Browse: Departments   Dates   Agencies  

CFR Citation: 40 CFR Part 63

Docket ID: [Docket ID No. OAR-2002-0085; FRL-7462-3]

RIN ID: RIN 2060-AH55

NOTICE: Part III

DOCUMENT ACTION: Final rule.

SUBJECT CATEGORY: National Emission Standards for Hazardous Air Pollutants for Coke Ovens: Pushing, Quenching, and Battery Stacks

EFFECTIVE DATES: April 14, 2003.

DOCUMENT SUMMARY: This action promulgates national emission standards for hazardous air pollutants (NESHAP) for coke ovens. The final standards establish emission limitations and work practice requirements for control of hazardous air pollutants (HAP) from pushing, quenching, and battery stacks at new and existing coke oven batteries. The HAP emitted from pushing, quenching, and battery stacks include coke oven emissions, as well as polycyclic organic matter (POM) and volatile organic compounds (VOC) such as benzene and toluene. Exposure to these substances has been demonstrated to cause chronic and acute health effects. These final standards will implement section 112(d) of the Clean Air Act (CAA) by requiring all major sources to meet HAP emission standards reflecting application of the maximum achievable control technology (MACT). The EPA previously promulgated emission standards addressing emissions from coke oven charging, topside leaks, and door leaks.

SUMMARY: Environmental Protection Agency,

SUPPLEMENTAL INFORMATION

Regulated Entities. Categories and entities potentially regulated by this action include:
Category NAICS * Example of regulated entities Industry............................. 331111, 324199......... Coke plants and integrated iron and steel mills. Federal government................... ....................... Not affected. State/local/tribal government........ ....................... Not affected. * North American Industry Classification System.

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 Sec. 63.7281 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 under Docket ID No. OAR20020085. The official public docket consists of the documents specifically referenced in this action, any public comments received, and other information related to this action. Although a part of the official docket, the public docket does not include Confidential Business Information or other information whose disclosure is restricted by statute. The official public docket is the collection of materials that is available for public viewing at the Air Docket in the EPA Docket Center (EPA/DC), EPA West, Room B102, 1301 Constitution Ave., 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) 5661744, and the telephone number for the Air Docket is (202) 5661742.

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 of the contents of the official public docket, and to access those documents in the public docket that are available electronically. Although not all docket materials may be available electronically, you may still access any of the publicly available docket materials through the docket facility in the above paragraph entitled ``Docket.'' Once in the system, select ``search,'' then key in the appropriate docket identification number.

Worldwide Web (WWW). In addition to being available in the docket, an electronic copy of the final rule will also be available on the WWW through the Technology Transfer Network (TTN). Following signature, a copy of the final 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) 5415384.

Judicial Review. This action constitutes final administrative action on the proposed NESHAP for coke oven pushing, quenching, and battery stacks (66 FR 35326, July 3, 2001). Under CAA section 307(b)(1), judicial review of the final rule is achievable only by filing a petition for review in the U.S. Court of Appeals for the District of Columbia Circuit by June 13, 2003. Under CAA section 307(b)(2), the requirements that are the subject of this document may not be challenged later in civil or criminal proceedings brought by EPA to enforce these requirements.

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

A. What Is the Source of Authority for NESHAP?

B. What Criteria Are Used in the Development of NESHAP?

C. How Did We Develop the Final Rule?

II. Summary of the Final Rule

A. What Are the Affected Sources and Emission Points?

B. What Are the Requirements for Pushing?

C. What Are the Requirements for Soaking?

D. What Are the Requirements for Quenching?

E. What Are the Requirements for Battery Stacks?

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

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

H. What Are the Compliance Deadlines?
III. Summary of Responses to Major Comments

[[Page 18009]]

A. Why Did We Select a Work Practice Standard for Fugitive Pushing Emissions?

B. What Changes Did We Make to the Work Practice Standard for Fugitive Pushing Emissions?

C. What Changes Did We Make to the Requirements for Pushing Emission Control Devices (PECD)?

D. What Changes Did We Make to the Requirements for Quenching?

E. What Were the Major Comments on the Proposed Standard for Battery Stacks?

F. What Changes Did We Make to the Requirements for Soaking?

G. What Changes Did We Make to the O&M Requirements?

H. Why Did We Change the Compliance Dates for Existing Sources? IV. Summary of Environmental, Energy, and Economic Impacts

A. What Are the Air Emission Reduction Impacts?

B. What Are the Cost Impacts?

C. What Are the Economic Impacts?

D. What Are the NonAir Health, Environmental and Energy Impacts?

V. Statutory and Executive Order Reviews

A. Executive Order 12866: Regulatory Planning and Review

B. Paperwork Reduction Act

C. Regulatory Flexibility Analysis

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 & 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
I. Background

A. What Is the Source of Authority for NESHAP?

Section 112 of the CAA requires the EPA to establish technology based regulations for all categories and subcategories of major and area sources emitting one or more of the HAP listed in section 112(b). Major sources are those that emit or have the potential to emit at least 10 tons per year (tpy) of any single HAP or 25 tpy of any combination of HAP. We previously listed the category of major sources covered by today's final rule, ``Coke Ovens: Pushing, Quenching, and Battery Stacks,'' on July 16, 1992 (57 FR 31576). This action is a rulemaking under section 307(d) of the CAA.

B. What Criteria Are Used in the Development of NESHAP?

Section 112 of the CAA requires that we establish NESHAP for the control of HAP from both new and existing major sources. The CAA requires the NESHAP to reflect the maximum degree of reduction in emissions of HAP that is achievable. This level of control is commonly referred to as MACT.

The MACT floor is the minimum level allowed for NESHAP and is defined under section 112(d)(3) of the CAA. In essence, the MACT floor ensures that the standard is set at a level that assures that all major sources achieve the level of control at least as stringent as that already achieved by the bettercontrolled and loweremitting sources in each source category or subcategory. For new sources, the MACT floor cannot be less stringent than the emission control that is achieved in practice by the bestcontrolled similar source. The MACT standards for existing sources cannot be less stringent than the average emission limitation achieved by the bestperforming 12 percent of existing sources (for which we have emissions information) in the category or subcategory or by the bestperforming 5 sources (for which we have or could reasonably obtain emissions information) for categories or subcategories with fewer than 30 sources.

