# NESHAP: Standards for Hazardous Air Pollutants for Hazardous Waste Combustors-Proposed Amendments > Briefs, arguments, decisions, and more. URL: https://www.frixlaw.com/law-library/documents/fr%3A01-16426 ## Record - **Collection:** Federal Register - **Document type:** Proposed Rule - **Published:** July 3, 2001 - **Citation:** 66 FR 35126 ## Text ENVIRONMENTAL PROTECTION AGENCY 40 CFR Parts 63, 264, 265, 266, and 270 [FRL-7001-9] RIN 2050-AE79 NESHAP: Standards for Hazardous Air Pollutants for Hazardous Waste Combustors—Proposed Amendments AGENCY: Environmental Protection Agency (EPA). ACTION: Proposed rule. SUMMARY: Under the Clean Air Act (CAA), EPA established new emissions standards for hazardous waste burning cement kilns, lightweight aggregate kilns, and incinerators on September 30, 1999 (NESHAP: Final Standards for Hazardous Air Pollutants for Hazardous Waste Combustors). Following promulgation of this final rule, the regulated community, through informal comments and through litigation, raised numerous issues related to specific requirements of the final rule. In response to relevant concerns, we are proposing and taking comment on certain targeted changes to the final rule. These regulatory changes do not propose to amend the numerical emission standards, but rather focus on improvements to the implementation of the emission standards, primarily in the areas of compliance, testing and monitoring. DATES: Comments must be submitted by August 17, 2001. ADDRESSES: If you wish to comment on this proposed rule, you must send an original and two copies of the comments referencing Docket Number F-2001-RC5P-FFFFF to: RCRA Information Center (RIC), Office of Solid Waste (5305G), U.S. Environmental Protection Agency Headquarters (EPA HQ), Ariel Rios Building, 1200 Pennsylvania Avenue, NW, Washington, D.C. 20460-0002; or, (2) if using special delivery, such as overnight express service: RIC, Crystal Gateway One, 1235 Jefferson Davis Highway, First Floor, Arlington, VA 22202. You may also submit comments electronically following the directions in the Supplementary Information section below. You may view public comments and supporting materials in the RIC. The RIC is open from 9 am to 4 pm Monday through Friday, excluding Federal holidays. To review docket materials, we recommend that you make an appointment by calling 703-603-9230. You may copy up to 100 pages from any regulatory document at no charge. Additional copies cost $ 0.15 per page. For information on accessing an electronic copy of the data base, see the Supplementary Information section. FOR FURTHER INFORMATION CONTACT: For general information, call the RCRA Call Center at 1-800-424-9346 or TDD 1-800-553-7672 (hearing impaired). Callers within the Washington Metropolitan Area must dial 703-412-9810 or TDD 703-412-3323 (hearing impaired). The RCRA Call Center is open Monday-Friday, 9 am to 4 pm, Eastern Standard Time. For more information on specific aspects of this proposed rule, contact Mr. Frank Behan at 703-308-8476, behan.frank@epa.gov, or write him at the Office of Solid Waste, 5302W, U.S. EPA, Ariel Rios Building, 1200 Pennsylvania Avenue, NW, Washington, D.C. 20460. SUPPLEMENTARY INFORMATION: Submittal of Comments You may submit comments electronically by sending electronic mail through the Internet to: rcra-docket@epamail.epa.gov. You should identify comments in electronic format with the docket number F-2001-RC5P-FFFFF. You must submit all electronic comments as an ASCII (text) file, avoiding the use of special characters or any type of encryption. The official record for this action will be kept in the paper form. Accordingly, we will transfer all comments received electronically into paper form and place them in the official record which will also include all comments submitted directly in writing. The official record is the paper record maintained at the RIC as described above. We may seek clarification of electronic comments that are garbled in transmission or during conversion to paper form. You should not electronically submit any confidential business information (CBI). You must submit an original and two copies of CBI under separate cover to: RCRA CBI Document Control Officer, Office of Solid Waste (5305W), U.S. EPA, Ariel Rios Building, 1200 Pennsylvania Avenue, NW, Washington, D.C. 20460. If you do not submit comments electronically, we are asking prospective commenters to voluntarily submit one additional copy of their comments on labeled personal computer diskettes in ASCII (text) format or a word processing format that can be converted to ASCII (text). It is essential that you specify on the disk label the word processing software and version/edition as well as the commenter's name. This will allow us to convert the comments into one of the word processing formats used by the Agency. Please use mailing envelopes designed to protect the diskettes. We emphasize that submission of diskettes is not mandatory, nor will it result in any advantage or disadvantage to any commenter. Acronyms Used in the Rule APCD—Air pollution control device ASME—American Society of Mechanical Engineers CAA—Clean Air Act CEMS—Continuous emissions monitors/monitoring system COMS—Continuous opacity monitoring system CFR—Code of Federal Regulations DOC—Documentation of Compliance DRE—Destruction and removal efficiency dscf—Dry standard cubic feet dscm—Dry standard cubic meter EPA/USEPA—United States Environmental Protection Agency gr—Grains HAP—Hazardous air pollutant HWC—Hazardous waste combustor MACT—Maximum Achievable Control Technology NESHAP—National Emission Standards for HAPs ng—Nanograms NIC—Notice of Intent to Comply NOC—Notification of compliance OPL—Operating parameter limit PM—Particulate matter POHC—Principal organic hazardous constituent ppmv—Parts per million by volume RCRA—Resource Conservation and Recovery Act TEQ—Toxicity equivalence Table of Contents Part One: Overview and Background for This Proposed Rule I. What Is the Purpose of This Proposed Rule? II. What Is the Phase I Rule? III. What Related Actions Have Been Taken Since Publication of the Phase I Rule? IV. How Can I Influence EPA's Thinking on this Rule? Part Two: NESHAP—Proposed Amendments to the HWC Final Rule I. Definition of Research, Development, and Demonstration Sources II. Identification of an Organics Residence Time that is Independent of and Shorter than the Hazardous Waste Residence Time III. Controls on APCDs after the Hazardous Waste Residence Time Has Expired IV. Instantaneous Monitoring of Combustion Zone Pressure V. Operator Training and Certification VI. Bag Leak Detection System VII. Time Extensions For Performance Testing if the Test Plan Has Not Been Approved VIII. Flexibility in Operations During Confirmatory Performance Testing for Dioxin/Furan IX. Waiving Operating Parameter Limits during Performance Testing X. Method 23 as an Alternative to Method 0023A for Dioxin/Furans XI. Calibration Requirements for Thermocouples XII. Alternative Approach to Establish Operating Parameter Limits XIII. Extrapolation of Operating Parameter Limits XIV. Limit on Minimum Combustion Chamber Temperature for Cement Kilns XV. Revisions to Operating Requirements for Activated Carbon Injection and Carbon Bed Systems XVI. Clarification of Requirements to Confirm Carbon Bed Age XVII. Revisions to Operating Parameter Limits for Wet Scrubbers XVIII. Reproposal of kVA Limits for Electrostatic Precipitators and Request for Comment on Approaches to Ensure Baghouse Performance XIX. How to Comply Temporarily with Alternative, Otherwise Applicable MACT Standards XX. RCRA Permitting Requirements for Sources Entering the RCRA Process Post-Rule Promulgation Part Three: Analytical and Regulatory Requirements I. Executive Order 12866 II. Regulatory Flexibility Act (RFA), as Amended by the Small Business Regulatory Enforcement Fairness Act of 1996 (SBREFA), 5 U.S.C. 601 et. seq. III. Executive Order 13045: “Protection of Children from Environmental Health Risks and Safety Risks” IV. Environmental Justice Executive Order 12898 V. Unfunded Mandates Reform Act VI. Executive Order 13132 (Federalism) VII. Consultation with Tribal Governments VIII. Paperwork Reduction Act IX. National Technology Transfer and Advancement Act of 1995 Part Four: State Authority Part One: Overview and Background for This Proposed Rule I. What Is the Purpose of This Proposed Rule? Today's notice proposes specific changes to the NESHAP: Final Standards for Hazardous Air Pollutants for Hazardous Waste Combustors (Phase I) rule, published September 30, 1999 (64 FR 52828). After promulgation, commenters (primarily the regulated community) raised numerous potential issues through informal comments and during litigation settlement discussions. After considering the issues raised, we have decided to propose for comment twenty amendments to the final rule, most of the proposed changes relating to compliance and implementation of the rule. The ability of facilities to meet the September 30, 2002 compliance date may be dependent upon when these proposed changes are made final. While we expect to complete the rulemaking process and publish final amendments in a timely manner, we request comments on how the timing of these rule changes could impact compliance. In addition, we solicit comments on solutions to address compliance problems should they arise (e.g., use of § 63.1206(b)(4) to obtain an extension of compliance with the emission standards of up to one year). In the “Rules and Regulations” section of the Federal Register , we are taking direct final action on thirteen additional amendments to the Phase I rule. If you wish to comment on those amendments, you must submit comments following the directions in the Addresses section of that action. The remaining sections of this part provide additional background information on the Phase I final rule. II. What Is the Phase I Rule? In the Phase I final rule, we adopted National Emissions Standards for Hazardous Air Pollutants to control toxic emissions from the burning of hazardous waste in incinerators, cement kilns, and lightweight aggregate kilns. These emission standards created a technology-based national cap for hazardous air pollutant emissions from the combustion of hazardous waste in these devices. Additional risk-based conditions necessary to protect human health and the environment may be imposed (assuming a proper, site-specific justification) under section 3005(c)(3) of the Resource Conservation and Recovery Act (RCRA). Section 112 of the CAA requires emissions standards for hazardous air pollutants to be based on the performance of the Maximum Achievable Control Technology (MACT). These standards apply to the three major categories of hazardous waste burners—incinerators, cement kilns, and lightweight aggregate kilns. For purposes of today's proposal, we refer to these three categories collectively as hazardous waste combustors (HWC). Hazardous waste combustors burn about 80% of the hazardous waste combusted annually within the United States. The Phase I HWC MACT standards are expected to achieve significant reductions in the amount of hazardous air pollutants being emitted each year. Additionally, the Phase I HWC MACT rule satisfies our obligation under RCRA (the main statute regulating hazardous waste management) to ensure that hazardous waste combustion is conducted in a manner protective of human health and the environment. By using both CAA and RCRA authorities in a harmonized fashion, we consolidate regulatory control of hazardous waste combustion into a single set of regulations, thereby minimizing the potential for conflicting or duplicative federal requirements. More information on the Phase I HWC MACT rule is available electronically from the World Wide Web at www.epa.gov/hwcmact. III. What Related Actions Have Been Taken Since Publication of the Phase I Rule? On November 19, 1999, we issued a technical correction to the Phase I HWC MACT final rule (64 FR 63209). It clarified our intent with respect to certain aspects of the Notification of Intent to Comply and Progress Report requirements of the 1998 “Fast Track” final rule (63 FR 33783). Additionally, specific to the Phase I HWC MACT final rule, we corrected several typographical errors and omissions. On July 10, 2000, we issued a second technical correction to the Phase I HWC MACT final rule (65 FR 42292). This action corrected additional typographical errors and clarified several issues to make the Phase I rule easier to understand and implement. This action also supplied one omission from the technical correction published on November 19, 1999, and made one correction to the related June 19, 1998 “Fast Track” final rule (63 FR 33783). On July 25, 2000, the Court of Appeals for the District of Columbia decided Chemical Manufacturers Association v. EPA, 217 F. 3d 861 (D.C. Cir. No. 99-1236). The court held that EPA had the legal authority to promulgate a requirement of early cessation of hazardous waste burning activity for those sources not intending to comply with the MACT emission standards. However, the court also held that we had not adequately explained our reasons for imposing the early cessation requirement. As a result, the court vacated the early cessation requirement and the related Notice of Intent to Comply (NIC) and Progress Report requirements. This vacature took effect on October 11, 2000. Since the requirements were not vacated until after sources were required to submit their NICs (on October 2, 2000), we determined that the court's action does not impact a source's ability to request a RCRA permit modification using the streamlined procedures of 40 CFR 270.42(j)(1). As long as a source complied with the NIC provisions (including filing the NIC before the provision was vacated), the source has met the requirements in 40 CFR 270.42(j)(1) and is therefore eligible for the streamlined RCRA permit modification process. The court's decision does not impact the emission standards or compliance schedule for the other requirements of the HWC NESHAP Subpart EEE. On November 9, 2000, we issued a third technical correction to the Phase I HWC MACT final rule (65 FR 67268). It clarified our intent with respect to the applicability of new source versus existing source standards for hazardous waste incinerators. This action also clarified three issues to make the Phase I rule easier to understand and implement. On May 14, 2001, we issued a final rule implementing two court orders that removed affected provisions of the Phase I HWC MACT final rule from the Code of Federal Regulations (66 FR 24270). This action removed the Notice of Intent to Comply provisions (discussed above) and certain operating parameter limits of baghouses and electrostatic precipitators. IV. How Can I Influence EPA's Thinking on This Rule? In developing this proposal, we tried to address the concerns of all our stakeholders. Your comments will help us improve this rule. We invite you to provide different views on options we propose, new approaches we haven't considered, new data, how this rule may effect you, or other relevant information. We welcome your views on all aspects of this proposed rule. Your comments will be most effective if you follow the suggestions below: • Explain your views as clearly as possible and why you feel that way. • Provide solid technical and cost data to support your views. • If you estimate potential costs, explain how you arrived at the estimate. • Tell us which parts you support, as well as those you disagree with. • Provide specific examples to illustrate your concerns. • Offer specific alternatives. • Refer your comments to specific sections of the proposal, such as the units or page numbers of the preamble, or the regulatory sections. • Make sure to submit your comments by the deadline in this notice. • Be sure to include the name, date, and docket number with your comments. Part Two: NESHAP—Proposed Amendments to the HWC Final Rule I. Definition of Research, Development, and Demonstration Sources Section 63.1200, Table 1, exempts research, development, and demonstration sources from the Part 63, Subpart EEE, hazardous waste combustor MACT standards. 