Vol. 76 Thursday, No. 33 February 17, 2011

Part II

Environmental Protection Agency

40 CFR Part 63 National Emission Standards for Hazardous Air Pollutants: Primary Lead Smelting; Proposed Rule

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00001 Fmt 4717 Sfmt 4717 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 9410 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules

ENVIRONMENTAL PROTECTION • Fax: (202) 566–9744, Attention about EPA’s public docket, visit the EPA AGENCY Docket ID Number EPA–HQ–OAR– Docket Center homepage at http:// 2004–0305. www.epa.gov/epahome/dockets.htm. 40 CFR Part 63 • Mail: U.S. Postal Service, send Docket. The EPA has established a comments to: EPA Docket Center, EPA docket for this rulemaking under Docket [EPA–HQ–OAR–2004–0305; FRL–9263–2] West (Air Docket), Attention Docket ID ID Number EPA–HQ–OAR–2004–0305. Number EPA–HQ–OAR–2004–0305, All documents in the docket are listed RIN 2060–AQ43 U.S. Environmental Protection Agency, in the http://www.regulations.gov index. National Emission Standards for Mailcode: 2822T, 1200 Pennsylvania Although listed in the index, some Hazardous Air Pollutants: Primary Ave., NW., Washington, DC 20460. information is not publicly available, Lead Smelting Please include a total of two copies. In e.g., CBI or other information whose addition, please mail a copy of your disclosure is restricted by statute. AGENCY: Environmental Protection comments on the information collection Certain other material, such as Agency (EPA). provisions to the Office of Information copyrighted material, is not placed on ACTION: Proposed rule. and Regulatory Affairs, Office of the Internet and will be publicly Management and Budget (OMB), Attn: available only in hard copy. Publicly SUMMARY: EPA is proposing Desk Officer for EPA, 725 17th Street, available docket materials are available amendments to the national emission NW., Washington, DC 20503. either electronically in http:// • standards for hazardous air pollutants Hand Delivery: U.S. Environmental www.regulations.gov or in hard copy at (NESHAP) for Primary Lead Smelting to Protection Agency, EPA West (Air the EPA Docket Center, EPA West, address the results of the residual risk Docket), Room 3334, 1301 Constitution Room 3334, 1301 Constitution Ave., and technology reviews conducted as Ave., NW., Washington, DC 20004, NW., Washington, DC. The Public required under sections 112(d)(6) and Attention Docket ID Number EPA–HQ– Reading Room is open from 8:30 a.m. to (f)(2) of the Clean Air Act (CAA). These OAR–2004–0305. Such deliveries are 4:30 p.m., Monday through Friday, proposed amendments include revisions only accepted during the Docket’s excluding legal holidays. The telephone to the emission limits for lead, the normal hours of operation, and special number for the Public Reading Room is addition of a lead concentration in air arrangements should be made for (202) 566–1744, and the telephone standard, and the modification and deliveries of boxed information. number for the EPA Docket Center is addition of testing and monitoring and Instructions. Direct your comments to (202) 566–1742. related notification, recordkeeping, and Docket ID Number EPA–HQ–OAR– Public Hearing. If a public hearing is reporting requirements. We are also 2004–0305. EPA’s policy is that all held, it will begin at 10 a.m. on March proposing to revise provisions comments received will be included in 4, 2011 and will be held at EPA’s addressing periods of startup, the public docket without change and campus in Research Triangle Park, shutdown, and malfunction to ensure may be made available on-line at http:// North Carolina, or at an alternate facility that they are consistent with a recent www.regulations.gov, including any nearby. Persons interested in presenting court decision. Finally, we are personal information provided, unless oral testimony or inquiring as to proposing revisions to the rule’s the comment includes information whether a public hearing is to be held applicability provision to make it claimed to be confidential business should contact Ms. Virginia Hunt, Office consistent with the definition of the information (CBI) or other information of Air Quality Planning and Standards, whose disclosure is restricted by statute. source category and proposing other Sector Policies and Programs Division, Do not submit information that you minor technical changes to the standard. Metals and Minerals Group (D243–02), consider to be CBI or otherwise We are also responding to a petition for U.S. Environmental Protection Agency, protected through http:// rulemaking filed on the standard with Research Triangle Park, North Carolina www.regulations.gov or e-mail. The regard to lead as a surrogate and 27711; telephone number: (919) 541– http://www.regulations.gov Web site is regulation of volatile organic 0832. an ‘‘anonymous access’’ system, which compounds (VOC) and acid gases. means EPA will not know your identity FOR FURTHER INFORMATION CONTACT: For DATES: Comments must be received on or contact information unless you questions about this proposed action, or before April 4, 2011. Under the provide it in the body of your comment. contact Ms. Sharon Nizich, Sector Paperwork Reduction Act, comments on If you send an e-mail comment directly Policies and Programs Division (D243– the information collection provisions to EPA without going through http:// 02), Office of Air Quality Planning and are best assured of having full effect if www.regulations.gov, your e-mail Standards, U.S. Environmental the Office of Management and Budget address will be automatically captured Protection Agency, Research Triangle (OMB) receives a copy of your and included as part of the comment Park, North Carolina 27711, telephone comments on or before March 21, 2011. that is placed in the public docket and (919) 541–2825; fax number: (919) 541– Public Hearing. If anyone contacts made available on the Internet. If you 5450; and e-mail address: EPA requesting to speak at a public submit an electronic comment, EPA [email protected]. For specific hearing by February 28, 2011, a public recommends that you include your information regarding the risk modeling hearing will be held on March 4, 2011. name and other contact information in methodology, contact Ms. Elaine ADDRESSES: Submit your comments, the body of your comment and with any Manning, Health and Environmental identified by Docket ID Number EPA– disk or CD–ROM you submit. If EPA Impacts Division (C539–02), Office of HQ–OAR–2004–0305, by one of the cannot read your comment due to Air Quality Planning and Standards, following methods: technical difficulties and cannot contact U.S. Environmental Protection Agency, • http://www.regulations.gov: Follow you for clarification, EPA may not be Research Triangle Park, North Carolina the on-line instructions for submitting able to consider your comment. 27711; telephone number: (919) 541– comments. Electronic files should avoid the use of 5499; fax number: (919) 541–0840; and • E-mail: [email protected], special characters, any form of e-mail address: Attention Docket ID Number EPA–HQ– encryption, and be free of any defects or [email protected]. For OAR–2004–0305. viruses. For additional information information about the applicability of

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00002 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules 9411

the NESHAP to a particular entity, contact the appropriate person listed in SUPPLEMENTARY INFORMATION: Table 1 to this preamble.

TABLE 1—LIST OF EPA CONTACTS FOR THE NESHAP ADDRESSED IN THIS PROPOSED ACTION

NESHAP for: OECA contact 1 OAQPS contact 2

Primary Lead Smelting ... Maria Malave, (202) 564–7027, [email protected] .. Sharon Nizich, (919) 541–2825, [email protected]. 1 EPA’s Office of Enforcement and Compliance Assurance. 2 EPA’s Office of Air Quality Planning and Standards.

I. Preamble Acronyms and NTTAA National Technology Transfer and C. Overview of the source category and Abbreviations Advancement Act MACT standards OAQPS EPA’s Office of Air Quality V. Analyses Results and Proposed Decisions Several acronyms and terms used to Planning and Standards A. What data were used in our risk describe industrial processes, data OECA EPA’s Office of Enforcement and analyses? inventories, and risk modeling are Compliance Assurance B. What are the results of the risk included in this preamble. While this OMB Office of Management and Budget assessments and analyses? may not be an exhaustive list, to ease PB-HAP Hazardous air pollutants known to C. What are our proposed decisions on risk be persistent and bio-accumulative in the the reading of this preamble and for acceptability and ample margin of environment reference purposes, the following terms safety? POM Polycyclic Organic Matter and acronyms are defined here: RACT Reasonably Available Control D. What are the results and proposed ADAF Age-dependent Adjustment Factors Technology decisions from the technology review? AERMOD Air dispersion model used by the RBLC RACT/BACT/LAER Clearinghouse E. Variability HEM–3 model RFA Regulatory Flexibility Act F. What other actions are we proposing? AEGL Acute Exposure Guideline Levels RfC Reference Concentration VI. Proposed Action ANPRM Advance Notice of Proposed RfD Reference Dose A. What actions are we proposing as a Rulemaking RTR Residual Risk and Technology Review result of the residual risk reviews? BACT Best Available Control Technology SAB Science Advisory Board B. What actions are we proposing as a CAA Clean Air Act SBA Small Business Administration result of the technology reviews? CBI Confidential Business Information SCC Source Classification Codes C. What other actions are we proposing? CEEL Community Emergency Exposure SF3 2000 Census of Population and D. Compliance Dates Levels Housing Summary File 3 VII. Request for Comments CEMS Continuous Emissions Monitoring SIP State Implementation Plan VIII.Submitting Data Corrections System SOP Standard Operating Procedures IX. Statutory and Executive Order Reviews CERMS Continuous Emission Rate SSM Startup, Shutdown, and Malfunction A. Executive Order 12866: Regulatory Monitoring System TOSHI Target Organ-Specific Hazard Index Planning and Review CFR Code of Federal Regulations TPY Tons Per Year EJ Environmental Justice B. Paperwork Reduction Act TRIM Total Risk Integrated Modeling C. Regulatory Flexibility Act EPA Environmental Protection Agency System ERPG Emergency Response Planning D. Unfunded Mandates Reform Act TTN Technology Transfer Network E. Executive Order 13132: Federalism Guidelines UF Uncertainty Factor F. Executive Order 13175: Consultation HAP Hazardous Air Pollutants UMRA Unfunded Mandates Reform Act and Coordination With Indian Tribal HI Hazard Index URE Unit Risk Estimate Governments HEM–3 Human Exposure Model version 3 VOC Volatile Organic Compounds HON Hazardous Organic National VOHAP Volatile Organic Hazardous Air G. Executive Order 13045: Protection of Emissions Standards for Hazardous Air Pollutants Children From Environmental Health Pollutants WWW Worldwide Web Risks and Safety Risks HQ Hazard Quotient H. Executive Order 13211: Actions IRIS Integrated Risk Information System Organization of this Document. The Concerning Regulations That Km Kilometer following outline is provided to aid in Significantly Affect Energy Supply, LAER Lowest Achievable Emission Rate the location of information in this Distribution, or Use LOAEL Lowest Observed Adverse Effect preamble. I. National Technology Transfer and Level Advancement Act MACT Maximum Achievable Control I. Preamble Acronyms and Abbreviations II. General Information J. Executive Order 12898: Federal Actions Technology To Address Environmental Justice in MACT Code Code within the NEI used to A. Does this action apply to me? Minority Populations and Low-Income identify processes included in a source B. Where can I get a copy of this document Populations category and other related information? MIR Maximum Individual Risk C. What should I consider as I prepare my II. General Information NAAQS National Ambient Air Quality comments for EPA? Standard III. Background A. Does this action apply to me? NAC/AEGL Committee National Advisory A. What is the statutory authority for this Committee for Acute Exposure Guideline action? The regulated industrial source Levels for Hazardous Substances B. How did we consider the risk results in category that is the subject of this NAICS North American Industry making decisions for this proposal? proposal is listed in Table 2 to this Classification System C. What other actions are we addressing in preamble. Table 2 is not intended to be NAS National Academy of Sciences this proposal? exhaustive, but rather provides a guide IV. Analyses Performed and Background for NATA National Air Toxics Assessment for readers regarding entities likely to be NESHAP National Emissions Standards for the Source Category and MACT Standard Hazardous Air Pollutants A. How did we estimate risks posed by the affected by this proposed action for the NEI National Emissions Inventory source category? source categories listed. This standard, NOAEL No Observed Adverse Effects Level B. How did we perform the technology and any changes considered in this NRC National Research Council review? rulemaking, would be directly

VerDate Mar<15>2010 17:37 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00003 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 9412 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules

applicable to sources as a Federal to, the following smelting processes: the 1992 listing will remain Primary program. Thus, Federal, State, local, and sintering, reduction, preliminary Lead Smelting (as in 1992 listing) we are tribal government entities are not treatment, and refining operations.1 As proposing to change the title of the rule affected by this proposed action. As discussed in section III. (C)(3), to be to refer to Primary Lead Processing. For defined in the source category listing consistent with the 1992 listing, EPA is clarification purposes, all references to report published by EPA in 1992, the proposing to change the applicability of lead emissions in this preamble means Primary Lead Smelting source category the Primary Lead Smelting NESHAP to ‘‘lead compounds’’ (which is a HAP) and is defined as any facility engaged in apply to any facility that produces lead all reference to lead production means producing lead metal from ore metal from lead ore concentrates. elemental lead (which is not a HAP) as concentrates; including, but not limited Although the source category name in provided under CAA 112(b)(7)).

TABLE 2—NESHAP AND INDUSTRIAL SOURCE CATEGORIES AFFECTED BY THIS PROPOSED ACTION

Source category NESHAP NAICS code 1 MACT code 2

Primary Lead Smelting ...... Primary Lead Processing ...... 331419 0204 1 North American Industry Classification System. 2 Maximum Achievable Control Technology.

B. Where can I get a copy of this inclusion in the public docket. If you commonly referred to as maximum document and other related submit a CD ROM or disk that does not achievable control technology (MACT) information? contain CBI, mark the outside of the standards. In addition to being available in the disk or CD ROM clearly that it does not MACT standards must require the docket, an electronic copy of this contain CBI. Information not marked as maximum degree of emission reduction proposal will also be available on the CBI will be included in the public through the application of measures, World Wide Web (WWW) through the docket and EPA’s electronic public processes, methods, systems, or Technology Transfer Network (TTN). docket without prior notice. Information techniques, including, but not limited Following signature by the EPA marked as CBI will not be disclosed to, measures which (A) Reduce the Administrator, a copy of this proposed except in accordance with procedures volume of or eliminate pollutants action will be posted on the TTN’s set forth in 40 CFR part 2. Send or through process changes, substitution of policy and guidance page for newly deliver information identified as CBI materials or other modifications; proposed or promulgated rules at the only to the following address: Roberto (B) enclose systems or processes to following address: http://www.epa.gov/ Morales, OAQPS Document Control eliminate emissions; (C) capture or treat ttn/atw/rrisk/rtrpg.html. The TTN Officer (C404–02), Office of Air Quality pollutants when released from a provides information and technology Planning and Standards, U.S. process, stack, storage, or fugitive exchange in various areas of air Environmental Protection Agency, emissions point; (D) are design, pollution control. Research Triangle Park, North Carolina equipment, work practice, or Additional information is available on 27711, Attention Docket ID Number operational standards (including the residual risk and technology review EPA–HQ–OAR–2004–0305. requirements for operator training or (RTR) Web page at http://www.epa.gov/ III. Background certification); or (E) are a combination of ttn/atw/rrisk/rtrpg.html. This the above. CAA section 112(d)(2)(A)– information includes source category A. What is the statutory authority for (E). The MACT standards may take the descriptions and detailed emissions and this action? form of design, equipment, work other data that were used as inputs to Section 112 of the Clean Air Act practice, or operational standards where the risk assessments. (CAA) establishes a two-stage regulatory EPA first determines either that, (A) a process to address emissions of pollutant cannot be emitted through a C. What should I consider as I prepare hazardous air pollutants (HAP) from conveyance designed and constructed to my comments for EPA? stationary sources. In the first stage, emit or capture the pollutants, or that Submitting CBI. Do not submit after EPA has identified categories of any requirement for, or use of, such a information containing CBI to EPA sources emitting one or more of the HAP conveyance would be inconsistent with through http://www.regulations.gov or listed in section 112(b) of the CAA, law; or (B) the application of e-mail. Clearly mark the part or all of section 112(d) of the CAA calls for us measurement methodology to a the information that you claim to be to promulgate NESHAP for those particular class of sources is not CBI. For CBI information on a disk or sources. ‘‘Major sources’’ are those that practicable due to technological and CD ROM that you mail to EPA, mark the emit or have the potential to emit 10 economic limitations. CAA sections outside of the disk or CD ROM as CBI tons per year (TPY) or more of a single 112(h)(1)–(2). and then identify electronically within HAP or 25 TPY or more of any The MACT ‘‘floor’’ is the minimum the disk or CD ROM the specific combination of HAP. For major sources, control level allowed for MACT information that is claimed as CBI. In these technology-based standards must standards promulgated under CAA addition to one complete version of the reflect the maximum degree of emission section 112(d)(3) and may not be based comment that includes information reductions of HAP achievable (after on cost considerations. For new sources, claimed as CBI, a copy of the comment considering cost, energy requirements, the MACT floor cannot be less stringent that does not contain the information and non-air quality health and than the emission control that is claimed as CBI must be submitted for environmental impacts) and are achieved in practice by the best-

1 USEPA. Documentation for Developing the Initial Source Category List—Final Report, USEPA/ OAQPS, EPA–450/3–91–030, July, 1992.

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00004 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules 9413

controlled similar source. The MACT doing so, EPA may adopt standards on Conference Report). We notified floors for existing sources can be less equal to existing MACT standards if Congress in the Residual Risk Report to stringent than floors for new sources, EPA determines that the existing Congress that we intended to use the but they cannot be less stringent than standards are sufficiently protective. As Benzene NESHAP approach in making the average emissions limitation stated in NRDC v. EPA, 529 F.3d 1077, CAA section 112(f) residual risk achieved by the best-performing 12 1083 (D.C. Dir. 2008), ‘‘If EPA determinations (EPA–453/R–99–001, p. percent of existing sources in the determines that the existing technology- ES–11). category or subcategory (or the best- based standards provide an ‘ample In the Benzene NESHAP, we stated as performing 5 sources for categories or margin of safety,’ then the Agency is an overall objective: subcategories with fewer than 30 free to readopt those standards during * * * in protecting public health with an sources). In developing MACT the residual risk rulemaking.’’ Section ample margin of safety, we strive to provide standards, we must also consider 112(f)(2) of the Clean Air Act further maximum feasible protection against risks to control options that are more stringent states that EPA must also adopt more health from hazardous air pollutants by (1) than the floor. We may establish stringent standards, if necessary, to protecting the greatest number of persons standards more stringent than the floor ‘‘prevent taking into consideration costs, possible to an individual lifetime risk level based on the consideration of the cost of energy, safety, and other relevant no higher than approximately 1-in-1 million; achieving the emissions reductions, any factors, an adverse environmental and (2) limiting to no higher than approximately 1-in-10 thousand [i.e., 100-in- non-air quality health and effect.’’ 2 1 million] the estimated risk that a person environmental impacts, and energy When Section 112(f)(2) of the CAA living near a facility would have if he or she requirements. was enacted in 1990, it expressly were exposed to the maximum pollutant The EPA is then required to review preserved our use of the two-step concentrations for 70 years. these technology-based standards and to process for developing standards to ‘‘ revise them ‘‘as necessary (taking into address any residual risk and our The Agency also stated that, The EPA account developments in practices, interpretation of ‘‘ample margin of also considers incidence (the number of processes, and control technologies)’’ no safety’’ developed in the National persons estimated to suffer cancer or less frequently than every 8 years, under Emission Standards for Hazardous Air other serious health effects as a result of CAA section 112(d)(6). In conducting Pollutants: Benzene Emissions from exposure to a pollutant) to be an this review, EPA is not obliged to Maleic Anhydride Plants, Ethylbenzene/ important measure of the health risk to completely recalculate the prior MACT Styrene Plants, Benzene Storage Vessels, the exposed population. Incidence determination. NRDC v. EPA, 529 F.3d Benzene Equipment Leaks, and Coke measures the extent of health risks to 1077, 1084 (D.C. Cir., 2008). By-Product Recovery Plants (Benzene the exposed population as a whole, by The second stage in standard-setting NESHAP) (54 FR 38044, September 14, providing an estimate of the occurrence focuses on reducing any remaining 1989). The first step in this process is of cancer or other serious health effects ’’ ‘‘residual’’ risk according to CAA section the determination of acceptable risk. in the exposed population. The Agency ‘‘ 112(f). This provision requires, first, that The second step provides for an ample went on to conclude that estimated EPA prepare a Report to Congress margin of safety to protect public health, incidence would be weighed along with discussing (among other things) which is the level at which the other health risk information in judging ’’ methods of calculating the risks posed standards are set (unless a more acceptability. As explained more fully (or potentially posed) by sources after stringent standard is required to in our Residual Risk Report to Congress, ‘‘ implementation of the MACT standards, prevent, taking into consideration costs, EPA does not define rigid line[s] of ’’ the public health significance of those energy, safety, and other relevant acceptability, but considers rather risks, and the recommendations factors, an adverse environmental broad objectives to be weighed with a regarding legislation of such remaining effect). series of other health measures and risk. EPA prepared and submitted this The terms ‘‘individual most exposed,’’ factors (EPA–453/R–99–001, p. ES–11). report (Residual Risk Report to ‘‘acceptable level,’’ and ‘‘ample margin of The determination of what represents an Congress, EPA–453/R–99–001) in March safety’’ are not specifically defined in ‘‘acceptable’’ risk is based on a judgment 1999. Congress did not act in response the CAA. However, CAA section of ‘‘what risks are acceptable in the to the report, thereby triggering EPA’s 112(f)(2)(B) preserves the interpretation world in which we live’’ (Residual Risk obligation under CAA section 112(f)(2) set out in the Benzene NESHAP, and the Report to Congress, p. 178, quoting the to analyze and address residual risk. Court in NRDC v. EPA, concluded that Vinyl Chloride decision at 824 F.2d Section 112(f)(2) of the CAA requires EPA’s interpretation of subsection 1165) recognizing that our world is not us to determine for source categories 112(f)(2) is a reasonable one. See NRDC risk-free. subject to certain MACT standards, v. EPA, 529 F.3d at 1083 (D.C. Cir. In the Benzene NESHAP, we stated ‘‘ whether the emissions standards 2008), which says ‘‘[S]ubsection that EPA will generally presume that if provide an ample margin of safety to 112(f)(2)(B) expressly incorporates the risk to [the maximum exposed] protect public health. If the MACT EPA’s interpretation of the Clean Air individual is no higher than standards that apply to a source Act from the Benzene standard, approximately 1-in-10 thousand, that ’’ category emitting a HAP that is complete with a citation to the Federal risk level is considered acceptable. 54 ‘‘ classified as a known, probable, or Register.’’ See also, A Legislative History FR 38045. We discussed the maximum possible human carcinogen do not of the Clean Air Act Amendments of individual lifetime cancer risk as being ‘‘ reduce lifetime excess cancer risks to 1990, volume 1, p. 877 (Senate debate the estimated risk that a person living the individual most exposed to near a plant would have if he or she emissions from a source in the category 2 ‘‘Adverse environmental effect’’ is defined in were exposed to the maximum pollutant or subcategory to less than one-in-one CAA section 112(a)(7) as any significant and concentrations for 70 years.’’ Id. We million,’’ EPA must promulgate residual widespread adverse effect, which may be explained that this measure of risk ‘‘is risk standards for the source category (or reasonably anticipated to wildlife, aquatic life, or an estimate of the upper bound of risk natural resources, including adverse impacts on subcategory) as necessary to provide an populations of endangered or threatened species or based on conservative assumptions, ample margin of safety to protect public significant degradation of environmental qualities such as continuous exposure for 24 health. CAA section 112(f)(2)(A). In over broad areas. hours per day for 70 years.’’ Id. We

