National Emission Standards for Hazardous Air Pollutants for Iron and Steel Foundries

AGENCY:

Environmental Protection Agency (EPA).

ACTION:

Final rule.

SUMMARY:

This action promulgates national emission standards for hazardous air pollutants (NESHAP) for iron and steel foundries. The EPA has identified iron and steel foundries as a major source of hazardous air pollutant (HAP) emissions. These standards implement section 112(d) of the Clean Air Act (CAA) by requiring all major sources to meet HAP emissions standards reflecting application of the maximum achievable control technology (MACT).

The HAP emitted by facilities in the iron and steel foundries source category include metal and organic compounds. For iron and steel foundries that produce low alloy metal castings, metal HAP emitted are primarily lead and manganese with smaller amounts of cadmium, chromium, and nickel. For iron and steel foundries that produce high alloy metal or stainless steel castings, metal HAP emissions of chromium and nickel can be significant. Organic HAP emissions include acetophenone, benzene, cumene, dibenzofurans, dioxins, formaldehyde, methanol, naphthalene, phenol, pyrene, toluene, triethylamine, and xylene. Exposure to these substances has been demonstrated to cause adverse health effects, including cancer and chronic or acute disorders of the respiratory, reproductive, and central nervous systems. When fully implemented, the final rule will reduce HAP emissions from iron and steel foundries by over 820 tons per year (tpy).

EFFECTIVE DATE:

April 22, 2004.

ADDRESSES:

The official public docket is available for public viewing at the EPA Docket Center, EPA West, Room B102, 1301 Constitution Ave., NW., Washington, DC 20460.

FOR FURTHER INFORMATION CONTACT:

Kevin Cavender, Metals Group (C439-02), Emission Standards Division, Office of Air Quality Planning and Standards, U.S. EPA, Research Triangle Park, NC 27711, telephone number (919) 541-2364, electronic mail (e-mail) address, cavender.kevin@epa.gov.

SUPPLEMENTARY INFORMATION:

Regulated Entities. Categories and entities potentially regulated by this action include:

This table is not intended to be exhaustive, but rather provides a guide for readers regarding entities likely to be regulated by this action. To determine whether your facility is regulated by this action, you should examine the applicability criteria in § 63.7682 of the final rule. If you have any questions regarding the applicability of this action to a particular entity, consult the person listed in the preceding FOR FURTHER INFORMATION CONTACT section.

Docket. The EPA has established an official public docket for this action including both Docket ID No. OAR-2002-0034 and Docket ID No. A-2000-56. The official public docket consists of the documents specifically referenced in this action, any public comments received, and other information related to this action. All items may not be listed under both docket numbers, so interested parties should inspect both docket numbers to ensure that they have received all materials relevant to the final rule. Although a part of the official public docket, the public docket does not include Confidential Business Information or other information whose disclosure is restricted by statute. The official public docket is available for public viewing at the EPA Docket Center (Air Docket), EPA West, Room B-102, 1301 Constitution Avenue, NW., Washington, DC. The EPA Docket Center Public Reading Room is open from 8:30 a.m. to 4:30 p.m., Monday through Friday, excluding legal holidays. The telephone number for the Reading Room is (202) 566-1744, and the telephone number for the Air Docket is (202) 566-1742.

Electronic Docket Access. You may access the final rule electronically through the EPA Internet under the Federal Register listings at http://www.epa.gov/fedrgstr/ .

An electronic version of the public docket is available through EPA's electronic public docket and comment system, EPA Dockets. You may use EPA Dockets at http://www.epa.gov/edocket/ to view public comments, access the index listing the contents of the official public docket, and to access those documents in the public docket that are available electronically. Once in the system, select “search,” then key in the appropriate docket identification number. Although not all docket materials may be available electronically, you may still access any of the publicly available docket materials through EPA Dockets. (See Docket No. A-2000-56 in the Air Docket).

Worldwide Web (WWW). In addition to being available in the docket, an electronic copy of today's final rule is also available on the WWW through the Technology Transfer Network (TTN). Following the Administrator's signature, a copy of the rule will be placed on the TTN's policy and guidance page for newly proposed or promulgated rules at http://www.epa.gov/ttn/oarpg. The TTN provides information and technology exchange in various areas of air pollution control. If more information regarding the TTN is needed, call the TTN HELP line at (919) 541-5384.

Judicial Review. This action constitutes final administrative action on the proposed NESHAP for iron and steel foundries (67 FR 78274, December 23, 2002). Under section 307(b)(1) of the CAA, judicial review of the rule is available only by filing a petition for review in the U.S. Court of Appeals for the District of Columbia Circuit by June 21, 2004. Only those objections to the NESHAP which were raised with reasonable specificity during the public comment period may be raised during judicial review. Under section 307(b)(2) of the CAA, the requirements that are the subject of today's final rule may not be challenged separately in civil or criminal proceedings brought by the EPA to enforce these requirements.

Outline. The information presented in this preamble is organized as follows:

I. Background

II. Summary of the Final Rule

A. What Is the Affected Source?

B. What Are the Emissions Limitations?

C. What Are the Operation and Maintenance (O&M) Requirements?

D. What Are the Requirements for Demonstrating Initial and Continuous Compliance?

E. What Are the Notification, Recordkeeping, and Reporting Requirements?

F. What Are the Compliance Deadlines?

III. Summary of Environmental, Energy, and Economic Impacts

A. What Are the Air Quality Impacts?

B. What Are the Cost Impacts?

C. What Are the Economic Impacts?

D. What Are the Non-air Health, Environmental, and Energy Impacts?

IV. Summary of Major Comments and Responses

A. Why Did We Revise the Proposed Affected Source Designation?

B. Why Did We Revise the Proposed Emissions Limits?

C. Why Did We Revise the Proposed Work Practice Standards?

V. Statutory and Executive Order Reviews

A. Executive Order 12866: Regulatory Planning and Review

B. Paperwork Reduction Act

C. Regulatory Flexibility Act

D. Unfunded Mandates Reform Act

E. Executive Order 13132: Federalism

F. Executive Order 13175: Consultation and Coordination with Indian Tribal Governments

G. Executive Order 13045: Protection of Children From Environmental Health and Safety Risks

H. Executive Order 13211: Actions That Significantly Affect Energy Supply, Distribution, or Use

I. National Technology Transfer Advancement Act

J. Congressional Review Act

VI. Statutory Authority

I. Background

Section 112(d) of the CAA requires us (the EPA) to establish national emission standards for all categories and subcategories of major sources of HAP and for area sources listed for regulation under section 112(c). Major sources are those that emit or have the potential to emit at least 10 tpy of any single HAP or 25 tpy of any combination of HAP. Area sources are stationary sources of HAP that are not major sources. Additional information on the NESHAP development process can be found in the preamble to the proposed rule (67 FR 78274).

We received a total of 83 comment letters on the proposed NESHAP from trade associations, individual plants, consultants, vendors, State agencies, environmental groups, and private citizens. We provided a 60-day comment period and held a public hearing on January 22, 2003 to provide the opportunity for oral presentations of data, views, or arguments concerning the proposed rule.

Today's final rule reflects our full consideration of all the comments we received. A detailed response to all the comments is included in the Background Information Document (BID) for the Promulgated Standards (Docket ID No. OAR-2002-0034).

II. Summary of the Final Rule

A. What Is the Affected Source?

The affected source is each new or existing iron and steel foundry that is a major source of HAP emissions. A new affected source is an iron and steel foundry for which construction or reconstruction began after December 23, 2002. An existing affected source is an iron and steel foundry for which construction or reconstruction began on or before December 23, 2002. The final rule defines an “iron and steel foundry” as:

A facility or portion of a facility that melts scrap, ingot, and/or other forms of iron and/or steel and pours the resulting molten metal into molds to produce final or near final shape products for introduction into commerce. Research and development facilities and operations that only produce non-commercial castings are not included in this definition.

The final rule covers emissions from metal melting furnaces, scrap preheaters, pouring areas, pouring stations, automated conveyor and pallet cooling lines that use a sand mold system, automated shakeout lines that use a sand mold system, and mold and core making lines. The final rule also covers fugitive emissions from foundry operations.

B. What Are the Emissions Limitations?

The final rule includes emissions limits for metal and organic HAP as well as operating limits for capture systems and control devices. Particulate matter (PM) and opacity serve as surrogate measures of metal HAP emissions; emissions limits for total metal HAP are included as alternatives to the PM limits. The final rule also includes emissions limits for volatile organic HAP (VOHAP) and triethylamine (TEA). Except for the fugitive emissions opacity limit, each of the emissions limits apply to emissions discharged to the atmosphere through a conveyance. The term “conveyance” means the system of equipment that is designed to capture pollutants, convey them through ductwork, and exhaust them using forced ventilation. The opacity limit for fugitive emissions applies to each building or structure housing any emissions source at the iron and steel foundry. The emissions limitations and work practice requirements are:

The final rule requires a capture system for those emissions sources subject to VOHAP or TEA limits. You (the owner or operator) must establish operating limits for identified capture system parameter (or parameters) that are appropriate for assessing capture system performance. At a minimum, the limits must indicate the level of ventilation draft and damper position settings. You must operate the capture systems at or above the lowest value or setting established in the operation and maintenance (O&M) plan.

If you use a wet scrubber to control PM or total metal HAP emissions from a metal melting furnace, scrap preheater, pouring area, or pouring station, the 3-hour average pressure drop and scrubber water flow rate must not fall below the minimum levels established during the initial (or subsequent) performance test. If you use a combustion device to control VOHAP emissions from a cupola metal melting furnace, the 15-minute average combustion zone temperature must not fall below 1,300 degrees Fahrenheit (°F). Periods when the cupola is off blast and for 15 minutes after going on blast from an off blast condition are not included in the 15-minute average. If you use a combustion device to control VOHAP emissions from a scrap preheater or TEA cold box mold or core making line, the 3-hour average combustion zone temperature must not fall below the minimum level established during the initial (or subsequent) performance test. If you use a wet acid scrubber to control TEA emissions, the 3-hour average scrubbing liquid flow rate must not fall below the minimum level established during the initial (or subsequent) performance test and the 3-hour average pH level of the scrubber blowdown (or the pH level during a production shift) must not exceed 4.5.

Operating limits do not apply to control devices for automated conveyor and pallet cooling lines or automated shakeout lines that use a sand mold system at a new iron and steel foundry. The final rule requires a continuous emissions monitoring system (CEMS) for these emissions sources. However, the final rule includes procedures for requesting alternative monitoring requirements. To obtain approval of alternative monitoring requirements, you must submit a monitoring plan containing information needed to demonstrate continuous compliance along with performance test results showing compliance with the emissions limit.

The final rule also includes work practice standards. Facilities must meet certification requirements for their charge materials or develop and implement a scrap selection and inspection program to minimize the amount or organics and HAP metals in furnace charge materials. The certification option requires the foundry to purchase and use only certified-metal ingots, pig iron, skittle, or other materials that do not include post-consumer automotive body scrap, post-consumer engine blocks, oil filters, oily turnings, lead components, mercury switches, plastics, or organic liquids. The scrap selection plan option requires scrap specifications, a certification that the scrap supplier has implemented procedures to remove mercury switches and lead components from automotive scrap, and visual inspection procedures to ensure materials meet the specifications.

The owner or operator of an existing iron and steel foundry must install, operate, and maintain a gas-fired preheater where the flame directly contacts the scrap charged. As alternative compliance options, the owner or operator may meet a 20 ppmv limit for VOHAP emissions or may charge to a preheater only materials subject to the scrap certification requirement. The owner or operator of a new iron and steel foundry must meet the 20 ppmv limit for VOHAP emissions and the operating limit for combustion devices. As an alternative compliance option for new scrap preheaters, the owner or operator must meet the scrap certification requirements.

Plants with a furan warm box mold or core making line at a new or existing iron and steel foundry must use a binder chemical formulation that contains no methanol, as listed in the Material Data Safety Sheet. This requirement applies to the catalyst portion (and not the resin portion) of the binder system.