1 We explained at promulgation that the hazardous waste combustor emission standards and compliance assurance requirements may not be appropriate for these sources because of their typically intermittent operations and small size. See 64 FR at 52839. 1 Hazardous waste research, development, and demonstration sources remain subject to RCRA permit requirements under § 270.65. See 64 FR at 52839. The rule defines research, development, or demonstration sources as those sources engaged in laboratory, pilot plant, or prototype demonstration operations: (1) Whose primary purpose is to conduct research, development, or short-term demonstration of an innovative and experimental hazardous waste treatment technology or process; and (2) where the operations are under the close supervision of technically-trained personnel. Stakeholders express concern that the definition of demonstration source and the provision to allow unlimited one-year time extensions to the exemption may result in commercial, production sources taking inappropriate advantage of the exemption. We request comment on approaches to preclude inappropriate use of the exemption for demonstration sources. Approaches that we are considering include: (1) Clearly distinguishing between research and development sources versus demonstration sources, and limiting the exemption for demonstration sources to one year or less; or (2) requiring documentation of how a source's demonstration of an innovative or experimental hazardous waste treatment technology or process is different from the waste management services provided by a commercial hazardous waste combustor. II. Identification of an Organics Residence Time That Is Independent of and Shorter Than the Hazardous Waste Residence Time “Hazardous waste residence time” is defined at § 63.1201(a) as the time elapsed from cutoff of the flow of hazardous waste into the combustor (including, for example, the time required for liquids to flow from the cutoff valve into the combustor) until solid, liquid, and gaseous materials from the hazardous waste, excluding residues that may adhere to combustion chamber surfaces, exit the combustion chamber. As stakeholders recognize, hazardous waste residence time has significant regulatory and enforcement implications. For example, if a source were to exceed an operating requirement or emission standard after the hazardous waste residence time has expired, it is not a violation if the exceedance occurred during start-up or shut-down, or because of a malfunction provided that the source follows the procedures and corrective measures prescribed in the start-up, shut-down, and malfunction plan. In addition, after the hazardous waste residence time has expired, sources may elect to comply with emission standards the Agency has promulgated under sections 112 and 129 of the Clean Air Act for source categories that do not burn hazardous waste in lieu of the hazardous waste combustor standards of Subpart EEE, Part 63. See § 63.1206(b)(1). 2 2 As discussed in Section XIX, if sources elect to comply temporarily with alternative section 112 or 129 MACT standards after the hazardous waste residence time has expired, sources nonetheless remain an affected source only under Subpart EEE for hazardous waste combustors. Since promulgation of the hazardous waste combustor rule, stakeholders have raised the issue of whether a hazardous waste organics residence time should be defined that is independent of and shorter than the bulk solids residence time. Industry stakeholders recommend an approach to calculate a hazardous waste organics residence time that defines when organic constituents in solid materials have been destroyed. 3 Although the concept has merit, several issues must be addressed prior to revising the rule to allow sources to petition the Administrator for case-by-case determinations of an organics residence time. We therefore are not proposing a change at this time but are requesting comment on the concept and implementation of an organic residence time. 3 Email from David Case, Environmental Treatment Council, to Bob Holloway, EPA, with attachment entitled “Proposed Method for Calculation of Hazardous Constituents Retention Time,” dated June 7, 2000. As contemplated by stakeholders, the hazardous waste organics residence time would be independent of and considerably shorter than the bulk hazardous waste residence time discussed above. As with the bulk hazardous waste residence time, an organics residence time would have significant regulatory and enforcement implications. After the hazardous waste organics residence time has expired, an exceedance of the carbon monoxide or hydrocarbon emission standard or an operating parameter limit associated with the destruction and removal efficiency (DRE) or dioxin/furan emission standards would not be a violation if the exceedance occurred during start-up or shut-down or were caused by a malfunction and sources comply with the procedures and corrective measures prescribed in the start-up, shut-down, and malfunction plan. In addition, it seems appropriate to allow sources to elect to comply with standards the Agency has promulgated under sections 112 or 129 of the Clean Air Act to control organic emissions for source categories that do not burn hazardous waste in lieu of the hazardous waste combustor standards of Subpart EEE, Part 63. As discussed in Section III below, however, providing only a partial transition from the hazardous waste combustor MACT standards of Subpart EEE may be problematic. A. What Is the Approach Stakeholders Recommend to Calculate Hazardous Waste Organics Residence Time? Stakeholders suggest that a hazardous waste organics residence time can be calculated as the sum of: (1) The time for the solid matrix containing the organic constituents to reach the target temperature required to destroy the organics; (2) the time for the organic constituent to be destroyed at the target temperature; and (3) the time for the gas to pass through the combustion chamber and exit the air pollution control system. The time required for the organic constituents within the solid matrix to reach the target temperature would be calculated using standard heat transfer equations which are available in chemical engineering references. 4 Stakeholders state that these equations can be applied to various materials, assuming the thermal conductivity of the material. These equations also can be applied easily to various geometries, such as a 55 gallon drum (right circular cylinder), or to irregular shaped items resulting from shredder feed. 4 Geankoplis, C.J., “Transport Processes and Unit Operations,” Chapters 3 and 4, Allyn and Bacon, Inc., Boston, 1978. Stakeholders state that once the solid is at the target temperature the time for the hazardous constituent to be destroyed can be calculated using equations that are readily available from Dr. Dellinger's work on developing the low oxygen thermal stability index for hazardous organic compound incinerability. 5 Using Dellinger's kinetic models under low oxygen conditions, the destruction time for hazardous constituents can be calculated. 5 B. Dellinger, et al, “Development of a Thermal Stability Based Index of Hazardous Waste Incinerability,” University of Dayton Research Institute Final Report Under EPA Cooperative Agreement CR-813938, November 15, 1991. Also, B. Dellinger, et al, “Development of a Thermal Stability Based Index of Hazardous Organic Compound Incinerability,” Environmental Science and Technology, 24, p.316, March 1990. To implement this approach to calculate a hazardous waste organics residence time, stakeholders suggest that sources should include the retention time evaluation and calculations in a report developed by an independent Professional Engineer with combustion engineering expertise. Sources would submit the report to the Administrator for review and approval. B. How Would Site-Specific Factors Be Addressed? Stakeholders state that the general approach can be readily applied to various scenarios as necessary on a site-specific basis. Stakeholders have considered how some scenarios could be addressed, as discussed below, and believe that approaches to address other scenarios would become apparent as the approach is applied to the site-specific situation. 1. How Would Various Geometric Shapes and Sizes of Solids Be Addressed? Stakeholders acknowledge that an incineration process can have several types of solid feed such as bulk solids, direct drum feed in various sizes, shredded waste feed, and other mechanisms. Each of these solid feed scenarios can be evaluated for the heat transfer step by assigning an appropriate geometry to the solid for use in the heat flux equations. Heat transfer will take place more rapidly in shredded waste feed, in which the particle size of the solids is reduced. At the other extreme is a monolith in a 55 gallon drum, which will require a longer time for the center point to reach the target temperature. The center point of the monolith can be considered the point where the organic constituent is located for ensuring a worst case for the heat transfer step. Site-specific feed can be modeled by evaluating the actual geometry and size of solid feed and post-shredder feed. 2. How Would the Thermal Conductivity of the Solid Be Determined? The time for the solid mass to reach temperature will depend on the thermal conductivity of the solid mass. The thermal conductivity is a key parameter in the heat transfer equation. The types of solid feed managed at a particular site can be used to select a worst case material for thermal conductivity. Stakeholders present as an example a facility that feeds certain polymeric monolithic materials in 55 gallon drums. Certain polymers may have a low thermal conductivity that can be used as a worst case. References such as Perry's Chemical Engineering Handbook can be consulted to provide a range of thermal conductivities for consideration. For example, stakeholders note that polypropylene has one of the lower thermal conductivities of 0.08 BTU/hr-sq.ft.-°F (see Table 23-10 in Perry's Handbook). This might provide a good worst case value to use for the solid mass thermal conductivity for this source. 3. How Would a Worst-Case Organic Constituent Be Selected? Stakeholders suggest that a worst-case hazardous organic constituent could be selected on the basis of its ranking in various incinerability indices, just as principal organic hazardous constituents (POHCs) are selected for demonstrating destruction and removal efficiency (DRE). A constituent that ranks high in both the heat of combustion and low oxygen thermal stability indices could be used. In addition, a few compounds with complex structures that would be expected to yield various decomposition byproducts could be modeled. Examples of such compounds are pentachlorophenol, perchloroethylene, and certain pesticides. Stakeholders suggest that Dr. Dellinger's work, cited above, can be consulted to select additional worst case constituents. 4. How Would the Target Destruction Temperature Be Selected? Stakeholders suggest that target destruction temperatures can be selected based on the kinetic studies of Dr. Dellinger. Stakeholders state that Dellinger has generally found that any organic chemical and its organic byproducts can be completely destroyed at 800°C. 6 Also, the range of destruction temperatures published by Dellinger can be consulted to select a target temperature on a site-specific basis for the types of wastes that are managed. 6 See Attachment 5 of ETC's Comments to the MACT Rule, Docket F-96-RCSP-FFFFF, filed August 19, 1996. 5. How Would Paralytic and Starved Air Conditions Be Addressed? Stakeholders acknowledge that certain solid geometries may result in the organic constituent being isolated from combustion air, such that pyrolytic conditions must be assumed. Nonetheless, stakeholders state that a destruction time can still be calculated and the low oxygen conditions can be incorporated into the kinetic model. Dellinger has published such calculations in developing the low oxygen thermal stability index for incinerability. 7 Stakeholders state that pyrolytic conditions would likely be required to be assumed for monolithic feed. They note that shredder-feed, however, substantially reduces the particle size of the solid feed, and mixing with combustion air is achieved. 7 Dellinger, B. et al, “PIC Formation Under Pyrolytic and Starved Air Conditions,” EPA Publication No. EPA/600/S2-86/006, July 1986. 