In developing MACT, we also consider control options that are more stringent than the floor. We may establish standards more stringent than the floor based on the consideration of cost of achieving the emissions reductions, nonair quality health and environmental impacts, and energy impacts.

C. How Did We Develop the Final Rule?

We proposed the NESHAP for the Coke Ovens: Pushing, Quenching, and Battery Stacks source category on July 3, 2001 (66 FR 35326). We provided a 90day comment period for the proposed rule. We received a total of 18 comment letters. A copy of each of these comment letters is available in the docket for this rulemaking (Docket No. OAR20020085).

The final rule reflects full consideration of all the comments we received. Major public comments on the proposed rule along with our responses to these comments are summarized in this document. A detailed response to all comments is included in the Background Information Document (BID) for the promulgated standards (Docket No. OAR2002 0085).

Since publication of the proposal, six coke plants with 12 batteries have permanently closed. The plants have closed primarily because of the distressed economic condition of the iron and steel industry, and none of the closures are due to the cost of installing emission control systems. The requirements in the final rule take into account the levels of control that have been demonstrated as achievable, including in some cases levels achieved by batteries that are no longer operating. We believe it is appropriate to consider all of the data collected and relied upon for the proposed rule. These data reflect the level of performance of batteries operating concurrently with this rulemaking effort, and provide useful and relevant information about the emission limits that such sources can achieve. II. Summary of the Final Rule

A. What Are the Affected Sources and Emission Points?

The affected source is each new or existing coke oven battery at a plant that is a major source of HAP emissions. A new affected source is one constructed or reconstructed after July 3, 2001. An existing affected source is one constructed or reconstructed on or before July 3, 2001. The final rule covers fugitive pushing emissions, emissions from control devices applied to pushing emissions, and emissions from quenching, soaking, and battery stacks.
B. What Are the Requirements for Pushing?

1. ByProduct Coke Oven Batteries With Vertical Flues

We proposed two options for controlling fugitive pushing emissionsnumerical opacity limits (Option 1) and a work practice standard (Option 2). Based on comments received on the proposed rule and further consideration of the proposed options, we are promulgating a work practice standard.

Under the work practice standard, owners or operators must observe and record the opacity from four consecutive pushes each operating day. If the average opacity of the six highest 15second consecutive readings for any individual push is more than 30 percent for a short battery or 35 percent for a tall battery, the owner or operator must take corrective action and/or increase coking time to fix the problem within a specified time frame. To demonstrate the corrective action and/or increased coking time was successful, the owner or operator must observe two additional daytime pushes for the oven after completing the corrective action. If the corrective action is not successful, the owner or operator must take additional corrective action. If the second attempt to fix the problem is not successful, the failure must be reported as a deviation, and the owner or operator must again take corrective action or increase the coking time. Each subsequent failure to
[[Page 18010]]
fix the problem on the same oven must also be reported as a deviation. We have included provisions to qualify an oven for decreased coking time after it is placed on increased coking time, which requires a demonstration that the opacity is 30 percent or less for a short battery or 35 percent or less for a tall battery when the oven is operated on the decreased coking time. If an oven fails to qualify for decreased coking time, it must be returned to the previously established increased coking time, or the owner or operator may implement some other corrective action or increased coking time. If the facility implements some other corrective action or increased coking time, it must confirm that the selected action was successful. If an individual oven fails to qualify for a decreased coking time in two or more consecutive attempts, the failure on the second and any subsequent attempts must be reported as a deviation.

The final rule requires that observers taking opacity readings to comply with the work practice standard for pushing must be certified according to Method 9 in 40 CFR part 60, appendix A. Opacity observations begin with the first detectable movement of the coke mass. The plant owner or operator must identify each oven that cannot be read using Method 9 due to obstructions, interferences, or sun angle and must propose alternative procedures to observe these ovens.

To demonstrate initial compliance, the plant owner or operator must certify, as part of the notification of compliance status, that the plant will meet each of the requirements in the work practice standard. Records of all observations and calculations are needed to document continuous compliance. Additional records are required in each instance where pushing emissions from an oven exceed 30 percent opacity for a short battery or 35 percent opacity for a tall battery.

2. ByProduct Coke Oven Batteries With Horizontal Flues

Plant owners or operators must prepare and implement a written plan to prevent incomplete coking. The plan must establish minimum flue temperatures at different coking times and a lowest acceptable minimum flue temperature. The minimum temperatures must be established based on a study conducted by the plant that establishes minimum flue temperatures at different minimum coking times and an absolute minimum flue temperature. The plan must be submitted to the Administrator for approval. The authority to approve the work practice plan is retained by the Administrator and is not delegated to the State, local, or tribal agency.

In implementing the plan, owners or operators must measure and record the temperature of all flues on two ovens per day within the 2 hours before the scheduled push time. If the measured temperature is below the minimum established for an oven's coking time, the coking time must be increased by the amount specified in the plant's written plan. If the flue temperature measurement is below the lowest acceptable minimum temperature, the oven must be removed from service for repairs. If a flue temperature is below the lowest acceptable minimum after return to service, the owner or operator must report the event as a deviation.

No performance test is required to demonstrate initial compliance with the work practice standards. The plant owner or operator must certify, as part of the notification of compliance status, that the plant has submitted the written plan to prevent incomplete coking and the supporting study to the Administrator for review and approval, and that the plant will meet each of the requirements in the work practice standard beginning no later than the first day that compliance is required according to Sec. 63.7283 of the final rule. If the plan is disapproved, the owner or operator must revise the plan as directed by the Administrator and resubmit it for approval. If an original or re submitted plan has not been approved by the applicable compliance date, the owner or operator must operate in accordance with the last plan submitted to the Administrator.