C. What Are the Operation and Maintenance Requirements?

All foundries must prepare and follow a written operation and maintenance (O&M) plan for capture systems and control devices. The plan must include operating limits for capture systems; requirements for inspections and repairs; preventative maintenance procedures and schedules; and procedures for operation of bag leak detection systems (including corrective action steps to be taken in the event of a bag leak detection system alarm). The plan also must contain procedures for igniting gases from mold vents in pouring areas and pouring stations that use sand mold systems. These procedures may consider the ignitability of the mold gases, accessibility to the molds, and safety issues associated with igniting the gases.

The final rule also requires a startup, shutdown, and malfunction plan that meets the requirements in § 63.6(e) of the NESHAP General Provisions. The plan must include procedures for operating and maintaining the emissions source during periods of startup, shutdown, and malfunction and a program of corrective action for malfunctioning process equipment, air pollution control systems, and monitoring systems. The final rule requires that the plan also include a description of the conditions that constitute a shutdown of a cupola and normal operating conditions following startup of a cupola. The owner or operator may use the standard operation procedures manual for the emissions source or other type of plan if it meets EPA's requirements. For more information on startup, shutdown, and malfunction plans, see the amendments to the NESHAP General Provisions published on May 30, 2003 (68 FR 32586).

D. What Are the Requirements for Demonstrating Initial and Continuous Compliance?

Emissions Limits

Foundries must demonstrate initial compliance by conducting performance tests for all emissions sources subject to an emissions limit. To determine compliance with the metal HAP emissions limits, EPA Methods 1 through 4, and either Method 5, 5B, 5D, 5F, or 5I, as applicable (to measure PM) or Method 29 (to measure total metal HAP) are required. To determine compliance with the organic HAP limits, foundries can use EPA Method 18 to measure VOHAP, Method 25 to measure total gaseous nonmethane organics (TGNMO) as hexane, or Method 25A to measure total organic compounds (TOC) as hexane. All of these methods are in appendix A to 40 CFR part 60.

The performance test requirements for automated conveyor and pallet cooling lines and automated shakeout lines at new foundries allow you to either meet the 20 ppmv emissions limit directly using the volatile organic compound (VOC) CEMS to measure total hydrocarbons (as a surrogate for VOHAP) or to establish a site-specific VOC limit for the CEMS that is correlated to the VOHAP emissions limit. The final rule also includes procedures for computing the flow-weighted average of multiple exhaust streams from automated conveyor and pallet cooling lines or automated shakeout lines, and for determining compliance for combined emissions streams. Procedures for establishing operating limits for capture systems and control devices, and revising the limits, if necessary or desired, after the initial performance test are given in § 63.7733 of the final rule. Previous performance tests (conducted since December 22, 2002) may be used to establish operating limits.

Monitoring of capture system and control device operating parameters is required to demonstrate continuous compliance with the operating limits. These requirements include bag leak detection systems for baghouses and continuous parameter monitoring systems (CPMS) for capture systems (unless damper positions are fixed) and control devices. For wet acid scrubbers, the final rule allows plants to measure the pH every 8 hours during process operations using a pH probe and meter as an alternative to a pH CPMS. The owner or operator of automated conveyor and pallet cooling lines or automated shakeout lines that use a sand mold system at a new iron and steel foundry must monitor organic HAP emissions using a CEMS unless they apply for alternative monitoring requirements. Technical specifications, along with requirements for installation, operation, and maintenance of CPMS and CEMS, are included in the final rule. Records are required to document compliance with the monitoring, inspection, and maintenance requirements for monitoring equipment. The final rule requires performance tests every 5 years to demonstrate continuous compliance with the PM (or total metal HAP), VOHAP, and TEA emissions limits and every 6 months to demonstrate continuous compliance with the opacity limit for fugitive emissions. Subsequent performance tests are not required for foundries that demonstrate continuous compliance using a CEMS.

Work Practice Standards

No performance test is required to demonstrate initial compliance with the work practice standards. Foundries must certify that they have prepared the required plans, have installed a direct flame contact gas-fired scrap preheater if applicable (or that they will comply by meeting the 20 ppmv emissions limit or by only preheating scrap that meets the scrap certification requirements), that they will meet each applicable work practice requirement, and that they have records documenting their certification.

Records are required to demonstrate continuous compliance with compliance certifications or to document conformance with their scrap inspection and selection plan. Foundries also must keep records of the chemical composition of all catalyst binder formulations applied in a furan warm box mold or core making line.

Operation and Maintenance Requirements

Foundries must certify in their notification of compliance status that they have prepared the O&M plan and that the plant will operate equipment according to the plan requirements. Records are required to demonstrate continuous compliance with other requirements in the O&M plan for capture systems, control devices, and bag leak detection system corrective actions. To demonstrate continuous compliance with the plan for mold vent ignition, foundries must make a compliance certification in each semiannual report that they have followed the procedures in their O&M plan.

E. What Are the Notification, Recordkeeping, and Reporting Requirements?

These requirements rely on the NESHAP General Provisions in 40 CFR part 63, subpart A. Table 1 to subpart EEEEE (the final rule) shows each of the requirements in the General Provisions (§§ 63.1 through 63.15) and whether they apply.

The major notifications include one-time notifications of applicability (due no later than 120 days of promulgation), performance tests (due at least 60 days before each test), performance evaluations, and compliance status. The notification of compliance status is required no later than 60 days after the compliance demonstration if a performance test is required or no later than 30 days after the compliance demonstration if no performance test is required.

Foundries are required to maintain records that are needed to document compliance, such as performance test results; copies of the startup, shutdown, and malfunction plan; O&M plan; scrap selection and inspection plan, and associated corrective action records; monitoring data; and inspection records. Records of annual usage, chemical composition, and HAP content are also required for chemical binders and coating materials. In most cases, records must be kept for 5 years, with records for the most recent 2 years kept onsite. However, the O&M plan; scrap selection and inspection plan; and startup, shutdown, and malfunction plan are to be kept onsite and available for inspection for the life of the affected source (or until the affected source is no longer subject to the rule requirements.)

All foundries must make semiannual compliance reports of any deviation from an emissions limitation (including an operating limit), work practice standard, or O&M requirement. If no deviation occurred and no monitoring systems were out of control, only a summary report is required. More detailed information is required in the report if a deviation did occur. An immediate report is required if actions taken during a startup, shutdown, or malfunction were not consistent with the startup, shutdown, and malfunction plan.

F. What Are the Compliance Deadlines?

Existing iron and steel foundries must comply with most requirements by April 23, 2007. The final rule requires existing foundries to comply with the work practice standards in § 63.7700(b) or (c), as applicable, by April 22, 2005. New or reconstructed iron and steel foundries that start up on or before April 22, 2004 must comply by April 22, 2004. New or reconstructed iron and steel foundries that start up after April 22, 2004 must comply upon initial startup.

III. Summary of Environmental, Energy, and Economic Impacts

A. What Are the Air Quality Impacts?

Most iron and steel foundries have had emissions controls in place for many years similar to those in the final rule. Overall, we expect the final rule to reduce HAP emissions by more than 820 tpy. The NESHAP will also reduce PM and VOC emissions by about 2,550 tpy. Implementation of scrap selection and inspection procedures is expected to reduce mercury emissions by 1.4 tpy—an 80 percent reduction from current levels.

B. What Are the Cost Impacts?

The total annualized cost of the final rule is estimated at $21 million, including costs for control equipment, compliance tests monitoring, recordkeeping, and reporting. This cost also includes the annualized cost of capital and the annual operating and maintenance costs for supplies, control equipment, monitoring devices, and recordkeeping media.

The nationwide total capital cost of the final rule is about $188 million. The capital costs associated with the final rule are primarily due to the costs of installing modular pulse-jet baghouse systems to control emissions of metal HAP and PM from cupolas currently controlled using venturi scrubbers. This capital cost is estimated at $175 million and includes the cost of removing the venturi scrubbers and installing modular pulse-jet baghouse systems. Based on information provided by the iron and steel foundry industry, we used a retrofit cost factor of 2.2 (i.e., the cost of installing a baghouse at an existing facility was estimated to be 2.2 times the cost of installing an identical baghouse at a new facility). This retrofit cost factor is considerably higher than the typical retrofit costs suggested by the literature (typical retrofit cost factors range from 1.2 to 1.5). As the cost of operating a baghouse is less than the cost of operating a PM wet scrubber due to lower energy consumption (lower pressure drop) of the baghouse system and the avoidance of wastewater treatment/disposal costs, the annual operating and maintenance cost of the final rule is actually estimated to be less than the cost of operating the current control equipment for cupolas. Therefore, there will be a net savings in the annual operating and maintenance costs for baghouses over venturi scrubbers of $6 million.

The cost impacts also include:

• The cost of installing and operating baghouses on currently uncontrolled electric induction metal melting furnaces;
• The cost of installing and operating baghouses on currently uncontrolled pouring stations;
• The cost of installing and operating wet acid scrubbers for currently uncontrolled TEA cold box mold and core making lines;
• The cost of installing and operating monitoring equipment (predominantly baghouse leak detection systems) for emissions sources; and
• The cost of electronic and paper recordkeeping media.

C. What Are the Economic Impacts?

We conducted a detailed assessment of the economic impacts associated with the final rule. The compliance costs are estimated to increase the price of iron and steel castings by 0.1 percent with domestic production declining by 8,400 tons in aggregate. The analysis also indicates no impact on the market price for foundry coke, which is used by cupolas in the production of iron castings. Foundry coke production is projected to decrease by less than 0.1 percent.

Through the market impacts described above, the final rule is predicted to have distributional impacts across producers and consumers of iron and steel castings. Consumers would incur $13.2 million of the overall regulatory burden of the final rule ($21.2 million) because of higher prices and forgone consumption. Domestic producers of iron and steel castings are expected to experience profit losses of $9.0 million due to compliance costs and lower output levels, while foreign producers may experience profit gains of $1 million associated with the higher prices. For more information, consult the economic impact analysis that is available in the docket.

D. What Are the Non-Air Health, Environmental, and Energy Impacts?

The final rule will generally provide positive secondary environmental and energy impacts. Replacing cupola wet scrubber control systems with baghouses will increase emissions of sulfur oxides by 370 tpy. However, due to the lower energy requirements for operating a baghouse versus a wet scrubber, which more than offset the energy requirements of the other new control equipment, the final rule is projected to result in a net reduction in annual energy consumption of 121,000 megawatt hours per year. This will lead to a reduction in emissions of nitrogen oxides and sulfur oxides from power plants of roughly 180 tpy and 370 tpy, respectively. Therefore, the final rule will have no net impact on emissions of sulfur oxides. There is uncertainty about the estimates of secondary emission reductions due to energy savings because we have not conducted a detailed analysis that identifies the fuel sources used at power plants from which the energy savings will be realized. Furthermore, the SO₂ emission reduction estimates may be overstated if the national cap on SO₂ emissions is binding. The replacement of wet scrubbers with baghouses is also responsible for the final rule's estimated 18.1 billion gallons per year reduction in water consumption and waste water disposal rates. Although baghouses have slightly higher dust collection efficiencies, the dust is collected in a dry form while PM collected using a wet scrubber contains significant water even after dewatering processes. Therefore, the total volume and weight of solids disposed under the final rule is estimated to be approximately the same as, if not less than, the current solid waste disposal rates.

IV. Summary of Major Comments and Responses

A. Why Did We Revise the Proposed Affected Source Designation?

Comment: Industry commenters felt the metal casting department should be separated into two separate affected sources: a melting department and a casting department. The commenters also suggested that we clarify that a foundry may contain multiple affected sources of a single type, such as more than one melting department, which may be operationally different and physically removed from each other. Some commenters felt that HAP emissions from melting are insignificant and suggested that this process either be excluded as an affected source or listed as a separate source category and then delisted.