6. How Would Heat Sink and Other Heat Consuming Factors Be Addressed? Stakeholders acknowledge that other factors in a given waste may consume energy, requiring another step or two to the retention time calculation. For example, a solid waste monolith that is a low melting point solid will go through a melting transition that will consume heat before the temperature of the mass rises past the transition point. Stakeholders state that this step can be easily added to the retention time calculation, if necessary. Similarly, a waste may contain a pocket of water or other low boiling point material, and a step for enthalpy of vaporization may need to be added. Stakeholders note that these calculations can also be performed as a form of sensitivity analysis to determine how conservative the retention time calculation is. B. What Are the Unresolved Issues About Stakeholders' Recommended Approach? We acknowledge that the residence time for organic constituents in a solid matrix is generally less than the residence time for the bulk hazardous waste residue. Thus, ideally, sources should be eligible for the reduced regulatory and enforcement burden discussed above once the organics residence time has expired. One promising feature of the stakeholders' approach is that it would conservatively predict how long it takes the waste monolith to heat up to volatilize the organic constituent. We are concerned, however, that their approach does not explicitly address how long it would take for: (1) The generated volatiles or their potentially toxic intermediates to diffuse to the surface of the monolith where oxygen is present for destruction; or (2) alternatively for oxygen in the bulk gas to diffuse from the surface of the monolith to reach the volatiles. In lieu of accounting for the time required to destroy organic constituents under oxidative conditions, stakeholders' approach assumes that destruction would occur within solid matrices under pyrolytic conditions. Further, stakeholders believe that calculations developed by Dellinger while developing the low oxygen thermal stability index could be used to model those pyrolytic reactions. We request comments on whether Dellinger's work on low oxygen destruction would adequately model destruction under the pyrolytic conditions that occur within solid matrices, with respect to either the time required for destruction of the initial organic constituent or the types of intermediates that would be formed and the time required to destroy the intermediates. Finally, we request comments on whether it is practicable to perform valid engineering calculations for multiple waste streams that are not homogenous and that contain multiple organic constituents of concern. We request comment on stakeholders' approach to calculate an organics residence time and specifically whether it can be revised to address our concerns. III. Controls on APCDs After the Hazardous Waste Residence Time Has Expired For sources equipped with a dry particulate matter control device, we propose to maintain the semivolatile metal, low volatile metal, and particulate matter 8 emission standards and the associated particulate matter control device operating requirements after the hazardous waste residence time has expired and until the control device undergoes a complete cleaning cycle (e.g., for all compartments of a baghouse; for all fields of an electrostatic precipitator). 9 For sources equipped with activated carbon injection, the dioxins/furans and mercury emission standards would also continue to apply after the hazardous waste residence time has expired until the control device undergoes a complete cleaning cycle. 8 The particulate matter emission standard is included because particulate matter is a surrogate for metal hazardous air pollutants other than mercury, the enumerated semivolatile metals, and the enumerated low volatile metals. 9 If sources comply with the semivolatile and low volatile metal emission standards without emissions testing by assuming all metals in feedstreams are emitted, and therefore do not rely on the particulate matter control device to comply with the emission standards for these metals, the proposed requirements to maintain compliance with the semivolatile and low volatile metals emission standards and control device operating parameter limits would not apply. A. What Concern Would This Requirement Address? Dry particulate matter emission control devices such as electrostatic precipitators and baghouses retain collected particulate matter in the device until the sections of the device are cleaned sequentially, e.g., rapping of electrostatic precipitator plates, bag cleaning. This retained particulate matter contains metal hazardous air pollutants other than mercury due to its volatility. In addition, if the source is equipped with activated carbon injection, the collected particulate matter also will contain mercury and dioxins/furans. Our concern is that these pollutants could be emitted at levels exceeding the hazardous waste combustor emission standards after the hazardous waste residence time has expired. After the hazardous waste residence time has expired, sources may choose to comply with MACT standards the Agency has promulgated under sections 112 and 129 of the Clean Air Act for source categories that do not burn hazardous waste in lieu of the Subpart EEE standards. See § 63.1206(b)(1). 10 If sources choose to comply with those otherwise applicable MACT standards, we are concerned that these standards may not adequately ensure that the hazardous waste-derived pollutants remaining in the dry particulate matter control device are controlled to the level required by the hazardous waste combustor rules of Subpart EEE. For example, if the alternative particulate matter standard were substantially higher than the hazardous waste combustor MACT standard, sources may be able to operate the control device under less stringent operating levels (e.g., lower power to a field of an electrostatic precipitator) which could cause the accumulated particulate matter (containing hazardous waste-derived pollutants) retained within the device to be reentrained in the stack gas. This could allow hazardous waste-derived pollutants to be emitted at levels exceeding the hazardous waste combustor MACT emission standards. When the particulate matter control device undergoes a complete cleaning cycle, the accumulated hazardous waste-derived pollutants are removed, thus precluding an exceedance of the hazardous waste combustor emission standards. 10 As discussed in Section XIX of the text, if sources choose to comply with otherwise applicable section 112 or 129 requirements (e.g., Subpart LLL for cement kilns) after the hazardous waste residence time has expired, sources remain an affected source under Subpart EEE only. Sources would comply with those otherwise applicable MACT standards under an alternative mode of operation that sources would specify under § 63.1209(q). B. Is It Necessary To Require Continued Compliance With the Limit on Gas Temperature at the Inlet to the Dry Particulate Matter Control Device? We considered whether increasing the gas temperature at the inlet to the dry particulate matter control device, absent a requirement to maintain the temperature operating limit, could cause hazardous waste-derived semivolatile metals (and mercury and dioxins/furans if sources use activated carbon injection) contained in the accumulated particulate matter to volatilize and be emitted at levels that exceed an emission standard. We initially conclude that, absent a gas inlet temperature limit, gas temperatures are not likely to increase to the levels necessary to volatilize enough semivolatile metals to cause an exceedance of the emission standards. This is particularly true if we consider that many sources should be able to complete a cleaning cycle of their electrostatic precipitator or baghouse within 30 minutes after the hazardous waste residence time has expired. We are concerned, however, that, for sources equipped with activated carbon injection, increases in inlet gas temperatures above the operating limit may potentially revolatilize captured mercury and dioxins/furans. We request comment on the extent to which mercury and dioxins/furans may revolatilize and be emitted because of increased gas temperatures in the short period after the hazardous waste residence time has expired and the cleaning cycle for the dry particulate matter control device has been completed. C. Would the Proposed Requirement Increase Cleaning Cycle Frequency and Potentially Emissions of Hazardous Air Pollutants? As discussed above, we propose to require continued compliance with the semivolatile metal, low volatile metal, and particulate matter standards (and the dioxin/furan and mercury standards if sources use carbon injection) after the hazardous waste residence time has expired until the dry particulate matter control device undergoes a complete cleaning cycle. However, we are concerned, that unless additional controls are applied, this requirement could potentially result in an increase in particulate emissions and associated hazardous air pollutants. When a dry particulate matter control device is cleaned, collection efficiency is temporarily degraded. For example, when the plates in the last field of an electrostatic precipitator are rapped, some of the resuspended particulate matter is unavoidably emitted. For baghouses, when the bags are cleaned, typically using a pulse of air, the collection efficiency of the cleaned bags is reduced until a layer of particulate matter reforms on the bags. Thus, increasing the cleaning frequency of a baghouse decreases its collection efficiency. To comply with the proposed requirement to clean the particulate matter control device before the Subpart EEE metals and particulate matter standards are waived in lieu of other standards, sources may want to initiate a cleaning cycle immediately after the hazardous waste residence time has expired. Further, they may want to restart the timing of the cleaning cycle beginning with the cleaning that occurs after the hazardous waste residence time has expired. Increasing the cleaning cycle frequency could potentially result in an exceedance of the emission standards, however, if compliance with the standards has not been demonstrated during performance testing at that cleaning cycle frequency. To ensure that the emission standards are not exceeded due to increased cleaning cycle frequency, sources may not increase the cleaning cycle frequency beyond the frequency used during the comprehensive performance test. D. How Would This Requirement Be Implemented? If sources elect to comply temporarily with the otherwise applicable section 112 or 129 Clean Air Act standards after the hazardous waste residence time has expired, sources would remain subject to certain Subpart EEE standards and associated compliance requirements until sources completed a cleaning cycle of the dry particulate matter control device: Particulate matter, semivolatile metals, low volatile metals, and, if sources use activated carbon injection, dioxin/furan and mercury. Given that sources remain an affected source only under Subpart EEE when sources elect to comply temporarily with otherwise applicable MACT standards, sources would identify this operating scenario as an alternative mode of operation under § 63.1209(q). 11 Consequently, sources would specify the applicable emission standards and compliance requirements for this alternative mode of operation as: (1) Those standards and compliance requirements of Subpart EEE that remain in effect; and (2) those otherwise applicable standards and compliance requirements established under section 112 or 129 (e.g., Subpart LLL for cement kilns). If an otherwise applicable section 112 or 129 standard or compliance requirement were more stringent than a Subpart EEE standard or compliance requirement that remains in effect, sources would comply with the more stringent standard or compliance requirement. 11 See Section XIX below in the text for additional discussion on using § 63.1209(q) to specify operations under otherwise applicable section 112 or 129 MACT standards. Exceedance of a Subpart EEE operating parameter limit (OPL) for a dry particulate matter control device after the hazardous waste residence time has expired but before a cleaning cycle of the device has been completed would be evidence of failure to maintain compliance with the Subpart EEE emission standards. Given that the hazardous waste residence time has expired, however, the exceedance need not be considered for the excessive exceedance reporting requirement under § 63.1206(c)(vi). Similarly, if the exceedance occurs because of a malfunction, the exceedance would not be evidence of failure to maintain compliance with an emission standard if the source followed the corrective measures prescribed in its startup, shutdown, and malfunction plan. Thus, the consequences of an exceedance would be the same after the hazardous waste residence time has expired whether the exceedance occurs before or after the cleaning cycle has been completed if the source chose to continue to comply with the Subpart EEE emission standards (i.e., in lieu of otherwise applicable MACT standards under a different mode of operation). Having equivalent consequences of an exceedance of an OPL after the hazardous waste residence time has expired irrespective of whether the cleaning cycle has been completed is appropriate. Our objective is simply to ensure that the Subpart EEE OPLs for the dry particulate matter control device are maintained until the cleaning cycle is completed to minimize emissions of hazardous waste-derived HAPs to below the Subpart EEE emissions standards. Our intent is not to penalize a source for exceedances that may be attributable to unavoidable malfunctions after the source has taken the preventative measures to minimize emissions of HAPs by cutting off the hazardous waste feed and allowing the hazardous waste residence time to expire. Some stakeholders have expressed initial concern with the technical feasibility of these proposed requirements. We will be considering these concerns prior to issuing a final rule. IV. Instantaneous Monitoring of Combustion Zone Pressure The final rule requires sources to control combustion system leaks by either: (1) Keeping the combustion zone sealed; (2) maintaining the maximum combustion zone pressure lower than the ambient pressure measured using an instantaneous monitor; or (3) using an alternative means to provide control of system leaks. See §§ 63.1201(a), 63.1206(c)(5)(ii), and 63.1209(p). The rule defines an “instantaneous monitor” as one that continuously samples, detects and records the regulated parameter without use of an averaging period. In today's notice, we propose to revise the combustion system leak requirements to better clarify the intent of this provision, and we are taking comment on whether we should allow sources to average pressure readings over short periods of time when demonstrating that their combustion system is maintained below ambient pressure. After publication of the final rule, stakeholders expressed concern that the requirement to maintain the combustion zone pressure lower than ambient pressure (option 2 above) could result in an overly prescriptive requirement. Stakeholders believe this regulatory language can be interpreted to require sources to monitor and record combustion zone pressure at a frequency of every 50 milliseconds. 12 Stakeholders state such an interpretation would be problematic because of the enormous number of data points that must be recorded and because such a frequent monitoring frequency would greatly increase the number of automatic waste feed cutoffs. Stakeholders also requested that we clarify that combustion system leaks refer to fugitive emissions resulting from the combustion of hazardous waste, and not fugitive emissions that originate from nonhazardous process streams (e.g., the clinker product at a cement kiln). 