions also occur from the wastewater collection and treatment system The blast furnace gas is heavily laden with particles (on the order of 30 g/scm) as it leaves the furnace. The gas is cleaned by passing it through a cyclone (called a dust catcher) and then directing it to venturi scrubbers for final cleaning. The direct contact water used in the scrubber dissolves HCN from the gas, and the HCN is subsequently stripped from the water when it passes through the cooling tower. 3.2.3 Estimates of Baseline Emissions 3-18 ------- The approach used in this section to estimate baseline emissions relies on estimates submitted by the individual companies and emission factors in EPA's AP-42 emission factor document.26 For metal HAP, Mn was the only HAP metal reported by most companies. Estimates of PM emissions are used with analyses of the dust for metals (expressed as percent Mn) to estimate Mn emissions. The only other HAP identified from blast furnace operations was HCN. HCN was measured and reported by two plants as being emitted from the blast furnace water.7'27 The estimates of HCN emissions from these two plants are applied to other plants to estimate HCN emissions. 3.2.3.1 Casthouse PM Emissions. The emission estimates for PM submitted by the companies are summarized in Table 3-3. The table indicates there was considerable variability in the emission factors used by the different plants even though similar controls are in place. For example, five plants that use flame suppression and covered runners used an emission factor of 0.3 to 0.6 Ib/ton, while other plants used lower values. The AP-42 emission factor for the casthouse roof monitor is 0.6 Ib/ton or 0.3 Ib/ton for the taphole and trough only. If local evacuation is used, an emission factor (prior to any control device) of 1.3 Ib/ton is recommended.26 Because 5 of the 11 plants in Table 3-4 that use suppression controls (e.g., flame suppression, covered runners) used the factor of 0.3 to 0.6 Ib/ton to estimate their emissions and it is consistent with the AP-42 number, a value of 0.6 Ib/ton (from AP-42) is used in this section to estimate emissions for plants without hoods and baghouses. The emission factors used by seven plants with hoods to capture the emissions and a control device (baghouse or a scrubber) to remove PM are also shown in Table 3-4. These factors range from 0.01 to 0.1 Ib/ton and average about 0.05 Ib/ton. Consequently, emissions from casthouse operations that use hoods to capture emissions and direct them to a control device will be estimated as 0.05 Ib/ton. 3.2.3.2 Miscellaneous Emission Points. The PM emission estimates provided for raw material handling are given in Table 3-4 and show a range of 0.0086 to 0.1 Ib/ton with an average of 0.04 Ib/ton, which will be used in this section to estimate baseline emissions. 3-19 ------- TABLE 3-3. CASTHOUSE PM EMISSION ESTIMATES SUBMITTED BY THE COMPANIES Plant (State) Acme Steel (IN)28'29 AK Steel (KY)30'31 AK Steel (OH)3-4 Bethlehem Steel (IN)5'6 Geneva Steel (UT)8 LTV Steel(OH)32 USX(AL)33'34 Rouge Steel (MI)35 USS/Kobe No. 436 LTV Steel H3 (IN)12'13 Weirton Steel (WV)19 Wheeling-Pitt No. 1 (WV)16 Control FS,CR FS,CR FS,CR FS,CR FS,CR FS,CR FS,CR FS,CR FS,CR FS,CR FS,CR FS,CR Production (106 tpy) 0.8 1.92 2.2 5.5 2.7 4.1 1.9 2.7 1.0 1.6 2.1 0.7 PM emissions (tpy) 53 288 330 1,250 189 147 570 122 300 48 80 14 Emission factor (Ib/ton) 0.1 0.3 0.3 0.5 0.1 0.07 0.6 0.09 0.6 0.06 0.08 0.04 Plants with local hoods vented to a control device Bethlehem Steel (MD)7 Inland Steel No.7 (IN)10'11 Inland Steel Nos. 5,6 (IN)10' n National Steel (X)37'38 USS/Kobe No. 336 USX (PA)33'34 Wheeling-Pitt 5,6 (OH)16 FS, ECR, Hood, BH ECR, Hood, BH ECR, Hood, Scrubber CR,Hood,BH ECR, Hood, BH FS, AC, BH FS, CR, AC, Hood, BH 3.5 4.0 2.5 2.4 1.3 22 1.6 208 146 48 94 57 29 8.7 0.1 0.07 0.04 0.08 0.087 0.03 0.01 FS = flame suppression, usually for covered runners and sometimes at the taphole. CR = covered runners. ECR = evacuated covered runners (vented to a control device). Hood = local hoods used to capture emissions at the tap hole and trough, and sometimes from the torpedo car, and subsequently ducted to a control device. BH = baghouse. AC = an air curtain that is used to contain emissions within a limited area of the casthouse where they are captured by a hood and sent to a control device. 3-20 ------- TABLE 3-4. RAW MATERIAL HANDLING, SLAG HANDLING,AND FURNACE SLIP PM EMISSION ESTIMATES SUBMITTED BY THE COMPANIES Plant Production (106tpy) PM emissions (tpy) Emission factor (Ib/ton) Raw material handling Acme Steel28'29 Geneva Steel8 Inland No. 710'11 USX (AL)33' 34 USS Kobe36 Weirton Steel19 Wheeling-Pitt No. I16 Wheeling-Pitt No. 5,616 0.8 2.7 4.0 1.9 2.3 2.1 0.7 1.6 30 39 45 71 118 16 O 15 Average 0.075 0.029 0.023 0.075 0.1 0.015 0.0086 0.019 0.04 Slag handling Acme Steel28'29 AK Steel (KY)30'31 National (IL)37'38 USX (AL)33' 34 USS Kobe36 Weirton Steel19 0.8 1.92 2.4 1.9 2.3 2.1 1.6 53 4.7 3.2 80.9 3.4 0.0040 0.056 0.004 0.0034 0.07 0.0032 Slips Acme Steel28'29 AK Steel (KY)30'31 Geneva Steel8 USX (AL)33' 34 0.8 1.92 2.7 1.9 1 295 69 20 0.0025 0.31 0.051 0.021 3-21 ------- 3-22 ------- The emission estimates for slag handling are also shown in Table 3-4 and range from 0.003 to 0.07 Ib/ton. The AP-42 estimate for emissions from slag handling (using a front-end loader) is 0.026 lb/ton.26 This value is used to provide a highly uncertain estimate of fugitive emissions from slag handling. Three plants provided estimates of PM emissions from furnace slips that were in the range of 0.0025 to 0.31 lb/ton (see Table 3-4), which spans a range of 2 orders of magnitude. Emissions from slips are highly variable and difficult to estimate. A median value between the extremes of 0.03 lb/ton is used to provide a highly uncertain estimate of emissions from slips. The emissions of PM from the blast furnace stoves were provided by several plants and are summarized in Table 3-5. The average value of 0.056 lb/ton will be used to estimate PM emissions from the blast furnace stove. These emission factors are applied to each emission point at each plant in Table 3-6 to estimate total PM emissions from blast furnace operations. The use of consistent emission factors for plants with similar controls should provide a better relative comparison among plants than the use of site-specific emission factors of unknown origin. 3-23 ------- TABLE 3-5. BLAST FURNACE STOVE PM EMISSION ESTIMATES SUBMITTED BY THE COMPANIES Plant (State) Acme Steel (IN)28' 29 AK Steel (KY)30'31 Geneva Steel (UT)8 LTV Steel(OH)32 LTV Steel H3 (IN)12' 13 LTV Steel H4 (IN)12' 13 National (IL)37' 38 Rouge Steel (MI)35 USX (PA)33' 34 USX (AL)33' 34 USS Kobe (OH)36 Weirton Steel (WV)19 Wheeling-Pitt No. 1 (WV)16 Wheeling-Pitt No. 5,6 (WV)16 Production (106 tpy) 0.8 1.9 2.7 4.1 1.6 2.0 2.4 2.7 2.2 1.9 2.3 2.1 0.7 1.6 PM emissions (tpy) 26.1 50 30 152 36 48 13 5.8 20 161 18.7 61 21 84 Average Emission factor (Ib/ton) 0.052 0.05 0.022 0.074 0.045 0.048 0.011 0.0043 0.018 0.17 0.016 0.058 0.060 0.11 0.05 3-24 ------- TABLE 3-6. ESTIMATES OF PM EMISSIONS FROM BLAST FURNACE OPERATIONS Plant Acme Steel, Riverdale, IL28 29 AK Steel, Ashland, KY3"1 AK Steel, Middletown, OH14 Bethlehem Steel, Burns Harbor, IN Bethlehem Steel, Sparrows Point, MD Geneva Steel, Orem, UT Gulf States Steel, Gadsden, AL Inland Steel, East Chicago, IN LTV Steel, Cleveland, OH LTV Steel, East Chicago, IN Furnace 1 A 3 C,D L* 1,2,3 2 7 5, 6 C1,C5 C6 H3 Capacit y (millio n tpy) 1.0 1.9 2.2 5.5 3.5 2.7 1.2 4.0 2.5 2.7 1.4 1.6 Casthouse control Flame suppression and covered runners Covered runners with flame suppression Flame suppression at taphole, torpedo car Flame suppression at taphole, covered runners, N2 over torpedo car Hood over tapping, evacuated runner covers, both to baghouse; flame suppression at torpedo car Partially covered runners with flame suppression No controls Hood over tapping, evacuated runner covers, both to baghouse Hood over tapping, evacuated runner covers, both to scrubber Covered runners with flame suppression Covered runners with flame suppression; fume suppression hoods for tapping Covered runners with flame suppression Particulate matter emissions (tpy) Casthouse 300 570 660 1,650 88 810 360 100 63 810 420 480 Raw materials 20 38 44 110 70 54 24 80 50 54 28 32 Slips 15 29 33 83 53 41 18 60 38 41 21 24 Stoves 25 48 55 138 88 68 30 100 63 68 35 40 Slag 13 25 29 72 46 35 16 52 33 35 18 21 Total 373 709 821 2,052 343 1,007 448 392 245 1,007 522 597 3-25 ------- TABLE 3-6. ESTIMATES OF PM EMISSIONS FROM BLAST FURNACE OPERATIONS Plant National Steel, Granite City, IL National Steel, Ecorse MI Rouge Steel, Dearborn, MI USX, Braddock, PA USX, Fairfield, AL USX, Gary, IN USX, Gary, IN USS/Kobe Steel, Lorain, OH WCI Steel, Warren, OH Weirton Steel, Weirton, WV Wheeling Pittsburgh Steel, Mingo Junction, OH Furnace H4 A,B A,B,D B, C 1, 3 8 4,6,8 13 3 4 1 1,3 IN Capacit y (millio n tpy) 2.0 2.4 4.9 2.7 2.3 2.0 3.4 2.7 1.3 1.0 1.6 2.1 0.7 Casthouse control Covered runners with flame suppression; hood over tapping and tilting spout to baghouse Hoods over tapping and torpedo car to baghouse; covered runners Hoods over tapping and tilting spout to baghouse Flame suppression at taphole & at covered runners Air curtain at tapping to baghouse, flame suppression Covered runners with flame suppression Flame suppression Hood and evacuated covered runners to baghouse Hood over tapping to baghouse, covered runners Flame suppression Hoods over tapping and pouring station to baghouse Covered runners with flame suppression for iron & slag Partially covered runners with flame suppression Particulate matter emissions (tpy) Casthouse 50 60 123 810 57 600 1,020 68 33 300 40 630 210 Raw materials 40 48 98 54 46 40 68 54 26 20 32 42 14 Slips 30 36 74 41 35 30 51 41 20 15 24 32 11 Stoves 50 60 137 76 64 56 95 76 36 28 45 59 20 Slag 26 31 64 35 30 26 44 35 17 13 21 27 9 Total 196 235 495 1,015 232 752 1,278 273 131 376 162 790 263 3-26 ------- TABLE 3-6. ESTIMATES OF PM EMISSIONS FROM BLAST FURNACE OPERATIONS Plant Furnace 5S Totals Capacit y (millio n tpy) 1.6 60.9 Casthouse control Hood in roof (with air curtain) for tapping, hood over torpedo car, both to baghouse; covered runners with flame suppression Particulate matter emissions (tpy) Casthouse 40 10,350 Raw materials 32 1,218 Slips 24 914 Stoves 45 1,601 Slag 21 792 Total 162 14,875 * H and J are operated only when L is down for a re-line. 3-27 ------- 3.2.3.3 Estimates of Mn Emissions. Several plants provided data from dust analyses to estimate Mg emissions from the blast furnace casthouse. A few plants reported the percent Mn in the dust from other operations. These data are summarized in Table 3-7. The table indicates that Mn was reported to range from 0.1 to 1.7 percent of PM with an average of 0.6 percent, which is the value used in Table 3-8 to estimate Mn emissions from the casthouse. This same value (0.6 percent) was also used to estimate Mn emissions from raw material handling, slag, and slips because the few data points for these sources are relatively consistent with the range seen for casthouse dust. A value of 0.2 percent Mn was used to estimate Mn emissions from blast furnace stoves (Table 3-8). Total Mn emissions from blast furnace operations are given in Table 3-8. 3.2.3.4 Estimates of HCN Emissions. Emissions of HCN were reported for the cooling tower used for the blast furnace scrubber water. Two plants calculated the amount of HCN stripped from the water in the cooling tower by analyzing the water for HCN concentration before and after cooling and measuring the wastewater flow rate. The decrease in HCN concentration times the water flow rate provides a measure of the HCN that was emitted. One plant reported the average value from several measurements as resulting in 51 tpy of HCN emissions and an emission factor based on iron production of 0.035 lb/ton.7 Another plant reported a range of 40 to 70 tpy of HCN emissions based on their measurements.27 For a typical production rate of 1.15 million tpy, this range results in an HCN emission factor of 0.07 to 0.12 lb/ton. The emission estimates for HCN presented in Table 3-9 are based on a midrange value 0.08 lb/ton. The Bethlehem Burns Harbor plant reported that there was essentially no HCN in their scrubber water and provided data from samples taken and analyzed by EPA. Differences in the furnace design and operation such as furnace top temperatures and pressures may explain why HCN is generated in some blast furnace operations and not in others. The Burns Harbor plant has indicated that HCN can be produced at certain times during the startup or shutdown of the blast furnace for a reline. No HCN emissions were estimated for the Burns Harbor plant to reflect the normal steady-state operation. If HCN is produced in the blast furnace operation, it will remain in the scrubber water under alkaline conditions because it will be in an ionized form. Under acidic conditions, HCN is in an un-ionized form and is stripped from the scrubber water as it goes through a cooling tower. The pH of blast furnace water systems is 3-28 ------- controlled at different levels depending on the corrosiveness and fouling potential of the system. A system that is either too basic or too acidic can result in damage to equipment or piping. 3.2.4 Uncertainties in the Emission Estimates There is inherent uncertainty in the estimates of emissions from blast furnace operations because of their fugitive nature. The limited data on casthouse emissions apparently were developed from tests in other countries for casthouses that were entirely evacuated to a control device. In addition, there are few data available on the effectiveness of covers and flame suppression or on the capture efficiency of local hoods that are used at some plants. Another uncertainty is the variation in the Mn content in the iron, which may be affected by the Mn content of the iron ore or other materials. One plant reported a significant decrease in Mn content, which would mean the Mn emitted with the PM may be less than the estimates provided here. Uncertainly is also introduced for the HCN emission estimates because data were available for only three plants. Data on the HCN concentration in the wastewater entering and leaving the scrubber for other plants would improve the HCN emission estimates. 3-29 ------- TABLE 3-7. Mn DATA PROVIDED BY THE COMPANIES Source Casthouse Raw material handling Slag Stove Slips Plant Acme Steel28'29 AK Steel (KY)30'31 BSC (MD)7 Geneva Steel8 Inland No. 710'11 Inland Nos. 5, 610' n LTVH312'13 LTVH412'13 National37' 38 Rouge Steel35 USX (PA)33' 34 USX (AL)33' 34 USS Kobe (OH)36 Average Acme Steel28'29 Weirton Steel19 National Steel37' 38 National Steel37' 38 USX (AL)33' 34 Geneva Steel8 Percent Mn in PM 0.81 1.7 0.29 0.25 0.43 1.2 0.55 0.52 0.14 0.13 0.52 0.11 0.3 0.6 0.60 1.2 0.64 0.25 0.20 0.25 3-30 ------- TABLE 3-8. ESTIMATES OF Mn EMISSIONS FROM BLAST FURNACE OPERATIONS Plant Acme Steel, Riverdale, IL AK Steel, Ashland, KY30- 31 AK Steel, Middletown, OH3' Bethlehem Steel, Burns Harbor, IN Bethlehem Steel, Sparrows Point, MD Geneva Steel, Orem, UT Gulf States Steel, Gadsden, AL Inland Steel, East Chicago, IN LTV Steel, Cleveland, OH LTV Steel, East Chicago, IN National Steel, Granite City, IL National Steel, Ecorse MI Rouge Steel, Dearborn, MI USX, Braddock, PA Furnace 1 A 3 C,D L 1,2,3 2 7 5, 6 C1,C5 C6 H3 H4 A,B A,B,D B, C 1, 3 Capacity (million tpy) 1.0 1.9 2.2 5.5 3.5 2.7 1.2 4.0 2.5 2.7 1.4 1.6 2.0 2.4 4.9 2.7 2.3 Mn emissions (tpy) Casthouse 1.8 3.4 4.0 9.9 0.5 4.9 2.2 0.6 0.4 4.9 2.5 2.9 0.3 0.4 0.7 4.9 0.3 Raw materials 0.12 0.23 0.26 0.66 0.42 0.32 0.14 0.48 0.30 0.32 0.17 0.19 0.24 0.29 0.59 0.32 0.28 Slips 0.090 0.171 0.198 0.495 0.315 0.243 0.108 0.360 0.225 0.243 0.126 0.144 0.180 0.216 0.441 0.243 0.207 Stoves 0.056 0.106 0.123 0.308 0.196 0.151 0.067 0.224 0.140 0.151 0.078 0.090 0.112 0.134 0.274 0.151 0.129 Slag 0.078 0.148 0.172 0.429 0.273 0.211 0.094 0.312 0.195 0.211 0.109 0.125 0.156 0.187 0.382 0.211 0.179 Total 2.1 4.1 4.7 11.8 1.7 5.8 2.6 2.0 1.2 5.8 3.0 3.4 1.0 1.2 2.4 5.8 1.1 3-31 ------- TABLE 3-8. ESTIMATES OF Mn EMISSIONS FROM BLAST FURNACE OPERATIONS Plant USX, Fairfield, AL USX, Gary, IN USX, Gary, IN USS/Kobe Steel, Lorain, OH WCI Steel, Warren, OH Weirton Steel, Weirton, WV Wheeling Pittsburgh Steel, Mingo Junction, OH Furnace 8 4,6,8 13 3 4 1 1,3 IN 5S Totals Capacity (million tpy) 2.0 3.4 2.7 1.3 1.0 1.6 2.1 0.7 1.6 60.9 Mn emissions (tpy) Casthouse 3.6 6.1 0.4 0.2 1.8 0.2 3.8 1.3 0.2 62.1 Raw materials 0.24 0.41 0.32 0.16 0.12 0.19 0.25 0.08 0.19 7.3 Slips 0.180 0.306 0.243 0.117 0.090 0.144 0.189 0.063 0.144 5.5 Stoves 0.112 0.190 0.151 0.073 0.056 0.090 0.118 0.039 0.090 3.4 Slag 0.156 0.265 0.211 0.101 0.078 0.125 0.164 0.055 0.125 4.8 Total 4.3 7.3 1.3 0.6 2.1 0.8 4.5 1.5 0.8 83.1 3-32 ------- TABLE 3-9. ESTIMATES OF HCN EMISSIONS FROM BLAST FURNACE WASTEWATER TREATMENT (COOLING TOWER) Plant Acme Steel, Riverdale, IL AK Steel, Ashland, KY30'31 AK Steel, Mddletown, OH3'4 Bethlehem Steel, Burns Harbor, IN Bethlehem Steel, Sparrows Point, MD Geneva Steel, Orem, UT Gulf States Steel, Gadsden, AL Inland Steel, East Chicago, IN LTV Steel, Cleveland, OH LTV Steel, East Chicago, IN National Steel, Granite City, IL** National Steel, Ecorse, MI Rouge Steel, Dearborn, MI Furnace 1 A 3 C,D L 1,2,3 2 7 5,6 C1,C5 C6 H3 H4 A,B A,B,D B,C Capacity (million tpy) 1.0 1.9 2.2 5.5 3.5 2.7 1.2 4.0 2.5 2.7 1.4 1.6 2.0 2.4 4.9 2.8 HCN emissions (tpy) 40 76 88 * 140 108 48 160 100 108 56 64 80 96 196 112 3-33 ------- TABLE 3-9. ESTIMATES OF HCN EMISSIONS FROM BLAST FURNACE WASTEWATER TREATMENT (COOLING TOWER) Plant USX, Braddock, PA USX, Fairfield, AL USX, Gary, IN USS/Kobe Steel, Lorain, OH WCI Steel, Warren, OH Weirton Steel, Weirton, WV Wheeling Pittsburgh Steel, Mngo Junction, OH Furnace 1,3 8 4,6,8,13 3 4 1 1,3 IN 5S Total Capacity (million tpy) 2.3 2.0 6.1 1.3 1.0 1.6 2.1 0.7 1.6 61 HCN emissions (tpy) 92 80 244 52 40 64 84 28 64 2,220 * This plant provided data showing essentially no HCN in the scrubber water. ** This plant does not have a cooling tower. HCN emissions are likely to occur at other steps in the wastewater treatment process. 3-34 ------- 3.3 BASIC OXYGEN PROCESS FURNACE ' 39 This section provides a brief description of the BOPF process; additional details on the process can be found in References 1 and 39. The BOPF shop is a cyclical batch operation, beginning when the molten iron is brought from the blast furnace in torpedo cars and transferred to a ladle. Each shop is comprised of several distinct operations including: (1) hot metal transfer of the molten iron received from the blast furnace; (2) deslagging of the hot metal; (3) desulfurization; (4) charging of hot metal and steel scrap to the BOPF vessel; (5) refining the hot metal into steel; (6) tapping the furnace; (7) deslagging; (8) ladle metallurgy, where additional alloy additions and final changes to the chemistry of the steel may be made; and (9) transfer of the steel to a continuous caster. The plants and their production capacities, process flow rates, and control devices for primary and secondary emissions are listed in Tables 3-10 and 3-11. Information on controls used for ancillary processes is given in Table 3-12. This information was obtained from survey responses listed in References 3 through 8, 10 through 13,15 through 17, and 27 through 38. There are a total of 23 BOPF shops at 20 plants that are owned by 14 companies with a nationwide capacity of about 68 million tpy. The plants are located in 10 different States, with the largest number in Ohio and Indiana with 6 shops each. 3.3.1 Reladling, Desulfurization, and Slag Skimming After the hot metal is produced in the blast furnace, it is transferred to the BOPF shop. Brick lined torpedo cars are preferred because of their insulating qualities and consequent lower heat loss from the iron. The hot metal is then reladled from the torpedo cars to the BOPF shop ladle. This transfer is accompanied by the emissions of kish, a mixture of fine iron oxide particles together with larger graphite particles. The reladling generally takes place under a hood to capture these emissions. Desulfurization of the hot metal is accomplished by means of various reagents such as oda ash, lime, and magnesium. Injection of the reagents is accomplished pneumatically with either dry air or nitrogen. Desulfurization may take place at various locations within the iron and steel making facility; however, if the location is the BOPF shop, then it is most often 3-35 ------- TABLE 3-10. BOPF SHOP EMISSION CONTROL SYSTEMS - CLOSED HOOD BOPF SHOPS Closed Hood BOPF Shops Plant AK Steel30-31 AK Steel3'4 Geneva Steel Inland Steel No. 2 Shop LTV Steel No. 2 Shop USS/Kobe Steel USS Fairfield Works Location Ashland, KY Middletown, OH Provo, UT East Chicago, IN Cleveland, OH Lorain, OH Fairfield, AL Capacity (tpy) 2,167,545 2,716,000 2,500,000 2,500,000 4,380,000 2,600,000 2,200,000 Flow (dscfm) 78,000 40,000 (#15) 51,000 (#16) 77,800 50-60,000 138,000 58,000 (L) 59,000 (N) 81,000 Top/bottom blown Top Top Bottom Top Top Top Bottom Primary control Scrubber Scrubber Scrubber Scrubber Scrubber Scrubber Scrubber Secondary control Baghouse None Baghouse Scrubber Baghouse Baghouse Baghouse 3-36 ------- TABLE 3-11. BOPF SHOP EMISSION CONTROL SYSTEMS - OPEN HOOD BOPF SHOPS Open Hood BOPF Shops Plant Acme Steel28-29 Bethlehem Steel Bethlehem Steel Gulf States Steel Inland Steel No. 4 LTV Steel LTV Steel No. 1 Shop National Steel National Steel Rouge Steel USS Edgar Thomson Works USS Gary Works USS Gary Works WCI Steel Weirton Steel Wheeling Pittsburgh Steel Location Riverdale, IL Bums Harbor, IN Sparrows Point, MD Gadsden, AL East Chicago, IN East Chicago, IN Cleveland, OH Granite City, IL Ecorse, MI Dearborn, MI Braddock, PA Gary, IN Gary, IN Warren, OH Weirton, WV Mingo Junction, OH Capacity (tpy) 1,290,000 5,353,500 4,000,000 1,300,000 2,740,000 4,161,000 3,340,000 2,575,440 4,100,000 3,309,000 2,760,000 2,933,935 3,992,812 1,728,000 3,200,000 2,600,000 Flow (dscfm) 288,000 339,600b 600,000C 327,000 310-380,000 458,100 550,000 330,000 500,000 500,000 174,000 267,858C 267,227C 480,000C 280,000 210,000 Top/bottom blown Top Top Top Top Top Top Top Top Top Top Top Top Bottom Top Top Top Primary control ESP Scrubber Scrubber ESP Scrubber ESP ESP ESP ESP ESP Scrubber Scrubber Scrubber ESP Scrubber Scrubber Secondary control3 None Noneb None None Baghouse None None Enclosure to primary system Baghouse None Baghouse None Baghouse None None None a Only systems with separate capture and control devices for fugitive emissions are listed; several plants use the primary control system for partial capture of charging and tapping emissions. b acfm total for 3; this shop has 1 closed hood and 2 open hood vessels. The closed hood vessel has a scrubber for secondary control. 0 acfm. 5-37 ------- TABLE 3-12. SUMMARY OF CONTROLS FOR ANCILLARY PROCESSES Plant Acme Steel, Riverdale, IL28- 29 AK Steel, Ashland, KY AK Steel, Middletown, OH Bethlehem, Bums Harbor, IN (3 vessels in 1 shop) — 3 stations Bethlehem, Sparrows Pt, MD Geneva Steel, Orem, UT Gulf States Steel, Gadsden, AL Inland Steel, East Chicago, (o) IN (2 shops) (c) LTV Steel, East Chicago, IN LTV Steel, Cleveland, OH (o) (2 shops) (c) National Steel, Granite City, IL National Steel, Ecorse, MI Rouge Steel, Dearborn, MI USX, Fairfield, AL USX, Gary, IN (2 shops) (o) (o) USX, Braddock, PA USS/Kobe Steel, Lorain, OH WCI Steel, Warren, OH Weirton Steel, Weirton, WV Wheeling Pittsburgh Steel, Mingo Junction, OH Other controls Hot Metal Reladle Baghouse Baghouse-1 Baghouse- 1 Baghouse-1 Baghouse-1 None Baghouse-1 Baghouse-1 Baghouse-1 Baghouse-1 Baghouse-1 Baghouse-1 Baghouse-1 Baghouse-1 Baghouse-1 Baghouse- 1,2 Baghouse Baghouse Baghouse-1 Flame suppression Flame supp. Baghouse-1 Baghouse-1 Hot Metal desulfurization Baghouse Baghouse-1 Baghouse-1 Baghouse-1 Baghouse-2 Baghouse-1 Baghouse-1 Baghouse-1 Baghouse-1 Baghouse-1 Baghouse-1 Baghouse-1 Baghouse-1 Baghouse-1 Baghouse-2 Baghouse- 1,2 Baghouse Baghouse Baghouse-1 Baghouse-1 Baghouse-1 Baghouse-2 Unknown Hot Metal deslagging Baghouse Baghouse-1 Baghouse-2 Baghouse-1 Baghouse-3 None Baghouse-1 None Baghouse-1 Baghouse-1 Baghouse-1 Baghouse-1 Baghouse-2 Baghouse-1 Baghouse-2 None None None None None Baghouse-1 Baghouse-2 Baghouse-1 Slag transfer None None None None None None None None None None None None None None None None None None None None None None Water spray Ladle metallurgy Baghouse Baghouse-2 Baghouse-3, Scrubber Baghouse-3 Baghouse-4 None Baghouse-2 Baghouse-2 Baghouse-2 Baghouse-2 Baghouse-2 & scrubber Baghouse-3 Baghouse-2 Baghouse-3,4 Baghouse-3,4 Baghouse Baghouse-1 Cyclone/ Baghouse-2 Baghouse-2 Baghouse-3 None 3-38 ------- o = open; c = closed * One ladle met station has no controls; a steam injector and condenser are used for vacuum degassing operations. 3-39 ------- accomplished at the reladling station to take advantage of the fume collection system at that location. Skimming of slag from the ladle of molten iron keeps this source of high sulfur out of the steelmaking process. Skimming results in the emissions of kish, and is therefore normally done under a hood.40 3.3.2 BOPF Shop The BOPF receives a charge composed of molten iron and scrap and converts it to molten steel. Each BOPF shop contains at least two BOPF vessels that may be operated alternately; in some shops, both vessels may be in use at different stages of the cycle. The distinct operations in the BOPF process are: (1) charging - the addition of molten iron and metal scrap to the furnace, (2) oxygen blow - introducing oxygen into the furnace to refine the iron, (3) turndown - tilting the vessel to obtain a sample and check temperature, (4) reblow — introducing additional oxygen, if needed, (5) tapping — pouring the molten steel into a ladle, and (6) deslagging - pouring residual slag into a slag pot. A jet of high purity oxygen oxidizes the carbon and the silicon in the molten iron in order to remove these products and to provide heat for melting the scrap. After the oxygen jet is started, lime is added to the top of the bath to provide a slag of the desired basicity. Fluorspar and mill scale are also added in order to achieve the desired slag fluidity. The oxygen combines with the unwanted elements (with the exception of sulfur) to form oxides, which leave the bath as gases or enter the slag. As refining continues and the carbon content decreases, the melting point of the bath increases. Sufficient heat must be generated from the oxidation reactions to keep the bath molten.1 The furnace is a large, open-mouthed vessel lined with a basic refractory material (the term "basic" refers to the chemical characteristic of the lining). There are currently three methods that are used to supply the oxidizing gas: (1) top blown, (2) bottom blown, and (3) combination blowing. These processes are described in detail below. The basic oxygen steelmaking process is a thermochemical process; computations are made to determine the necessary percentage of molten iron, scrap, flux materials, and alloy additions. Various steelmaking fluxes are added during the refining process to reduce the sulfur and phosphorus content of the metal to the 3-40 ------- proscribed level. The oxidation of silicon, carbon, manganese, phosphorus, and iron provide the energy required to melt the scrap, form the slag, and raise the temperature of the bath to the desired temperature. After the steel is refined, alloy or other additions are made in the vessel as necessary, and the vessel is then turned down and tapped. If the analysis is correct, the heat is tapped; however, if the analysis is off, then it may be necessary to either blow with additional oxygen to elevate the temperature and/or cool the steel by coolant additions to the bath. In most shops, the steel is transferred to a ladle metallurgy station for further alloy additions and then to a continuous caster. A few facilities may still teem some of their steel, pouring the molten steel into ingot molds, but most facilities have switched to the more modern and efficient process of continuous casting. The BOPF shop is generally arranged with three parallel aisles. The first aisle, the charging aisle, has one or more cranes for handling charge materials to the furnace as well as handling ladles of molten slag away from the furnace. The second aisle, the furnace aisle, contains the furnaces, collection hood for the fumes, lances for injecting oxygen into the bath, and overhead bins for storing and metering out the various flux materials and alloy additions. The third aisle, the pouring aisle, handles the finished heats of steel. This aisle has one or more overhead cranes and facilities for receiving heats of steel into ingot molds or continuous casting machines. During the oxygen blow in the top blown process, the oxygen lance is lowered through a special hole in the top wall of the hood, is stopped a short distance above the bath of steel, and the oxygen flow is initiated. The vessel is upright during the blow and the fumes have a direct access from the mouth of the furnace into the mouth of the hood. At other times in the process, the vessel may be tilted so that the mouth of the vessel does not align with the opening in the hood and capture of the fumes becomes more difficult. The vessel is tilted toward the charging aisle for charging with scrap, charging with molten iron, sampling the heat for analysis, and dumping the slag. The vessel is tilted toward the pouring aisle when pouring the finished heat of steel from the furnace into the steel ladles. These operations are controlled by a secondary capture system at some facilities in the industry. The desired specifications of the end product are usually accomplished by the additions of suitable alloying materials to the ladle of finished steel as it is filled. The gases which evolve from the steelmaking operation are captured by the hood and then enter a gas cleaning system consisting of a electrostatic precipitator or a wet scrubber.40 3-41 ------- 3.3.2.1 Bottom Blown Furnace. An alternative to the use of an oxygen lance is found in the Q-BOP process. In this process, oxygen and natural gas are injected through tuyeres in the bottom of the vessel. The metallurgy of the process, the ancillary equipment, and the fume suppression system are generally the same as for the BOPF. The principal advantage of the Q-BOP is that it requires less headroom in the furnace aisle than the BOPF. This has allowed the Q-BOP to be installed in an existing open hearth building, saving cost in construction. The Q-BOP is also capable of producing steel at a somewhat faster rate than does the BOPF. When the Q-BOP is tilted to receive scrap and molten iron, or to sample the steel for analysis, it is necessary to maintain a flow through the tuyeres so that they do not become blocked. In normal practice, the oxygen and natural gas are turned off when the vessel is tilted and these gases are replaced by a flow of nitrogen. Because of this, there is an unrestrained flow of emissions of fumes from the mouth of the vessel due to the gas flow from the tuyeres. For this reason, the Q-BOP is more fully enclosed at the level of the charging floor than many BOPF vessels. In order to direct the gases back into the collection system, a pair of large horizontally sliding doors are provided; these doors are opened to permit the addition of scrap and molten iron but are closed at all other times.40 3.3.2.2 Combination Blowing. Combination blowing processes utilize oxygen through a top lance and an inert gas through tuyeres or permeable elements in the furnace bottom to stir the bath. A second class of combination blowing processes uses some of the oxygen through a top lance or tuyeres mounted in the top cone of the vessel, and the balance of the oxygen through Q-BOP type tuyeres in the vessel bottom. These processes can usually switch the bottom gas from oxygen to argon or nitrogen for stirring purposes.1 3.3.3 Ladle Metallurgy The purpose of ladle metallurgy (also referred to as secondary steelmaking) is to produce steel which satisfies stringent requirements of surface, internal, and microcleanliness quality and mechanical properties. Ladle metallurgy is a secondary step of the steelmaking process often performed in a ladle after the initial refining process in the primary BOPF is completed. This secondary step enables plants to exercise control over many processing conditions contributing to a higher quality of steel including: 1. Teeming temperature, especially for continuous casting operations; 2. Deoxidation; 3. Decarburization (ease of producing steels to low carbon levels of less than 0.03 percent); 3-42 ------- 4. Additional adjustment for chemical composition; 5. Increasing production rates by decreasing refining times in the furnace.