Response: We considered splitting the metal casting department into a melting department and a casting processing department. This further classification of the affected sources might have been appropriate because the melting furnaces (melting department) are often separate from the pouring, cooling, and shakeout lines (casting processing department). However, most commenters requesting a change in the affected source or separate source categories thought that we could then either de-list melting departments or that the emissions from the melting department could be excluded from emissions limitations. Even if the melting department were a separate source category or affected source, these sources would still be co-located at major source facilities, and we would still be required to develop MACT standards for them. Furthermore, we do not consider emissions exceeding 100 tpy of metal HAP from melting furnaces to be de minimis as suggested by industry. Consequently, it is necessary and appropriate to establish MACT standards for these emissions sources.

A secondary rationale for requesting a change in the affected source was the fear of triggering new source MACT requirements. However, upon clarification that defining the melting department as a separate source would not eliminate the requirements to control melting furnace emissions, these commenters supported a broad definition of the affected source.

Therefore, in response to these comments, we have written the final rule to include a broader definition of the affected source (i.e., the iron and steel foundry). This broad definition eliminates a somewhat artificial separation of the mold and core making processes, which can often occur in close proximity, if not in conjunction with the casting (pouring) operations. This approach also avoids instances where an existing foundry might make minor equipment changes that might subject one process or a single piece of equipment subject to the new source emissions limits. This could occur if the affected source was defined as each “metal melting department” which could be delineated as each melting furnace at the foundry.

B. Why Did We Revise the Proposed Emissions Limits?

Metal Melting Furnaces

Comment: Most industry commenters opposed the proposed PM limit for melting furnaces and scrap preheaters, especially at a new affected source (i.e., the 0.001 gr/dscf). According to the commenters, the limit cannot be maintained on a continuous basis, will not be guaranteed by vendors, will result in high costs, will be subject to measurements errors, and stretches the capability of Method 5 (40 CFR part 60, appendix A). Several commenters stated that the emissions reductions that would be achieved did not warrant the costs associated with the PM limits. Five commenters stated that the MACT floor determination did not adequately account for inherent variability and operation under the worst foreseeable conditions. Another commenter stated that it was inappropriate to apply any variablity factor in establishing the MACT floor emissions limits. One commenter noted that a limit based on the 95th percentile of performance would suggest that the unit is out of compliance 5 percent of the time.

Several commenters stated that the EPA should not specify the control equipment in establishing the new source PM emissions limits, that the facility EPA used for new source MACT for cupolas was not representative, or that the more stringent limit was a disincentive to modernize plants. Two commenters noted that the vendor guarantee for the facility is 0.0016 gr/dscf (instead of 0.001 gr/dscf as reported by EPA) because the guarantee was 0.001 in grains per actual cubic feet. While two equipment vendors stated that they could not guarantee a long term performance of 0.001 gr/dscf, a representative for control device vendors stated that the 0.001 gr/dscf PM emissions limit for new sources is reasonable and appropriate and that a variety of fabric collector designs can achieve similar results. Most commenters recommended a limit of 0.005 gr/dscf or 0.0052 gr/dscf (which was proposed as the limit for certain new operations at integrated iron and steel plants). One commenter suggested a limit of 0.002 gr/dscf because baghouses achieving an average outlet PM concentration of 0.001 gr/dscf would be out of compliance with a limit of 0.001 gr/dscf about half the time.

Response: The CAA directs EPA to set limits that are at least as stringent as the MACT floor. For existing units, the MACT floor is the average emissions limitation achieved in practice by the best performing 12 percent of the existing units (for which we have emissions information). The MACT floor for new sources must not be less stringent than the emission control that is achieved in practice by the best-controlled similar source. Consequently, the comments related to vendor guarantees and high costs are not relevant in establishing the MACT floor for new and existing sources.

We disagree that the limit will result in significant measurement errors or that it stretches the capability of Method 5 (40 CFR part 60, appendix A). We require a minimum gas volume of 60 dry standard cubic feet (dscf) to ensure that sufficient PM is collected to evaluate the compliance of the emissions source with the PM emissions limits. In addition, the practical quantification limit for Method 5 is a filterable PM catch of 3 milligrams (mg), which is 0.0463 grains (gr). At the practical quantification limit of 3 mg of PM collected from 60 dscf of gas, the practical quantification limit of Method 5 as required in the rule is less than 0.0008 gr/dscf. If less than 3 mg of dust is collected during a test in which at least 60 dscf of gas are collected, we have reasonable assurance that the emissions source is in compliance with a 0.001 gr/dscf PM emissions limit. Without a minimum gas volume of 60 dscf, we could not confidently establish that an emissions source meets the 0.001 gr/dscf emissions limit.

As noted by the commenters, the emissions limits must be achieved at all times, and it is important that the MACT floor limit adequately account for the normal and unavoidable variability in the process and in the operation of the control device. The choice of selecting the 90th, 95th, or 99th percentile performance value depends largely on the adequacy of the data. As there were only 10 to 15 emissions tests for a given type of unit or source with which to assess the performance and variability of baghouse control systems, selecting a higher percentile range is appropriate to reflect additional uncertainty. In response to comments concerning the potential variability in process and control system performance and in recognition of the fact that the available emissions data are from a fairly limited number of short-term tests, we have re-evaluated the MACT floor determination using the 99th percentile of performance. This approach is designed to account for the different sources of variability, including variations in how the process is operated, changes in control device parameters, and variability associated with sampling and analysis.

By selecting the 99th percentile, we have sufficiently accounted for process operation, control device performance, and measurement variability. The 99th percentile is appropriate in this case because it accounts for the extreme end of the range of performance that could occur. Based on this re-evaluation of the MACT floor limits, we have adjusted the floor for cupolas at existing sources from 0.005 gr/dscf to 0.006 gr/dscf. We have adjusted the floor for cupola and electric arc furnaces at new sources from 0.001 gr/dscf to 0.002 gr/dscf. This new source limit of 0.002 gr/dscf is consistent with the vendor guarantee when corrected from actual to standard conditions (0.0016 gr/dscf).

We do not believe that setting a limit at the 95th or 99th percentile means that the emissions source will be out of compliance 5 percent or 1 percent of the time. Through proper operation and maintenance of the control device and process equipment, the owner or operator can avoid periods of poor performance. As such, a properly operated and maintained control device applied to normal process operations should not experience performance levels that exceed the limit. In the rare event of an unavoidable failure such as a malfunction, the owner or operator can continue to demonstrate compliance by following the procedures in the startup, shutdown, and malfunction plan. If the limit is exceed as a result of variability that can and should be controlled (i.e., a preventable event), then the event is a deviation.

We understand industry concerns over the representativeness of the test data for one of the foundries that was mentioned. Fortunately, emissions test data are available for an equivalent control system that does not control an additional process which might dilute the emissions. The performance level for this system is also a PM emissions limit of 0.002 gr/dscf. Consequently, the limit for new sources is not dependent only on the source test data from the one facility cited by the commenters.

Unlike cupolas and electric arc furnaces, the furnace control system that represents MACT for electric induction furnaces at new sources is a traditional baghouse, followed by a cartridge filter, followed by a high energy particulate air filter. The limit for this system is still 0.001 gr/dscf when evaluated at the 99th percentile. Therefore, the data clearly support that MACT for electric induction furnaces at new sources is 0.001 gr/dscf.

In the final rule, we have established emissions limits for the emissions sources and do not require a specific type of control device. Foundry owners or operators may use any control measure that will meet the applicable standard. In trying to understand the differences in the performance achieved by certain systems, we evaluated and compared baghouse design, cleaning mechanism, flow rate, temperature, fabric material, and air-to-cloth ratio for each system as operated during the emissions source test. Certainly a number of these factors influence the performance of a fabric filter control system. In evaluating the performance of the cupola control systems, the horizontally-designed baghouses exhibited the best performance of the systems tested. The description regarding these systems was provided primarily to document why the low outlet PM concentrations observed were real and not the result of an unknown source testing error. We do not endorse any specific baghouse design.

Because the affected sources will be required to comply with the emissions limits at all times, the limits established must account for the normal and unavoidable variability inherent in the process and in the control device operation. The emissions rate for a given emissions source does vary over time, as is demonstrated by the variability seen between individual test runs and repeat tests. As such, the MACT floor should not be developed based on the stack test data without accounting for variability. For each facility for which we have stack test emissions data, we have estimated the emissions limitation that the facility can achieve on a continuous basis by applying statistics to the available emissions data to estimate the emissions rate that facility can achieve at least 99 percent of the time.

In summary, we have established emissions limits for both new and existing emissions sources and have not specified the type of control system that must be used. For cupolas and electric arc furnaces, MACT for new sources is 0.002 gr/dscf, reflecting the 99th percentile level of performance of the median unit in the top 12 percent of best-performing units. The MACT floor for cupolas at existing foundries is 0.006 gr/dscf, reflecting the 99th percentile level of control of the median unit in the top 12 percent of best-performing units. These limits reflect our conclusion that the proposed 0.001 gr/dscf limit for cupolas and electric arc furnaces at new foundries and the 0.005 gr/dscf limit for cupolas at existing foundries did not adequately account for the variability expected in the actual performance of the units that were used to establish the MACT floor for these emissions sources. The 0.001 gr/dscf limit for electric induction furnaces and the 0.002 gr/dscf emissions limit for cupolas and electric arc furnaces at new foundries represent emissions limits that the best-performing sources can and do meet under the most adverse circumstances which can reasonably be expected to recur.

Comment: Three commenters recommended that the final rule include emissions limits for individual metal HAP. The commenters suggested that PM is not a good surrogate for lead (which is a semi-volatile metal) or mercury (which typically has low collection efficiencies in baghouses) and does not consider the hazard of the emitted pollutants. In addition, the metal HAP in the PM from some emissions sources comprise only a small portion of emissions from the emissions source or the overall foundry and has not been characterized for other emissions sources.

Response: As described in our MACT floor documentation, metal HAP emissions reductions tracked well with PM emissions reductions for the cupola control systems we tested. Reductions in lead emissions also tracked well with PM emissions reductions. Mercury emissions were a small component of the total metal HAP emissions, but both control systems tested by EPA were ineffective in reducing mercury emissions. Therefore, we do not consider these add-on control devices to be control technologies for the purpose of reducing mercury emissions. The only effective method for reducing mercury emissions at iron and steel foundries is scrap metal selection and inspection to prevent mercury contamination of the furnace charge. For all other metal HAP emissions from metal melting furnaces, the test data show that effective PM emissions control also provides effective metal HAP emissions control. In addition, PM is a reasonable surrogate for metal HAP emissions control effectiveness because MACT is a technology-based standard, and the technologies currently used by foundries that reduce metal HAP emissions are those specifically designed to control PM. Additionally, it is clear from our data the greater the PM reductions are for a specific unit, the greater are the HAP reductions. Thus, we have concluded that it is appropriate to use PM as a surrogate for HAP metals because the unit that achieves the greatest level of control of PM will also achieve the greatest level of control of metal HAP. As discussed in the following response, we have also developed an alternative limit for total metal HAP. Finally, to the extent that it is feasible to reduce metal HAP emissions by means other than operation of emission control devices, we are requiring such measures. Specifically, we are requiring a scrap selection and inspection program to reduce lead and mercury emissions. These requirements combined with the PM limits accurately reflect the MACT level of control.

Comment: Two commenters oppose the use of PM as a surrogate because some foundries melt only high quality steel with very low tramp metal content in the induction furnaces rather than scrap iron. Consequently, their uncontrolled melting furnaces may have lower HAP emissions than those from a baghouse on a furnace melting scrap with higher levels of tramp metals. We also received comments that some foundry operations, such as dry scrubbing for sulfur dioxide control, may result in disproportionately high PM emissions without correspondingly high metal HAP emissions.