12 The final rule preamble states that typical pressure transducers in use today are capable of responding to pressure changes once every 50 milliseconds. See 64 FR 52920. After careful review of the regulatory language and after considering our original intent, we agree that the final rule is ambiguous and may be conservatively interpreted to require sources to monitor and record combustion zone pressure at a frequency of every 50 milliseconds. Therefore, in today's notice, we clarify that our intent is to require sources to use a pressure monitor and recording frequency that is adequate to detect combustion system leak events. We also clarify that the intent of the combustion system leak requirement is to prevent fugitive emissions from the combustion of hazardous waste, not fugitive emissions that originate from nonhazardous process streams. To make these clarifications, we propose to modify the § 63.1201(a) definition of an instantaneous pressure monitor to read as follows: “Instantaneous monitoring for combustion system leak control means detecting and recording pressure without use of an averaging period, at a frequency adequate to detect combustion system leak events from hazardous waste combustion ” (emphasis added). 13 We also propose to revise the § 63.1209(p) automatic waste feed cutoff regulatory language to read as follows: “If you comply with the requirements for combustion system leaks under § 63.1206(c)(5) by maintaining the maximum combustion chamber zone pressure lower than ambient pressure to prevent combustion system leaks from hazardous waste combustion, sources must perform instantaneous monitoring of pressure and the automatic waste feed cutoff system must be engaged when negative pressure is not maintained” (emphasis added). 13 Note that this newly proposed definition removes the word “sampling” from the definition of instantaneous pressure monitor because a pressure monitor is not thought to physically withdraw a combustion gas sample. We do not specify the monitoring and recording frequencies in the regulations, however, because sources differ in design and operation such that different monitoring and recording frequencies may be needed to ensure that fugitive emissions do not occur. Rather, sources and permit officials should determine on a site-specific basis what frequency of monitoring and recording would be appropriate. Each source should describe in the comprehensive performance test workplan and Notification of Compliance how their compliance method will ensure that fugitive emissions will not occur. We propose that this description specify the monitoring and recording frequency and how the monitoring approach will be integrated into the automatic waste feed cutoff system. Stakeholders also suggest that we allow averaging of the pressure readings over short periods of time, e.g., a 5-second rolling average updated every second, in demonstrating the combustion system is maintained below ambient pressure. Averaging of pressure readings is less stringent than the current final rule instantaneous monitoring requirements. We request comment on whether such a monitoring approach is appropriate, and specifically, whether averaging pressure readings can adequately detect pressure excursion events that result in combustion system leaks. V. Operator Training and Certification On July 10, 2000, we issued a technical correction to the operator training and certification requirements of § 63.1206(c)(6) to clarify which employees are subject to the training and certification requirements and to note that the training and certification program should be tailored to the responsibilities of the employee. See 65 FR at 42295. Subsequent to this technical correction, incinerator stakeholders raised concerns about the requirement for incinerator control room operators and shift supervisors to be trained and certified under the American Society of Mechanical Engineers (ASME) Standard Number QHO-1-1994. Although the rule allows incinerator control room operators to be trained and certified under either a State program or ASME's program, stakeholders note that they are required to use the ASME program because there are no State programs at this time. Stakeholders raise the following concerns: (1) The scope of the ASME training and certification program is too broad; (2) the ASME certification program is problematic for new sources and newly hired operators because it requires 6 months of operating experience at the source before full certification may be awarded; (3) the ASME control room operator training and certification program is not necessary for shift supervisors; and (4) the ASME training and certification program cannot be implemented by the regulatory compliance date. We provide below our reasons for preferring the ASME training and certification program over site-specific, source-implemented programs, but acknowledge stakeholders' concerns that the program may be more comprehensive than necessary to ensure compliance with the requirements of Subpart EEE. Accordingly, we propose to allow incinerator control room operators to be trained and certified under: (1) A site-specific, source-developed and implemented program; (2) the ASME program; or (3) a State program. We also conclude that it may be difficult for sources that choose to use the ASME program to fully certify their control room operators by the compliance date. Therefore, we propose to require only provisional certification by the compliance date for such sources. In addition, for sources that choose to use the ASME program, only provisional certification would be required for new employees and employees at new facilities prior to their assuming duties. Finally, we propose that control room operator training and certification is not necessary for shift supervisors to help ensure that the source operates within the limits established under the rule and that emissions of hazardous air pollutants are minimized. A. How Do We Address Concerns About the ASME Training and Certification Program? 1. Is the Scope of the ASME Program Too Broad? Incinerator stakeholders state that the scope of the ASME training and certification program for incinerator control room operators is too broad to apply generically to all control room operators. They prefer a tailored, site-specific, source-developed and implemented training and certification program. The ASME program requires that control room operators be trained and certified to ensure a broad knowledge of operational, preventive maintenance, safety procedures, and practices for various types of incineration systems, emission control systems, and continuous emissions monitoring systems. Incinerator stakeholders state that there is no obvious benefit of requiring a broad knowledge of incineration issues; knowledge of only the equipment and operations at the operator's site are important. They question the benefit of, for example, an operator of a small liquid waste incinerator equipped with a wet scrubber knowing how to operate a rotary kiln incinerator equipped with a baghouse. They note further that it is unnecessarily time-consuming and stressful for operators that are unfamiliar with equipment they have never operated to undergo a rigorous training and certification program for that equipment. In addition, they note that the ASME standard was developed as a voluntary standard. Finally, they note that cement kiln and lightweight aggregate kiln control room operators may be trained and certified under a site-specific program. The ASME program is comprised of a broad training curriculum that is implemented by each source followed by a provisional certification that is administered by ASME. Provisional certification is awarded after the operator passes a comprehensive, generic written test addressing operations of various types of incinerators and control systems. Operators with provisional certification may apply to ASME for full certification. Full certification is awarded after passing an on-site, site-specific oral examination. We continue to believe that a broad training and certification program can be beneficial. A broad training program may enable control room operators to recommend modifications to existing equipment or make recommendations for new equipment, which may reduce HAP emissions. In addition, certification under a broad training curriculum would avoid the retraining and recertification that would be required if the source modifies the design or operation of the unit in a manner that could affect compliance with the emission standards and operating requirements of Subpart EEE. Nonetheless, we agree with incinerator stakeholders that the broad scope of the ASME program may not be necessary to ensure compliance with the provisions of Subpart EEE. Accordingly, we propose to allow sources to use site-specific, source-developed and implemented training and certification programs, as discussed under Section B below. 2. Full Certification Under the ASME Program Cannot Be Achieved by the Compliance Date The rule currently requires full ASME (or State) certification by the compliance date. We agree with stakeholders that this is not workable because ASME does not have the resources to implement the site-specific oral examination requisite for full certification by the compliance date. After passing the written examination and achieving provisional certification, control room operators must apply to ASME for the oral examination. Stakeholders indicate that it will take one half day per control room operator to administer the site-specific oral examination. For many facilities, the ASME oral examination team 14 will require approximately one week, including travel time, to administer the exam to all control room operators. Although ASME may train several examination teams, it is unlikely that full certification examinations can be implemented at all 149 hazardous waste incineration facilities prior to the compliance date. 14 The examination team is comprised of representatives from ASME, the hazardous waste industry, the operator's facility, and/or the regulatory agency or jurisdictional authority applicable to the facility. To address this concern, we propose to require only provisional certification by the compliance date for sources that choose to use the ASME certification program. In addition, the operator would be required to submit an application to ASME for full certification and be scheduled for the certification examination. Finally, the operator would be required to achieve full certification within 1 year of the compliance date. We hope that providing this flexibility in the deadline for full certification will encourage use of the ASME program. We specifically request comment on whether the proposed deadlines for implementing the ASME certification program are appropriate. 3. Requiring Six Months of Operating Experience at the Source Before Full Certification Is Problematic The ASME standard requires that control room operators have six months of operating experience at the source before they can be fully certified. This is a problem for new sources and for newly hired operators. We propose to preclude this problem by requiring only provisional certification before operators at new sources and newly hired operators could assume their duties. Also, we would require that provisionally certified operators apply to ASME for, and be scheduled for, full certification before they assume their duties. In addition, we would require that they achieve full certification within one year of assuming their duties. This will ensure that full certification will be achieved in a timely manner. B. What Would Be the Requirements for Site-Specific, Source-Developed and Implemented Training and Certification Programs? Under today's proposed rule, a source could choose to develop and implement a site-specific training and certification program in lieu of the ASME program or a State program. Certification under a site-specific program would be required by the compliance date given that the source will implement both the training and certification (i.e., written examination at a minimum). We note that cement and lightweight aggregate kiln sources are currently allowed to use site-specific training and certification programs because there is no ASME or other standard for these sources that addresses their hazardous waste burning activities. Because the requirements discussed below are appropriate for these sources as well, we propose to require that the requirements also apply to cement and lightweight aggregate kilns. We propose to specify a training curriculum to ensure that the scope of the training is sufficient to ensure the control room operator can maintain compliance with the requirements of Subpart EEE. The certification program (i.e., written examination at a minimum) would be required to address the topics in the training curriculum. The training curriculum would be required to include the following topics: (1) Environmental concerns, including types of emissions; (2) basic combustion principals, including products of combustion; (3) operation of the specific type of combustor used by the operator, including proper startup, waste firing, and shutdown procedures; (4) combustion controls and continuous monitoring systems; (5) operation of air pollution control equipment and factors affecting performance; (6) inspection and maintenance of the combustor, continuous monitoring systems, and air pollution control devices; (7) actions to correct malfunctions or conditions that may lead to malfunctions; (8) residue characteristics and handling procedures; and (9) applicable Federal, state, and local regulations, including Occupational Safety and Health Administration workplace standards. This training curriculum is modeled after the requirements the Agency recently promulgated for commercial and industrial solid waste incinerators. See 65 FR 75338 (December 1, 2000). We believe this training is also appropriate for hazardous waste combustors. To maintain certification, an operator would be required to complete an annual review or refresher course covering, at a minimum, the following topics: (1) Update of regulations; (2) combustor operation, including startup and shutdown procedures, waste firing, and residue handling; (3) inspection and maintenance; (4) responses to malfunctions or conditions that may lead to malfunction; and (5) operating problems encountered by the operator. These are the same requirements the Agency recently promulgated for commercial and industrial solid waste incinerators at § 60.2085, and we believe they are also appropriate for hazardous waste combustors. C. Control Room Operator Training and Certification Would Not Be Required for Shift Supervisors The final rule requires the same level of training and certification for shift supervisors and control room operators. Incinerator stakeholders question whether shift