* Nearly all of the integrated iron and steel facilities have ladle metallurgy operations. Several ladle metallurgy processes are commonly used, including vacuum degassing, ladle refining, argon/oxygen decarburization, and lance powder injection. 3.3.4 Emission Points and Factors Affecting Emissions The emission points associated with the BOPF shop are shown in Figure 3-3. The most significant sources of emissions are from charging, tapping, and the oxygen blow portions of the furnace cycle. Auxiliary processes including hot metal transfer, desulfurization, slag skimming, and ladle treatment also contribute to the total emissions. Emissions from desulfurization and ladle metallurgy are captured and controlled by a series of one or more control devices at most plants. Emissions from slag removal, slag transfer and disposal, and from transfer to the continuous caster or ingot molds are generally uncontrolled. 3-43 ------- 1 Ei Hot metal Additives \ t Torpedo car transfer V Hot metal 1 1 C Baghouse ~N or ESP J 1 — 4 2. Desulfurization emissions Transfer to continuous Cast steel missions from transfer 3. Emissions from 5. Emissions from I skimming 4. Charging emissions oxygen blow i A i ^ "N { Baghouse j ^ ) 1 1 Baghouse i i ___ Slag skimming i ! i ! 1 ESP, scrubber | f Scrubber | 1 or baghouse II or ESP I i 1 i 1 i I i 1 1 1 i ! I I metal . ^ Basic oxygen furnace Slag »- 1 " t Scrap 1 Additives Flux Molten steel Oxygen 1 V Ladle i i i 1 1 Tapping /" "X 6. Emissions from slag removal 4 i i Cone, scrubber^ ESP, or baghouse J 1 1 _ Deslagging \ ' Slag transfer, disposal i t None or baghouse i T 10. Transfer emissions Baghouse or scrubber i 9. Emissions from ladle metallurgy ESP, scrubber or baghouse i 8. Emissions from tapping None or •water spray I 7. Emissions from slag transfer, disposal FIGURE 3-3. SCHEMATIC OF BOPF SHOP EMISSION POINTS AND TYPICAL CONTROLS 3-44 ------- The major HAP reported to be emitted from the BOPF process is Mn; some Pb has also been reported, as have very small quantities of chromium, copper, mercury, nickel, and selenium.7'8'37'38 Emission control performance for this operation has been determined traditionally based on PM emissions. There are differences in BOPF and types of control devices among the various shops. The primary emission capture and control system for the BOPF vessel is either an open hood directed to an ESP or wet scrubber, or a closed hood ducted to a wet scrubber. In the closed hood system, the diameter of the hood is approximately the same as the diameter of the vessel, and the lower portion of the hood is a skirt that can be lowered onto the mouth of the vessel. This seals off the space between the hood and the vessel, limiting the amount of air that can enter the system. In contrast, the open hood is loose-fitting and draws in dilution air with the emissions captured from the BOPF. The volume of gas collected in the closed hood system is reduced by 80 to 85 percent as compared to the open hood system. Because there is less danger from explosion in the open hood system, the vessels may be connected to a common gas cleaning system. In an closed hood system, each vessel has a separate scrubber system because of the potential explosion hazard from leakage of air into the system from an idle furnace. There are currently 7 closed hood BOPF shops and 16 open hood BOPF shops in operation in the U.S. BOPF vessels are also differentiated as either "top blown" or "bottom blown." There are currently 3 bottom-blown shops and 20 top-blown shops in the US As discussed above, differences in process design and operation affect the quantity and concentration of pollutants that escape capture and are emitted as fugitive emissions. In addition, charging emissions are affected by the quality, quantity, and composition of scrap charged to the furnace as well as the pour rate and pouring technique used to charge the hot metal. After refining in the BOPF vessel, the steel may be sent to a ladle metallurgy station for further refining or alloy additions before subsequent transfer to the continuous caster. All of the BOPF shops in the U.S. have a ladle metallurgy station, although the actual process varies from plant to plant. Emissions from these operations are affected by the type of capture device used and the surface area of molten metal that is exposed to the atmosphere. 3-45 ------- 3.3.5 Estimates of Baseline Emissions The approach used in this section to estimate baseline emissions relies on estimates submitted by the individual companies and emission factors in EPA's AP-42 emission factor document.26 For metal HAP, Mn was the HAP metal most reported by the facilities. Estimates of PM are used with analyses of the dust for metals (expressed as percent Mn) to estimate Mn emissions. 3.3.5.1 BOPF Charging, Oxygen Blow, and Tapping PM Emissions. Emission estimates for PM from BOPF charging, oxygen blow, and tapping, are provided in Table 3-13 from estimates provided by each company in response to survey questionnaires. Many companies apparently reported only emissions from the discharge stacks of emission control devices, and only a few attempted to estimate fugitive emissions that escape through the roof monitor. To put the estimates on a common basis, the AP-42 emission factors26 were used in an attempt to account for both primary system emissions and fugitive emissions from certain processes that escape through the roof monitor. Several plants submitted emission measurements for the primary control system; consequently, when measurements were available, the measurements were used instead of the AP-42 emission factors. 3.3.5.2 Miscellaneous Emission Points. The PM emission estimate factors used for hot metal transfer, desulfurization, charging, oxygen blow, and tapping, are summarized in Table 3-14. The PM emission measurements for ladle metallurgy shown in Table 3-13 are very low relative to emissions from other points because this process is controlled by baghouses at almost all plants. Consequently, ladle metallurgy operations contribute very little to HAP metal emissions (i.e., Mn) from the BOPF shop. All of the emission factors except that for the primary control system for closed hood BOPFs are from AP-42. For closed hood shops, the emission factor of 0.0068 Ib/ton was not consistent with the test measurements submitted by three plants with closed hood shops. Their measurements given below were used to derive an emission factor of 0.035 Ib/ton. 3-46 ------- Plant Capacity (tpy) Emissions (Ib/yr) 1 1,700,000 64,000 2 2,500,000 134,000 3 4,400,000 102,000 Total 8,600,000 300,000 Emission factor (closed hood BOPFs) = 300,000/8,600,000 = 0.035 Ib/ton 3.3.5.3 Estimates of Mn Emissions. More than half of the plants provided data on Mn emissions from the BOPF shop, and they typically estimated these emissions from the percent of PM that was Mn. Emissions of Mn far exceeded all other FtAP metals combined; consequently, estimates of FtAP emissions from this process will focus on Mn as the FtAP of interest. The data from various companies ranged from 0.14 to 10.7 percent of PM; however, the vast majority was in the range of about 0.5 to 1.5 percent. Neglecting the two outliers on the extreme end of the range, the overall average of the data indicated that Mn was 0.95 percent of PM. This value of 0.95 percent should be accurate within a factor of two or less for most plants and was used to estimate Mn emissions from charging, oxygen blow, and tapping operations for the BOPF shop. The emission estimates for PM and Mn are provided in Tables 3-15 and 3-16. 3-47 ------- TABLE 3-13. PM EMISSIONS FROM THE BOPF SHOP REPORTED BY THE COMPANIES Plant Acme Steel (IL) AK Steel (KY) AK Steel (OH) Bethlehem Steel (IN) Bethlehem Steel (MD) Geneva Steel (UT) Gulf States Steel (AL) Inland Steel #2 Inland Steel #4 LTV Steel (IN) LTV Steel (OH) #1 LTV Steel (OH) #2 National Steel (IL) National Steel (MI) Rouge Steel (MI) USS/Kobe Steel (OH) USX (AL) USX BOPF (IN) USX Q-BOP (IN) USX (PA) WCI Steel (OH) Weirton Steel (WV) Wheeling-Pitt (OH) Capacity (million tpy) 1.3 2.2 2.7 5.3 4.0 2.5 1.3 2.5 2.7 4.2 3.3 4.4 2.6 4.1 3.3 2.6 2.2 2.9 4.0 2.8 1.7 3.2 2.6 Reported PM Emissions (tpy) Desulf- urization 24 7 19 119d 33g 6d 5d 4r 2g 76d 31k 9.8 1 2.5d 12> 11 Charge 29 193 93 22 50e 89e 159e 161 5.5 496e 149e 18 O2 Blow 51 9 58 51° 226 28 67f 400 307h 1583'4 23 Oc 208C 17.4 23 203 96 Tap 64 220a 394 119 32 301 4.2 52 Ladle Met. 0 13 0.3b 0.6 8.4 5 2 10 39b 10.3 5.6 2 9 "includes transfer, desulfurization, skimming, charging bfrom control device only 'includes charging, tapping; from control device only Includes hot metal transfer Includes charging; 354 tpy from roof 'includes transfer, skim, tap, charge 5-48 ------- dincludes transfer, slag skimming "includes tapping Includes slag skimming Includes roof monitor 'roof monitor only 'reported 84 tpy from roof monitor TABLE 3-14. EMISSION FACTORS USED FOR THE BOPF SHOP Emission point Open hood with ESP Open hood with scrubber Closed hood with scrubber Q-BOP with scrubber Charging fugitives with baghouse Tapping fugitives with baghouse Uncontrolled charging fugitives Uncontrolled tapping fugitives Hot metal transfer at roof monitor Desulfurization with baghouse PM in Ib/ton 0.13 0.09 0.035 0.056 0.0006 0.0026 0.142 0.29 0.056 0.009 Source AP-42 AP-42 Derived (see text) AP-42 AP-42 AP-42 AP-42 AP-42 AP-42 AP-42 5-49 ------- TABLE 3-15. ESTIMATES OF PM EMISSIONS FROM THE BOPF SHOP Plant Capacity (106tpy) Estimates of PM emissions (tpy) from AP-42 factors3 Transfer, desulfurization Charge O2 blow Tap Total Open hood shops with no secondary controls b Acme Steel (IL) Bethlehem Steel (IN) Bethlehem Steel (MD) Gulf States Steel (AL) LTV Steel (IN) LTV Steel (OH) #1 National Steel (IL) Rouge Steel (MI) USX Gary (IN) WCI Steel (OH) Weirton Steel (WV) Wheeling-Pitt (OH) 1.3 5.3 4.0 1.3 4.2 3.3 2.6 3.3 2.9 1.7 3.2 2.6 42 174 130 42 135 109 84 107 95 56 104 85 92 380 284 92 295 237 1 234 208 123 227 185 85 51C 180 85 270 158C 168 215 132 113 203C 96C 189 776 580 189 603 484 3 479 425 251 464 377 408 1,381 1,174 408 1,303 988 256 1,035 860 543 998 743 Open hood shops with secondary controls Inland Steel #4 National Steel (MI) USX (PA) 2.7 3.5 2.8 89 114 90 1 1 1 400° 159= 124 4 5 4 494 279 218 Closed hood shops with secondary controls AK Steel (KY)30'31 Inland Steel #2 LTV Steel (OH) #2 USS/Kobe Steel (OH) 2.2 2.5 4.4 2.6 71 81 142 48 1 1 1 0 90 67C 51C 26 o 3 o 3 6 2 83 152 200 76 5-50 ------- TABLE 3-15. ESTIMATES OF PM EMISSIONS FROM THE BOPF SHOP Plant Capacity (106tpy) Estimates of PM emissions (tpy) from AP-42 factors3 Transfer, desulfurization Charge O2 blow Tap Total Closed hood shops with no secondary controls AK Steel (OH)3'4 2.7 88 192 230° 392 902 Q-BOPs with secondary controls Geneva Q-BOP (UT) USXQ-BOP(AL) USX Q-BOP (IN) Total 2.5 2.2 4.0 68 81 72 130 2,169 0.8 0.7 1.2 2,558 28C 62 112 3,024 3.3 2.9 5.2 5,249 113 138 248 13,001 a Estimated from the emission factors in Table 3-14 unless otherwise noted. b Assumes no capture and control by the primary system; most open hood shops control some of the fugitive emissions by the primary capture and control system. c These are based on emission measurements submitted by the companies. 5-51 ------- TABLE 3-16. ESTIMATES OF Mn EMISSIONS FROM THE BOPF SHOP Plant Capacity (106tpy) Estimates of Mn emissions (tpy) a Transfer, desulfurization Charge O2 blow Tap Total Open hood shops with no secondary controls Acme Steel (IL) Bethlehem Steel (IN) Bethlehem Steel (MD) Gulf States Steel (AL) LTV Steel (IN) LTV Steel (OH) #1 National Steel (IL) Rouge Steel (MI) USX Gary (IN) WCI Steel (OH) Weirton Steel (WV) Wheeling-Pitt (OH) 1.3 5.3 4.0 1.3 4.2 3.3 2.6 3.3 2.9 1.7 3.2 2.6 0.4 2 1 0.4 1 1.0 0.8 1 0.9 0.5 1.0 0.8 0.9 4 3 0.9 3 2.3 0.0 2 2.0 1.2 2.2 1.8 0.8 0 2 0.8 3 1.5 1.6 2 1.3 1.1 1.9 0.9 1.8 7 6 1.8 6 4.6 0.0 5 4.0 2.4 4.4 3.6 3.9 13 11 3.9 12 9.4 2.4 10 8.2 5.2 9.5 7.1 Open hood shops with secondary controls Inland Steel #4 National Steel (MI) USX (PA) 2.7 3.5 2.8 0.8 1.1 0.9 0.0 0.0 0.0 3.8 1.5 1.2 0.0 0.0 0.0 4.7 2.6 2.1 Closed hood shops with secondary controls AK Steel (KY) Inland Steel #2 LTV Steel (OH) #2 USS/Kobe Steel (OH) 2.2 2.5 4.4 2.6 0.7 0.8 1.3 0.5 0.0 0.0 0.0 0.0 0.1 0.6 0.5 0.2 0.0 0.0 0.1 0.0 0.8 1.4 1.9 0.7 5-52 ------- TABLE 3-16. ESTIMATES OF Mn EMISSIONS FROM THE BOPF SHOP Plant Capacity (106tpy) Estimates of Mn emissions (tpy) a Transfer, desulfurization Charge O2 blow Tap Total Closed hood shops with no secondary controls AK Steel (OH) 2.7 0.8 1.8 22 3.7 8.6 Q-BOPs with secondary controls Geneva Q-BOP (UT) USXQ-BOP(AL) USX Q-BOP (IN) Total 2.5 2.2 4.0 95 0.8 0.7 1.2 21 0.0 0.0 0.0 24 0.3 0.6 1.1 29 0.0 0.0 0.0 50 1.1 1.3 2.4 124 * Based on 0.95 percent Mn in the PM (Table 3-15). ------- 3.4 REFERENCES 1. United States Steel. The Making, Shaping, and Treating of Steel. Published by the Association of Iron and Steel Engineers (AISE). Available from AISE at Suite 2350, Three Gateway Center, Pittsburgh, PA. 2. Carpenter,B., D. VanOsdell, D. Coy, and R. Jablin. Pollution Effects of Abnormal Operations in Iron and Steel Making - Volume n. Sintering, Manual of Practice. EPA-600/2-78-118b. June 1978. pp. 12-15. 3. Francis, S.L., AK Steel, Middletown, OH to B. Jordan, EPA. Responding to section 114 request. September 5, 1991. 4. Felton, S.S., AK Steel, Middletown, OH, to P. Mulrine, EPA. Comments on draft background information document for integrated iron and steel plants. November 30, 1998. 5. Riley, W.J., Bethlehem Steel, Burns Harbor, IN to B. Jordan, EPA. Response to section 114 request. February 14, 1994. 6. Ossman, G.A., Bethlehem Steel, to P. Mulrine, EPA. Comments on draft background information document for integrated iron and steel plants, February 1, 1999. 7. Anderson, D.A., Bethlehem Steel, Sparrows Point, MD to B. Jordan, EPA. Response to section 114 request. August 29, 1991. 8. Starley, J.R., Geneva Steel, Provo, UT to B. Jordan, EPA. Response to section 114 request. October 29, 1993. 9. Christiansen, R, Geneva Steel to J. Calcagni, Research Triangle Institute. Sinter plant test results for the new baghouse. March 12, 1996. 10. Shoup, S.P., Inland Steel, East Chicago, IN to B. Jordan, EPA. Response to section 114 request. November 12, 1993. 11. Allie, G.R., Inland Steel, East Chicago, IN to P. Mulrine, EPA. Comments on draft background information document for integrated iron and steel plants. December 1, 1998. 12. Thomas, M.J., LTV Steel, East Chicago, IN to B. Jordan, EPA. Response to section 114 request. November 29, 1993. 5-54 ------- 13. Piccirillo, B.L., LTV Steel, East Chicago, IN to P. Mulrine, EPA. Comments on draft background information document for integrated iron and steel plants. April 13, 1999. 14. Summaries of Stack Test Results Provided by USS Gary Works to EPA Region V for the Sinter Plant. Total of 42 Runs Performed Between April 1982 and October 1984. 15. Stack Test Results for WCI Sinter Baghouse Provided by T. Shepker, WCI Steel: Envisage Environmental, Inc. on July 10, 1991 and CSA Company on May 27, 1992. 16. Samples, W.R., Wheeling-Pittsburgh Steel, to B. Jordan, EPA. Response to section 114 request. November 12, 1993. 17. Summary of Emission Testing Conducted by EPA at LTV Steel's Sinter Plant in East Chicago, Indiana. June 25-27, 1997. 18. Summary of Emission Testing Conducted by EPA at WCI Steel's Sinter Plant in Youngstown, Ohio. August 12-15, 1997. 19. Current, G. P., Weirton Steel, Weirton, WV, to B. Jordan, EPA. Response to section 114 request. January 6, 1994. 20. Mostardi-Platt Associates, Inc. Paniculate Metals and Gaseous Emissions Study: Sinter Plant Windbox Baghouse Stack. Inland Steel, East Chicago, IN. May 16-17, 1995. 21. Mostardi-Platt Associates, Inc. Particulate and Gaseous Emission Study Performed for LTV Steel Company at the Sinter Plant Stack. East Chicago, IN. May 29, 1992. 22. Mostardi-Platt Associates, Inc. Diagnostic Gaseous Study Performed for Bethlehem Steel Company at the Sinter Plant Scrubber Stack. Burns Harbor, IN. March 9-11, 1992. 23. Jablin, R., D. Coy, et al. Pollution Effects of Abnormal Operations in Iron and Steel Making - Volume HI. Blast Furnace Ironmaking Manual of Practice. EPA-600/2-78-118c. June 1978. 93pp. 24. Blast Furnace Coal Injection, Long Proven... Now Economically Justified, The Armco/Babcock & Wilcox Coal Injection System. 25. McManus, George J. Coal Gets a New Shot, Iron Age, Jan. 1989, p. 31-38. 26. U. S. Environmental Protection Agency. Compilation of Air Pollution Emission Factors. Publication AP-42. Section 12.5 Iron and Steel Production. July 1995. 3-55 ------- 27. Stewart, E.M., Gulf States Steel, Gadsden, AL to B. Jordan, EPA. Response to section 114 request. February 11, 1994. 28. Zibble, D., Acme Steel, Riverdale, IL to B. Jordan, EPA. Response to section 114 request. June 2, 1994. 29. Wentz, J. Acme Steel, Riverdale, IL to P. Mulrine, EPA. Comments on draft background information document for integrated iron and steel plants. January 6, 1999. 30. Felton, S.S., AK Steel to B. Jordan, EPA. Response to section 114 request for Ashland Works. November 15, 1993. 31. S. S. Felton, AK Steel, Mddletown, OH to P. Mulrine, EPA. Comments on draft background information document for integrated iron and steel plants. November 30, 1998. 32. Nemeth, R.L., LTV Steel, Cleveland, OH to B. Jordan, EPA. Response to section 114 request. December 16, 1993. 33. Dilanni, L.G, USX to B. Jordan, EPA. Response to section 114 request for USX Fairfield Works. December 28, 1993. 34. W.S. Kubiak, U.S. Steel to P. Mulrine, EPA. Forwarding information on U.S. Steel's sinter, iron and steel production facilities. December 4, 1998. 35. J. Earl, Rouge Steel, Dearborn, MI to P. Mulrine, EPA. Comments on draft background information document for integrated iron and steel plants. May 10, 1999. 36. Ames, H.C., USS/Kobe Steel, Lorain, OH to P. Mulrine, EPA. Comments on the draft background information document for integrated iron and steel plants. December 15, 1999. 37. Heintz, J.K., National Steel to B. Jordan, EPA. Response to section 114 request for National Steel's plants in Granite City, IL and Ecorse, MI. January 31, 1994. 38. J.K. Heintz, National Steel, Mishawaka, IN to P. Mulrine, EPA. December 14, 1998. 39. Revised Standards for Basic Oxygen Process Furnaces — Background Information for Proposed Standards. EPA-450/3-82-005a. December 1982. 40. Jablin, R, D. Coy, et al. Pollution Effects of Abnormal Operations in Iron and Steel Making - Volume VI. Basic Oxygen Process, Manual of Practice. EPA-600/2-78-118f June 1978. 5-56 ------- 4.0 EMISSION CONTROL TECHNIQUES AND EQUIPMENT This chapter presents an overview of the techniques typically used to capture and control PM emissions from integrated iron and steel processes, including sinter plants, blast furnaces, and BOPF shops. This overview describes equipment design parameters, operating conditions, application of these control techniques in the industry, and factors that determine the effectiveness of these techniques in reducing emissions. Each section includes a discussion of the various capture systems and control techniques, performance of controls, and pollution prevention opportunities. Detailed descriptions of a few systems in place at actual plants are given to provide more insight into operation and design considerations. 4.1 SINTER PLANT 4.1.1 Windbox The sinter plant windbox serves as the capture system for the sintering machine and is the most critical source of emissions in the sinter plant because of the number and variety of pollutants to be controlled and the high volume flowrate of the exhaust air. After sinter materials are mixed, they are ignited on the surface by gas burners. As the materials move through the sinter bed, air is pulled down through the mixture to bum the fuel by downdraft combustion through a series of windboxes, and evacuated to a control device.1 Baghouses and wet scrubbers are the principal means for controlling emission from the sinter plant windbox. Four plants use a baghouse and five plants use a wet scrubber to control windbox emissions. The final control unit may be preceded by a mechanical collector to remove large, heavy, and abrasive particles.2 The control of emissions from the windbox is made more difficult by factors such as the high volume rate of gas, the sometimes high resistivity of the dust, and the presence of hydrocarbon vapors. Table 4-1 presents various operating parameters for the windbox control systems. 4-1 ------- TABLE 4-1. EMISSION CONTROLS FOR SINTER PLANT WINDBOXES Plants with baghouses Plant Geneva Steel3 Inland Steel4'5 USS Gary Works6-7 WCI Steel8 State UT IN IN OH Capacity (tpy) 803,000 1,000,000 4,400,000 840,000 Flow (dscfm) 306,800 400,000 675,000 (estimate) 400,000 Air/cloth ratio (acfm/ft2) 4.0 1.4 3.9 • P(in- water) 6 4.9 - Cleaning Pulse jet Reverse air Pulse jet Filter Material Polyester Polyester Nomex Location Windbox Windbox Windbox Plants with wet scrubbers Plant AK Steel9- 10 Bethlehem11'12 Bethlehem13' 14 LTV Steel15'16 Wheeling Pittsburgh17 State OH IN MD IN WV Capacity (tpy) 895,000 2,922,000 4,000,000 1,927,000 500,000 Flow (dscfm) 219,000 485,000 600,000 265,000 141,000 L/G (gal/ 1000 acf) 8.2 12 12 9 9.4 • P(in- water) 50-55 60-70 35 40-55 80 Scrubber Type Venturi Venturi Venturi Venturi Venturi Demister Mist eliminator Chevrons Chevrons Chevrons - Location Windbox Windbox Windbox Windbox Windbox air/cloth ratio = ratio of air flow to cloth area in actual cubic feet per minute per square foot of cloth • p = pressure drop in inches of water L/G = liquid to gas ratio in gallons of water per 1,000 actual cubic feet of gas. 4-2 ------- 4.1.1.1 Baghouses. In a baghouse, the particle-laden gas flows through a number of filter bags placed in parallel, leaving the dust retained by the fabric. The type of filter material used in a baghouse depends on the specific application in terms of chemical composition of the gas, operating temperature, dust loading, and the physical and chemical characteristics of the particulate. The type of filter material used will limit the maximum operating gas temperature for the baghouse. Extended operation of a baghouse requires that the dust be periodically cleaned off the cloth surface and removed from the baghouse; this is commonly accomplished in the sinter plant by reverse air or pulse jet cleaning; shaker cleaning may also used in certain circumstances, and is used for several baghouses on the discharge end of the sinter plant. After a new fabric goes through a few cycles of use and cleaning, it retains a residual layer of dust that becomes the filter medium; this phenomenon is responsible for highly efficient filtering of small particles. In reverse air cleaning, gas flow to the bags is stopped in the compartment being cleaned, and a reverse flow of air is directed through the bags. This reversal of air gently collapses the bags and the shear forces developed remove dust from the surface of the bags. The reverse air for cleaning comes from a separate fan capable of supplying clean, dry air for one or two compartments at an air-to-cloth ratio similar to that of the forward air flow.18 In pulse jet cleaning, a burst of air is forced down through the bag expanding it violently. The fabric reaches its extension limit, and the dust separates from the bag. The filtering flows are opposite in direction when compared with reverse air designs. Bags are mounted on wire cages to prevent collapse while the dusty gas flows through them. The top of the bag and cage assembly is attached to the baghouse structure, whereas the bottom end is loose and tends to move in the turbulent gas flow. Pulse jet baghouses may be compartmented; the bags are cleaned by compartment, with one compartment off-line at a time. Where they are not compartmented, bags are cleaned by rows when a timer initiates the burst of cleaning air through a quick-opening valve. A pipe above each row of bags carries the compressed air. The pipe is pierced above each bag so that cleaning air exits directly downward into the bag. In shaker cleaning, inside-to-outside air flow is used and cleaning is accomplished by suspending the bag from a motor-driven hook or framework that oscillates. The motion creates a sine 4-: ------- wave along the fabric, which dislodges the previously collected dust. Chunks of agglomerated dust fall into a hopper below the compartment. The compartments operate in sequence so that one compartment at a time is cleaned. Parameters that affect cleaning include the amplitude and frequency of the shaking motion and the tension of the mounted bag. The vigorous oscillations tend to stress the bags and require heavier and more durable fabrics.18 Baghouses have been installed on the sinter plant windbox at four plants. Two of the baghouses are pulsejet and one is a reverse air cleaning system; the remaining baghouse is a dry injection baghouse. One of these systems is described in greater detail below. The baghouses generally have an air flowrate of 300,000 to 400,000 scfrn, an air-to-cloth ratio of 1.0 to 4.0 acfm/ft2, and a pressure drop of 4 to 9 inches of water. Two of the windbox baghouses have polyester bags and one has Nomex® bags; information is not currently available on the fourth baghouse since it was only recently brought on-line. The plants with baghouses have strict limits on the amount of oil in the sinter feed or the amount of oily mill scale that can be used because organic condensibles from the process can foul ("blind") the fabric used to filter PM. The oil content of the sinter is measured so that it does not exceed a level of approximately 0.1 to 0.2 percent oil in the mill scale feed material. Plants with wet scrubbers can use more oily mill scale in their mix because the hydrocarbon vapors do not interfere with the scrubber's control of PM. Baghouse Installation at WCI Steel.