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00005 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 9414 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules

acknowledge that maximum individual uncertainties, and any other relevant and considered risk estimation lifetime cancer risk ‘‘does not factors. Considering all of these factors, uncertainties. EPA is providing this necessarily reflect the true risk, but the Agency will establish the standard same type of information in support of displays a conservative risk level which at a level that provides an ample margin the proposed actions described in this is an upper-bound that is unlikely to be of safety to protect the public health, as Federal Register notice. exceeded.’’ Id. required by CAA section 112(f). 54 FR The Agency is considering all Understanding that there are both 38046. available health information to inform benefits and limitations to using our determinations of risk acceptability maximum individual lifetime cancer B. How did we consider the risk results and ample margin of safety under CAA risk as a metric for determining in making decisions for this proposal? section 112(f). Specifically, as explained acceptability, we acknowledged in the As discussed in section III.A of this in the Benzene NESHAP, ‘‘the first step 1989 Benzene NESHAP that preamble, we apply a two-step process judgment on acceptability cannot be ‘‘consideration of maximum individual for developing standards to address reduced to any single factor’’ and thus risk * * * must take into account the residual risk. In the first step, EPA ‘‘[t]he Administrator believes that the strengths and weaknesses of this determines if risks are acceptable. This acceptability of risk under section 112 is measure of risk.’’ Id. Consequently, the determination ‘‘considers all health best judged on the basis of a broad set presumptive risk level of 100-in-1 information, including risk estimation of health risk measures and million (1-in-10 thousand) provides a uncertainty, and includes a presumptive information.’’ 54 FR at 38046. Similarly, benchmark for judging the acceptability limit on maximum individual lifetime with regard to making the ample margin of maximum individual lifetime cancer [cancer] risk (MIR) 3 of approximately 1- of safety determination, as stated in the risk, but does not constitute a rigid line in-10 thousand [i.e., 100-in-1 million].’’ Benzene NESHAP ‘‘[I]n the ample for making that determination. 54 FR 38045. In the second step of the margin decision, the Agency again The Agency also explained in the process, EPA sets the standard at a level considers all of the health risk and other 1989 Benzene NESHAP the following: that provides an ample margin of safety health information considered in the ‘‘In establishing a presumption for MIR ‘‘in consideration of all health first step. Beyond that information, [maximum individual cancer risk], information, including the number of additional factors relating to the rather than a rigid line for acceptability, persons at risk levels higher than appropriate level of control will also be the Agency intends to weigh it with a approximately 1-in-1 million, as well as considered, including cost and series of other health measures and other relevant factors, including costs economic impacts of controls, factors. These include the overall and economic impacts, technological technological feasibility, uncertainties, incidence of cancer or other serious feasibility, and other factors relevant to and any other relevant factors.’’ Id. health effects within the exposed each particular decision.’’ Id. The Agency acknowledges that population, the numbers of persons In past residual risk actions, EPA has flexibility is provided by the Benzene exposed within each individual lifetime presented and considered a number of NESHAP regarding what factors EPA risk range and associated incidence human health risk metrics associated might consider in making within, typically, a 50-kilometer (km) with emissions from the category under determinations and how they might be exposure radius around facilities, the review, including: the MIR; the numbers weighed for each source category. In science policy assumptions and of persons in various risk ranges; cancer responding to comment on our policy estimation uncertainties associated with incidence; the maximum non-cancer under the Benzene NESHAP, EPA the risk measures, weight of the hazard index (HI); and the maximum explained that: ‘‘The policy chosen by scientific evidence for human health acute non-cancer hazard (72 FR 25138, the Administrator permits consideration effects, other quantified or unquantified May 3, 2007; 71 FR 42724, July 27, of multiple measures of health risk. Not health effects, effects due to co-location 2006). In our most recent proposals (75 only can the MIR figure be considered, of facilities, and co-emission of FR 65068, October 21, 2010 and 75 FR but also incidence, the presence of non- pollutants.’’ Id. 80220, December 21, 2010), EPA also cancer health effects, and the In some cases, these health measures presented and considered additional uncertainties of the risk estimates. In and factors taken together may provide measures of health information, this way, the effect on the most exposed a more realistic description of the including: estimates of ‘‘facility-wide’’ individuals can be reviewed as well as magnitude of risk in the exposed risks (risks from all HAP emissions from the impact on the general public. These population than that provided by the facility at which the source category factors can then be weighed in each maximum individual lifetime cancer is located); 4 demographic analyses individual case. This approach complies risk alone. As explained in the Benzene (analyses of the distributions of HAP- with the Vinyl Chloride mandate that NESHAP, ‘‘[e]ven though the risks related risks across different social, the Administrator ascertain an judged ‘‘acceptable’’ by EPA in the first demographic, and economic groups acceptable level of risk to the public by step of the Vinyl Chloride inquiry are living near the facilities); and estimates employing [her] expertise to assess already low, the second step of the of the risks associated with the available data. It also complies with the inquiry, determining an ‘‘ample margin maximum level of emissions which Congressional intent behind the CAA, of safety,’’ again includes consideration might be allowed by the current MACT which did not exclude the use of any of all of the health factors, and whether standards (see, e.g., 75 FR 65068, particular measure of public health risk to reduce the risks even further.’’ In the October 21, 2010 and 75 FR 80220, from the EPA’s consideration with ample margin of safety decision process, December 21, 2010). EPA also discussed respect to CAA section 112 regulations, the Agency again considers all of the and, thereby, implicitly permits health risks and other health 3 Although defined as ‘‘maximum individual consideration of any and all measures of information considered in the first step. risk,’’ MIR refers only to cancer risk. MIR, one health risk which the Administrator, in Beyond that information, additional metric for assessing cancer risk, is the estimated [her] judgment, believes are appropriate factors relating to the appropriate level risk were an individual exposed to the maximum to determining what will ‘protect the level of a pollutant for a lifetime. of control will also be considered, 4 EPA previously provided estimates of total public health.’ ’’ 54 FR at 38057. including costs and economic impacts facility risk in a residual risk proposal for coke oven For example, the level of the MIR is of controls, technological feasibility, batteries (69 FR 48338, August 9, 2004). only one factor to be weighed in

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00006 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules 9415

determining acceptability of risks. It is of adverse non-cancer health effects in C. What other actions are we addressing explained in the Benzene NESHAP that a population, the exposures resulting in this proposal? ‘‘an MIR of approximately 1-in-10 from emissions from the facility in 1. Startup, Shutdown and Malfunction thousand should ordinarily be the upper combination with emissions from all of end of the range of acceptability. As the other sources (e.g., other facilities) to This proposed action would amend risks increase above this benchmark, which an individual is exposed may be the provisions of the existing NESHAP that apply to periods of startup, they become presumptively less sufficient to result in increased risk of shutdown, and malfunction (SSM). The acceptable under CAA section 112, and adverse non-cancer health effects. In would be weighed with the other health proposed revisions of these provisions May 2010, the Science Advisory Board result from a Court decision that vacated risk measures and information in (SAB) advised us ‘‘* * * that RTR making an overall judgment on portions of two provisions in EPA’s assessments will be most useful to acceptability. Or, the Agency may find, ‘‘General Provisions’’ regulation under in a particular case, that a risk that decision makers and communities if CAA section 112, governing the includes MIR less than the results are presented in the broader emissions of HAP during periods of presumptively acceptable level is context of aggregate and cumulative SSM. The current Primary Lead unacceptable in the light of other health risks, including background Smelting MACT includes references to risk factors.’’ Id. at 38045. Similarly, concentrations and contributions from the vacated provisions in the General with regard to the ample margin of other sources in the area.’’ 5 Provisions rule. safety analysis, EPA stated in the While we are interested in placing We are proposing to revise the Benzene NESHAP that: ‘‘* * * EPA source category and facility-wide HAP Primary Lead Smelting MACT standard to require affected sources to comply believes the relative weight of the many risks in the context of total HAP risks with the emission limitations at all factors that can be considered in from all sources combined in the times and during periods of SSM. selecting an ample margin of safety can vicinity of each source, we are only be determined for each specific Specifically, we are proposing several concerned about the uncertainties of revisions to subpart TTT including source category. This occurs mainly doing so. At this point, we believe that because technological and economic revising Table 1 to indicate that the such estimates of total HAP risks will factors (along with the health-related requirements of the General Provisions factors) vary from source category to have significantly greater associated pertaining to SSM do not apply and to source category.’’ Id. at 38061. uncertainties than for the source revise language in § 63.1547 (g)(1) and EPA wishes to point out that certain category or facility-wide estimates, and (2) to remove the exemption for bag leak health information has not been hence compounding the uncertainty in detection alarm time attributable to SSM considered to date in making residual any such comparison. This is because from total allowed alarm time. For risk determinations. In assessing risks to we have not conducted a detailed reasons discussed below, we are also populations in the vicinity of the technical review of HAP emissions data proposing to promulgate an affirmative facilities in each category, we present for source categories and facilities that defense to civil penalties for estimates of risk associated with HAP have not previously undergone an RTR exceedances of emission standards emissions from the source category review or are not currently undergoing caused by malfunctions, as well as alone (source category risk estimates) such review. We are requesting criteria for establishing the affirmative and HAP emissions from the entire comment on whether and how best to defense. These changes would go into facility at which the covered source estimate and evaluate total HAP effect upon the effective date of category is located (facility-wide risk exposure in our assessments, and, in promulgation of the final rule. The United States Court of Appeals estimates). We do not attempt to particular, on whether and how it might characterize the risks associated with all for the District of Columbia Circuit be appropriate to use information from vacated portions of two provisions in HAP emissions impacting the EPA’s National Air Toxics Assessment populations living near the sources in EPA’s CAA Section 112 regulations (NATA) to support such estimates. We these categories. That is, at this time, we governing the emissions of HAP during are also seeking comment on how best do not attempt to quantify those HAP periods of SSM. Sierra Club v. EPA, 551 risks that may be associated with mobile to consider various types and scales of F.3d 1019 (D.C. Cir. 2008), cert. denied, source emissions, natural source risk estimates when making our 130 S. Ct. 1735 (U.S. 2010). Specifically, emissions, persistent environmental acceptability and ample margin of safety the Court vacated the SSM exemptions pollution, or atmospheric determinations under CAA section contained in 40 CFR 63.6(f)(1) and 40 transformation in the vicinity of the 112(f). Additionally, we are seeking CFR 63.6(h)(1), that are part of a sources in these categories. comments and recommendations for regulation commonly known as the The Agency understands the potential any other comparative measures that ‘‘General Provisions Rule,’’ that EPA had importance of considering an may be useful in the assessment of the promulgated under section 112 of the individual’s total exposure to HAP in distribution of HAP risks across CAA. When incorporated into CAA addition to considering exposure to potentially affected demographic section 112(d) regulations for specific HAP emissions from the source category groups. source categories, these two provisions and facility. This is particularly exempt sources from the requirement to important when assessing non-cancer comply with the otherwise applicable risks, where pollutant-specific exposure 5 EPA’s responses to this and all other key CAA section 112(d) emission standard levels (e.g., Reference Concentration recommendations of the SAB’s advisory on RTR during periods of SSM. (RfC)) are based on the assumption that risk assessment methodologies (which is available We are proposing the elimination of thresholds exist for adverse health at: http://yosemite.epa.gov/sab/sabproduct.nsf/ the SSM exemption in this rule. effects. For example, the Agency 4AB3966E263D943A8525771F00668381/$File/EPA- Consistent with Sierra Club v. EPA, EPA SAB-10-007-unsigned.pdf) are outlined in a memo recognizes that, although exposures to this rulemaking docket from David Guinnup is proposing standards in this rule that attributable to emissions from a source entitled, EPA’s Actions in Response to the Key apply at all times. We are also proposing category or facility alone may not Recommendations of the SAB Review of RTR Risk several revisions to Table 1 (the General indicate the potential for increased risk Assessment Methodologies. Provisions Applicability table). For

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00007 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 9416 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules

example, we are proposing to eliminate situations. After a certain point, the Maintenance, and Malfunctions (Feb. the incorporation of the General transgression of regulatory limits caused 15, 1983).) EPA is therefore proposing to Provisions’ requirement that the source by ‘uncontrollable acts of third parties,’ add to the final rule an affirmative develop an SSM plan. We also are such as strikes, sabotage, operator defense to civil penalties for proposing to eliminate or revise certain intoxication or insanity, and a variety of exceedances of emission limits that are recordkeeping and reporting that relate other eventualities, must be a matter for caused by malfunctions. See 40 CFR to the SSM exemption. EPA has the administrative exercise of case-by- 63.1542 (defining ‘‘affirmative defense’’ attempted to ensure that we have not case enforcement discretion, not for to mean, in the context of an included in the proposed regulatory specification in advance by enforcement proceeding, a response or language any provisions that are regulation.’’). defense put forward by a defendant, inappropriate, unnecessary, or Further, it is reasonable to interpret regarding which the defendant has the redundant in the absence of the SSM section 112(d) as not requiring EPA to burden of proof, and the merits of which exemption. We are specifically seeking account for malfunctions in setting are independently and objectively comment on whether there are any such emission standards. For example, we evaluated in a judicial or administrative provisions that we have inadvertently note that CAA section 112 uses the proceeding). We also are proposing incorporated or overlooked. concept of ‘‘best performing’’ sources in other regulatory provisions to specify In proposing standards in this rule, defining MACT, the level of stringency the elements that are necessary to EPA has taken into account startup and that major source standards must meet. establish this affirmative defense; the shutdown periods and, for the reasons Applying the concept of ‘‘best source must prove by a preponderance explained below, has not proposed performing’’ to a source that is of the evidence that it has met all of the different standards for those periods. malfunctioning presents significant elements set forth in § 63.1551. (See 40 Information on periods of startup and difficulties. The goal of best performing CFR 22.24.) The criteria ensure that the shutdown in the industry indicate that sources is to operate in such a way as affirmative defense is available only emissions during these periods do not to avoid malfunctions of their units. where the event that causes an increase. Furthermore, all processes are Moreover, even if malfunctions were exceedance of the emission limit meets controlled by either control devices or considered a distinct operating mode, the narrow definition of malfunction in work practices and these controls would we believe it would be impracticable to 40 CFR 63.2 (sudden, infrequent, not not typically be affected by an SSM take malfunctions into account in reasonably preventable and not caused event. Also, compliance with the setting CAA section 112(d) standards for by poor maintenance and/or careless standard already requires averaging of Primary Lead Smelting. As noted above, operation). For example, to successfully emissions over a three month period, by definition, malfunctions are sudden assert the affirmative defense, the source which accounts for the variability of and unexpected events and it would be must prove by a preponderance of the difficult to set a standard that takes into emissions that may result during evidence that excess emissions ‘‘[w]ere account the myriad different types of periods of startup and shutdown. caused by a sudden, short, infrequent, malfunctions that can occur across all Therefore, separate standards for and unavoidable failure of air pollution sources in the category. Moreover, periods of startup and shutdown are not control and monitoring equipment, malfunctions can vary in frequency, being proposed. process equipment, or a process to Periods of startup, normal operations, degree, and duration, further operate in a normal or usual manner and shutdown are all predictable and complicating standard setting. ***’’ The criteria also are designed to routine aspects of a source’s operations. In the unlikely event that a source ensure that steps are taken to correct the However, by contrast, malfunction is fails to comply with the applicable CAA malfunction, to minimize emissions in defined as a ‘‘sudden, infrequent, and section 112(d) standards as a result of a not reasonably preventable failure of air malfunction event, EPA would accordance with §§ 63.1543(i) and pollution control and monitoring determine an appropriate response 63.1544(e) and to prevent future equipment, process equipment, or a based on, among other things, the good malfunctions. For example, the source process to operate in a normal or useful faith efforts of the source to minimize must prove by a preponderance of the ‘‘ manner * * *’’ (40 CFR 63.2). EPA has emissions during malfunction periods, evidence that [r]epairs were made as determined that malfunctions should including preventative and corrective expeditiously as possible when the not be viewed as a distinct operating actions, as well as root cause analyses applicable emission limitations were mode and, therefore, any emissions that to ascertain and rectify excess being exceeded * * *’’ and that ‘‘[a]ll occur at such times do not need to be emissions. EPA would also consider possible steps were taken to minimize factored into development of CAA whether the source’s failure to comply the impact of the excess emissions on section 112(d) standards, which, once with the CAA section 112(d) standard ambient air quality, the environment promulgated, apply at all times. In was, in fact, ‘‘sudden, infrequent, not and human health * * *.’’ In any Mossville Environmental Action Now v. reasonably preventable’’ and was not judicial or administrative proceeding, EPA, 370 F.3d 1232, 1242 (D.C. Cir. instead ‘‘caused in part by poor the Administrator may challenge the 2004), the court upheld as reasonable maintenance or careless operation.’’ 40 assertion of the affirmative defense and, standards that had factored in CFR 63.2 (definition of malfunction). if the respondent has not met its burden variability of emissions under all Finally, EPA recognizes that even of proving all of the requirements in the operating conditions. However, nothing equipment that is properly designed and affirmative defense, appropriate in section 112(d) or in case law requires maintained can sometimes fail and that penalties may be assessed in accordance that EPA anticipate and account for the such failure can sometimes cause or with section 113 of the Clean Air Act innumerable types of potential contribute to an exceedance of the (see also 40 CFR part 22.77). malfunction events in setting emission relevant emission standard. (See, e.g., Specifically, we are proposing the standards. See, Weyerhaeuser v. Costle, State Implementation Plans: Policy following changes to the rule. 590 F.2d 1011, 1058 (D.C. Cir. 1978)(‘‘In Regarding Excessive Emissions During • Added general duty requirements in the nature of things, no general limit, Malfunctions, Startup, and Shutdown §§ 63.1543 and 63.1544 to replace individual permit, or even any upset (Sept. 20, 1999); Policy on Excess General Provision requirements that provision can anticipate all upset Emissions During Startup, Shutdown, reference vacated SSM provisions.

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00008 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules 9417

• Added replacement language that the control of metal HAP are the same to produce lead metal. Under the eliminates the reference to SSM as those used to control lead emissions. current applicability provisions, the exemptions applicable to performance Therefore, EPA expected that the affected sources include any sinter tests in § 63.1546. standards requiring control of lead machine, blast furnace, dross furnace, • Added paragraphs in § 63.1549(e) would achieve similar control of the process fugitive source, and fugitive requiring the reporting of malfunctions other metal HAP emitted from primary dust source located at a primary lead as part of the affirmative defense lead smelters. No adverse comments smelter and excludes secondary lead provisions. were received regarding EPA’s proposed smelters, lead refiners, or lead remelters. • Added paragraphs in § 63.1549(b) rationale for relying on lead as a Combined with the current definition requiring the keeping of certain records surrogate for other metal HAP emitted for ‘‘primary lead smelter,’’ the current during malfunctions as part of the by these sources and EPA adopted that rule effectively only applies to facilities affirmative defense provisions. rationale in the final rule promulgating that produce lead metal from lead • Revised Table 1 to reflect changes the Primary Lead Smelting MACT. The sulfide ore concentrates using in the applicability of the General petitioners do not have any substantive pyrometallurgical techniques. While the Provisions to this subpart resulting from basis as to why EPA’s rationale is not only processes available for the a court vacatur of certain SSM supported. Nor do they claim that there production of lead from lead ore requirements in the General Provisions. is any new information that would concentrate at the time the MACT rule 2. Lead as a Surrogate and Regulation of support re-opening this issue. Thus they was developed were pyrometallurgical Volatile Organic Compounds (VOC) and fail to present a basis for re-opening this techniques, that applicability language Acid Gas Emissions issue. is narrower than the primary lead The petitioners also insist that EPA smelting source category description In a January 14, 2009, petition for should have set standards for VOC and rulemaking filed by the Natural EPA identified in its source category acid gases that are HAP because lead listing issued pursuant to CAA section Resources Defense Council and Sierra would not be a surrogate for these Club, the petitioners claim that for the 112(c)(1), Documentation for pollutants. EPA noted in the original Developing the Initial Source Category Primary Lead Smelting MACT, EPA proposal that due to small amounts of relied on lead as a surrogate for all HAP List (EPA–450/3–91–030, July 1992). In coke fed to the blast furnace, organic the source category listing, EPA defined and they claim that it was inappropriate HAP (VOC) was emitted at a rate so low for EPA to do so in absence of a showing the primary lead smelting source as to be infeasible to reduce. Again, no category as follows: ‘‘The Primary Lead that lead is an appropriate surrogate for adverse comments were received on all other HAP (such as mercury, acid Smelting source category includes any EPA’s proposed conclusions, which facility engaged in producing lead metal gases, and volatile organic compounds were adopted in the final rule, and the (VOC)). The petitioners asserted that from ore concentrates. The category petitioners do not now provide includes, but is not limited to, the EPA should set standards for other HAP substantive support for their claim. Nor absent a showing that lead is an following smelting processes: sintering do they explain why any such claim reduction, preliminary treatment, and appropriate surrogate for these HAP. could not have been raised during the They also assert that EPA’s PM standard refining operations. The sintering initial rulemaking. Thus, they fail to process includes an updraft or does not reflect the emission level present a basis for re-opening the rule achieved by the best performing sources downdraft sintering machine. The on this issue. reduction process includes the blast and that EPA must re-open the rule to Finally, petitioners claim that the ‘‘PM furnace, electric smelting furnace with a set floors for PM in accordance with standard does not reflect the emission converter or reverberatory furnace, and CAA section 112(d)(3). A copy of the level achieved by the best performing petition is included in the docket. sources.’’ This claim is unclear as there slag fuming furnace process units. The As part of this rulemaking, EPA is is no PM standard in the Primary Lead preliminary treatment process includes responding to the claims made by the Smelting MACT. The monitoring the drossing kettles and dross petitioners regarding the Primary Lead provisions provide that PM should be reverberatory furnace process units. The Smelting MACT. measured in relation to a predetermined refining process includes the refinery As an initial matter, the petitioners PM level as one test for indicating process unit.’’ The definition is clear are incorrect in their claim that EPA baghouse performance. However, the that the primary intent was to cover considers lead as a surrogate for all PM levels are not enforceable emission sources that produce lead metal from HAP. Rather, EPA used lead as a limits, but merely an indication that the ore concentrates, which would surrogate only for other metal HAP baghouse may not be operating ‘‘include’’ the use of a pyrometallurgical compounds in establishing the properly. Again, these provisions were process, but would not be limited to emissions limit in the current MACT clearly explained in the proposed and such. As noted previously, at the time standard for this source category (63 FR final Primary Lead Smelting MACT we promulgated the MACT standard, 19206 and 64 FR 30195). EPA rulemakings. Any claims concerning the the only method of producing lead determined in the 1999 rule that lead, appropriateness of these monitoring metal from ore concentrates was a nonvolatile metal HAP, is an requirements should have been raised through use of pyrometallurgical appropriate surrogate for other during the initial rulemaking process. techniques and we adopted an nonvolatile metal HAP including Petitioners do not claim any new applicability provision that focused on antimony, arsenic, chromium, nickel, grounds for raising this issue now. that process. manganese, and cadmium. In the Thus, the petition fails to provide a However, information provided by the proposed rule for the Primary Lead basis for re-opening the MACT. sole operating primary lead smelting Smelting MACT (63 FR 19206), EPA facility indicates that lead production is discussed the use of lead as a surrogate 3. Modification of the Applicability likely to continue at the current Doe for metal HAP emissions and explained Provision Run facility, although using a process that strong correlations exist between EPA is proposing to amend the other than a pyrometallurgical emissions of lead and other metal HAP applicability section to apply to any technique. The new lead facility would and that the technologies identified for facility processing lead ore concentrate continue to process lead ore concentrate

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00009 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 9418 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules

in order to produce lead metal. Based on soliciting comment on whether it A. How did we estimate risks posed by the current applicability section and should apply. the source category? definitions, it could be interpreted that We are also taking this opportunity to the future lead producing process, using clarify the reference to ‘‘lead refiners’’ in The EPA conducted a risk assessment techniques other than the second sentence of the applicability that provided estimates of the MIR pyrometallurgical, would not be subject section, which provides that the MACT posed by the HAP emissions from the to the NESHAP for primary lead standard does not apply to ‘‘secondary one source in the source category, the smelters. Such a limited interpretation lead smelters, lead refiners, or lead distribution of cancer risks within the is not consistent with EPA’s intent as remelters.’’ The intent of this provision exposed populations, cancer incidence, evidenced by the broader definition in was to make clear that secondary lead HI for chronic exposures to HAP with the source category list. Therefore, EPA smelters would not be subject to the rule the potential to cause non-cancer health is proposing to amend the applicability because secondary lead smelters were effects, hazard quotients (HQ) for acute section to specify that the MACT listed as a separate source category and exposures to HAP with the potential to applies to any lead processing facility addressed in a separate MACT standard. cause non-cancer health effects, and an that produces lead metal from lead ore With regard to lead refiners and lead evaluation of the potential for adverse concentrate. Consistent with the remelters, the intent was to provide that environmental effects. The risk proposed revision to the applicability these activities, to the extent that they assessments consisted of seven primary section, we are proposing to remove the are not located at facilities that produce steps, as discussed below. definition of ‘‘primary lead smelter’’ and lead from lead ore concentrate, would The docket for this rulemaking add a definition of ‘‘primary lead not be subject to the Primary Lead contains the following document which processor’’ which means any facility Smelting MACT. However, it was not provides more information on the risk engaged in the production of lead metal the intention of the rule to exempt kettle assessment inputs and models: Draft from lead sulfide ore concentrates refining operations included as part of Residual Risk Assessment for the through the use of pyrometallurgical or a primary lead processing facility. Primary Lead Smelting Source Category. other techniques. In addition, we are Therefore, EPA is proposing to add 1. Establishing the Nature and proposing to replace ‘‘primary lead definitions for secondary lead smelters, Magnitude of Actual Emissions and smelter’’ with ‘‘primary lead processor’’ lead refiners, and lead remelters in the Identifying the Emissions Release throughout 40 CFR subpart TTT. definitions section of this NESHAP in Characteristics (§ 63.1541 through § 63.1545, § 63.1547 order to further clarify the exemption in through § 63.1549). We are specifically the applicability provisions with regard For the Primary Lead Smelting source asking for comment on this proposed to these types of facilities. As this category, we compiled a preliminary change in the definition. change only clarifies an existing dataset using readily available Because there is only one primary provision in the rule, there will be no information, reviewed the data, and lead processing facility in the U.S., there impact to the number of facilities made changes where necessary. The will be no impact of this change on the covered by the rule. preliminary dataset was based on data number of existing facilities covered by 4. Other Changes in the 2002 National Emissions the MACT. Inventory (NEI) Final Inventory, Version We note, however, that although we The following lists additional minor 1 (made publicly available on February are changing the applicability section to changes we are proposing. This list 26, 2006). The NEI is a database that clarify that the MACT applies to all includes rule changes that address contains information about sources that processes for producing lead metal from editorial errors and plain language emit criteria air pollutants, their ore concentrates, we are not today revisions. precursors, and HAP. The NEI database proposing a specific MACT standard • As part of EPA’s effort to includes estimates of annual air that would apply to the as-yet incorporate plain language into its pollutant emissions from point, non- undemonstrated hydrometallurgical ‘‘ ’’ regulations, replaced the word shall point, and mobile sources in the 50 process which Doe Run has indicated with ‘‘must.’’ (§ 63.1543 through that it plans to build at the current Doe States, the District of Columbia, Puerto § 63.1550) Rico, and the Virgin Islands. The EPA Run facility. If and when that process • Correction to the original rule collects this information and releases an begins operation, we will consider (‘‘thru’’ replaced with ‘‘through’’ in the updated version of the NEI database whether to revise the MACT standard to definition of ‘‘tapping location’’). every 3 years. specifically address that process or any (§ 63.1542) other new processes. However, the • Minor wording change to definition On December 4, 2009, a CAA Section limits applicable to specific emission of ‘‘fugitive dust source’’ to clarify 114 Information Collection Request sources currently in operation as meaning. (§ 63.1542) (ICR) was issued requesting information specified in the MACT and as revised from the one facility in this source under CAA sections 112(d)(6) and (f)(2) IV. Analyses Performed and category. An updated dataset was in this rulemaking would continue to Background of the Source Category and created through incorporation of apply to any emission source at the MACT Standard changes to the dataset from the ICR data facility that continues in operation, such As discussed above, in this proposed review process and additional as the refinery. In addition, to the extent rule we are proposing action to address information gathered by EPA. The that we establish a final air lead the RTR requirements of CAA sections updated dataset contains information concentration limit as proposed in 112(d)(6) and (f)(2) for the Primary Lead for the one facility in the source § 63.1544, those limits would also Smelting MACT standard. In this category and was used to conduct the continue to apply to the facility. We also section, we describe the analyses risk assessment and other analyses that are proposing that the plant-wide performed to support the proposed form the basis for the proposed risk and emission limit we are proposing today decisions for the RTR for this source technology reviews. A copy of the should continue to apply to any facility category and we also include dataset used and documentation of the that meets the revised applicability background information on the source risk assessment can be found in the definition, but we are specifically category and the MACT standard. docket.