supervisors need to meet these training and certification requirements. Stakeholders note that shift supervisors often have administrative duties that are not closely related to the technical knowledge required to operate and maintain a combustor. After reconsideration, we agree with stakeholders' reasons for not requiring that shift supervisors be trained and certified to the level of a control room operator. Accordingly, we propose to require that shift supervisors, like personnel other than control room operators, be trained and certified to the technical level commensurate with the employee's job duties. D. A Certified Control Room Operator Must Be on Duty At All Times We propose to revise the rule to clarify that a certified control room operator must be on duty at the source at all times the source is in operation. Having a certified operator present at all times is necessary to ensure compliance with the emission standards and operating requirements, and to take appropriate corrective measures when malfunctions occur. VI. Bag Leak Detection System Section 63.1206(c)(7)(ii) of the hazardous waste combustor rule prescribes baghouse operation and maintenance requirements for incinerators and lightweight aggregate kilns, including a requirement for the continuous operation of a bag leak detection system as a continuous monitor. Since promulgation of the rule, stakeholders have raised two issues: (1) Can less sensitive bag leak detectors be approved under the alternative monitoring provisions; and (2) why did we conclude that opacity monitors meeting revised Performance Specification 1 are not likely to be acceptable bag leak detectors. A. Can Less Sensitive Bag Leak Detectors Be Approved Under the Alternative Monitoring Provisions? Section 63.1206(c)(7)(ii)(D)( 1 ) requires the bag leak detector system to be capable of continuously detecting and recording mass changes in particulate matter emissions at concentrations of 1.0 milligrams per actual cubic meter or less. Stakeholders state that monitors with higher detection limits are able to detect subtle changes in baseline, normal emissions as well as catastrophic events, and question whether these monitors can be approved under the alternative monitoring petitioning procedures of § 63.1209(g)(1). We support the use of monitors with higher detection limits provided the monitor is sensitive enough to detect subtle increases in baseline, normal emissions, and we plan to develop guidance recommendations on this issue. We request comment on whether § 63.1206(c)(7)(ii)(D)( 1 ) should be revised to explicitly allow the use of monitors with higher detection limits, or whether the existing alternative monitoring provisions coupled with guidance recommendations is sufficient. In addition, we request comment on how a source would document that a bag leak detection system, with a detection level higher than 1.0 milligrams per actual cubic meter, can detect subtle changes in baseline, normal mass emissions of particulate matter. For example, should we require site-specific tests to document that alternative detectors provide a measurable and repeatable change in opacity output with an increase in particulate matter mass emissions? B. Why Did We Conclude That Opacity Monitors Meeting Revised Performance Specification 1 Are Not Likely To Be Acceptable Bag Leak Detectors? EPA promulgated a significantly improved Performance Specification 1 (PS1) for opacity monitors on August 10, 2000. See 65 FR at 48914. We considered whether to allow use of opacity monitors meeting PS1 as bag leak detectors, but conclude that they are not likely to be sensitive enough to detect subtle increases in mass particulate matter emissions from a source equipped with a well designed and operated baghouse. Revised PS1 includes additional design and performance specifications as well as new test procedures that provide a profound improvement on opacity data accuracy and precision. Collectively, the additional measures provide a comprehensive, in-depth functional test of the complete measurement procedure, thereby eliminating many of the performance problems associated with previous opacity monitors. The revisions go far beyond the previous version of PS1, drawing on recent technological advancements in optics, electronics, and information transfer. There are similar specifications for such monitors in Europe. The stipulation of automatic self-diagnosing capability is one of the many modern features incorporated into the new PS1. Taken together, the additional measures reflect a distinct new generation in the state-of-the-art of opacity monitors. Notwithstanding the improvements that revised PS1 requires, opacity monitors are generally not acceptable for use as a bag leak detector because they are not sensitive enough to detect subtle increases in baseline, normal emissions. Baghouse emission opacity levels are very nearly zero at particulate matter concentrations below emission standards and are very near the lower detection limits of a continuous opacity monitoring system (COMS). COMS manufacturers have collectively raised the concern about COMS sensitivity limitations at low opacity levels. (See ASTM D-6216-98, Standard Practices for Opacity Monitor Manufacturers to Certify Conformance with Design and Performance Specifications.) Although the increase in particulate matter mass emissions that would trigger a measurable opacity change that a COMS could detect is usually site-specific and would depend on the particle size and reflective and refractive properties. We are concerned that particulate matter emission concentrations may have to double or triple before a COMS could detect a significant opacity change at the low opacity levels associated with baghouse emissions. For these reasons, we conclude that COMS meeting Performance Specification 1 are not likely to be suitable as bag leak detectors. Nonetheless, as discussed above, we request comment on whether an approach could be developed to allow use of bag leak detectors that have detection limits above 1.0 milligrams per actual cubic meter. Moreover, we note that electrodynamic and triboelectric bag leak detectors have proven to be much more sensitive and cost about the same or less than COMS to install and operate. In addition, some particulate matter continuous emissions monitors (CEMS) have been shown to be able to detect very small changes in particulate matter mass emissions at low emission levels. If sources were to use a particulate matter CEMS as a bag leak detector, sources need not correlate the detector to particulate matter emission concentrations. Rather, sources would be required to document that the CEMS provides a measurable and repeatable change in output with an increase in particulate matter mass emissions. VII. Time Extensions For Performance Testing if the Test Plan Has Not Been Approved During the comment period of the final rule and after promulgation, stakeholders raised the concern that the rule requires sources to commence performance testing within 180 days after September 30, 2002, even if the test plan has not been approved. Although the rule requires submittal of the test plan 12 months prior to the scheduled test date to provide a nine-month review period, stakeholders are concerned that the test plan may not be approved at the conclusion of that review period. Stakeholders state that they may spend hundreds of thousands of dollars to conduct a test under an unapproved test plan, only to learn after the test that EPA or the state may not accept the results as a valid demonstration of compliance with the emission standards due to differences of opinion on test design. In the preamble to the final rule, we address this issue by stating that “If permit officials nevertheless fail to act within the nine-month review and approval period, a source could argue that this failure is tacit approval of the plan and that later “second-guessing” is not allowable.” See 64 FR at 52912. However, stakeholders are concerned that this preamble language does not guarantee that they will not have to repeat the test. Stakeholders recommend revising the rule to allow sources to wait until a test plan has been approved before conducting a performance test. We are reluctant to deviate from the Part 63 General Provision's six-month deadline after the compliance date for conducting the initial comprehensive performance test. We continue to believe that an open-ended test date will not provide an incentive for either sources or regulatory officials to resolve differences related to a test plan, thereby unnecessarily delaying testing. Nonetheless, we acknowledge that there may be situations where a source and regulatory officials are making genuine efforts to complete review of the test plan, but for reasons beyond their control, the review cannot be completed prior to the testing deadline. Accordingly, we propose to revise the rule to address these particular situations. Under today's proposal, a source may petition the Administrator, under the authority of § 63.7(h), to obtain a “waiver” of any performance test—initial or periodic performance test; comprehensive or confirmatory test. The “waiver” would not eliminate the test, but would be used to grant an extension of time to conduct the performance test. To qualify for the waiver, a source must make a good faith effort to comply with the testing requirements in a timely manner. First, as currently required, sources must submit a site-specific emissions testing plan and a continuous monitoring system performance evaluation test plan at least one year before a comprehensive performance test is scheduled to begin (see § 63.1207(c)(1)), or at least 60 days before a confirmatory performance test is scheduled to begin (see § 63.1207(d)). Sources also must submit all other documentation required by Subpart EEE to be included with the performance test plans. The submitted test plans must fulfill the substantive content requirements of §§ 63.1207(f) and 63.8(e). Second, a source must make a good faith effort to accommodate the Administrator's comments on the test plans. Finally, the Administrator must not take final action, through a notification of intent to deny (see § 63.7(c)(3)(i)(B)), to deny the source's test plan(s). Under today's proposal, sources must submit to the Administrator a waiver petition or request to renew the petition under § 63.7(h), separately for each source, at least 60 days prior to the scheduled date of the performance test. The Administrator would approve or deny the petition within 30 days of receipt and promptly notify the source of the decision. The Administrator would not approve extensions of the test date for a duration exceeding 6 months, and the Administrator would include in granted petitions a sunset provision to end the waiver within 6 months. To renew a waiver, we are proposing that sources must submit a revised petition under § 63.7(h)(3)(iii) at least 60 days prior to the end date of the most recently approved waiver petition. The Administrator could approve a revised petition for a total waiver period up to 12 months. A performance test could not be delayed more than a total of 12 months, irrespective of the status of approval of the test plan. If the Administrator denies a § 63.7(h) waiver petition, we are proposing that the source must commence the performance test, with or without approved test plans, by either the deadline provided by Subpart EEE or by the expiration date of their most recently approved waiver petition, whichever is later. Sources would also need to address, in the waiver petition, the following requirements of § 63.7(h). A source must provide documentation to enable the Administrator to determine if “the source is meeting the relevant standard(s) on a continuous basis * * *.” See § 63.7(h)(2). For extension requests of the initial comprehensive performance test, a source must submit the Documentation of Compliance to assist the Administrator in making this determination. In addition, § 63.7(h)(3)(iii) requires sources to “include information justifying the owner or operator's request for a waiver, such as the technical or economic infeasibility, or the impracticality, of the affected source performing the required test.” In order to continue to keep the public informed of the source's compliance status, the source would need to notify the public (i.e., the source's public mailing list) of their § 63.7(h) petition to “waive” a performance test. The following is an example time line indicating how the proposed § 63.7(h) waiver petitioning process would work for the initial comprehensive performance test. All end dates should be read as “no later than” X number of months. The time line assumes the source has submitted its performance test plans (i.e., for emissions testing and continuous monitoring system evaluation) on the deadline date—one year before the performance test must be conducted (i.e., sources submit the test plans 6 months prior to the compliance date). 0 time—Submittal of performance test plans for review (1 year prior to test date; 6 months prior to compliance date). 9 months—Administrator does not approve or deny test plans, even though the source has acted in good faith to obtain approval 10 months—Submittal of performance test waiver petition and notify public (2 months prior to test date). 11 months—Administrator approves or denies the performance test waiver (1 month after receipt of waiver). 12 months—Commence performance test if the Administrator denies waiver. 12 months + ≤6months—Extended performance test commencement date if the Administrator approves waiver. 16 months—If needed, submit performance test waiver renewal petition and notify public (2 months prior to sunset of latest approved waiver). 17 months—Administrator approves or denies renewal petition (1 month after receipt of renewal petition). 18 months—Maximum extension of test date for unrenewed performance test waivers. 18 months + ≤6months—Extended performance test commencement date with renewed waiver. 