* A system of four baghouses is used to control emissions from the sinter plant at WCI Steel, known as the strand, A, C, and cooler baghouses. The system was modified from an ESP system that was previously used to control emissions. No new building or major structural changes were required to modify the system. A new instrument/control room was built for the new control device system. The baghouse on the strand was manufactured by Environmental Elements. It is a pulse jet baghouse with 14 compartments, utilizing Nomex® bags. Air is pulled down through 21 windboxes and evacuated to the baghouse. The flow to the baghouse is approximately 400,000 cfm. The baghouse has an air to cloth ratio of 3.90 acfm/ft2. A preheat burner is used to minimize condensation and to bring the gas up to the desired inlet temperature of 275 • F. The dust is removed from the 4-4 ------- baghouse by rotary screw to bins where it is stored on the ground to gather moisture and is blended back into the sinter feed. When the strand baghouse system was first brought online, there were problems with sparks and burning bags in the baghouse. In order to decrease the likelihood of fires occurring in the baghouse, a few changes were made to the system. Spark deflectors were added to the baghouse inlet, and the inlet temperature to the baghouse was decreased from 325* F to 275* F. The molecular size of the hydrocarbons was increased by lowering the inlet temperature so that the bags would not ignite as easily. Additional deflector plates were also added to the baghouse. Based on the present performance of the system, several changes would have been made if the system was redesigned from the beginning: (1) baffles would be added to the baghouse; (2) the baghouse would be set further away and would have a longer system of duct work and an expansion chamber to drop out sparks before they reach the baghouse; (3) the air-to-cloth ratio would be lowered from 3.9 to 2.5 acfm/ft2; and (4) spark deflectors would have been added to the system from the beginning. The C baghouse was manufactured by Bahnson-Hawley and is a pulse jet baghouse that utilizes polyester bags. It serves the material handling bins and the conveyors that transfer the sinter mix to the sinter machine. The A baghouse was also manufactured by Bahnson-Hawley. It is a pulse jet baghouse with four compartments, utilizing polyester bags. The system serves the discharge end, including the sinter production bins, sinter breaker, hot and cold screens, and 30-40 transfer points. The cooler baghouse was manufactured by Ohio Ferroalloy. It is a shaker baghouse with 9 compartments, utilizing Nomex® bags. Eight of the compartments are used for the cooler and one compartment is used for the truck loadout station. There are four 200 horsepower fans on the sinter cooler. The first fan is the dirtiest fan and is directed back to hoods on the sinter machine and sent back through as preheat air. The other 3 fans are ducted to the baghouse. The truck loadout station has a 70,000 cfm fan. 4.1.1.2 Scrubbers. High energy scrubbers are used to control emissions from the sinter plant windbox at five plants. Four of the units are high energy venturi scrubbers and one is an impingment 4-5 ------- scrubber. The impingment scrubber and three of the venturi scrubbers are preceded by a cyclone to remove the heavy particles. The wet scrubbers generally have an air flowrate of 140,000 to 600,000 scfm, a liquid to gas ratio of 2.0 to 12.0 gallons per thousand actual cubic feet, and a pressure drop of 35 to 80 inches of water. In general, the wet scrubbers do not have the same limitations as the baghouse systems in the amount of oily mill scale that they can handle. While the use of larger quantities of mill scale will not foul up the scrubber systems, the level of control achievable for hydrocarbons and organic compounds depends on a number of factors. High energy scrubbers offer good control of particulate condensible hydrocarbons, and, in addition, offer control of the fluorides and sulfur dioxide contained in sinter plant windbox gases. Control of hydrocarbons has been shown to depend on three factors: the concentration of the hydrocarbons in the inlet gas; the particle size of the hydrocarbon mist; and the pressure drop across the venturi throat. The most critical factor in controlling oil emissions when using a high energy scrubber is the control of oily emissions from the sinter strand itself. The efficiency of the oil removal from the scrubber system has rarely been shown to exceed 80 percent.2 4.1.2 Discharge End Emission points on the discharge end include sinter discharge, crusher, hot screen, sinter cooler, and cold screen. These emission points are generally hooded individually with an enclosed hood or a suspended hood and evacuated to one or more control devices; the majority of facilities use a series of one or more baghouses. Scrubbers and rotoclones are also used by several plants to control emissions from these sources. The sinter product is generally cooled by air, although water sprays are occasionally used. The baghouse is the best demonstrated emission control device for discharge end emission control. In designing a suitable baghouse, the high abrasion characteristics and temperature of the dust require special consideration. Approximately ten baghouses are in use to control emissions from the various discharge emission points, handling one or more emission points. The most common cleaning mechanism is pulse jet, although shaker and reverse air systems are also used. Most of the baghouses 4-6 ------- use polyester bags, but Nomex® and fiberglass baghouses are also used at some facilities. The baghouses generally have an air flowrate of 32,000 to 350,000 scfm, an air-to-cloth ratio of 1.5 to 6.0 acfm/ft2, and a pressure drop of 4 to 12 inches of water. Venturi scrubbers and cyclones are also used to control discharge end emission points at several plants. The venturi scrubbers generally have an air flowrate of approximately 100,000 scfm and a pressure drop of 35 inches of water. The cyclones generally have an air flowrate of 5,000 to 33,000 scfm and a pressure drop of approximately 5 inches of water. Emissions from the discharge end consist mainly of PM and metals. Table 4-2 shows the various control technologies used for sinter discharge emission points at each plant in the industry. TABLE 4-2. SINTER DISCHARGE AND COOLER CONTROL TECHNOLOGIES Plant Bethlehem, IN Inland Steel, IN LTV Steel, IN U.S. Steel, IN Bethlehem, MD AK Steel, OH WCI Steel, OH Wheeling-Pitt, OH Geneva Steel, UT Discharge Baghouse Baghouse Scrubber Baghouse Baghouse Baghouse Baghouse Baghouse Rotoclone Crusher Baghouse Baghouse Scrubber Baghouse Baghouse Baghouse Baghouse N/A N/A Hot Screen Baghouse Baghouse Scrubber Baghouse Baghouse Baghouse Baghouse N/A Rotoclone Cooler None Baghouse None None Cyclone Baghouse Baghouse Water sprays N/A Cold Screen N/A None None Baghouse Baghouse Water sprays Baghouse Water sprays N/A * Certain transfer points are controlled by the discharge baghouse. 4.1.3 Materials Handling Emissions from material handling are generally fugitive emissions and are usually uncontrolled. These emissions result from material storage, materials mixing, and sinter storage. Fugitive emissions escaping the raw material handling equipment are normally confined within the building in which they are processed, and primarily affect the worker environment. Only one sinter plant in the country uses a baghouse to control emissions from material storage and handling; the remaining plants use no control. Emissions from mixing are also generally uncontrolled, although they are also normally contained within 4-7 ------- the building. One plant, however, uses water sprays to wet the materials at the various transfer points. While water sprays by themselves may be effective on materials such as dry ore, they are not effective in controlling hot fines. Emissions from sinter storage are generally uncontrolled, although one plant uses chemical dust suppression on the product. 4.1.4 Capture and Control System Performance. Windbox capture efficiencies were reported by six companies in a 1993 industry survey and by one company in a 1991 screening survey response. These efficiencies range from 93 to 99.9 percent based on engineering estimates. Control device efficiencies varied considerably, ranging from 96.2 to 99.5 percent for a baghouse and from 70 to 99+ percent for a wet scrubber. 4.1.5 Pollution Prevention Pollution from sinter plants is generated by particulate emissions from various emission points and by organic emissions from the windbox. Sinter plants serve as a means of recycling waste iron- bearing materials that would otherwise be landfilled from other processes at an integrated iron and steel facility and within the sinter plant itself. The use of sinter plants is an effective pollution prevention measure, but significant quantities of particulate and organic compounds are generated as a result of the recycling process. One of the major sources of organic emissions in the sinter plant is from oily mill scale blended into the feed materials. One way to reduce organic emissions in the sinter plant would be to set a limit for the oil content of the sinter mixture or for the amount of oily mill scale that a plant may use. Even though a high energy wet scrubber may be able to handle larger quantities of oil than a comparable baghouse system, limiting the amount of oil for all plants may reduce organic emissions. Another option may be to de-oil the mill scale prior to recycling the scale in the sinter plant. 4-8 ------- 4.2 BLAST FURNACE 4.2.1 Casthouse Emissions from molten iron and slag occur primarily at the tap hole of the blast furnace and in the iron trough immediately adjacent to it. Emissions also result from the runners that transport the iron and slag and from the ladle that receives the molten iron. These emissions include flakes of graphite (carbon) called "kish" that is released as the metal cools (because the solubility of carbon in the metal decreases as it cools) and metal oxides that form when the reduced metal (e.g., iron, manganese) reacts with oxygen in the air.19 Factors affecting these emissions include the duration of tapping, exposed surface area of metal and slag, length of runners, and the presence/absence of runner covers and flame suppression, which reduce contact with air. Table 4-3 presents the capture and control systems in place on each furnace in the industry. Three furnaces at three facilities did not report the presence of capture or control systems for emissions from the casthouse. A combination of flame suppression and covered runners is most commonly used at the remaining furnaces in the industry; in addition, more than one-third of the furnaces evacuate emissions to a control device, most commonly a baghouse. Flame suppression consists of blowing natural gas over the iron runners and torpedo cars. The combustion of the gas consumes oxygen, which suppresses emissions. In addition to flame suppression, many facilities use covered runners on the iron and slag runners. Most furnaces have a removable cover over the iron trough; the cover is removed during drilling of the furnace and is quickly put back into place when the molten iron starts to flow. The cover is removed again at the end of the tap to plug the taphole with refractory clay. One method of controlling emissions from the casthouse is to totally enclose the casthouse and evacuate it to a baghouse. Alternatively, there may be localized hooding over the iron trough, iron and slag runners, and hot metal ladles that are evacuated to a baghouse. Two furnaces at one facility use a vertical rod-type scrubber to control casthouse emissions. 4-9 ------- TABLE 4-3. CASTHOUSE CAPTURE AND CONTROL SYSTEMS Plant Acme Steel20- 21 AK Steel22- 23 AK Steel9- 10 Bethlehem Steel13' 14 Bethlehem Steel11-12 Geneva SteeP Gulf States SteeP4 Inland SteeP' 5 LTV Steel15- 16 LTV SteeP5 National SteeP6- 27 National SteeP6- 27 Rouge SteeP8' 29 USX SteeP0' 31 USX SteeP USX SteeP2' 33 USS/Kobe SteeP4' 35 WCI Steel8 Weirton SteeP6 Wheeling Pittsburgh Steel17 Location IL KY OH MD IN UT AL IN IN OH IL MI MI AL IN PA OH OH WV WV Furnace A A 3 L C,D 1,2,3 2 5,6 7 H3 H4 C1,C5,C6 A3 AB,D B,C 1 4,6,8 13 1,3 3 4 1 1 3 1 5 Casthouse FSa Yes Yes Yes No Yes Yes None No No Yes Yes Yes No No Yes Yes Yes No Yes No Yes Yes Yes Yes Yes Yes CRb Yes Yes No Yes Yes Yes None Yes Yes Yes Yes Yes Yes Yes Yes No No Yes Yes Yes No Yes Yes Yes Yes Yes Control None None None Baghouse None None None Scrubber Baghouse None Baghouse None Baghouse Baghouse None Baghouse None Baghouse Baghouse Baghouse None Baghouse Baghouse None None Baghouse 3 Flame suppression b Covered runners 4-10 ------- The most common baghouse cleaning mechanism is pulse jet, although shaker and reverse air systems are also used. Most of the casthouse baghouses use polyester or polypropylene bags. The baghouses generally have an air flowrate of 125,000 to 400,000 scfrn, an air-to-cloth ratio of 2.0 to 7.0 acfm/ft2, and a pressure drop of 3 to 14 inches of water. Table 4-4 presents the operating parameters for various control systems used on blast furnaces in the U.S. Gaseous and particulate emissions occur from slag handling as the slag is discharged and allowed to cool. Particulate emissions also occur when the solidified slag is later broken up and removed. These emissions are generally uncontrolled, although some facilities use covered runners. No. 7 Blast Furnace at Inland Steel. The No. 7 blast furnace at Inland Steel has four holes for tapping. One taphole is always open and the hot metal is removed continuously. To stop tapping, clay is injected into the taphole under pressure to seal the hole. The molten iron and slag that leave the furnace after tapping are separated in troughs and runners. The slag is diverted outside the casthouse and is sprayed with water to cool. The molten iron is transferred to Pugh ladles to be sent to the BOPF. There are covers over the runners for the molten metal and slag as well as canopies above the tapholes, which are evacuated to route the emissions to the baghouse. The casthouse is controlled by two baghouses, a new baghouse with computerized control that can concentrate on specific sources during the various phases of operating practice, and an older general baghouse that serves as back-up. Dust from the baghouses is currently stored for later recycle.37 4-11 ------- TABLE 4-4. EMISSIONS CONTROLS FOR BLAST FURNACE CASTHOUSES Furnaces with baghouses Plant Bethlehem Steel13- 14 Inland Steel4'5 LTV Steel15'16 Rational Steel26' Rational Steel26' USX Steel32'33 USX Steel6 USS/Kobe34'35 State MD IN IN IL MI PA IN OH Capacity (tpy) 3,450,000 4,000,000 1,971,000 2,372,500 2,000,000 900,000 2,000,000 1,200,000 1,100,000 3,440,000 1,300,000 Furnace 1 7 H4 A B A B D 1 3 13 3 Flow (dscfm) 420,000 acfm @170-200 - 250,000- 275,000 220,000 369,000 100,000 acfm 400,000 170,000 275,000 140,000 140,000 600,000 acfm 224,000 Air/cloth ratio (acfm/ft2) 4.0 - 4.2 4.4 6.88 5.82 5.15 9.0 5.38 - - 4.8 6.28 • P(in- water) 8 - 7 7 14 10 3-8 4-8 3-6 3-12 3-12 <8 3-10 Cleaning pulse jet - pulse jet pulse jet pulse jet shaker reverse air pulse jet pulse jet - - pulse jet pulse jet Filter material polyester - polyester polyester polyester polyester polyester needle felt polyester felt polyester woven - - polyester felt polyester Location Runner covers Canopies over 4 notches Iron trough and tilting spout "A" & "B" taphole Torpedo cars Iron trough/ tilting spout Casthouse Casthouse Casthouse 4-12 ------- TABLE 4-4. EMISSIONS CONTROLS FOR BLAST FURNACE CASTHOUSES (continued) Furnaces with baghouses Plant WCI Steel8 Wheeling- Pittsburgh17 State OH OH Capacity (tpy) 1,500,000 1,682,000 Furnace 1 5 Flow (dscfm) 125,000 103,200 Air/cloth ratio (acfm/ft2) 1.98-2.23 4.5 • P(in- water) - 4-6 Cleaning shaker pulse jet Filter material - polyester felt Location Casthouse Furnaces with wet scrubbers Plant Inland Steel4'5 State IN Capacity (tpy) 1,253,000 1,253,000 Furnace 5 6 Flow (dscfm) 40,000 acfm @250«F 40,000 acfm @250«F L/G (gal/1000 acf) 10.0 10.0 • P (in. water) 24-30 35 Scrubber type Multi-element fixed throat vertical rod type scrubber (2 scrubbers) Multi-element fixed throat (1 scrubber) Demister vanes in tank Location Local hoods over notch, iron and slag runners, and pugh ladles air/cloth ratio = ratio of air flow to cloth area in actual cubic feet per minute per square foot of cloth • p = pressure drop in inches of water 4-13 ------- 4.2.2 Gas Cleaning Blast furnace gas is primarily CO and is heavily laden with particles (on the order of 30 g/scm) as it leaves the furnace. The gas is cleaned and is used as fuel in the blast furnace stoves and other operations at the plant. Emissions occur from the stove stack when this gas is burned; these emissions are generally uncontrolled at all facilities in the industry. Most furnaces are equipped with a multistage dust collection consisting of a dry cyclone and a wet collection. The gas is cleaned by passing it through the cyclone (called a dust catcher) and then directing it to venturi scrubbers for final cleaning. The preferred method of cleaning the gas is the venturi scrubber. Gases in the venturi scrubber are accelerated in the convergent section of the venturi throat in order to impact at high velocity with the injected scrubber water. The wetted particles of dust are agglomerated to form droplets in the venturi diffuser due to decreasing velocity and surface tension. The water droplets containing the pollutants are then separated from the gas in the subsequent gas separator. Most modern venturi scrubbers are designed with an adjustable throat section to compensate for varied rates of gas flow from the blast furnace. Wear in the throat of the venturi is minimized by the provision of a hardened lining and by a protecting film of water on the convergent inner wall. Two of the major consumers of blast furnace gas, blast furnace stoves and the underfiring jets of coke ovens, require that the gas be as free of PM as possible. Any excess PM that might remain in the gas would tend to deposit in the combustion spaces of these units causing premature outages and failures. Because the units are essential to the ironmaking process and require a high investment of capital, the plants find it necessary to maintain and operate the gas cleaning equipment at maximum efficiency.19 Table 4-5 presents the various gas cleaning systems used at integrated iron and steel facilities. 4.2.3 Wastewater The direct contact water used in the scrubber dissolves HCN from the gas, and the HCN is subsequently stripped from the water when it passes through the cooling tower. Cooling tower emissions are not controlled. 4-14 ------- 4-15 ------- TABLE 4-5. GAS CLEANING SYSTEMS FOR EACH FURNACE Plant Acme Steel20'21 AK Steel22' 23 AK Steel9' 10 Bethlehem Steel13' 14 Bethlehem Steel11' 12 Geneva Steel3 Gulf States Steel24 Inland Steel4' 5 LTV Steel15' 16 LTV Steel25 National Steel26' 27 National Steel26'27 Rouge Steel28 USX Steel30 USX Steel6 State IL KY OH MD IN UT AL IN IN OH IL MI MI AL IN ID A A 3 L C,D 1,2,3 2 5,6 7 H3,H4 C1,C5, C6 A B A B D B,C 1 4,6,8,1 3 Gas cleaning system Dry dust catcher, variable throat venturi scrubber, mist eliminator Dust catcher, Bischoff venturi scrubber Dust catcher, Bischoff venturi scrubber, mist eliminator Dust catcher, venturi scrubber 1) Dust catcher; 2) primary wet scrubber; 3) water separator; 4) 3 cone scrubber; 5) water separator; 6) gas cooler; 7) mist eliminator Dust collector, venturi scrubber, gas washer Dust collector, venturi scrubber Dust catcher, venturi scrubber Dust catcher, Bischoff scrubber Dust catcher, fixed orifice scrubber, variable throat scrubber Mechanical dust collector, gas washer and cooler, venturi scrubber, gas recirculation stoves Mechanical dust collector, Bischoff variable throat anulus wet scrubber Dust collector, variable throat venturi scrubber Dust catcher, variable throat venturi scrubber Dust catcher, fixed orifice scrubber, gas washer, cooler tower Dust catcher, fixed orifice scrubber, variable throat venturi gas cooler/scrubber, demister Mechanical collector, venturi scrubber Dust catcher, quencher, scrubber Mechanical collector, gas cleaning 4-16 ------- Plant USX Steel32 USS/Kobe Steel34 WCI Steel8 Weirton Steel36 Wheeling- Pittsburgh17 State PA OH OH WV OH ID 1,3 3,4 1 1,3 1N,5S Gas cleaning system Dry scrubber, wet scrubber Dust catcher, quencher, venturi scrubber Dust collector, primary orifice scrubber, secondary venturi scrubber, spray chamber-type gas cooler Mechanical dust collector, venturi scrubber Dust catcher, variable throat venturi scrubber, gas cooler 4.2.4 Capture and Control System Performance Casthouse capture efficiencies were reported by several companies in a 1993 industry survey. These efficiencies range from 50 to 99 percent based on engineering estimates. Control device efficiencies were on the order of 99 percent. 4.3 BOPF SHOP38 4.3.1 Primary Furnace Controls Primary emissions refer to those emissions leaving the mouth of the furnace vessel during the oxygen blow that are captured by the primary hood. Primary emission control systems are divided into two basic types: open full combustion and closed suppressed combustion; partial combustion systems also exist. Use of high energy venturi scrubbers and ESP have been the traditional, best demonstrated control technologies for controlling BOPF primary emissions. More recently, use of fabric filters has been proven to be effective, although this technology is not currently in use at any facility in the U.S. CO is emitted from the vessel mouth during the oxygen blow phase of the furnace cycle. The gas temperature is sufficiently hot to promote combustion of CO if air is permitted to mix with the waste gas. A design decision must be made to determine how much air, if any, is allowed to mix with the gas, so that hood cooling capacity can be matched to the needs of the system. Some air must be admitted to obtain sufficient capture velocity necessary to contain fume emissions within the hood. Capture velocities generally run 14 to 58 feet per second. 4-17 ------- Many BOPF furnace installations use ESP for controlling PM emissions. Because of the potential for igniting the CO/air mixture by precipitator sparking, it is necessary to use an open hood to admit large quantities of excess combustion air at the hood and to facilitate the complete combustion of CO. This design decision leads to larger gas volumes to be treated for control of particulate emissions than is necessary for closed hood furnaces. More recent designs have incorporated limited or partial combustion of CO (closed hood design), reducing the heat generated in the hood and the volume of gas to be treated. Careful control of the amount of air entering the hood allows 10 to 50 percent combustion of CO. Gas cleaning in closed hood systems is exclusively venturi scrubbers to reduce explosion hazards. The advantages of suppressed combustion (closed hood systems) are reduced energy consumption for gas cleaning as compared to full combustion and the potential for recovering CO as a low-grade fu el source. Ten BOPF shops and one vessel in an open hood BOPF shop currently operate with suppressed combustion hoods; however, none of the plants are recovering the CO, and the gas is generally flared before discharging it to the atmosphere. 4.3.1.1 Open Hood Designs. Both wet scrubbers and ESP are used to control emissions from open hood systems. In this system, the hood skirt is in a fixed position and no precautions for leakage into the system are necessary. Control systems may be shared between furnaces and multiple fans operating in a parallel flow arrangement may be used. When an ESP is used, gas cooling down stream from the hood skirt is continued by the use of water sprays located in the upper part of the hood. These sprays are generally controlled by time and temperature to turn on and off at various points in the operating cycle. The intent is to limit the gas temperature reaching the precipitator and to moisture condition the gases for better precipitation. Emissions during the oxygen blow are captured by the open hood, enter a hood cooling section, and pass through a conditioning chamber where the gas is cooled and humidified to the required levels for proper ESP operation. The gas cleaning system commonly consists of precipitators, fans, dust handling equipment, and a stack for carrying away the cleaned gases. ESP can be used with open hoods because the combustible CO generated during the oxygen blow burns at the mouth of the vessel, 4-18 ------- reducing the risk of explosions which could be set off by sparks in the precipitator. Alternatively, a venturi scrubber may be used to control emissions. Because there is less danger of explosion in the open hood system as compared to the closed hood system (most of the CO has been converted to carbon dioxide), all of the vessels in the shop may be connected to a common gas cleaning system. Control device parameters for open hood BOPF systems are presented for each facility in Table 4-6. The venturi scrubbers on open hood systems generally have an air flowrate of approximately 210,000 to 600,000 scfm and a pressure drop of 25 to 55 inches of water. The ESP on open hood systems generally have an air flowrate of 230,000 to 720,000 scfm and a plating area of approximately 80,000 to 650,000 ft2. 4-19 ------- TABLE 4-6. OPEN HOOD BOPF SHOP PRIMARY CONTROL SYSTEM Wet Scrubber Control Technology Plant Bethlehem11' 12 Bethlehem13' 14 Inland (#4)4' 5 USX, Gary6 USX, Gary (Q-BOP)6 USX, Braddock32 Weirton Steel36 Wheeling- Pittsburgh17 State IN MD IN IN IN PA WV OH Capacity (million tpy) 5.35 4.00 2.74 2.9 4.0 2.76 3.20 2.95 Flow (dscfm) 1 13,200 x O 600,000a 310,000- 380,000 268,000 267,000 174,000 280,000 210,000 L/G (gal/1000 ac 0 20 8 1.0 13.1 34.7 -- -- 10 - p (in. water) 55 50 25 70-75 70 68-76 50 50 Scrubber type Venturi Venturi Venturi Venturi Venturi Venturi Venturi Venturi Demister Pall rings Chevrons Yes Yes Yes Yes Wood Yes Location 3 scrubbers for 2 vessels (#1 & 2) 4 Scrubbers for 2 vessels 2 vessels 3 vessels 3 vessels 2 vessels 1 Scrubber for 2 vessels North & south scrubbers 4-20 ------- TABLE 4-6. OPEN HOOD BOPF SHOP PRIMARY CONTROL SYSTEM (continued) ESP Control Technology Plant Acme Steel20 Gulf States24 LTV Steel15 LTV (#1)25 National26 National26 Rouge Steel28 WCI Steel8 State IL AL IN OH IL MI MI OH Capacity (million tpy) 1.29 1.30 4.16 3.34 3.58 4.1 3.3 1.73 Flow (dscfm) 288,000 327,000 458,000 550,000 410,000 500,000a 500,000 400,000 ESP type Single Stage Single stage Single stage — — — — — Plate area ft2 92,000 150,000 650,000 255,000 -- 80,200 -- 114,000 # of fields in series O 8 5 4 4 4 4 6 Type bottom Dry Dry Dry Dry Dry Dry — — Cleaning method Rapping Rapping Rapping Rapping Rapping Rapping Rapping -- Conditioning Agents Water Water/steam Water/steam Water Water/steam Water/steam Humidification — aacfm 4-21 ------- Gulf States Steel ESP upgrade:'9 Gulf States Steel has an open hood BOPF shop with an 550,000 acfm primary gas cleaning system. Extensive developments were carried out to improve the effectiveness of the system. The system became operational in 1994 and has proven to be effective in reducing stack emissions to within regulatory limits. Air atomized spray nozzles were used to replace direct pressure nozzles in the spray chamber. The improved atomization reduced moisture and dust build-up in the off-gas ducting, as well as the ESP and dust handling system. These nozzles also improved the moisture content of the off-gas, lowering the dust resistivity and improving collection efficiency. However, during low temperature periods of the blowing cycle, the desired cleaning efficiency was not being achieved. Therefore, the plant decided to install a new precipitator system in parallel with the existing units. To determine the additional collection plate area, the precipitator performance was predicted during the entire blowing system for the existing system, 50 percent and 100 percent expansion. Based on stack opacity, the 100 percent expansion was required to provide acceptable stack opacity levels (10 percent) throughout the oxygen blowing cycle. The expanded system increased the specific collection area from 285 to 560 ft2/l,000 acfm. The expanded system increased the collection efficiency from 99 to 99.93 percent, and the outlet particulate concentration was reduced from 0.059 to 0.004 gr/acf (0.14 to 0.01 gr/dscf). 4.3.1.2 Closed Hood Designs ,38 In a closed hood system, the diameter of the hood face is roughly the same as the diameter of the mouth of the vessel. The hood usually fits close to the furnace mouth to restrict the inflow of combustion air. Because a completely closed hood would restrict vessel tilting necessary for charging and tapping the furnace, the hood skirt must be movable. The lower portion of the hood is a skirt that can be lowered onto the mouth of the vessel, sealing off the space between the hood and the vessel, thereby limiting the amount of air that can enter the system. The gas, mainly CO, is collected in an uncombusted state. The volume of gas collected in a closed hood system is reduced by as much as 80 to 85 percent as compared to that of an open hood system. In addition, there is a need to limit the amount of air infiltration downstream of the hood. Normal points of leakage 4-22 ------- in an open hood system such as the lance port and flux chutes must be sealed and purged of nitrogen before use in the closed hood system. Gas cleaning is performed by a scrubber to minimize the risk of explosion. The cleaned gas is usually flared at the stack. Because of the potential explosion hazard from leakage of air into the system from an idle furnace, the closed hood system must have a separate scrubber system for each vessel. Control device parameters for closed hood BOPF systems are presented for each facility in Table 4-7. Initial cooling of the gas leaving the furnace is carried out using a water-cooled hood. Cooling is continued by the use of a spark box or quencher, in which grit and coarse particles resulting from refractory and chunks of slag or metal are separated from the gas stream. From the quencher, the waste stream flows to a high energy scrubbing device where the removal of fine particles occurs. The most common scrubber type is a venturi with an adjustable throat. The venturi is opened or closed to increase or decrease gas velocity, i.e., pressure drop through the throat. A critical part of the scrubbing unit is a moisture-separating device to knock out drops of water carried out of the throat. The device may be a series of baffles or a centrifugal chamber in which the gas rotates, causing the drops to impinge on the chamber walls. An after cooling chamber is occasionally used, in which the used cooling water is sprayed to further reduce the gas temperature. At cooler temperatures, moisture condenses from the gas, reducing the volume of gas to be handled by the fan. The system may have multiple venturi throats, but draft is provided only by a single fan. The gas cleaning facilities are not shared between adjacent furnace vessels; each furnace has an independent gas cleaning system. All closed hood systems in the U.S flare the CO-rich waste gas stream generated during oxygen blowing. The venturi scrubbers on closed hood systems generally have an air flowrate of approximately 40,000 to 268,000 scfm, a pressure drop of 40 to 80 inches of water, and a liquid-to-gas ratio of 2.6 to 34.7 gal/1,000 acf 4-23 ------- TABLE 4-7. OPERATING PARAMETERS OF CLOSED HOOD BOPF SYSTEMS-VENTURI SCRUBBERS Plant AK Steel22 AK Steel9 Bethlehem Steel11 Geneva (Q- BOP)3 Inland Steel (No. 2)4 LTV Steel (No. 2)25 USS/Kobe34 USS Steel30 State KY OH IN UT IN OH OH AL Capacity (million tpy) 2.17 2.71 -- 2.5 2.5 4.38 2.6 2.2 Vessel 1 2 15 16 3 1 2 1,2 1 2 L N U X c Flow (dscfm) 78,000 78,000 40,000 51,000 197,000a 78,300 77,300 50,000- 60,000 55,000 55,000 58,000 59,000 — 76,000 76,000 L/G (gal/1000 acf) 11.5 11.5 2.9 2.6 21 -- -- 10 -- -- -- -- -- -- -- - p (in. water) 60 60 45-50 40-50 55 70-80 70-80 55 -- -- -- -- 60-95 51-92 59-96 Scrubber type Venturi Venturi Venturi Venturi Venturi Venturi Venturi Venturi Venturi Venturi Venturi Venturi Venturi Venturi Venturi Demister Yes Yes No No -- -- -- Yes -- -- Yes Yes -- -- -- Efficiency (%) 99+ (E) 99+ (E) 99+ 99+ — 99 99 99.8 (E) 99.9 99.9 99+ 99+ — — — aacfm 4-24 ------- 4.3.2 Secondary Sources of Emissions Secondary sources of emissions within a BOPF shop include hot metal transfer, desulfurization, slag skimming, charging, turndown, tapping, deslagging, teeming, ladle maintenance, flux handling, slag handling and disposal, and ladle metallurgy operations. Most facilities use a combination of one or more baghouses or, less frequently, wet scrubbers, to control secondary BOPF shop emissions. Capture and control systems are described in detail in the following sections. Following the general description of the controls, Table 4-8 presents the controls currently used for each emission point at the various facilities and Table 4-9 presents the operating parameters for each control device. 