Plant owners or operators must demonstrate continuous compliance by: (1) Measuring and recording flue temperatures for two ovens a day and for all ovens in each battery at least once a month, and (2) recording the time each oven is charged and pushed and the net coking time. Plant owners or operators must keep additional records to show that the correct procedures were followed if any measured flue temperature is below the minimum flue temperature or the lowest acceptable minimum temperature.

3. NonRecovery Coke Oven Batteries

The final work practice standards require plant owners or operators to visually inspect each oven prior to pushing by opening the door damper and observing the bed of coke. The oven cannot be pushed unless the visual inspection confirms that there is no smoke in the open space above the coke bed, and that there is an unobstructed view of the door on the opposite side of the oven. Plant owners or operators must demonstrate initial compliance by certifying in their initial notification of compliance status that they will follow the work practice standards. Plant owners or operators must demonstrate continuous compliance by maintaining records of each visual inspection. 4. Emission Control Devices

We are establishing emission limits for particulate matter (PM) as a measure of control device performance. Plant owners or operators that currently use capture and control equipment must continue to use such equipment and must meet the applicable PM emission limits. The limits differ in form and numerical value depending on the type of capture system used (cokeside shed or moveable hood) and whether the control device is stationary (landbased) or mobile. Where a cokeside shed is used as the capture system, the PM limit is 0.01 grain per dry standard cubic foot (gr/dscf). If a moveable hood vented to a stationary control device is used to capture emissions, the PM emission limit is 0.02 pound per ton (lb/ton) of coke pushed. For mobile scrubber cars that do not capture emissions during travel, the emission limits are 0.03 lb/ ton of coke for short batteries and 0.01 lb/ton of coke for tall batteries. For mobile scrubber cars that capture emissions during travel, the limit is 0.04 lb/ton of coke.

We have also established operating limits for control devices and capture systems applied to pushing emissions. If a venturi scrubber is used, the daily average pressure drop and scrubber water flow rate must remain at or above the minimum level established during the initial performance test. The final rule provides two options for a capture system applied to pushing emissions: (1) Maintain the daily average fan motor amperes at or above the minimum level established during the initial performance test, or (2) maintain the daily average volumetric flow rate at the inlet of the control device at or above the minimum level established during the initial performance test.

The final rule requires a performance test for each control device to demonstrate it meets the emission limit. The concentration of PM is to be measured using EPA Method 5 or 5D in 40 CFR part 60, appendix A. The testing requirements also include procedures for establishing operating limits for venturi scrubbers and capture systems and for revising the limits, if needed, after the performance test. To demonstrate continuous compliance with the applicable emission limit, plant owners or operators must conduct
[[Page 18011]]
performance tests for each control device at least twice during each term of their title V operating permit (at midterm and renewal).

If a baghouse is applied to pushing emissions, plant owners or operators must monitor the relative change in PM loading using a bag leak detection system and make inspections at specified intervals. The basic inspection requirements include daily, weekly, monthly, or quarterly inspections of specified parameters or mechanisms with monitoring of bag cleaning cycles. Each bag leak detection system must be capable of detecting PM at concentrations of 10 milligrams per actual cubic meter or less and provide an output of relative PM loading, and be installed and operated according to EPA guidance.\1\ If the system does not work based on the triboelectric effect, it must be installed and operated consistent with the manufacturer's written specifications and recommendations. In addition, the bag leak detection system must be equipped with an alarm system that will alert operators if PM is detected above a preset level. The proposed requirement that a bag leak detection system must not sound for more than 5 percent of the time in a semiannual period has been deleted from the final rule. \1\ ``Fabric Filter Bag Leak Detection Guidance,'' EPA 454/R98 015, September 1997, available on the TTN at http://www.epa.gov/ttnemc01/cem/tribo.pdf_____________________________________

To demonstrate continuous compliance, the final rule requires plant owners or operators to maintain records of corrective actions taken in response to bag leak detection system alarms. They must also keep records documenting conformance with the inspection and maintenance requirements.

If a venturi scrubber is applied to pushing emissions, plant owners or operators must monitor the daily average pressure drop and scrubber water flow rate using continuous parameter monitoring systems (CPMS). The CPMS must measure and record the pressure drop and scrubber water flow rate at least once per push and determine and record the daily average of the readings. To demonstrate continuous compliance with the operating limits, plant owners or operators must maintain the daily average pressure drop and scrubber water flow rate at levels no lower than those established during the performance test. Valid monitoring data must be available for all pushes.

Section 63.7331 of the rule establishes requirements for the installation, operation, and maintenance of continuous monitoring systems. The final rule requires owners or operators to prepare a site specific monitoring plan for CPMS that addresses installation, performance, operation and maintenance, quality assurance, and recordkeeping and reporting procedures. These requirements replace the more detailed performance specifications contained in the proposed rule.

For a capture system applied to pushing emissions, plant owners or operators are required to check the fan motor amperes or the volumetric flow rate at least once each 8hour period to verify the daily average is at or above the level established during the initial performance test and to record the results of each check.

C. What Are the Requirements for Soaking?

The final rule contains a work practice standard to address emissions that occur during soaking, which is the period prior to pushing when an oven is dampered off the collecting main and vented to the atmosphere through an open standpipe to relieve oven pressure. Plant owners or operators must prepare and implement a plan to mitigate potential soaking emissions. Each plan must include measures and procedures to train topside workers to identify the cause of soaking emissions and to take corrective measures to reduce or eliminate such emissions.

If soaking emissions are caused by leaks from the collecting main, actions must be taken to eliminate the emissions, such as reseating the damper, cleaning the flushing liquor piping, applying aspiration, putting the oven back on the collecting main, or igniting the emissions. If soaking emissions are not caused by leaks from the collecting main, a designated responsible party must be notified, who must then determine whether the cause of emissions is incomplete coking. If so, the oven must either be put back on the collecting main until coking is complete, or the emissions must be ignited.