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00010 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules 9419

2. Establishing the Relationship MACT standards required only 92 (i.e., eight three-month periods) from Between Actual Emissions and MACT– percent control, we would estimate that April 1997 through June 1999. To obtain Allowable Emissions Levels MACT-allowable emissions from that one year of meteorological data, we used The available emissions data in the emission point type could be as much the middle portion of these data, the NEI and from other sources typically as 4 times higher (8 percent allowable year 1998, in our modeling. A second represent the estimates of mass of emissions compared with 2 percent library of United States Census Bureau 7 emissions actually emitted during the actually emitted), and the ratio of census block internal point locations specified annual time period. These MACT-allowable to actual would be 4:1 and populations provides the basis of ‘‘actual’’ emission levels are often lower for this emission point type. After human exposure calculations (Census, than the emission levels that a facility developing these ratios for each 2000). In addition, for each census might be allowed to emit and still emission point type at the one facility block, the census library includes the comply with the MACT standards. The in this source category, we next applied elevation and controlling hill height, these ratios to the maximum chronic emissions level allowed to be emitted by which are also used in dispersion risk estimates from the inhalation risk calculations. A third library of pollutant the MACT standards is referred to as the assessment to obtain maximum risk unit risk factors and other health ‘‘MACT-allowable’’ emissions level. This estimates based on MACT-allowable benchmarks is used to estimate health represents the highest emissions level emissions. The estimate of MACT- risks. These risk factors and health that could be emitted by the facility allowable emissions for the Primary benchmarks are the latest values without violating the MACT standards. We discussed the use of both MACT- Lead Smelting source category is recommended by EPA for HAP and described in section V of this preamble. allowable and actual emissions in the other toxic air pollutants. These values are available at http://www.epa.gov/ttn/ final Coke Oven Batteries residual risk 3. Conducting Dispersion Modeling, atw/toxsource/summary.html and are rule (70 FR 19998–19999, April 15, Determining Inhalation Exposures, and discussed in more detail later in this 2005) and in the proposed and final Estimating Individual and Population section. Hazardous Organic NESHAP (HON) Inhalation Risks In developing the risk assessment for residual risk rules (71 FR 34428, June Both long-term and short-term chronic exposures, we used the 14, 2006, and 71 FR 76609, December inhalation exposure concentrations and estimated annual average ambient air 21, 2006, respectively). In those health risks from the source category concentration of each of the HAP previous actions, we noted that addressed in this proposal were emitted by each source for which we assessing the risks at the MACT- estimated using the Human Exposure have emissions data in the source allowable level is inherently reasonable Model (Community and Sector HEM–3 category. The air concentrations at each since these risks reflect the maximum version 1.1.0). The HEM–3 performs nearby census block centroid were used level sources could emit and still three of the primary risk assessment as a surrogate for the chronic inhalation comply with national emission activities listed above: (1) Conducting exposure concentration for all the standards. But we also explained that it dispersion modeling to estimate the people who reside in that census block. is reasonable to consider actual concentrations of HAP in ambient air, We calculated the MIR for the one emissions, where such data are (2) estimating long-term and short-term facility as the cancer risk associated available, in both steps of the risk inhalation exposures to individuals with a lifetime (70-year period) of analysis, in accordance with the residing within 50 km of the modeled exposure to the maximum concentration Benzene NESHAP. (54 FR 38044, sources, and (3) estimating individual at the centroid of an inhabited census September 14, 1989.) It is reasonable to and population-level inhalation risks block. Individual cancer risks were consider actual emissions because using the exposure estimates and calculated by multiplying the estimated sources typically seek to perform better quantitative dose-response information. lifetime exposure to the ambient than required by emission standards to The dispersion model used by HEM– concentration of each of the HAP (in provide an operational cushion to 3 is AERMOD, which is one of EPA’s micrograms per cubic meter) by its Unit accommodate the variability in preferred models for assessing pollutant Risk Estimate (URE), which is an upper manufacturing processes and control concentrations from industrial bound estimate of an individual’s 6 device performance. facilities. To perform the dispersion probability of contracting cancer over a As described above, the actual modeling and to develop the lifetime of exposure to a concentration emissions data were compiled based on preliminary risk estimates, HEM–3 of 1 microgram of the pollutant per the NEI, information gathered from the draws on three data libraries. The first cubic meter of air. In general, for facility and State, and information is a library of meteorological data, residual risk assessments, we use URE received in response to the ICR. To which is used for dispersion values from EPA’s Integrated Risk estimate emissions at the MACT- calculations. This library includes 1 Information System (IRIS). For allowable level, we developed a ratio of year of hourly surface and upper air carcinogenic pollutants without EPA MACT-allowable to actual emissions for observations for 130 meteorological IRIS values, we look to other reputable each source type (i.e., the individual stations, selected to provide coverage of sources of cancer dose-response values, stacks and the aggregate fugitive the United States and Puerto Rico. often using California Environmental emissions) for the one facility in the However, in this instance, site-specific Protection Agency (CalEPA) URE source category. This ratio is based on meteorological data for the one facility values, where available. In cases where the level of control required by the in this source category were supplied by new, scientifically credible dose- MACT standards compared to the level the state of Missouri and used for the response values have been developed in of reported actual emissions and modeling. The data provided by the a manner consistent with EPA available information on the level of state of Missouri were for eight quarters guidelines and have undergone a peer control achieved by the emissions review process similar to that used by controls in use. For example, if there 6 U.S. EPA. Revision to the Guideline on Air was information to suggest that an Quality Models: Adoption of a Preferred General Purpose (Flat and Complex Terrain) Dispersion 7 A census block is generally the smallest emission point type was being Model and Other Revisions (70 FR 68218, geographic area for which census statistics are controlled by 98 percent while the November 9, 2005). tabulated.

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00011 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 9420 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules

EPA, we may use such dose response Screening estimates of acute 9.9 This analysis is provided in values in place of, or in addition to, exposures and risks were also evaluated Appendix 4 of the Draft Residual Risk other values, if appropriate. In this for each of the HAP at the point of Assessment for Primary Lead Smelting review, IRIS values were available for highest off-site exposure for each facility which is available in the docket for this both carcinogenic pollutants (cadmium (i.e., not just the census block centroids) action. Considering this analysis, unless and arsenic) emitted by the facility in assuming that a person was located at specific process knowledge or data are this source category, and therefore IRIS this spot at a time when both the peak available to provide an alternate value, values were used in the assessment. (hourly) emission rate and hourly to account for more than 99 percent of Incremental individual lifetime dispersion conditions occurred. In the peak hourly emissions, we apply a cancer risks associated with emissions general, acute HQ values were conservative screening multiplication from the one source in the source calculated using best available, short- factor of 10 to the average annual hourly category were estimated as the sum of term dose-response value. These acute emission rate in these acute exposure the risks for each of the carcinogenic dose-response values include REL, screening assessments. For the Primary HAP (including those classified as Acute Exposure Guideline Levels Lead Smelting source category, this carcinogenic to humans, likely to be (AEGL), and Emergency Response factor of 10 was applied. In cases where all acute HQ values carcinogenic to humans, and suggestive Planning Guidelines (ERPG) for 1-hour from the screening step were less than evidence of carcinogenic potential 8) exposure durations. Notably, for HAP or equal to 1, acute impacts were emitted by the modeled source. Cancer emitted from this source category, REL deemed negligible and no further incidence and the distribution of values were the only such dose- analysis was performed. In the cases individual cancer risks for the response values available. As discussed where an acute HQ from the screening population within 50 km of the source below, we used conservative step was greater than 1, additional site- were also estimated for the source assumptions for emission rates, specific data were considered to category as part of these assessments by meteorology, and exposure location for develop a more refined estimate of the summing individual risks. A distance of our acute analysis. potential for acute impacts of concern. 50 km is consistent with both the As described in the CalEPA’s Air Ideally, we would prefer to have analysis supporting the 1989 Benzene Toxics Hot Spots Program Risk continuous measurements over time to NESHAP (54 FR 38044) and the Assessment Guidelines, Part I, The see how the emissions vary by each limitations of Gaussian dispersion Determination of Acute Reference hour over an entire year. Having a models, including AERMOD. Exposure Levels for Airborne Toxicants, an acute REL value (http:// frequency distribution of hourly To assess risk of non-cancer health www.oehha.ca.gov/air/pdf/acuterel.pdf) emission rates over a year would allow effects from chronic exposures, we is defined as ‘‘the concentration level at us to perform a probabilistic analysis to summed the HQ for each of the HAP or below which no adverse health estimate potential threshold that affects a common target organ effects are anticipated for a specified exceedances and their frequency of system to obtain the HI for that target exposure duration is termed the REL. occurrence. Such an evaluation could organ system (or target organ-specific REL values are based on the most include a more complete statistical HI, TOSHI). The HQ is the estimated sensitive, relevant, adverse health effect treatment of the key parameters and exposure divided by the chronic reported in the medical and elements adopted in this screening reference value, which is either the EPA toxicological literature. REL values are analysis. However, we recognize that RfC, defined as ‘‘an estimate (with designed to protect the most sensitive having this level of data is rare, hence uncertainty spanning perhaps an order individuals in the population by the our use of the multiplier (i.e., factor of of magnitude) of a continuous inclusion of margins of safety. Since 10) approach in our screening analysis. inhalation exposure to the human margins of safety are incorporated to population (including sensitive 4. Conducting Multipathway Exposure address data gaps and uncertainties, and Risk Modeling subgroups) that is likely to be without exceeding the REL does not an appreciable risk of deleterious effects automatically indicate an adverse health The potential for significant human during a lifetime,’’ or, in cases where an impact. health risks due to exposures via routes RfC is not available, the Agency for To develop screening estimates of other than inhalation (i.e., Toxic Substances and Disease Registry acute exposures, we first developed multipathway exposures) and the (ATSDR) chronic Minimal Risk Level estimates of maximum hourly emission potential for adverse environmental (MRL) or the CalEPA Chronic Reference rates by multiplying the average actual impacts were evaluated in a three-step Exposure Level (REL). The REL is annual hourly emission rates by a factor process. In the first step, we determined defined as ‘‘the concentration level at or to cover routinely variable emissions. whether any facilities emitted any HAP below which no adverse health effects We chose the factor to use based on known to be persistent and bio- are anticipated for a specified exposure process knowledge and engineering accumulative in the environment (PB– duration.’’ judgment and with awareness of a Texas HAP). There are 14 PB–HAP study of short-term emissions compounds or compound classes 8 These classifications also coincide with the variability, which showed that most identified for this screening in EPA’s Air terms ‘‘known carcinogen, probable carcinogen, and peak emission events, in a heavily- Toxics Risk Assessment Library possible carcinogen,’’ respectively, which are the (available at http://www.epa.gov/ttn/ terms advocated in the EPA’s previous Guidelines industrialized 4-county area (Harris, _ _ for Carcinogen Risk Assessment, published in 1986 Galveston, Chambers, and Brazoria fera/risk atra vol1.html). They are (51 FR 33992, September 24, 1986). Summing the Counties, Texas) were less than twice cadmium compounds, chlordane, risks of these individual compounds to obtain the the annual average hourly emission rate. chlorinated dibenzodioxins and furans, cumulative cancer risks is an approach that was dichlorodiphenyldichloroethylene, recommended by the EPA’s SAB in their 2002 peer The highest peak emission event was 74 review of EPA’s NATA entitled, NATA—Evaluating times the annual average hourly heptachlor, hexachlorobenzene, the National-scale Air Toxics Assessment 1996 emission rate, and the 99th percentile 9 Data—an SAB Advisory, available at: http:// ratio of peak hourly emission rate to the See http://www.tceq.state.tx.us/compliance/ yosemite.epa.gov/sab/sabproduct.nsf/214C6E915BB field_ops/eer/index.html or docket to access the 04E14852570CA007A682C/$File/ecadv02001.pdf. annual average hourly emission rate was source of these data.

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00012 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules 9421

hexachlorocyclohexane, lead account all of the elements of the is because the secondary lead NAAQS, compounds, mercury compounds, NAAQS for lead. That is, our estimated which has the same averaging time, methoxychlor, polychlorinated 3-month lead concentrations are form, and level as the primary standard, biphenyls, POM, toxaphene, and calculated in a manner that is consistent was set to protect the public welfare trifluralin. Emissions of two PB HAP with the indicator, averaging time, and which includes among other things were identified in the emissions form of the NAAQS for lead, and those soils, water, crops, vegetation and inventory for the Primary Lead Smelting estimates are compared to the actual wildlife. CAA section 302(h). Thus, source category: Lead compounds and level of the lead NAAQS (0.15 μg/m3). ambient lead concentrations above the cadmium compounds. The NAAQS value, a public health NAAQS for lead also indicate the Cadmium emissions were evaluated policy judgment, incorporated the potential for adverse environmental for potential non-inhalation risks and Agency’s most recent health evaluation effects. adverse environmental impacts using of air effects of lead exposure for the For additional information on the our recently developed screening purposes of setting a national standard. multi-pathway analysis approach, see scenario that was developed for use In setting this value, the Administrator the residual risk documentation as with the TRIM.FaTE model. This promulgated a standard that was referenced in section IV.A of this screening scenario uses environmental requisite to protect public health with preamble. The EPA solicits comment media outputs from the peer-reviewed an adequate margin of safety. We generally on the modeling approach TRIM.FaTE to estimate the maximum consider values below the level of the used herein to assess air-related lead potential ingestion risks for any primary NAAQS to protect against risks, and specifically on the use of the specified emission scenario by using a multi-pathway risks because, as lead NAAQS in this analytical generic farming/fishing exposure mentioned above, the primary NAAQS construct. scenario that simulates a subsistence is set as to protect public health with an 5. Assessing Risks Considering environment. The screening scenario adequate margin of safety. However, Emissions Control Options retains many of the ingestion and ambient air lead concentrations above scenario inputs developed for EPA’s the NAAQS are considered to pose the In addition to assessing baseline Human Health Risk Assessment potential for increased risk to public inhalation risks and screening for Protocols (HHRAP) for hazardous waste health. We consider this NAAQS potential multi-pathway risks, we also combustion facilities. In the assessment to be a refined analysis estimated risks considering the potential development of the screening scenario a given the numerous health studies, emission reductions that would be sensitivity analysis was conducted to detailed risk and exposure analyses, and achieved by the particular control ensure that its key design parameters level of external peer and public review options under consideration. The were established such that that went into the development of the expected emissions reductions were environmental media concentrations primary NAAQS for lead, combined applied to the specific HAP and were not underestimated, and to also with the site-specific dispersion emissions points in the source category minimize the occurrence of false modeling analysis performed to develop dataset to develop corresponding positives for human health endpoints. the ambient concentration estimates due estimates of risk reductions. See Appendix 3 of the risk assessment to emissions from the one Primary Lead 6. Conducting Other Risk-Related document for a complete discussion of Processing facility being addressed in Analyses, Including Facility-Wide the development and testing of the this RTR. It should be noted, however, Assessments and Demographic Analyses screening scenario, as well as for the that this comparison does not account values of facility-level de minimis for possible population exposures to a. Facility-Wide Risk emission rates developed for screening lead from sources other than the one To put the source category risks in potentially significant multi-pathway being modeled; for example, via context, for our residual risk review, we impacts. For the purpose of developing consumption of water from untreated also examine the risks from the entire de minimis emission rates for our local sources or ingestion of locally ‘‘facility,’’ where the facility includes all cadmium multi-pathway screening, we grown food. Nevertheless, the HAP-emitting operations within a derived emission levels for cadmium at Administrator judged that such a contiguous area and under common which the maximum human health risk standard, would protect, with an control. In other words, we examine the would be 1-in-1 million for lifetime adequate margin of safety, the health of HAP emissions not only from the source cancer risk. children and other at-risk populations category of interest, but also emissions In evaluating the potential air-related against an array of adverse health of HAP from all other emission sources multi-pathway risks from the emissions effects, most notably including at the facility. In this rulemaking, for the of lead compounds from the one facility neurological effects, particularly sole facility in the Primary Lead in this source category, rather than neurobehavioral and neurocognitive Smelting source category, there are no developing a de minimis emission rate, effects, in children. 73 FR 67007. The other significant HAP emission sources we compared its maximum modeled 3- Administrator, in setting the standard, present. With the exception of organic month average atmospheric lead also recognized that no evidence-or risk HAP sources determined to present concentration at any off-site location based bright line indicated a single insignificant risk, all HAP sources have with the current primary National appropriate level. Instead a collection of been included in the risk analysis. Ambient Air Quality Standard (NAAQS) scientific evidence and other Therefore, the facility-wide risks are the for lead (promulgated in 2008), which is information was used to select the same as the source category risk and no set to a level of 0.15 micro-grams per standard from a range of reasonable separate facility-wide analysis was cubic meter (μg/m3) based on a rolling values. 73 FR 67006. necessary. 3-month period with a not-to-be- We further note that comparing exceeded form, and which will require ambient lead concentrations to the b. Demographic Analysis attainment by 2016. 73 FR 66964. NAAQS for lead, considering the level, To examine the potential for any Notably, in making these comparisons, averaging time, form and indicator, also environmental justice issues that might we estimated maximum rolling 3-month informs whether there is the potential be associated with HAP emissions with ambient lead concentrations taking into for adverse environmental effects. This this source category, we evaluated the

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00013 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 9422 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules

distributions of HAP-related cancer and includes an average of 26 blocks or an ensures that our decisions are health- non-cancer risks across different social, average of 1,350 people). Where census protective. A brief discussion of the demographic, and economic groups data are available at the block group uncertainties in the emissions dataset, within the populations living near the level but not the block level, we dispersion modeling, inhalation one facility in this source category. The assumed that all census blocks within exposure estimates, and dose-response development of demographic analyses the block group have the same relationships follows below. A more to inform the consideration of distribution of ages and incomes as the thorough discussion of these environmental justice issues in EPA block group. uncertainties is included in the risk rulemakings is evolving. EPA offers the We focused the analysis on those assessment documentation (Draft demographic analyses in this census blocks where source category Residual Risk Assessment for Primary rulemaking to inform the consideration risk results show either estimated Lead Smelting) available in the docket of potential environmental justice lifetime inhalation cancer risks above 1- for this action. issues, and invites public comment on in-1 million or chronic non-cancer a. Uncertainties in the Emissions the approaches used and the indices above 1. In addition, in this case Dataset interpretations made from the results, we also focused on those census blocks with the hope that this will support the where estimated ambient lead Although the development of the RTR refinement and improve the utility of concentrations were above the level of dataset involved quality assurance/ such analyses for future rulemakings. the lead NAAQS. For each of these quality control processes, the accuracy For the demographic analyses, we cases, we determined the relative of emissions values will vary depending focus on the populations within 50 km percentage of different racial and ethnic on the source of the data, the degree to of any facility with emission sources groups, different age groups, adults with which data are incomplete or missing, subject to the MACT standard (identical and without a high school diploma, the degree to which assumptions made to the risk assessment). Based on the people living in households below the to complete the datasets are accurate, emissions for the source category or the national median income, and for people whether and to what extent errors were facility, we then identified the living below the poverty line within made in estimating emissions values, populations that are estimated to have those census blocks. The specific census and other factors. The emission exposures to HAP which result in: (1) population categories included: estimates considered in this analysis are Cancer risks of 1-in-1 million or greater, • Total population; annual totals provided by the facility • (2) non-cancer HI of 1 or greater, and/ White; that do not reflect short-term or (3) ambient lead concentrations above • African American (or Black); • fluctuations during the course of a year the level of the NAAQS for lead. We Native Americans; or variations from year to year. In compare the percentages of particular • Other races and multiracial; • contrast, the estimates of peak hourly demographic groups within the focused Hispanic or Latino; emission rates for the acute effects • People living below the poverty populations to the total percentages of screening assessment were based on line; those demographic groups nationwide. multiplication factors applied to the The results, including other risk • Children 18 years of age and under; • average annual hourly emission rates metrics, such as average risks for the Adults 19 to 64 years of age; • (the default factor of 10 was used for exposed populations, are documented Adults 65 years of age and over; • Primary Lead Smelting), which is in a technical report in the docket for Adults without a high school diploma. intended to account for emission the source category covered in this fluctuations due to normal facility 10 It should be noted that these proposal. operations. The basis for the risk values used in categories overlap in some instances, the demographic analyses for the one resulting in some populations being b. Uncertainties in Dispersion Modeling facility subject to the Primary Lead counted in more than one category (e.g., While the analysis employed EPA’s Smelting MACT was the modeling other races and multiracial and recommended regulatory dispersion results based on actual emissions levels Hispanic). In addition, while not a model, AERMOD, we recognize that obtained from the HEM–3 model specific census population category, we there is uncertainty in ambient described above. The risk values for also examined risks to ‘‘Minorities,’’ a concentration estimates associated with each census block were linked to a classification which is defined for these any model, including AERMOD. In database of information from the 2000 purposes as all race population circumstances where we had to choose decennial census that includes data on categories except white. between various model options, where race and ethnicity, age distributions, The methodology and the results of possible, we selected model options poverty status, household incomes, and the demographic analyses for this (e.g., rural/urban, plume depletion, education level. The Census Department source category are included in the chemistry) that provided an Landview® database was the source of technical report available in the docket overestimate of ambient concentrations the data on race and ethnicity, and the for this action. (Risk and Technology of the HAP rather than an data on age distributions, poverty status, Review—Analysis of Socio-Economic underestimate. However, because of household incomes, and education level Factors for Populations Living Near practicality and data limitation reasons, were obtained from the 2000 Census of Primary Lead Smelting Operations). some factors (e.g., building downwash) Population and Housing Summary File 7. Considering Uncertainties in Risk have the potential in some situations to 3 (SF3) Long Form. While race and Assessment overestimate or underestimate ambient ethnicity census data are available at the impacts. Despite these uncertainties, we census block level, the age and income Uncertainty and the potential for bias believe that at off-site locations and census data are only available at the are inherent in all risk assessments, census block centroids, the approach census block group level (which including that performed for the source category addressed in this proposal. considered in the dispersion modeling 10 Risk and Technology Review—Analysis of Although uncertainty exists, we believe analysis should generally yield Socio-Economic Factors for Populations Living Near the approach that we took, which used overestimates of ambient HAP Primary Lead Smelting Operations. conservative tools and assumptions, concentrations.