4.3.2.1 Furnace Controls ,38 Emissions that occur during the steps of the furnace cycle that require the vessel to be tipped out from under the hood include scrap charging, hot metal charging, sampling, tapping, and deslagging. These sources are often poorly controlled by the primary system. When the BOPF vessel is tipped out from under the hood of the primary control system, whether for charging, sampling, or tapping refined steel, the primary control system may be rendered entirely ineffective. Secondary furnace emissions are typically produced by unconfmed sources such as leaks from the primary furnace hood or the open top of a ladle. These emissions may be captured by enclosures or hoods and ducted to a paniculate control device. Capture techniques for secondary furnace emissions include furnace enclosures, local hoods, full or partial building evacuation, and, in the case of open hood systems, adapting the primary furnace hooding to also capture secondary emissions. Particulate removal techniques that are currently in use include baghouses and wet scrubbers. These systems are described in detail below. Furnace enclosures. A furnace enclosure is a structure that may partially (on at least two sides) or fully (on four sides plus the top) enclose a furnace vessel. Most recently constructed BOPF vessels are enclosed. A partial enclosure may be designed to shield the BOPF from most drafts, other natural convection, permitting hoods within or adjacent to the enclosure to be more effective at lower air flow rates. In comparison to a full enclosure, a partial enclosure is less expensive, easier to retrofit (possibly without interrupting production), and less likely to impede operations. 4-25 ------- In a total enclosure system, the enclosure can be relatively simple on two sides because the vessel is designed routinely to tilt about only one horizontal axis. The enclosure roof is usually penetrated by the primary exhaust duct, and it must be high enough to permit maneuvering the hood in a closed system. Similarly, the flux chute and the oxygen lance of top blown vessels must penetrate either the roof of the enclosure or the primary hood. Within the enclosure, and sometimes as part of the enclosure, there may be charging and tapping hoods. The enclosure can extend partially or completely to the operating floor at the rear-facing tapping aisle. Tapping is carried out at and below the level of the vessel, and there is a tendency for hot, dusty gases to escape in the natural draft induced by the process heat. A hood that is either permanently arranged so that it does not interfere with operations or that is otherwise retractable to collect tapping emissions is preferred. Most of the complications resulting from full enclosure arise in the front facing charging aisle. This side of the enclosure includes a door that is moved out of the way while charging scrap and hot metal. Because these operations occur at and above the vessel, natural convection will permit a plume of hot dusty gas to escape into the building. The secondary control system (capture plus paniculate removal) may be an extension of the primary control system. Hoods designed to capture charging and tapping emissions may be ducted to the primary system. Gas flow may also be adjusted for the differing demands of several parts of the cycle. In a closed hood system, the typical arrangement is to duct the charging and tapping hoods in the furnace enclosure to a secondary control unit, most commonly a baghouse. Furnace operations dictate the necessity for opening and closing the doors on a furnace enclosure. For a total enclosure, the charging of scrap and hot metal to the furnace requires the doors to be open; immediately following hot metal charging, the doors may be closed. As the oxygen blow portion of the cycle is completed, it is necessary to take a metal sample and measure the metal temperature; most furnaces must be turned down to do this. Another opening in the enclosure door may be provided to insert a thermocouple and sampling spoon. Where such an opening has not been provided, it is necessary to open the doors at least partially, which may cause poor control of emissions during the sampling period. If the doors are left open for the remainder of the production cycle, 4-26 ------- generally poorer capture of secondary furnace emissions can be expected. Doors on the tapping side of the enclosure generally do not need to be opened except for maintenance. Primary control systems used for secondary emission control. Consent decrees negotiated between EPA and steel companies have included provisions for reducing roof monitor discharges from BOPF shops. In several instances, roof monitor emissions have been decreased to levels complying with consent decree terms by using primary emission control systems to capture charging and tapping emissions. The use of the primary system to achieve compliance has been strengthened by the adoption of operating practices conducive to lesser fume generation and by the modification of, and in addition to, process equipment and pollution control. Those shops with relatively large flow capacity in their primary control system are better suited to achieving low roof monitor emissions from furnace operations. Higher flow capacity means that higher indraft velocities can be achieved to capture fugitive emissions at a given distance from the hood. In addition, the use of clean, non-oil-bearing, non-galvanized scrap, the positioning of the hot metal ladle with respect to the hood face and furnace mouth, and the proper furnace tilt angle are all means of reducing charging emissions. Extension (flanges) from the primary hood into the charging and tapping aisles helps to provide more draft closer to the points of emission. Similarly, an extension of the pouring spout on the hot metal charging ladle will move the emission generation point closer to or under the hood. Canopy or roof hoods, partial building evacuation. The design of hoods for BOPF shop secondary emissions is complicated by cross drafts that develop within the building, interfering with fume capture. A hood that is located close to the source and intended to reduce cross drafts may get in the way of crane operations. Every design is a compromise between hood and vessel clearance and the clearance necessary for crane operations. In addition to emissions that are collected regularly at fixed locations, certain necessary maintenance operations generate dust that is less susceptible to collection by local hoods. The canopy hood is one method for collecting some emissions that have either not been provided for or that inevitably escape the local hoods. A canopy hood will not interfere with furnace 4-27 ------- operations, can collect the fine, entrained particles at relatively low velocities, and can be ducted continuously to a collecting device. Disadvantages to canopy hoods include: (1) cross drafts in the shop that displace rising fume so that it evades the hood or the face; (2) a significantly larger volume of gas to be cleaned; and (3) when added to an existing system, canopy hoods may reduce draft in the rest of the system to the point that air velocity in the other hoods is too low to capture fume effectively. One method of reducing the impact of cross drafts and avoiding the problem of the plume's becoming larger than the hood face dimensions is to use partial building evacuation. The building structure becomes the hood for a particular portion of the operation. Partition walls may be installed between building columns to prevent lateral movement of the plume into adjacent portions of the building. These partition walls may extend as low as crane operations will permit and may extend as high as the roof. Sheeting or partitions may also be used to seal the roof area to prevent the escape of emissions by natural thermal draft. One or more duct connections may be made into the sealed portion of the building to extract contaminated air for gas cleaning. 4.3.2.2 Ancillary Operations. Ancillary operations, including hot metal transfer, desulfurization, and slag skimming are usually controlled by hooding ducted to a control device separate from the primary control device, although one facility uses the primary furnace ESP to control secondary emissions in the BOPF shop. Inland Steel No. 2 BOPF shop39 The hot metal transfer baghouse at Inland's No. 2 BOPF shop was upgraded in June 1994 in order to optimize the existing equipment. The 400,000 acfm negative pressure shaker baghouse operated at excessively high pressure drop, reducing system flow capacity and causing dust to bleed through the bags. As part of the overall secondary emission control system upgrade, a baghouse appraisal study was completed to help define the problems. The investigation indicated that the absence of hopper air lock valves and leaks in the screw conveyor dust disposal system caused dust reintrainment which prevented regular dust disposal and resulted in a slow, steady rise in bag pressure drop, even with proper cleaning. The primary cause of bag failure was identified as abrasion resulting from under- tensioning of the bags in their attachment to the shaker mechanism. The strap bag attachment induced 4-28 ------- the bags to fold during shaking, which restricted dust removal. Ingress of moisture through poorly sealing access doors allowed bags to get wet, resulting in crust formation on the bags. A high degree of shaker maintenance was attributed to generally poor mechanical design and aggravated by wear on a knife edge support at the far ends of the shaker logs. In addition to mechanical problems, the original hard-wired relay control system was found to be unreliable and too difficult to maintain. A new bag design, complete with a spring-tensioned attachment, and top and bottom sewn rings was installed in a test compartment and operated for several weeks. The new bag design was subsequently installed in all 18 compartments. A new screw conveyor system was installed utilizing a rotary air lock at each compartment hopper to eliminate reintrainment of dust through the hopper discharge conveyors. Other modifications included replacement of all compartment doors with a new design that provided better sealing, the replacement of butterfly outlet dampers with poppet dampers on all compartments, and a new PLC baghouse control system. Baghouse performance was greatly improved as a result of the modifications. The bag pressure drop was reduced to 6 in. water and the system capacity was restored to the original design. 4.3.3 Ladle Metallurgy Operations After hot metal is refined into steel in the BOPF vessel, further alloy additions and refining of the steel occur during ladle treatment and vacuum degassing. Most BOPF shops have a separate ladle metallurgy station. Emissions are generally captured and controlled from ladle metallurgy operations using a baghouse, although one facility uses a wet scrubber. Several facilities also use a wet scrubber to control emissions from vacuum degassing operations. The control device parameters for each facility are presented in Table 4-10. 4-29 ------- TABLE 4-8. SECONDARY EMISSION CONTROL SYSTEMS IN THE BOPF SHOP Plant Acme Steel, Riverdale, IL20 AK Steel, Ashland, KY22 AK Steel, Middletown, OH9 Bethlehem, Bums Harbor, IN (3 vessels in 1 shop/1 Bethlehem, Sparrows Pt, MD13 Geneva Steel, Orem, UT3 Gulf States Steel, Gadsden, AL20 Inland Steel, East Chicago, (o) IN (2 shops)4 (c) LTV Steel, East Chicago, IN15 LTV Steel, Cleveland, OH (2 (o) shops)" (c) National Steel, Granite City, IL26 National Steel, Ecorse, MP6 Rouge Steel, Dearborn, MP8 USX, Fairfield, AL30 USX, Gary, IN (2 shops)6 (o) (0*) USX, Braddock, PA32 USS/Kobe Steel, Lorain, OH34 WCI Steel, Warren, OH8 Weirton Steel, Weirton, WV36 Wheeling Pittsburgh Steel, Mingo Junction, OH17 Secondary Emission Controls HMReladle HM desulf Skimming Charging Tapping Baghouse with canopy hoods Baghouse with canopy hoods Baghouse Baghouse Baghouse Baghouse None Baghouse Baghouse Baghouse None Baghouse with canopy hoods Baghouse None Baghouse Baghouse with side draft hoods Baghouse None !• Scrubber !• Scrubber Baghouse, doghouse !• ESP Baghouse Scrubber !• ESP Flame suppression and tapside enclosure !• ESP Baghouse with multiple hoods controlled by dampers Baghouse Baghouse Baghouse Baghouse Baghouse Baghouse (2) Fume supp. Fume supp. Baghouse- 1 Baghouse- 1 Baghouse Baghouse Flame supp. Baghouse Baghouse Baghouse None None None Hood to !• ESP !• ESP, doghouse Baghouse !• ESP Baghouse** 1 • scrubber 1 • scrubber None Enclosure to baghouse Baghouse None Baghouse Baghouse Baghouse !• SCR Baghouse, enclosure !• ESP None Slow pour, !• SCR !• SCR 4-30 ------- 1 • = primary furnace control BH = baghouse o = open SCR = scrubber c = closed * Bottom blown ** To be installed by 2000. 4-31 ------- TABLE 4-9. SECONDARY CONTROL DEVICE PARAMETERS Plant Acme Steel20 AK Steel22 AK Steel9 Bethlehem Steel13 Bethlehem Steel11 Geneva Steel3 Gulf States Steel24 Inland Steel4 State IL KY OH MD IN UT AL IN Capacity (million tpy) 1.29 2.17 2.71 4.0 5.4 2.5 1.3 2.5 4.7 Shop 1 1 1 1 1 1 1 2 4 Flow (dscfm) 227,500 450,000 149,000 40,000 200,0001 SO^OO1 40,0001 135,000- leo^oo1 30,700 ISO^OO1 288,000 167,000 193,000 470,000 Air/cloth ratio (acfm/ft2) 4.0 4.8 5.0 4.9 4.3 5.0 4.9 4.1-5.1 -- 3.9 3 5.2 4.3 5 - p (in. water) 6-11 5 2-5 4-12 10 4 6 4-18 -- 5 -- 7.2 10 6 Cleaning Pulse jet Pulse jet Pulse jet Pulse jet Pulse jet Pulse jet Pulse jet Pulse jet Shaker Pulse jet Shaker Pulse jet Pulse jet Pulse jet Filter material Polyester Felt Polyester — Polyester Polyester Polyester Polyester Polyester Polyester Polyester Polyester Polyester Polyester Nomex Location HMT, DS, SS, C, T HMT, DS, SS, C, T HMT,DS SS HMT DS SS HMT, DS, SS DS HMT, DS, SS HMT, DS, SS HMT,DS HMT,DS C, T, F 4-32 ------- TABLE 4-9. SECONDARY CONTROL DEVICE PARAMETERS (continued) Plant LTV Steel15 LTV Steel25 National Steel26 National Steel26 Rouge Steel28 USS/Kobe34 USX30 USX6 USX32 State IN OH IL MI MI OH AL IN PA Capacity (million tpy) 4.2 3.3 4.4 2.6 3.5 3.4 2.6 2.2 2.9 4.0 2.8 Shop 1 1 2 1 1 1 1 1 BOPF Q-BOP 1 Flow (dscfm) 220,000! 163,000! 680,000! 90,000 30,000 210,000 500,000 106,000 68,000 349,000 126,000 480,000 -- -- 124,600 Air/cloth ratio (acfm/ft2) 5.0 2.0 5.4 2.8 3.4 2.8 2.8 2.1 -- 5.45 3.3 2.6 -- -- -- - p (in. water) 10 6 4-6 10.0 11.7 6-8 -- -- -- 3-10 -- 1.172 -- -- -- Cleaning Pulse jet Shaker Pulse jet Shaker Shaker Shaker Shaker -- -- Pulse Jet -- -- -- -- -- Filter material Polyester felt Polyester Nomex Polyester Orion Polyester Polyester — — Polyester Dacron Nomex — — — Location HMT, DS, SS HMT, DS, SS HMT, DS, SS, C, T HMT,DS SS HMT, DS, SS C, T HMT DS, SS C, T, HMT HMT,DS HMT,DS HMT,DS HMT, C, DS HMT,DS 4-33 ------- TABLE 4-9. SECONDARY CONTROL DEVICE PARAMETERS (continued) Plant WCI Steel8 Weirton Steel36 Wheeling Pittsburgh17 Plant Inland Steel4 State OH WV OH State IN Capacity (million tpy) 1.73 3.2 2.6 Capacity (million tpy) 2.5 Shop 1 1 1 Shop 2 Flow (dscfm) 450,700 -- 100,000! 150,000! 181,200 80,000 Air/cloth ratio (acfm/ft2) -- -- 4.5 5 -- -- - p (in. water) -- -- 5 6 3-5 5-6 Cleaning -- -- Pulse jet Pulse jet Pulse jet Pulse jet Plants with wet scrubbers Flow (dscfm) 100,000 L/G (gal/1000 ac f) -- - p (in. water) 35-45 Scrubber type Venturi Filter material — — Nomex Polyester Nomex Nomex Demister No Location C HMT,DS HMT DS, SS HMT DS, SS Location C, T 1acfin 2in. Hg HMT= hot metal transfer DS = desulfurization SS= slag skimming C = charging T = tapping 4-34 ------- TABLE 4-10. LADLE METALLURGY STATION CONTROL DEVICE PARAMETERS LMF stations with baghouses Plant Acme Steel20 AK Steel22 AK Steel9 Bethlehem Steel11 Bethlehem Steel13 Gulf States Steel24 Inland Steel4 LTV Steel15 LTV Steel25 State IL KY OH IN MD AL IN IN OH Capacity (million tpy) 1.29 2.17 2.71 5.35 4.00 1.30 2.50 4.16 3.34 4.38 Shop 1 1 1 1 1 1 2 1 1 2 Flow (dscfm) 110,000 40,000 6,000 13,500a 45,000a 120,000a 70,000a 45,000- 120,000 144,000 192,800a 120,000a Air/cloth ratio (acfm/ft2) 3.9 5.5 -- 5.4 6.7 3.8 -- 5 3.9 4.3 -- - p (in. water) 6-11 5-6 -- varies 8 4 -- 2-8 5 0-6 5 Cleaning Pulse jet Pulse jet Pulse jet Pulse jet Pulse jet Pulse jet Pulse jet Pulse jet Pulse jet Pulse jet Pulse jet Filter material Polyester felt Polyester felt — Polyester Polyester Polyester Polyester Polyester felt Polyester felt Polyester Nomex Location LMF at Continuous Caster LMF Cas ob Material handling LMF LMF LMF LMF LMF LMF LMF 4-35 ------- TABLE 4-10. LADLE METALLURGY STATION CONTROL DEVICE PARAMETERS (continued) LMF stations with baghouses Plant National Steel26 National Steel26 Rouge Steel28 USS/Kobe34 USX30 USX (Q- BOP)6 WCI Steel8 Weirton Steel36 Wheeling- Pittsburgh17 State IL MI MI OH AL IN OH WV OH Capacity (million tpy) 2.58 3.50 3.3 1.4 1.2 2.20 4.0 1.73 3.20 2.60 Shop 1 1 1 1 2 1 2 1 1 1 Flow (dscfm) 60,000a 165,000 144,000 37,400 37,700 60,000 -- -- -- 8,000a 40,000a Air/cloth ratio (acfm/ft2) 3.5 2.9 -- -- -- 5.7 2.4 -- -- 4.5 5 - p (in. water) -- -- -- -- 3-12 4-12 -- -- -- 5 6-8 Cleaning Shakeout pulse Shaker Pulsejet Pulsejet Pulsejet Pulsejet Pulsejet -- -- Pulsejet Pulsejet Filter material Woven polyester Polyester — — Nomex Gortex Nomex — — Nomex Nomex Location LMF LMF LMF LMF LMF LMF LMF LMF LMF LMF LMF 4-36 ------- TABLE 4-10. LADLE METALLURGY STATION CONTROL DEVICE PARAMETERS (continued) LMF Stations with wet scrubbers Plant AK Steel9 Inland4 LTV Steel25 State OH IN OH Capacity (million tpy) 1.71 2.74 4.38 Shop 1 4 2 Flow (dscfm) 2,200 3,100a 72,000a L/G (gal/1000 ac f) — — — - p (in. water) -- -- -- Scrubber type Condenser Condenser Hot well Demister -- -- -- Location Vacuum degassing Vacuum degassing Vacuum degassing ' acfm; LMF = ladle metallurgy 4-37 ------- 4.4 REFERENCES 1. United States Steel. The Making, Shaping, and Treating of Steel. Published by the Association of Iron and Steel Engineers (AISE). Available from AISE at Suite 2350, Three Gateway Center, Pittsburgh, PA. 2. Carpenter,B., D. VanOsdell, D. Coy, and R. Jablin. Pollution Effects of Abnormal Operations in Iron and Steel Making - Volume H. Sintering, Manual of Practice. EPA-600/2-78-118b. June 1978. pp. 12-15. 3. Starley, J.R., Geneva Steel, Provo, UT to B. Jordan, EPA. Response to section 114 request. October 29, 1993. 4. Shoup, S.P., Inland Steel, East Chicago, IN to B. Jordan, EPA. Response to section 114 request. November 12, 1993. 5. Allie, G, Inland Steel, East Chicago, IN to P. Mulrine, EPA. Comments on draft background information document for integrated iron and steel plants. December 1, 1998. 6. Moniot, J.D., USS Technical Center, to B. Jordan, EPA. Transmitting final U.S. EPA screening information request for USS Gary Works. Prepared by ENSR Consulting and Engineering. September 5, 1991. 7. Calcagni, J., RTI, to P. Mulrine, EPA. Trip Report for Visit to USX Gary Works, Gary, IN. August 31, 1995. 8. Calcagni, J., RTI, to P. Mulrine, EPA. Trip Report for Visit to WCI Steel, Warren, OH. August 31, 1993. 9. Francis, S.L., AK Steel, Middletown, OH to B. Jordan, EPA. Response to section 114 request. September 5, 1991. 10. Felton, S., AK Steel, Middletown, OH to P. Mulrine, EPA . Comments on draft background information document for integrated iron and steel plants. November 30, 1998. 11. Riley, W. J., Bethlehem Steel, Burns Harbor, IN to B. Jordan, EPA. Response to section 114 request. February 14, 1994. 12. Ossman, G., Bethlehem Steel to P. Mulrine, EPA. Comments on draft background information document for integrated iron and steel plants. 4-38 ------- 13. Anderson, D.A., Bethlehem Steel, Sparrows Point, MD to B. Jordan, EPA. Response to section 114 request. August 29, 1991. 14. Ossman, G., Bethlehem Steel, to P. Mulrine, EPA. Comments on draft background information document for integrated iron and steel plants. 15. Thomas, M.I, LTV Steel, East Chicago, IN to B. Jordan, EPA. Response to section 114 request. November 29, 1993. 16. Piccarillo, B., LTV Steel, East Chicago, IN to P. Mulrine, EPA. Comments on draft background information document for integrated iron and steel plants. April 13, 1999. 17. Samples, W.R., Wheeling-Pittsburgh Steel, to B. Jordan, EPA. Response to section 114 request. November 12, 1993. 18. Benitez, J., Process Engineering and Design for Air Pollution Control, pp. 414-438, 1993. 19. Jablin, R, D. Coy, et al. Pollution Effects of Abnormal Operations in Iron and Steel Making - Volume HI. Blast Furnace Ironmaking Manual of Practice. EPA-600/2-78-118c. June 1978. 93pp. 20. Zibble, D., Acme Steel, Riverdale, IL to B. Jordan, EPA. Response to section 114 request. June 2, 1994. 21. Zibble, D., Acme Steel, Riverdale, IL to B. Jordan, EPA. Response to section 114 request. June 2, 1994. 22. Felton, S.S., AK Steel to B. Jordan, EPA. Response to section 114 request for AK Steel's Ashland Works. November 15, 1993. 23. Felton, S.S., AK Steel, Middletown, OH to P. Mulrine, EPA. Comments on draft background information document for integrated iron and steel plants. Noverber 30, 1998. 24. Stewart, E.M., Gulf States Steel, Gadsden, AL to B. Jordan, EPA. Response to section 114 request. February 11, 1994. 25. Nemeth, R.L., LTV Steel, Cleveland, OH to B. Jordan, EPA. Response to section 114 request. December 16, 1993. 26. Heintz, J.K., National Steel to B. Jordan, EPA. Response to section 114 request for National Steel's plants in Granite City, IL and Ecorse, MI. January 31, 1994. 4-39 ------- 27. Heintz, J.K., National Steel, Mishawaka, IN to P. Mulrine, EPA. Comments on draft background information document for integrated iron and steel plants. 28. Earl, J., Rouge Steel, Dearborn, MI to P. Mulrine, EPA. Response to section 114 request. November 16, 1993. 29. Earl, J., Rouge Steel, Dearborn, MI to P. Mulrine, EPA. Comments on draft background information document for integrated iron and steel plants. May 10, 1999. 30. Dilanni, L.G., USX to B. Jordan, EPA. Response to section 114 request for USX Fairfield Works. December28, 1993. 31. Kubiak, W. U.S. Steel to P. Mulrine, EPA. Forwarding information on U.S. Steel's sinter, iron and steel production facilities. December 4, 1998. 32. Dilanni, L.G., USX to B. Jordan, EPA. Response to section 114 request for the USX plant in Braddock, PA. December 28, 1993. 33. Kubiak, W., U.S. Steel to P. Mulrine, EPA. Forwarding information on U.S. Steel's sinter, iron and steel production facilities. December 4, 1993. 34. Stinson, R., USS/Kobe Steel, Lorain, OH to B. Jordan, EPA. Response to section 114 request. November 16, 1993. 35. Ames, H., USS/Kobe Steel, Lorain, OH to P. Mulrine, EPA. Comments on draft background information document for integrated iron and steel plants. December 15, 1999. 36. Current, G.P., Weirton Steel, Weirton, WV to B. Jordan, EPA. Response to section 114 request. January 6, 1994. 37. Calcagni, J., RTI, to J. Myers, EPA. Trip Report for visit to Inland Steel, East Chicago, IN. JuneS, 1993. 38. Revised Standards for Basic Oxygen Process Furnaces - Background Information for Proposed Standards. EPA-450/3-82-005a.. December 1982. 39. Cesta, T., Optimization of BOPF Air Emission Control Systems. Iron and Steel Engineer, July, 1995, pp. 23-31. 4-40 ------- 5.0 EXISTING STATE REGULATIONS 5.1 SINTER PLANT 5.1.1 Windbox There are nine sinter plants in the U.S.; however, only seven were operating in 2000. The windbox exhaust is controlled by a baghouse at four plants and by a venturi scrubber at five plants. State emission limits for the windbox are given in Table 5-1. Most of the limits are in concentration units of gr/dscf; however, two States have limits in Ib/hr, and one has a limit in Ib/ton. TABLE 5-1. SINTER PLANTS IN THE U.S. Plant Inland USS Geneva* WCI Steel LTV Bethlehem Bethlehem Wheeling-Pittsburgh* AK Steel State IN IN UT OH IN IN MD WV OH Control Baghouse Baghouse Baghouse Baghouse Scrubber Scrubber Scrubber Scrubber Scrubber PM emission limit 0.007 gr/dscf 0.01 gr/dscf 0.0122 gr/dscf; 27 Ib/hr 501b/hr 0.02 gr/dscf 0.277 Ib/ton 0.03 gr/dscf 0.03 gr/dscf 501b/hr * These plants were not operating in 1999 - 2000. 5.1.2 Discharge End The sinter plant discharge end is comprised of sinter breakers (crushers), hot screens, conveyors, and transfer points that are designed to separate undersize sinter and to transfer the hot sinter to the cooler. In most cases, these discharge end operations are housed in a building. Emissions are usually controlled by local hooding and ventilation to one or more baghouses or wet scrubbers. Seven plants use baghouses and two plants use wet scrubbers. Details on existing limits are given in 5-1 ------- Table 5-2. For comparison purposes, the equivalent concentration limits were estimated for plants with limits expressed as a mass rate (Ib/hr) based on the typical volumetric flow rate. The PM limits for control devices vary substantially from plant to plant both in terms of format and numerical values. Four plants have concentration limits for total PM (0.01, 0.02, 0.02, and 0.03 gr/dscf), one has concentration limits for PM10, and three have mass rate limits (42.9, 50, and 50 Ib/hr). Existing State regulations also include both building opacity standards to limit releases of fugitive emissions (those escaping capture). As shown in Table 5-3, five of the seven operating sinter plants are subject to a building opacity limit. One plant is subject to a 10 percent limit (6-minute average), and four plants are subject to 20 percent limits (6-minute average). 5.1.3 Sinter Cooler Sinter plant coolers are large diameter circular tables through which ambient air is drawn to cool the hot sinter after screening. Seven plants operate sinter coolers to cool the sinter product prior to storage. Two plants that are not currently operating have no cooler and stockpile hot sinter directly. Of the seven plants with coolers, three vent directly to the atmosphere, one vents to a cyclone, two vent to a baghouse, and one vents half of the cooler exhaust to a baghouse with the remainder vented directly to the atmosphere. Five plants have emission limits expressed as concentration or mass rate while two plants have no emission limits (see Table 5-4). 5-2 ------- TABLE 5-2 . CONTROLS AND EMISSION LIMITS FOR THE DISCHARGE END Plant AK Steel, OH Bethlehem, MD Bethlehem, IN Geneva, UT Ispat-Inland, IN LTV, IN USX Gary, IN WCI, OH Wheeling- Pittsburgh, WV Control Baghouse Baghouse Baghouse Rotoclones (scrubbers) Baghouse Scrubber Baghouse 1 Baghouse 2 Baghouse A Baghouse Emission Points discharge, crusher, hot screen, cooler discharge, crusher, hot screen, cold screen discharge, crusher, hot screen discharge discharge, crusher, hot screen, 1A cooler discharge discharge, crusher hot and cold screens, conveyors discharge, crusher, hot screen, cold screen discharge Emission limit 50.0 Ib/hr 0.03 gr/dscf 42.9 Ib/hr 0.0096 gr/dscf PM10 0.01 gr/dscf 0.02 gr/dscf 0.02 gr/dscf PM10 0.0052 gr/dscf PM10 50.0 Ib/hr 0.02 gr/dscf Flow rate (dscfm) 112,000 340,000 212,000 105,000 122,000 100,000 161,322 180,000 141,470 32,900 Best estimate of TSP (gr/dscf 0.05 0.03 0.024 — 0.01 0.02 — — 0.04 0.02 5-: ------- TABLE 5-3. DISCHARGE END FUGITIVE EMISSIONS: OPACITY LIMITATIONS Plant Bethlehem, Sparrows Point, MD Ispat-Inland, East Chicago, IN LTV Steel, East Chicago, IN USX Steel, Gary, IN Geneva Steel, Provo, UT Limit for sinter building and fugitives 10% (6-min average) 20% (6-min average) 20% (6-min average) 20% (6-min average) 20% (6-min average) TABLE 5-4. SINTER COOLER DESCRIPTIONS AND LIMITS Plant Ispat-Inland WCI Steel Bethlehem, Sparrows Point USS, Gary AK Steel, OH Bethlehem, Bums Harbor LTV, East Chicago Description Baghouse controls the discharge, scrubber, hot screen and !/2 of cooler (one quadrant where the sinter is transferred to the cooler and one quadrant where it is removed); the other half is covered and vents through an uncontrolled stack. 20 minute residence time. Baghouse flow is 120,000 dscfm. Baghouse with forced air at 189,000 dscfm Cyclone at 320,000 dscfm and 0.02 gr/dscf; 90 to 120 min residence time 3 coolers, uncontrolled; with hood and stack; 360,000 dscfm each Baghouse controls discharge, crusher, hot screen and cooler; flow of 1 12,000 dscfm Uncontrolled, with hood over cooler; 30-ft diameter and 575,000 dscfm; 60 min residence time Uncontrolled; 60-ft diameter and 320,000 dscfm; 100 min residence time Limit 0.01 gr/dscf (for controlled portion) 42.9 Ib/hr (about 0.027 gr/dscf) 0.03 gr/dscf 0.03 gr/dscf 50 Ib/hr (about 0.05 gr/dscf) no limit no limit 5-4 ------- Geneva Steel Wheeling-Pittsburgh These plants do not have coolers. Sinter is transferred from the hot screen to a storage pile and cooled by ambient air. Wheeling-Pittsburgh also uses water sprays. 5.2 BLAST FURNACE The casthouse is a building or structure that encloses the section of the blast furnace where hot metal and slag are tapped from the furnace. These emissions are controlled in one of two fundamentally different ways, flame suppression or conventional ventilation practices and control. Flame suppression consists of blowing natural gas over the iron runners and torpedo cars. The combustion of the gas consumes oxygen, which retards (suppresses) the formation of emissions. Ventilation practices employed include the use of localized hooding and ventilation applied at the iron trough and iron and slag runners. Alternatively, the casthouse may be totally enclosed and evacuated. Eighteen of the 39 blast furnaces have capture and control systems, 16 are controlled by baghouses and two are controlled by one wet scrubber. As a means for limiting fugitive emissions of PM from the casthouse during hot metal tapping, most States have developed visible emission standards that limit the opacity of emissions discharged from the casthouse roof monitor or other openings. As shown in Table 5-5, the most common limit is 20 percent (6-minute average), which is applied to 24 of the 39 casthouses. States also apply particulate limits on gases discharged from control devices used to capture tapping emissions. The most common form is a concentration limit, typically on the order of 0.01 gr/dscf (Table 5-6). 