To demonstrate initial compliance, the plant owner or operator must certify, as part of the notification of compliance status, that the plant has submitted the written plan for soaking to their permitting authority for review and approval, and that each of the requirements in the work practice standard will be met beginning no later than the first day that compliance is required according to Sec. 63.7283 of the final rule. To demonstrate continuous compliance, plant owners or operators must keep records documenting conformance with these requirements.

D. What Are the Requirements for Quenching?

The equipment and work practice standards for quenching apply to all coke oven batteries. Each quench tower must be equipped with baffles such that no more than 5 percent of the cross sectional area of the tower may be uncovered or open to the sky. Baffles must be cleaned each day that the quench tower is used except when the highest measured ambient temperature during the day is below 30 degrees Fahrenheit. Each quench tower must be inspected at least monthly for damaged or missing baffles and blockage. If the monthly inspection reveals any damaged or missing baffles, plant owners or operators must initiate repairs within 30 days and complete repairs as soon as practicable.

The final rule also limits the total dissolved solids (TDS) content of water used for quenching to 1,100 milligrams per liter (mg/L). The final rule includes an alternative to the TDS limit that achieves an equivalent level of HAP control. The plant owner or operator may establish a sitespecific constituent limit for the HAP that are characteristic of coke oven emissions (benzene, benzo(a)pyrene, and naphthalene). The constituent limit is based on analyses of at least nine samples of the quench water for TDS, benzene, benzo(a)pyrene, and naphthalene. The HAP limit is the highest sum of the concentrations of the HAP in any single sample that meets the TDS limit of 1,100 mg/L. We also replaced the definition of ``clean water'' with a definition of ``acceptable makeup water,'' which includes surface water from a river, lake, or stream; water meeting drinking water standards; storm water runoff and production area clean up water except for water from the by product recovery plant area; process wastewater treated to meet effluent limitations guidelines; any of these types of water that have been used only for noncontact cooling or in water seals; or water from scrubbers used to control pushing emissions.

To demonstrate initial compliance, the plant owner or operator must certify, as part of the notification of compliance status, that the equipment standard has been met, and that the work practice requirements regarding baffle repair and cleaning will be met beginning no later than the first day that compliance is required according to Sec. 63.7283 of the final rule. The owner or operator must also conduct an initial performance test to demonstrate that the TDS content of quench water does not exceed 1,100 mg/
[[Page 18012]]
L or that the concentration of benzene, benzo(a)pyrene, and naphthalene does not exceed the sitespecific constituent limit. To demonstrate continuous compliance, plant owners or operators are required to maintain baffles in each quench tower to meet the rule requirements, test quench water for TDS at least weekly or at least monthly for benzene, benzo(a)pyrene, and naphthalene, and keep records documenting conformance with the work practice requirements regarding baffle repair and cleaning.

Backup quench stations at existing coke oven batteries that are used for less than 5 percent of the quenches in a 12month calendar period are not subject to the baffle requirements for quench towers. However, backup quench stations at new batteries are subject to the requirements for baffles.

E. What Are the Requirements for Battery Stacks?

The final rule requires plant owners or operators to monitor the opacity of emissions from each battery stack using a continuous opacity monitoring system (COMS) and to meet specified opacity limits at all times. The opacity limits are a daily average of 15 percent for a by product coke oven battery on a normal coking cycle and a daily average of 20 percent for a byproduct coke oven battery on a batterywide extended coking cycle. A battery is on batterywide extended coking if the average coking time for all ovens in a battery is increased by 25 percent or more over the manufacturer's specified design rate.

Initial compliance must be demonstrated through a performance test using a COMS. The opacity of emissions from each battery stack must be monitored for 24 hours and the daily average determined. A performance evaluation is also required to show that the COMS meets Performance Specification (PS) 1 in appendix B to 40 CFR part 60. To demonstrate continuous compliance, plant owners or operators must monitor opacity using the COMS and determine and record the 24hour average opacity. F. What Are the Operation and Maintenance (O&M) Requirements?

All plant owners or operators are required to prepare and implement a written startup, shutdown, and malfunction plan according to the O&M requirements in 40 CFR 63.6(e). Operation and maintenance plans are also required for each byproduct coke oven battery and for capture systems and control devices applied to pushing emissions.

The plan for general O&M of each byproduct coke oven battery must address procedures (and frequency of measurements, where appropriate) for underfiring gas parameters, flue and crosswall temperatures, preventing ovens from being pushed before they are fully coked, preventing overcharging and undercharging of ovens, and inspecting flues, burners, and nozzles.

The O&M plan for capture systems and control devices applied to pushing emissions must describe procedures for monthly inspections of capture systems, preventative maintenance requirements for control devices, and corrective action requirements for baghouses. In the event of a bag leak detection system alarm, the plan must include specific requirements for initiating corrective action to determine the cause of the problem within 1 hour, initiating corrective action to fix the problem within 1 working day, and completing all corrective actions needed to fix the problem as soon as practicable.

To demonstrate initial compliance, plant owners or operators must certify in their notification of compliance status that they have prepared the plan according to the rule requirements and that the plant will operate according to the plan beginning no later than the first day that compliance is required under Sec. 63.7283 of the final rule. To demonstrate continuous compliance, plant owners or operators must adhere to the requirements in the plan and keep records documenting conformance with these requirements.
G. What Are the Notification, Recordkeeping, and Reporting Requirements?

The notification, recordkeeping, and reporting requirements rely on the NESHAP General Provisions in 40 CFR part 63, subpart A. Table 1 of the final rule (subpart CCCCC) shows each of the requirements in the General Provisions (Sec. Sec. 63.2 through 63.15) and whether they apply.

The final rule requires the owner or operator to submit each initial notification in the NESHAP General Provisions that applies to them. An initial notification of applicability with general information about the plant must be submitted within 120 days of April 14, 2003 (or for a new affected source, 120 days after becoming subject to the rule). A notification of performance tests must be provided at least 60 calendar days before each test. A notification of compliance status must be submitted within 60 calendar days of the compliance demonstration if a performance test is required or within 30 calendar days if no performance test is required. For the work practice standard for pushing for a byproduct coke oven battery with horizontal flues, plant owners or operators must provide prior written notification of the date the study of flue temperatures will be initiated. Other notification requirements that may apply are shown in Table 1 of the final rule (subpart CCCCC).