Response: As discussed in our previous response, PM is a good surrogate for HAP metals other than mercury. However, we recognize that the metal HAP content of the PM can vary significantly depending on the type of metal cast. Some foundries may have very low metal HAP emissions due to very low HAP content in their casting. We also recognize that it is infeasible for all foundries to use scrap with very low HAP metal content because of the limited supply of such scrap and because various levels of certain elements are needed in certain grades and types of iron and steel casting. Also, when foundries use scrubbing techniques for reducing sulfur dioxide emissions, they may have unusually high PM emissions without correspondingly high HAP emissions. Therefore, while PM is a good surrogate with which to judge the performance of a control system to reduce metal HAP emissions, we realize that it is only a surrogate and not a direct measure of HAP emissions, and that in some cases the PM limit may have unwarranted consequences. For the above reasons, we are establishing alternative total metal HAP emissions limits that are equivalent to the PM limits. The alternative metal HAP limits are based on, and are dependent on the MACT limits for PM.

Having identified the appropriate level of control based on PM performance, we re-examined our data on metal HAP emissions and evaluated the metal HAP emissions as a percent of the PM emissions. We evaluated metal HAP emissions to project the range of metal HAP emissions as a percent of PM associated with the performance of the type of control system used by the unit identified as the MACT floor emissions unit. That is, by normalizing the HAP emissions data by the PM emissions and aggregating these data for the various emissions sources being regulated, we calculated a reasonable estimate of the magnitude and variability of the HAP content as a percent of PM for these sources. By applying this information to the specific system that established the MACT floor PM emissions limits for each source type, we developed a total metal HAP emissions limit for each source type that is based on the performance of the MACT floor unit. Each total metal HAP limit is equivalent to the corresponding MACT floor PM emissions limit. We used this calculation to develop alternative limits for total metal HAP for melting furnaces and pouring operations.

The basis of this alternative emissions limit is the MACT floor determination for PM emissions. Because we lack sufficient test data for metal HAP, we could not otherwise derive a metal HAP emissions limit without first identifying the MACT floor unit on the basis of its PM emissions performance. Therefore, we concluded that this total metal HAP emissions limit is an alternative to the PM emissions limit, and not an additional MACT floor requirement.

We developed a distribution of the PM emissions for each emissions source based on the actual performance of the unit identified as the 6th percentile unit and the same 0.4 relative standard deviation used to determine the MACT floor performance limits. A separate distribution based on the available metal HAP emissions data was developed to characterize the total metal HAP content of the emitted PM. Using Monte Carlo techniques, 5,000 randomizations were generated for each of these distributions and the projected metal HAP emissions were calculated for each of the 5,000 randomizations. This is a common statistical approach for establishing a distribution for a parameter that is dependent on multiple, variable parameters.

As with the MACT floor determination of PM emissions performance, we selected the 99th percentile metal HAP concentrations determined from these distributions. These metal HAP emissions limits were equivalent to approximately 8 percent of the 99th percentile PM emissions limit (i.e., the MACT floor PM emissions limit) for each of the emissions sources. That is, this analysis indicated that the total metal HAP emissions limit that is equivalent to the MACT floor PM emissions limit can be calculated by multiplying the PM emissions limit by 0.08 (i.e., assuming the PM is 8 percent metal HAP). The final metal HAP emissions limits were rounded to one significant digit in keeping with the relative accuracy of the assessment.

As the identification of the unit that represents the MACT floor is solely dependent on the PM emissions performance, these metal HAP emissions limits do not represent a separate MACT floor that must be met at all emissions sources, but rather an alternative emissions limit that is equivalent to the MACT floor PM emissions limit. The alternative metal HAP emissions limits provide foundry operators with more flexibility in meeting the metal HAP emissions limits (for example, by adopting a scrap program that is more stringent than the MACT requirement, in conjunction with PM emissions controls to further reduce metal HAP emissions). This alternative also avoids, in some cases, the need for replacing well-performing venturi wet scrubbers with high efficiency baghouses to achieve a required PM emissions reduction when other measures might be used to achieve the desired metal HAP emissions reduction. The alternative also accommodates facilities that may have disproportionate PM emissions but low HAP emissions, as in the case for dry scrubbers used to control sulfur dioxide.

Comment: More than twenty industry commenters opposed the proposed carbon monoxide (CO) emissions limit for cupolas (200 ppmv). Several of these commenters stated that CO data from CEMS and CO monitors show that the limit cannot be achieved. They explained that the cupola operation is a dynamic process that is affected by changes in the melt rate and iron chemistry, which requires the CO combustor to adjust and seek a new equilibrium; CO concentrations are highly variable even in the best afterburner systems. The material being melted, coke sources, and seasonal adjustments also affect CO emissions. One vendor stated that his company could not guarantee equipment that can meet the 200 ppmv CO emissions limit. The commenters also suggested that the CO limit is based on the Illinois emissions standard, which was found to be improperly derived and never enforced.

Five commenters stated that EPA failed to provide sufficient data that maintaining a CO concentration of 200 ppmv is an effective surrogate for organic HAP destruction, while two commenters supported the use of CO as a surrogate for HAP. One commenter asked why VOC was not used as the surrogate for organic HAP emissions from the cupolas.

Response: The proposed CO emissions limit was based upon the emissions source test data for CO emissions from cupolas; it was not based upon the Illinois CO emissions limit. Two of the CO emissions tests used to develop the 200 ppm CO emissions limit were from foundries located in New Jersey, where CO CEMS are required. Therefore, EPA requested CO CEMS emissions data from these foundries to verify the performance of these systems and to better understand the variability associated with the process. Data were received from one of these foundries which supported the assertion that the 200 ppmv limit did not adequately accommodate the variability in the process operations and control device performance. Additionally, emissions test data were also received from a cupola-afterburner system that measured CO and VOC (minus methane) emissions concurrently. For the individual runs of this test, the average outlet CO concentrations were 701, 1470, and 849 ppmv, while the average VOC emissions were 3.4, 4.2 and 5.1 ppmv as propane. This limited data supports the industry commenters' assertion that organic HAP emissions (as indicated by VOC emissions) are not well correlated, although there is a limited range of CO and VOC emissions considered in this single emissions test.

As discussed in the preamble to the proposed rule, CO is an indicator of good (complete) combustion, but, at some lower level of CO, further reductions in CO concentrations do not necessarily result in further reductions of organic HAP. That is, we recognize that CO is not a perfect surrogate for organic HAP emissions from the best-performing units, but it is a surrogate for which emissions data were available and one that provides a reasonable indication of adequate combustion characteristics. However, based on the comments and the additional data received, we agree that we do not have sufficient data to support the establishment of a specific CO concentration limit as a surrogate for the organic HAP emissions performance of a cupola afterburner system.

We reviewed the submitted data and other data in the docket for VOC and organic HAP for the best-controlled cupolas (those using afterburners). These data are too limited to identify the level of performance of the best-performing units or to establish a specific organic HAP or VOC emissions limit. Therefore, we rely on our experience with the performance of thermal destruction systems such as these afterburners. This experience clearly indicates that these units should be able to meet a 98 percent destruction efficiency or an outlet concentration of 20 ppmv (as the chemical emitted), whichever is less stringent. However, due to safety issues associated with typical equipment configurations, sampling between the cupola chamber and the afterburner is impracticable and unsafe. Therefore, we provide only the 20 ppmv exhaust concentration alternative. The limited available data on organic HAP emissions from cupola afterburners suggest that the 20 ppmv emissions limit is achievable and reflects the level of performance of the best controlled units, and that the 98 percent reduction alternative is not needed for this application.

Furthermore, we establish this emissions limit as the sum of all volatile organic HAP (or VOHAP) emitted, thereby eliminating the need to select a surrogate. However to provide flexibility in conducting the performance tests, we are providing compliance alternatives to allow for demonstration of compliance using test methods to measure TGNMO or TOC concentrations (in ppmv as hexane). These test method alternatives will measure both HAP and non-HAP compounds, and will, therefore, ensure that a unit is meeting an emissions level as stringent or more stringent than the VOHAP emissions limit. However, these test methods are cheaper and easier to perform, and therefore, these options may be desirable for some sources. Hexane was selected for the concentration equivalency because the primary HAP expected to be emitted are C6 hydrocarbons or higher (e.g., benzene, toluene, and xylenes).

Comment: While one commenter supported the proposed rule requirement for direct measurement of CO emissions from cupolas using a CEMS, many industry commenters were opposed. They argued that the final rule should include an operating limit for the afterburner temperature measured by a CPMS. According to the commenters, a CO CEMS is not technically feasible or reliable because of the harsh conditions of the gas stream, and it is costly while achieving minimal benefit.

Response: We have deleted the requirement for a CO CEMS from the final rule because the CO limit has been replaced by a limit for VOHAP emissions. The autoignition temperature of the organic HAP present in the cupola exhaust stream (primarily benzene, toluene, and xylenes) is lower than the autoignition temperature of CO, which is 1,300 °F. Therefore, an adequately designed afterburner operating at a minimum of 1,300 °F will effectively ensure combustion of the organic HAP. Once a performance test indicates that the cupola afterburner is sufficiently engineered (in terms of excess air flow, residence time and mixing) to achieve the required VOHAP emissions limit, then continuous monitoring of combustion zone temperature will provide adequate assurance of continuous compliance. Therefore, we require foundry operators to install and operate a CPMS for combustion zone temperature, and we require that the 15-minute average combustion zone temperature must not fall below 1,300 °F. Periods when the cupola is off blast and for 15 minutes after going on blast from an off blast condition are not included in the 15-minute average.

Comment: Several industry commenters objected to the proposed VOC emissions limit for scrap preheaters (20 ppmv as propane or 98 percent reduction). The commenters contended that the VOC limit based on afterburning technology does not meet the requirements for determining the MACT floor because only 4 or 5 of 169 preheaters nationwide (3 percent) currently use afterburners. The commenters stated that there is no basis for the proposed limit, there are no data indicating the presence of organic HAP in preheater emissions, and improvements in direct flame preheaters have made the afterburners an outdated technology. Commenters also stated the existing units cannot achieve 20 ppmv because of process variability and the likely presence of uncombusted methane from the preheater, which can contribute significantly to the VOC concentration, especially when measured as propane.

Response: Based on the information available at the time the proposed rule was developed, it appeared that more than 6 percent of the scrap preheaters were controlled by afterburners. However, we have confirmed that, as the commenters suggested, one foundry that had reported using afterburners had subsequently upgraded their material handling system and installed direct flame preheater systems. With this change, the median of the top 12 percent of units is no longer a unit using an afterburner, but a unit using a direct flame preheater.

There are two basic types of preheater designs: direct flame contact preheaters and hot gas flow preheaters. Direct flame contact preheaters primarily use gas-fired burners where the flame impinges on the scrap. The primary heating mechanism for direct flame contact preheaters is the burner flames contacting the scrap. Hot gas flow preheaters may use gas-fired burners or electricity to heat air and the hot air (and combustion gases from the burner, if applicable) is used to preheat the scrap. In hot gas flow preheaters, the scrap is not heated by direct contact with a high temperature flame. Preheaters are used primarily to remove water and organic contaminants that could cause explosions or other hazards when the scrap is melted in induction furnaces. Although both types of preheaters are effective for this purpose, the different preheater designs have different HAP emissions potentials.

For preheaters generally, we require a scrap selection and inspection program to limit, to the extent practicable, the amount of organic HAP precursors (i.e., oils and other organic liquids) entering a scrap preheater, and we are establishing a work practice standard to require either preheaters with direct flame contact or application of an afterburner. Because the scrap selection and inspection program cannot completely exclude the potential presence of tramp organic materials, scrap preheaters are a potential source of organic HAP emissions. Furthermore, we could not identify specific scrap selection and inspection programs for these types of scrap materials that would be more effective than those proposed. Therefore, the primary variable affecting the organic HAP emissions from scrap preheaters is the preheater design. Additionally, it is not feasible to capture and convey emissions from all preheaters at existing foundries because of certain design and operational constraints, such as preheaters with moving grates, interferences with overhead moving cranes, and lack of space. However, preheaters at new foundries can be designed to capture and convey emissions prior to construction.