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00014 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules 9423

c. Uncertainties in Inhalation Exposure near future. If this happens, chronic complete detailed discussion of The effects of human mobility on risks based on the continuation of uncertainties and variabilities in dose- exposures were not included in the current emission levels will be over response relationships is given in the assessment. Specifically, short-term estimated. residual risk documentation which is mobility and long-term mobility The exposure estimates used in these available in the docket for this action. between census blocks in the modeling analyses assume chronic exposures to Cancer URE values used in our risk domain were not considered.11 As a ambient levels of pollutants. Because assessments are those that have been result, this simplification will likely most people spend the majority of their developed to generally provide an upper bias the assessment toward time indoors, actual exposures may not bound estimate of risk. That is, they overestimating the highest exposures. In be as high, depending on the represent a ‘‘plausible upper limit to the addition, the assessment predicted the characteristics of the pollutants true value of a quantity’’ (although this modeled. For many of the HAP, indoor is usually not a true statistical chronic exposures at the centroid of 13 each populated census block as levels are roughly equivalent to ambient confidence limit). In some surrogates for the exposure levels, but for very reactive pollutants or circumstances, the true risk could be as concentrations for all people living in larger particles, these levels are low as zero; however, in other typically lower. This factor has the circumstances the risk could be that block. Using the census block 14 centroid to predict chronic exposures potential to result in an overstatement of greater. When developing an upper tends to over-predict exposures for 25 to 30 percent of exposures.12 bound estimate of risk and to provide people in the census block who live In addition to the uncertainties risk values that do not underestimate farther from the facility, and under- highlighted above, there are several risk, health-protective default predict exposures for people in the factors specific to the acute exposure approaches are generally used. To err on census block who live closer to the assessment that should be highlighted. the side of ensuring adequate health- facility. Thus, using the census block The accuracy of an acute inhalation protection, EPA typically uses the upper centroid to predict chronic exposures exposure assessment depends on the bound estimates rather than lower may lead to a potential understatement simultaneous occurrence of bound or central tendency estimates in or overstatement of the true maximum independent factors that may vary our risk assessments, an approach that may have limitations for other uses (e.g., impact for any one individual, but is an greatly, such as hourly emissions rates, priority-setting or expected benefits unbiased estimate of average risk and meteorology, and human activity patterns. In this assessment, we assume analysis). incidence. Chronic non-cancer reference (RfC The assessments evaluate the that individuals remain for 1 hour at the and RfD) values represent chronic projected cancer inhalation risks point of maximum ambient exposure levels that are intended to be associated with pollutant exposures concentration as determined by the co- health-protective levels. Specifically, over a 70-year period, which is the occurrence of peak emissions and worst- these values provide an estimate (with assumed lifetime of an individual. In case meteorological conditions. These uncertainty spanning perhaps an order reality, both the length of time that assumptions would tend to overestimate of magnitude) of a continuous modeled emissions sources at facilities actual exposures since it is unlikely that inhalation exposure (RfC) or a daily oral actually operate (i.e., more or less than a person would be located at the point exposure (RfD) to the human population 70 years), and the domestic growth or of maximum exposure during the time (including sensitive subgroups) that is decline of the modeled industry (i.e., the of worst-case impact. likely to be without an appreciable risk increase or decrease in the number or d. Uncertainties in Dose-Response of deleterious effects during a lifetime. size of United States facilities), will Relationships To derive values that are intended to be influence the future risks posed by a ‘‘without appreciable risk,’’ the given source or source category. There are uncertainties inherent in methodology relies upon an uncertainty Depending on the characteristics of the the development of the dose-response factor (UF) approach (U.S. EPA, 1993, industry, these factors will, in most values used in our risk assessments for 1994) which includes consideration of cases, result in an overestimate both in cancer effects from chronic exposures both uncertainty and variability. When individual risk levels and in the total and non-cancer effects from both there are gaps in the available estimated number of cancer cases. chronic and acute exposures. Some information, UF are applied to derive However, in rare cases, where a facility uncertainties may be considered reference values that are intended to maintains or increases its emission quantitatively, and others generally are protect against appreciable risk of levels beyond 70 years, residents live expressed in qualitative terms. We note deleterious effects. The UF are beyond 70 years at the same location, as a preface to this discussion a point on commonly default values,15 e.g., factors and the residents spend most of their dose-response uncertainty that is days at that location, then the risks brought out in EPA’s 2005 Cancer Guidelines; namely, that ‘‘the primary 13 IRIS glossary (http://www.epa.gov/NCEA/iris/ could potentially be underestimated. _ goal of EPA actions is protection of help gloss.htm). Annual cancer incidence estimates from 14 An exception to this is the URE for benzene, exposures to emissions from these human health; accordingly, as an which is considered to cover a range of values, each sources would not be affected by Agency policy, risk assessment end of which is considered to be equally plausible, and which is based on maximum likelihood uncertainty in the length of time procedures, including default options that are used in the absence of scientific estimates. emissions sources operate. For the 15 According to the NRC report, Science and specific source in this source category data to the contrary, should be health Judgment in Risk Assessment (NRC, 1994) ‘‘[Default] we anticipate significant reduction in protective.’’ (EPA 2005 Cancer options are generic approaches, based on general activities and emissions in the relatively Guidelines, pages 1–7.) This is the scientific knowledge and policy judgment, that are approach followed here as summarized applied to various elements of the risk assessment process when the correct scientific model is 11 Short-term mobility is movement from one in the next several paragraphs. A unknown or uncertain.’’ The 1983 NRC report, Risk microenvironment to another over the course of Assessment in the Federal Government: Managing hours or days. Long-term mobility is movement 12 U.S. EPA. National-Scale Air Toxics the Process, defined default option as ‘‘the option from one residence to another over the course of a Assessment for 1996. (EPA 453/R–01–003; January chosen on the basis of risk assessment policy that lifetime. 2001; page 85.) Continued

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00015 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 9424 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules

of 10 or 3, used in the absence of (3) uncertainty in lowest observed We also note the uncertainties compound-specific data; where data are adverse effect (exposure) level to no associated with the health-based (i.e., available, UF may also be developed observed adverse effect (exposure) level primary) NAAQS are likely less than the using compound-specific information. adjustments; and (4) uncertainty in uncertainties associated with dose- When data are limited, more accounting for an incomplete database response values developed for many of assumptions are needed and more UF on toxic effects of potential concern. the other HAP, particularly those HAP are used. Thus, there may be a greater Additional adjustments are often for which no human health data exist. tendency to overestimate risk in the applied to account for uncertainty in In 1988, EPA’s IRIS program reviewed sense that further study might support extrapolation from observations at one the health effects data regarding lead development of reference values that are exposure duration (e.g., 4 hours) to and its inorganic compounds and higher (i.e., less potent) because fewer derive an acute reference value at determined that it would be default assumptions are needed. another exposure duration (e.g., 1 hour). inappropriate to develop an RfD for However, for some pollutants, it is As further discussed below, there is these compounds, saying, ‘‘A great deal no RfD or other comparable chronic possible that risks may be of information on the health effects of health benchmark value for lead underestimated. lead has been obtained through decades ‘‘ ’’ compounds. Thus, to address While collectively termed UF, these of medical observation and scientific factors account for a number of different multipathway human health and research. This information has been quantitative considerations when using environmental risks associated with assessed in the development of air and observed animal (usually rodent) or emissions of lead from this facility, water quality criteria by the Agency’s human toxicity data in the development ambient lead concentrations were Office of Health and Environmental of the RfC. The UF are intended to compared to the NAAQS for lead. In Assessment (OHEA) in support of account for: (1) Variation in developing the NAAQS for lead, EPA susceptibility among the members of the considered human health evidence regulatory decision-making by the human population (i.e., inter-individual reporting adverse health effects Office of Air Quality Planning and variability); (2) uncertainty in associated with lead exposure, as well Standards (OAQPS) and by the Office of extrapolating from experimental animal as an EPA conducted multipathway risk Drinking Water (ODW). By comparison data to humans (i.e., interspecies assessment that applied models to to most other environmental toxicants, differences); (3) uncertainty in estimate human exposures to air-related the degree of uncertainty about the extrapolating from data obtained in a lead and the associated risk (73 FR health effects of lead is quite low. It study with less-than-lifetime exposure 66979). EPA also explicitly considered appears that some of these effects, (i.e., extrapolating from sub-chronic to the uncertainties associated with both particularly changes in the levels of chronic exposure); (4) uncertainty in the human health evidence and the certain blood enzymes and in aspects of extrapolating the observed data to exposure and risk analyses when children’s neurobehavioral obtain an estimate of the exposure developing the NAAQS for lead. For development, may occur at blood lead associated with no adverse effects; and example, EPA considered uncertainties levels so low as to be essentially (5) uncertainty when the database is in the relationship between ambient air without a threshold. The Agency’s RfD incomplete or there are problems with lead and blood lead levels (73 FR Work Group discussed inorganic lead the applicability of available studies. 66974), as well as uncertainties between (and lead compounds) at two meetings Many of the UF used to account for blood lead levels and loss of IQ points (07/08/1985 and 07/22/1985) and variability and uncertainty in the in children (73 FR 66981). considered it inappropriate to develop development of acute reference values In considering the evidence and risk an RfD for inorganic lead.’’ EPA’s IRIS are quite similar to those developed for analyses and their associated assessment for Lead and compounds chronic durations, but they more often uncertainties, the EPA Administrator (inorganic) (CASRN 7439–92–1), http:// use individual UF values that may be noted his view that there is no evidence- www.epa.gov/iris/subst/0277.htm. or risk-based bright line that indicates a less than 10. UF are applied based on We also note that because of the chemical-specific or health effect- single appropriate level. Instead, he noted, there is a collection of scientific multi-pathway, multi-media impacts of specific information (e.g., simple evidence and judgments and other lead, the risk assessment supporting the irritation effects do not vary appreciably information, including information NAAQS considered direct inhalation between human individuals, hence a about the uncertainties inherent in exposures and indirect air-related multi- value of 3 is typically used), or based on many relevant factors, which needs to pathway exposures from industrial the purpose for the reference value (see be considered together in making this sources like primary and secondary lead the following paragraph). The UF public health policy judgment and in smelting operations. It also considered applied in acute reference value selecting a standard level from a range background lead exposures from other derivation include: (1) Heterogeneity of reasonable values (73 FR 66998). In sources (like contaminated drinking among humans; (2) uncertainty in so doing, the Administrator decided water and exposure to lead-based extrapolating from animals to humans; that, a level for the primary lead paints). In revising the NAAQS for lead, standard of 0.15 μg/m3, in combination we note that the Administrator placed appears to be the best choice in the absence of data more weight on the evidence-based to the contrary’’ (NRC, 1983a, p. 63). Therefore, with the specified choice of indicator, default options are not rules that bind the Agency; averaging time, and form, is requisite to framework and less weight on the rather, the Agency may depart from them in protect public health, including the results from the risk assessment, evaluating the risks posed by a specific substance health of sensitive groups, with an although he did find the risk estimates when it believes this to be appropriate. In keeping to be roughly consistent with and with EPA’s goal of protecting public health and the adequate margin of safety (73 FR 67006). environment, default assumptions are used to A thorough discussion of the health generally supportive of the evidence- ensure that risk to chemicals is not underestimated evidence, risk and exposure analyses, based framework applied in the NAAQS (although defaults are not intended to overtly and their associated uncertainties can be determination. 73 FR 67004. Thus, overestimate risk). See EPA, 2004, An Examination when revising the NAAQS for lead to of EPA Risk Assessment Principles and Practices, found in EPA’s final rule revising the EPA/100/B–04/001 available at: http:// lead NAAQS (73 FR 66970–66981, protect public health with an adequate www.epa.gov/osa/pdfs/ratf-final.pdf. November 12, 2008). margin of safety, EPA considered both

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00016 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules 9425

the health evidence and the risk f. Uncertainties in the Facility-Wide In addition to looking back at assessment, albeit to different extents. Risk Assessment practices, processes, or control In addition to the uncertainties We did not conduct a separate technologies reviewed at the time we discussed above with respect to chronic, facility-wide risk assessment for this developed the MACT standards, we cancer, and the lead NAAQS reference proposal because all of the HAP reviewed a variety of sources of data to values, there are also uncertainties emission sources at the one facility aid in our evaluation of whether there associated with acute reference values. subject to the MACT are covered by the were additional practices, processes, or Not all acute reference values are MACT standard under review. Thus, the controls to consider. One of these developed for the same purpose, and level of the facility-wide HAP emissions sources of data was subsequent air care must be taken when interpreting is the same as the level of emissions toxics rules. Since the promulgation of the results of an acute assessment of from the emissions sources subject to the MACT standard for the primary lead human health effects relative to the the MACT standard under review. smelting source category addressed in reference value or values being this proposal, EPA has developed air exceeded. Where relevant to the g. Uncertainties in the Demographic toxics regulations for a number of estimated exposures, the lack of short- Analysis additional source categories. We term dose-response values at different Our analysis of the distribution of reviewed the regulatory requirements levels of severity should be factored into risks across various demographic groups and/or technical analyses associated the risk characterization as potential is subject to the typical uncertainties with these subsequent regulatory uncertainties. associated with census data (e.g., errors actions to identify any practices, processes, and control technologies Although every effort is made to in filling out and transcribing census forms), as well as the additional considered in these efforts that could identify peer-reviewed reference values possibly be applied to emission sources for cancer and non-cancer effects for all uncertainties associated with the extrapolation of census-block group data in the primary lead smelting source pollutants emitted by the sources category. included in this assessment, some (e.g., income level and education level) down to the census block level. We also consulted EPA’s RACT/ hazardous air pollutants continue to BACT/LAER Clearinghouse (RBLC). The have no peer-reviewed reference values B. How did we perform the technology terms ‘‘RACT,’’ ‘‘BACT,’’ and ‘‘LAER’’ are for cancer or chronic non-cancer or review? acronyms for different program acute effects. Since exposures to these requirements under the CAA provisions pollutants cannot be included in a Our technology review is focused on the identification and evaluation of addressing the national ambient air quantitative risk estimate, an quality standards. Control technologies, understatement of risk for these developments in practices, processes, and control technologies. If a review of classified as RACT (Reasonably pollutants at environmental exposure Available Control Technology), BACT levels is possible. available information identifies such developments, then we conduct an (Best Available Control Technology), or Additionally, chronic reference values analysis of the technical feasibility of LAER (Lowest Achievable Emission for several of the compounds included these developments, along with the Rate) apply to stationary sources in this assessment are currently under impacts (costs, emission reductions, risk depending on whether the sources are EPA IRIS review (e.g., cadmium and reductions, etc.). We then make a existing or new, and on the size, age, nickel), and revised assessments may decision on whether it is necessary to and location of the facility. BACT and determine that these pollutants are more amend the regulation to require any LAER (and sometimes RACT) are or less potent than the current value. We identified developments. determined on a case-by-case basis, may re-evaluate residual risks for the Based on specific knowledge of the usually by state or local permitting final rulemaking if, as a result of these primary lead smelting source category, agencies. EPA established the RBLC to reviews, a dose-response metric changes we began by identifying known provide a central database of air enough to indicate that the risk developments in practices, processes, pollution technology information assessment supporting this notice may and control technologies. For the (including technologies required in significantly understate human health purpose of this exercise, we considered source-specific permits) to promote the risk. any of the following to be a sharing of information among e. Uncertainties in the Multipathway ‘‘development’’: permitting agencies and to aid in • and Environmental Effects Assessment Any add-on control technology or identifying future possible control other equipment that was not identified technology options that might apply We generally assume that when and considered during MACT broadly to numerous sources within a exposure levels are not anticipated to development; category or apply only on a source-by- adversely affect human health, they also • Any improvements in add-on source basis. The RBLC contains over are not anticipated to adversely affect control technology or other equipment 5,000 air pollution control permit the environment. For each source (that was identified and considered determinations that can help identify category, we generally rely on the site- during MACT development) that could appropriate technologies to mitigate specific levels of PB–HAP emissions to result in significant additional emission many air pollutant emission streams. determine whether a full assessment of reduction; We searched this database to determine the multi-pathway and environmental • Any work practice or operational whether any practices, processes, or effects is necessary. For PB–HAPS other procedure that was not identified and control technologies are included for the than lead (i.e., cadmium), site-specific considered during MACT development; types of processes covered by the PB–HAP emission levels were far below and primary lead smelting MACT. levels which would trigger a refined • Any process change or pollution We also requested information from assessment of multi-pathway impacts, prevention alternative that could be the facility regarding developments in thus we are confident that these types broadly applied that was not identified practices, processes, or control of impacts are insignificant for the one and considered during MACT technology. Finally, we reviewed other facility in this source category. development. information sources, such as state or

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00017 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 9426 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules

local permitting agency databases and furnaces. HAP expected in process 63.1543(c). The MACT standard also industry-supported databases. emissions are metals (mostly lead requires the use of bag leak detection compounds, but also some arsenic, systems for continuous monitoring of C. Overview of the Source Category and cadmium, and other metals) and also baghouses. 40 CFR 63.1547(c)(9). For MACT Standards may include small amounts of organic fugitive dust sources, as defined in 40 1. Source Category and MACT Standard compounds that result from incomplete CFR 63.1544, the MACT standard The National Emission Standard for combustion of coke, which is charged requires that the owner or operator Primary Lead Smelting (or MACT rule) along with sinter to the blast furnace. prepare and operate at all times was promulgated on June 4, 1999 (64 FR Process fugitive emissions occur at according to a standard operating 30194) and codified at 40 CFR part 63, various points during the smelting procedures (SOP) manual. The SOP subpart TTT. As promulgated in 1999, process (such as during charging and manual must describe in detail the the MACT standard applies to affected tapping of furnaces) and the only HAP measures used to control fugitive dust sources of HAP at primary lead emitted are metal HAP. Fugitive dust emissions from plant roadways, material smelters.16 The MACT defines ‘‘Primary emissions result from the entrainment of storage and handling areas, sinter lead smelters’’ as ‘‘any facility engaged dust due to material handling, vehicle machine areas, blast and dross furnace in the production of lead metal from traffic, and wind erosion from storage areas, and refining and casting lead sulfide ore concentrates through piles and the only HAP emitted are operations areas. Existing work practice the use of pyrometallurgical metal HAP. manual(s) that describe the measures in techniques.’’ 40 CFR 63.1542. The The MACT standard (40 CFR part 63, place to control fugitive dust sources MACT standard for the Primary Lead subpart TTT) applies to process required as part of a state Smelting source category does not apply emissions (stack) from sinter machines, implementation plan for lead satisfy this to secondary lead smelters, lead blast furnaces, and dross furnaces; requirement. process fugitive emissions from sinter, remelters, or lead refiners (§ 63.1541). 2. MACT as it Applies to Doe Run Today there is one facility (The Doe Run blast furnace, drossing and refining processes, concentrate handling, and Company Primary Lead Smelter, Company in Herculaneum, Missouri) Herculaneum, Missouri operating that is subject to the MACT locations around such processes; and As stated above, the Doe Run Smelter standards (See Section V.A. below). fugitive dust emission sources, such as At the time of promulgation of the roadways, storage piles and the plant in Herculaneum, Missouri, is the sole Primary Lead Smelting MACT rule, yard. Process emissions of lead remaining lead processing facility in the there were three operating lead smelters. compounds from sinter machines, blast United States subject to the MACT. The Due to economic pressures (decreased furnaces, and dross furnaces, and 1999 MACT rule established a plant- market demand for lead) and regulatory process fugitive emissions from the blast wide lead emission limit of 1 lb of lead pressures, two of the lead smelting furnace and dross furnace charging, per ton of lead produced that applies to facilities subject to the MACT standard blast furnace and dross furnace tapping, the aggregation of emissions from have since been permanently closed, and the sinter machine (charging, specific sources that discharge from air leaving one primary lead smelter discharging, crushing, and sizing) are pollution control devices. Compliance currently operating in the United States. limited to 500 grams (g) of lead with the plant-wide emission limit is No new primary lead smelters have emissions per mega gram (Mg) of lead demonstrated by annual stack testing. been built in the last 20 years, and no produced (500 g/Mg), which is equal to The rule lists nine sources as subject to new primary lead processing facilities 1.0 pound (lb) of lead emissions per ton the plant-wide limit including: (1) using pyrometallurgical techniques are of lead produced (1 lb/ton). 40 CFR Sinter machine, (2) blast furnace, (3) anticipated in the foreseeable future. 63.1542(a). A plant-wide limit format dross furnace, (4) dross furnace charging The one operating lead smelter is not was used for MACT because it was location, (5) blast furnace and dross collocated with other sources of HAP consistent with SIPs, the commingling furnace tapping location, (6) sinter emissions. of exhaust gases from processes to a machine charging location, (7) sinter Lead is used to make various single stack made it impossible to set machine discharge end, (8) sinter construction and consumer products limits for individual sources, it gave the crushing and sizing equipment, and (9) such as batteries, paint, glass, piping, facilities more flexibility in complying sinter machine area. At the Doe Run and filler. Lead sulfide (PbS) ore with the standard, and it promoted plant, lead emissions from these sources concentrates are the main feed material pollution prevention by giving each are controlled by baghouses that exhaust to primary lead smelters. The primary facility the ability to meet the emission through two stacks. The sources in the lead smelting process consists of lead limit through any combination of source sinter operation, the blast furnace, and sulfide concentrate storage and reduction and control technology the dross furnace are controlled by three handling, sintering of ore concentrates, options. (63 FR 19208). baghouses all of which discharge sinter crushing and handling, smelting In addition to being subject to the through one emission point, which is of sinter to lead metal, drossing (i.e., plant-wide emission limit of the designated as the main stack. The removing the solid oxide deposits), standard, process fugitive emissions building that houses the blast furnace refining and alloying of lead metal, and must be captured by a hood and and dross kettles is vented to a separate ventilated to a baghouse or equivalent smelting of the drosses. baghouse (#7) which discharges through HAP are emitted from primary lead control device and the hood design and a separate stack, designated as the smelting as process emissions (stack), ventilation rate must be consistent with furnace area stack. process fugitive emissions, and fugitive American Conference of Governmental Under the 1999 MACT rule, all other dust emissions. Process emissions are Industrial Hygienists recommended sources of process fugitive and fugitive associated with the exhaust gases from practices. 40 CFR 63.1543(b). In dust emissions are required to follow sinter machines and blast and dross addition, the sinter machine area work practice standards detailed in the fugitives must be enclosed in a building plant’s standard operating procedures 16 As provided above in section III(C)(3), we are that is ventilated to a baghouse at a rate (SOP) manual. proposing to change the standard to apply to that maintains a positive in-draft The HAP emitted in the largest Primary Lead Processors. through any doorway opening. 40 CFR quantities from the Doe Run facility are

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00018 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules 9427

lead compounds, which account for improvements to existing emission from paved surfaces to reduce emissions over 99 percent of the total HAP control technology, adding or upgrading from traffic, maintaining a minimum emissions by mass. The remaining HAP enclosures, process changes and water content percentage for ore emissions are arsenic, antimony, limitations, and work practices. These concentrate and for baghouse dust that cadmium, cobalt, nickel and trace requirements are summarized below. is loaded into railcars, and inspecting organic HAP. Negligible levels of Point Source Requirements—As the siding that encloses buildings organic HAP are also emitted from required under the SIP, lead emissions (followed by prompt repairs if needed). natural gas-fired space heating at the from the refining kettles and refining Missouri requires Doe Run to report facility and the incomplete combustion building emissions must be captured all metal HAP emissions annually based of coke in the blast furnace. Further and vented to baghouses. Doe Run on a speciation analysis that was discussions of the emission profile for implemented these controls and vents performed.19 The state also requires an this facility is included in the Technical the emissions to baghouses #8 and #9 annual emissions inventory based on Support Document in the docket. and the exhaust from the baghouse #9 the stack tests for the point discharges is combined with baghouse #7 exhaust and AP–42 or facility-specific emission 3. Missouri SIP and the Lead NAAQS as and vented to a common stack. factors for fugitive emissions. They Apply to Doe Run Company, Although the MACT standard does not As a result of the implementation of Herculaneum, Missouri require Doe Run to do so, it has the emission control requirements in the In addition to the MACT standard, the included emissions from refining currently approved 2002 SIP, and the Doe Run Company’s primary lead Baghouses #8 and #9 in their additional requirements adopted by the smelter in Herculaneum, Missouri is demonstrations of compliance with the state, as discussed above, the Doe Run subject to a SIP for the purpose of MACT plant-wide lead emission limit. facility has achieved a significant attaining and maintaining the lead Under the 2007 SIP submittal, Doe reduction of lead and metal HAP NAAQS.17 The current SIP, which was Run was required to make emissions since 2000 through a approved in 2002, addresses the former improvements to existing baghouse combination of reduced production lead ambient air concentration limit of controls including the installation of levels and improved emissions controls. 1.5 μg/m3 NAAQS. In addition, the 2007 pleated filters and lowering the air-to- Based on emissions inventory data SIP submittal from the State includes cloth ratio for baghouses, increased submitted to the Missouri Department of requirements addressing lead emissions ventilation and improved ventilation Natural Resources (DNR), total HAP from the Doe Run facility and can be hoods at the blast furnace, and using emissions have been reduced from an found at http://www.dnr.mo.gov/env/ reverse flow technology for baghouse estimated 140 tons in 2000 to 20 tons in apcp/docs/2009drh-leadsip.pdf. cleaning. The 2007 SIP submittal also 2008, and the majority of the 20 tons are In 2008, EPA revised the lead NAAQS required the installation of enclosures lead compound emissions. The 2008 from 1.5 μg/m3 to 0.15 μg/m3. In and/or partial enclosures for unloading reported emissions reflect November 2010, EPA identified or ore concentrate, sinter storage, and the implementation of all emission controls ‘‘designated’’ several areas as not sides of the sinter machine (which will stipulated in the 2002 SIP and the 2007 meeting the lead NAAQS. These be evacuated to a baghouse). SIP revision. ‘‘nonattainment’’ designations include Process Requirements—Process 4. Other Federal and State Actions portions of Jefferson County, Missouri changes to reduce emissions required by Affecting Doe Run Company surrounding the Doe Run facility. the SIP included a process control Missouri is required by the Act to take system for the injection of air through More recently, the 2008 revision to steps to further control pollution in this the blast furnace tuyeres located at the the lead NAAQS has resulted in Doe area, and to detail these steps in a bottom of the blast furnace, limitations Run Company deciding that it is not revision to the SIP. The revised SIP is on individual process and overall plant feasible for the facility to reduce due to EPA within eighteen months throughputs, and limiting specific emissions further to the level necessary after the effective date of the operations to only certain times of the to meet the newly revised NAAQS designation, or by June 2012, and day when the impact on ambient air without closure of the current smelting attainment of the NAAQS should be concentrations is less. The SIP also operations. As a result of past and achieved by 2016. stipulates that emissions from ongoing regulatory compliance issues at malfunctions will be reduced by alarms The SIP and the pending 2007 SIP the facility, the facility has entered into that sound when the baghouse fan submittal contain specific measures to a consent decree with U.S. EPA Region malfunctions, an interlock system to be implemented by the Doe Run plant VII and the State of Missouri. Under the restrict air flow into the blast furnace to reduce lead emissions. The State of consent decree, the facility will, among when the baghouse is not operating Missouri revised the control other things, close the existing smelter properly, and cameras for the dross and requirements for the Doe Run facility in operation and remediate the site to an refinery kettles to detect kettle failure 2001 and 2007, requiring numerous agreed-upon level. The consent decree (i.e., when a plume of smoke is detected emissions-reducing measures and requires that all support operations for from the stack, the kettle burner can be improvements to add-on control the smelter cease by December 31, 2013 immediately shut off and the problem devices, processes, and work and that the blast furnace cease corrected). practices.18 These included operations by April 1, 2014. Fugitive Dust Requirements—Under Remediation of the site is required to both the current SIP and the 2007 SIP 17 EPA most recently approved the Missouri SIP commence following approval of a plan for Herculaneum in 2002 (67 FR 18497, April 16, submittal, work practices are required to to be submitted to EPA in January 2013. 2002). Missouri Department of Natural Resources reduce fugitive dust emissions. (MDNR) substantially revised the requirements for Under the consent decree, the existing Requirements include road watering refining, casting and alloying operations the smelter in 2007. EPA has proposed approval of and automatic sprinklers, using new this revision, but has not yet taken final action. 18 EPA most recently approved the Missouri SIP regenerative sweepers to remove dust 19 Doe Run Company submits annual emissions for Herculaneum in 2002 (67 FR 18497, April 16, inventories to MDNR that report speciated metals 2002). MDNR substantially revised the proposed approval of this revision, but has not yet using speciation factors for each metal/source requirements for the smelter in 2007. EPA has taken final action. derived in the late 1990s through emissions testing.