The final rule requires plant owners or operators to maintain the records required by the NESHAP General Provisions that are needed to document compliance, such as performance test results; copies of startup, shutdown, and malfunction plans and associated corrective action records; monitoring data; and inspection records. All records must be kept for a total of 5 years, with the records from the most recent 2 years kept onsite. The final rule also requires that the current O&M plans be kept onsite and available for inspection upon request for the life of the affected source or until the affected source is no longer subject to the rule requirements.

We revised the reporting requirement for battery stacks from monthly to quarterly in response to comments. For other affected sources, semiannual reports are required for any deviation from an emission limitation (including an operating limit), work practice standard, or O&M requirement. Each report is due no later than 30 days after the end of the reporting period. If no deviation occurred and no continuous monitoring systems were out of control, only a summary report is required. If a deviation did occur, more detailed information is required.

An immediate report is required if there were actions taken during a startup, shutdown, or malfunction that were not consistent with the startup, shutdown, and malfunction plan. Deviations that occur during a period of startup, shutdown, or malfunction are not violations if the owner or operator demonstrates to the permitting authority that the source was operating in accordance with the startup, shutdown, and malfunction plan.

H. What Are the Compliance Deadlines?

We revised the compliance date for an existing affected source from 2 years to 3 years after April 14, 2003. New or reconstructed sources that startup on or before April 14, 2003. New or reconstructed sources that startup after April 14, 2003 must comply upon initial startup. [[Page 18013]]
III. Summary of Responses to Major Comments
A. Why Did We Select a Work Practice Standard for Fugitive Pushing Emissions?

We proposed an opacity standard for fugitive pushing emissions as one potential option for controlling sources in the category. Because we were uncertain about the feasibility of an opacity standard for this emission point, we also proposed a work practice standard. We refer to the opacity standard as Option 1 and the work practice standard as Option 2. Both options would require observing four consecutive pushes per day and determining the average opacity of each push. The opacity limits proposed were 20 percent for short batteries and 25 percent for tall batteries based on the average of four pushes. We proposed a work practice standard that would be triggered if the average opacity of any single push exceeded 30 percent for short batteries and 35 percent for tall batteries.

Comment: Four commenters stated a preference for a work practice standard. Two commenters said that EPA has not and cannot adequately subcategorize batteries to account for the range in performance achievable by batteries implementing a stateoftheart O&M program for the minimization of green pushes. The commenters stated there are not enough data to set standards for each subcategory reflecting the performance of the top sources over time and under the worst foreseeable conditions. Therefore, the opacity standard (Option 1) must be rejected.

One commenter prefers an opacity standard over a work practice standard because he believes a work practice standard could cause several problems: (1) It would not allow them to effectively manage their longterm wall and end flue replacement program; (2) the constant change from taking ovens out of service and putting them back into service would result in damage to the battery; and (3) many of the actions required by the work practice standard would disrupt the heating system, damage refractory, and increase emissions in other areas of the battery.

Three commenters urged EPA to combine the opacity standard with the work practice standard. One commenter noted that the opacity standard does not require that an oven be repaired, and the work practice standard may not be sufficient to keep a problem oven from continuing to operate. Two commenters prefer a combination because it would more closely approach their existing State standards. Another commenter prefers the opacity standard but would support combining it with the work practice standard if it improved compliance.

Response: The insight provided by several commenters and further consideration of the two options we proposed lead us to conclude that a work practice standard that requires owners or operators to take appropriate corrective action and to confirm that they have successfully addressed problem ovens is the most effective approach to control fugitive pushing emissions. A work practice standard is appropriate because pushing emissions are fugitive in nature and are not emitted through a conveyance designed to capture and control HAP. Moreover, there is no practicable measurement methodology to determine the mass emission rate of HAP in these fugitive emissions. Section 112(h) of the CAA explicitly permits a work practice standard in lieu of an emission standard when emissions cannot be emitted through a conveyance.

We concluded an opacity limit as proposed would not be appropriate because coke oven batteries cannot entirely avoid green pushes. While facilities can significantly reduce the frequency of green pushes by carefully monitoring emissions and responding quickly to diagnose and repair problem ovens, they cannot eliminate them altogether. (For example, a flue may become plugged unexpectedly during coking.) Any steps that we might take to allow for the periodic exceedance of an emission limit (such as averaging across several pushes) would undermine the purpose of the standard by allowing malfunctioning ovens to continue operating without diagnosis or repair. Therefore, the most meaningful approach is to establish a work practice standard that requires coke oven facilities to identify and successfully remedy problems that result in increased emissions. Accordingly, considering the nature of the pushing operation, it is appropriate for EPA to establish a work practice standard that uses opacity observations to identify problem ovens (those which produce green pushes) and requires corrective action to diagnose and correct the problem.

There was a fundamental flaw in the opacity standard as proposed in that it would not ensure that an oven producing green pushes is repaired. If the fourpush average exceeds the opacity standard, one or more of the ovens may have serious problems that require immediate attention to prevent subsequent green pushes. However, these problem ovens would not have to be observed again for 90 days, and during that 90day period, many green pushes could occur.

Additionally, an opacity standard based on the average of four pushes does not reliably indicate when a green push has or has not occurred. We analyzed data from two batteries that had frequent green pushes to compare the effectiveness of the opacity standard and work practice standard in identifying green pushes. We found cases where the fourpush average had one oven with a green push (an opacity of more than 30 percent), but the proposed opacity standard was not exceeded because the other pushes had low opacity. We also found cases where the 20 percent opacity standard was marginally exceeded, but none of the pushes were green (i.e., all four pushes were less than 30 percent).