5-5 ------- TABLE 5-5. CASTHOUSE EMISSION CONTROLS AND OPACITY LIMITS Plant Acme Steel, IL AK Steel, KY AK Steel, OH Bethlehem Steel, IN Bethlehem Steel, MD Geneva Steel, UT Gulf States Steel, AL Inland Steel, IN LTV Steel, OH LTV Steel, IN National Steel, IL National Steel, MI Rouge Steel, MI USX, PA USX, AL USX, IN USS/Kobe Steel, OH WCI Steel, OH Furnace A Amanda 3 C D L 1 2 3 1 7 5 6 Cl C5 C6 H3 H4 A B A B D B C 1 3 8 4 6 8 13 3 4 1 Casthouse control Flame suppression (FS), covered runners FS, covered runners Flame suppression Inert suppression, FS Inert suppression, FS Baghouse, evacuated runner covers & hoods FS, partially covered runners FS, partially covered runners FS, partially covered runners No controls Baghouse Scrubber Scrubber FS, covered runners FS, covered runners Fume suppression hoods FS, covered runners FS, covered runners, baghouse Baghouse, covered runners Baghouse, covered runners Baghouse Baghouse Baghouse Covered runners, FS Covered runners, FS Baghouse Baghouse Covered runners, Baghouse FS FS FS Baghouse, covered runners, evac. hood Baghouse, covered runners FS Baghouse Casthouse opacity limit 20%, 6 minute average 20%, 6 minute average Covered under a "bubble" No opacity limit No opacity limit 5%, 6 minute average, 20% drilling, O2 lance and mudding For all 3: 20%, except for any aggregate of 3 min. (12 readings) in any 60 min. None 15%, 6 minute average 20%, 6 minute average 20%, 6 minute average 20%, 6 minute average 15%, 6 min., w/ exceptions to 20% 20%, 6 minute average 20%, 6 minute average 20%, 6 minute average 20%, 6 minute average 20%, 6 minute average 20%, 6 minute average 20%, 6 minute average 20%, 6 minute average 20%, 6 minute average 20%, 6 minute average For both: Not to equal or exceed 20% except for 12 readings per hour 20%, 6 minute average 20%, 6 minute average 20%, 6 minute average 20%, 6 minute average 20%, 6 minute average 15%, 6 minute average 20%, 6 minute average 20%, 6 minute average 5-6 ------- Plant Weirton Steel, WV Wheeling Pittsburgh Steel, OH Furnace l 4 1 5 Casthouse control Covered runners, FS, baghouse Covered runners, FS Covered runners, FS Covered runners, FS, baghouse Casthouse opacity limit 20%, except 40% for 5 minutes/hour 20%, except 40% for 5 minutes/hour 20%, 6 minute average 5% to 20% 5-7 ------- TABLE 5-6. EMISSION LIMITS FOR CASTHOUSE CONTROL DEVICES Plant Bethlehem Steel, MD Ispat-Inland, IN LTV Steel, IN National Steel, IL National Steel, MI USX, PA USS/Kobe, OH WCI Steel, OH Wheeling- Pittsburgh, OH Furnace L 7 H4 A B A B D 1 3 3 1 5 Control Baghouse Baghouse 1 Baghouse 2 Baghouse Baghouse #1 Baghouse #2 Baghouse Baghouse Baghouse Baghouse Baghouse Baghouse Baghouse Baghouse Capture Points Evacuated runner covers & hoods Canopy hood Runners Hood over tilting spout & iron trough Suspended hood 6 air hoods, 3 at each furnace with damper control Hoods over trough & pouring spouts — each furnace Air curtain Evacuated runner covers & hoods Trough hood, covered runners, hood at tilting runners Emission Limit 0.03 gr/dscf 0.003 gr/dscf 0.0 11 gr/dscf No limit 0.01 gr/dscf 0.01 gr/dscf 0.0075 gr/dscf 0.02 Ib PM/1000 Ib exhaust 0.0052 gr/dscf No limit No limit 0.0052 gr/dscf 0.03 Ib/ton 0.31 Ib/hr; proposed PM10limitof5.93 Ib/hr 5-i ------- 5.3 BOPF SHOP 5.3.1 Primary Control Devices There are 50 BOPF located in 23 BOPF shops. The 50 BOPF include 34 furnaces with open hood systems at 16 shops and 16 furnaces with closed hood systems at 8 shops. All of the BOPF have capture and control systems for the primary emissions. For the open hood systems, 8 shops are controlled by venturi scrubbers and 8 shops are controlled by ESP. All 8 of the closed hood shops are controlled by venturi scrubbers. Open and closed hood vessels are very different in terms of operation, pollutant loading, and emissions. Open hood systems are characterized by very high primary exhaust air flowrates due to the large quantities of combustion air introduced at the furnace mouth to support CO combustion. In contrast, closed hood systems, which include hoods that are tightly fitted to the vessel to suppress CO combustion, are characterized by much lower exhaust air flowrates. Typical flowrates for open hood shops are 200,000 to 500,000 acfm, while closed hood designs are usually less than 100,000 acfm. Each shop is subject to existing State limits with a wide variety of formats, including concentration limits in gr/dscf and lb/1,000 Ib gas for PM or PM10, mass emission rate limits in Ib/hr, and process weighted limits in Ib/ton of steel. In addition, the emission test period required for compliance with the existing State limits varies from testing over the steel production cycle, only during the oxygen blow, for 1-hour runs, and for 2-hour runs. Emission limits are summarized in Tables 5-7 and 5-8. 5.3.2 BOPF Secondary Controls Secondary or fugitive emissions occur from the BOPF when the molten iron and scrap metal are charged to the furnace and when the molten steel and slag are tapped from the furnace. The emissions generated are primarily metal oxides formed when oxygen in the air reacts with the molten iron or steel. Twelve of the 23 BOPF shops have a separate capture and control system for BOPF charging and tapping emissions. Ten of these shops use baghouses and the other two use scrubbers. Existing State limits for the control devices are shown in Tables 5-9 and 5-10 and range from 0.0052 to 0.015 gr/dscf and the NSPS limit is 0.01 gr/dscf. The most common limit is 0.01 gr/dscf. 5-9 ------- TABLE 5-7. EMISSION LIMITS FOR PRIMARY CONTROL - OPEN HOOD Open Hood BOPF Shops Plant Acme Steel Bethlehem Steela Bethlehem Steel Gulf States Steel Ispat-Inland No. 4 LTV Steel LTV No. 1 Shop National Steel National Steel Rouge Steel USX Gary (BOPF) USX Gary(Q-BOP) USX Edgar Thomson WCI Steel Weirton Steel Wheeling-Pittsburgh State IL IN MD AL IN IN OH IL MI MI IN IN PA OH WV OH Control ESP Scrubber Scrubber ESP Scrubber ESP ESP ESP ESP ESP Scrubber Scrubber Scrubber ESP Scrubber Scrubber Emission Limit 0.028 gr/dscf 0.09 Ib/ton liquid steel 0.03 gr/dscf — 0.1871b/ton 0.018 gr/dscf PM10 39.8 Ib/hr 60.0 Ib/hr or 0.255 Ib/ton 0.057 lb/1000 Ib gas 0.02 gr/dscf PM10 0.02 gr/dscf PM10 Process rate 62.90 Ib/hr 0.03 gr/dscf 21.40 Ib/hr; 7.09 Ib/hr PM10 (pending) Two furnaces are open hood and one is closed hood. 5-10 ------- TABLE 5-8. EMISSION LIMITS FOR PRIMARY CONTROL - CLOSED HOOD Closed Hood BOPF Shops Plant AK Steel AK Steel Geneva Steel Inland No. 2 LTV No. 2 USS/Kobe Steel USX Fairfield State KY OH UT IN OH OH AL Control Scrubber Scrubber Scrubber Scrubber Scrubber Scrubber Scrubber Emission Limit 0.03 gr/dscf 1141b/hra 0.02 gr/dscf PM10 0.058 Ib/ton 15 Ib/hr (for each of 2 stacks) 45.0 Ib/hr 0.022 gr/dscf; c process rated 3 Both vessels combined b During oxygen blow c Furnace C, subject to NSPS, Subpart NN, which is 0.022 gr/dscf for closed hood shops d Furnaces X & U TABLE 5-9. LIMITS FOR SECONDARY CONTROL DEVICES AT CLOSED HOOD BOPF SHOPS Closed Hood BOPF Shops Plant Bethlehem Steel Geneva Steel Inland No. 2 Shop LTV No. 2 Shop USS/Kobe Steel USX Fairfield State IN UT IN OH OH AL Control Scrubber Baghouse Scrubber Baghouse Baghouse Baghouse Limit 0.05 Ib/ton liquid steel (#3) 0.002 gr/dscf 0.015 Ib/ton TSP 0.010 gr/dscf 0.0 12 gr/dscf 0.010 gr/dscf 5-11 ------- TABLE 5-10. STATE EMISSION LIMITS FOR SECONDARY CONTROL DEVICES AT OPEN HOOD BOPF SHOPS Open Hood BOPF Shops Plant Acme Steel Inland No. 4 Shop USX, Gary (Q-BOP) USX, Braddock State IL IN IN PA Control Baghouse Baghouse Baghouse Baghouse Actual Limit 10.22 Ib/hr, 0.0052 gr/dscf 0.006 gr/dscfTSP 0.0052 gr/dscf Process weight limit 1 gr/dscf PM 10 5.3.3 Hot Metal Transfer, Desulfurization, Slag Skimming, and Ladle Metallurgy There are several different ancillary operations performed within the BOPF shop: (1) operations associated with the molten iron before it is charged to the BOPF (hot metal transfer, desulfurization, and slag skimming), and (2) treatment of the molten steel after tapping (various ladle metallurgy operations). The emissions from these operations are primarily metal oxides formed when oxygen in the air reacts with the molten iron or steel. Molten iron is transported from the blast furnace casthouse to the BOPF shop in a torpedo car and transferred to a vessel at the reladling (or hot metal) station, where it is usually desulfurized and slag is skimmed from the surface. Emissions from these operations are captured by local hooding and controlled by a baghouse. Existing State emission limits for these operations shown in Table 5-11 range from 0.0052 to 0.04 gr/dscf, but most are on the order of 0.01 gr/dscf. The steel from the BOPF is usually transferred to a ladle where final adjustments in temperature and chemistry are made in an operation known as ladle metallurgy. Emissions from ladle metallurgy are captured by a close fitting hood and ducted to a baghouse. Existing State limits for ladle metallurgy shown in Table 5-12 are a mixture of mass emission rates in Ib/hr and concentration limits in gr/dscf. The mass emission rate limits range from 0.42 to 7.5 Ib/hr and the concentration limits range from 0.0052 to 0.02 gr/dscf. 5-12 ------- 5.3.4 BOPF Shop Roof Monitor The BOPF shop is a building or structure that houses several operations involved in steelmaking. These include hot metal transfer, desulfurization, slag skimming stations; one or more BOPF for refining iron into steel; and ladle metallurgy stations. Fugitive emissions from these operations in the BOPF shop exit through the roof monitor. States have set roof monitor opacity standards to limit these fugitive emissions (see Table 5- 13). The most stringent existing limit is the NSPS opacity limit of 10 percent (6-minute average, with one exception per cycle up to 20 percent). The most common standard is a 20 percent limit (3-minute average) that is applied to 14 of the 23 BOPF shops. In addition, there is an NSPS limit of 10 percent opacity during the steel production cycle of any top-blown BOPF or during hot metal transfer or skimming operations for any bottom-blown BOPF; except that an opacity greater than 10 percent but less than 20 percent may occur once per steel production cycle. 5-13 ------- TABLE 5-11. STATE LIMITS FOR TRANSFER, DESULFURIZATION, AND SLAG SKIMMING-ALL BAGHOUSES Plant Acme Steel AK Steel AK Steel Bethlehem Steel Geneva Steel Inland Steel No. 2 Inland Steel, No. 4 LTV Steel National Steel Rouge Steel National Steel USS, Edgar USS, Fairfield USS Gary Works, USS Gary USS/Kobe Steel WCI Steel Weirton Steel Wheeling-Pittsburgh Steel State IL KY OH IN UT IN IN IN IL MI MI PA AL IN IN OH OH WV OH Process Transfer, desulfurization, skimming Transfer, desulfurization, skimming Transfer and desulfurization Deslagger Transfer, desulfurization, skimming Desulfurization Buildings 1& 2 Reladle and desulfurization Reladle and desulfurization Reladle and desulfurization Transfer, desulfurization, skimming Transfer and desulfurization Hot metal transfer Reladle and desulfurization Reladle and desulfurization Desulfurization Reladle and desulfurization Transfer and desulfurization Desulfurization Hot metal transfer Desulfurization Hot metal transfer Desulfurization Hot metal transfer backup Emission Limit 10.2 Ib/hr 0.01 gr/dscf 58 Ib/hr 0.03 gr/dscf 23.1 Ib/hr 0.011 gr/dscf PM,n 0.011 gr/dscf 0.0052 gr/dscf 0.008 gr/dscf PM,n 0.01 gr/dscf 0.007 gr/dscf Process weight rate 0.01 gr/dscf 0.01 gr/dscf 0.0052 gr/dscf PM,n 0.03 gr/dscf 0.04 gr/dscf 0.01 gr/dscf 5.97 Ib/hr 5.01 Ib/hr (proposed) 6.41 Ib/hr (proposed) 5-14 ------- TABLE 5-12. STATE LIMITS FOR LADLE METALLURGY PROCESS Plant Acme Steel AK Steel AK Steel AK Steel Inland Steel, No. 2 LTV Steel National Steel National Steel National Steel National Steel National Steel Rouge Steel Rouge Steel USS Fail-field USS Gary Q-BOP USS Gary Q-BOP USS/Kobe Weirton Steel Wheeling-Pittsburgh Wheeling-Pittsburgh State IL KY OH OH IN IN IL IL MI MI MI MI MI AL IN IN OH WV OH OH Control Baghouse Baghouse Baghouse Baghouse3 Baghouse Baghouse Baghouse 1 Baghouse 2 Baghouse lb Baghouse 2a Baghouse 3C Baghouse 1 Baghouse 2 Baghouse Baghouse 1 Baghouse 2 Baghouse Baghouse Baghouse Baghouse Emission Limit 0.037 lb PM10/ton 3.8 Ib/hr 0.02 gr/dscf 0.03 gr/dscf 0.0052 gr/dscf 0.004 gr/dscf PM10 0.01 gr/dscf 0.01 gr/dscf 1.261b/hr 2.131b/hr 1.1 Ib/hr 7.50 Ib/hr 1.6 Ib/hr 0.02 gr/dscf 0.01 gr/dscf PM10 0.01 gr/dscf PM10 0.002 gr/dscf 0.42 Ib/hr 0.54 lb/hrd 2.3 Ib/hr, 0.02 gr/dscf Vacuum degassing Ladle metallurgy, No. 2 argon stirring No. 1 argon stirring station Proposed limit 5-15 ------- TABLE 5-13. SUMMARY OF BOPF ROOF MONITOR OPACITY LIMITS Plant Acme Steel, Riverdale, IL AK Steel, Ashland, KY AK Steel, Middletown, OH Bethlehem, Bums Harbor, IN (3 vessels in 1 shop) Bethlehem, Sparrows Point, MD Geneva Steel, Orem, UT Gulf States, Gadsden,AL Inland Steel, East Chicago, IN (2 shops) LTV, Cleveland, OH (2 shops) LTV, East Chicago, IN National, Granite City, IL National, Ecorse, MI Rouge Steel, Dearborn, MI USX,Braddock,PA USX,Fairfield,AL USX Gary, IN (2 shops) USS/Kobe, Lorain, OH WCI Steel, Warren, OH Weirton Steel Weirton, WV Wheeling-Pittsburgh, OH Open or closed Open Closed Closed Open(2) Closed(l) Open Closed* Open Closed Open Open Closed Open Open Open Open Open Closed* Open Open* Closed Open Open Open Primary control ESP Scrubber Scrubber Scrubber Scrubber Scrubber ESP Scrubber Scrubber ESP Scrubber ESP ESP ESP ESP Scrubber Scrubber Scrubber Scrubber Scrubber ESP Scrubber Scrubber Secondary control Baghouse Baghouse None None Scrubber None Baghouse None Scrubber Baghouse Baghouse None None Baghouse None Baghouse Baghouseb Baghouse Baghouse None None None Roof monitor opacity limit 20%, 3 minute average 20% except for 3 min/hr Covered under "bubble" 40%, 6 minute average; <60% for 15-min in 6 In- S-day roll avg of 15% (6-min avg), except 3 min/hr 10%, 6 minute average 20%, 3 minute average 20%, 3 minute average 20%, 3 minute average 20%, 3 minute average 20%, 3 minute average 20%, 3 minute average 20%, 3 minute average 20%, 3 minute average 20%, 3 minute average Not to equal or exceed 20% except for 12 readings per hour. 20%, 6 minute average 20%, 3 minute average 20%, 3 minute average 20%, 3 minute average None 20% 20%, 3 minute average a Bottom blown b Canopy hood baghouse controls emissions from "C" furnace only; new secondary control system under construction. 0 The NSPS for the roof monitor is 10 percent opacity based on 6-minute averages, except one period per cycle can go to 20 percent. 5-16 ------- 6. CONTROL COSTS 6.1 APPROACH The costs associated with improved emission control are based on what each plant may have to do with respect to upgrading or replacing emission control equipment. The estimates are worst case or upper bound estimates because they assume in several cases that plants will have to replace existing control equipment, when if fact, it may be possible to upgrade existing controls. The cost estimates are derived from industry survey responses, information from vendors, and procedures in EPA's manual for estimating costs. 6.2 BOPF PRIMARY CONTROL SYSTEMS Two plants were identified as candidates for upgrading or replacing their venturi scrubbers used as the primary control devices for BOPF. Ispat-Inland's Number 4 BOPF shop has three venturi scrubbers that are over 30 years old and were designed with a lower pressure drop (25 inches of water) than most scrubbers that are currently used. The company had performed an engineering analysis in 1990 to estimate the cost of replacing these scrubbers with higher pressure scrubbers.1 The estimate is based on an entirely new emission control system that includes three venturi scrubbers and three new capture hoods for the BOPF. The capital cost estimates are presented below and are indexed to 1998 dollars: Item Capital cost (millions of dollars) Three venturi scrubbers 11 Three new B OFF hoods 6.6 Engineering 0.7 Miscellaneous 0.4 Total ($1990) 18.7 Total ($1998) index = 389.5/357.6 20 The increase in operating cost for the new scrubbers is primarily the cost of increased energy (electricity) due to operating at the higher pressure drop. A cost function is provided in EPA's cost manual2 that expresses electricity cost as a function of the volumetric flow rate and pressure drop: Electricity cost ($/yr) = 0.00018 x acfm x • p x hrs/yr x $/kW-hr Estimates of electrical costs are given below for pressure drops of 25 and 50 inches of water based on 600,000 acfm , 8,760 hrs/yr, and $0.059/kW-hr: 6-1 ------- * p (in. water) Cost ($ millions/yr) 25 1.4 50 2.8 The increase in operating cost for the higher pressure drop scrubbers is estimated as $1.4 million per year. Test data indicated that the venturi scrubbers at AK Steel (Mddletown, OH) may require a minor upgrade to improve emission control. These scrubbers were designed with an adequate pressure drop (50 to 60 inches of water). However, the water supply system may need to be upgraded, and the scrubbers do not have demisters. Estimates obtained from a vendor (Coastal Technologies, Inc.) indicated that two demisters for two 72-inch diameter stacks would cost about $7,000 (316 stainless steel chevrons). The cost of new water supply piping2 for venturi scrubbers of this size was estimated as $10,600 for a total equipment cost of $17,600. Based on a retrofit factor of 1.3 and an indirect cost factor (from the cost manual2) of 36 percent of the purchased equipment cost, the total installed capital cost for the minor scrubber upgrade is estimated as $31,000. 6.3 SECONDARY CAPTURE AND CONTROL SYSTEMS Capture and control systems are used for fugitive emissions in many blast furnace casthouses and BOPF shops. Table 6-1 summarizes the capital and operating costs reported by several plants that use a baghouse as the control device. Only one plant reported no controls for their casthouse - Gulf States Steel in Gadsden, Alabama. This plant may be able to use flame suppression and covered runners to provide adequate control to meet an opacity limit for the casthouse. However, a worst case approach is used by assuming that a capture system and baghouse may need to be installed. Based on the cost for such a system as reported by USS/Kobe Steel in Table 6-1, costs are estimated as an installed capital cost of $3.3 million, an operating cost of $0.7 million per year, and a total annualized cost of $ 1.0 million per year (includes capital recovery based on a 20-year life and 7 percent interest rate.) 6-2 ------- TABLE 6-1. BAGHOUSE COSTS Plant Process Date installed Type Air: cloth Flow (acfin) Temperature (• F) Bag type Installed capital cost ($1998) Annual operating cost($1998/yr) Geneva Steel3 sinter windboxes 1993 pulse jet 4.0 540,000 275 polyester $4,300,000 $610,000 WCI Steel4 sinter windboxes 1991 pulse jet 4.0 400,000 300 Nomex® $4,700,000 $1,000,000 USS/Kobe5 blast furnace fugitives 1992 pulse jet 6.3 300,000 250 polyester $3,300,000 $730,000 Geneva Steel3 Q-BOP fugitives 1991 pulse jet 4.8 440,000 145 Nomex® $3,400,000 $460,000 AK Steel6 BOPF fugitives 1992 pulse jet 4.8 880,000 275 polyester — $500,000 Gulf States7 hot metal transfer, slag skimming, desulfurization 1992 pulse jet 3.9 150,000 250 polyester $4,300,000 -- 6-3 ------- AK Steel has a closed hood BOPF shop in Middletown, OH that does not have a secondary capture and control system. The cost of a new system, including a baghouse control device, is estimated from the costs reported by two plants in Table 6-1 (Geneva Steel and AK Steel in Kentucky): capital cost of $3.4 million, an operating cost of $0.5 million per year, and a total annualized cost of $0.8 million per year (includes capital recovery based on a 20-year life and 7 percent interest rate.) The MACT technology for secondary capture and control systems is a baghouse, and all plants except two use baghouses. Ispat-Inland and Bethlehem Steel (Burns Harbor, IN) use scrubbers as the control device for secondary emissions in the BOPF shop. There is uncertainty about the level of emission control these scrubbers can achieve. As a worst case, assume these scrubbers must be replaced by a baghouse at a capital cost of $3.4 million. There would be no increase in operating cost (the operating cost for baghouses would be less than the current operating costs for the scrubbers). 6.4 BAG LEAK DETECTION SYSTEMS Each baghouse will be equipped with a bag leak detection system. These systems have an installed capital cost of $9,000 each with an annual operating cost of $500/year8. There are approximately 88 baghouses at the 20 iron and steel plants. Consequently, the total capital cost for bag leak detectors is $0.8 million with an annual operating cost of $44,000/year. 6.5 TOTAL NATIONWIDE COSTS The nationwide costs are summarized in Table 6-2 and represent a somewhat worst case estimate because some of these plants may not have to install new controls. The nationwide capital cost is estimated as $34 million with a total annualized cost of $5.9 million/year. 6-4 ------- TABLE 6-2 NATIONWIDE COST ESTIMATES Source Gulf States, baghouse for casthouse AK Steel (Mddletown, OH), baghouse for secondary BOPF system AK Steel, BOPF scrubber upgrade Ispat Inland, new primary scrubbers and hoods for No. 4 BOPF shop (50" • p) Ispat-Inland, baghouse to replace scrubber for secondary BOPF system Bethlehem, Burns Harbor, baghouse to replace scrubber for secondary BOPF system Bag leak detection systems Total Capital ($ million) 3.3 3.4 0.03 20 3.4 3.4 0.8 34 Operating ($ million/yr) 0.7 0.5 0 1.4 0 0 0.04 2.6 Total annual ($ million/yr) 1.0 0.8 0.003 3.3 0.3 0.3 0.2 5.9 6.6 REFERENCES 1. Carson,! No. 4 EOF Gas Cleaning Upgrade (dated 9/21/90). Provided on April 6, 2000. 2. U.S. Environmental Protection Agency. OAQPS Control Cost Manual. 5th edition. EPA453/B- 96-001. February 1996. 3. Shaw, K.C. Geneva Steel's response to pollution control equipment cost survey. January 26, 1996. 4. Shepker, T. WCI Steel's response to pollution control equipment cost survey. January 12, 1996. 5. Stinson, R. USS/Kobe Steel's response to pollution control equipment cost survey. March 6, 1996. 6. Bradley, L. AK Steel's response to pollution control equipment cost survey. January 1996. 7. Stewart, E.M. Gulf States Steel's response to pollution control equipment cost survey. January 17, 1996. 5-5 ------- EPA. Supporting Statement for the Primary Lead MACT Information Collection Request. April 1998. 6-6 ------- 7. ENVIRONMENTAL IMPACTS 7.1 EMISSION REDUCTIONS There are four integrated iron and steel plants that may be impacted by MACT. Each plant, emission points, controls, and assumptions for emission reductions are described below. Gulf States Steel Gulf States Steel has neither suppression controls nor a capture and control system for the blast furnace casthouse. Assume as a worst case the installation of a capture and control systems that will achieve a 90-percent reduction in fugitive emissions. Ispat-Inland. Ispat-Inland's Number 4 BOPF shop has venturi scrubbers that operate at a pressure drop of about 25 inches of water. Assume these scrubbers must be replaced by higher energy scrubbers that will achieve a 50 percent reduction in emissions. The plant uses scrubbers to control secondary emission from the BOPF, and most plants use baghouses. Assume as a worst case that the scrubbers may be replaced by baghouses for secondary emissions and will result in a 50 percent decrease in baseline emissions. AK Steel. AK Steel (Mddletown, Ohio) may have to upgrade the venturi scrubbers in their BOPF shop. Assume the upgrade will result in a 50 percent reduction in emissions. The plant does not have a capture and control system for secondary emissions from the BOPF. Assume a capture system and baghouse will be installed that will achieve a 90 percent reduction in secondary emissions. Bethlehem Steel. Bethlehem Steel (Burns Harbor) has a scrubber for the control of emissions from hot metal transfer, desulfurization, charging, and tapping in their BOPF shop. Assume the scrubber may be replaced by a baghouse and will result in an emission reduction of 50 percent. Table 7-1 summarizes the baseline emissions and expected reductions based on reductions of 50 percent for upgrading control systems and 90 percent for new capture and control systems for fugitive emissions. Details on the estimates of baseline emissions are given in Chapter 3. 7-1 ------- TABLE 7-1. ESTIMATES OF EMISSION REDUCTIONS Plant Gulf States3 Ispat-Inland No. 4 BOPFb AK Steel (OH) BOPFb Bethlehem (Bums Harbor) BOPFb Source blast furnace casthouse oxygen blow secondary emissions oxygen blow secondary emissions transfer, desulfurization charging tapping Totals Baseline emissions (tpy) PM 360 400 94 230 672 58 127 259 2,200 HAP 2.2 3.4 0.9 2.2 6.4 0.6 1.2 2.5 19 Emission reduction (tpy) PM 324 200 47 115 605 29 64 130 1,514 HAP 2.0 1.7 0.5 1.1 5.8 0.3 0.6 1.3 13 a Estimates of baseline emissions are from Tables 3-6 and 3-8. b Estimates of baseline emissions are from Tables 3-15 and 3-16. 7.2 SECONDARY IMPACTS Secondary impacts include the increased generation of solid waste or wastewater and increased energy usage as a result of upgrading or installing new pollution control equipment. From Table 7-1, the installation of baghouses will result in an increase in dust generation of 1,200 tpy. Upgrading venturi scrubbers will result in a reduction in PM emissions of 320 tpy. Assuming 10 percent solids in the sludge generated, the increase in sludge to be disposed of is 3,200 tpy. The largest increase in energy usage will be from the venturi scrubbers at Ispat-Inland if the pressure drop is increased from 25 to 50 inches of water. The minor scrubber upgrade at AK Steel is associated with improving water supply and not pressure drop because the scrubbers already operate at 50 inches of water. Baghouses for uncontrolled sources will also result in increased energy usage; however, baghouses that replace existing scrubbers will reduce energy usage because scrubbers require more energy. 7-2 ------- The OAQPS control cost manual1 provides the following empirical equation for estimating a fan's energy usage for capture and control systems based primarily on the system's pressure drop and the volumetric flow rate. kw-hr/yr = 0.00018 xacfmx pressure drop (inches of water) x hrs/yr For Ispat-Inland's venturi scrubber with a flow rate of 600,000 acfm, an increased pressure drop of 25 inches of water, and operation 8,760 hrs/yr, the increased energy usage would be: kw-hr/yr = 0.00018x 600,000x 25x 8,760 = 24x 106 kw-hr/yr The increased energy usage for the two new baghouses at Gulf States Steel and AK Steel is more than offset by the replacement of venturi scrubbers with baghouses at Ispat-Inland and Bethlehem Steel. Consequently, there is no net increase in energy usage from the installation of baghouses. 7.3 REFERENCES 1. U.S. Environmental Protection Agency. OAQPS Control Cost Manual. 5th edition. EPA 453/B-96-001. February 1996. 