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00019 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 9428 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules

will be allowed to continue operation. 26, 2006), and the 2005 National Lead compounds account for about 99 Notice of the consent decree was Emissions Inventory (NEI), version 2.0 percent of the HAP emissions from the published for public comment on (made publicly available in October source category, or about 20 tons in October 15, 2010, (75 FR 63506). Once 2008). The 2005 NEI was updated to 2008. The facility also reported small finalized, the consent decree is federally develop the 2005 National Air Toxics emissions of five other metal HAP, and enforceable among the parties. Assessment (NATA) Inventory. NATA trace levels of 25 organic HAP. Prior to closure of the current smelter, inventory updates for the primary lead The emissions data, calculations and the Doe Run Company may build and smelting category included SIP data risk assessment inputs for the Primary bring to full operation a new provided by the state of MO to EPA. The Lead Smelting source category are hydrometallurgical process that will 2005 NATA inventory was used with described further in the Technical produce lead from lead sulfide ore, updated 2008 data received in an Support Document for this action which potentially adjacent to the current Information Collection Request (ICR) is available in the docket for this smelter. The hydrometallurgical process response from the Doe Run facility. The proposed rulemaking. uses chemical reactions involving NEI is a database that contains fluboric acid which allows recovery of information about sources that emit We used the 2008 production lead metal through leaching, criteria air pollutants, their precursors, information as the basis for calculating electrowinning, and co-product and HAP. The NEI database includes the MACT allowable ratio (allowable to treatment processes. Some of the lead estimates of annual air pollutant actual) because the 2008 emissions are from the new process is likely to emissions from point and volume the most recent reported emissions that undergo further processing at the sources, emission release characteristic also reflect implementation of the existing refinery, primarily for data such as emission release height, requirements of the 2007 SIP revision. remelting/casting purposes. Based on temperature, velocity, and location For more information on the ratio of limited data from a demonstration latitude/longitude coordinates. We actual to MACT-allowable emissions, project, Doe Run expects that lead reviewed the NEI datasets, checked see the Technical Support Document in emissions from the hydrometallurgical geographic coordinates, and made the docket for this action describing the process will be minimal. changes based on available information. emission data information and We also reviewed the emissions and estimation of MACT-allowable emission V. Analyses Results and Proposed other data to identify data anomalies levels and associated risks and impacts. Decisions that could affect risk estimates. This section of the preamble provides The risk assessment was based on B. What are the results of the risk a description of the dataset used in the estimates of the actual emissions and assessments and analyses? RTR analysis, the results of our RTR for allowable emissions. The estimates of For the Primary Lead Smelting source the source category, and our proposed actual emissions were for the year 2008 category, we conducted an inhalation decisions concerning changes to the and were based on data from the ICR risk assessment for all HAP emitted. We Primary Lead Smelting MACT standard. along with data from our NEI dataset. also conducted a multi-pathway As noted previously, all references to These estimates included both stack and analysis for cadmium and lead. With lead emissions in this proposal means fugitive emission sources. Fugitive dust respect to lead, we used the recently- ‘‘lead compounds,’’ which is the sources include material handling promulgated lead NAAQS to evaluate regulated HAP under CAA section 112. (concentrate, sinter, fume and dross), the potential for multi-pathway and All reference to lead production means plantwide resuspension (roadways, environmental effects. Furthermore, we the production of element lead. storage piles and plant yard) and other miscellaneous sources (vents and heat conducted a demographic analysis of A. What data were used in our risk stacks). The material handling sources population risks. Details of the risk analyses? contribute approximately 84 percent of assessments and additional analyses can For the Primary Lead Smelting source the total fugitive dust emissions, while be found in the residual risk category, we compiled a preliminary plantwide resuspension and documentation referenced in section dataset using readily available miscellaneous sources contribute IV.A of this preamble, which is information, reviewed the data, and approximately 11 and 5 percent, available in the docket for this action. made changes where necessary. The respectively. The estimates of allowable 1. Inhalation Risk Assessment Results preliminary dataset was based on data emissions were calculated using in the 2002 National Emissions production data from the ICR response Table 3 provides an overall summary Inventory (NEI) Final Inventory, Version combined with the current emissions of the results of the inhalation risk 1 (made publicly available on February limits in the MACT standard. assessment.

TABLE 3—PRIMARY LEAD SMELTING INHALATION RISK ASSESSMENT RESULTS

Maximum individual cancer risk Maximum chronic non-cancer (in 1 million) 1 Estimated Estimated annual TOSHI 2 Maximum off-site population at risk cancer incidence refined acute Actual Allowable ≥ 1-in-1 million (cases per year) Actual Allowable non-cancer HQ 3 emissions level emissions level emissions level emissions level

30 30 4,900 0.0008 1 1 0.6 1 Estimated maximum individual excess lifetime cancer risk. 2 Maximum TOSHI. The target organ with the highest TOSHI for the Primary Lead Smelting source category is the kidney. 3 The maximum acute HQ value shown uses the only available acute dose-response value for arsenic, which is the REL. See section IV.A of this preamble for explanation of acute dose-response values.

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00020 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules 9429

The results of the chronic inhalation assessment and estimated a maximum emission scenarios, emissions from the cancer risk assessment indicate that, off-site HQ value of 0.6. main stack do not result in lead levels based on estimates of actual emissions The results of a multipathway above the NAAQS within the 50 km from the base year 2008, the maximum screening analysis for cadmium radius that was modeled. This is likely individual lifetime cancer risk could be emissions from this facility were well due to the height of the stack (500 feet), as high as 30-in-1 million with fugitive below the de minimis emission rate that which would result in broader and dust emissions of cadmium dominating would indicate a non-negligible risk of further dispersal of lead emissions. the risk. The total estimated cancer adverse health effects from However, results of the analysis did incidence from this source category multipathway exposures. We estimate indicate that modeled ambient air lead based on actual emission levels is the specific multipathway de minimis concentrations resulting from this emission rate for cadmium to be 0.65 0.0008 excess cancer cases per year or facility’s fugitive dust emissions could TPY and only 0.1 TPY is emitted from one case in every 1,250 years. exceed the NAAQS for lead by as much Approximately 200 people were the one facility in this source category. as 50-fold at the property boundary estimated to have cancer risks above Thus, there appears to be little, if any, based on both actual and allowable 10-in-1 million and approximately 4,900 multipathway risk associated with people were estimated to have cancer cadmium emissions from this facility. emissions. Moreover, results indicate risks above 1-in-1 million. When In evaluating the potential multi- that modeled emissions from the considering the maximum levels of pathway risks from emissions of lead furnace area stack could result in emissions allowed under the current compounds, we compared modeled NAAQS exceedances under both actual MACT standard, the MIR remains 30-in- maximum 3-month rolling average 2008 and MACT-allowable emissions 1 million. The MIR remains the same atmospheric concentrations with the scenarios. In addition, the actual since the fugitive dust emissions are NAAQS for lead. Table 4 presents the estimated emissions from the refining governed by work practices, which results of our lead impact analysis stacks, which were put into place based under § 63.1544 are defined as the broken down by emission point on requirements adopted by the State measures that will be ‘‘put into place to considering actual 2008 emissions as for purposes of the SIP, could result in control fugitive dust emissions.’’ Thus, well as the maximum emissions of lead NAAQS exceedances. We were unable the actual emissions, which reflect the that the MACT standard would have to calculate a ‘‘MACT allowable’’ measures that have been put in place, allowed based on production rates for emission level for the refinery should be equivalent to the allowable calendar year 2008. For purposes of our emissions, which under the MACT are emissions. analysis, we determined separately the included as fugitive emissions. This The maximum chronic noncancer risk from each of the types or processes/ analysis also indicates that within TOSHI value is 1, with fugitive emissions sources regulated by the 50 km of this facility, approximately emissions of cadmium dominating those current MACT, with one exception. 1,900 people could be exposed to impacts. When considering MACT Under the MACT, emissions from the ambient air lead concentrations allowable emissions, the maximum refining and casting area were exceeding the level of the NAAQS for chronic noncancer TOSHI value considered fugitive emissions subject to lead. remains 1 since, for the reasons work practice standards under provided above, MACT-allowable § 63.1544. Since then, pursuant to As mentioned above, to evaluate the fugitive emissions are equal to actual requirements that the 2002 State SIP potential for adverse environmental fugitive emissions. adopted for purposes of meeting the effects, we also compared maximum Based on the acute REL value for 1.50 μg/m3 lead NAAQS, Doe Run 3-month rolling average atmospheric arsenic, an off-site screening-level acute enclosed the refining and casting area concentrations with the current HQ value from this facility could be as and vents those emissions to the secondary NAAQS for lead, which is the high as 6. However, the emissions factor refinery stacks. We considered these same as the primary standard. Thus, the of 10 times the average hourly emissions stack emissions separate from the analyses presented in Table 4 also rate is not appropriate in this instance, fugitive dust emissions. Thus, the four indicate the potential for adverse given that fugitive emissions are emission process/sources we evaluated environmental effects from emissions of minimized during the meteorological for risk were: (1) The main stack, (2) the lead. Note that modeling performed for conditions associated with the worst- furnace area stack, (3) the refinery stack, this analysis is based on different inputs case short-term impacts (i.e., during and (4) fugitive emissions. than SIP modeling done for the one low-wind, stable atmospheric The analysis indicates that under both remaining primary lead facility, and conditions). Thus, we refined the actual 2008 or MACT allowable thus results differ.

TABLE 4—SUMMARY OF MODELED LEAD CONCENTRATIONS RELATIVE TO THE NAAQS BASED ON ESTIMATED ACTUAL 2008 AND MACT ALLOWABLE EMISSIONS

Actual Allowable 2008 Maximum impact— 1 Maximum impact— Emission point emissions actual emissions emissions allowable emissions (TPY) (TPY)

Main stack 2 ...... 13.31 0.05 times the NAAQS .... 65.8 0.25 times the NAAQS. Refining stacks ...... 2.74 3 times the NAAQS ...... NA NA. Furnace area stack: (controlled blast and drossing fugitives) .. 1.81 2 times the NAAQS ...... 8.94 10 times the NAAQS. Fugitive dust 3 ...... 2.85 50 times the NAAQS ...... 2.85 50 times the NAAQS. 1 Allowable emissions for the main stack and furnace area emission points are based on 1 lb of Pb/ton production (MACT limit); Refinery emis- sions are included as fugitive emissions under MACT but are now vented to a stack because of SIP requirements; therefore, we were unable to calculate a ‘‘MACT allowable’’ emission level. 2 Main stack is the emission point for sinter machine, blast furnace and drossing operations.

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00021 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 9430 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules

3 Fugitive dust emissions are covered by work practices under current MACT and were calculated via emission factors assuming compliance with the MACT. The site of maximum ambient air lead concentration resulting from fugitive dust emissions occurs in close proximity to the south- east boundary of the facility (see Figure 3.1–1 of the risk assessment document). Note that this maximum result and its location are based on modeling 2008 emissions using 1998 site-specific meteorology, and that these may differ from inputs used for other types of modeling (e.g., SIP modeling.)

2. Facility-wide Risk Assessment maximum 3-month rolling average lead modeling to estimate ambient pollutant Results concentrations at model receptor concentrations and considering that the locations in close proximity to these meteorological data used to develop the Our screening analysis determined monitoring sites. These monitor model estimates were from a different that the organic HAP emissions from locations were chosen because they year than the actual monitoring and facility represented negligible risk and represented the closest offsite monitors emissions data (i.e., meteorological data were determined to be insignificant to the Herculaneum primary lead with regard to this risk analysis. As a used in the AERMOD simulation was smelter. Thus, lead measurements at from 1998 while the emissions estimates result, all significant HAP emissions these monitoring sites would likely be from the one facility in this category are and the monitoring data were from dominated by emissions from this 2008). However, results do indicate that reflected in the risk analyses presented facility which is important given that above; therefore, facility-wide risks are the maximum 3-month average lead AERMOD estimates of ambient air lead concentration across the group of equivalent to those of the source concentrations only considered lead monitors nearest the facility is category. emissions from this facility (i.e., only approximately equal to the maximum 3- lead emissions from the Herculaneum 3. Model to Monitor Comparison month average lead concentration primary lead smelter were used as In addition to the results presented inputs into AERMOD). estimated by AERMOD across the group above, we also compared maximum Results of this analysis are presented of these monitoring sites (i.e., both the AERMOD estimates of ambient air lead in Table 5 and indicate that with respect Main Street monitor and the South concentrations with those measured at 4 to the Main Street and Circle Street Cross AERMOD estimate indicate the monitors in close proximity to the monitors, AERMOD underestimates 3- maximum 3-month average lead Herculaneum Primary Lead Smelting month maximum lead concentrations by concentration to be approximately 2.1 Facility for calendar year 2008. More approximately 2.8- and 4.2-fold, μg/m3). Taken together, these results specifically, we compared maximum 3- respectively. While these monitor to indicate that AERMOD estimates of month rolling average lead model comparisons are not in complete ambient air lead concentration provide concentrations (for calendar year 2008) agreement on a point-by-point basis, we a reasonable representation of the calculated from data reported at the note that this would not be expected measured 3-month maximum lead Main Street, Circle Street, South Cross, given the general uncertainties concentrations present in the ambient and Church Street monitors to the associated with using dispersion air near this facility.

TABLE 5—COMPARISON OF AERMOD MODELED TO AMBIENT AIR LEAD CONCENTRATIONS REPORTED BY FOUR MONITORS SURROUNDING THE HERCULANEUM PRIMARY LEAD SMELTING FACILITY

Maximum AEMOD Maximum monitored Location modeled 3-month 3-month lead con- Model to lead concentration monitor ratio 21 centration 20 (μg/m3) (μg/m3)

Main Street ...... 0.47 3.14 ¥4.6 Circle Street ...... 0.38 1.14 ¥3.0 South Cross ...... 2.13 0.75 2.8 Church Street ...... 1.99 0.47 4.2

4. Demographic Risk Analysis Results distribution of: (1) Cancer risks at or summarized in Table 5 and are based on above 1-in-1 million and (2) risks from modeling using estimated actual Demographic analyses were ambient air lead concentrations above emissions levels for the population performed to investigate the population the NAAQS for lead. Results are living within 50 km of this facility.

TABLE 6—PRIMARY LEAD SMELTING DEMOGRAPHIC RISK ANALYSIS RESULTS

Population with Population with ambient air lead Nationwide cancer risk concentrations greater than exceeding the 1 in a million NAAQS

Total population ...... 285,000,000 4,900 1,900

Race by percent

White ...... 75 96 96

20 Maximum 3-month monitored concentrations 21 Negative sign denotes an underestimation of estimated concentrations were based on the 2008 were calculated for the year 2008 based on data AERMOD modeled ambient lead concentrations, emissions estimates described in section V.A. submitted to EPA’s Air Quality System (AQS). relative to monitored concentrations. AERMOD

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00022 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules 9431

TABLE 6—PRIMARY LEAD SMELTING DEMOGRAPHIC RISK ANALYSIS RESULTS—Continued

Population with Population with ambient air lead Nationwide cancer risk concentrations greater than exceeding the 1 in a million NAAQS

All Other Races ...... 25 4 4

Race by percent

White ...... 75 96 96 African American ...... 12 4 3 Native American ...... 0.9 0.2 0 Other and Multiracial ...... 12 1 0.8

Ethnicity by percent

Hispanic ...... 14 1 0 .3 Non-Hispanic ...... 86 99 99.7

Income by percent

Below poverty level ...... 13 15 15 Above poverty level ...... 87 85 85

Results of the risk assessment indicate concentrations above the NAAQS is not indicate the potential for that there are approximately 4,900 only slightly higher at 28% (see Risk significantly disproportionate heath people exposed to a cancer risk greater and Technology Review—Analysis of impacts (see above, section V(3)(c)). than 1-in-1 million, and 1,900 people in Socio-Economic Factors for Populations Thus, risks associated with the non-lead areas with ambient air lead Living Near Primary Lead Smelting emissions from the Primary Lead concentrations above the NAAQS for Facilities in the docket for this proposed Smelting source category for cancer, lead. In both instances, the rulemaking), a difference which is likely acute and chronic non-cancer health demographics analysis estimates that not significant. effects and environmental effects are about 4 percent of these populations can C. What are our proposed decisions on considered acceptable. be classified as a minority (listed as ‘‘all risk acceptability and ample margin of However, since ambient air lead Other Races’’ in the table), which is well safety? concentrations resulting from emissions below the national percentage of 25. from this facility were modeled to be in Similarly, in the cancer and lead 1. Risk Acceptability excess of the NAAQS for lead, the risks demographic analyses, the percentage of As noted in section III.B of this associated with lead emissions from this ‘‘African American,’’ ‘‘Hispanic,’’ ‘‘Native preamble, we weigh all health risk facility were judged to be significant. American,’’ and ‘‘Other and Multiracial’’ factors in our risk acceptability Our analysis estimated that modeled off- population groups are well below the determination, including cancer risks to site ambient air lead concentrations corresponding national percentages. the individual most exposed, risk (based on actual 2008 emissions) With respect to the percentage of those estimation uncertainty, and other health resulting from this facility could be as ‘‘Below the Poverty Level,’’ in both information. For the Primary Lead high as 50 times the NAAQS for lead demographic analyses there is a small (2 Smelting source category, the risk based on fugitive dust emissions, and percent) increment above the analysis indicates that the cancer risks that approximately 1,900 individuals corresponding national percentage. to the individual most exposed could be could be exposed to lead concentrations However, given that the total population as high as 30-in-1 million due to actual in excess of the NAAQS. Given that the affected is small (i.e., 4,900 individuals or MACT-allowable emissions. These NAAQS for lead was set to ‘‘provide for cancer risk greater than 1-in-1 risks are considerably less than 100-in- increased protection for children and million and 1,900 individuals in areas 1 million, which is the upper bound of other at-risk populations against an with lead concentrations above the the presumptive range of acceptability. array of adverse health effects, most NAAQS), we do not think this indicates The incidence of cancer is very low— notably including neurological effects in any significant potential for disparate 0.0008 excess cancer cases per year; or children, including neurocognitive and impacts to the specific demographic one case every 1,250 years. Similarly, neurobehavioral effects (73 FR 67007)’’, groups analyzed. the risks of chronic non-cancer health we are proposing that risks associated Moreover, given the extent to which effects from HAP emissions other than with lead emissions from this source lead may impact children’s health, we lead were low, with a maximum HQ of category are unacceptable. further note that our demographic 1. Moreover, while an initial screening As noted above, our risk analysis for analysis doesn’t indicate the presence of analysis suggested that fugitive lead was based on modeled 3-month a higher percentage of children than one emissions of arsenic had the potential to rolling average lead concentrations in would normally expect around this create a risk of acute health effects, a ambient air in comparison to the facility. That is, while the national refined analysis based on our primary lead NAAQS. We believe that percentage of children 18 years and knowledge of this emission source in order to provide an acceptable level younger is 27%, the percentage of indicated that the risk was low (HQ = of risk, lead concentrations in the children living near this facility who are 0.6). In addition to these health ambient air must be reduced to the level estimated to be exposed to lead analyses, a demographics analysis did of the lead NAAQS. Thus, we

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00023 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 9432 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules

considered specific emission limits for necessary to meet the primary lead inspection of building siding or the three emission sources/points that NAAQS level of 0.15 μg/m 3. Thus, our damages and openings. The SOP would were modeled to result in lead evaluation of risks based on actual be required to include procedures, concentrations in excess of the NAAQS emissions already considered emissions including recordkeeping, to ensure that (see Table 4); refinery stack, furnace with these controls largely in place. In the work practices are being area stack, and fugitive dust emissions, order to ensure that site remediation implemented at a frequency and in a with the majority of fugitive dust efforts, or any other efforts the source manner that would ensure that fugitive impacts from material handling sources. may choose to undertake, will result in dust emissions are being minimized. To Based on our analysis, we conclude that sufficient emission reductions to determine whether the work practices in order to meet the NAAQS for lead at address the unacceptable level of risk, described in the SOP are reducing all model receptors, fugitive dust we are proposing to establish a lead emissions sufficient to comply with the emissions would have to be reduced by concentration in air limit of 0.15 μg/m 3 lead concentration in air limit, the approximately 98 percent to 0.064 TPY, to be measured at locations approved by owner or operator would be required refinery stack emissions and furnace the Administrator. This lead once a year to model the fugitive dust area stack emissions would have to be concentration in air limit would be emissions using measurement data or reduced by approximately 80 percent to established as the enforceable emission factors according to an a total of 0.91 TPY (the maximum requirement to address fugitive approved fugitive dust emissions impacts of refinery and furnace emissions under the MACT standard.22 modeling plan. At a minimum, EPA emission points occur at the same Because we are proposing a would require that this modeling plan location.) Further, because the concentration limit to address fugitive include a detailed description of each maximum ambient air impacts of the dust emissions, we no longer believe it fugitive dust emission source; a detailed refinery/furnace emissions, the fugitive is necessary for the affected facility to description of the control practices or dust emissions, and the main stack do provide a plan to the Administrator techniques used to limit fugitive dust not significantly overlap each other, we describing work practices that will be emissions from each source; the estimate that lead emissions from all used to reduce fugitive emissions. emission factors, test data or other emission points other than the main Therefore, we are proposing to remove methods used to characterize and stack would have to be limited to a total the requirement to develop and submit quantify lead emissions from each of approximately 0.97 TPY in order to a work practice standard operating source; a description of the emissions ensure 3-month rolling average ambient procedure (SOP) manual as required in modeling that will be used to estimate air lead concentrations do not exceed § 63.1544(a). the concentrations of lead in air at or the lead NAAQS level of 0.15 μg/m3. As As an alternative to proposing near the property boundary as noted above, emissions from the main compliance monitoring requirements for contributed by each source as well as stack (i.e., emission point for sinter demonstration of compliance with the cumulatively contributed by all sources; machine, blast furnace and drossing lead concentration in air limit, we a description of process or other operations) did not result in ambient air considered retaining the current fugitive conditions that would indicate the need lead concentrations in excess of the lead dust emissions requirement to develop to demonstrate compliance more often NAAQS at modeled locations within 50 and submit to the Administrator or than annually; the calculations to be km of the property boundary and thus delegated authority a work practices used to show compliance with the air we are not proposing any reductions at SOP. Using this alternative approach, lead concentration limit that consider we believe it would be necessary to the main stack in order to ensure an the highest modeled air lead modify the current general requirements acceptable level or risk. concentrations from the modeled for an SOP by specifying the minimum fugitive dust sources and any Once we determined the emissions work practice requirements that the contributions from background lead reductions necessary to achieve an plan must include. For example, under concentrations in air; and a description acceptable level of risk, we investigated this alternative approach, we would of the records that will be kept. We are available emissions control options and require that the SOP must include, at a seeking comments on the proposed their ability to reduce emissions and minimum, detailed descriptions of all requirements to monitor air lead health risks for fugitive dust and for measures that would be used to control concentrations versus the alternative stack emissions from both the refining fugitive dust emissions from plant approach described above, of requiring and furnace area stacks. Control options roadways; material storage, transfer and extensive work practices and a work considered for reducing fugitive dust handling areas; sinter machine areas; practice SOP in conjunction with emissions and associated risks include furnace areas; refining and casting areas; emissions modeling, to demonstrate improved or additional work practices, and other areas the Administrator may compliance with the air lead site remediation, application of identify. Further, EPA would require concentration limit. additional capture/control measures, that the SOP contain detailed Measures available for reducing lead and lead production limitations. With descriptions of work practices including emissions from the refining and furnace the exception of site remediation, all of road watering and automatic sprinklers, area stacks include upgrading existing these control measures have been methods to remove dust from paved baghouses by replacing the existing implemented to varying degrees at the surfaces to reduce emissions from fabric bags with high efficiency Doe Run facility in response to the traffic, maintenance of minimum water membrane bag filters. Another option Missouri SIP, as revised in 2002 and the content for ore concentrate and for would be to add extra in-line baghouses 2007 revisions submitted for approval to baghouse dust that will be handled or after existing baghouses. Such measures the SIP. As such, because the actual transferred, and procedures for the would reduce lead emissions and emissions for 2008 reflect the associated risk to within acceptable implementation of those control 22 Under the consent decree, of which we sought levels. measures, requiring those controls public comment last fall, fugitive dust sources will In summary, our analysis indicates under the MACT would be unlikely to be addressed by site remediation; however, some that in order to ensure that lead fugitive dust emissions will remain during the yield the additional 98 percent remediation of the site, which will likely extend emissions from this source do not pose reduction in fugitive emissions beyond April 2014. an unacceptable risk, emissions from