In contrast, the work practice standard is triggered by opacity observations of individual ovens. When a green push occurs, the problem oven is identified. This oven is then placed on a remedial track that requires appropriate repairs in a reasonable period of time. Consequently, the work practice standard will not allow green pushes to occur unabated.

Several commenters urged us to combine the performance standard (an opacity limit) with the work practice standard. While we are not adopting a specific performance standard in the form of a hard and fast opacity limit, and we do not believe that such a standard would provide a feasible mechanism for identifying and remediating individual problem ovens, we do recognize the benefits of having a mechanism to prevent ongoing failure to repair problem ovens.

Therefore, we have revised the work practice standard to ensure that ovens are properly repaired. As proposed, the work practice standard could have allowed individual problem ovens to continue to operate, while cycling through corrective actions without ever being properly repaired. Consequently, we revised the work practice standard to require an owner or operator to report a deviation after two consecutive unsuccessful attempts at corrective action and/or increased coking time and after two consecutive unsuccessful attempts to decrease coking time on the same oven. In addition, subsequent consecutive failures to repair or remediate the same oven must be reported as deviations. There is adequate time provided to correct any problems during the two attempts20 days or more. An owner or operator may also remove an oven from service for as long as necessary to conduct repairs.
[[Page 18014]]
This approach accurately reflects the performance of the best controlled facilities in the category that already implement oven diagnosis and repair programs to successfully identify and remedy problems that lead to increased emissions. Most of the bestcontrolled batteries will seldom have an oven that enters the ovendirected program, and our data show that none have had the types of continuing problems that would result in a deviation under the final rule.

We believe that the work practice standard can be coordinated with a longterm repair program. The batteries upon which the MACT floor is based have a longterm repair program to address major repairs. This longterm program includes procedures for minimizing impacts on adjacent ovens and preventing excess emissions when ovens must be removed from service. In addition, these batteries have effective procedures for identifying problem ovens and making shortterm repairs. There is no legitimate reason why this type of approach cannot be implemented at other coke oven batteries.
B. What Changes Did We Make to the Work Practice Standard for Fugitive Pushing Emissions?

Comment: Four commenters requested revisions to the work practice standard. They requested that the final rule require that all pushes be read exactly according to EPA Method 9 (40 CFR part 60, appendix A). They suggested that Method 9 observations begin with the first detectable movement of the coke mass because this would ensure that observations are made through the entire pushing sequence and would be consistent with how the data were generated for the proposed rule. They also requested that we not require ``independent certified observers'' because all Method 9 certified observers are qualified and should be treated the same.

The commenters asked that we allow the observation of more than four pushes per day so that every oven can be observed at least once every 3 months. In addition, the commenters asked that we clarify that the pushing schedule can be changed for operational reasons, but not ``solely'' for the purpose of changing the sequence of observations. They suggested we add a definition for ``increased coking time'' to prevent confusion with ``batterywide extended coking time,'' which is a term used only in the provisions for battery stacks.

Response: We agree with some of these suggested revisions and do not agree with others. We do not agree that all ovens must be read exactly as required by Method 9 (40 CFR part 60, appendix A) because we are aware that the view of opacity from some ovens may be obstructed within the sector required by the method. In this situation, the observer may need to find an alternative position to make opacity observations. We added a provision to the final rule requiring plant owners or operators to identify ovens that cannot be observed according to Method 9 and develop alternative procedures to determine if green pushes are occurring on those ovens. The alternative procedures must be submitted to the permitting authority for review and approval. Facilities must operate according to these procedures beginning no later than the applicable compliance date. Based on the information we received, there are only a few ovens that fall into this category.

We have written the final rule to state that Method 9 observations should begin with the first detectable movement of the coke mass. In addition, we agree that any Method 9 certified observer is qualified to make Method 9 opacity observations and have changed the provision to reflect this. We also agree that more than four ovens may be observed each day because doing so provides more scrutiny of performance and greater assurance that every oven can be observed at least once every 90 days.

With respect to the comment on changing pushing schedules, we do not believe that the precise language that the commenter suggests is appropriate (specifically the word ``solely'' would create an extraordinarily difficult burden of proof for purposes of enforcement). However, we do agree with the general idea underlying the commenter's recommendation, and we have written the final rule to acknowledge that there may be legitimate operational reasons for changing the pushing schedule. If an oven's pushing schedule is changed and that oven was previously scheduled to be one of the four consecutive ovens to be observed, the operator must keep records to document the legitimate operational reason for changing the schedule. We have added a definition for ``increased coking time'' to prevent confusion with ``batterywide extended coking time,'' which is a term used only in the provisions for battery stacks.

Comment: Several commenters said that the rule should not mandate that an oven be taken out of service if corrective actions are unsuccessful. In addition, commenters requested that after taking corrective actions or extending the coking time, we allow two coking cycles before requiring the facility to demonstrate that the action was successful. They believe it is necessary to observe only one push rather than two to show the action was successful. Finally, the commenters asked that we drop the requirement to obtain the permitting authority's permission to return an oven to service and instead change this to a notification requirement.

Response: We added a provision that requires plant owners or operators to report a deviation after two unsuccessful attempts at repair, and with this requirement, we believe that it is not necessary to require that an oven be removed from service. Our goal at proposal was to require that an oven be removed from service for repair to avoid endless cycling of unsuccessful repairs. This is accomplished in the final rule by requiring that the owner or operator repair the problem oven, and by requiring the owner or operator to bring any two or more consecutive failures to repair the same oven to the attention of the permitting authority by reporting the failure(s) as a deviation.