Based on an engineering assessment of the scrap preheater designs and control systems, units that operate with an external combustion system (afterburner) are expected to be the best performing for organic HAP emissions. The next most effective control is the use of direct flame contact preheaters, which have lower organic HAP emissions than hot gas flow (indirect heating) preheaters because organic contaminants in the scrap are thermally destroyed by direct contact with the preheater flame. We ranked scrap preheater systems according to their projected organic HAP destruction efficiency based on the heating methods that are used. From this analysis, we identified the MACT floor unit as one that uses natural gas, direct flame, scrap preheating (used at well over 12 percent of existing sources). The direct flame contact provides efficient destruction of organic HAP, and organic HAP control is improved when combined with the requirements of the scrap selection and inspection program. Moreover, many of the direct flame contact preheaters use an open burner design where the burners are directed onto the scrap, even when the preheater uses a moving grate system where it is not feasible to collect the emissions through a conveyance. Therefore, we believe a work practice standard is appropriate, and we are requiring foundry owners and operators to use direct flame contact preheaters. However, we are allowing foundries to use a properly designed and operated afterburner as a compliance option for the preheater MACT standard because an afterburner on either a direct flame or indirect flame preheater will result in better control of organic emissions than the use of direct flame preheating alone. This option is reflected by an alternative standard of 20 ppmv VOHAP. Furthermore, we also conclude that afterburners are not a cost-effective “beyond-the-floor” technology for existing preheaters based both on the costs associated with redesigning the burner configuration to allow capture and control of the emissions and the small amount of additional emissions reductions achieved by the additional afterburner control.

The MACT floor for scrap preheaters at new sources, however, is still based on an afterburner control system. As discussed when considering the performance limits for cupola afterburners, we believe that a 20 ppmv emissions limits is still appropriate, but that the 20 ppmv limit should be based on specific VOHAP and should not necessarily include uncombusted methane emissions.

We have acknowledged that all foundries cannot completely eliminate organic contaminants from their scrap. However, some foundries use only scrap that can be certified to be free of the organic contaminants. In the final rule, we distinguish two general grades of scrap in the scrap selection and inspection program. Under a certification program, foundries can certify that they use only certified-metal ingots, pig iron and similar material that do not contain organic contaminants. Foundries that use scrap without organic contaminants will not generate organic HAP emissions from their scrap, regardless of the type of preheater used. Most foundries that use this type of material are small production foundries, and most of these are not major sources of HAP emissions. However, this may be a potentially viable alternative for some major source foundries as well. Therefore, we provide a compliance option for scrap preheaters that charge only clean scrap as described by the certification alternative in the scrap selection and inspection program. The compliance option for scrap preheaters that charge clean scrap at new and existing iron and steel foundries is the work practice of charging only material that has been certified to comply with the scrap certification alternative in the scrap selection and inspection program.

In summary, based on comments received and changes in the control configurations used at the top 12 percent of scrap preheaters, we revised the organic HAP MACT floor for scrap preheaters. The MACT floor for scrap preheaters at existing sources is the work practice of using a gas-fired preheater in which the gas flame directly contacts the scrap. Alternatively, scrap preheaters at existing sources can meet a 20 ppmv VOHAP emissions limit (with alternatives of measuring TGNMO or TOC as hexane as a surrogate for VOHAP). MACT for scrap preheaters at new iron and steel foundries is the 20 ppmv VOHAP emissions limit. Also, we provide an alternative compliance option for preheaters at new and existing foundries that charge only clean scrap as described in the certification alternative of the scrap selection and inspection program. In this case, owners or operators need only certify that their preheater heats only scrap as described in the scrap certification alternative.

Comment: Several commenters opposed the requirement for direct measurement of VOC emissions from scrap preheaters and pouring, cooling, and shakeout (PCS) lines. The commenters believed that CEMS are not practical for scrap preheaters or justifiable (technically or economically) for PCS lines. Some commenters noted that VOC measurements for scrap preheaters and PCS lines would be more accurate with calibration by xylene or toluene rather than propane. One commenter explained that most HAP emitted from foundries have six carbons or more. Therefore, the VOC measurement should be calibrated with toluene or xylene as these would provide a better measure of VOC emissions than propane.

Response: The point concerning the representativeness of propane to characterize the HAP emissions is well-taken. Even though a wide variety of HAP are expected to be emitted from these sources, an analysis of the available VOHAP emissions data indicate that the average carbon number for the VOC emitted from these operations is six. Additionally, the historical documents where EPA has established the 20 ppm VOC emissions limit indicates that it was established by compound exit concentration rather than by a specified indicator of VOC, such as propane. Therefore, based on the available data and a review of the basis for VOC measurements, we have adjusted the organic HAP emissions limits to either measure VOHAP concentrations directly or to measure TOC using hexane as the calibration gas (i.e., measure VOC outlet concentrations as hexane or C6 equivalents) as a surrogate for VOHAP. These organic HAP emissions limits now apply to cupolas (at new and existing foundries), scrap preheaters (at new foundries and as an alternative at existing foundries), and automated conveyor and pallet cooling lines and automated shakeout lines that use sand mold systems (at new foundries).

Although a VOC CEMS is technically feasible for these applications, especially for new foundries, a review of the relative costs associated with these monitoring requirements compared to the control equipment costs to achieve the emissions limits does not appear to justify the requirement to install and operate VOC CEMS for cupola afterburners or scrap preheaters. Furthermore, for cupolas and scrap preheaters which use thermal destruction, the combustion zone (or flame) temperature provides an excellent indicator of on-going control device performance. Therefore, alternative continuous parameter monitoring requirements for these emissions sources can be used that will ensure continuous compliance with the emissions limit without undue additional costs. No alternative continuous parameter monitoring requirement could be identified for the cooling and shakeout operations. As the organic HAP emissions limits only apply to automated conveyor and pallet cooling lines and automated shakeout lines that use a sand mold system at a new iron and steel foundry, we maintained the VOC CEMS requirement for these emissions sources. We provide options to either meet the 20 ppmv VOHAP limit directly using the VOC CEMS (measuring total hydrocarbons as hexane) or to develop an equivalent site-specific VOC CEMS emissions limit based on the results of the VOHAP emissions measured during the performance test. The VOC CEMS actually measures total hydrocarbons, which includes non-HAP compounds. As a result, using a VOC CEMS to directly measure total hydrocarbons may be more stringent than the site-specific VOC limit correlated to measured VOHAP emissions.

We also included procedures in the final rule that will allow other monitoring methods to demonstrate compliance with the VOHAP emissions limit. For example, if you use a carbon adsorption system to control organic HAP emissions, appropriate monitoring parameters may include carbon breakthrough by replacing the carbon at specified frequencies. Other compliance methods, such a pollution prevention (P2) techniques, also may be used to meet the VOHAP emissions limit. If you use P2 techniques, appropriate monitoring methods may include measuring loss on ignition or recording the type of binder formulation used, total chemical usage rate, and/or chemical usage rate per volume of sand. If through P2 measures you can eliminate all HAP emissions from the emissions source or you can demonstrate continued HAP emissions reductions equal to or better than the MACT level of control, you may be eligible for a P2 compliance alternative under amendments to the NESHAP General Provisions (40 CFR part 63, subpart A). These amendments were proposed on May 15, 2003 (68 FR 26249).

The procedures in the final rule require that you submit a monitoring plan that includes a description of the control technique (or P2 measures), a description of the continuous monitoring system or method (including appropriate operating parameters to be monitored), test results demonstrating compliance with the emissions limit, operating limit(s) if applicable determined according to the test results, and the frequency of measuring and recording to establish continuous compliance. If applicable, you also must include operation and maintenance requirements for the monitor(s).

Pouring, Cooling, and Shakeout

Comment: Several commenters requested that we clarify the applicability of the emissions limits with regard to “pouring areas” and “shakeout.” In general, large area casting producers requested that we remove reference in the definition of “pouring area” to maintaining the molds in a stationary position through cooling. One commenter requested that the definition for “shakeout” be revised to indicate that it is a mechanical operation, typically automated, and does not include manual operations that dismantle or separate castings from molds as seen in pouring areas. The change is needed because otherwise such manual operations may be subject to the requirements for new lines; however, it is infeasible to capture and control these operations, especially when they involve large castings in a pouring area.

Other commenters pointed out that centrifugal and permanent molds have very low organic content compared to sand molds. The commenters recommended that these systems be subcategorized and stated that the MACT floor for pouring, cooling, and shakeout for these operations at new sources would be no control.

Response: We agree with some of the commenters suggestions for clarifying definitions. We examined the data and found that no cooling lines associated with floor or pit molding operations are currently controlled for organic HAP emissions. Of the three cooling lines that have end-of-pipe controls, two are automated conveyor lines and one is a pallet line. One of the foundries that has a carbon adsorption unit performs both pallet and floor molding; however, only the pallet cooling line is controlled.

Based on this information and in response to comments, we removed the proposed rule definition of “pouring, cooling, and shakeout line” and adjusted the proposed rule definition of “pouring area” to clarify that it includes floor and pit molding processes. In addition, the molds in a pouring area do not have to remain stationary for the duration of mold cooling. We also adjusted the proposed definition of “pouring station” to clarify that it means the fixed location to which molds are brought by an automated conveyor or pallet molding line. We added a definition for “automated conveyor and pallet cooling line” (i.e., cooling lines associated with pouring stations) and “floor and pit cooling operation” (i.e., a cooling operation associated with a pouring area). We also removed the proposed rule definition of “shakeout” and added a definition for “automated shakeout line” that distinguishes automated shakeout operations from manual knockout operations. The purpose of these revisions is to clarify that the 20 ppmv VOHAP limit for a new iron and steel foundry applies only to automated conveyor and pallet cooling lines and to automated shakeout lines.

As discussed in the BID for the final standards, permanent and centrifugal molds have significantly lower organic HAP emissions than green sand molds. Our re-evaluation of new source MACT for organic HAP demonstrates the need for a subcategorization of permanent and centrifugal molds for cooling and shakeout. For this reason, we also adjusted the VOHAP limit for new foundries to apply only to lines (automated conveyor and pallet cooling lines and automated shakeout lines) that use a sand mold system.

Capture Systems

Comment: Several commenters stated that the requirement of a minimum face velocity of 200 feet per minute (ft/min) has no underlying MACT floor basis and that it does not account for variability. Numerous commenters stated that a blanket requirement of 200 ft/min is not universally applicable and it is not consistent with good engineering design. Other commenters stated that the capture requirements creates a safety hazard, increases energy requirements (for building heating and air conditioning), and creates defects in the castings (especially during pouring).

Several commenters noted that indoor air quality is regulated by other agencies and stated that when a process is operated in a manner that limits worker exposure (e.g., so as to comply with standards established by the Occupational Safety and Health Administration), then there is no basis for requiring stricter capture and ventilation standards. Another commenter noted that adjustments to individual fans for workers, which were installed for worker comfort, can change air flow in the surrounding area and impact face velocity, making it difficult to maintain compliance with the standard. Consequently, the requirement to maintain a minimum of 200 ft/min face velocity would require much higher design and operating face velocities in order to ensure continuous compliance, increasing energy consumption with no demonstrable environmental benefit.

A few commenters stated that it was technically infeasible to install close capture hoods on their induction furnaces, pouring stations, or pouring areas due to process configurations and accessibility limitations. The only option would be to evacuate the entire building at huge costs and energy requirements for very limited HAP emissions reduction.

One commenter noted that their foundry has reduced VOC and HAP emissions by judicious reductions in capture and collection, and that the prescriptive ventilation requirement would reduce operator flexibility and may increase HAP emissions. Another commenter noted that they had received a patent for controllers that limit air ventilation at times of lower emissions, which saves heating and energy costs without impairing air quality.

Most of the commenters recommended that the final rule require that existing capture systems be operated consistent with good engineering practices and consistent with the facility's operation and maintenance plan. Two commenters recommended requiring a best engineering design based on the “Industrial Ventilation Manual of Recommended Practice.”

Response: Due to the comments received regarding the capture system requirements, we have decided to eliminate the 200 ft/min capture velocity requirement. In the final rule, we require that capture systems be designed and operated according to accepted engineering practices, such as the “Industrial Ventilation Manual of Recommended Practice.” Periodic inspection, maintenance, and continuous parametric monitoring are required to ensure they are properly operated and maintained on a continuing basis.