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00024 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules 9433

this facility would need to be reduced limit, at which time monitoring will be implement all technically feasible to a level that would ensure that these required to resume based on the initial controls in order to ensure that the emissions would not result in air lead plan approved by the Administrator ambient air meets the level of the lead levels greater than the 0.15 μg/m 3 for until another three years of consecutive NAAQS, which is the level that we have any 3-month period at all of the monitoring below 50 percent of the air determined will ensure an acceptable modeled locations. Further, we lead concentration limit is achieved. level of risk. Because all feasible conclude that in order to achieve the The compliance requirements discussed controls will need to be adopted in 0.15 μg/m 3 level (for any 3-month above were designed to allow for order to meet that proposed standard, rolling average) at all modeled locations, flexibility, prevention of redundant there are no additional controls to fugitive dust emissions would need to requirements, and also to provide consider for the three emission sources: be reduced by 98 percent and the consistency with current monitoring Fugitive dust emissions, the furnace emissions from the furnace area and required at the site. We are soliciting area stack, and the refinery stacks. We refining operation stacks would need to comment on this approach. For existing further note that the same controls we be reduced by 80 percent. We have facilities, compliance with the emission have proposed for the three emission identified emission reduction and limit for the furnace area and refinery points to reduce lead emissions are the control options for achieving the stacks would be demonstrated through same controls that would reduce risks required reductions, which include stack testing conducted on a quarterly from cadmium and all other metal HAP implementation of site remediation, basis. All performance testing will be known to be emitted from this source work practices, and upgrade of existing consistent with the existing MACT category. Thus, we are proposing that baghouses with membrane bags and/or testing requirements, with the exception the controls required to ensure that risk addition of an additional in-series of frequency. As provided in § 63.153(e) from lead emissions from those three baghouse. of the current rule, the facility can emission points is acceptable also We are proposing the following reduce compliance testing frequency if protect public health with an ample requirements to ensure that risk is the most recent three compliance tests margin of safety with regard to reduced to an acceptable level. demonstrated compliance. We are • emissions from all metal HAP from A stack lead emission cap of 0.91 maintaining this provision, however, these three emission points. Notably, TPY that would apply to the furnace because this proposed rule increases the after these standards are in place, we area stack and the refining operation testing frequency to quarterly, the estimate that the MIR cancer risk due to stacks. number of most recent tests necessary to the non-lead HAP will be less than 1-in- • An air lead concentration limit of comply with this provision will be 1 million. 0.15 μg/m3 based on 3-month rolling increased from three to 12. New primary Our risk analysis indicates that the average (to be measured at locations lead processing facilities would be main stack emissions do not result in approved by the Administrator) to required to demonstrate compliance ambient air lead levels exceeding the ensure that fugitive dust emission levels using a lead continuous emission NAAQS based on either actual or will not exceed the NAAQS. monitoring systems (CEMS). However, allowable emission levels. We The proposed limits apply to both since the Agency has not finalized the new and existing facilities. Any facility performance specification for the use of determined, as discussed section V.D. subject to the MACT would be required these instruments, we are deferring the below, that it is technologically feasible to meet these requirements for each effective date of the requirement to to reduce emissions from the main stack emission unit it is operating that is install, correlate, maintain and operate to a level well below the allowable level subject to the limit. In order to address lead CEMS until these actions can be of the MACT, since those levels are any fugitive dust emissions, the facility, completed. The lead CEMS installation currently being achieved, and thus we regardless of whether it is operating all deadline will be established through are proposing to require such controls or just some of the emission sources future rulemaking, along with other under CAA section 112(d)(6). We covered by this action, would be pertinent requirements. In the event evaluated whether there were additional required to meet the air lead operations commence at a new facility controls to further reduce emissions concentration emission limit. prior to promulgation of the from the main stack and determined For both new and existing facilities, performance specification, compliance that lead emissions from the main stack compliance with the air lead would be demonstrated through could be further reduced by replacing concentration limit would be quarterly stack testing until the standard cloth bags with membrane demonstrated using lead compliance promulgation of the lead CEMS bags at a capital cost of approximately monitoring devices and would be based performance specification. $2 million and an annual cost of $0.3 on a rolling 3-month average million. Assuming a 50 percent concentration. The proposed rule 2. Ample Margin of Safety reduction from 2008 main stack requires development of a monitoring Reducing lead emissions to meet the emissions, the cost of reducing lead plan for approval by the Administrator NAAQS would ensure that emissions of emissions would be about $40,000 to that includes the minimum sampling all HAP do not pose an unacceptable $229,000 per ton of lead. (See the and analysis methods and compliance risk. Once we ensure that the risk is Technical Support Document included demonstration criteria provided in the acceptable, we then look to determine in the docket for a complete discussion rule. A provision is included in this whether further reductions are of this analysis.) Because the highest proposed rule that allows for reduced appropriate to ensure an ample margin ambient air lead concentration resulting monitoring if the facility demonstrates of safety. In this part of our analysis, we from the emissions from the main stack an air lead concentration for three again consider the health factors we already is more than 20 times below the consecutive years at less than 50 percent considered to determine whether the level that is considered acceptable, it of the air lead concentration limit. The risks are acceptable but we also consider was determined that although monitoring can be reduced to once the cost of controls. additional controls such as membrane every six months unless one of the 6- With regard to lead emissions, we are bags could result in additional emission month monitoring events exceeds 50 proposing to require most of the reductions, the additional controls are percent of the air lead concentration emission sources at the facility to not warranted since they would not

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00025 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 9434 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules

appreciably reduce risk. We are developments in practices, processes stringent requirements that have been proposing that the MACT standard, with and control technologies on lead implemented in accordance with the the changes we are proposing under the emissions, emissions data from 2008 approved SIP and the 2007 SIP section 112(d)(6) technology review as were compared with emissions data revisions; we are not proposing described in section V.D. below will from 2000. Data from 2008 were additional requirements for these stacks provide an ample margin of safety with selected because they reflect the many as part of our technology review because regard to emissions of lead and other improvements that have been we have already proposed that these HAP from the main stack. implemented at the facility since stacks implement all feasible controls, promulgation of the MACT rule. regardless of cost, in order to ensure that D. What are the results and proposed Emissions data from earlier years would the risks due to these emission points decisions from the technology review? not reflect all of the emission-reducing are acceptable. Thus, there are no We evaluated developments in changes that have been implemented at additional developments in practices, practices, processes, and control the Doe Run facility given that some of processes and control technologies technologies applicable to emission the improvements were not beyond those which are reflected in the sources subject to the Primary Lead implemented until 2007 and 2008. As emission limits we have proposed to Smelting MACT. This included a search described above, technological meet CAA section 112(f)(2), above. of the RBLC Clearinghouse, the improvements to baghouses and To be consistent with the existing California BACT Clearinghouse, the processes that have been implemented MACT standard, EPA is proposing to internet, and correspondence with state at the facility since the MACT rule was retain the plant-wide pound per ton of agencies and industry. We have promulgated have resulted in production format that currently applies determined that there have been substantially lower emissions from to the aggregate emissions from the advances in emission control measures these sources at this facility. These main stack and the furnace area stack. since the Primary Lead Smelting MACT improvements include upgrade of cloth Because there are also stacks for the standard was originally promulgated in bags and ventilation improvements. In refining and casting operations, we are 1999. 2008, lead emissions from the main proposing to include those emissions as The 1999 MACT limit was set using stack, which vents emissions from the part of the plant-wide emission limit. the lead emission limits from the lead sintering operation and the blast and Thus we are proposing a plant-wide SIPs for the three states in which dross furnace, were 13.31 TPY. In lead emission limit of 0.22 pounds of primary lead smelting sources were addition, emissions from the furnace lead per ton of lead produced based on operational at the time of the area stack (i.e., the blast furnace and the proposed reductions due to the rulemaking. EPA took each of the three dross plant building which vent to section 112 (f)(2) risk review for the lead SIP limits, in lb/day, divided them baghouse 7) were 1.81 TPY, for a total furnace area and refining operations by the corresponding lead production of approximately 15.1 TPY. At the 2008 stacks (discussed above in Section C) capacity, in tons/day, and calculated a lead production rate of 149,500 tons, the and the reduction in emissions from the lead emission rate in lb/ton. The results lead emission rate for these sources at main stack (sinter/blast furnace/ were as follows: Doe Run was about 0.2 lb/ton, or 80 drossing operations) based on this Section 112(d)(6) technology review ASARCO—Missouri 1.0 lb/ton percent less than the current MACT This proposed plant-wide lead emission ASARCO—Montana 1.0 lb/ton limit of 1 lb/ton. Based on this limit was determined by summing the Doe Run—Missouri 0.84 lb/ton demonstrated performance, EPA believes that under Section 112(d)(6), 15 TPY for the main stack and the 0.91 The values were ranked and the the MACT standard should be revised to TPY for the furnace area and the median value (1.0 lb/ton) was selected reflect the reduction achieved in refining operation, and dividing by the as representative of the MACT floor. practice. annual production from 2008 of 149,564 Since the MACT standard was Because we believe that the 2008 tons. We note that variability was only promulgated, the industry has emissions of 13.31 TPY from the main applied in establishing technology- undergone significant changes. Two of stack (or combined sintering/blast based emissions from the main stack in the three facilities have shut down. The furnace/drossing operations) reflect the order to establish a plant-wide emission only remaining primary lead smelting annual rate of emissions achievable as a limit. Because the emission levels facility is the Doe Run smelter at result of the technological required from the refining operation and Herculaneum, Missouri, which is improvements that have been made furnace area stacks are based on subject to control requirements under since 1999, we are proposing an acceptable risk, we conclude it is not the Missouri SIP for lead. The existing emission limit based on the actual 2008 appropriate to consider variability in SIP, as well as a 2007 SIP revision annual emissions that vent to the main establishing limits for these emission submitted by the State and proposed for stack (i.e., sintering, blast furnace and points. approval by EPA require numerous drossing operations). In order to account We are proposing that the plant-wide emissions-reducing measures and for variability in the operation and lead emission limit apply to new and improvements to add-on control emissions, recent stack tests were used existing facilities that are subject to the devices, processes, and work practices. to calculate the 95 percent upper MACT. By default this would include We considered these developments in predictive limit (UPL). The 95 percent any new, controlled lead processing practices, processes, and control UPL for the main stack is 15 TPY. source not currently covered, including technologies in our technology review. Variability in the operations and lead processing by other than the Recent emissions tests (2000 through emission for this source are discussed in current techniques. We are requesting 2008) at the Doe Run facility support more detail in Section E below. comment on the appropriateness of that these improvements have resulted Although we believe that there have applying the plant-wide lead emission in significantly lower emissions and been developments in processes, limit to any future new lead processing demonstrate that actual lead emissions practices and control technologies with technique. from the facility are much lower than regard to the furnace area stack and with For the existing facility, compliance are allowed under the current MACT regard to refining and casting with the plant-wide stack emission limit rule. To assess the impacts of operations, as reflected by the more would be demonstrated in the same

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00026 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules 9435

manner as discussed above in section performance of the control device for represents the value which we can V.C.1 for the furnace area and refining which both run-to-run and test-to-test expect the mean of future observations stack limit (i.e., stack testing on a variability must be accounted.23 (3-run average for lead) to fall below quarterly basis). We are proposing stack In determining the contribution to a within a specified level of confidence, testing on a quarterly basis as opposed plant-wide emission limit of the main based upon the results of an to testing on an annual basis since this stack, we considered annual emissions independent sample from the same allows the facility the opportunity to discharged from the air pollution population. In other words, if we were adjust their emissions throughout the control devices that control lead to randomly select a future test year to be in compliance, rather than to emissions. For this rule, we used the condition from any of these sources (i.e., find they are out of compliance at the 2008 emissions reported by Doe Run to average of 3 runs or 30-day average) we end of the year, thereby risking the State of Missouri. can be 95 percent confident that the violations. This schedule also coincides We assessed variability using a reported level will fall at or below the with other quarterly monitoring and statistical formula designed to estimate UPL value. Use of the UPL is reporting required of the facility. Also as an emissions level that is equivalent to appropriate in this rulemaking because discussed in section V.C.1, new primary the source’s performance based on it sets a limit any single or future source lead processing facilities would be future compliance tests. Specifically, can meet based on the sources past required to demonstrate compliance the calculated limit is an upper performance. using lead continuous emission prediction limit (UPL) calculated with the Student’s t-test using the TINV This formula uses a pooled variance monitoring systems (CEMS). 2 function in Microsoft Excel®. The (in the s term) that encompasses all the E. Variability Student’s t-test has also been used in data-point to data-point variability. In assessing sources’ performance, other EPA rulemakings (e.g., NESHAP Where variability was calculated using EPA may consider variability both in for Portland Cement Manufacturing [75 the UPL statistical approach, we used identifying which performers are ‘‘best’’ FR 54970, September 9, 2010]; NSPS for the sample standard deviation and in assessing their level of Hospital/Medical/Infectious Waste calculated from the emissions data performance. Brick MACT, 479 F. 3d at Incinerators [74 FR 51368, October 6, distributions for lead. The standard 881–82; see also Mossville Envt’l Action 2009]; NESHAP for Industrial, deviation is the common measure of the Now v. EPA, 370 F.3d 1232, 1241–42 Commercial, and Institutional Boilers dispersion of the data set around an (D.C. Cir 2004) (EPA must exercise its and Process Heaters-Proposed [75 FR average. We note here that the judgment, based on an evaluation of the 32006, June 4, 2010]) in accounting for methodology accounts for both short- relevant factors and available data, to variability. A prediction interval for a term and long-term variability and determine the level of emissions control future observation is an interval that encompasses run-to-run and test-to-test that has been achieved by the best will, with a specified degree of variability. performing sources considering these confidence, contain the next (or some We adopted a form of the UPL sources’ operating variability). other pre-specified) randomly selected equation that has been used in more Variability in lead producers’ observation from a population. In other recent rulemakings. See 75 FR 54970 performance has a number of causes. words, the prediction interval estimates (September 9, 2010), 75 FR 32020 (June For emissions of lead compounds that what the upper bound of future values 4, 2010) and 75 FR 31905 (June 4, 2010). are controlled by baghouses, the will be, based upon present or past The UPL used in this proposed rule is variability is chiefly due to variations in samples taken. The UPL consequently calculated by:

Where: F. What other actions are we proposing? promulgate an affirmative defense to x¯ = 2008 annual emissions civil penalties for exceedances of As discussed in Section III.C. above, n = the number of test runs emission limits caused by malfunctions, EPA is proposing to remove provisions m = the number of test runs in the as well as criteria for establishing the in the existing standard that would have compliance average affirmative defense. s2 = observed variance exempted sources from complying with EPA has attempted to ensure that we t = student t distribution statistic the standard during periods of startup, shutdown and malfunction. Specifically have not included in the proposed This calculation was performed using we are proposing revisions to subpart regulatory language any provisions that the following Excel functions: 95 TTT Table 1 and rule provisions to are inappropriate, unnecessary, or percent UPL = 2008 annual emissions + remove applicability of the General redundant in the absence of the SSM [STDEV (Test Runs) × TINV (2 × Provisions with regard to SSM and exemption. We are specifically seeking probability, n-1 degrees of freedom) × remove the exemption for bag leak comment on whether there are any such SQRT ((1/n) + (1/m))], for a one-tailed detection alarm time attributable to SSM provisions that we have inadvertently t-value, probability of 0.05, and sample events from determining compliance incorporated or overlooked. size of n. with the total alarm time limit. In addition, we are proposing to

23 Run-to-run variability is essentially within-test imprecision of stack test methods and laboratory in the maintenance cycle in which a fabric filter is variability, and encompasses variability in analysis. 72 FR 54877 (Sept. 27, 2007). Test-to-test tested). Test-to-test variability can be termed long- individual runs comprising the compliance test, variability results from variability in pollution term variability. 72 FR 54878. and includes uncertainties in correlation of device control efficiencies over time (depending on monitoring parameters and emissions, and many factors, including for fabric filters the point

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00027 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 EP17FE11.000 9436 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules

VI. Proposed Action that will reduce emissions of these other would be demonstrated using lead metal HAP. The cancer risk from CEMS. A. What actions are we proposing as a cadmium emissions will be reduced result of the residual risk reviews? C. What other actions are we proposing? from 30-in-1 million to less than 1-in-1 Consistent with CAA section 112 million. Therefore, we are proposing As described above, we are proposing (f)(2), we are proposing to amend the that the existing MACT, as it would be to amend the applicability section for MACT standard for primary lead modified based on our proposed the MACT rule to tailor it to the processing to include a lead requirements for lead emissions, would definition of the source category we μ concentration in air limit of 0.15 g/m3 provide an ample margin of safety with established under CAA section (based on 3-month rolling averages)to be respect to emissions from all metal 112(c)(1). See ‘‘Documentation for measured at locations approved by the HAP. Development of Initial Source Category Administrator to address the risks from With regard to lead emissions from List—Final Report’’, USEPA/OAQPS, all fugitive dust emissions addressed in the main stack, we have identified EPA–450/3–91–030, July, 1992. In 40 CFR 63.1544. We are also proposing developments in practices, processes support of this applicability provision to remove refining and casting and control technologies since clarification, we are also proposing to ‘‘ operations from § 63.1544 and to require promulgation of the MACT standard in replace the definition of primary lead ’’ ‘‘ that emissions from these operations be 1999, and are proposing a reduced smelter with a definition of primary ’’ ‘‘ vented to one or more stacks. Finally, emission limit for the main stack based lead processor . The primary lead processor’’ definition would include any we are proposing to establish an on these improvements. Since the main emission cap of 0.91 TPY for the furnace facility that produces lead from stack does not pose an unacceptable risk area stack and the refining operation processing of lead sulfide ore by at its current emissions level, we are not stacks. These limits were established pyrometallurgical (smelting) or any proposing reductions for this emission based on the level of reductions in lead other technique. We are also proposing point under 112(f)(2). However, we are emissions from the three sources that to add definitions of ‘‘secondary lead proposing a reduced emission limit are necessary to show that the lead smelters’’, ‘‘lead refiners’’, and ‘‘lead under 112(d)(6) due to the NAAQS will not be exceeded within the remelters’’ to clarify the meaning of improvements we identified. 50 km modeled domain. We believe the those terms in the second sentence of NAAQS level represents an acceptable B. What actions are we proposing as a the applicability provision. level of risk and that the proposed limits result of the technology reviews? We propose to amend the Primary are necessary to ensure that risks from Lead Smelting MACT standards to these sources are acceptable. We are For the Primary Lead Smelting source remove the language that exempts bag proposing that the risk posed by lead category, we have determined that there leak detection system alarm time emissions from the main stack and by have been developments in practices, incurred during periods of SSM from emissions of all other HAP is processes, or control technologies since inclusion in the allowable alarm time. acceptable. the promulgation of the MACT This change is being made to ensure the We are proposing that compliance standards that are feasible for the one rule is consistent with the court’s ruling with the emission limits applicable to facility in this source category to in Sierra Club v. EPA, 551 F.3d 1019 the furnace area and refinery stacks implement at the main stack. The (D.C. Cir. 2008). We are also proposing would be demonstrated based on stack proposed limit is consistent with the minor modifications throughout the rule testing for existing facilities and, for current demonstrated performance of to incorporate plain language and to new facilities, using CEMS after the facility based on obligations adopted make editorial and clarifying revisions. promulgation of performance by the State and reflected in the 2002 In addition, we are proposing changes to specifications for a CEMS capable of SIP and 2007 SIP revision for Doe Run. Table 1 of the rule to reflect revisions measuring lead emissions. We are proposing that a performance to SSM requirements. We are proposing that compliance of 15.11 TPY has been demonstrated for with the lead concentration in air limit emissions from the main stack, taking D. Compliance Dates would be demonstrated using a into consideration variability of We are proposing that the compliance monitoring system emissions from that stack. The existing requirements under CAA section approved by the Administrator. MACT lead emissions standard that is 112(f)(2) for the one existing source, if We are also proposing that the applicable to emissions from the main finalized, must be implemented no later Primary Lead Smelting standard, as we stack is a plant-wide emission limit that than two years after the effective date of have proposed to revise it to ensure an also applies to emissions from the this rule. Consistent with CAA section acceptable level of risk, will also protect furnace-area stack. We are proposing to 112(f)(4)(B), we are proposing that a public health with an ample margin of revise the plant-wide limit to reflect the two-year compliance period is necessary safety. With regard to lead emissions 15.11 TPY limit for the main stack as so the facility has adequate time to from fugitive dust sources and from the well as the emissions limits we are install additional controls and furnace and refining area stacks, we proposing for the furnace-area and demonstrate compliance, including the have not identified any feasible controls refinery stacks under CAA section time necessary to purchase, install and beyond those needed to meet the 112(f)(2). Thus, we are proposing to test replacement bags, or if the facility proposed emission limits that will revise the plant-wide emissions limit decides to add a new baghouse in series provide an acceptable level of risk . The from 1 pound of lead per ton of lead with an existing baghouse, seek bids, standards we are proposing to ensure an produced, to 0.22 pound of lead per ton select a vendor, install and test the new acceptable level of risk for lead of lead produced and the new limit equipment; prepare and submit the emissions will also reduce the risk from would include emissions from the required monitoring plan to monitor cadmium and will also reduce refinery stack as well as emissions from lead concentrations in air; purchase, emissions of all other metal HAP known the main stack and the furnace area install and conduct quality assurance to be emitted from this source category stack. Compliance with this limit would and quality control measures on because the controls that will reduce be demonstrated quarterly with stack compliance monitoring equipment and; lead emissions are the same controls testing. For new facilities, compliance conduct site remediation necessary to

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00028 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules 9437

reduce fugitive emissions. A two-year comments received during the comment http://www.epa.gov/ttn/atw/rrisk/ compliance period is also consistent period will be considered. In addition to rtrpg.html. The data files include with the schedule of required actions general comments on this proposed detailed information for each HAP contained in the Consent Decree. actions, we are also interested in any emissions release point for the facility In addition, we are proposing that the additional data that may help to reduce included in the source category. plant-wide limit that would reflect the uncertainties inherent in the risk If you believe the data are not reductions required for the main stack assessments. We are specifically representative or are inaccurate, please pursuant to CAA section 112(d)(6) and interested in receiving corrections to the identify the data in question, provide for the furnace area and refinery stacks dataset used for risk modeling. Such your reason for concern, and provide pursuant to CAA section 112(f)(2) must data should include supporting any ‘‘improved’’ data that you have, if be met no later than two years after the documentation in sufficient detail to available. When you submit data, we effective date of this rule. Because these allow characterization of the quality and request that you provide documentation limits reflect the reductions from the representativeness of the data or of the basis for the revised values to furnace area and refinery stacks information. Please see the following support your suggested changes. To required under section 112(f)(2), we section for more information on submit comments on the data believe a two-year compliance submitting data. downloaded from the RTR Web page, complete the following steps: timeframe is needed for the same VIII. Submitting Data Corrections reasons provided above. (1) Within this downloaded file, enter The facility-specific data used in the suggested revisions to the data fields VII. Request for Comments source category risk analyses and appropriate for that information. The We are soliciting comments on all demographic analyses are available for data fields that may be revised include aspects of this proposed action. All download on the RTR Web Page at the following:

Data element Definition

Control Measure ...... Are control measures in place? (yes or no). Control Measure Comment ...... Select control measure from list provided, and briefly describe the control meas- ure. Delete ...... Indicate here if the facility or record should be deleted. Delete Comment ...... Describes the reason for deletion. Emission Calculation Method Code For Revised Emissions ... Code description of the method used to derive emissions. For example, CEM, material balance, stack test, etc. Emission Process Group ...... Enter the general type of emission process associated with the specified emis- sion point. Fugitive Angle ...... Enter release angle (clockwise from true North); orientation of the y-dimension relative to true North, measured positive for clockwise starting at 0 degrees (maximum 89 degrees). Fugitive Length ...... Enter dimension of the source in the east-west (x-) direction, commonly referred to as length (ft). Fugitive Width ...... Enter dimension of the source in the north-south (y-) direction, commonly re- ferred to as width (ft). Malfunction Emissions ...... Enter total annual emissions due to malfunctions (TPY). Malfunction Emissions Max Hourly ...... Enter maximum hourly malfunction emissions here (lb/hr). North American Datum ...... Enter datum for latitude/longitude coordinates (NAD27 or NAD83); if left blank, NAD83 is assumed. Process Comment ...... Enter general comments about process sources of emissions. REVISED Address ...... Enter revised physical street address for MACT facility here. REVISED City ...... Enter revised city name here. REVISED County Name ...... Enter revised county name here. REVISED Emission Release Point Type ...... Enter revised Emission Release Point Type here. REVISED End Date ...... Enter revised End Date here. REVISED Exit Gas Flow Rate ...... Enter revised Exit Gas Flowrate here (ft3/sec). REVISED Exit Gas Temperature ...... Enter revised Exit Gas Temperature here (F). REVISED Exit Gas Velocity ...... Enter revised Exit Gas Velocity here (ft/sec). REVISED Facility Category Code ...... Enter revised Facility Category Code here, which indicates whether facility is a major or area source. REVISED Facility Name ...... Enter revised Facility Name here. REVISED Facility Registry Identifier ...... Enter revised Facility Registry Identifier here, which is an ID assigned by the EPA Facility Registry System. REVISED HAP Emissions Performance Level Code ...... Enter revised HAP Emissions Performance Level here. REVISED Latitude ...... Enter revised Latitude here (decimal degrees). REVISED Longitude ...... Enter revised Longitude here (decimal degrees). REVISED MACT Code ...... Enter revised MACT Code here. REVISED Pollutant Code ...... Enter revised Pollutant Code here. REVISED Routine Emissions ...... Enter revised routine emissions value here (TPY). REVISED SCC Code ...... Enter revised SCC Code here. REVISED Stack Diameter ...... Enter revised Stack Diameter here (ft). REVISED Stack Height ...... Enter revised Stack Height here (Ft). REVISED Start Date ...... Enter revised Start Date here. REVISED State ...... Enter revised State here. REVISED Tribal Code ...... Enter revised Tribal Code here. REVISED Zip Code ...... Enter revised Zip Code here. Shutdown Emissions ...... Enter total annual emissions due to shutdown events (TPY). Shutdown Emissions Max Hourly ...... Enter maximum hourly shutdown emissions here (lb/hr).