7-: ------- APPENDIX A SUMMARY OF SINTER PLANT TESTING A.1 LTV Steel's Sinter Plant at East Chicago, Indiana LTV Steel's sinter plant at their Indiana Harbor Works was constructed in 1959 and is a part of the integrated iron and steel plant that also includes blast furnaces, BOPF, ladle metallurgy, continuous casting, rolling mills, and galvanizing lines. The sinter plant has a maximum rated capacity of 5,280 tpd and operates 24 hours per day, 7 days a week. Typically, the plant produces 3,800 tpd and operates 24 hours per day for about 310 days per year. Emission testing was performed June 25-27, 1997. Emissions are generated in the process as sinter dust and combustion products and are discharged through the grates and windboxes to a common collector main. Coarse dust particles settle out of the air stream in the collector main and are discharged through flapper valves to a conveyor belt. This conveyor also receives the returns from a series of hoppers that collect any particles that fall under the sinter machine. This material is returned by conveyor to the sinter mix feed for recycle to the process. The exhaust then passes through a battery of cyclones and a series of chambers (originally designed for an ESP that is no longer used). The cyclones and chambers remove dust particles, which are also deposited onto a conveyor (through air actuated valves) for recycle to the process. The exhaust is moved by a 6,000 horsepower fan to the primary control device, which is a double-throat Kinpactor scrubber designed by American Air Filter. The parameters associated with the scrubber that are monitored include the pressure drop across the scrubber, flow rate of water to the scrubber, exhaust fan draft and amperage, and the scrubber water blowdown rate. A. 1.1 Parameter Montoring The operating parameters associated with the process and control device were recorded at 15- minute intervals throughout each test day. The process parameters that were monitored included the feed rate from each of the 10 bins that were used in the sinter mix, the temperatures and the fan draft for the windboxes, percent water in the feed, sinter machine speed, and the sinter production rate. The emission control device parameters that were monitored included the pressure drop across the scrubber, the water flow rate, blowdown rate, fan draft, and fan amps. Tables A-l and A-2 present a summary of the range of values for these parameters for each test period. A-l ------- The process and control device appeared to be stable throughout the three test days; consequently, sampling was conducted under normal and representative conditions. The feed rates of mill scale and other materials were typical of the historical rates in recent years that had been reported by the plant. In addition, the oil content of the mill scale was typical (target is 0.2 percent, maximum) with an average of 0.21 percent oil (a range of 0.17 to 0.24 percent) based on the analysis of 5 samples. An examination of the monitoring data showed that the average pressure drop across the scrubber was 43.1, 42.8, and 42.4 inches of water for the 3 test days. The coke rate seemed to be the most variable parameter during the tests because adjustments were made frequently to change the sintering temperature. The coke rate for the 3 tests averaged 1.7, 1.15, and 0.67 ton per hour; consequently, the emission test results may provide some insight into the effect of coke rate on emissions. The windbox temperatures also varied somewhat during the tests. Using Windbox 20 as an example, the average temperatures during the 3 tests were 538, 567, and 443'F. A.1.2 Analysis of Monitoring and Test Results Table A-3 summarizes the emission results for each run along with selected parameters that were monitored during the test. Only a few comparisons can be made because the process operated stably and consistently during the 3 test runs. One difference is that the coke (fuel) rate during Run 3 was only 39 percent of the rate during Run 1 and only 58 percent of the rate during Run 2. The lower fuel rate during Run 3 is reflected in the lower windbox temperature during Run 3, which was about 100* F lower than in the previous 2 runs. The pollutants most likely to be affected by the change in combustion conditions are D/F and PAH.. During Run 3, the emission rates for all of these compounds were lower than in the previous 2 runs. The highest emissions of PM and Pb occurred during Run 3. The cause is not conclusive, but some of the possible factors affecting this, perhaps in combination, were that Run 3 had the highest sinter feed and production rate and the lowest average pressure drop across the scrubber. In addition, Table A- 1 indicates that Run 3 had a higher feed rate of fines (pellet fines and BOPF slag fines) than that recorded during the previous 2 runs. Service water was used in the scrubber during Run 1 and recycled blast furnace water was used during Runs 2 and 3. There is no obvious difference in emissions that can be clearly attributed to the type of scrubber water. A-2 ------- The major metal HAP that was found was Pb, which accounted for over 97 percent of the total metal HAP emissions. Discussions with the plant and examination of data from the analysis of blast furnace fines and sludge indicated that a likely source of the Pb emissions was from this fine material recycled from the blast furnace. Data in the literature showed that the Pb content of blast furnace dust and sludge was generally in the range of 0.01 to 0.1 percent. At a typical feed rate for the dust and sludge of 28,000 Ib/hr (14 tph), these materials would introduce 2.8 to 28 Ib/hr of Pb into the process, which could easily account for the Pb that was found entering the scrubber (4.2 Ib/hr). In addition, the small particle size of these pollution control residues from the blast furnace may increase the probability that they become airborne, and the volatility of Pb and some Pb compounds from combustion processes may tend to increase the concentration of Pb in the windbox emissions. Table A-4 through A-6 presents a summary of the annual emissions and the emission factors derived from this test. A-3 ------- TABLE A-l. PROCESS PARAMETER VALUES DURING THE TESTS Parameter Run 1 (6/25/97) Run 2 (6/26/97) Run 3 (6/27/97) Feed rate (tph): Mil scale BOPF slag/filter cake Fines Pellet chips Pellet fines- blend Limestone Cold fines Coke breeze Flue dust BOPF slag fines Other parameters: Percent water Grate speed Windbox 20 temperature (• F) Windbox draft (in. water) Feed rate (tph) Sinter production (tph) 25.2 (24.8 - 25.5) 16.7(16.1 -17.9) 16.7(16.1 -17.6) 77.4 (75.9 - 78.8) 9.5 (8.5 - 10.2) 27.2 (26.9 - 27.7) 19.6(17.6-21.4) 1.7(1.5-1.9) 5.9 (5.8 - 6.0) 7.9 (7.6 - 8.2) 6.7-7.5 70-76 453 - 656 13.6-17.4 205-210 155 - 158 25.2 (24.9 - 25.5) 16.9(15.9-18.2) 16.4(15.9-18.0) 77.7 (76.2 - 79.0) 10.7(10.1 -11.4) 27.5 (26.8 - 27.8) 17.2(15.2-19.5) 1.2(0.9-1.5) 5.9 (5.8 - 6.0) 9.3(9.4-10.1) 6.5-7.4 70-76 474 - 659 13.3-18.2 201 -212 153 - 161 25.2 (24.8 - 25.6) 16.9(15.5-17.9) 16.7 (15.3 - 18.0) 77.6 (76.5 - 79.5) 12.3(11.3-13.6) 27.7 (27.4 - 28.8) 17.8(16.8-23.2) 0.7(0.34-1.1) 5.9 (5.8 - 6.0) 10.0(9.8-10.1) 7.2 - 8.2 70-82 334 - 571 14.2- 18.2 209 - 213 159-161 TABLE A-2. CONTROL DEVICE OPERATING PARAMETERS DURING THE TESTS Parameter Pressure drop (in. water) Water flow (gal/min) Slowdown (gal/min) Fan amps Fan draft (in. water) Type of water Run 1 (6/25/97) 38.4-46.6 3,040 - 3,085 236 - 239 663 - 695 3.1 -5.8 service (lake) Run 2 (6/26/97) 39.4 - 46.3 3,080 - 3,130 242 - 246 685 - 700 3.2-5.8 Run 3 (6/27/97) 39.8-47.0 3,080-3,110 241 - 244 700 - 730 3.8-5.1 recycled blast furnace A-4 ------- TABLE A-3. VENTURI SCRUBBER: RESULTS FOR EACH TEST RUN Parameter PMa - inlet PM - outlet PM efficiency PM - inlet PM - outlet HAP metals - inb HAP metals - outb Metals efficiency D/F congeners0 D/FTEQd 7PAHe 16 PAH TOTAL PAH Sinter production Scrubber* p Windbox 20 temperature Units Ib/hr Ib/hr percent gr/dscf gr/dscf Ib/hr Ib/hr percent •g/hr •g/hr g/hr g/hr g/hr tons/hr in. water •F Runl 419 34 92 0.20 0.014 4.5 3.8 16 810 93 1.9 69 83 156 43.1 538 Run 2 479 38 92 0.23 0.017 4.5 3.7 18 768 91 2.0 78 92 159 42.8 567 Run3 550 43 92 0.25 0.019 4.9 3.9 20 694 79 1.4 61 73 160 42.4 443 Average 483 38 92 0.23 0.017 4.6 3.8 17 757 88 1.7 69 83 158 42.8 516 1 PM = participate matter b Mostly lead c D/F congeners are those dioxins and furans that have a toxicity equivalent factor relative to 2,3,7,8-TCDD. d D/F TEQ is the toxicity equivalent expressed relative to 2,3,7,8-TCDD. e PAH = polycyclic aromatic hydrocarbons. A-5 ------- TABLE A-4. VENTURI SCRUBBER: SUMMARY OF RESULTS FOR PM AND HAP METALS Pollutant Particulate matter Concentration (gr/dscf) Inlet 0.23 Outlet 0.017 Emission rate (Ib/hr) Inlet 483 Outlet 38 Efficiency (%) 92 Annual rate (tpy)a Inlet 1,800 Outlet 142 Emission factor (Ib/t sinter) Inlet 3.1 Outlet 0.24 Pollutant: HAP metals Mercury Arsenic Beryllium Cadmium Cobalt Chromium Manganese Nickel Lead Antimony Selenium Total HAP metals Concentration (• g/dscm) Inlet 0.96 4.3 0.054 20 0.30 24 400 23 4,500 2.6 13 5,000 Outlet 1.5 1.1 0.052 17 0.050 5.2 17 22 3,700 1.6 8.7 3,800 Emission rate (g/hr) Inlet 0.41 1.8 0.023 8.4 0.18 9.9 171 9.8 1,900 1.1 5.5 2,100 Outlet 0.69 0.50 0.023 7.8 0.023 2.4 7.9 9.9 1,690 0.75 4.0 1,700 Efficiency (%) 0 73 0 7.4 87 76 95 0 11 32 28 18 Annual rate (tpy) Inlet 3.3 x lO'3 l.SxlO-2 1.9xlO-4 6.9 xlO'2 l.SxlO-3 S.lxlO'2 1.4 8.0 xlO'2 16 9.0 xlO'3 4.5 x lO'2 17 Outlet 5.7 x lO'3 4.1xlQ-3 1.9xlO-4 6.4 xlO'2 1.9xlO-4 1.9xlO-2 6.4 xlO'2 8.1xKr2 1.4xlO+1 6.1xlO-3 3.2 xlO'2 1.4xlO+1 Emission factor (Ib/ton sinter) Inlet 5.7 x lO'6 2.5 x 10'5 3.2 xlO'7 1.2xlQ-4 2.5 xlO'6 1.4xlO-4 2.4 xlO'3 1.4xlQ-4 2.7 xlO'2 l.SxlO'5 7.7 x lO'5 2.9 xlO'2 Outlet 9.7 x lO'6 7.0 xlO'6 3.3 x lO'7 l.lxlO'4 3.3 x ID'7 3.3 x ID'5 l.lxlO'4 1.4xlQ-4 2.4 xlO'2 l.OxlO-5 5.5 x lO'5 2.4 xlO'2 Based on operation for 24 hours per day for 310 days per year. A-6 ------- TABLE A-5. VENTURI SCRUBBER: RESULTS FOR PAH AND D/F Pollutant: PAH3 Benzo(a)anthracene Benzo(a)pyrene Benzo(b)fluoranthene Benzo(k)fluoranthene Chrysene Dibenzo(a,h)anthracene Indeno(l ,2,3-cd)pyrene Total 7 PAH Acenaphthene Acenaphthylene Anthracene Benzo(g.h,l)perylene Fluoranthene Fluorene Naphthalene Phenanthrene Pyrene Total 16 PAH 2-Methylnaphthalene 2-Chloronaphthalene Benzo(e)pyrene Perylene Total -all PAH D/F D/FTEQC D/F Congeners'1 Concentration (• g/dscm) 0.53 0.23 1.2 0.22 1.3 0.097 0.26 3.9 3.5 7.6 1.8 0.36 6.9 5.4 78 43 3.0 153 29 0.039 0.76 0.058 183 Concentration (ng/dscm) 0.19 1.7 Emission rate (g/hr) 0.24 0.11 0.54 0.10 0.60 0.044 0.12 1.7 1.6 3.4 0.81 0.16 3.1 2.4 35 19 1.4 69 13 0.018 0.30 0.026 83 Emission rate (• g/hr) 88 757 Emissions'" (tpy) 0.0019 0.00086 0.0044 0.00082 0.0049 0.00036 0.00096 0.014 0.013 0.028 0.0067 0.0013 0.026 0.020 0.29 0.16 0.011 0.57 0.11 0.00015 0.0028 0.00022 0.68 g/yr 0.66 5.6 Ib/ton sinter 3.3 x 10- 6 l.SxlO'6 7.5 xlO'6 1.4xlO-6 8.4 xlO'6 6.1xlO'7 1.6xlO-6 2.4 x 10 5 2.2 xlO'5 4.8 xlO'5 l.lxlO'5 2.2 xlO'6 4.3 x ID'5 3.4 xlO'5 4.9 xlO'4 2.7 x 10'4 1.9xlO-5 9.7 x 10 4 l.SxlO'4 2.5 xlO'7 4.8 xlO'6 3.7 x 10'7 1.2 x 10 3 g/ton 5.5 x 10'7 5.0 xlO'6 a PAH = poly cyclic aromatic hydrocarbons. b Based on operation for 24 hours per day for 310 days per year. 0 D/F TEQ is the toxicity equivalent expressed relative to 2,3,7,8-TCDD. A-7 ------- 1D/F congeners are those dioxins and furans that have a toxicity equivalent factor relative to 2,3,7,8-TCDD. A-8 ------- A.2 Youngstown Sinter Company's Sinter Plant The Youngstown sinter plant is operated by Youngstown Sinter Company, a wholly owned subsidiary of WCI Steel. The plant was purchased from LTV Steel Company and was brought on line in June 1991. The sinter plant is located a few miles from the WCI Steel integrated iron and steel plant in Warren, OH. The integrated plant includes one blast furnace, a BOPF shop containing two BOPF vessels, ladle metallurgy, continuous casting, rolling mills, and galvanizing lines. The sinter plant has a capacity of 60,000 tons per month (tpm) and operates 24 hours per day with 2 days scheduled downtime every seven days for routine maintenance. Testing was performed August 12-15, 1997. Emissions are generated in the process as sinter dust and combustion products are discharged through the grates and the 21 windboxes to a common collector main and are then collected by the strand baghouse. The pulse jet baghouse is manufactured by Environmental Elements and uses Nomex® bags that are coated with an acid-resistant finish. There are 14 modules, each containing 306 bags. The bags are 6 inches in diameter and 15 feet in length, and the total cloth area for each module is 7,215 square feet. The gross air-to-cloth ratio is 3.96 acfm/ft2 and the net air-to-cloth ratio, with one module off-line for cleaning is 4.26 acfm/ft2. The flow to the baghouse is approximately 400,000 cfm. A preheat burner is used to minimize condensation and to bring the gas up to the desired inlet temperature. The dust is removed from the baghouse by rotary screw to bins where it is stored on the ground to gather moisture and is blended back into the sinter feed. The parameters associated with the baghouse that are monitored include the pressure drop across the baghouse, inlet temperature, stack temperature, damper percent, and fan amps. Three additional baghouses are used to control emissions from the sinter plant. The C baghouse, a pulse jet baghouse utilizing polyester bags, is used to control emissions from the material handling bins and the conveyors that transfer the sinter mix to the sinter machine. The cooler baghouse controls emissions from the sinter cooler and from the main truck loadout station. The baghouse is a shaker baghouse that utilizes Nomex® bags and contains nine compartments. Eight of the compartments are used for the cooler and one compartment is used for the truck loadout station. There are four 200 horsepower fans on the sinter cooler. The first fan is the dirtiest fan and is directed back to hoods on the sinter A-9 ------- machine and sent back through as preheat air. The other three fans are ducted to the baghouse. In addition, the truck loadout station has a 70,000 cfm fan. These baghouses were not evaluated as part of this test program. The A baghouse that serves the discharge end of the sinter plant was evaluated as part of this test program. This baghouse controls emissions from discharge end emission points, including the hood before the sinter machine; the hood over sinter discharge; the sinter breaker and hot screen which is enclosed by a cloth curtain; the tail end of the sinter cooler; emissions from each of the ten sinter feed bins; a variety of transfer points for the transport of sinter, dust, and fines; and emissions from sinter bins located in the sinter overflow storage area. At any point where there is hot sinter, emissions are first ducted to a cyclone before going to the baghouse. All of the baghouses are monitored on a weekly basis by an outside contractor to check the operation and for any visible opacity. A whole compartment is dye- tested if there is more than 5 percent visible emissions observed, and the broken bags are then replaced. Every other month, a complete compartment of either the strand or cooler baghouse is replaced; each compartment is replaced approximately every 3 years. A.2.1 Monitoring Results During the Tests The operating parameters associated with the process and control device were recorded at 15- minute intervals throughout each test day. The process parameters that were monitored included the temperatures and the fan draft for the windboxes, percent water in the feed, sinter machine speed, and the temperature of each of the four cooling fans. In addition, the turn supervisor's report provided additional information, including tons per hour of pre-blend, and tons per 8-hour turn of limestone, dolomite, coke fines, and cold fines. The emission control device parameters that were monitored included the pressure drop across the baghouse, damper percent, inlet temperature, stack temperature, fan amps, and the pressure drop of each of the 14 compartments of the baghouse. Tables A-6 and A-7 present a summary of the range of values for these parameters for each test period. Table A-8 presents a summary of the pressure drops of each compartment of the baghouse for the four days of testing. A-10 ------- The process and control device appeared to be stable throughout the four test days; consequently, sampling was conducted under normal and representative conditions. An examination of the monitoring data showed that the average pressure drop across the baghouse was 10.8, 12.0, 12.9 and 13.5 inches of water for the 4 test days. The pressure drop across the baghouse did increase slightly during each day of testing. On the third day, the compartments were double cleaned to try to reduce the pressure drop. The temperatures and draft of the windboxes varied somewhat during the tests; plant operators stated that the temperature of windboxes 19 and 20, should generally be 475-500 • F to achieve proper burnthrough of the sinter bed. During each run of testing performed on A baghouse, the pressure drops of each compartment and the pressure drop across the baghouse were monitored periodically, generally every 20 to 30 minutes. The plant does not monitor any other parameters on A baghouse; since the A baghouse is responsible for the capture and control of dust sources throughout the sintering process, malfunctions are readily apparent. Table A-9 presents a summary of the pressure drops of each compartment and the pressure drop across the baghouse during each test period. A.2.2 Analysis of Monitoring and Test Results Table A-10 summarizes the emission results for each run for key pollutants from the outlet of the control device on the sinter strand, along with selected parameters that were monitored during the test. Only a few comparisons can be made because the process operated stably and consistently during the 3 test runs. One difference is that the pressure drop across the strand baghouse increased over the four days of testing, from an average of 10.78 on the first day of testing, to an average of 13.48 on the final day of testing. However, the results were fairly stable and did not appear to be impacted by the increased pressure drop over the course of testing. Table A-l 1 presents emission results for each run for key pollutants from the baghouse that controls emissions from the discharge end ("A" baghouse). Paniculate matter and HAP metal emissions were fairly steady over three runs. One interesting factor is that while paniculate matter emissions during Run 2 were three times lower than during Run 1, and two times lower than during Run 3, HAP metal emissions were steady over the course of the three A-ll ------- runs. The major metal HAPs that were found were Pb and Mn; both were effectively captured and controlled by both the Strand baghouse and A baghouse. A-12 ------- TABLE A-6. PROCESS PARAMETER RANGES DURING THE TESTS Parameter Testl (8/12/97) Test 2 (8/13/97) Tests (8/14/97) Test 4 (8/15/97) Feed rate: Pre-blend (ore) (tons/hour) Limestone (tons/turn) Dolomite (tons/turn) Coke fines (tons/turn) Cold fines (tons/turn) 120 144 43 19 1738 120 114 39 17 1545 120 167 43 18 1787 120 — — — — Other parameters: Percent water Grate speed (feet/min) Windbox 1 temperature (• F) Windbox 1 draft (in. H2O) Windbox 3 temperature (• F) Windbox 3 draft (in. H2O) Windbox 13 temperature (*F) Windbox 13 draft (in. H2O) Windbox 18 temperature (*F) Windbox 18 draft (in. H2O) Windbox 19 temperature (*F) Windbox 19 draft (in. H2O) Windbox 20 temperature (• F) Windbox 20 draft (in. H2O) Windbox 21 temperature (*F) Windbox 21 draft (in. H2O) 7.0-7.2 — 177-211 18.0-22.1 167-195 16.2-20.3 187-266 — 327-463 14.7-18.3 396-542 16.4-21.1 373-580 14.5-18.9 — 14.9-17.7 6.7-7.6 — 150-202 20.3-23.5 108-186 18.6-21.5 184-233 — 251-459 16.6-19.9 357-513 18.4-21.9 391-546 17.0-20.7 360-465 15.7-19.3 6.8-7.0 — 157-207 19.5-22.3 149-181 18.1-20.5 169-231 — 288-457 15.7-18.5 350-460 18.0-20.4 372-496 16.2-18.9 332-429 15.1-17.5 6.7-6.8 6.3 -7.0 166-220 19.5-21.8 159-198 18.0-20.1 165-342 — 301-521 16.0-17.8 363-545 17.2-20.5 385-545 16.5-18.6 355-443 15.3-17.2 A-13 ------- Cooling Fan Temperatures (• F) A B C D 420-463 505-546 430-460 185-243 411-460 405-544 205-458 116-237 395-415 456-530 372-440 157-200 376-413 456-507 385-435 172-192 TABLE A-7. CONTROL DEVICE PARAMETERS DURING THE TESTS - WINDBOX BAGHOUSE Parameter Pressure drop (in. H2O) Inlet Temp. (• F) Stack Temp. (• F) Fan amps Damper (%) Testl (08/12/97) 9.30-11.87 242 - 265 243 - 248 684-735 88.9-90.1 Test 2 (08/13/97) 10.60-12.59 217-253 231-248 667-690 89.5-91.2 Tests (08/14/97) 11.61-13.57 211-245 216-243 667-694 88.8-90.9 Test 4 (08/15/97) 12.09-14.12 217-236 227-248 659-690 89.0-90.8 TABLE A-8. PRESSURE DROP ACROSS EACH COMPARTMENT OF THE WINDBOX BAGHOUSE Compartment Pressure Drop 1 2 3 4 5 6 7 8 Testl (08/12/97) 7.0-8.6 8.2-9.2 7.1-8.6 5.6-8.0 7.1-8.5 6.6-7.9 6.4-8.0 6.7-8.4 Test 2 (08/13/97) 6.8-9.3 6.7-9.6 8.6-9.8 6.8-8.8 8.0-9.8 7.8-9.3 7.1-9.4 6.0-8.8 Tests (08/14/97) 7.0-9.6 6.9-9.8 9.4-10+ 7.4-9.8 9.1-10+ 8.3-9.9 8.9-10.0 7.7-9.7 Test 4 (08/15/97) 8.6-9.9 8.0-10.0 9.9-10+ 7.9-10+ 10.0-10+ 8.9-10+ 9.7-10+ 7.2-10+ A-14 ------- Compartment Pressure Drop 9 10 11 12 13 14 Total Testl (08/12/97) 7.6-9.4 7.1-9.0 6.8-8.9 7.6-9.4 6.4-9.0 6.4-9.2 9.9-11.5 Test 2 (08/13/97) 8.6-9.9 7.8-9.7 7.3-9.4 8.8-10+ 7.6-10+ 7.6-10+ 10.0-11.5 Tests (08/14/97) 9.4-10+ 9.3-10+ 8.5-10+ 9.6-10+ 9.8-10+ 9.4-10+ 11.4-12.3 Test 4 (08/15/97) 9.5-10+ 9.9-10+ 8.2-10+ 10+ 10.0-10+ 8.5-10+ 12.0-13.0 TABLE A-9. PRESSURE DROP ACROSS EACH COMPARTMENT OF DISCHARGE END BAGHOUSE ("A") Compartment 1 2 O 4 Total Test 1 (08/15/97) 2.6-3.8 2.8-3.7 4.7-5.5 4.4-6.0 7.7-8.1 Test 2 & 3 (08/16/97) 3.0-4.7 3.7-5.5 1.5-2.0 5.5-7.4 7.9-10.9 A-15 ------- TABLE A-10. WINDBOX BAGHOUSE: RESULTS FOR EACH TEST RUN Parameter PM — inlet PM — outlet PM efficiency PM — inlet PM — outlet HAP metals - inleta HAP metals - outlet Metals efficiency Parameter Dioxin/furan congenersb Dioxin/furan TEQC 7 PAH 16 PAH Total PAH Sinter production Baghouse • P Windbox 20 Temp. Units Ib/hr Ib/hr % gr/dscf gr/dscf Ib/hr Ib/hr % Units • g/hr • g/hr g/hr g/hr g/hr tons/hr in.H2O •F Run 1 1,960 2.35 99.9 0.68 0.001 11.7 0.063 99.5 Runs 1 & 2 unacceptable leak checks 110 10.8 474 Run 2 1,120 0.71 99.9 0.41 0.0003 11.9 0.120 99.0 Run 3 2,142 342 28.9 510 691 110 12.0 467 Run 3 1,490 1.30 99.9 0.52 0.00055 11.8 0.068 99.4 Run 4 2,444 404 34.8 457 634 110 12.9 446 Runs 4 & 5 - -- -- -- -- - - Run 5 2,186 375 33.9 575 755 110 13.5 457 Average 1,520 1.45 99.9 0.54 0.0006 11.8 0.084 99.3 Average 2,257 374 32.5 514 693 110 12.3 461 a Mostly Pb. b D/F congeners are those dioxins and furans that have a toxicity equivalent factor relative to 2,3,7,8-TCDD 0 D/F TEQ is the toxicity equivalent expressed relative to 2,3,7,8-TCDD A-16 ------- TABLE A-ll. DISCHARGE END BAGHOUSE ("A") - RESULTS FOR EACH RUN Parameter PM — outlet Mn — outlet HAP metals — outlet Units Ib/hour Ib/hour Ib/hour Runl 0.53 0.0033 0.012 Run 2 0.67 0.036 0.046 Run 3 0.26 0.016 0.028 Average 0.48 0.019 0.029 A surprising result is the emission rate of PAH that was measured during the testing. Emissions for PAH were slightly higher than PM emissions from the outlet of the strand baghouse. These results were consistent over all test runs; even though the first two test runs resulted in questionable data, the results still are consistent with the remaining three test runs. It is not known if the higher emissions were present in the inlet stream or if the baghouse performed poorly in the capture and control of PAH emissions, since inlet testing for PAH was not performed. The major PAH present in the outlet stream were naphthalene and 2-methylnaphthalene, with 3,660 and 2,920 Ibs/yr being emitted respectively. Table A-12 presents a summary of PM and metal HAP results for the strand baghouse, including concentrations, efficiencies, annual emission rates, and emissions factors for each metal HAP. Table A-13 presents similar results for PAH and D/F. Table A-14 presents a summary of results for the discharge end baghouse for PM and metal HAP . A-17 ------- TABLE A-12. WINDBOX BAGHOUSE: SUMMARY OF RESULTS FOR PM AND HAP METALS Pollutant Particulate matter Concentration (gr/dscf) Inlet 0.54 Outlet 0.0006 Emission rate (Ib/hr) Inlet 1,520 Outlet 1.45 Efficiency (%) 99.9 Annual rate (tpy)a Inlet 5,700 Outlet 5.4 Emission factor (Ib/t sinter) Inlet 13.8 Outlet 0.013 Pollutant: HAP metals Mercury Arsenic Beryllium Cadmium Cobalt Chromium Manganese Nickel Lead Antimony Selenium Total HAP metals Concentration (• g/dscm) Inlet 6.23 8.27 0.075 32.2 9.35 90.2 2230 18.3 7153 2.48 23.1 9,573 Outlet 5.02 0.452 0.038 0.180 0.135 4.47 29.1 2.07 21.3 1.21 18.0 82 Emission rate (g/hr) Inlet 3.5 4.6 0.04 18.0 5.2 50.5 1,247 10.2 4,001 1.4 12.9 5,354 Outlet 2.35 0.21 0.02 0.08 0.06 2.09 13.62 0.97 9.97 0.57 8.42 38 Efficiency (%) 32.5 95.4 57.7 99.5 98.8 95.9 98.9 90.5 99.8 59.3 34.7 99.3 Annual rate (tpy) Inlet 0.03 0.04 0.00 0.15 0.04 0.41 10.16 0.08 32.61 0.01 0.11 44 Outlet 0.02 0.00 0.00 0.00 0.00 0.02 0.11 0.01 0.08 0.00 0.07 0.31 Emission factor (Ib/t sinter) Inlet 7.0 xlO'5 9.3 xlO'5 8.4 xlO'7 3.6 xlO'4 1.0 xlO'4 1.0 xlO'3 2.5 xlO'2 2.0 xlO'4 8.0 xlO'2 2.8 xlO'5 2.6 xlO'4 l.lxlO-1 Outlet 4.7 x 10'5 4.2 xlO'6 3.6 xlO'7 1.7xlO-6 l.SxlO'6 4.2 xlO'5 2.7 xlO'4 1.9xlO-5 2.0 xlO'4 l.lxlO-5 1.7xlO'4 7.7 xlO'4 ' Based on operation for 24 hours per day, 6 days per week, 52 weeks per year (7400 hours/year). A-18 ------- TABLE A-13. WINDBOX BAGHOUSE: RESULTS FOR PAH AND D/F Pollutant: PAH Benzo(a)anthracene Benzo(a)pyrene Benzo(b)fluoranthene Benzo(k)fluoranthene Chrysene Dibenzo(a,h)anthracene Indeno(l ,2,3-cd)pyrene Total 7 PAH Acenaphthene Acenaphthylene Anthracene Benzo(g.h,l)perylene Fluoranthene Fluorene Naphthalene Phenanthrene Pyrene Total 16 PAH 2-Methylnaphthalene 2-Chloronaphthalene Benzo(e)pyrene Perylene Total -all PAH D/F D/FTEQb D/F Congeners0 Concentration (• g/dscm) 21.2 2.07 8.81 2.79 34.6 0.590 0.433 70.7 19.0 34.5 44.2 0.419 122 40.3 478 250 54.8 1114 382 1.74 4.27 0.557 1503 Concentration (ng/dscm) 0.81 4.9 Emission rate (g/hr) 9.79 0.956 4.07 1.29 16.0 O.273 O.200 32.6 8.80 16.0 20.4 O.194 56.3 18.8 221 115 25.3 514 176 0.804 1.98 O.257 693 Emission rate (• g/hr) 374 2,257 Emissions3 tpy 0.0799 0.0078 0.0332 0.0105 0.1305 0.0022 0.0016 0.266 0.072 0.1305 0.1664 0.0016 0.459 0.1534 1.80 0.938 0.206 4.19 1.44 0.0066 0.0162 0.0021 5.65 g/yr 2.8 16.7 Ib/ton sinter 1.96xlO-4 1.92xlO'5 8.16xlO-5 2.58xlO-5 3.21xlO'4 5.47X10'6 4.01xlO-6 6.53xl04 1.76xlO-4 3.21xlO-4 4.09xlO-4 3.89X10'6 l.lSxlO'3 3.77xlO-4 4.43xlO'3 2.30xlO-3 5.07xlO-4 1.03xl02 3.53xlO-3 1.61xlO-5 3.97xlO'5 S.lSxlO'6 1.39xl02 g/ton 3.4 xlO'6 2.1xlO'5 1 PAH = polycyclic aromatic hydrocarbons. b Based on operation for 24 hours per day for 310 days per year. c D/F TEQ is the toxicity equivalent expressed relative to 2,3,7,8-TCDD. d D/F congeners are those dioxins and furans that have a toxicity equivalent factor relative to 2,3,7,8-TCDD. A-19 ------- TABLE A-14. DISCHARGE END BAGHOUSE ("A") - RESULTS FOR PM AND METAL HAP Pollutant — PM PM Pollutant — Metal HAP Arsenic Beryllium Cadmium Cobalt Chromium Mercury Manganese Nickel Lead Antimony Selenium HAP metals Outlet Ib/hr 0.48 gr/dscf 0.0007 Outlet g/hr 0.10 0.013 0.017 0.039 1.2 0.29 8.4 1.0 1.1 0.48 0.43 13.1 • g/dscm 0.755 0.098 0.126 0.292 8.92 2.13 62.3 7.59 7.88 3.57 3.21 96.9 Emissions a tpy 1.8 Emissions a tpy 0.0008 0.0001 0.0001 0.0003 0.0099 0.0024 0.070 0.0084 0.0086 0.0040 0.0036 0.11 Emission Factor Ib/ton sinter 0.0044 Emission Factor Ib/ton sinter 2.4 xlO-6 2.6 xlO-7 3.4 xlO-7 7.8 x 10-7 2.4 xlO-5 5.8 x 10-6 1.7 xlO-4 2.0 xlO-5 2.2 xlO-5 9.6 xlO-6 8.6 xlO-6 2.6 xlO-4 3 Based on operation for 24 hours per day, 6 days per week, 52 weeks per year (7400 hours/year) A-20 ------- APPENDIX B DOCUMENTATION FOR THE MACT FLOOR This appendix documents the data analyses used to develop the MACT floor for integrated iron and steel manufacturing facilities. The data are presented and referenced, and all references are available in EPA Docket Number A-2000-44. The proposal preamble provides details on the rationale for selection of the floor and MACT, and all data summarized in the preamble are presented in detail in this appendix. Additional details on existing State limits are given in Chapter 5. B.1 SINTER PLANT WINDBOXES The sintering process converts fine-sized raw materials, including iron ore, coke breeze, limestone, mill scale, and flue dust, into an agglomerated product (sinter) of suitable size for charging into the blast furnace. There are nine sinter plants in the U.S.; however, only seven are currently operating. The windbox exhaust is controlled by a baghouse at four plants and by a venturi scrubber at five plants (see Table B-l). B.1.1 PM Emission Control Performance Useful test data on actual PM emissions are given in Table B-2 and are available for six of the nine plants, two equipped with baghouses and four equipped with venturi scrubbers. In each case, the data reflect the results of performance tests comprised of the average of three test runs, expressed in terms of total PM. Details for each run are given in Table B-3 and are plotted in Figure 1. An initial characterization of achievable performance based on concentration (gr/dscf) suggested that baghouses perform substantially better than do scrubbers. Concentration values recorded for the two baghouses are two to nearly four times lower than those recorded for the four scrubbers. Upon closer scrutiny, the results show that much of the difference in perceived performance is due to the fact that baghouses require the addition of relatively large quantities of ambient air to cool the hot windbox exhaust gases prior to control, whereas scrubbers do not. To correct for this difference, the test results were transformed into a pounds of emissions per ton of sinter (Ib/ton) format. The test results expressed in terms of the hourly mass rate were converted to annual emissions assuming 8,760 hours per operating year. The resultant annual emissions were then divided by a best estimate of annual sinter production for each plant (average for the 5-year period from 1995 through 1999). The results range from 0.26 to 0.33 Ib PM/ton of sinter. B-l ------- TABLE B-l. SINTER PLANTS IN THE U.S. Plant Inland USS Geneva* WCI Steel LTV Bethlehem Bethlehem Wheeling-Pittsburgh* AK Steel State IN IN UT OH IN IN MD WV OH these plants were not operating in 19 Control Baghouse Baghouse Baghouse Baghouse Scrubber Scrubber Scrubber Scrubber Scrubber PM emission limit 0.007 gr/dscf 0.01 gr/dscf 0.0122 gr/dscf; 27 Ib/hr 501b/hr 0.02 gr/dscf 0.277 Ib/ton 0.03 gr/dscf 0.03 gr/dscf 501b/hr 99 - 2000. TABLE B-2. TEST RESULTS FOR SINTER PLANT WINDBOX EXHAUST Sinter plant WCI Steel, Youngstown, OH Ispat-Inland, East Chicago, IN Bethlehem Steel, Sparrows Point, MD LTV Steel, East Chicago, IN AK Steel, Middletown, OH Bethlehem Steel, Burns Harbor, IN Control device Baghouse Baghouse Venturi scrubber Venturi scrubber Venturi scrubber Venturi scrubber Average of the top five PM emissions (Ib/ton of sinter) 0.26 0.26 0.30 0.31 0.32 0.33 0.29 Concentration (gr/dscf) 0.009 0.007 0.026 0.017 0.017 0.025 ~ Flow rate (dscfm) 270,000 470,000 530,000 270,000 220,000 460,000 - B-2 ------- FIGURE B-l. SINTER PLANT - LB/TON FOR INDIVIDUAL RUNS 0.45 0.40 0.35 0.30 = 0.25 ,2 -0.20 0.15 0.10 0.05 0.00 Baahouses Scrubbers The mean and median of the top 5 = 0.3 Ib/ton WCI Steel Inland Steel Bethlehem LTV Steel AK Steel Bethlehem Sparrows Pt Burns Harbor Plant ------- TABLE B-3. SUMMARY OF SINTER PLANT TEST DATA WCI Steel3 (05/27/92)1 Method 5 lun lunl Iun2 Iun3 Average Flow, dscfm 261,036 264,472 274,274 266, 594 gr/dscf 0.0072 0.0107 0.0084 0.0088 Ib/hr 16.07 24.22 19.72 20.00 Ib/ton 0.20 0.31 0.25 0.26 [nland Steelb (05/1 7/95)2 Method 5 lunl Iun2 Iun3 Average 438,188 484,168 477,703 466,686 0.0088 0.0038 0.0083 0.0070 33.19 15.58 33.87 27.55 0.31 0.15 0.32 0.26 Bethlehem Steel, Sparrows Pointc (07/23/9 1)3 Method MD1005 Morth stack Average South stack Average ioth stacks 1 ioth stacks 2 3oth stacks 3 Overall 246,980 255,087 251,521 251,196 278,081 280,540 279,311 -- -- -- — 0.0243 0.0285 0.0257 0.0262 0.0272 0.0256 0.0264 -- -- -- 0.0263 51.4 62.3 55.5 56.4 64.8 61.5 63.2 116.0 123.8 118.7 119.5 — -- -- -- — -- -- 0.29 0.31 0.30 0.30 LTV Steel -EPA test1 (06/25/97)4 Method 29 lunl Iun2 Iun3 Average 271,569 268,850 260,870 267,085 0.0140 0.0170 0.0190 0.0170 34.00 38.00 43.00 38.00 0.27 0.30 0.34 0.31 4K Steel6 (11/22/93)5 Method 5 lunl Iun2 Iun3 4verage 218,090 225,994 220,965 221,683 0.0146 0.0139 0.0219 0.0168 27.38 26.88 41.42 31.89 0.27 0.27 0.42 0.32 B-4 ------- lun Flow, dscfm gr/dscf Ib/hr Ib/ton Bethlehem Steel, Burns Harbo^ (03/1 1/92)6 Method 5 lunl Iun2 Iun3 4verage 458,088 457,195 477,942 464,408 0.0208 0.0273 0.0254 0.0245 81.85 106.90 104.21 97.65 0.28 0.36 0.36 0.33 1 Based on the average annual production of 686,828 tpy. d Based on the average annual production of 1,103,202 tpy b Based on the average annual production of 935,743 tpy. 'Based on the average annual production of 874,112 tpy. 