Additionally, we agree that there are process configurations and designs for which capture is infeasible, impractical, and ineffective. For example, capture systems at some iron and steel foundries would interfere with the movement of overhead cranes used to move large molds. Some pouring areas cover several thousand square feet, which makes capture impractical because of the enormous evacuation rate that would be needed. Physical constraints and space limitations, such inadequate clearance between equipment and structural columns, also pose problems for installing capture systems. For operations that cannot feasibly be captured, the emissions from the operation are released into the interior of foundry buildings and may be emitted as fugitive emissions through roof vents, doors, and other openings. We specifically require control of such fugitive emissions as described above.

Opacity Limit

Comment: Several commenters recommended that fugitive emissions from miscellaneous sources not be included because the control of these emissions would be costly and will not contribute to a significant reduction in HAP emissions. These commenters do not believe an opacity limit for fugitive emissions is necessary or appropriate. One commenter noted that an opacity limit of 5 percent would be beyond the MACT floor. The commenter stated that they have two plants regulated under a single permit that included a 5 percent opacity limit as a condition to proposed modifications. Modifications have been completed to one of the plants to meet this limit and modifications are planned at the other plant (at an investment of $3 to $11 million) to enable them to meet the permit limit by December 2004.

On the other hand, two commenters stated that EPA needs to set a limit for fugitive emissions and also develop work practices to control fugitive emissions. One of the commenters submitted a summary of dust analysis results surrounding a steel foundry indicated elevated levels of several HAP, including chromium (total), lead, manganese, and nickel, near the foundry. The commenter suggested that these elevated metal HAP emissions are due largely to uncontrolled fugitive emissions from the foundry.

Response: The CAA directs EPA to establish standards under section 112(d) to reduce emissions of HAP from stationary sources, and expressly includes fugitive emissions. Our data indicate that there are significant sources of fugitive HAP emissions at iron and steel foundries. Fugitive HAP emissions from iron and steel foundries include un-captured metal fumes from metal melting and pouring operations. The available emissions data clearly demonstrates that metal fumes from these sources contain metal HAP including manganese, lead, and other heavy metals. Additionally, commenters have submitted data regarding the elevated HAP content in dust surrounding one foundry, and suggested that fugitive emissions may have contributed to these high HAP concentrations. In general, it is clear that fugitive emissions contribute to the overall HAP emissions from foundry operations. Moreover, such fugitive emissions are often subject to emission limitations.

Our evaluation indicates that these fugitive emissions have been effectively regulated by establishing opacity limits. We examined State regulations for fugitive emissions and found that almost all States apply an opacity limit for the buildings that house the process equipment. We ranked the regulations and chose the most stringent (Michigan's limit of 20 percent with one exception per hour up to 27 percent) because at least 6 percent of the foundries are subject to this limit. This opacity limit represents the MACT floor for existing sources and is the primary standard for fugitive emissions.

This opacity limit is indicative of the achievable performance of these foundries under the most adverse circumstances that can reasonably be expected to recur. Based on observations of visual emissions at a number of iron and steel foundries, this opacity limit can be achieved at well controlled foundries. Furthermore, we know of no facility that is currently subject to, and able to meet, a more stringent opacity limit. One commenter appears to be in the process of trying to meet a 5 percent opacity, but the overall regulated facility (which consists of two plants) has yet to be able to meet this limit, and as such, we do not consider the 5 percent opacity limit achieved. Therefore, we conclude that the MACT floor for fugitive emissions from new sources is the same as for existing sources (20 percent opacity except for one 6-minute average per hour not to exceed 27 percent) because this is the emissions limit required of the best performing facility, and we believe this emissions limit is indicative of the actual emissions limitations achieved by these facilities under the most adverse circumstances that can reasonably be expected to recur. The opacity limit applies specifically to fugitive emissions from the foundry buildings, and fugitive emissions are defined as all releases to the atmosphere that are not discharged through a conveyance.

Mold and Core Making

Comment: Several industry representatives commented that the scrubbers evaluated for MACT appeared to be operating with fresh acid solution with a pH below 2. However, contractors who recycle used TEA will not accept material with a pH less than 2. One commenter felt that recyclers would not accept the scrubber solutions because of the low pH that would result from the 1 ppmv emissions limit. Commenters also questioned the technical validity of the 1 ppmv emissions limit, especially for systems with high inlet TEA concentrations. The commenters recommended that we adjust the proposed operating limit for wet acid scrubbers to require operating within manufacturer's specifications, maintaining the pH at 4.5 or less, and assess performance in terms of percent removal as specified by the manufacturer.

Response: The commenters' point regarding the test data being representative of TEA scrubber performance with fresh acid solution is well-founded. All of the available TEA scrubber performance data was generated from tests that used fresh acid solution (pH of 2 or less). Discussions with control equipment vendors indicate that the scrubbers are designed to operate at a scrubbing solution pH of 4.5 or lower. Discussions with foundry operators, as well as the public comments received, indicate that these foundries replace the scrubbing solution when the pH reaches either 4.5 or 5, depending on the foundry. As recycling of the TEA in the scrubbing solution is environmentally beneficial, we do not want to preclude the recycling of TEA by establishing a very low pH operating limit during the performance test. Also, because the performance limits were derived from test data of systems with fresh acid solution, it is not necessarily appropriate to require foundries to meet an emissions limit with spent acid solution (i.e., a pH nearing 4.5) when the data used to establish the performance limit of the scrubbers were all based on performance with fresh acid solution (i.e., a pH of 2 or less). From the information collected regarding the operation of these systems, at least 12 percent of the units replace the scrubbing solution at a pH of 4.5 or less (rather than at a pH of 5 or less). No units were identified that replaced the scrubbing solution at a pH of 4.0 or less. Therefore, replacing the scrubber solution at a pH of 4.5 or less is representative of MACT floor operating conditions for these scrubbing systems at new and existing iron and steel foundries.

The data used to establish the performance of the wet scrubber systems were also limited in that we have no data for systems with inlet TEA concentrations greater than 250 ppmv. Based on comments received from both foundry and TEA scrubber vendor representatives, the TEA systems are designed to achieve a percent removal of TEA and that the 1 ppmv limit is not achievable for systems with inlet TEA concentrations in the 1,000 ppmv range or higher. We believe that these are valid concerns and that a percent reduction alternative is warranted for systems with high TEA concentrations. After reviewing the source test data and the operating parameters associated with the TEA scrubber at the best-performing sources, we concluded that the MACT floor performance of the TEA scrubbers is correctly defined as a 99 percent or more TEA removal efficiency or an outlet TEA concentration of 1 ppmv or less, as determined when the system is operated with fresh scrubbing media. These emissions limits are consistent with the available data that establish the MACT floor level of control, and the operating limits are consistent with the operation of the best-performing TEA acid scrubbers.

For these reasons, we adjusted the proposed emissions limit to require the owner or operator to reduce TEA emissions from a TEA cold box mold or core making line at a new or existing foundry by at least 99 percent or to a level that does not exceed 1 ppmv, as determined when scrubbing with fresh acid solution. We also adjusted the proposed operating limit to require that the 3-hour average pH of the scrubber blowdown not exceed 4.5. We also added compliance provisions to implement these new requirements. Plants must conduct an initial performance test to establish that the TEA scrubber is correctly designed to meet the required emissions limit and to establish the minimum flow rate of scrubbing media that must be maintained. Continuous compliance is established by maintaining the scrubber media flow rate at or above the limit established during the performance test and maintaining the pH of the scrubbing media at or below a pH of 4.5.

C. Why Did We Revise the Proposed Work Practice Standards?

Scrap Selection and Inspection

Comment: We received about 20 comments from foundries and recyclers on the proposed work practice standards. Most believed that the requirements are unnecessary because the emissions limits for organic HAP already require capture and control. They stated that cupolas are both designed for and capable of handling some of the restricted material, such as oily scrap, and a cupola is the most environmentally acceptable process in which to recycle these materials.

Response: We proposed a single scrap selection and inspection requirement regardless of the type of melting furnace used. Upon consideration of the public comments and data submitted regarding used oil filter recycling, we agree that a cupola, properly controlled with an afterburner, provides a safe and environmentally beneficial means of recycling oily scrap. That is, our test data and engineering analyses indicate that the afterburner will destroy organic compounds resulting from the melting of oily scrap. Therefore, we have included a specific provision that allows oily scrap in cupolas as long as it is drained of free liquids and an afterburner is used that meets specific design and operating requirements to ensure destruction of organic compounds.

Comment: Several commenters recommended that we include additional specifications or a requirement to ensure that no mercury switches are included in the scrap. These requirements are needed to reduce mercury emissions from the furnaces. These commenters provided information on programs to remove mercury switches from automobile scrap and the potential reductions in mercury emissions when this scrap is melted. Other commenters stated that restrictions on HAP metals in scrap were unnecessary because the melting furnaces have PM controls and are subject to emissions limits for PM.

Response: Although there are provisions for metal HAP emissions control for all furnace types, mercury is not well-controlled by these control systems because of its volatility. We agree with the commenters that removing mercury switches from automobile scrap is the best technique to reduce mercury emissions from melting furnaces. We researched programs currently in place for the removal of mercury switches. We found that there are some mandatory and voluntary programs that are being implemented by the States to remove mercury switches from end of life vehicles. However, we could not confirm that the removal of mercury switches would be part of the floor of a scrap inspection program for iron and steel foundries because some programs were voluntary and others affected scrap recyclers rather than foundries. We evaluated the costs and emissions reductions of mercury switch removal and found that the removal of mercury switches associated with convenience lighting was cost effective. The switches are readily accessible, and for automobiles manufactured in 2001 and earlier, they account for the vast majority of mercury in automobile components. We estimate that such a program could achieve annual mercury reductions of 2,800 pounds at an annual cost of only $3.6 million. This evaluation indicates that it is a reasonable and cost effective beyond-the-floor alternative. Consequently, we incorporated requirements into the scrap inspection program to address the removal of mercury switches from under hoods and trunks.

We also considered the feasibility of the removal of the small amount of mercury that may be used in flat panel displays used in entertainment and navigation systems and in some headlamps. These uses of mercury comprise only 1 percent of that used in automobiles historically, such as convenience light switches. The small amount of mercury, poor accessibility to the mercury, and the costs of removal indicated that removal of mercury from these small applications was not a cost effective alternative for beyond the MACT floor.

There are several other efforts underway to reduce the use of mercury switches in automobiles and to remove them from end of life vehicles. The U.S. automobile industry has committed to removing mercury convenience lighting switches from new automobiles. The Alliance of Automobile Manufacturers (a trade association of car and light truck manufacturers) reports that the use of mercury in automobile components has been reduced to 1 percent of the level used in the 2001 calendar year. Several States and EPA have initiated programs, such as legislative efforts, pilot projects, and outreach campaigns to facilitate the removal of mercury switches from automobile scrap, which is particularly important for vehicles manufactured in 2001 and earlier. These efforts supplement the scrap inspection program in the final rule and will help to ensure continued reductions in mercury emissions in the future.

Several commenters also expressed concerns that lead may not necessarily be well-controlled by these systems depending on the operating temperatures of the control system. Although the data for the two cupola control systems that we tested indicated excellent control of lead emissions, experience with a variety of PM control systems at other industries (but similar types of emissions) indicate that lead removal efficiency may be reduced at higher temperatures. In addition, many plants already limit and inspect for lead components, and many such components are identifiable in scrap. Our analysis of the practices currently used by iron and steel foundries indicates that preventing or removing identifiable lead components in scrap is part of the MACT floor. Therefore, we have included requirements restricting lead components in scrap. However, we have eliminated restrictions for other metal components, such as galvanized parts, both because it is difficult to distinguish these parts from other scrap metals and because the metal HAP that might be released during the melting process are low in volatility and are well controlled by PM control devices over the range of temperatures that these devices operate.