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00029 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 9438 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules

Data element Definition

Stack Comment ...... Enter general comments about emission release points. Startup Emissions ...... Enter total annual emissions due to startup events (TPY). Startup Emissions Max Hourly ...... Enter maximum hourly startup emissions here (lb/hr). Year Closed ...... Enter date facility stopped operations.

2. Fill in the commenter information Smelting source category in the form of Send comments to OMB at the Office of fields for each suggested revision (i.e., monitoring for lead concentrations in air Information and Regulatory Affairs, commenter name, commenter and increased frequency for stack Office of Management and Budget, 725 organization, commenter e-mail address, testing as described in 40 CFR 17th Street, NW., Washington, DC commenter phone number, and revision 63.1547(k) (compliance monitoring) and 20503, Attention: Desk Office for EPA. comments). 40 CFR 63.1546 (stack testing). These Since OMB is required to make a 3. Gather documentation for any requirements are described in section decision concerning the ICR between 30 suggested emissions revisions (e.g., VI.A and B. Although these are and 60 days after February 17, 2011, a performance test reports, material additional requirements under today’s comment to OMB is best assured of balance calculations, etc.). proposed rule, they are consistent with having its full effect if OMB receives it 4. Send the entire downloaded file existing monitoring and testing by March 21, 2011. The final rule will ® with suggested revisions in Microsoft currently conducted by the facility to respond to any OMB or public Access format and all accompanying meet MACT and SIP requirements. comments on the information collection documentation to Docket ID Number Therefore, we do not believe that the requirements contained in this proposal. EPA–HQ–OAR–2004–0305 (through one additional paperwork required by these of the methods described in the proposed changes would constitute an C. Regulatory Flexibility Act ADDRESSES section of this preamble). To undue burden to the facility. expedite review of the revisions, it We estimate one regulated entity is The Regulatory Flexibility Act (RFA) would also be helpful if you submitted currently subject to subpart TTT and generally requires an agency to prepare a copy of your revisions to the EPA will be subject to all proposed a regulatory flexibility analysis of any directly at [email protected] in addition to standards. This facility will have no rule subject to notice and comment submitting them to the docket. capital costs associated with the rulemaking requirements under the 5. If you are providing comments on information collection requirements in Administrative Procedure Act or any a facility, you need only submit one file the proposed rule. other statute unless the agency certifies for that facility, which should contain The estimated recordkeeping and that the rule will not have a significant all suggested changes for all sources at reporting burden after the effective date economic impact on a substantial that facility. We request that all data of the proposed rule is estimated to be number of small entities. Small entities revision comments be submitted in the 1,323 labor hours at a cost of $465,503. include small businesses, small form of updated Microsoft® Access files, This estimate includes the cost of organizations, and small governmental which are provided on the http:// reporting, including reading jurisdictions. For purposes of assessing www.epa.gov/ttn/atw/rrisk/rtrpg.html instructions, and information gathering. the impacts of today’s proposed rule on Web page. Recordkeeping cost estimates include small entities, small entity is defined as: IX. Statutory and Executive Order reading instructions, planning activities, (1) A small business that is a small Reviews monitoring plan development, industrial entity as defined by the Small conducting compliance monitoring, Business Administration’s (SBA) A. Executive Order 12866: Regulatory sampling and analysis and maintenance regulations at 13 CFR 121.201; (2) a Planning and Review of rolling 3-month average data. The small governmental jurisdiction that is a Under Executive Order 12866 (58 FR average hours and cost per regulated government of a city, county, town, 51735, October 4, 1993), this action is a entity would be 1,323 hours and school district or special district with a significant regulatory action because it $465,503 based on one facility response population of less than 50,000; and (3) raises novel legal and policy issues. per year. Burden is defined at 5 CFR a small organization that is any not-for- Accordingly, EPA submitted this action 1320.3(b). profit enterprise which is independently to the Office of Management and Budget An agency may not conduct or owned and operated and is not (OMB) for review under Executive sponsor, and a person is not required to dominant in its field. respond to, a collection of information Order 12866 and any changes made in After considering the economic response to OMB recommendations unless it displays a currently valid OMB control number. The OMB control impacts of today’s proposed rule on have been documented in the docket for small entities, I certify that this action this action. numbers for EPA’s regulations in 40 CFR are listed in 40 CFR part 9. will not have a significant economic B. Paperwork Reduction Act To comment on the Agency’s need for impact on a substantial number of small entities. This proposed rule will not The information collection this information, the accuracy of the impose any requirements on small requirements in this rule have been provided burden estimates, and any entities. This proposed rule is currently submitted for approval to the Office of suggested methods for minimizing applicable to one operating facility that Management and Budget (OMB) under respondent burden, EPA has established the Paperwork Reduction Act, 44 U.S.C. a public docket for this rule, which does not meet the definition of a small 3501 et seq. The Information Collection includes this ICR, under Docket ID entity. Request (ICR) document prepared by number EPA–HQ–OAR–2004–0305. We continue to be interested in the EPA has been assigned EPA ICR number Submit any comments related to the ICR potential impacts of the proposed rule 1856.07. to EPA and OMB. See ADDRESSES on small entities and welcome We are proposing new paperwork section at the beginning of this notice comments on issues related to such requirements to the Primary Lead for where to submit comments to EPA. impacts.

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00030 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules 9439

D. Unfunded Mandates Reform Act has concluded that this proposed rule unless to do so would be inconsistent This proposed rule does not contain will not have tribal implications, as with applicable law or otherwise a federal mandate under the provisions specified in Executive Order 13175. It impractical. Voluntary consensus of Title II of the Unfunded Mandates will not have substantial direct effect on standards are technical standards (e.g., materials specifications, test methods, Reform Act of 1995 (UMRA), 2 U.S.C. tribal governments, on the relationship between the federal government and sampling procedures, and business 1531–1538 for State, local, or tribal Indian tribes, or on the distribution of practices) that are developed or adopted governments or the private sector. The power and responsibilities between the by voluntary consensus standards proposed rule would not result in Federal government and Indian tribes, bodies. The NTTAA directs EPA to expenditures of $100 million or more as specified in Executive Order 13175. provide Congress, through OMB, for State, local, and tribal governments, Thus, Executive Order 13175 does not explanations when the Agency decides in aggregate, or the private sector in any apply to this action. not to use available and applicable 1 year. The proposed rule imposes no EPA specifically solicits additional voluntary consensus standards. enforceable duties on any State, local or comment on this proposed action from This proposed rulemaking does not tribal governments or the private sector. tribal officials. involve technical standards. Therefore, Thus, this proposed rule is not subject EPA is not considering the use of any G. Executive Order 13045: Protection of to the requirements of sections 202 or voluntary consensus standards. 205 of the UMRA. Children From Environmental Health This proposed rule is also not subject Risks and Safety Risks J. Executive Order 12898: Federal to the requirements of section 203 of This proposed rule is not subject to Actions To Address Environmental UMRA because it contains no regulatory Executive Order 13045 (62 FR 19885, Justice in Minority Populations and requirements that might significantly or April 23, 1997) because it is not Low-Income Populations uniquely affect small governments economically significant as defined in Executive Order 12898 (59 FR 7629, because it contains no requirements that Executive Order 12866. However, the February 16, 1994) establishes federal apply to such governments nor does it Agency does believe there is a executive policy on environmental impose obligations upon them. disproportionate risk to children. justice. Its main provision directs E. Executive Order 13132: Federalism Modeled ambient air lead federal agencies, to the greatest extent concentrations from the one facility in practicable and permitted by law, to This proposed rule does not have this source category are in excess of the make environmental justice part of their federalism implications. It will not have NAAQS for lead, which was set to mission by identifying and addressing, substantial direct effects on the States, ‘‘provide increased protection for as appropriate, disproportionately high on the relationship between the national children and other at-risk populations and adverse human health or government and the States, or on the against an array of adverse health environmental effects of their programs, distribution of power and effects, most notably including policies, and activities on minority responsibilities among the various neurological effects in children, populations and low-income levels of government, as specified in including neurocognitive and populations in the United States. Executive Order 13132. None of the neurobehavioral effects.’’ 73 FR 67007. To examine the potential for any facilities subject to this action are However, the control measures environmental justice issues that might owned or operated by State proposed in this notice will result in be associated with each source category, governments, and, because no new lead concentration levels that are in we evaluated the distributions of HAP- requirements are being promulgated, compliance with the lead NAAQS, related cancer and non-cancer risks nothing in this proposal will supersede thereby mitigating the risk of adverse across different social, demographic, State regulations. Thus, Executive Order health effects to children. and economic groups within the 13132 does not apply to this proposed populations living near the facilities rule. H. Executive Order 13211: Actions where these source categories are In the spirit of Executive Order 13132, Concerning Regulations That located. The methods used to conduct and consistent with EPA policy to Significantly Affect Energy Supply, demographic analyses for this rule are promote communications between EPA Distribution, or Use described in section IV.A of the and State and local governments, EPA This action is not a ‘‘significant energy preamble for this rule. The development specifically solicits comment on this action’’ as defined under Executive of demographic analyses to inform the proposed rule from State and local Order 13211, ‘‘Actions Concerning consideration of environmental justice officials. Regulations That Significantly Affect issues in EPA rulemakings is an evolving science. The EPA offers the F. Executive Order 13175: Consultation Energy Supply, Distribution, or Use’’ (66 demographic analyses in today’s and Coordination With Indian Tribal FR 28355, May 22, 2001), because it is rulemaking as examples of how such Governments not likely to have significant adverse effect on the supply, distribution, or use analyses might be developed to inform Subject to the Executive Order 13175 of energy. This action will not create such consideration, and invites public (65 FR 67249, November 9, 2000) EPA any new requirements for sources in the comment on the approaches used and may not issue a regulation that has tribal energy supply, distribution, or use the interpretations made from the implications, that imposes substantial sectors. results, with the hope that this will direct compliance costs, and that is not support the refinement and improve required by statute, unless the Federal I. National Technology Transfer and utility of such analyses for future government provides the funds Advancement Act rulemakings. necessary to pay the direct compliance Section 12(d) of the National In the case of Primary Lead costs incurred by tribal governments, or Technology Transfer and Advancement Processing, we focused on populations EPA consults with tribal officials early Act of 1995 (‘‘NTTAA’’), Public Law within 50 km of the one facility in this in the process of developing the 104–113, 12(d) (15 U.S.C. 272 note) source category with emission sources proposed regulation and develops a directs EPA to use voluntary consensus subject to the MACT standard. More tribal summary impact statement. EPA standards in its regulatory activities specifically, for these populations we

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00031 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 9440 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules

evaluated exposures to HAP which subpart do not apply to secondary lead materials (including ore concentrate, could result in cancer risks of 1-in-1 smelters, lead refiners, or lead remelters. sinter, granulated lead, dross, slag, and million or greater, or population (b) Table 1 of this subpart specifies flue dust) are stored or handled between exposures to ambient air lead the provisions of subpart A of this part process steps, including areas in which concentrations above the level of the that apply and those that do not apply materials are stored in piles, bins, or NAAQS for lead. We compared the to owners and operators of primary lead tubs, and areas in which material is percentages of particular demographic processors. prepared for charging to a sinter groups within the focused populations 3. Section 63.1542 is amended by: machine or smelting furnace or other to the total percentages of those a. Adding in alphabetical order lead processing operation. demographic groups nationwide. The definitions for ‘‘Affirmative defense,’’ * * * * * results of this analysis are documented ‘‘Lead refiner,’’ ‘‘Lead remelter,’’ Plant roadway means any area of a in section V.B.1 (see Table 6), as well as ‘‘Primary lead processor,’’ and primary lead processor that is subject to in a technical report located in the ‘‘Secondary lead smelter’’. vehicle traffic, including traffic by b. Removing the definition for docket for this rulemaking. In brief, forklifts, front-end loaders, or vehicles ‘‘Primary lead smelter’’. although our analyses show that there is carrying ore concentrates or cast lead c. Revising the definitions for the potential for adverse environmental ingots. Excluded from this definition are ‘‘Fugitive dust source,’’ ‘‘Furnace area,’’ and human health effects from employee and visitor parking areas, ‘‘Malfunction,’’ ‘‘Materials storage and emissions of lead, it does not indicate provided they are not subject to traffic handling area,’’ ‘‘Plant roadway,’’ any significant potential for disparate by vehicles carrying lead-bearing ‘‘Process fugitive source,’’ ‘‘Refining and impacts to the specific demographic materials. casting area,’’ Sinter machine area,’’ and groups analyzed (see section V.B.1). Primary lead processor means any ‘‘Tapping location’’. Notably however, the proposed rule facility engaged in the production of would require additional control § 63.1542 Definitions. lead metal from lead sulfide ore measures to address the identified * * * * * concentrates through the use of environmental and health risks and Affirmative defense means, in the pyrometallurgical or other techniques. would therefore, decrease risks to any context of an enforcement proceeding, a Process fugitive source means a populations exposed to these sources. response or defense put forward by a source of hazardous air pollutant List of Subjects in 40 CFR Part 63 defendant, regarding which the emissions at a primary lead processor that is associated with lead smelting, Environmental protection, Air defendant has the burden of proof, and the merits of which are independently processing or refining but is not the pollution control, Reporting and primary exhaust stream and is not a recordkeeping requirements, Lead. and objectively evaluated in a judicial or administrative proceeding. fugitive dust source. Process fugitive Dated: January 31, 2011. * * * * * sources include sinter machine charging Lisa P. Jackson, Fugitive dust source means a locations, sinter machine discharge Administrator. stationary source of hazardous air locations, sinter crushing and sizing For reasons set out in the preamble, pollutant emissions at a primary lead equipment, furnace charging locations, title 40, chapter I, of the Code of Federal processor resulting from the handling, furnace taps, and drossing kettle and Regulations is proposed to be amended: storage, transfer, or other management refining kettle charging or tapping of lead-bearing materials where the locations. PART 63—[AMENDED] source is not part of a specific process, Refining and casting area means any process vent, or stack. Fugitive dust area of a primary lead processor in 1. The authority citation for part 63 which drossing or refining operations continues to read as follows: sources include roadways, storage piles, materials handling transfer points, and occur, or casting operations occur. Authority: 42 U.S.C. 7401 et seq. materials transport areas. Secondary lead smelter means any 2. Section 63.1541 is revised to read Furnace area means any area of a facility at which lead-bearing scrap as follows: primary lead processor in which a blast material, primarily, but not limited to, furnace or dross furnace is located. lead-acid batteries, is recycled into § 63.1541 Applicability. Lead refiner means any facility that elemental lead or lead alloys by (a) The provisions of this subpart refines lead metal that is not located at smelting. apply to any facility engaged in a primary lead processor. * * * * * producing lead metal from ore Lead remelter means any facility that Sinter machine area means any area concentrates. The category includes, but remelts lead metal that is not located at of a primary lead processor where a is not limited to, the following smelting a primary lead processor. sinter machine, or sinter crushing and processes: Sintering, reduction, Malfunction means any sudden, sizing equipment is located. preliminary treatment, refining and infrequent, and not reasonably * * * * * casting operations, process fugitive preventable failure of air pollution Tapping location means the opening sources, and fugitive dust sources. The control and monitoring equipment, through which lead and slag are sinter process includes an updraft or process equipment, or a process to removed from the furnace. downdraft sintering machine. The operate in a normal or usual manner 4. Section 63.1543 is revised to read reduction process includes the blast which causes, or has the potential to as follows: furnace, electric smelting furnace with a cause, the emission limitations in an converter or reverberatory furnace, and applicable standard to be exceeded. § 63.1543 Standards for process and slag fuming furnace process units. The Failures that are caused in part by poor process fugitive sources. preliminary treatment process includes maintenance or careless operation are (a) No owner or operator of any the drossing kettles and dross not malfunctions. existing, new, or reconstructed primary reverberatory furnace process units. The Materials storage and handling area lead processor shall discharge or cause refining process includes the refinery means any area of a primary lead to be discharged into the atmosphere process unit. The provisions of this processor in which lead-bearing lead compounds in excess of 0.22

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00032 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules 9441

pounds per ton of lead metal produced detection system required under requirements of this subpart no later from the aggregation of emissions § 63.1547(c)(8) does not sound for more than [DATE TWO YEARS FROM discharged from air pollution control than five percent of the total operating PUBLICATION OF THE FINAL RULE devices used to control emissions at time in a 6-month reporting period. IN THE FEDERAL REGISTER]. primary lead processing facilities, (h) The owner or operator of a (b) Each owner or operator of a new including the sources listed in primary lead processor must record the primary lead processor must achieve paragraphs (a)(1) through (a)(10) of this date and time of a bag leak detection compliance with the requirements of section. system alarm and initiate procedures to this subpart no later than [DATE 60 (1) Sinter machine; determine the cause of the alarm DAYS AFTER PUBLICATION OF THE (2) Blast furnace; according to the corrective action plan FINAL RULE IN THE FEDERAL (3) Dross furnace; required under § 63.1547(f) within 1 REGISTER] or startup, whichever is (4) Dross furnace charging location; hour of the alarm. The cause of the later. (5) Blast furnace and dross furnace alarm must be corrected as soon as 7. Section 63.1546 is revised to read tapping location; practicable. as follows: (6) Sinter machine charging location; (i) At all times, the owner or operator (7) Sinter machine discharge end; must operate and maintain any affected § 63.1546 Performance testing. (8) Sinter crushing and sizing source, including associated air (a) The following procedures must be equipment; pollution control equipment and used to determine quarterly compliance (9) Sinter machine area; and monitoring equipment, in a manner with the emissions standard for lead (10) Refining and casting, and furnace consistent with safety and good air compounds under § 63.1543(a) and (b) area. pollution control practices for for existing sources: (b) No owner or operator of any minimizing emissions. Determination of (1) Each owner or operator of existing existing, new, or reconstructed primary whether such operation and sources listed in § 63.1543(a)(1) through lead processor shall discharge or cause maintenance procedures are being used (10) must determine the lead compound to be discharged into the atmosphere will be based on information available emissions rate, in units of pounds of lead compounds in excess of 0.91 tons to the Administrator which may lead per hour according to the following per year from the air pollution control include, but is not limited to, test methods in appendices of part 60 of devices used to control emissions from monitoring results, review of operation this chapter: furnace area and refining and casting and maintenance procedures, review of (i) Method 1 to appendix A–1 of 40 operations. operation and maintenance records, and CFR part 60 must be used to select the (c) The process fugitive sources listed inspection of the source. sampling port location and the number in paragraphs (a)(4) through (a)(8) of this 5. Section 63.1544 is revised to read of traverse points. section must be equipped with a hood as follows: (ii) Methods 2 and 2F of appendix and must be ventilated to a baghouse or A–1 and Method 2G of appendix A–2 of equivalent control device. The hood § 63.1544 Standards for fugitive dust 40 CFR part 60 must be used to measure design and ventilation rate must be sources. volumetric flow rate. consistent with American Conference of (a) No owner or operator of any (iii) Methods 3, 3A, 3B of appendix Governmental Industrial Hygienists existing, new or reconstructed primary A–2 of 40 CFR part 60 must be used for recommended practices. lead processor shall discharge or cause gas analysis. (d) The sinter machine area must be to be discharged into the atmosphere (iv) Method 4 of appendix A–3 of 40 enclosed in a building that is ventilated lead compounds that cause the CFR part 60 must be used to determine to a baghouse or equivalent control concentration of lead in air to exceed moisture content of the stack gas. device at a rate that maintains a positive 0.15 μg/m3 on a 3-month rolling average (v) Method 12 of appendix A–5 or in-draft through any doorway opening. measured at locations approved by the Method 29 of appendix A–8 of 40 CFR (e) Except as provided in paragraph (f) Administrator. part 60 must be used to determine lead of this section, following the initial tests (b) At all times, the owner or operator emissions rate of the stack gas. to demonstrate compliance with must operate and maintain any affected (2) A performance test shall consist of paragraphs (a)and (b) of this section, the source, including associated air at least three runs. For each test run owner or operator of a primary lead pollution control equipment and with Method 12 of appendix A–5 or processor must conduct compliance monitoring equipment, in a manner Method 29 of appendix A–8 of 40 CFR tests for lead compounds on an consistent with safety and good air part 60, the minimum sample time must quarterly basis (no later than 100 days pollution control practices for be 60 minutes and the minimum following any previous compliance minimizing emissions. Determination of volume must be 1 dry standard cubic test). whether such operation and meter (35 dry standard cubic feet). (f) If the 12 most recent compliance maintenance procedures are being used (3) Performance tests shall be tests demonstrate compliance with the will be based on information available completed quarterly, once every 3 emission limit specified in paragraphs to the Administrator which may months, to determine compliance. (a) and (b) of this section, the owner or include, but is not limited to, (4) The lead emission rate in pounds operator of a primary lead processor monitoring results, review of operation per quarter is calculated by multiplying shall be allowed up to 12 calendar and maintenance procedures, review of the quarterly lead emission rate in months from the last compliance test to operation and maintenance records, and pounds per hour by the quarterly plant conduct the next compliance test for inspection of the source. operating time, in hours as shown in lead compounds. 6. Section 63.1545 is revised to read Equation 1: (g) The owner or operator of a primary as follows: lead processor must maintain and operate each baghouse used to control § 63.1545 Compliance dates. emissions from the sources listed in (a) Each owner or operator of an Where:

paragraphs (a)(1) through (a)(10) of this existing primary lead processor must EPb = quarterly lead emissions, pounds per section such that the alarm on a bag leak achieve compliance with the quarter;