'Based on the average annual production of 3,460,737 tpy. f Based on the average annual production of 2,568,117 tpy. B.I.2 Organic HAP The windbox exhaust gas can contain appreciable quantities of organic HAP, including both volatile and semivolatile compounds. There is strong evidence that demonstrates that the quantity of organic HAP emitted is directly related to the quantity and oil content of the mill scale component of the sinter feed. United States sinter plants limit organic emissions by carefully monitoring and limiting the oil content of the sinter feed. This pollution prevention control measure is an effective method for preventing, and thus reducing, emissions of organic HAP. Two plants in Indiana have performed testing to relate oil content with emissions of VOC. The test results show a strong correlation between oil content and potential VOC emissions. One of the organic pollutants of concern that has been related to oil content is a family of compounds called poly chlorinated dibenzodioxins and furans (D/F). A 1994 paper7 identified sinter plants in Germany as one of the most important industrial sources of D/F. Tests showed an average concentration in the windbox exhaust of 47 ng toxicity equivalent (TEQ)/m3 and annual emissions of 122 g TEQ. The D/F emissions were attributed to high levels of oils and chlorinated organics in the waste materials recycled to the sinter plant. EPA conducted emission tests at two representative facilities to characterize D/F emissions from U.S. sinter plants, one that uses a venturi scrubber as the windbox control device and one that uses a baghouse. The test results are presented in Appendix A. The tests were performed in 1997 on the venturi scrubber at LTV Steel in East Chicago, IN4 and on the baghouse at WCI Steel in Youngstown, OH.8 These plants routinely monitor the oil content of their sinter feed, which averages B-5 ------- 0.014 percent oil at the East Chicago, IN facility and 0.025 percent oil at the Youngstown, OH facility. The average D/F concentration from three 4-hour runs at each plant ranged from 0.2 ng TEQ/m3 at the East Chicago, IN facility to 0.8 ng TEQ/m3 at the Youngstown, OH facility, both far below the levels reported for the German sinter plant. Assuming typical operation of each plant (310 days/yr), annual emissions would range from 0.7 to 2.8 g TEQ/yr, well below the levels indicated by the German data. Based upon emission factors derived from these test results, nationwide emissions from all U.S. sinter plants are estimated to be 26 g TEQ/yr, which corresponds to less than one percent of current estimates of the national inventory from all sources. The operators of all seven active sinter plants as well as the two inactive plants were surveyed to obtain information on the oil content of their sinter feed. As shown in Table B-4, four of the active plants provided data that ranged in magnitude from 976 samples collected over one year (sampling about three times per day) to 14 samples collected over 14 months (monthly sampling). All four plants carefully monitor their sinter feed for oil to minimize emissions of volatile organic compounds. In addition, plants with baghouses are motivated to limit oil content due to concerns over blinding of bags and possible fire hazards. The other three active plants and the two inactive plants provided little data since none routinely monitor oil content. The four plants providing data reported long-term averages of 0.014, 0.02, 0.02 and 0.025 percent, respectively. TABLE B-4. HISTORICAL DATA FOR FOUR PLANTS WITH LOW OIL CONTENT* Plant LTV, IN9 Bethlehem, IN10 USS, IN11 WCI, OH12 Percent oil in sinter feed Average 0.014 0.02 0.02 0.025 Range 0.001 to 0.03 0.00 to 0.086 0.003 to 0.086 0.01 to 0.046 Description Plant samples routinely three times per day; results based on 976 samples Plant samples routinely once per month; results based on 48 samples Plant samples twice per week when blending with purchased scale; results are for 69 samples taken over one year (1999) Plant samples routinely once per month; results based on 14 samples over 14 months B-6 ------- * The oil content results for Ispat Inland13 and Bethlehem Steel (MD)14 were not considered representative because they were collected over a short period of time for special purposes. No data were available for AK Steel, Geneva Steel, and Wheeling-Pittsburgh Steel B.2 SINTER PLANT DISCHARGE END The sinter plant discharge end is comprised of sinter breakers (crushers), hot screens, conveyors, and transfer points that are designed to separate undersize sinter and to transfer the hot sinter to the cooler. In most cases, these discharge end operations are housed in a building. Emissions are usually controlled by local hooding and ventilation to one or more baghouses or wet scrubbers. Seven plants use baghouses and two plants use wet scrubbers. Details are given in Table B-5. Existing State regulations include both building opacity standards to limit releases of fugitive emissions (those escaping capture) and PM emission standards assigned to control devices. Five of the seven operating sinter plants are subject to a building opacity limit. One plant is subject to a 10 percent limit (6-minute average), and four plants are subject to 20 percent limits (6-minute average). The PM limits for control devices vary substantially from plant to plant both in terms of format and numerical values. Four plants have concentration limits for total PM (0.01, 0.02, 0.02, and 0.03 gr/dscf), one has concentration limits for PM10, and three have mass rate limits (42.9, 50, and 50 Ib/hr). Credible source test data on actual emissions were available from only one plant - the refurbished sinter plant in Youngstown, Ohio (Table B-6). Captured emissions from the discharge end are ventilated to a relatively new baghouse (1991) for control. No data were available on the opacity of fugitive emissions that escape capture from the discharge end. As noted above, five plants are subject to standards that limit the opacity of visible emissions released from the discharge end building (Table B-7). These range from 10 percent (one plant) to 20 percent (four plants) with a median value of 20 percent opacity based on a 6-minute average. For control devices, the top five most stringent existing emission limits for total PM are shown in Table B-8. These include the four concentration limits cited above and a fifth value derived from the lowest mass rate limit to which a plant is subject (42.9 Ib/hr), which is equivalent to 0.02 gr/dscf. The average of these five values is 0.02 gr/dscf. B-7 ------- TABLE B-5 . CONTROLS AND EMISSION LIMITS FOR THE DISCHARGE END3 Plant AK Steel, OH Bethlehem, MD Bethlehem, IN Geneva, UT Ispat-Inland, IN LTV, IN USX Gary, IN WCI, OH Wheeling- Pittsburgh, WV Control Baghouse Baghouse Baghouse Rotoclones (scrubbers) Baghouse Scrubber Baghouse 1 Baghouse 2 BaghouseA Baghouse Emission Points discharge, crusher, hot screen, cooler discharge, crusher, hot screen, cold screen discharge, crusher, hot screen discharge discharge, crusher, hot screen, /^ cooler discharge discharge, crusher hot and cold screens, conveyors discharge, crusher, hot screen, cold screen discharge Emission limit 50.0 Ib/hr 0.03 gr/dscf 42.9 Ib/hr 0.0096 gr/dscf PM10 0.01 gr/dscf 0.02 gr/dscf 0.02 gr/dscf PM10 0.0052 gr/dscf PM10 50.0 Ib/hr 0.02 gr/dscf Flow Rate (dscfm) 112,000 340,000 212,000 105,000 122,000 100,000 161,322 180,000 141,470 32,900 Best estimate of PM (gr/dscf 0.05 0.03 0.02 b 0.01 0.02 b b 0.04 0.02 a Compiled from 1993 industry survey b No equivalent PM limit could be estimated because the existing limit is expressed in terms of PM10. ------- TABLE B-6. SUMMARY OF EMISSIONS DATA FOR THE DISCHARGE END15 Plant WCI - baghouse for discharge, crusher, hot screen, cold screen (1991) average gr/dscf 0.0059 0.0053 0.0059 0.0057 Ib/hr 7.5 6.3 7.0 7.0 TABLE B-7. DISCHARGE END FUGITIVE EMISSIONS: TOP FIVE LIMITATIONS Plant Bethlehem, Sparrows Point, MD Ispat-Inland, East Chicago, IN LTV Steel, East Chicago, IN USX Steel, Gary, IN Geneva Steel, Provo, UT Median Limit for sinter building and fugitives 10% (6-min average) 20% (6-min average) 20% (6-min average) 20% (6-min average) 20% (6-min average) 20% (6-min average) TABLE B-8. DISCHARGE END CONTROL DEVICE: TOP FIVE LIMITATIONS Plant Ispat-Inland, East Chicago, IN Wheeling-Pittsburgh LTV Steel, East Chicago, IN Bethlehem, IN Bethlehem, MD Average gr/dscf 0.01 0.02 0.02 0.02* 0.03 0.02 *Estimated from Ib/hr limit and volumetric flowrate. B-9 ------- B-10 ------- B.3 SINTER PLANT COOLER Sinter plant coolers are large diameter circular tables through which ambient air is drawn to cool the hot sinter after screening. Seven plants operate sinter coolers to cool the sinter product prior to storage. Two plants that are not currently operating have no cooler and stockpile hot sinter directly. Of the seven plants with coolers, three vent directly to the atmosphere, one vents to a cyclone, two vent to a baghouse, and one vents half of the cooler exhaust to a baghouse with the remainder vented directly to the atmosphere. Five plants have emission limits expressed as concentration or mass rate while two plants have no emission limits (see Table B-9). TABLE B-9. SINTER COOLER DESCRIPTIONS AND LIMITS3 Plant Ispat-Inland WCI Bethlehem, Sparrows Point USS, Gary AK Steel, OH Bethlehem, Burns Harbor LTV, East Chicago Geneva Steel Description Baghouse controls the discharge, scrubber, hot screen and l/2 of cooler (one quadrant where the sinter is transferred to the cooler and one quadrant where it is removed); the other half is covered and vents through an uncontrolled stack. 20 minute residence time. Baghouse flow is 120,000 dscfm. Baghouse with forced air at 189,000 dscfm Cyclone at 320,000 dscfm and 0.02 gr/dscf; 90 to 120 minute residence time 3 coolers, uncontrolled; with hood and stack; 360,000 dscfm each Baghouse controls discharge, crusher, hot screen and cooler; flow of 1 12,000 dscfm Uncontrolled, with hood over cooler; 30-ft diameter and 575,000 dscfm; 60 minute residence time Uncontrolled; 60-ft diameter and 320,000 dscfm; 100 minute residence time Limit 0.01 gr/dscf (for controlled portion) 42.9 Ib/hr (about 0.03 gr/dscf) 0.03 gr/dscf 0.03 gr/dscf 50 Ib/hr (about 0.05 gr/dscf) no limit no limit These plants do not have coolers . Sinter is transferred from the hot screen LU a snjiayc puc tuiu tuuicu uy cunuieiii an. vviieeiiiiy-riusuuiyn aisu uses water sprays. B-ll ------- Wheeling-Pittsburgh Compiled from 1993 industry survey Information on actual releases is limited to one source test of controlled emissions from the cooler located at Youngstown, Ohio that is equipped with a baghouse (Table B-10). TABLE B-10. SUMMARY OF EMISSIONS DATA FOR THE SINTER COOLER 16 Description WCI- baghouse (1991) Run 1 2 3 average gr/dscf 0.018 0.0050 0.0052 0.0093 Ib/hr 29 8.3 8.0 15 As shown in Table B-9, three plants have concentration limits (0.01, 0.03, and 0.03 gr/dscf), and two plants have mass rate limits. The mass rates in Ib/hr were converted to equivalent concentration limits in gr/dscf based on the volumetric flow rate through the coolers. The two mass rate limits resulted in equivalent concentration values of 0.03 and 0.05 gr/dscf. The average of the five concentration limits shown in Table B-l 1 is 0.03 gr/dscf. TABLE B-ll. SINTER COOLER: TOP FIVE LIMITATIONS Plant Ispat-Inland, East Chicago, IN WCI Bethlehem, MD USS Gary AK Steel, OH Average gr/dscf 0.01 0.03* 0.03 0.03 0.05* 0.03 B-12 ------- * Estimated from the Ib/hr limit and volumetric flowrate. B.4 BLAST FURNACE CASTHOUSE The casthouse is a building or structure that encloses the section of the blast furnace where hot metal and slag are tapped from the furnace. The emissions from the blast furnace casthouse are fugitive emissions that escape through the roof monitor and other building openings during tapping. The emissions are primarily metal oxide fumes that are formed when air contacts the surface of the molten metal. Factors affecting these emissions include the duration of tapping, the exposed surface area of metal and slag, and the presence or absence of runner covers and flame suppression, which reduce contact with air. These emissions are controlled in one of two fundamentally different ways, flame suppression or conventional ventilation practices and control. Flame suppression consists of blowing natural gas over the iron runners and torpedo cars. The combustion of the gas consumes oxygen, which retards (suppresses) the formation of emissions. Ventilation practices employed include the use of localized hooding and ventilation applied at the iron trough and iron and slag runners. Alternatively, the casthouse may be totally enclosed and evacuated. Eighteen of the 39 blast furnaces have capture and control systems, 16 are controlled by baghouses, and two are controlled by wet scrubbers (see Table B-12). As a means for limiting fugitive emissions of PM from the casthouse during hot metal tapping, most States have developed visible emission standards that limit the opacity of emissions discharged from the casthouse roof monitor or other openings. These limits are given in Table B-12. The most common limit is 20 percent (6-minute average), which is applied to 24 of the 39 casthouses. States also apply particulate limits on gases discharged from control devices used to capture tapping emissions (see Table B-13). The most common form is a concentration limit, typically on the order of 0.01 gr/dscf. B-13 ------- As shown in Table B-14, the most stringent opacity limit is 15 percent (6-minute average) and is applied to two casthouses. The next most stringent limit is 20 percent (6-minute average), which is applied to 24 casthouses. B-14 ------- TABLE B-12. CASTHOUSE EMISSION CONTROLS AND OPACITY LIMITS Plant Acme Steel, IL AK Steel, KY AK Steel, OH Bethlehem Steel, IN Bethlehem Steel, MD Geneva Steel, UT Gulf States Steel, AL Ispat-Inland Steel, IN LTV Steel, OH LTV Steel, IN National Steel, IL National Steel, MI Rouge Steel, MI USX, PA USX, AL USX, IN USS/Kobe Steel, OH WCI Steel, OH Furnace A Amanda 3 C D L 1 2 3 1 7 5 6 Cl C5 C6 H3 H4 A B A B D B C 1 3 8 4 6 8 13 3 4 1 Casthouse control Flame suppression, covered runners Flame suppression, covered runners Flame suppression Flame suppression, inert suppression Flame suppression, inert suppression Baghouse, covered runners Flame suppression, covered runners Flame suppression, covered runners Flame suppression, covered runners No controls Baghouse Scrubber Scrubber Flame suppression, covered runners Flame suppression, covered runners Flame suppression, covered runners Flame suppression, covered runners Baghouse, flame suppression, covered runners Baghouse, covered runners Baghouse, covered runners Baghouse Baghouse Baghouse Flame suppression, covered runners Flame suppression, covered runners Baghouse Baghouse Baghouse, covered runners Flame suppression Flame suppression Flame suppression Baghouse, covered runners Baghouse, covered runners Flame suppression Baghouse Casthouse opacity limit 20%, 6 minute average 20%, 6 minute average Covered under a "bubble" No opacity limit No opacity limit 5%, 6 minute average, 20% drilling, O2 lance and mudding For all: 20%, except for any aggregate of 3 min. (12 readings) in any 60 min. 15%, 6 minute average 20%, 6 minute average 20%, 6 minute average 20%, 6 minute average 15%, 6 min.avg, exceptions to 20% 20%, 6 minute average 20%, 6 minute average 20%, 6 minute average 20%, 6 minute average 20%, 6 minute average 20%, 6 minute average 20%, 6 minute average 20%, 6 minute average 20%, 6 minute average 20%, 6 minute average For both: Not to equal or exceed 20% except for 12 readings per hr. 20%, 6 minute average 20%, 6 minute average 20%, 6 minute average 20%, 6 minute average 20%, 6 minute average 15%, 6 minute average 20%, 6 minute average 20%, 6 minute average B-15 ------- Plant Weirton Steel, WV Wheeling Pittsburgh Steel, OH Furnace l 4 1 5 Casthouse control Baghouse, flame suppression, covered runners Flame suppression, covered runners Flame suppression, covered runners Baghouse, flame suppression, covered runners Casthouse opacity limit 20%, except 40% for 5 min/hour 20%, except 40% for 5 min/hour 20%, 6 minute average 5% to 20% TABLE B-13. EMISSION LIMITS FOR CASTHOUSE CONTROL DEVICE Plant Bethlehem Steel, MD Ispat-Inland, IN LTV Steel, IN National Steel, IL National Steel, MI USX, PA USS/Kobe, OH WCI Steel, OH Wheeling- Pittsburgh, OH Furnace L 7 H4 A B A B D 1 3 3 1 5 Control Baghouse Baghouse 1 Baghouse 2 Baghouse Baghouse 1 Baghouse 2 Baghouse Baghouse Baghouse Baghouse Baghouse Baghouse Baghouse Baghouse Capture Points Evacuated runner covers & hoods Canopy hood Runners Hood over tilting spout & iron trough Suspended hood 6 air hoods, 3 at each furnace with damper control Hoods over trough & pouring spouts — each furnace Air curtain Evacuated runner covers & hoods Trough hood, covered runners, hood at tilting runners Emission Limit 0.03 gr/dscf 0.003 gr/dscf 0.011 gr/dscf No limit 0.01 gr/dscf 0.01 gr/dscf 0.0075 gr/dscf 0.02 lb/1000 Ib gas 0.0052 gr/dscf No limit No limit 0.0052 gr/dscf 0.03 Ib/ton 0.311b/hr; proposed PM10 Iimitof5.931b/hr B-16 ------- B-17 ------- TABLE B-14. BLAST FURNACE CASTHOUSE: TOP FIVE LIMITATIONS Plant Ispat-Inland, East Chicago, IN USS/Kobe, Lorain, OH WCI Steel, Warren, OH Acme, Riverdale, IL AK Steel, KY (and several others) Furnace 7 O 1 A A Median Opacity limit 15% (6-min average) 15% (6-min average) 20% (6-min average) 20% (6-min average) 20% (6-min average) 20% (6-min average) B-18 ------- As shown in Table B-12, there are 18 casthouses equipped with hooding and ventilation equipment to limit fugitive emissions. Sixteen use a baghouse for the control of captured emissions. Industry survey information on the baghouses indicate they are similar in design and performance (Table B-15). Most are pulse jet baghouses with air to cloth ratios of around 4 acfm/ft2. Performance test data were available for four of the 16 baghouses and are presented in Figure B-2 and Table B-16. The database includes a total of eight source tests; four tests at one facility, two tests at another facility, and single tests at the two other facilities. Each performance test is comprised of three individual test runs. The three run averages for each of the eight tests range from 0.002 to 0.009 gr/dscf. Results from individual runs range from 0.001 to 0.009 gr/dscf. The highest emitting unit is the one at National Steel in Granite City, IL facility for which there are four independent performance tests. The performance tests range from 0.006 to 0.009 gr/dscf with individual runs ranging from 0.003 to 0.009 gr/dscf. Three tests were conducted in 1988 and one in 1985, and all tests met the facility's State limit of 0.01 gr/dscf. B-19 ------- FIGURE B-2. TEST RESULTS FOR BAGHOUSES IN BLAST FURNACE CASTHOUSES u> U.U IUU n nnan •. .UUoU .uu/ u .UUoU .uuou .UU4U .UUoU .UUzU .UU IU n nnnn - < 1 4 < > 1 r > > i i r , 4 1 1 <• s 1 1 ^ 1 II 0 Bethlehem, WCI Steel (1) WCI Steel (1) National, National, National, National, Wheeling Pitt. SP (L) Granite City Granite City Granite City Granite City (5) (A & B) (A & B) (A & B) (A & B) Plant TABLE B- 15. CONTROL DEVICE PARAMETERS3 B-20 ------- Furnaces with baghouses Plant Bethlehem Steel Inland Steel LTV Steel National Steel National Steel USX Steel USX Steel USS/Kobe State MD IN IN IL MI PA IN OH Capacity (tpy) 3,450,000 4,000,000 1,971,000 2,372,500 2,000,000 900,000 2,000,000 1,200,000 1,100,000 3,440,000 1,300,000 Furnace 1 7 H4 A B A B D 1 o J 13 o J Flow (dscfm) 420,000 acfm @170-200 - 250,000-275,000 220,000 369,000 100,000 acfm 400,000 170,000 275,000 140,000 140,000 600,000 acfm 224,000 Air/cloth ratio (acfm/ft2) 4.0 - 4.2 4.4 6.88 5.82 5.15 9.0 5.38 - - 4.8 6.28 • P(in- water) 8 - 7 7 14 10 3-8 4-8 3-6 3-12 3-12 <8 3-10 Cleaning pulse jet - pulse jet pulse jet pulse jet shaker reverse air pulse jet pulse jet - - pulse jet pulse jet Filter material polyester - polyester polyester polyester polyester polyester needle felt polyester felt polyester woven - - polyester felt polyester Location Runner covers Canopies over 4 notches Iron trough & tilting spout "A" & "B" taphole Torpedo cars Iron trough/ tilting spout Casthouse Casthouse Casthouse B-21 ------- TABLE B- 15. CONTROL DEVICE PARAMETERS (continued) Furnaces with baghouses Plant WCI Steel Wheeling- Pittsburgh State OH OH Capacity (tpy) 1,500,000 1,682,000 Furnace 1 5 Flow (dscfm) 125,000 103,200 Air/cloth ratio (acfm/ft2) 1.98-2.23 4.5 • P(in- water) - 4-6 Cleaning shaker pulse jet Filter material - polyester felt Location Casthouse Furnaces with wet scrubbers Plant Inland Steel State IN Capacity (tpy) 1,253,000 1,253,000 Furnace 5 6 Flow (dscfm) 40,000 acfm @250«Fx(2) 40,000 acfm @250«F L/G (gal/1000 acf) 10.0 10.0 * P (in. water) 24-30 35 Scrubber type Multi-element fixed throat vertical rod type scrubber (2 scrubbers) Multi-element fixed throat (1 scrubber) Demister vanes in tank Location Local hoods over notch, iron and slag runners, and pugh ladles Compiled from 1993 industry survey air/cloth ratio = ratio of air flow to cloth area in actual cubic feet per minute per square foot of cloth • p = pressure drop in inches of water B-22 ------- TABLE B-16. PM DATA FOR CASTHOUSE CONTROL DEVICE National Steel, Granite City (A & B)17 24 August 1988 Runl Run 2 Run 3 average 4 May 1988 Runl Run 2 Run 3 average 23 February 1988 Runl Run 2 Run 3 average 4 January 1985 Runl Run 2 Run 3 average dscfm 333,787 324,833 333,206 330,609 dscfm dscfm dscfm 340,906 360,747 373,219 358,291 gr/dscf 0.0087 0.0058 0.0033 0.0059 gr/dscf 0.0093 0.0093 0.0091 0.0093 gr/dscf 0.0074 0.0088 0.0052 0.0071 gr/dscf 0.0092 0.0061 0.0039 0.0064 Ib/hr 25.00 16.23 9.41 16.88 Ib/hr 25.68 24.49 25.21 25.13 Ib/hr 21.14 26.06 14.93 20.71 Ib/hr WCI Steel18 29 May 1996 Runl Run 2 Run 3 average 17 November 1992 Runl Run 2 Run 3 average dscfm 268,200 266,220 264,960 266,460 dscfm 193,700 gr/dscf 0.0030 0.0020 0.0070 0.0040 gr/dscf 0.0040 0.0010 0.0028 0.0026 Ib/hr 7.03 5.04 15.64 9.24 Ib/hr 7.66 1.45 4.45 4.52 Wheeling Pittsburgh (5)19 August 1999 Runl dscfm 102,840 gr/dscf 0.0029 Ib/hr 2.56 B-23 ------- Run 2 Run 3 average 106,120 100,640 103,200 0.0012 0.0032 0.0024 1.09 2.76 2.14 Bethlehem, SP "L" Furnace20 18 June 1996 Runl Run 2 Run 3 average dscfm 140,016 140,474 140,897 140,462 gr/dscf 0.0037 0.0006 0.0016 0.0019 Ib/hr 4.38 0.67 1.87 2.30 B-24 ------- B.5 BOPF PRIMARY EMISSIONS Primary emissions from the BOPF refer to the particulate emissions generated during the steel production cycle which are captured and controlled by the furnace's primary emission control system. The majority of the emissions occur during the oxygen blow. The oxygen blow is the period in the steel production cycle when oxygen is lanced or injected into the vessel. Some shops operate open hood furnaces and others use closed hood systems. Open and closed hood vessels are very different in terms of operation, pollutant loading, and emissions. Open hood systems are characterized by very high primary exhaust air flowrates due to the large quantities of combustion air introduced at the furnace mouth to support CO combustion. In contrast, closed hood systems, which include hoods that are tightly fitted to the vessel to suppress CO combustion, are characterized by much lower exhaust air flowrates. Typical flowrates for open hood shops are 200,000 to 500,000 acfm, while closed hood designs are usually less than 100,000 acfm. There are 50 BOPF located in 23 BOPF shops. The 50 BOPF include 34 furnaces with open hood systems at 16 shops and 16 furnaces with closed hood systems at 8 shops. All of the BOPF have capture and control systems for the primary emissions. For the open hood systems, 8 shops are controlled by venturi scrubbers and 8 shops are controlled by electrostatic precipitators (ESPs). All 8 of the closed hood shops are controlled by venturi scrubbers. Each shop is subject to existing State limits with a wide variety of formats, including concentration limits in gr/dscf and lb/1,000 Ib gas for PM or PM10, mass emission rate limits in Ib/hr, and process weighted limits in Ib/ton of steel. In addition, the emission test period required for compliance with the existing State limits varies from testing over the steel production cycle, only during the oxygen blow, for 1-hour runs, and for 2-hour runs. Emission limits are summarized in Tables B-17 and B-18. B-25 ------- TABLE B-17. EMISSION LIMITS FOR PRIMARY CONTROL - OPEN HOOD Open Hood BOPF Shops Plant Acme Steel Bethlehem Steela Bethlehem Steel Gulf States Steel Inland No. 4 LTV Steel LTV No. 1 Shop National Steel National Steel Rouge Steel USX Gary (BOPF) USX Gary(Q-BOP) USX Edgar Thomson WCI Steel Weirton Steel Wheeling-Pittsburgh State IL IN MD AL IN IN OH IL MI MI IN IN PA OH WV OH Control ESP Scrubber Scrubber ESP Scrubber ESP ESP ESP ESP ESP Scrubber Scrubber Scrubber ESP Scrubber Scrubber Emission Limit 0.028 gr/dscf 0.09 Ib/ton liquid steel 0.03 gr/dscf — 0.1871b/ton 0.018 gr/dscf PM10 39.8 Ib/hr 60.0 Ib/hr or 0.255 Ib/ton 0.057 lb/1000 Ib gas 0.02 gr/dscf PM10 0.02 gr/dscf PM10 Process rate 62.90 Ib/hr 0.03 gr/dscf 21.40 Ib/hr; 7.09 Ib/hr PM10 (pending) 3 Two furnaces are open hood and one is closed hood. For the data analysis, the control performance for open and closed hood furnaces was evaluated separately due to the operational differences and volumetric air flowrates between the two designs as discussed previously. This is consistent with the development of separate standards for open and closed hood vessels for the NSPS in 40 CFR Part 60, Subpart N. The NSPS for open hood B-26 ------- BOPF is a PM limit of 0.022 gr/dscf and for closed hood the PM standard is 0.030 gr/dscf. For both types of furnaces the NSPS PM limit is measured during the primary oxygen blow. B-27 ------- TABLE B-18. EMISSION LIMITS FOR PRIMARY CONTROL - CLOSED HOOD Closed Hood BOPF Shops Plant AK Steel AK Steel Geneva Steel Inland No. 2 LTV No. 2 USS/Kobe Steel USX Fairfield State KY OH UT IN OH OH AL Control Scrubber Scrubber Scrubber Scrubber Scrubber Scrubber Scrubber Emission Limit 0.03 gr/dscf 114 Ib/hr1 0.02 gr/dscf PM10 2 0.058 Ib/ton 15 Ib/hr (for each of 2 stacks) 45.0 Ib/hr 0.022 gr/dscf;3 process rate4 1 Both vessels combined 2 During oxygen blow 3 Furnace C, subject to NSPS, Subpart NN, which is 0.022 gr/dscf for closed hood shops 4 Furnaces X & U B.5.1 Open Hood BOPF Control devices applied to primary emissions at open hood shops include both ESP and venturi scrubbers (see Table B-19). Source test data and design information are available for seven of the 16 open hood shops, five with ESP and two with venturi scrubbers. The test data indicate that the ESP perform better than the venturi scrubbers. All the test data (based on charge-to-tap measurements) for the ESP are less than 0.019 gr/dscf (see Figure B-3 and Table B-20). All of the ESP are similar in design and operation. All have three to five fields in series and operate at specific collection areas greater than 300 ft2/l,000 cfm. Data for the two plants with venturi scrubbers, operating at pressure drops of 25 to 35 inches of water, averaged 0.025 and 0.035 gr/dscf, respectively. B-28 ------- TABLE B-19. OPERATING PARAMETERS OF OPEN HOOD PRIMARY CONTROL Wet Scrubber Control Technology Plant Bethlehem Bethlehem Inland (#4) USS Steel (BOPF) USS Steel (Q- BOP) USS Steel Weirton Steel Wheeling- Pittsburgh State IN MD IN IN IN PA WV OH Capacity (million tpy) 5.4 4.0 2.7 2.9 4.0 2.8 3.2 2.95 Flow (dscfm) 11 3,200^ 3 600,000! 310,000- 380,000 268,000 x 3 267,000 x 3 174,000 x 2 280,000 210,000 L/G (gal/1,000 acf) 20 8 1.0 13.1 34.7 ~ ~ 10 • p (in. water) 55 50 25 70-75 70 68-76 502 50 ESP Control Technology Plant Acme Steel Gulf States LTV Steel LTV (#1) National National Rouge Steel WCI Steel State IL AL IN OH IL MI MI OH Capacity (million tpy) 1.3 1.3 4.2 3.3 3.6 4.1 3.3 1.7 Flow (dscfm) 288,000 327,000 847,000! 550,000 330,000 500,000! 500,000 440,000 Collectio n plate area (ft2' 92,000 150,000 650,000 255,000 ~ 80,200 ~ 114,000 No. of fields in series 3 4 (2 sets) 5 4 4 4 4 6 SCA (ft2/l,000 cfm)3 320 460 770 560 ~ 160 ~ 260 1 acfm 2 Scrubber upgrade increased the pressure drop from 35 to 50 inches. 