Comment: Numerous commenters recommended that we write the final rule to include specifications with restrictions on the amount of free liquids, grease, oil, and plastic parts; procedures to inspect a representative number of scrap shipments (e.g., 10 percent), and procedures to ensure that oily turnings are properly drained of free liquids. These commenters also stated that the requirement to perform the inspections at the best vantage point was nebulous and makes compliance difficult to ensure. One commenter requested that we write the final rule to exempt any foundry from the scrap inspection and recordkeeping requirements if they use certified metal ingots that do not contain HAP.

Response: We reconsidered the practicality and, in some cases, the vagueness of the proposed scrap inspection program. These commenters have offered several suggestions that will improve the program, and we have written the scrap selection and inspection requirements to incorporate many of these suggestions. For example, we realize it is impractical and almost impossible to inspect all shipments, so we require inspection of representative shipments (but not less than 10 percent of the shipments). The undefined best vantage point for performing the inspections has been revised to a reasonable vantage point. We also clarified that a continuing scrap inspection program is not necessary for those foundries that do not use scrap containing the HAP generating contaminants if they meet compliance certification requirements for their furnace charge materials. These adjustments and the resulting requirements are consistent with the practices at the best-controlled foundries and are representative of the MACT floor.

Comment: Several commenters requested that EPA require foundries to implement the work practice requirements that will reduce mercury emissions (i.e., scrap selection and inspection program) within 1 year of the effective date. The commenters pointed out that most foundries already have these programs in place and no control equipment is needed that might require more time to install. Implementing these requirements sooner would result in greater reductions in mercury emissions especially considering the phase out of mercury switches in new automobiles.

Response: We agree with the commenters' suggestions and see no reason why foundries can not implement the scrap selection and inspection program or certification requirements sooner. While owners or operators of iron and steel foundries are provided 3 years after the effective date of the final rule to comply with other requirements, we are requiring that existing iron and steel foundries comply with the scrap selection and inspection program in § 63.7700(b) or the certification requirements in § 63.7700(c) within 1 year of the effective date of the final rule.

Mold and Core Making

Comment: Several commenters opposed the proposed requirement to manually light off molds because some molds do not produce gases that will support combustion, and they would automatically ignite if they were combustible. It is not practical to inspect each mold vent at high production foundries, and in some cases, hoods or enclosures make it impractical and unsafe to manually ignite and inspect vents. Some commenters stated that the requirements are burdensome and unclear with respect to how to demonstrate compliance (e.g., how quickly they must be lit, how long must they burn, and does the requirement depend on mold size and binder type). Others stated that EPA has not demonstrated that mold light off represents the MACT floor and presented no data to show that HAP emissions would be reduced.

Response: From our observations of foundry operations, ignition of mold vents was a standard operating procedure, although we recognize that ignition of mold vent gasses generally occurs spontaneously. In reviewing the public comments, it is evident that the requirements, as proposed, had several significant short-comings. For foundries with mold vents that are not ignitable, there must be a mechanism to document this fact, they should not be required to try to manually ignite every mold vent, and it should not be necessary to keep records of which mold vents did not ignite. In addition, we did not intend to endanger the safety of the workers through this requirement. Finally, we did not intend to limit mold light off to only manual means. The use of natural gas pilot flames in automated cooling lines to light off mold vents is certainly acceptable; consequently, we adjusted the requirement to manually ignite the gases.

There is no doubt that mold vent gases contain HAP and that the ignition of the mold vent gases will reduce the HAP emissions that occur due to mold off-gassing. Therefore, we have not eliminated requirements for mold vent light off, but we have significantly revised the requirements. The final rule incorporates the mold vent ignition requirements into the O&M plan. The plan must include procedures for providing an ignition source to mold vents unless the owner or operator determines the gases either are not ignitable, ignite automatically, or cannot be ignited due to legitimate accessibility or safety reasons. Criteria are included for determining ignitability. The final rule requires that foundries document and maintain records of this determination.

Coating and Binder Formulations

Comment: We received one comment supporting the proposed requirement for non-HAP coating formulations. We also received many comments from industry representatives opposing the total elimination of HAP. Most of these commenters asked us to allow HAP compounds in small percentages in coatings when they are needed to achieve the physical and chemical properties required by the coating specifications. One commenter explained that there is a small but specialized need for methanol-based coatings. The methanol-based coatings are designed for light off in which the flammable components are consumed so that minimal methanol is released to the environment. Methanol used as a carrier in the coating could be replaced, but not methanol used as an active ingredient in the coating. While methanol has been replaced in many cases by water, methanol in small quantities is needed in coatings as a biocide or surfactant. Several commenters suggested that Material Safety Data Sheets be used to satisfy recordkeeping requirements.

Response: After considering the numerous comments and the technical details associated with this issue, we concluded that we could not show that prohibiting methanol in this application would be a cost-effective beyond-the-floor option. In addition, we cannot show that it is technically feasible in all cases, considering the specialized use of methanol in some applications and the unknown effect on the quality of certain products that must meet coating specifications. For these reasons, we deleted the proposed requirement for non-HAP coating formulations from the final rule. Consistent with our intent to have foundries consider the HAP content and potential HAP emissions from their coating formulations, we are applying recordkeeping requirements to HAP used in coatings. These include requirements to record annual chemical usage rates for each binder system, annual HAP specific usage rates for each binder system, and total HAP usage rate by the foundry. These records will identify those systems with the highest HAP usage rates and make it easier for foundries to focus on opportunities to reduce the HAP content.

Comment: Several commenters said the no methanol requirements placed on furan warm box binder systems should be removed because they were beyond the floor and had not been justified. Also, there is no assurance that binders without methanol can provide the quality of castings that is needed. The commenters explained that the catalyst portion of the binder system is water-based in most current formulations, but the resin portion of the binder system typically contains up to 5 percent methanol as a stabilizer for the resin. Therefore, the no methanol requirement for furan warm box systems should be clarified to limit the requirement of no methanol only to the catalyst and should allow up to 5 percent methanol in the resin material. One commenter recommended that EPA defer all specific binder reformulation requirements until residual risk standards; this will allow time to complete testing on low-emitting binder systems. Another commenter recommended that all specific binder reformulation requirements be deleted because they limit greener alternatives from being evaluated.

Response: The proposed no methanol requirement was not based on a beyond-the-floor analysis; it was based on the fact that over 40 percent of the mold and core making lines using the furan warm box system (based on responses to a detailed industry survey) had switched from a methanol-based catalyst. However, it appears that we mischaracterized the extent to which methanol can be eliminated from the furan warm box system. The survey responses used to establish the MACT floor specifically indicated that the conversion was performed only for the catalyst portion of the binder system. The comments we received verify that conversion to a no-methanol or water-based catalyst is technically feasible. Therefore, we revised the requirement for furan warm box binder systems to indicate that foundries must use a furan warm box catalyst that does not include methanol as a specific ingredient as listed in the Material Data Safety Sheet. We also revised this provision to clarify that the requirement does not apply to the resin portion of the binder system. Methanol is allowed in the resin portion of the binder system. The final rule also requires plants to maintain records of all catalyst binder formulations.

Comment: While one commenter supported the proposed requirement for naphthalene-depleted solvents in binders for phenolic urethane cold box or nobake mold or core making lines, several commenters opposed the requirement. According to these commenters, EPA should delete the requirement because it is beyond the floor and unjustified. Three commenters stated that naphthalene-depleted solvents may increase VOC emissions and that EPA had underestimated the cost. One commenter added that the proposed requirement would be ineffective because naphthalene-depleted solvents contain other HAP. The proposed requirement may require expensive tooling modifications and product testing if cores are changed, and there is no assurance that binders without naphthalene will be capable of providing the quality of castings that is needed, will work at all foundries, or will be available for all major source foundries. Some commenters recommended that EPA encourage environmentally friendly resins using New Source Review Clean Technology concepts and have foundries report on the results. Others recommended requiring a study or deferring the requirement until the residual risk is evaluated.

Response: Based on a review of the comments and upon further analysis, we determined that the requirement for naphthalene-depleted solvents is not warranted. First, the naphthalene-depleted solvent does not provide the same characteristics as the traditional phenolic urethane base solvent and, therefore, may not achieve acceptable quality castings in all applications. Second, we feel we underestimated the cost of the required binder system substitution by not considering the cost to recertify the castings through a production parts approval process. Third, we may have overestimated the amount of HAP emissions reductions that are achievable by the use of the naphthalene-depleted solvent. Therefore, we feel that we cannot require that all phenolic urethane binder systems be converted to a naphthalene-depleted solvent. In addition, the requirement to convert solvents is not a cost-effective alternative; consequently, we rejected the use of naphthalene-depleted solvents as a beyond-the-floor requirement. Therefore, this specific requirement has been removed from the final rule. With this change, almost all of the concerns expressed by the commenters have been addressed.

Comment: Several commenters recommended that the binder system evaluation requirements be deleted. The mold and core binder assessment is a beyond-the-floor requirement with no economic cost-effectiveness demonstration, imposes a heavy burden on the foundry, and is written in a manner subject to interpretation and potential compliance actions. The MACT floor is mostly no change in formulation. Most of these commenters state that EPA does not have the authority to require a re-evaluation every 5 years because MACT standards are to represent a one-time identification of the technologies currently available.

Response: We felt that foundries routinely evaluated alternative binder systems to identify systems that might help to reduce costs, speed production, improve casting quality, and reduce defects. Primarily, we wanted foundries to include in this process an evaluation of the potential HAP emissions and factor in these HAP emissions reductions in the process of selecting an appropriate binder system. However, as proposed, the requirement was too broad (evaluate all binder systems) and too vague (what is a reduced-HAP binder system?) to be practically implemented. As we attempted to craft this requirement into something that could be reasonably implemented without undue burden, we still struggled with numerous questions: what is a reduced-HAP binder system; do we consider emissions only from mold curing or from both mold making and subsequent releases from cooling and shakeout; and how do we define what is technically and economically feasible?

After considering the numerous comments and the technical details associated with this issue, we concluded that any prescriptive requirement we developed would not be a cost-effective beyond-the-floor option. Consistent with our intent to have foundries consider the HAP content and potential HAP emissions from their binder formulations, we are requiring foundries to record the annual chemical usage rates for each binder system employed at the foundry, the annual HAP specific usage rates for each binder system, and the total annual HAP usage rate by the foundry. These records will identify those systems with the highest HAP usage rates and make it easier for foundry owners or operators to focus on opportunities to reduce HAP content. This information can also be considered when alternative binder systems are routinely evaluated for reasons related to production, cost, and quality. In addition, these data will also help to further address mold and core making emissions, if necessary, under section 112(f) for residual risk.

V. Statutory and Executive Order Reviews

A. Executive Order 12866: Regulatory Planning and Review

Under Executive Order 12866 (58 FR 51735, October 4, 1993), the EPA must determine whether the regulatory action is “significant” and, therefore, subject to review by the Office of Management and Budget (OMB) and the requirements of the Executive Order. The Executive Order defines a “significant regulatory action” as one that is likely to result in a rule that may:

(1) Have an annual effect on the economy of $100 million or more or adversely affect in a material way the economy, a sector of the economy, productivity, competition, jobs, the environment, public health or safety, or State, local, or tribal governments or communities;

(2) create a serious inconsistency or otherwise interfere with an action taken or planned by another agency;

(3) materially alter the budgetary impact of entitlement, grants, user fees, or loan programs or the rights and obligations of recipients thereof; or

(4) raise novel legal or policy issues arising out of legal mandates, the President's priorities, or the principles set forth in the Executive Order.

Pursuant to the terms of Executive Order 12866, it has been determined that the final rule is a “significant regulatory action” because it may raise novel legal or policy issues. As such, this action was submitted to OMB for review. Changes made in response to OMB suggestions or recommendations will be documented in the public record.

B. Paperwork Reduction Act

The information collection requirements in the final rule have been submitted for approval to OMB under the Paperwork Reduction Act, 44 U.S.C. 3501 et seq. The information collection requirements are not enforceable until OMB approves them.