24 months—Maximum extension of test date for renewed performance test waivers. VIII. Flexibility in Operations During Confirmatory Performance Testing for Dioxin/Furan During the confirmatory performance test, the final rule requires sources to operate so that carbon monoxide or hydrocarbon levels, and operating parameter limits associated with the dioxin/furan emission standard, are within the range of the average values over the previous 12 months. Sources also must stay within the maximum or minimum value, as appropriate, that is allowed. See § 63.1207(g)(2). These requirements ensure that during the confirmatory performance test, dioxin/furan emissions are within the range of the normal to the highest allowable emissions. Stakeholders express concern that it may be difficult to “dial in” operation of the combustor to the required range for each operating parameter simultaneously. Sources are particularly concerned about having to operate within a potentially narrow range of carbon monoxide levels for sources that normally operate close to the 100 ppmv limit. This is because carbon monoxide levels are dependent on many combustion-related factors and cannot be directly “dialed in” as can be done for other parameters (e.g., activated carbon injection federate). Although this is not likely to be a widespread problem, we acknowledge there may be a problem in some situations. Accordingly, we propose today to revise the rule to: (1) Allow approval in the test plan for operations under a wider operating range for a particular parameter based on information justifying that operating within the required range may be problematic; and (2) allow the Administrator to accept test results during the finding of compliance based on operations outside of the range specified in the confirmatory test plan. Allowing the Administrator to accept test results based on operations outside of the range specified in the test plan would address when a source did not anticipate a problem in maintaining the operating levels within the required range (and therefore did not request advance approval to do so), but because of unforeseen factors, were unable to maintain the required range. This provision would give permit writers discretion to accept emissions data obtained when operating outside of the prescribed range so that sources would not have to incur the costs of an additional confirmatory test. In determining whether to accept test results from operations outside of the range specified in the test plan, permit writers would consider factors including: (1) The magnitude and duration of the deviation from the required range; (2) the historical range of the parameter (e.g., the range between the 10th and 90th percentile time-weighted average values for the parameter); (3) the proximity of the emission test results to the standard; and (4) the reason for not maintaining the required range. These factors determine whether the operations are reasonably representative of normal operations and how important it may be that test operations be truly representative of normal operations. IX. Waiving Operating Parameter Limits During Performance Testing Section 63.1207(h) automatically waives the operating parameter limits (OPLs) during subsequent comprehensive performance tests under an approved performance test plan. After promulgation, stakeholders raised two concerns. They believe that: (1) OPLs defined in the Documentation of Compliance should be waived during the initial comprehensive performance test and associated pretesting; and (2) OPLs should be waived during testing and pretesting irrespective of whether the test plan has been approved. A. Should We Waive OPLs During the Initial Comprehensive Performance Test? Section 63.1211(d) requires sources to include in the operating record a Documentation of Compliance (DOC) that establishes limits on the operating parameters under § 63.1209 that, based on an engineering evaluation, will ensure compliance with the emission standards. The DOC may be revised at any time prior to submitting the Notification of Compliance. If additional engineering information becomes available that leads sources to conclude that they can operate under less stringent OPLs during the initial comprehensive performance test and demonstrate compliance with the emission standards, the DOC may be revised accordingly. Therefore, we do not believe that additional regulatory language is needed to enable source to operate during pretesting or the initial comprehensive performance test under OPLs less stringent than those identified in the DOC. We specifically request comment on this issue. B. Should We Allow the OPLs To Be Waived if the Test Plan Has Not Been Approved? Section 63.1207(h) waives the OPLs during performance testing under an approved test plan. We required pretesting and testing operations to be conducted under an approved test plan as a prerequisite for the waiver. This will ensure that operations, when the OPLs are waived, are likely to remain in compliance with the emission standards. In retrospect, however, we acknowledge that stack emissions measurements will be taken during both pretesting (see § 63.1207(h)(2)) and testing. Given that there will be documentation of any exceedance of an emissions standard during a performance test, potentially indicating a violation during such testing, it is not necessary to require that the test plan be approved before the OPLs can be waived. Similarly, if a source records the results of pretesting, the OPLs should be waived without approval of the test plan. Accordingly, we propose to revise the rule to waive the OPLs during pretesting (if the source records the results of the pretest) and performance testing. See proposed § 63.1207(h). Although stakeholders have raised concerns about testing under an unapproved test plan (see Section VII above), there may be instances where a source may choose to test under such conditions. Consequently, the regulatory revision appears to be warranted. X. Method 23 as an Alternative to Method 0023A for Dioxin/Furans The final rule requires use of Method 0023A to determine compliance with the dioxin and furan emission standard. See § 63.1208(b)(1). Based on discussions with stakeholders after promulgation of the final rule, we believe it is appropriate to request comment on amending the final rule to allow petitions for the use of Method 23 in lieu of Method 0023A. 15 15 Sources can currently petition EPA to use alternative test methods pursuant to § 63.7(f). The petition process that we are requesting comment on would not require sources to submit the results of a Method 301 validation process as is required under § 63.7(f). Method 23 is the Clean Air Act dioxin/furan air emission test method found in 40 CFR Part 60, Appendix A. Method 0023A is the RCRA dioxin/furan air emission test method found in SW-846. 16 The final rule requires use of Method 0023A because this method is the updated version of Method 23. At the time of final rule publication, we believed that the improvements to the updated method warranted use of Method 0023A. 16 Method 0023A was proposed on July 25, 1995 (see 60 FR 37974). EPA received comments on Method 0023A and later incorporated the method into SW 846 in a final rule on June 13, 1997 (see 62 FR at 32451). Stakeholders request that we give sources the option to use Method 23 or Method 0023A because: (1) The dioxin/furan standard is based on emissions data that was collected using Method 23 procedures; (2) Method 0023A is more expensive because of additional analytical costs; and (3) Method 0023A results in higher detection limits. Method 23 and Method 0023A are similar methods. Method 23 combines the front half of the filter and probe rinse with the back half of the sorbent and rinses to perform a single extraction and analysis. Recovery of spiked standards into the sorbent are used to serve as an indicator of overall recovery. Method 0023A differs from Method 23, primarily in that Method 0023A uses the addition of standards to both the filter (front half) and sorbent (back half), and then separates the front half and back half for analysis in order to determine the recovery from each half. They are separated in order to better quantify recoveries for both the back half and front half fractions. This is important, because low recoveries in Method 23 are sometimes associated with dioxin/furan contained in solid phase particulate that may go unnoticed due to the combined front half and back half analysis. This may be of particular importance for sources that use activated carbon injection, or sources whose particulate matter contains significant levels of carbonaceous material. In other words, Method 0023A was designed as an improvement to Method 23 by incorporating separate recovery, spiking, and analysis of front half and back half samples to improve the quality assurance of the front half and back half analysis. The benefits of Method 0023A compared to Method 23 include accurate recovery data and known data quality. The downsides to Method 0023A include higher analytical costs and possibly higher detection limits. Although the detection limits of Method 0023A may be higher than Method 23 detection limits, we do not believe that these higher detection limits will adversely affect a source's ability to adequately demonstrate compliance with the dioxin/furan standard, as we explained in the technical support document to the final rule. 17 This is true because analytical detection limits have decreased over recent years. 17 See “Final Technical Support Document for HWC Standards, Volume IV: Compliance with HWC MACT Standards,” Chapter 16, July 1999. We request comment on whether we should amend the final rule to give the option to use Method 23 in lieu of Method 0023A. We are considering allowing sources to petition the authorized regulatory agency to use Method 23 in lieu of Method 0023A. Under such an approach, a source would have to justify why the use of Method 23 is warranted. Factors that could be considered by the regulatory official in reviewing these petitions include: (1) The carbonaceous content of the particulate that is emitted from the source; (2) analytical costs; (3) data quality; and (4) detection limits. For example, under this approach, we believe that a source could address Method 23 data quality concerns by submitting previous Method 0023A results to the regulatory official that document: (1) The recovery percentages of the front and back half of the analysis; and (2) the amount of dioxin and furans present in the front half. Method 0023A results that indicate good front half recoveries could support a source's claim that Method 23 is an appropriate method to demonstrate compliance with the dioxin/furan emission standard. 18 The added data quality checks associated with Method 0023A may not be needed if the results of previous Method 0023A analyses indicate good front half recovery percentages. Method 23 may also be warranted if dioxin and furans are not detected, or are detected at low levels in the front half of Method 0023A. 18 This assumes, however, that method recoveries do not significantly vary at a source for different emissions tests. Any petition to use Method 23 should address whether method recoveries are expected to change from one emission test to another. XI. Calibration Requirements for Thermocouples Section 63.1209(b)(2)(i) of the final rule requires verification of the calibration of each thermocouple or other temperature sensor at least once every three months. Stakeholders express concern that recalibration of each temperature measurement device every three months is a significant undertaking. Stakeholders explain that, for example, temperature measurement devices on the air pollution control train are typically flanged onto the process piping and/or vessels. To recalibrate these devices without shutting the combustor down is an involved process. Removing these measurement devices for recalibration would require the operator to enter a static value in the automatic waste feed cutoff system to avoid a cutoff, and have a technician equipped with appropriate personal protective equipment receive the appropriate line and equipment opening permits, and then try to safely remove the instrument from the process while the combustor is still running. For configurations that have pressurized portions of the air pollution control train, the combustor would be required to shut down to avoid release of fugitive emissions. Stakeholders question whether the benefits outweigh the burden of recalibrating each temperature device every three months. Stakeholders also state that recalibration of pyrometers is particularly problematic. Optical pyrometers are often sealed at the factory to prevent adjustment of the calibration. To check calibration on an optical pyrometer is difficult and stakeholders believe it is not a task that should be undertaken every three months unless there are clear benefits. It is not clear to us that recalibration of all types of temperature measurement devices every three months is as burdensome an undertaking as stakeholders suggest. Thermocouples are the most common temperature measurement device used for compliance assurance. We believe that their calibration can generally be confirmed without removing them from the combustor. Thermocouples may malfunction either by a failure in the circuit ( e.g. , the junction between the two wires at the bead may break) or the electronics may drift. If the circuit fails, the thermocouple will give clearly erroneous readings. Drift in the electronics can be corrected without removing the thermocouple. We specifically request comment on whether thermocouples can be recalibrated without removing them from the combustor. Although it may be impractical to calibrate the internal operations of a pyrometer every three months, as stakeholders suggest, there are other maintenance activities such as cleaning of the optics and alignment checks that will help ensure that the pyrometer is performing correctly. We specifically request comment on whether the rule should require that these and other maintenance activities should be performed every three months. If based on review of comments to this notice and reevaluation we determine that recalibration of temperature monitoring devices every three months is not practicable, we would revise the rule to delete § 63.1209(b)(2)(i). In lieu of a generic recalibration requirement that applies to all temperature monitors, we would require that you develop an appropriate calibration procedure and frequency and include that information in the evaluation plan required by § 63.8(e)(3)(i). XII. Alternative Approach To Establish Operating Parameter Limits The rule requires sources to establish most operating parameter limits as the average of the test run averages of the comprehensive performance test. Each test run average is calculated by summing all the one-minute readings within the test run and dividing that sum by the number of one-minute readings. See 64 FR at 52922. Stakeholders state that this is an unreasonably conservative approach to establish operating parameter limits in that sources would not be allowed to operate in the way that they did 50% of the time during the performance test (when demonstrating compliance with the emission standards). This may overstate the conservatism inherent in this approach. 19 Nevertheless, we believe that a conservative approach is warranted because: (1) These parameters can have a significant effect on emissions; and (2) the approach is consistent with how manual method emissions results are determined ( i.e. , manual method emission test results for each run represent average emissions over the entire run). 