3 SCA = specific collection area B-29 ------- FIGURE B-3. OPEN HOOD SHOPS WITH ESPs (charge to tap testing) 0.025 0.015 -• U) 0.010 -- 0.005 -- 0.000 H 1 1 1 h H h H 1 h LTV, OH LTV, OH LTV, OH LTV, OH LTV, OH LTV, OH LTV, OH WCI WCI WCI LTV, IN Acme National, IL Plant B-30 ------- TABLE B-20. BOPF TEST DATA: OPEN HOOD SHOPS WITH ESP AND CHARGE-TO-TAP TESTING Acme Steel21 November 1998 Runl Run 2 Run 3 average dscfm 129,149 139,031 140,072 136,084 gr/dscf 0.0059 0.0096 0.0042 0.0066 Ib/hr 6.49 11.43 5.02 7.65 LTV, Cleveland #1 BOPF shop22 26 October 1989 Runl Run 2 Run 3 average dscfm 510,238 531,430 580,559 540,742 gr/dscf 0.0085 0.0064 0.0036 0.0062 Ib/hr 37.10 29.34 17.78 28.07 LTV Cleveland #1 BOPF shop23 20 November 1986 Runl Run 2 Run 3 average dscfm 486,825 493,755 504,465 495,015 gr/dscf 0.0248 0.0063 0.0050 0.0120 Ib/hr 104.46 26.97 21.49 50.97 LTV Cleveland #1 BOPF shop24 25 November 1985 Runl Run 2 Run 3 average dscfm 544,252 573,541 531,181 549,658 gr/dscf 0.0079 0.0053 0.0064 0.0065 Ib/hr 36.78 26.28 29.40 30.82 LTV Cleveland #1 BOPF shop25 8 April 1985 Runl Run 2 Run 3 average dscfm 503,922 466,345 463,267 477,845 gr/dscf 0.0089 0.0082 0.0078 0.0083 Ib/hr 34.81 30.36 26.95 30.71 B-31 ------- TABLE B-20. BOPF TEST DATA: OPEN HOOD SHOPS WITH ESP AND CHARGE-TO-TAP TESTING LTV Cleveland #1 BOPF shop26' 27 18 October 1984 Runl Run 2 Run 3 average dscfm 337,400 348,300 356,700 347,467 gr/dscf 0.0047 0.0042 0.0027 0.0038 Ib/hr 13.51 12.40 8.16 11.36 LTV Cleveland #1 BOPF shop28 Particulate emissions with 7 of 8 sections of ESP 3 January 1983 Runl Run 2 Run 3 Run 4 Run5 Run 6 Run? Run8 average dscfm 384,000 388,000 372,800 388,300 363,700 334,800 347,400 378,800 369,725 gr/dscf 0.0090 0.0095 0.0070 0.0080 0.0085 0.0080 0.0085 0.0075 0.0083 Ib/hr 30.20 32.00 23.70 25.80 24.90 23.60 19.90 24.10 25.53 LTV Cleveland #1 BOPF shop29 9 December 1982 Runl Run 2 Run 3 Run 4 average dscfm 426,000 439,200 441,400 425,900 433,125 gr/dscf 0.0136 0.0141 0.0080 0.0050 0.0102 Ib/hr 48.60 52.60 29.50 17.90 37.15 LTV, East Chicago30 20 August 1992 Runl Run 2 Run 3 average dscfm 490,329 433,827 450,196 458,117 gr/dscf 0.0059 0.0251 0.0140 0.0150 Ib/hr 24.83 93.26 54.05 57.38 B-32 ------- National Steel, Granite City31 30 March 1989 Runl Run 2 Run 3 average dscfm 349,127 332,540 337,902 339,856 gr/dscf 0.0216 0.0190 0.0170 0.0192 Ib/hr 64.64 54.16 49.17 55.99 WCI Steel18 32 33 12 April 1996 Runl Run 2 Run 3 average 25 August 1993 Runl Run 2 Run 3 average 17 May 1990 Runl Run 2 Run 3 average dscfm 460,970 436,470 423,450 440,297 dscfm 380,200 391,552 413,012 394,921 dscfm 371,888 372,305 368,611 370,935 gr/dscf 0.0130 0.0140 0.0160 0.0143 gr/dscf 0.0165 0.0114 0.0147 0.0142 gr/dscf 0.0076 0.0109 0.0027 0.0071 Ib/hr 52.07 52.28 56.34 53.56 Ib/hr 53.91 38.27 52.05 48.08 Ib/hr 24.31 34.91 8.41 22.54 B-33 ------- B.5.2 Closed Hood BOPF All 16 of the furnaces at the 8 closed hood shops use high-energy venturi scrubbers. Closed hood systems produce an exhaust gas high in CO which precludes the use of other types of control devices (such as baghouses or ESP) due to potential explosion or fire hazards. Information on the design and operation of these scrubbers shown in Table B-21 were obtained through an industry survey. These scrubbers operate at a pressure drop of 50 inches of water or more, and most have liquid-to-gas ratios greater than 10 gallons per thousand cubic feet of gas. Recent test data were available for only one of the eight closed hood shops with testing during the oxygen blow. However, performance test data were available from five other furnaces that were used to develop the NSPS. All tests include three test runs and all were performed only during the oxygen blow. Each of these plants use a high-energy venturi scrubber with a pressure drop of 50 inches of water or more. The three run averages for each of the six tests range from 0.015 to 0.024 gr/dscf. Results from individual runs range from 0.013 to 0.031 gr/dscf. The data are presented in Figure B-4 and Table B-22. B-34 ------- TABLE B-21. OPERATING PARAMETERS OF CLOSED HOOD VENTURI SCRUBBERS Plant AK Steel AK Steel Bethlehem Steel Geneva (Q- BOP) Inland Steel (No. 2) LTV Steel (No. 2) USS/Kobe US Steel State KY OH IN UT IN OH OH AL Capacity (million tpy) 2.17 2.71 ~ 2.5 2.5 4.38 2.6 2.2 Vessel 1 2 15 16 3 1 2 1 2 1 2 L N U X c Flow (dscfm) 78,000 78,000 40,000 51,000 197,000a 78,300 77,300 50,000- 60,000 50,000- 60,000 55,000 55,000 58,000 59,000 ~ 76,000 76,000 L/G (gal/1,000 acf) 11.5 11.5 2.9 2.6 21 ~ ~ 10.0 10.0 ~ ~ ~ ~ ~ ~ ~ • p (in. water) 60 60 45-50 40-50 55 70-80 70-80 55 55 ~ ~ ~ ~ 60-95 51-92 59-96 aacfm B-35 ------- FIGURE B-4. CLOSED HOOD BOPF TEST DATA (all for the oxygen blow) 0.030 0.025 ^•••••••^••••••••••••••••••••••••••••••^••••••••••••••••••••••••••••••••••••••••••••iV"i Or\i~tr\ I 1 J** yv .UZU T I M ^^ O | 0.015 O) O 0.010 Q.005 0.000 Geneva USS Fairfield USS Fairfield Kaiser (Furnace 5) Kaiser (Furnace 6) Armco Middletown (Furnace C) (Furnace X) Plant B-36 ------- TABLE B-22. BOPF TEST DATA: CLOSED HOOD SHOPS WITH VENTURI SCRUBBERS AND TESTING DURING THE OXYGEN BLOW Geneva Steel34 16 June 1992 Scrubber 1 Scrubber 2 average dscfm 82,000 77,000 79,500 gr/dscf 0.024 0.019 0.022 Ib/hr 16.90 12.60 14.80 USS Fairfield, Furnace C35 October 1978 Runl Run 2 Run 3 average dscfm 74,600 76,600 77,600 76,300 gr/dscf 0.021 0.021 0.023 0.022 Ib/hr 13.58 13.86 15.43 14.29 USS Fairfield, Furnace X35 December 1978 Runl Run 2 Run 3 average dscfm 78,600 76,100 74,600 76,400 gr/dscf 0.019 0.024 0.021 0.021 Ib/hr 12.67 15.39 13.21 13.76 Kaiser No. 535 December 1978 Runl Run 2 Run 3 average dscfm 82,000 87,000 90,400 86,571 gr/dscf 0.021 0.020 0.018 0.020 Ib/hr 6.44 5.46 6.02 5.97 Kaiser No. 635 December 1978 Runl Run 2 Run 3 average dscfm 79,000 68,000 83,000 76,500 gr/dscf 0.015 0.013 0.017 0.015 Ib/hr 4.68 3.48 3.75 3.97 Armco Steel35 October 1971 Runl Run 2 dscfm 37,000 32,000 gr/dscf 0.021 0.031 Ib/hr -- -- B-37 ------- Run 3 average 49,000 39,000 0.020 0.024 -- -- B.6 SECONDARY BOPF EMISSION CONTROL Secondary or fugitive emissions occur from the BOPF when the molten iron and scrap metal are charged to the furnace and when the molten steel and slag are tapped from the furnace. The emissions generated are primarily metal oxides formed when oxygen in the air reacts with the molten iron or steel. Twelve of the 23 BOPF shops have a separate capture and control system for BOPF charging and tapping emissions. Ten of these shops use baghouses and the other two use scrubbers. Existing State limits for the control devices are summarized in Table B-23 and range from 0.0052 to 0.015 gr/dscf, and the NSPS limit is 0.01 gr/dscf. The most common limit is 0.01 gr/dscf. TABLE B-23. STATE LIMITS FOR BOPF SECONDARY CONTROLS Closed Hood BOPF Shops Plant Bethlehem Steel Geneva Steel Inland No. 2 Shop LTV No. 2 Shop USS/Kobe Steel USX Fan-field State IN UT IN OH OH AL Control Scrubber Baghouse Scrubber Baghouse Baghouse Baghouse Limit 0.05 Ib/ton liquid steel (#3) 0.002 gr/dscf PM10 0.015 Ib/ton 0.010 gr/dscf 0.012 gr/dscf 0.010 gr/dscf Open Hood BOPF Shops Plant Acme Steel Inland No. 4 Shop USX, Gaiy (Q-BOP) USX Braddock State IL IN IN PA Control Baghouse Baghouse Baghouse Baghouse Actual Limit 10.22 Ib/hr, 0.0052 gr/dscf 0.006 gr/dscf 0.0052 gr/dscf PM10 Process weight limit The top five most stringent existing emission limits for total PM are given in Table B-24. The five plants with the most stringent secondary BOPF emission State limits are subject to concentration B-38 ------- limits of 0.0052, 0.006, 0.01, 0.01 and 0.012 gr/dscf. Each of these is associated with a facility with baghouse controls. The median of the five values is 0.01 gr/dscf. Available data on secondary BOPF emissions (Table B-25) is limited to one test run at a facility using a baghouse. This one test run includes measurements of multiple baghouse modules and averaged 0.001 gr/dscf. It is not likely that one test run will adequately reflect the full range of performance of a particular technology, and the results of the one available test run appear to represent, at most, what this type of control is able to achieve under very favorable circumstances. TABLE B-24. BOPF SECONDARY CONTROLS: TOP FIVE LIMITATIONS Plant Acme Steel, IL Inland, IN LTV, OH USX,AL USS/Kobe Steel, OH Shop 1 4 2 1 1 Median gr/dscf 0.0052 0.006 0.01 0.01 0.012 0.01 TABLE B-25. BOPF SECONDARY BAGHOUSE TEST AT USX, BRADDOCK, PA36 (October 12-13,1993) Baghouse module 1 2 O 4 5 6 7 dscfm 66,700 59,800 64,000 63,400 61,400 65,200 66,400 Weighted average gr/dscf gr/dscf 0.00157 0.00008 0.00075 0.00011 0.00151 0.00163 0.00233 0.001 B-39 ------- B-40 ------- B.7 HOT METAL TRANSFER, DESULFURIZATION, SLAG SKIMMING, AND LADLE METALLURGY There are several different ancillary operations performed within the BOPF shop: (1) operations associated with the molten iron before it is charged to the BOPF (hot metal transfer, desulfurization, and slag skimming), and (2) treatment of the molten steel after tapping (various ladle metallurgy operations). The emissions from these operations are primarily metal oxides formed when oxygen in the air reacts with the molten iron or steel. Molten iron is transported from the blast furnace casthouse to the BOPF shop in a torpedo car and transferred to a vessel at the reladling (or hot metal) station, where it is usually desulfurized and slag is skimmed from the surface. Emissions from these operations are captured by local hooding and controlled by a baghouse. Existing State emission limits for these operations range from 0.0052 to 0.04 gr/dscf, but most are on the order of 0.01 gr/dscf (see Table B-26). The steel from the BOPF is usually transferred to a ladle where final adjustments in temperature and chemistry are made in an operation known as ladle metallurgy. Emissions from ladle metallurgy are captured by a close fitting hood and ducted to a baghouse. Existing State limits for ladle metallurgy are a mixture of mass emission rates in Ib/hr and concentration limits in gr/dscf. The mass emission rate limits range from 0.42 to 7.5 Ib/hr and the concentration limits range from 0.0052 to 0.02 gr/dscf (Table B-27). B-41 ------- TABLE B-26. STATE LIMITS FOR TRANSFER, DESULFURIZATION, AND SLAG SKIMMING-ALL BAGHOUSES Plant Acme Steel AK Steel AK Steel Bethlehem Steel Geneva Steel Inland Steel, No. 2 Inland Steel, No. 4 LTV Steel National Steel Rouge Steel National Steel USS, Edgar USS, Fairfield USS Gary USS Gary USS/Kobe Steel WCI Steel Weirton Steel Wheeling-Pittsburgh Steel State IL KY OH IN UT IN IN IN IL MI MI PA AL IN IN OH OH WV OH Process Transfer, desulfurization skimming Transfer, desulfurization skimming Transfer and desulfurization Deslagger Transfer, desulfurization skimming Desulfurization Buildings 1&2 Reladle and desulfurization Reladle and desulfurization Reladle and desulfurization Transfer, desulfurization skimming Transfer and desulfurization Hot metal transfer Reladle and desulfurization Reladle and desulfurization Desulfurization Reladle and desulfurization Transfer and desulfurization Desulfurization Hot metal transfer Desulfurization Hot metal transfer Desulfurization Hot metal transfer backup Emission Limit 10.2 Ib/hr 0.01 gr/dscf 58 Ib/hr 0.03 gr/dscf 23.1 Ib/hr 0.011 gr/dscf PM,n 0.011 gr/dscf 0.0052 gr/dscf 0.008 gr/dscf PM,n 0.01 gr/dscf __ 0.007 gr/dscf Process weight rate 0.01 gr/dscf 0.01 gr/dscf 0.0052 gr/dscf PM,n 0.03 gr/dscf 0.04 gr/dscf 0.01 gr/dscf 5.97 Ib/hr 5.01 Ib/hr (proposed) 6.41 Ib/hr (proposed) B-42 ------- TABLE B-27. STATE LIMITS FOR LADLE METALLURGY PROCESS Plant Acme Steel AK Steel AK Steel AK Steel Inland Steel, No. 2 LTV Steel National Steel National Steel National Steel National Steel National Steel Rouge Steel Rouge Steel USS Fail-field USS Gary Q-BOP USS Gary Q-BOP USS/Kobe Weirton Steel Wheeling-Pittsburgh Wheeling-Pittsburgh State IL KY OH OH IN IN IL IL MI MI MI MI MI AL IN IN OH WV OH OH Control Baghouse Baghouse Baghouse Baghouse3 Baghouse Baghouse Baghouse 1 Baghouse 2 Baghouse lb Baghouse 2a Baghouse 3C Baghouse 1 Baghouse 2 Baghouse Baghouse 1 Baghouse 2 Baghouse Baghouse Baghouse Baghouse Emission Limit 0.037 lb PM10/ton 3.8 Ib/hr 0.02 gr/dscf 0.03 gr/dscf 0.0052 gr/dscf 0.004 gr/dscf PM10 0.01 gr/dscf 0.01 gr/dscf 1.261b/hr 2.131b/hr 1.1 Ib/hr 7.50 Ib/hr 1.61b/hr 0.02 gr/dscf 0.01 gr/dscf PM10 0.01 gr/dscf PM10 0.002 gr/dscf 0.42 Ib/hr 0.54 lb/hrd 2.3 Ib/hr, 0.02 gr/dscf Vacuum degassing Ladle metallurgy, No. 2 argon stirring No. 1 argon stirring station Proposed limit B-43 ------- Source test data were available for three of the 23 baghouses that control emissions from hot metal transfer and desulfurization and for seven of the 20 baghouses that control emissions from ladle metallurgy. These data are shown in Figures B-5 and B-6, and data for each run are given in Tables B- 28 and B-29. Each performance test is comprised of three individual runs. The three run averages for the ten tests range from 0.001 to 0.012 gr/dscf. Results from individual runs range from 0.001 to 0.021 gr/dscf. The highest three run averages and highest individual runs were examined more closely. In this case, both were obtained on the same baghouse, 0.012 and 0.021 gr/dscf. An examination of the test results on all 10 baghouses indicates that these results are 2 to 2.5 times higher than those obtained on the next highest emitting unit, suggesting that this baghouse is either an underperformer or that the test results include an outlier. Eliminating the 0.021 gr/dscf value from the three run average produces an average of 0.007 gr/dscf which is in line with the next highest emitting unit's three run average of 0.006 gr/dscf and the highest individual run of 0.0085 gr/dscf. Consequently, the 0.021 gr/dscf value is an outlier and does not reflect the level of performance demonstrated to be achievable for a baghouse applied to emissions from hot metal transfer, desulfurization, and ladle metallurgy operations. B-44 ------- FIGURE B-5. TRANSFER AND DESULFURIZATION TEST DATA - ALL BAGHOUSES u.uzo • 0.02 • 0.015 • •5 in •a u> 0.01 • 0.005 • n . One run outlier— not characteristic of a i baghouse Tested 13 modules for transfer and desulfurization; one run pjr. module i 1 1 Inland No. 4(1979) Wheel-Pitt (1992) Plant USS, PA (1993) B-45 ------- FIGURE B-6. LADLE METALLURGY TEST DATA - ALL BAGHOUSES u.u iuu - 0.0090 - 0.0080 - 0.0070 - 0.0060 - •5 | 0.0050 - U) 0.0040 - 0.0030 - 0.0020 - 0.0010 - n nnnn - lance injection, desulfurization, add alloys \ < lance injection, lance injection, fslofH" rnm^inn ctti/^ reheat, add alloys stirring, add alloys capped argon huJhbling 4 I I I I I $ argon stirring, fefieat 1 1 1 1 1 i I s 1 - o lance injection, ar.gQ.n.s.tir.ring add alloys 1 USS/Kobe#1 Acme Inland, No. 2 WC I Steel Plant LTV, IN Wheeling-Pitt LTV, OH No. 2 B-46 ------- TABLE B-28. TEST DATA FOR METAL TRANSFER, DESULFURIZATION Inland #4 (July 1979)37 Runl Run 2 Run 3 Flowrate, dscfm 165,000 165,000 163,000 3 -run average gr/dscf 0.0022 0.0017 0.0010 0.0016 Ib/hr 3.11 2.41 1.40 2.31 Wheeling-Pittsburgh (October 1992 - desulfurization)19' 38 Runl Run 2 Run 3 average Flowrate, dscfm 69,930 65,030 69,070 68,010 gr/dscf 0.0058 0.0207 0.0085 0.0117 Ib/hr 3.5 11.5 5.0 6.7 Wheeling-Pittsburgh (July 1980 - hot metal transfer)19' 38 Runl Run 2 Run 3 average Flowrate, dscfm 182,336 176,416 179,656 179,469 gr/dscf 0.0051 0.0016 0.0016 0.0027 Ib/hr 8.0 2.4 2.5 4.3 TABLE B-29. TEST DATA FOR LADLE METALLURGY Acme Steel, Chicago, IL21 Runl Run 2 Run 3 average Flowrate, dscfm 71,923 74,924 78,618 75,155 gr/dscf 0.0085 0.0035 0.0046 0.0055 Ib/hr 5.24 2.25 3.10 3.53 Inland, No. 2 BOPF shop39 1 1 Sept 86 Runl Run 2 Run 3 average Flowrate, dscfm 46,920 47,490 44,080 46,163 gr/dscf 0.0043 0.0015 0.0019 0.0026 Ib/hr 1.70 0.60 0.70 1.00 B-47 ------- LTV, E. Chicago40 15 Jun 89 Runl Run 2 Run 3 average Flowrate, dscfm 130,324 125,203 134,437 129,988 gr/dscf 0.0055 0.0041 0.0035 0.0044 Ib/hr 6.18 4.35 4.02 4.85 LTV Cleveland, No. 2 BOPF shop41 21 Apr 93 Runl Run 2 Run 3 average Flowrate, dscfm 127,872 147,083 125,950 133,635 gr/dscf 0.0078 0.0034 0.0028 0.0047 Ib/hr 8.59 4.28 3.02 5.30 USS/Kobe Steel, #2 LMF42 5 Nov 97 Runl Run 2 Run 3 Average Flowrate, dscfm gr/dscf 0.0011 0.0014 0.0013 0.0012 Ib/hr 0.55 0.72 0.65 0.64 WCI Steel43 4 Nov 91 Runl Run 2 Run 3 average Flowrate, dscfm 71,139 85,810 76,195 77,715 gr/dscf 0.0050 0.0028 0.0027 0.0035 Ib/hr 3.07 2.09 1.79 2.32 Wheeling Pittsburgh38 29 Sept 95 Runl Run 2 Run 3 average Flowrate, dscfm 39,400 36,830 39,330 38,540 gr/dscf 0.0010 0.0016 0.0037 0.0021 Ib/hr 0.35 0.50 1.24 0.70 B-48 ------- B.8 BOPF SHOP FUGITIVE EMISSIONS The BOPF shop is a building or structure that houses several operations involved in steelmaking. These include hot metal transfer, desulfurization, slag skimming stations; one or more BOPF's for refining iron into steel; and ladle metallurgy stations. Fugitive emissions from these operations in the BOPF shop exit through the roof monitor and other building openings. Table B-30 summarizes existing opacity limits for BOPF shops. Top and bottom blown furnaces were evaluated independently based on operational differences between the two designs. For top blown furnaces, the most stringent and also the most common State standard is a 20 percent limit (3-minute average) that is applied to 13 of the 20 BOPF shops that operate top blown furnaces. For bottom blown furnaces, the BOPF shop with the most stringent standard (Geneva Steel) is subject to a 10 percent opacity limit (6-minute average, with one exception per cycle up to 20 percent). A second shop (USX Gary) has three furnaces subject to a 20 percent limit (3-minute average). A third shop (USX Fairfield) has two furnaces subject to a 20 percent limit (6-minute average), and a third furnace subject to a 10 percent limit (3-minute average), with one 3-minute average greater than 10 percent but less than 20 percent applied only during hot metal transfer or skimming operations. Similar to the existing State standards, the NSPS for top blown furnaces applies during the entire production cycle. However, the NSPS for bottom blown furnaces applies only during periods of hot metal transfer and slag skimming. Both standards limit opacity to less than 10 percent (3-minute average), except that one 3-minute average greater than 10 percent but less than 20 percent can occur during each applicable performance period. B-49 ------- TABLE B-30. SUMMARY OF BOPF SHOP OPACITY LIMITS BOPF Shop Acme Steel, Riverdale, IL AK Steel, Ashland, KY AK Steel, Middletown, OH Bethlehem, Bums Harbor, IN (3 vessels in 1 shop) Bethlehem, Sparrows Point, MD Gulf States, Gadsden, AL Inland Steel, East Chicago, IN (2 shops) LTV, Cleveland, OH (2 shops) LTV, East Chicago, IN National, Granite City, IL National, Ecorse, MI Rouge Steel, Dearborn, MI USX Braddock, PA USX Gary, IN USS/Kobe,Lorain,OH WCI Steel, Warren, OH Weirton Steel Weirton, WV Wheeling-Pittsburgh, OH Geneva Steel, Orem, UT USX Fan-field, AL USX Gary, IN Type Top Top Top Top (2) Top (1) Top Top Top Top Top Top Top Top Top Top Top Top Top Top Top Top Bottom Bottom Bottom Primary control ESP Scrubber Scrubber Scrubber Scrubber ESP Scrubber Scrubber ESP Scrubber ESP ESP ESP ESP Scrubber Scrubber Scrubber ESP Scrubber Scrubber Scrubber Scrubber Scrubber Secondary control Baghouse Baghouse None None Scrubber None None Scrubber Baghouse Baghouse None None Baghouse None Baghouse Baghouse None None None Baghouse Baghouse Baghouse Roof monitor opacity limit 20%, 3 minute average 20% except for 3 min/hr Covered under "bubble" 40%, 6 minute average; <60% for 1 5-min in 6 hr 3 -day roll avg of 15% (6-min avg), except 3 min/hr 20%, 3 minute average 20%, 3 minute average 20%, 3 minute average 20%, 3 minute average 20%, 3 minute average 20%, 3 minute average 20%, 3 minute average 20%, 3 minute average 20%, 3 minute average Not to equal or exceed 20% except fo: 1 2 readings per hour. 20%, 3 minute average 20%, 3 minute average None 20% 20%, 3 minute average 10%, 6 minute average3 10%, 3-min avg/20%, 6-min avg* 20%, 3 minute average a Allows one 6-min average per steel production cycle up to 20%. b One furnace has a limit of 10% (3-min average) for hot metal transfer and skimming with one 3-min average per cycle over 10% but less than 20%; the other 2 furnaces have a 20% (6-min average) limit. B-50 ------- B.9 COST ESTIMATES FOR BAGHOUSES APPLIED TO SINTER PLANT DISCHARGE END AND COOLER The cost estimates are based on guidance provided in the OAQPS Cost Manual (Chapter 5: Fabric Filters)44 and the associated spreadsheet.45 The baghouse design is a pulse jet unit with an air- to-cloth ratio of 3 acfm/ft2. For the discharge end, a typical ventilation rate of 120,000 acfm is used, and a typical rate of 200,000 acfm is used for the sinter cooler. B.9.1 Capital and Total Annual Costs The capital cost elements for the two baghouses are given in Table B-31. The list of items associated with annual operating costs from the OAQPS cost manual are given in Table B-32 and are used to estimate the total annualized costs presented in Table B-3 3. TABLE B-31. CAPITAL COST ELEMENTS Item Baghouse Bags Cages Auxiliaries (hoods, ductwork, fans, stacks) Total Purchased equipment cost (1.18) Index (1.02 for 1998 to 1999) Retrofit factor (2) Total capital investment, including installation (2.17) Capital cost (120,000 acfm) 368,827 73,784 25,460 209,287 678,359 800,463 816,472 1,630,000 3,500,000 Capital cost (200,000 acfm) 607,352 122,973 42,436 280,525 1,053,286 1,242,872 1,267,729 2,536,000 5,500,000 B-51 ------- TABLE B-32. OPERATING COST ELEMENTS Item Operating hours per year Operating labor rate Maintenance labor rate Labor overhead Operating labor required Maintenance labor required Supervisory labor Maintenance materials Electricity usage Electricity cost Compressed air cost Dust disposal Taxes, insurance, administration Interest rate Bag life Capital recovery factor for bags Control system life Capital recovery factor for control system Value 8,760 $17.27/hr $17.74/hr 60% 2 hr/shift 1 hr/shift 15% equal to maintenance labor kw-hr/yr = 0.00018 x acfm x • p x 8,760 $0.0671 kw-hr $0.25/1,000 scf $25/ton 4% 7% 2 years 0.553 20 years 0.0944 B-52 ------- TABLE B-33. ESTIMATES OF TOTAL ANNUAL COSTS Item Operating labor Supervisory labor Maintenance labor Maintenance materials Labor overhead Electricity Compressed air Bag replacement Dust disposal Tax, insurance, administration Capital recovery Total annual cost Annual cost ($/yr for 120,000 acfm) 37,282 5,592 19,159 19,159 48,715 100,593 31,104 70,416 44,434 141,773 322,541 840,800 Annual cost ($/yr for 200,000 acfm) 37,282 5,592 19,159 19,159 48,715 167,656 51,840 117,362 74,057 220,130 499,439 1,260,000 B.9.2 Emission Reduction and Cost Effectiveness Emission reductions and cost effectiveness are presented for two cases: (1) installing a baghouse on the discharge end to reduce emissions of PM from 0.02 gr/dscf (the MACT floor) to 0.01 gr/dscf and (2) installing a baghouse on the sinter cooler to reduce emissions of PM from 0.03 gr/dscf (the MACT floor) to 0.01 gr/dscf. Data from two plants showed that the HAP content of dust from the discharge end ranged from 0.3 percent46 of PM to 1.2 percent.47 For this estimate, use a midrange value of 0.75 percent for both the discharge end and cooler because the dust from the cooler should be similar in composition to that from the discharge end. B-53 ------- B-54 ------- The PM and HAP emission reductions for the two cases are given in Table B-34. The cost effectiveness ranges from $1.2 to $2.5 million per ton of HAP reduced. TABLE B-34. EMISSION REDUCTIONS AND COST EFFECTIVENESS Emission point Discharge end (120,000 acfm) Cooler (200,000 acfm) PM emission reduction (tpy) 45 150 HAP emission reduction (tpy) 0.34 1.1 Total annual cost ($million/yr) 0.84 1.3 Cost effectiveness ($million/ton HAP) 2.5 1.2 B.10 REFERENCES 1. Stack test Results for WCI Sinter Baghouse Provided by T. Shepkar, WCI Steel: Envisage Environmental Inc. and CSA Company on May 27, 1992. 2. Test Report. Particulate, metals, and gaseous emissions study, performed for Inland Steel Company, Sinter Plant, Windbox Baghouse Stack, East Chicago, Indiana, on May 16-17, 1995, Mostardi-Platt Associates, July 18, 1995. 3. Test report. Emission testing for particulate, sulfur dioxide, and sulfuric acid mist for north and south sinter plant windbox stacks. Bethlehem Steel, Sparrows Point, MD. By Entropy Environmentalists. July 23-24, 1991. 4. Integrated Iron and Steel Industry: Test Report for the Sinter Plant at LTV Steel Company, East Chicago, Indiana. Final Report. Prepared by Eastern Research Group, Inc. for EPA's Emission Measurement Center. September 1997. 5. Letter. M. Fischer, Hamilton County Environmental Services, to J. Calcagni, RTI, February 11, 1998. Enclosing stack test reports for the basic oxygen furnace and sinter plant windbox at AK Steel Corporation. 6. Test Report. Mostardi-Platt Associates, Inc., Particulate Compliance Study Performed for Bethlehem Steel Corporation at the Burns Harbor Plant, Burns Harbor, Indiana Sinter Plant Scrubber Stack March 9 and 11, 1992, submitted April 6, 1992. 7. Article, U. Lahl and S. Lindenstr, Sinter Plants of Steel Industry - PCDD/F Emissions Status and Perspective, Chemosphere, 1994,2(9-11): 1939-1945. B-55 ------- 8. Integrated Iron and Steel Industry: Test Report for the Sinter Plant at Youngstown Sinter Co. (WCI Steel). Youngstown, Ohio. Final Report. Prepared by Eastern Research Group, Inc. for EPA's Emission Measurement Center. January 1998. 9. Letter. R. Zavoda, LTV Steel, to P. Mulrine: EPA:OAQPS:ESD, April, 2000. Completed sinter feed oil content survey for the Indiana Harbor Division of LTV Steel. 10. Letter. T. Easterly, Bethlehem Steel, to P. Mulrine, EPA:OAQPS:ESD, March 21, 2000. Enclosing the completed sinter feed oil content survey for the Bums Harbor Division of the Bethlehem Steel Corporation. 11. Letter. W. Kubiak, U.S. Steel, to P. Mulrine, EPA:OAQPS:ESD, April 2, 2000. Enclosing response to information request regarding the oil content of sinter feed materials at US Steel's sinter plant at Gary, Indiana facility. 12. Letter . T. Shepkar, Youngstown Sinter Company, to P. Mulrine, EPA:OAQPS:ESD, March 17, 2000. Enclosing response to "percent oil in sinter feed" questionnaire along with the analytical method being used to analyze for oil in sinter scale. 13. Letter. G Allie, Ispat Inland, to P. Mulrine, EPA:OAQPS:ESD, March 17, 2000. Enclosing response to request for information concerning oil analysis in sinter plant feeds. 14. Letter. J. Schindler, Bethlehem Steel, to P. Mulrine, EPA:OAQPS:ESD, March 24, 2000. Response to information request for the sinter feed oil content - for the No. 7 sinter plant at Bethlehem's Sparrows Point plant. 15. Test Report. CSA Company, Source Emissions Test at Warren Consolidated Industries Warren, Ohio, Sinter Plant "A" Baghouse Outlet, October 24, 1991. 16. Test Report. CSA Company, Source Emissions Test at Warren Consolidated Industries Warren, Ohio, Sinter Cooler Baghouse, November 20, 1991. 17. Siebenberger, L., Naational Steel, Granite City, IL to B. Jordan, EPA. Screening survey response. 1991. 18. Test data for WCI Steel & USS Kobe Steel from Northeast Ohio EPA, testing conducted in 1996-1997. 19. Letter. H. Strohmeyer to J. Calcagni, RTI, February 19, 1998. Enclosing summaries of recent stack tests performed at Wheeling-Pittsburgh Steel Corporation, Steubenville. B-56 ------- 20. Letter. L. Daniel, Maryland Department of the Environment, to P. Mulrine, EPA:OAQPS:ESD, March 30, 1998. Enclosing Stack Test for "L" Blast Furnace Baghouse at Bethlehem Steel, Sparrows Point, conducted September 3, 1996. 21. Letter. Wentz, Jeffrey, Acme Steel, Riverdale, IL to P. Mulrine, EPA. Comments on draft background information document for integrated iron and steel plants. January 6, 1999. 22. Test Report. LTV Steel Company, West Cleveland, Ohio, Basic Oxygen Furnace - Electrostatic Precipitator, Particulate Emission Evaluation, Envisage Environmental Incorporated, October 17, 25-26, 1985. 23. Test Report. LTV Steel Company, EOF #1 Cleveland, Ohio, EPA Methods 1-5, Emissions Evaluation BOF, ESP Exhaust, Envisage Environmental Incorporated, November 18-20, 1986. 24. Test Report. LTV Steel Company, Cleveland, Ohio, BOF #1, Cleveland West, EPA Methods 1-5, Particulate Emission Evaluation, Prepared by Envisage Environmental Incorporated, November 25-27, 1985. 25. Test Report. Envisage Environmental Incorporated, LTV Steel Corporation Cleveland West #1 BOF Shop ESP, EPA Methods 1-5 Particulate Compliance Test, April 8 & 9, 1985. 26. Test Report. Supplement to particulate compliance testing, BOF nos. 94 and 95 precipitator, LTV Steel Corporation, Cleveland, Ohio, October 19, 1984, submitted by WFI Sciences Company, November 28, 1984. 27. Letter. R Nemeth, LTV Steel Company, to Chief, EPA:AMD:ACB and Commissioner of Cleveland Division of Air Pollution Control, December 5, 1984. Enclosing final report of the electrostatic precipitator compliance test conducted October 16-18, 1984 at LTV No. 1 BOF shop, and summaries of opacity meter calibration, hot metal analysis and flux, BOF heat analysis, and opacity recorder charts. 28. Test Report. Particulate emissions with seven sections, BOF nos. 94 and 95 precipitator, prepared by WFI Science Company for Jones and Laughlin Steel Corporation, January 28, 1983. 29. Test Report. Compliance testing of stack emissions, BOF nos. 94 and 95, Jones and Laughlin Steel Corporation, Cleveland, Ohio, conducted on December 9, 10, 13, and 14, 1982, WFI Sciences Company, January 14, 1983. 30. Test Report. Mostardi-Platt Associates, Inc., Precipitator Performance Test Program Performed for LTV Steel Company at the Indiana Harbor Works BOF East Chicago, Indiana, August 18 and 20, 1992. Submitted November 5, 1992. B-57 ------- 31. Letter. Heintz, J.K., National Steel, Mishawaka, IN to P. Mulrine, EPA. Comments on draft background information document for integrated iron and steel plants. December 14, 1998. 32. Test Report. Source Emission Test at WCI Steel, Inc., Warren, Ohio, BOF Precipitator Stack, performed by CSA Company, August 25, 1993. 33. Test Report. Warren Consolidated Industries, Inc., Warren, Ohio, Basic Oxygen Furnace Electrostatic Precipitator Particulate Emission Evaluation, performed by Envisage Environmental Incorporated May 17, 1990. 34. Letter. M. Maxell, State of Utah department of Environmental Quality, Division of Air Quality, to J. Calcagni, RTI, August 17, 1993. Enclosing copies of the State Implementation Plan, emissions test data, and the approval orders for Quelle Basic Oxygen Process (Q-BOP), the Sinter Plant, the Blast Furnace, and the Primary Mill Hot Scarfing Machine Facility. 35. Revised Standards for Basic Oxygen Process Furnaces - Background Information for Proposed Standards. EPA-450/3-82-005a. December 1982. pp 4-47 to 4-50. 36. Letter and attachments. A Lorenzi, U.S. Steel, to J. Calcagni, RTI, February 16, 1998. Enclosing summary pages from compliance demonstrations performed in 1987 and 1993, and a copy of the report Basic Oxygen Process Compliance Demonstration, U.S. Steel Corporation, Mon Valley Works, Edgar Thompson Plant, Braddock, Pennsylvania, prepared by Geraghty & Miller, Inc. for U.S. Steel, January 1997. 37. Letter. H. Taylor, The Almega Corporation, to Inland Steel Company, July 3, 1979, summarizing test methods and results for the Baghouse Exhaust Particulate Emission Testing of the Fume Emission Control System for the Hot Metal Transfer Station at Inland Steel, East Chicago, Illinois on June 28, 1979; enclosing summary of emission test data and process weight rate summary data. 38. Letter . H. Strohmeyer, Wheeling Pittsburgh Steel Corporation, to P. Mulrine, EPA:OAQPS:ESD, April 25, 2000. Enclosing corrections to the MACT Floor Analysis for Integrated Iron and Steel Plants - Sinter Plants, Blast Furnaces, and Basic Oxygen Furnace Shops. 39. Test Report. Report on Particulate Emissions, Prepared by Clean Air Engineering, Incorporated for Inland Steel Company, September 17, 1986. 40. Test Report. Mostardi-Platt Associates, Inc., Baghouse Performance Study Performed for LTV Steel Company at the LMF Baghouse of Indiana Harbor Works Indiana Harbor, Indiana, June 15, 1989. B-58 ------- 41. Test Report. LTV Steel Company, Cleveland, Ohio, #2 EOF, LMF Baghouse exhaust, paniculate emissions evaluation, Envisage Environmental Incorporated, April 21, 1993. 42. Letter. Ames, H., USS/Kobe to P. Mulrine. Enclosing comments on draft background information document for integrated iron and steel plants. December 15, 1998. 43. Test Report. CSA Company, Source Emissions Test at Warren Consolidated Industries Warren, Ohio, LMF Baghouse Outlet, October 31 and November 4, 1991. 44. U.S. Environmental Protection Agency. OAQPS Control Cost Manual. 5th edition. Chapter 5: Fabric Filters. EPA453/B-96-001. February 1996. 4 5. Available at http ://www. epa. gov/ttn/catc 46. Anderson, D.A., Bethlehem Steel, Sparrows Point, MD to B. Jordan, EPA. Response to section 114 request. August 29, 1991. 47. Shoup, S.P., Inland Steel, East Chicago, IN to B. Jordan, EPA. Response to section 114 request. November 12, 1993. B-59 ------- TECHNICAL REPORT DATA (Please read Instructions on reverse before completing) 1. REPORT NO. 2. EPA-453/R-0 1-005 4. TITLE AND SUBTITLE National Emission Standards for Hazardous Air Pollutants (NESHAP) for Integrated Iron and Steel Plants - Background Information for Proposed Standards 7. AUTHOR(S) Marvin Branscome and Stacey Molinich, RTI and Phil Mulrine, EPA 9. PERFORMING ORGANIZATION NAME AND ADDRESS U.S. Environmental Protection Agency Office of Air Quality Planning and Standards Research Triangle Park, NC 2771 1 12. SPONSORING AGENCY NAME AND ADDRESS John Seitz, Director Office of Air Quality Planning and Standards Office of Air and Radiation U.S. Environmental Protection Agency Research Triangle Park, NC 2771 1 3. RECIPIENT'S ACCESSION NO. 5. REPORT DATE January 2001 6. PERFORMING ORGANIZATION CODE 8. PERFORMING ORGANIZATION REPORT NO. 10. PROGRAM ELEMENT NO. 11. CONTRACT/GRANT NO. 68-D6-0014 13. TYPE OF REPORT AND Final PERIOD COVERED 14. SPONSORING AGENCY CODE EPA/200/04 15. SUPPLEMENTARY NOTES 16. ABSTRACT This report provides the background information for the proposed NESHAP to control metal and organic hazardous air pollutants (HAPs) from integrated iron and steel plants. The emission control techniques, estimates of emissions, control costs, and environmental impacts are presented. 17. KEY WORDS AND DOCUMENT ANALYSIS a. DESCRIPTORS emission controls environmental impacts estimates of air emissions 18. DISTRIBUTION STATEMENT Release Unlimited b. IDENTIFIERS/OPEN ENDED TERMS Air Pollution Control Iron and Steel Plants Hazardous Air Pollutants 19. SECURITY CLASS (Report) Unclassified 20. SECURITY CLASS (Page) Unclassified c. COSATI Field/Group 21. NO. OF PAGES 205 22. PRICE EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDITION IS OBSOLETE ------- |