The information requirements in the final rule are based on notification, recordkeeping, and reporting requirements in the NESHAP General Provisions (40 CFR part 63, subpart A), which are mandatory for all operators subject to NESHAP. The records and reports required by the final rule are necessary for EPA to: (1) Identify major sources and new or reconstructed sources subject to the rule, (2) ensure that MACT is being properly applied, and (3) ensure that the emissions control devices are being properly operated and maintained on a continuous basis. Based on the reported information, EPA can decide which plants, records, or processes should be inspected. These recordkeeping and reporting requirements are specifically authorized by section 112 of the CAA (42 U.S.C. 7414). All information submitted to the EPA pursuant to the recordkeeping and reporting requirements for which a claim of confidentiality is made is safeguarded according to Agency policies in 40 CFR part 2, subpart B.

The annual average public reporting and recordkeeping burden for this collection of information over the first three years of the information collection request (ICR) is estimated to total 22,325 labor hours per year. This includes 10 responses per year from 98 respondents for an average of 22.7 hours per response. The total annualized cost burden to the facility is estimated at $1,626,649, including labor, capital, and operation and maintenance. The capital cost of monitoring equipment is estimated at $293,700; the estimated annual cost for operation and maintenance of monitoring equipment is $133,300.

Burden means the total time, effort, or financial resources expended by persons to generate, maintain, retain, or disclose or provide information to or for a Federal agency. This includes the time needed to review instructions; develop, acquire, install, and utilize technology and systems for the purpose of collecting, validating, and verifying information, processing and maintaining information, and disclosing and providing information; adjust the existing ways to comply with any previously applicable instructions and requirements; train personnel to respond to a collection of information; search data sources; complete and review the collection of information; and transmit or otherwise disclose the information.

An Agency may not conduct or sponsor, and a person is not required to respond to a collection of information unless it displays a currently valid OMB control number. The OMB control number for EPA's regulations in 40 CFR part 63 are listed in 40 CFR part 9. When the ICR is approved by OMB, the Agency will publish a technical amendment to 40 CFR part 9 in the Federal Register to display the OMB control number for the approved information collection requirements contained in the final rule.

C. Regulatory Flexibility Act

The EPA has determined that it is not necessary to prepare a regulatory flexibility analysis in connection with this final rule. The EPA has also determined that the final rule will not have a significant economic impact on a substantial number of small entities. For purposes of assessing the impacts of the final rule on small entities, small entity is defined as: (1) a small business according to the U.S. Small Business Administration size standards for NAICS codes 331511 (Iron Foundries), 331512 (Steel Investment Foundries), and 331513 (Steel Foundries, except Investment) of 500 or fewer employees; (2) a small governmental jurisdiction that is a government of a city, county, town, school district or special district with a population of less than 50,000; and (3) a small organization that is any not-for-profit enterprise which is independently owned and operated and is not dominant in its field.

After considering the economic impacts of today's final rule on small entities, EPA has concluded that this action will not have a significant economic impact on a substantial number of small entities. Based on SBA size definitions for the affected industries and reported sales and employment data, we identified 20 of the 63 companies incurring compliance costs as small businesses. These small businesses are expected to incur $3.3 million in compliance costs, or 15 percent of the total industry compliance costs of $21.2 million. The mean annual compliance cost as a share of sales for small businesses is estimated at 0.40 percent, and the median is 0.26 percent, with a range of 0.04 to 1.04 percent. We estimate that one small business may experience an impact between 1 and 3 percent of sales, but no small business is expected to experience an impact greater than 3 percent of sales. No significant impacts on their viability to continue operations and remain profitable is expected.

Although the final rule will not have a significant economic impact on a substantial number of small entities, we have nonetheless worked to minimize the impact of the final rule on small entities, consistent with our obligations under the CAA. We have discussed potential impacts and opportunities for emissions reductions with company representatives, and company representatives have also attended meetings held with industry trade associations to discuss the final rule. By changing the proposed requirements for capture systems and revising our initial MACT floor determinations, we have minimized the final rule impacts on small entities to the maximum extent allowable under the CAA.

D. Unfunded Mandates Reform Act

Title II of the Unfunded Mandates Reform Act of 1995 (UMRA), Public Law 104-4, establishes requirements for Federal agencies to assess effects of their regulatory actions on State, local, and tribal governments and the private sector. Under section 202 of the UMRA, the EPA generally must prepare a written statement, including a cost-benefit analysis, for proposed and final rules with Federal mandates that may result in expenditures by State, local, and tribal governments, in the aggregate, or by the private sector, of $100 million or more in any 1 year. Before promulgating an EPA rule for which a written statement is needed, section 205 of the UMRA generally requires the EPA to identify and consider a reasonable number of regulatory alternatives and adopt the least costly, most cost-effective, or least-burdensome alternative that achieves the objectives of the rule. The provisions of section 205 do not apply when they are inconsistent with applicable law. Moreover, section 205 allows the EPA to adopt an alternative other than the least-costly, most cost-effective, or least-burdensome alternative if the Administrator publishes with the final rule an explanation why that alternative was not adopted. Before the EPA establishes any regulatory requirements that may significantly or uniquely affect small governments, including tribal governments, it must have developed under section 203 of the UMRA a small government agency plan. The plan must provide for notifying potentially affected small governments, enabling officials of affected small governments to have meaningful and timely input in the development of EPA regulatory proposals with significant Federal intergovernmental mandates, and informing, educating, and advising small governments on compliance with the regulatory requirements.

Today's final rule contains no Federal mandate (under the regulatory provisions of the UMRA) for State, local, or tribal governments. The EPA has determined that the final rule does not contain a Federal mandate that may result in estimated costs of $100 million or more to either State, local, or tribal governments, in the aggregate, or to the private sector in any 1 year. Thus, today's final rule is not subject to sections 202 and 205 of the UMRA. The EPA has also determined that the final rule contains no regulatory requirements that might significantly or uniquely affect small governments. Thus, today's final rule is not subject to the requirements of section 203 of the UMRA.

E. Executive Order 13132: Federalism

Executive Order 13132 (64 FR 43255, August 10, 1999) requires EPA to develop an accountable process to ensure “meaningful and timely input by State and local officials in the development of regulatory policies that have federalism implications.” “Policies that have federalism implications” is defined in the Executive Order to include regulations that have “substantial direct effects on the States, on the relationship between the national government and the States, or on the distribution of power and responsibilities among the various levels of government.”

The final rule does not have federalism implications. It will not have substantial direct effects on the States, on the relationship between the national government and the States, or on the distribution of power and responsibilities among the various levels of government, as specified in Executive Order 13132. None of the affected facilities are owned or operated by State governments. Thus, Executive Order 13132 does not apply to the final rule.

F. Executive Order 13175: Consultation and Coordination With Indian Tribal Governments

Executive Order 13175 (65 FR 67249, November 9, 2000) requires EPA to develop an accountable process to ensure “meaningful and timely input in the development of regulatory policies on matters that have tribal implications.”

The final rule does not have tribal implications, as specified in Executive Order 13175. It will not have substantial direct effects on tribal governments, on the relationship between the Federal government and Indian tribes, or on the distribution of power and responsibilities between the Federal government and Indian tribes. No tribal governments own or operate facilities subject to the NESHAP. Thus, Executive Order 13175 does not apply to the final rule.

G. Executive Order 13045: Protection of Children From Environmental Health and Safety Risks

Executive Order 13045 (62 FR 19885, April 23, 1997) applies to any rule that: (1) Is determined to be “economically significant,” as defined under Executive Order 12866, and (2) concerns an environmental health or safety risk that EPA has reason to believe may have a disproportionate effect on children. If the regulatory action meets both criteria, the EPA must evaluate the environmental health or safety effects of the planned rule on children and explain why the planned regulation is preferable to other potentially effective and reasonably feasible alternatives considered by the Agency.

The EPA interprets Executive Order 13045 as applying only to those regulatory actions that are based on health or safety risks, such that the analysis required under section 5-501 of the Executive Order has the potential to influence the regulation. The final rule is not subject to Executive Order 13045 because it is based on control technology and not on health or safety risks.

H. Executive Order 13211: Actions That Significantly Affect Energy Supply, Distribution, or Use

This final rule is not a “significant energy action” as defined in Executive Order 13211 (66 FR 28355, May 22, 2001) because it is not likely to have a significant adverse effect on the supply, distribution, or use of energy. Further, we have concluded that the final rule is not likely to have any adverse energy effects.

I. National Technology Transfer Advancement Act

Section 12(d) of the National Technology Transfer and Advancement Act (NTTAA) of 1995 (Public Law 104-113; 15 U.S.C. 272 note) directs EPA to use voluntary consensus standards in its regulatory activities unless to do so would be inconsistent with applicable law or otherwise impractical. Voluntary consensus standards are technical standards (e.g., materials specifications, test methods, sampling procedures, business practices) developed or adopted by one or more voluntary consensus bodies. The NTTAA directs EPA to provide Congress, through annual reports to the OMB, with explanations when the Agency decides not to use available and applicable voluntary consensus standards.

The final rule involves technical standards. The final rule uses EPA Methods 1, 1A, 2, 2A, 2C, 2D, 2F, 2G, 3, 3A, 3B, 4, 5, 5D, 12, and 18, 25, or 25A in 40 CFR part 60, appendix A. Consistent with the NTTAA, EPA conducted searches to identify voluntary consensus standards in addition to these EPA methods. No applicable voluntary consensus standards were identified for EPA Methods 1A, 2A, 2D, 2F, 2G, 5D, and 12. The search and review results have been documented and are placed in the docket for the final rule.

The search for emissions measurement procedures identified 17 voluntary consensus standards applicable to the final rule. Three of the 17 voluntary consensus standards were not available at the time of promulgation and EPA determined that 14 of these 17 standards were impractical alternatives to EPA test methods. Therefore, EPA is not adopting these standards in the final rule. The reasons for this determination are in docket for the final rule.

The following three of the 17 voluntary consensus standards identified in this search were not available at the time the review was conducted for the purposes of the final rule because they are under development by a voluntary consensus body: ASME/BSR MFC 13M, “Flow Measurement by Velocity Traverse,” for EPA Method 2 (and possibly 1); ASME/BSR MFC 12M, “Flow in Closed Conduits Using Multiport Averaging Pitot Primary Flowmeters,” for EPA Method 2; and ISO/DIS 12039, “Stationary Source Emissions—Determination of Carbon Monoxide, Carbon Dioxide, and Oxygen—Automated Methods,” for EPA Method 3A. While we are not including these standards in today's rule, the EPA will consider the standards when they are finalized.

J. Congressional Review Act

The Congressional Review Act, 5 U.S.C. 801 et seq., as added by the Small Business Regulatory Enforcement Act of 1996, generally provides that before a rule may take effect, the agency promulgating the rule must submit a rule report, which includes a copy of the rule, to each House of the Congress and to the Comptroller General of the United States. The EPA has submitted a report containing the final rule and other required information to the U.S. Senate, the U.S. House of Representatives, and the Comptroller General of the United States prior to the publication of the final rule in today's Federal Register. The final rule is not a “major rule” as defined by 5 U.S.C. 804(2).

VI. Statutory Authority

The statutory authority for this action is provided by sections 112, 114, 116, and 301 of the CAA, as amended (42 U.S.C. 7401 et seq.) This rulemaking is subject to the provisions of section 307(d) of the CAA.

List of Subjects in 40 CFR Part 63

• Environmental protection
• Air pollution control
• Hazardous substances
• Reporting and recordkeeping requirements

For the reasons stated in the preamble, title 40, chapter I, part 63 of the Code of Federal Regulations is amended as follows:

PART 63—[AMENDED]

1. The authority citation for part 63 continues to read as follows:

Authority: 42 U.S.C. 7401, et seq.

Subpart A—[Amended]

2. Part 63 is amended by adding subpart EEEEE to read as follows:

Subpart EEEEE—National Emission Standards for Hazardous Air Pollutants for Iron and Steel Foundries

What this Subpart Covers

Editorial Note:

This document was received in the Office of the Federal Register on September 5, 2003.