Based on the comments requesting more time to fix problem ovens before they are removed from service, we investigated the time that might reasonably be required to take corrective action and to demonstrate that it was successful. We discovered that there can be some situations in which it would be difficult to obtain valid opacity observations within the time period in the proposed rule. For example, the opportunity to make opacity observations according to the prescribed procedures depends on coking time, number of daylight hours, sun angle, and other factors. In some cases, it may take several days to meet the criteria in the opacity procedures for a specific oven, especially during the winter months for ovens with 22 to 26 hour coking times. Consequently, we have written the final rule to require that the opacity observations to demonstrate that corrective action and/or increased coking time was successful be made on the first two pushes that can be observed according to the procedures for opacity observations after the allowed number of days. We decreased the time period to complete corrective action or increase coking time because the time period no longer includes the demonstrative observations. We have written the final rule to allow either 10 days or the number of days determined using an equation, whichever is greater. Depending on coking time, the time period allows batteries 10 to about 20 days to diagnose the problem,
[[Page 18015]]
implement corrective action or increased coking time, and stabilize oven temperatures. After that period, the next two pushes that can be observed according to the procedures must be observed to evaluate the success of corrective action. Days during which the oven is removed from service do not count in the allowed number of days. We also revised the standard to allow two attempts at repair in case the problem is not initially diagnosed properly or in case a second independent problem develops.

We do not agree that two coking cycles are always necessary to stabilize an oven after corrective actions are taken. We believe there is one case in which two coking cycles are needed to allow the oven temperature to stabilizewhen an oven that was placed on increased coking time has been repaired and the owner or operator attempts to qualify for decreased coking time. We have written the final rule to reflect this. There is adequate time within the allowed number of days following corrective action or increased coking time to allow the oven temperatures to stabilize. Adequate time is also provided for ovens removed from service because the time during which the oven is not operating is not counted in the allowed number of days. Relative to the comment that only one observation is needed to demonstrate the problem has been corrected, we continue to believe that two pushes should be observed rather than one to provide assurance that the repair was successful.

We agree that it is not necessary for a permitting authority to approve returning an oven to service, and the permitting authority may not be able to act within a time frame that is consistent with the legitimate needs of the operator. In addition, this requirement places a burden on the permitting authority that they may not want and may not have the resources or expertise to implement.

Comment: Three commenters stated that batteries with horizontal flues would be subject to significantly less stringent standards than batteries with vertical flues. They requested that these batteries be subject to the same pushing requirements as byproduct batteries with vertical flues.

Response: As stated in the proposal preamble, unlike vertical flue batteries which include 25 to 37 individual flues along each oven wall, the horizontal flue system of the Semet Solvay design includes only five horizontal flues which convey the combustion gases from top to bottom in serpentine fashion. Because the hot combustion products flow from one flue to the next, the heat control of each upper flue materially affects the heating conditions in the next flue down. Each flue in the horizontal design affects a larger percentage of the total coke mass than for the vertical flue design. Consequently, the occurrence of a heating or combustion problem in any of the single horizontal flues could have a significant adverse effect on the degree and uniformity of coking across the entire length of the coke bed. Therefore, since these differences in design and operation affect pushing emissions, we developed a separate subcategory for batteries with horizontal flues. There are two batteries with this design, and the work practice standard is based on the procedures used by these batteries to prevent green pushes. We have received no technical information that indicates this subcategorization was inappropriate.

However, after we reviewed the proposed work practice standard, we concluded a revision was needed to ensure that a source would not be permitted to operate its ovens below the lowest acceptable minimum flue temperature. The source is required to evaluate coking time, coking temperature, and factors associated with incomplete coking to develop minimum flue temperatures and coking times. The source must then submit to the Administrator (or delegated authority) for review and approval a written plan that establishes minimum flue temperatures for different coking times, and that establishes the lowest acceptable minimum flue temperature for oven operation. The plan must also include appropriate operation and maintenance procedures to ensure compliance upon plan implementation.
C. What Changes Did We Make to the Requirements for Pushing Emission Control Devices (PECD)?

Comment: Two commenters stated that there is no legal basis for setting MACT standards for PECD given EPA's conclusion at proposal that PECD are not part of the MACT floor for pushing. One commenter also stated that EPA has no legal authority to set operating limits for PECD because they are simply a surrogate for the underlying emission limits. In addition, PECD should not be regulated because the emissions do not contain HAP. The commenter said the limits and monitoring are not necessary and are duplicative of other existing requirements, including State implementation plans, title V permits, and the compliance assurance monitoring program.

Response: We believe emission limits for PECD are appropriate and warranted. As we explained in the preamble to the proposed rule, there are several reasons we do not believe it is appropriate to include PECD as a component of the MACT floor for pushing. However, we also indicated at proposal that operation of these controls does have some HAP reduction benefits (although we are unable to specifically quantify these benefits in terms of either HAP or PM), and there is little doubt that these devices help to reduce HAP emissions, including POM and trace metals. Thus, while minimizing the frequency of green pushes is the basis for the MACT floor, and achieving this objective will significantly decrease the emission benefits of the addon control devices, these devices will continue to reduce HAP emissions to some degree on a continuing basis. The EPA has reasonably concluded that it is important to ensure that the benefits related to the operation of these controls are maintained, and the appropriate way to accomplish this is to require that coke plants operate existing PECD at all times in a manner consistent with good air pollution control practices. Accordingly, today's requirements establish emission limitations for existing control devices that reflect the performance of welloperated PECD. The costs associated with the PECD requirements include those for periodic Method 5 testing, parametric monitoring (such as bag leak detection systems), and monthly inspections of capture and control systems. These costs are only $4,600 per year for a typical coke plant, which is a minimal cost relative to the overall costs of the final rule (less than 0.5 percent). While we are not able to quantify the HAP emission reductions associated with operation of PECD or with the PECD requirements in the final rule, we believe the requirements preserving these existing benefits of PECD's and ensuring proper operation of control devices is warranted. For example, bag leak detection systems and monthly inspections will ensure that corrective actions are taken promptly when the systems are not operating properly, and these actions will reduce excess emissions that might have occurred in the absence of the continuous monitoring.

We do not believe that the limits will duplicate existing State requirements because the limits are generally equivalent to or more stringent than those currently required by State agencies or contained in existing operating permits. By establishing these limits in national standards, we will
[[Page 18016]]
ensure that emissions from PECD do not increase in the future if existing State limits are relaxed or if operating permits are modified.