20 19 For example, if the duration of each run of the performance test were 60 minutes, establishing parameter limits based on the average of the run averages allow sources to continue to operate as during the performance test. This is because 1-minute values that are higher than the average would be off-set by 1-minute values that are correspondingly lower than the average. Because most performance test runs have a duration longer than 60 minutes, however, the “average of the run averages” approach coupled with an hourly rolling average averaging period for most parameter limits, will require that sources operate more conservatively than during the performance test as a practical matter. 20 Petitioners in litigation challenging the underlying rule have maintained that the one-hour averaging time to demonstrate compliance with the dioxin standard effectively amends the standard. The argument goes that the one-hour averaging period is shorter than that used in the source's performance test. EPA disagrees; the dioxin standard does not prescribe any particular averaging time, or other monitoring regime, for achieving a temperature level, so that using a one-hour averaging time does not amend the standard. However, even if (against our view) the temperature monitoring requirement is considered to change the emission standard, it appears justifiable as a beyond the floor standard under CAA section 112(d)(2). First, the standard is readily achievable technically. Spray quenching, the means of control, merely requires turning of a control valve to allow quenching. USEPA, “Final Technical Support Document for Hazardous Waste Combustor MACT Standards, Volume IV: Compliance with the Hazardous Waste Combustor Standards,” July 1999, p. 2-16. Operators can readily determine when quenching is needed, since thermocouples report instantaneous temperature changes, allowing immediate reaction to temperature changes. Ibid, p. 2-10. Second, EPA has already considered this cost ( i.e. , the cost of quenching) in determining the standards for HWCs. EPA does not believe that there would be any incremental cost associated with the one-hour averaging requirement, because it is based on the same spray quenching technology which is the basis for the standards already adopted. See also 64 FR at 52892 (finding that the cost of spray quenching technology for lightweight aggregate kilns is reasonable, in adopting the beyond-the-floor standard for dioxin/furans). In addition, the one-hour averaging requirement is needed to prevent exceedances of the emission standard itself, see Ibid, at 2-8 to 2-9 and 3-8 to 3-9 (documenting how net dioxin/furan emissions would increase over the amounts allowed by the emission standard without this requirement, but further explaining why the ten-minute averaging time that EPA initially proposed is not essential). See also EPA's Brief in CKRC v. EPA , no. 99-1457 (D.C. Cir. 2001) at pp. 113-120 (a copy of this brief is part of the record for this proposal). Finally, we do not believe there are any adverse non-air or energy impacts associated with the averaging requirement (and again, EPA has already assessed energy impacts and waste generation impacts of the standard when promulgating the standard in the first place). See generally USEPA, “Final Technical Support Document for Hazardous Waste Combustor MACT Standards, Volume V: “Emissions Estimates and Engineering Costs,” July 1999 (RC2F-S0011) chapter 10. Stakeholders also maintain that it is not technically practicable to establish some operating parameter limits using the average of the test run averages. Stakeholders present examples including cement kiln minimum combustion chamber temperature (see discussion in Section XIV below), and secondary power input to an ionizing wet scrubber or wet electrostatic precipitator. In light of stakeholders' concerns, we are considering an alternative approach to establish operating parameter limits that provides assurance of compliance with the emission standard: establishing multiple limits for a given parameter that ensures that the profile of the parameter does not exceed the profile documented during the comprehensive performance test. We call this the “matching-the-profile” approach. Under the matching-the-profile approach, a source would establish multiple limits for a given parameter that ensure that the profile of the parameter does not exceed the profile documented during the comprehensive performance test. This approach has the advantage of allowing operations at parameter levels above the average level of the performance test for the same period of time and at the same levels, as shown during the performance test. Provided that the source operates below the average level of the performance test for the same period of time, and at the same levels, as during the performance test. One disadvantage is that, to effectively implement the approach, sources would be required to establish multiple operating limits for a single parameter. As an example of how this matching-the-profile approach would work for establishing the gas temperature operating limit at the inlet to an electrostatic precipitator, consider the following hypothetical gas temperature data for three runs of a comprehensive performance test. The individual run times are presented, and the total of the run times is nine hours. Table 1.—Alternative Approach To Establish an OPL Whereby the Parameter Profile Documented During the Comprehensive Performance Test Cannot Be Exceeded [Example Parameter: Gas temperature at the inlet to an ESP.] [Assume Run Times as Follows: Run 1-180 minutes; Run 2-150 minutes; Run 3-210 minutes. Total time of 540 minutes (9 hrs).] Percent of time 1-Min avg temperature that was not exceeded the specified % of Time (°F) Run 1 Run 2 Run 3 Average of run averages Time that avg run avg can be exceeded in any 9-hr block 100% (max T) 405 415 425 415 0 min 90% 395 398 390 394 54 min 50% 375 380 375 377 270 min 25% 370 350 360 360 405 min In this example, we have assumed that four operating limits would be needed to ensure adequately that the performance test profile is not exceeded: a maximum temperature that could not be exceeded, and three temperature limits that could be exceeded for prescribed periods of time during each 9-hour block average. In practice, the number of parameter limits would be established on a site-specific basis and would be a function of factors including: (1) The variability of the parameter during the test (i.e., range from the high to low value 21 ); (2) whether the performance test emission levels were close to or well below the emission standard; and (3) the relationship between the parameter and emission levels. 22 21 The greater the range of values for a parameter within a percentile, the less certain we can be that the performance test profile (and emission levels) will be maintained. This is because a source could theoretically operate for extended periods of time (i.e., longer than during the performance test) at the upper end of the range (or the lower end for parameters for which minimum limits are established). 22 For example, more rather than fewer parameter limits would be appropriate to characterize the profile for gas temperature at the inlet to an electrostatic precipitator given that dioxin/furan emissions relate exponentially to inlet gas temperature. In the example presented above, 1-minute average temperature levels are ranked from highest to lowest for each run, and the temperature associated with various time percentiles (i.e., 100%, 90%, 50%, and 25%) are identified. In Run 1, for example, temperatures below 395°F were achieved 90% of the time. Then, a time-weighted average temperature across the runs is calculated for each of the percentiles. Finally, the time percentiles are converted to the number of minutes in a block period of time (corresponding to the time required to conduct all runs of the performance test). We now have a series of temperature limits that can be exceeded only for a specified period of time. Compliance with these time/temperature limits should ensure that the temperature profile of the performance test is not exceeded during normal operations, and that the emission standard is not exceeded. We request comment on whether this approach to establish operating parameter limits as an alternative to calculating the limit as the average of the test run averages would be less burdensome for regulated sources while ensuring compliance with the emission standards. We also note that sources can request alternative monitoring approaches under § 63.1209(g)(1) and may request to use this (or other) alternative approach whether or not EPA finalizes this proposal. We request comment on whether we should explicitly include this approach in the rule, or use this discussion as guidance recommendations. Explicitly defining the approach in the rule may better facilitate efforts by sources to adopt the approach to their needs, and review and approval of the approach by regulatory officials. XIII. Extrapolation of Operating Parameter Limits Stakeholders suggest that the rule inappropriately penalizes sources that achieve comprehensive performance test emission levels well below the standard by establishing operating parameter limits based on performance test operations at those low emission levels. Operating under conditions to artificially increase emissions during testing (e.g., by detuning emission control equipment) may not be feasible or desirable from a worker/public health and cost perspective. To address this concern, we request comment on whether the rule should allow extrapolation of an operating parameter limit 23 established as currently required to a higher limit (or lower limit if the parameter limit is a minimum limit) using a site-specific, empirically-derived relationship between the parameter and emissions of the pollutant in question. 24 An example is extrapolation of the gas temperature limit at the inlet to the dry particulate matter control device to a higher limit based on the relationship between gas temperature and dioxin/furan emissions. To use this approach, a source must document the relationship empirically for their source. To remain in compliance with the emission standard, however, the temperature limit could be extrapolated to levels higher than 400 °F only if the extrapolated dioxin/furan emissions were below 0.2 ng TEQ/dscm. 23 Requests to extrapolate metal feedrates would continue to be considered under § 63.1209(n)(2)(ii). 24 In addition to using site-specific, empirically-derived relationships, we also request comment on whether the rule should allow use of established engineering principles that define the relationship between operating parameters and emissions to extrapolate operating limits and emissions. Sources could not take advantage of this extrapolated gas temperature limit in this example without also extrapolating the gas temperature limit for compliance assurance with the semivolatile and low volatile metals standards. 25 This is because gas inlet temperature is a compliance parameter for both dioxin/furans and semivolatile and low volatile metals. We would also consider allowing extrapolation of the metals compliance assurance inlet gas temperature limit using engineering calculations to a temperature limit that would correspond to metals levels close to the emission standards. 25 A gas inlet temperature limit is not required, however, if the source feeds low levels of metals and complies with the semivolatile and low volatile metals standards without emissions testing by documenting compliance with the emissions standards assuming all metals that are fed are emitted. In addition, even if the source were required to comply with a lower gas temperature limit for compliance assurance for metals, there may still be advantages to establishing an extrapolated temperature limit for compliance with the dioxin/furan standard. For example, if the source had a performance test-based temperature limit (i.e., metals temperature limit) exceedance that did not exceed the extrapolated dioxin/furan-based limit, the temperature exceedance would not represent failure to maintain compliance with the dioxin/furan emission standard. We believe that extrapolated limits should be less than 100% of the standard. Such conservatism is important because sources would not have actually demonstrated compliance with the emission standards at the extrapolated operating parameter limit. We request comment on what upper level of extrapolation would be appropriate (e.g., 75%, 80%) and whether the upper level of extrapolation should vary depending on the level of confidence in the empirical relationship or other approach that is used to calculate the extrapolation. The Administrator would grant (or deny) a petition to extrapolate an operating parameter limit on a case-by-case basis considering factors including whether: (1) The operating parameter values during the performance test were at the upper (or lower for minimum limits) range of historical, normal levels; (2) the extrapolated level sources request is warranted considering historical levels of the parameter; (3) it is impracticable to demonstrate compliance with the emission standard when operating at the desired (i.e., extrapolated) operating limit during the performance test; and (4) the extrapolation procedure will conservatively predict the relationship between the operating parameter and emissions. To determine if the extrapolation procedure conservatively relates the operating parameter to emissions, the Administrator would consider factors including how far the source requests to extrapolate the limit beyond the value calculated from the performance test and how close the emissions during the performance test were to the standard. We also note that sources can request alternative monitoring approaches under § 63.1209(g)(1) and may request to use this (or other) alternative approach prior to promulgation of a final rule. We request comment on whether we should explicitly include this approach in the rule, or use this discussion as guidance recommendations. Explicitly defining the approach in the rule may better facilitate efforts by sources to adopt the approach to their needs, and review and approval of the approach by regulatory officials. XIV. Limit on Minimum Combustion Chamber Temperature for Cement Kilns Stakeholders have expressed concern that it is technically impracticable for cement kilns to establish a minimum combustion chamber temperature based on the average of the test run averages for each run of the comprehensive performance test. Stakeholders state that combustion chamber temperatures cannot be maintained at low enough levels for the duration of the comprehensive performance test to establish workable operating limits that would allow them to burn hazardous waste fuels economically without frequent waste feed cutoffs because of potential exceedences of the limit. Stakeholders indicate that combustion chamber temperature levels are fairly constant within a narrow range and note that there is a very narrow range of temperatures and feed composition in which a cement kiln must operate in order to produce quality clinker and a marketable product. Stakeholders further note that they must take extreme actions under the current RCRA requirements to establish an economically viable minimum combustion chamber limit based on the average of the lowest hourly rolling averages for each run. Stakeholders relate that during one hour of each run of the RCRA compliance test, they must take unusual and potentially equipment-damaging steps to lower temperatures. 26 Those problems are compounded by the requirement in the MACT rule to establish the limit based on the average temperature level. 26 We note that allowing sources to establish operating limits under current RCRA regulations based on the average minimum or maximum hourly rolling average (rather than the average of the average values as required under Subpart EEE) is intended to address routine deviations that can occur even though steady-state operating conditions are maintained. Modifying operating conditions during compliance testing to induce temporary, artificial perturbations is inappropriate. Such operations are not representative of operations under the test condition. In addition, stakeholders note that it is difficult to accurately monitor combustion chamber temperature in a cement kiln. We already acknowledge this concern and, accordingly, the rule allows measurement of the temperature at a location that best represents, as practicable, the bulk gas temperature in the combustion zone. See § 63.1209(j)(1)(i). The rule also allows sources to petition the permit writer to request approval of an alternative temperature monitoring approach. See § 63.1209(g)(1). We have responded, in the final rule Comment Response Document, to stakeholder's questions about the need for monitoring combustion chamber temperature by noting that combustion chamber temperature is a principal factor in ensuring combustion efficiency and destruction of toxic organic compounds. Although we acknowledge that a cement kiln inherently controls the kiln temperature to produce clinker, this inherent control may not be adequate to assure compliance with the dioxin/furan and destruction and removal efficiency emission standards. For example, we understand that cement kilns occasionally undergo upsets and produce substandard clinker. If lower than normal combustion chamber temperatures can result from an upset, we do not know how compliance with the emission standards can be assured. Notwithstanding these reservations, and in light of stakeholders' continued concerns, we request comment on whether the rule should continue to require cement kilns to establish and comply with a minimum combustion chamber temperature limit. Stakeholders have indicated that they have produced additional data supporting their views. We also request comment on whether the alternative approaches discussed above which can be used to establish alternative operating parameter limits (i.e., match-the-profile and extrapolation) would address some of stakeholders' concerns with establishing a minimum combustion chamber temperature limit for cement kilns. We note, again, that sources may use § 63.1209(g)(1) to request alternative monitoring approaches and need not wait for the Agency's final determinations subsequent to this notice. XV. Revisions to Operating Requirements for Activated Carbon Injection and Carbon Bed Systems The final rule requires sources using carbon beds or activated carbon injection systems to limit particulate matter emissions to the level achieved during the comprehensive performance test. See §§ 63.1209(k)(5) and 63.1209(l)(3). We have since determined that: (1) It is inappropriate to explicitly require a site-specific particulate matter limit if a carbon injection system is used; and (2) particulate matter control downstream of a carbon bed is not a critical operating parameter to ensure compliance with the dioxin/furan and mercury emission standards. We propose, therefore, to delete the site-specific particulate matter limit requirement for activated carbon injection systems. We also propose to delete the requirement for sources equipped with carbon beds to establish particulate matter operating parameter limits to ensure compliance with the dioxin/furan and mercury emission standards. Dioxin/furan and mercury will adhere to the solid carbon used in an activated carbon injection systems. The final rule requires a site-specific particulate matter limit for this type of control system because an increase in particulate matter emissions could also correspond to an increase in dioxin/furan and mercury emissions. After considering stakeholder comments, we believe it is inappropriate to explicitly require a site-specific particulate matter limit if a carbon injection system is used because the rule does not require continuous monitoring of particulate matter emissions with a continuous emission monitor. The use of a site-specific particulate matter limit was originally thought to go in tandem with the requirement to use particulate matter CEMS. Since we do not require sources to use particulate matter CEMS for compliance purposes, we believe it is inappropriate to require site-specific particulate matter limits. 27 Particulate matter emissions are instead controlled by complying with operating limits on the particulate matter control devices (e.g., minimum power to an electrostatic precipitator). Therefore, we propose to revise § 63.1209(k)(5) to require sources to establish operating limits on the particulate matter control device consistent with the approach used to control particulate emissions for compliance assurance with the semivolatile and low volatile metals emission standards. 27 The issue of the use of site-specific particulate matter limits to assure compliance with metal and dioxin/furan standards, when sources use a particulate matter CEM, will likely be addressed in any future particulate matter CEM proposal. We also believe that particulate matter control downstream of a carbon bed is not a critical operating parameter to ensure compliance with the dioxin/furan and mercury emission standards. We note that most, if not all, carbon bed systems in use today are positioned downstream from particulate matter control devices to minimize particulate buildup in the carbon bed. Carbon beds are also designed so that carbon leakage into the flue gas is minimized. 28 We, therefore, propose to delete the language in § 63.1209(k)(5) that requires control of particulate matter emissions to ensure compliance with the dioxin/furan and mercury standards for sources with a carbon bed. 28 See memo from S. Schliesser to M. Galbraith, June 7, 2000, regarding “Carbon Bed Reentrainment Issue” for more information. XVI. Clarification of Requirements to Confirm Carbon Bed Age When demonstrating compliance with the dioxin/furan (and mercury) emission standard during the initial comprehensive performance test, sources may use the manufacturer's specification for the limit on carbon bed age rather than the actual age of the bed during the performance test. If using the manufacturer's specification for carbon bed age, § 63.1209(k)(7)(i)(C) requires sources to recommend in the initial comprehensive performance test plan a schedule for subsequent dioxin/furan emissions testing, prior to the confirmatory performance test, that will be used to document to the Administrator that the initial limit on maximum bed age ensures compliance with the dioxin/furan emission standard. Stakeholders express several concerns with these requirements: (1) How much testing and what type of testing is required to confirm bed life; (2) if the manufacturer's specification for bed life is such that it extends beyond the deadline to conduct the dioxin/furan confirmatory test, testing to confirm bed life should not be required before that dioxin/furan confirmatory test; and (3) given that a carbon bed controls mercury as well as dioxin/furan, testing to confirm bed life should be required to demonstrate compliance with both the dioxin/furan and mercury emission standards. We address each of these issues below. A. How Much Testing and What Type of Testing Is Required to Confirm Bed Life? We intended that testing equivalent to the dioxin/furan confirmatory test would be required to confirm the life of the carbon bed. Therefore, a test comprised of at least three runs would be required. The operating conditions would be the same as required for the dioxin/furan confirmatory test under § 63.1207(g)(2). 29 29 Given that carbon bed removal efficiency is closely related to combustion gas temperature at the inlet to the bed system, we request comment on whether the carbon bed life confirmatory test should be conducted at inlet gas temperatures at or near the maximum allowed (i.e., rather than at levels within the range of normal levels to the maximum allowed). B. What Happens If Bed Life Extends Beyond the Deadline for Dioxin/Furan Confirmatory Testing? If the manufacturer's specification calls for a bed life beyond the deadline for confirmatory testing, the source must conduct the dioxin/furan confirmatory test by the deadline and also conduct the bed life confirmatory test at any time prior to the manufacturer's specification for bed life. We are proposing to revise the rule so that in this situation bed life confirmatory testing would not be required prior to dioxin/furan confirmatory testing. If, for example, the manufacturer's specification for bed life was 3.5 years and the bed was installed just prior to the comprehensive performance test, the source must conduct the dioxin/furan confirmatory test within 2.5 years after the comprehensive performance test. In addition, the source must conduct a bed life confirmatory test within 3.5 years of the comprehensive performance test. Of course, sources may elect to forgo the additional year of bed life to avoid the expense of conducting the carbon bed life confirmatory test. C. Should Bed Life Confirmatory Testing Include Testing To Confirm Compliance With Both the Dioxin/Furan and Mercury Emission Standards? Given that carbon beds control both dioxin/furan and mercury emissions, bed life confirmatory testing must document compliance with both the dioxin/furan and mercury emissions standards. Not requiring mercury testing during bed life confirmatory testing was an oversight when we promulgated the rule. We are proposing to revise the rule accordingly. See proposed revision to § 63.1209(l)(4). The bed life confirmatory testing for mercury must be conducted under normal conditions for the operating parameters used to control mercury emissions. See § 63.1209(l). This is the same concept that is used to confirm bed life for dioxins/furans, and for the dioxin/furan confirmatory test. Thus, the parameters specified under § 63.1209(l) must be held within the range of the average value over the previous 12 months and the maximum or minimum, as appropriate, that is allowed. XVII. Revisions to Operating Parameter Limits for Wet Scrubbers The final rule controls mercury emissions from hazardous waste combustors by: (1) Controlling the feedrate of mercury; (2) wet scrubbing to remove soluble mercury (e.g., mercuric chloride); and (3) carbon adsorption. See § 63.1209(l). There are specific operating parameter limits that apply to each control technology. For hazardous waste combustors using wet scrubbers to control mercury, the operating parameter limits are identical to those that are required to assure compliance with the hydrochloric acid/chlorine gas emission standard. Specifically, those requirements include establishing hourly rolling average limits on minimum pH of the scrubber water based on operations during the comprehensive performance test. The hourly rolling average is established as the average of the test run averages. The pH of the scrubber water is an important parameter for chlorine control because, at low pH, the scrubber solution is more acidic and removal efficiency of hydrochloric acid and chlorine gas decreases. A. What Is the Issue With the Minimum Operating Parameter Limit Requirement for Wet Scrubbers With Regard to Mercury Control? Since promulgation of the rule, we've become aware of evidence that the scrubber liquid pH can have an important effect on the control and fate of mercury in wet scrubbers. In particular, various wet scrubber manufacturers and operators have observed that low pH (acidic) scrubber liquid solutions improve the control of mercury in hazardous waste combustor stack gases. There also is some recent work supporting the idea that scrubber liquid pH is an important factor in mercury capture and removal. In addition to low pH liquids increasing the control of elemental mercury, there's also evidence that high pH liquids may tend to reduce the captured soluble mercury back to the elemental form of mercury, which would then be re-released with the liquid during the liquid recycle. This line of thinking suggests that it may be necessary to establish a maximum scrubber liquid pH during the compliance test to ensure sufficient mercury control. However, a maximum scrubber liquid pH is opposite to the minimum liquid pH limit that is set and used to control chlorine emissions. B. How Would a Low pH Scrubber Liquid Improve Mercury Control? There are a number of reasons why a low pH scrubber liquid is thought to improve mercury control: —Most elemental mercury formed in combustion is thought to favor conversion (or oxidation) to ionic mercury, such as HgCl 2 , mercuric chloride, under typical air pollution control device conditions on hazardous and other waste combustors (e.g., see Lee and Kilgroe (1998), Hall (1991)). —Oxidized mercury is very soluble in low pH scrubbing solutions (e.g., see Siret et al. (1997)). Also, strong reducing agents in scrubber liquid (which are more likely in higher pH scrubbing liquids) will reduce and revolatilize captured mercury. —Scrubber liquids with high pH (i.e., added NaOH) may inhibit the oxidation of elemental mercury, and its subsequent absorption into the scrubber liquid (i.e., the ability to be controlled by the scrubber), see Soelberg (1998). —In high pH reducing liquids, captured soluble ionic mercury may be reduced back to elemental forms in the scrubber liquid and then re-released during the liquid during recycle. The use of low pH solutions minimizes this possibility by favoring the formation of stable HgCl 4 (e.g., see Krivanek (1993)). Ionic mercury with a (Hg +2 ) oxidation state is very soluble in water, especially in l [Text truncated at 120,000 characters. The full text is on the page linked above.] --- Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/fr%3A01-16426. Public record. Not legal advice.