Comment: One commenter stated that the proposed emission limits are based on very limited data and that the limits are not achievable. In support of this claim, the commenter submitted statistical analyses that indicate that their ``statisticallyderived values'' are much higher than the proposed limits and should be used in lieu of the proposed values. Several commenters submitted additional test data for EPA to consider and asked for higher limits.

Response: We reviewed the additional test data submitted by the commenters. These new data include additional tests on mobile scrubber cars used on short batteries and baghouses applied to cokeside sheds. We also reexamined our approach for selecting appropriate emission limits. We believe that it is not necessary to use statistical analyses to account for variability because these control devices operate uniformly over time, and the data indicate there is little variability when the device is operating properly. In addition, we have data for most of the affected control devices, including multiple tests for some units. We believe the large database inherently accounts for variability and choosing the highest threerun average means that 100 percent of the test results are below the limit. However, to account for inherent variability in the performance of the control devices (to more accurately reflect the actual performance of existing controls over time), we established the limits in the final rule by rounding the highest test results to two decimal places.

The two additional tests for mobile scrubber cars used on short batteries include one result slightly below the proposed limit and another slightly higher than the proposed limit. The tests were conducted using approved methods and appear to be representative of normal operation. In addition, the results expanded the database for this subcategory from three tests to five tests. The averages for the five tests ranged from 0.012 to 0.025 lb/ton of coke. We rounded 0.025 lb/ton to 0.03 lb/ton and established this value as the limit for mobile scrubber cars for short batteries.

We also reviewed additional test data for three batteries equipped with a cokeside shed and baghouse, including three tests conducted on a 6meter battery at one plant and four tests conducted on two 4meter batteries designated Batteries 1 and 4 at a second plant. The proposed limit for existing cokeside sheds and baghouses was 0.004 gr/dscf. With the additional data, we now have results for ten tests at five batteries with cokeside sheds and baghouses. All three tests on the 6 meter battery are below the proposed limit of 0.004 gr/dscf with values of 0.0009, 0.0024, and 0.0013 gr/dscf.

The additional data for the two 4meter batteries plus one test result which we previously had gives us a total of five tests for that plant, four tests for Battery 1 and one test for Battery 4. The company acknowledged that a 1984 test which averaged 0.02 gr/dscf was performed under unrepresentative conditions because of operational problems with the baghouse during the test. We examined the other test reports for Battery 1 and found that a test conducted in 1984 averaged 0.004 gr/ dscf, a 1988 test averaged 0.0036 gr/dscf, and a 1998 test averaged 0.01 gr/dscf. The test reports indicate that sampling was performed under representative conditions. Consequently, we revised the emission limit for batteries with cokeside sheds to 0.01 gr/dscf to reflect the level that has been demonstrated as achievable.

No additional data were submitted for two types of capture and control systems: mobile scrubber cars on tall batteries and mobile scrubber cars that capture during travel. We chose as limits the highest threerun average for each of these systems0.01 lb/ton for mobile scrubber cars on tall batteries and 0.04 lb/ton for mobile scrubber cars that capture during travel. We believe the data show that these limits are achievable because they have been achieved at several different batteries over time.

Comment: One commenter requested that the 5 percent operating limit for bag leak detection system alarms be deleted. The commenter argued that the 5 percent of the operating time limit on alarms is arbitrary. In addition, the commenter stated that EPA had not demonstrated that a bag leak detection system is workable for pushing emissions given the intermittent operation of PECD (e.g., 1 to 2 minutes during a push, which occurs every 15 to 20 minutes).

Response: We reexamined the proposed operating limit of 5 percent for bag leak detection systems and concluded it was not applicable for PECD. The proposed limit was adopted from other rules and was not based on data associated with baghouses applied to pushing emissions. We do not believe we can establish an appropriate limit in this application because of the intermittent operation of baghouses. For most systems, the device operates only during the push, which is 1 to 2 minutes every 10 to 15 minutes. In addition, we have no information on the effect of the initial surge when full evacuation is applied at the beginning of the push. Thus, given that emissions from PECD are not the major focus of today's final rule and are not included as part of the MACT floor calculation, we believe it is appropriate to delete the 5 percent operating limit for bag leak detection systems. However, we are requiring that corrective actions be initiated within 1 hour of an alarm.
D. What Changes Did We Make to the Requirements for Quenching?

Comment: One commenter stated that the definition of ``clean water'' needs to be clarified because it would be difficult or impossible for plant owners or operators to prove that some sources of water meet the definition. As proposed, ``clean water'' is defined to mean surface water from a river, lake, or stream; water meeting drinking water standards; water that has been used for noncontact cooling; or process wastewater that has been treated to remove organic compounds and/or dissolved solids. The commenter recommended that the definition be revised to state that any water can be used except untreated process wastewater from the byproduct plant. Another commenter agreed and further stated that plant owners or operators should be allowed to use any source of makeup water that has been used historically and previously deemed acceptable by EPA. One commenter requested that the definition include water that is used in seals on standpipes; otherwise, the plant owner or operator would have to draw an additional 200,000 gallons per day from Lake Michigan and treat the same amount of water before discharge. Another commenter requested that storm water and wash down water associated with nonrecovery plants be added. The commenter stated that this water does not pick up toxic chemicals at nonrecovery plants, and using this water for quenching eliminates discharge to the watershed and reduces the amount of water drawn from the water supply.

Other commenters requested that the proposed definition of ``clean water'' be tightened by developing minimum quality standards for quench water. Two commenters suggested that ``clean water'' be defined as meeting Federal safe drinking water standards. Two other commenters asked that EPA establish a limit for TDS because the solids contain metals. Commenters also noted that the definition includes process water that has been treated to remove organic compounds and/or [[Page 18017]]
dissolved solids. They stated that removal of both solids and organics should be required, and EPA must establish appropriate levels of treatment. If a

FOR FURTHER INFORMATION CONTACT Lula Melton, Metals Group (C439-02), Emission Standards Division, U.S. EPA, Research Triangle Park, NC 27711, telephone number (919) 5412910, electronic mail (email) address, melton.lula@epa.gov.