VerDate Mar<15>2010 17:37 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00033 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 EP17FE11.001 9442 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules

ERPb = quarterly lead emissions rate, pounds normal operation with all remaining (6) Quarterly confirmation of the per hour; and doorways in their customary position physical integrity of the baghouse QPOT = quarterly plant operating time, hours during normal operation. through visual inspection of the per quarter. (4) Do not determine doorway in-draft baghouse interior for air leaks. (5) The lead production rate, in units when ambient wind speed exceeds 2 (7) Quarterly inspection of fans for of tons per quarter, must be determined meters per second. wear, material buildup, and corrosion based on production data for the (c) Performance tests shall be through visual inspection, vibration previous quarter according to the conducted under such conditions as the detectors, or equivalent means. procedures detailed in paragraphs Administrator specifies to the owner or (8) Except as provided in paragraph (a)(5)(i) through (iv) of this section: operator based on representative (h) of this section, continuous operation (i) Total lead products production performance of the affected source for of a bag leak detection system. multiplied by the fractional lead content the period being tested. Upon request, (d) The procedures specified in the must be determined in units of tons. the owner or operator shall make standard operating procedures manual (ii) Total copper matte production available to the Administrator such for maintenance must, at a minimum, multiplied by the fractional lead content records as may be necessary to include a preventative maintenance must be determined in units of tons. determine the conditions of schedule that is consistent with the (iii) Total copper speiss production performance tests. baghouse manufacturer’s instructions multiplied by the fractional lead content 8. Section 63.1547 is revised to read for routine and long-term maintenance. must be determined in units of tons. as follows: (e) The bag leak detection system (iv) Total quarterly lead production required by paragraph (c)(8) of this must be determined by summing the § 63.1547 Monitoring requirements. section must meet the specifications and values obtained in paragraphs (a)(5)(i) (a) Owners and operators of primary requirements of (e)(1) through (e)(8) of through (a)(5)(iii) of this section. lead processors must prepare, and at all this section. (6) To determine compliance with the times operate according to, a standard (1) The bag leak detection system production-based lead compound operating procedures manual that must be certified by the manufacturer to emission rate in § 63.1543(a), the describes in detail the procedures for be capable of detecting particulate quarterly production-based lead inspection, maintenance, and bag leak matter emissions at concentrations of 10 compound emission rate, in units of detection and corrective action for all milligram per actual cubic meter (0.0044 pounds of lead emissions per ton of lead baghouses that are used to control grains per actual cubic foot) or less. produced, is calculated as shown in process, process fugitive, or fugitive (2) The bag leak detection system Equation 2 by dividing lead emissions dust emissions from any source subject sensor must provide output of relative by lead production. to the lead emission standards in particulate matter loadings, and the §§ 63.1543 and 63.1544, including those owner or operator must continuously used to control emissions from general record the output from the bag leak ventilation systems. detection system. (3) The bag leak detection system (b) The standard operating procedures must be equipped with an alarm system Where: manual for baghouses required by CE = quarterly production-based lead that will sound when an increase in Pb paragraph (a) of this section must be relative particulate loading is detected compound emission rate, in units of submitted to the Administrator or pounds of lead emissions per ton of lead over a preset level, and the alarm must delegated authority for review and be located such that it can be heard or produced; approval. EPb = quarterly lead emissions, pounds per otherwise determined by the quarter; and (c) The procedures specified in the appropriate plant personnel. PPb = quarterly lead production, tons per standard operating procedures manual (4) Each bag leak detection system quarter. for inspections and routine maintenance that works based on the triboelectric (7) To determine quarterly must, at a minimum, include the effect must be installed, calibrated, and compliance with the emissions standard requirements of paragraphs (c)(1) maintained in a manner consistent with for lead compounds under § 63.1543(b), through (c)(8) of this section. guidance provided in the U.S. sum the lead compound emission rates (1) Weekly confirmation that dust is Environmental Protection Agency for the current and previous three being removed from hoppers through guidance document ’’Fabric Filter Bag quarters for the sources in § 63.1543 visual inspection or equivalent means of Leak Detection Guidance’’ (EPA–454/R– (a)(10) to determine compliance with ensuring the proper functioning of 98–015). Other bag leak detection § 63.1543(b), as determined in removal mechanisms. systems must be installed, calibrated, accordance with paragraphs (a)(1) (2) Daily check of compressed air and maintained in a manner consistent through (a)(4) of this section. supply for pulse-jet baghouses. with the manufacturer’s written (b) Owner and operators must perform (3) An appropriate methodology for specifications and recommendations. an initial compliance test to monitoring cleaning cycles to ensure (5) The initial adjustment of the demonstrate compliance with the sinter proper operation. system must, at a minimum, consist of building in-draft requirements of (4) Monthly check of bag cleaning establishing the baseline output by § 63.1543(d) at each doorway opening in mechanisms for proper functioning adjusting the sensitivity (range) and the accordance with paragraphs (b)(1) through visual inspection or equivalent averaging period of the device, and through (b)(4) of this section. means. establishing the alarm set points and the (1) Use a propeller anemometer or (5) Quarterly visual check of bag alarm delay time. equivalent device. tension on reverse air and shaker-type (6) Following initial adjustment, the (2) Determine doorway in-draft by baghouses to ensure that bags are not owner or operator must not adjust the placing the anemometer in the plane of kinked (kneed or bent) or laying on their sensitivity or range, averaging period, the doorway opening near its center. sides. Such checks are not required for alarm set points, or alarm delay time, (3) Determine doorway in-draft for shaker-type baghouses using self- except as detailed in the approved SOP each doorway that is open during tensioning (spring loaded) devices. required under paragraph (a) of this

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00034 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 EP17FE11.002 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules 9443

section. In no event shall the sensitivity initiate procedures to determine the and must be calibrated according to be increased by more than 100 percent cause of the alarm. manufacturer’s instructions. or decreased more than 50 percent over (3) The percentage of time the alarm (ii) During the initial demonstration of a 365-day period unless a responsible on the bag leak detection system sounds sinter building in-draft, and at any time official certifies that the baghouse has must be calculated as the ratio of the the owner or operator wishes to re- been inspected and found to be in good sum of alarm times to the total operating establish the baseline ventilation operating condition. time multiplied by 100. parameters, the owner or operator must (7) For negative pressure, induced air (h) Baghouses equipped with HEPA continuously record the volumetric flow baghouses, and positive pressure filters as a secondary filter used to rate through each separately ducted baghouses that are discharged to the control process or process fugitive hood, or continuously record the atmosphere through a stack, the bag leak sources subject to the lead emission volumetric flow rate at the control detector must be installed downstream standards in § 63.1543 are exempt from device inlet of each exhaust system of the baghouse and upstream of any the requirement in paragraph (c)(8) of ventilating the building and record wet acid gas scrubber. this section to be equipped with a bag exhaust system damper positions. The (8) Where multiple detectors are leak detector. The owner or operator of owner or operator must determine the required, the system’s instrumentation an affected source that uses a HEPA average volumetric flow rate(s) and alarm may be shared among filter must monitor and record the corresponding to the period of time the detectors. pressure drop across the HEPA filter in-draft compliance determinations are (f) The standard operating procedures system daily. If the pressure drop is being conducted. manual required by paragraph (a) of this outside the limit(s) specified by the (iii) The owner or operator must section must include a corrective action filter manufacturer, the owner or maintain the volumetric flow rate(s) at plan that specifies the procedures to be operator must take appropriate or above the value(s) established during followed in the event of a bag leak corrective measures, which may the most recent in-draft determination at detection system alarm. The corrective include, but not be limited to, the all times the sinter machine is in action plan must include at a minimum, following: operation. Volumetric flow rate(s) must procedures to be used to determine the (1) Inspecting the filter and filter be calculated as a 15-minute average. cause of an alarm, as well as actions to housing for air leaks and torn or broken (iv) If the volumetric flow rate is be taken to minimize emissions, which filters. monitored at the control device inlet, may include, but are not limited to, the (2) Replacing defective filter media, or the owner or operator must check and following. otherwise repairing the control device. record damper positions daily to ensure (3) Sealing off a defective control (1) Inspecting the baghouse for air they are in the positions they were in device by routing air to other leaks, torn or broken bags or filter during the most recent in-draft comparable control devices. media, or any other condition that may determination. (4) Shutting down the process (3) An owner or operator may request cause an increase in emissions. producing the particulate emissions. an alternative monitoring method by (2) Sealing off defective bags or filter (i) Owners and operators must following the procedures and media. monitor sinter machine building in-draft requirements in § 63.8(f) of the General (3) Replacing defective bags or filter to demonstrate continued compliance Provisions. media, or otherwise repairing the with the operating standard specified in (j) Each owner or operator of new or control device. § 63.1543(d) in accordance with either modified sources listed under § 63.1543 (4) Sealing off a defective baghouse paragraph (i)(1), (i)(2), or (i)(3) of this (a)(1) through (a)(10) must install, compartment. section. calibrate, maintain, and operate a (5) Cleaning the bag leak detection (1) Owners and operators must check continuous emission monitoring system system probe, or otherwise repairing or and record on a daily basis doorway in- (CEMS) for measuring lead emissions maintaining the bag leak detection draft at each doorway in accordance and a continuous emission rate system. with the methodology specified in monitoring system (CERMS) subject to (6) Shutting down the process § 63.1546(b). Performance Specification 6 of producing the particulate emissions. (2) Owners and operators must Appendix B to part 60. (g) The percentage of total operating establish and maintain baseline (1) Each owner or operator of a source time the alarm on the bag leak detection ventilation parameters which result in a subject to the emissions limits for lead system sounds in a 6-month reporting positive in-draft according to paragraphs compounds under § 63.1543(a) and (b) period must be calculated in order to (i)(2)(i) through (i)(2)(iv) of this section. must install a CEMS for measuring lead determine compliance with the five (i) Owners and operators must install, emissions within 180 days of percent operating limit in § 63.1543(h). calibrate, maintain, and operate a promulgation of performance The percentage of time the alarm on the monitoring device that continuously specifications for lead CEMS. bag leak detection system sounds must records the volumetric flow rate through (i) Prior to promulgation of be determined according to paragraphs each separately ducted hood; or install, performance specifications for CEMS (g)(1) through (g)(3) of this section. calibrate, maintain, and operate a used to measure lead concentrations, an (1) For each alarm where the owner or monitoring device that continuously owner or operator must use the operator initiates procedures to records the volumetric flow rate at the procedure described in § 63.1546(a)(1) determine the cause of an alarm within control device inlet of each exhaust through (a)(7) of this section to 1 hour of the alarm, 1 hour of alarm system ventilating the building. The determine compliance. time must be counted. flow rate monitoring device(s) can be (ii) [Reserved] (2) For each alarm where the owner or installed in any location in the exhaust (2) If a CEMS used to measure lead operator does not initiate procedures to duct such that reproducible flow rate emissions is applicable, the owner or determine the cause of the alarm within measurements will result. The flow rate operator must install a CERMS with a 1 hour of the alarm, alarm time will be monitoring device(s) must have an sensor in a location that provides counted as the actual amount of time accuracy of plus or minus 10 percent representative measurement of the taken by the owner or operator to over the normal process operating range exhaust gas flow rate at the sampling

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00035 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 9444 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules

location of the CEMS used to measure include descriptions of the sampling must submit this notification no later lead emissions, taking into account the and analytical methods used. The plan than 180 days before startup of the manufacturer’s recommendations. The may take into consideration existing constructed or reconstructed primary flow rate sensor is that portion of the monitoring being conducted under a lead processor, but no sooner than system that senses the volumetric flow state monitoring plan in accordance September 2, 1999. rate and generates an output with part 58 of this chapter. (c) The owner or operator of a primary proportional to that flow rate. (2) The owner or operator must lead processor must submit the (i) The CERMS must be designed to submit a written plan describing and compliance monitoring network plan measure the exhaust gas flow rate over explaining the basis for the design and required under § 63.1547(k)(2) to the a range that extends from a value of at adequacy of the compliance monitoring Administrator or delegated authority least 20 percent less than the lowest network, the sampling, analytical, and along with a notification that the expected exhaust flow rate to a value of quality assurance procedures, and any primary lead processor is seeking at least 20 percent greater than the other related procedures, and the review and approval of the plan. highest expected exhaust gas flow rate. justification for any seasonal, Owners or operators of existing primary (ii) The CERMS must be equipped background, or other data adjustments lead processors must submit this with a data acquisition and recording within 45 days after the effective date of notification no later than 45 days after system that is capable of recording this subpart. promulgation of this subpart. The owner values over the entire range specified in (3) The Administrator at any time may or operator of a new, reconstructed, or paragraph (b)(2)(i) of this section. require changes in, or expansion of, the modified primary lead processor must (iii) Each owner or operator must monitoring program, including submit this notification no later than perform an initial relative accuracy test additional sampling and analytical 180 days before startup of the of the CERMS in accordance with the protocols and network design. constructed or reconstructed primary applicable Performance Specification in (l) If all rolling three-month average lead processor. Appendix B to part 60 of the chapter. concentrations of lead in air measured 10. Section 63.1549 is revised to read (iv) Each owner or operator must by the compliance monitoring system as follows: operate the CERMS and record data are less than 50 percent of the lead during all periods of operation of the concentration in air limit in § 63.1544(a) § 63.1549 Recordkeeping and reporting affected facility including periods of for three consecutive years, the owner or requirements. startup, shutdown, and malfunction, operator may submit a revised plan to (a) The owner or operator of a primary except for periods of monitoring system reduce the monitoring sampling and lead processor must comply with the malfunctions, repairs associated with analysis frequency (e.g., from daily to recordkeeping requirements of § 63.10 monitoring system malfunctions, and weekly). For any subsequent period, if of subpart A, General Provisions as required monitoring system quality any rolling three-month average lead specified in Table 1 of this subpart. assurance or quality control activities concentration in air measured at any (b) In addition to the general records (including, as applicable, calibration monitor in the monitoring system required by paragraph (a) of this section, checks and required zero and span exceeds 50 percent of the concentration each owner or operator of a primary adjustments. limit in § 63.1544(a), the owner or lead processor must maintain for a (3) Each owner or operator must operator must resume monitoring period of 5 years, records of the calculate the lead emissions rate in tons pursuant to paragraph (k)(1) of this information listed in paragraphs (b)(1) per year by summing all hours of CEMS section at all monitors until another through (b)(10) of this section. data for a year to determine compliance three consecutive years of lead (1) Production records of the weight with 63.1543(b). concentration in air measurements less and lead content of lead products, (i) When the CERMS are unable to than 50 percent of the lead copper matte, and copper speiss. provide quality assured data the concentration in air limit is (2) Records of the bag leak detection following applies: demonstrated. system output. (A) When data are not available for 9. Section 63.1548 is revised to read (3) An identification of the date and periods of up to 48 hours, the highest as follows: time of all bag leak detection system recorded hourly emission rate from the alarms, the time that procedures to previous 24 hours must be used. § 63.1548 Notification requirements. determine the cause of the alarm were (B) When data are not available for 48 (a) The owner or operator of a primary initiated, the cause of the alarm, an or more hours, the maximum daily lead processor must comply with the explanation of the actions taken, and the emission rate based on the previous 30 notification requirements of § 63.9 of date and time the cause of the alarm was days must be used. subpart A, General Provisions as corrected. (ii) [Reserved] specified in Table 1 of this subpart. (4) Any recordkeeping required as (k) The owner or operator of each (b) The owner or operator of a primary part of the requirements described in source subject to § 63.1544(a) must lead processor must submit the standard the compliance monitoring system plan operate a continuous monitoring system operating procedures manual for required under § 63.1547(k)(2). for the measurement of lead compound baghouses required under § 63.1547(a) (5) Any recordkeeping required as concentrations in air. to the Administrator or delegated part of the practices described in the (1) The owner or operator must authority along with a notification that standard operating procedures manual operate compliance monitors sufficient the primary lead processor is seeking for baghouses required under in number, location, and frequency of review and approval of the manual and § 63.1547(a). sample collection to detect expected procedures. Owners or operators of (6) If an owner or operator chooses to maximum concentrations of lead existing primary lead processors must demonstrate continuous compliance compounds in air due to emissions from submit this notification no later than with the sinter building in-draft the affected source(s) in accordance November 6, 2000. The owner or requirement under § 63.1543(d) by with a written plan as described in operator of a primary lead processor that employing the method allowed in (k)(2) of this paragraph and approved by commences construction or § 63.1547(i)(1), the records of the daily the Administrator. The plan must reconstruction after April 17, 1998, doorway in-draft checks, an

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00036 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules 9445

identification of the periods when there (3) The reports must include a source to minimize emissions in was not a positive in-draft, and an calculation of the percentage of time the accordance with §§ 63.1543(i) and explanation of the corrective actions alarm on the bag leak detection system 63.1544(b), including actions taken to taken. sounded during the reporting period correct a malfunction. (7) If an owner or operator chooses to pursuant to § 63.1547(g). 11. Section 63.1550 is revised to read demonstrate continuous compliance (4) If an owner or operator chooses to as follows: with the sinter building in-draft demonstrate continuous compliance § 63.1550 Delegation of authority. requirement under § 63.1543(d) by with the sinter building in-draft employing the method allowed in requirement under § 63.1543(d) by (a) In delegating implementation and § 63.1547(i)(2), the records of the output employing the method allowed in enforcement authority to a State under from the continuous volumetric flow § 63.1547(i)(1), the reports must contain section 112(1) of the Act, the authorities monitor(s), an identification of the an identification of the periods when contained in paragraph (b) of this periods when the 15-minute volumetric there was not a positive in-draft, and an section must be retained by the flow rate dropped below the minimum explanation of the corrective actions Administrator and not transferred to a established during the most recent in- taken. State. draft determination, and an explanation (5) If an owner or operator chooses to (b) Authorities which will not be of the corrective actions taken. demonstrate continuous compliance delegated to States: No restrictions. (8) If an owner or operator chooses to with the sinter building in-draft 12. Section 63.1551 is added to read demonstrate continuous compliance requirement under § 63.1543(d) by as follows: employing the method allowed in with the sinter building in-draft § 63.1551 Affirmative defense for requirement under § 63.1543(d) by § 63.1547(i)(2), the reports must contain an identification of the periods when exceedance of emission limit during employing the method allowed in malfunction. § 63.1547(i)(2), and volumetric flow rate the 15-minute volumetric flow rate(s) dropped below the minimum In response to an action to enforce the is monitored at the baghouse inlet, standards set forth in this subpart you records of the daily checks of damper established during the most recent in- draft determination, and an explanation may assert an affirmative defense to a positions, an identification of the days of the corrective actions taken. claim for civil penalties for exceedances that the damper positions were not in (6) If an owner or operator chooses to of such standards that are caused by the positions established during the demonstrate continuous compliance malfunction, as defined in 40 CFR 63.2. most recent in-draft determination, and with the sinter building in-draft Appropriate penalties may be assessed, an explanation of the corrective actions requirement under § 63.1543(d) by however, if you fail to meet your burden taken. employing the method allowed in of proving all the requirements in the (9) Records of the occurrence and § 63.1547(i)(2), and volumetric flow rate affirmative defense. The affirmative duration of each malfunction of is monitored at the baghouse inlet, the defense shall not be available for claims operation (i.e., process equipment) or reports must contain an identification of for injunctive relief. the air pollution control equipment and the days that the damper positions were (a) To establish the affirmative monitoring equipment. not in the positions established during defense in any action to enforce such a (10) Records of actions taken during the most recent in-draft determination, limit, you must timely meet the periods of malfunction to minimize and an explanation of the corrective notification requirements in paragraph emissions in accordance with actions taken. (b) of this section, and must prove by a §§ 63.1543(i) and 63.1544(e), including (7) The reports must contain a preponderance of evidence that: corrective actions to restore summary of the records maintained as (1) The excess emissions: malfunctioning process and air part of the practices described in the (i) Were caused by a sudden, short, pollution control and monitoring standard operating procedures manual infrequent, and unavoidable failure of equipment to its normal or usual for baghouses required under air pollution control and monitoring manner of operation. § 63.1547(a), including an explanation equipment, process equipment, or a (c) Records for the most recent 2 years of the periods when the procedures process to operate in a normal or usual of operation must be maintained on site. were not followed and the corrective manner; and Records for the previous 3 years may be actions taken. (ii) Could not have been prevented maintained off site. (8) The reports must contain a through careful planning, proper design (d) The owner or operator of a summary of the compliance monitoring or better operation and maintenance primary lead processor must comply results for the required reporting period, practices; and with the reporting requirements of including an explanation of any periods (iii) Did not stem from any activity or § 63.10 of subpart A, General Provisions when the procedures outlined in the event that could have been foreseen and as specified in Table 1 of this subpart. compliance monitoring system plan avoided, or planned for; and (e) In addition to the information required by § 63.1547(k)(2) were not (iv) Were not part of a recurring required under § 63.10 of the General followed and the corrective actions pattern indicative of inadequate design, Provisions, the owner or operator must taken. operation, or maintenance; and provide semi-annual reports containing (9) If there was a malfunction during (2) Repairs were made as the information specified in paragraphs the reporting period, the report shall expeditiously as possible when the (e)(1) through (e)(9) of this section to the also include the number, duration, and applicable emission limitations were Administrator or designated authority. a brief description for each type of being exceeded. Off-shift and overtime (1) The reports must include records malfunction which occurred during the labor were used, to the extent of all alarms from the bag leak detection reporting period and which caused or practicable to make these repairs; and system specified in § 63.1547(e). may have caused any applicable (3) The frequency, amount and (2) The reports must include a emission limitation to be exceeded. The duration of the excess emissions description of the actions taken report must also include a description of (including any bypass) were minimized following each bag leak detection actions taken by an owner or operator to the maximum extent practicable system alarm pursuant to § 63.1547(f). during a malfunction of an affected during periods of such emissions; and

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00037 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 9446 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules

(4) If the excess emissions resulted (8) At all times, the facility was Administrator by telephone or facsimile from a bypass of control equipment or operated in a manner consistent with (FAX) transmission as soon as possible, a process, then the bypass was good practices for minimizing but no later than two business days after unavoidable to prevent loss of life, emissions; and the initial occurrence of the severe personal injury, or severe (9) A written root cause analysis has malfunction, if it wishes to avail itself property damage; and been prepared to determine, correct and of an affirmative defense to civil (5) All possible steps were taken to eliminate the primary causes of the penalties for that malfunction. The minimize the impact of the excess malfunction and the excess emissions owner or operator seeking to assert an emissions on ambient air quality, the resulting from the malfunction event at affirmative defense shall also submit a environment and human health; and issue. The analysis shall also specify, written report to the Administrator using best monitoring methods and within 30 days of the initial occurrence (6) All emissions monitoring and engineering judgment, the amount of of the exceedance of the standard in this control systems were kept in operation excess emissions that were the result of subpart to demonstrate, with all if at all possible; and the malfunction. necessary supporting documentation, (7) All of the actions in response to (b) Notification. The owner or that it has met the requirements set forth the excess emissions were documented operator of the facility experiencing an in paragraph (a) of this section. by properly signed, contemporaneous exceedance of its emission limit(s) 12. Table 1 to Subpart TTT of Part 63 operating logs; and during a malfunction shall notify the is revised to read as follows:

TABLE 1 TO SUBPART TTT OF PART 63—GENERAL PROVISIONS APPLICABILITY TO SUBPART TTT

Applies to subpart Reference TTT Explanation

§ 63.1 ...... Yes. § 63.2 ...... Yes. § 63.3 ...... Yes. § 63.4 ...... Yes. § 63.5 ...... Yes. § 63.6(a), (b), (c) ...... Yes. § 63.6 (d) ...... No ...... Section reserved. § 63.6(e)(1)(i) ...... No ...... See § 63.1543(i) and § 63.1544(b) for general duty requirement. § 63.6(e)(1)(ii) ...... No. § 63.6(e)(1)(iii) ...... Yes. § 63.6(e)(2) ...... No ...... Section reserved. § 63.6(e)(3) ...... No. § 63.6(f)(1) ...... No. § 63.6(g) ...... Yes. § 63.6(h) ...... No ...... No opacity limits in rule. § 63.6(i) ...... Yes. § 63.6(j) ...... Yes. § 63.7(a)–(d) ...... Yes. § 63.7(e)(1) ...... No ...... See § 63.1546(c). § 63.7(e)(2)–(e)(4) ...... Yes. § 63.7(f), (g), (h) ...... Yes. § 63.8(a)–(b) ...... Yes. § 63.8(c)(1)(i) ...... No. § 63.8(c)(1)(ii) ...... Yes. § 63.8(c)(1)(iii) ...... No. § 63.8(c)(2)–(d)(2) ...... Yes. § 63.8(d)(3) ...... Yes, except for last sentence. § 63.8(e)–(g) ...... Yes. § 63.9(a), (b), (c), (e), (g), (h)(1) through (3), (h)(5) and (6), (i) and (j) ...... Yes. § 63.9(f) ...... No. § 63.9(h)(4) ...... No ...... Reserved. § 63.10(b)(2)(i) ...... No. § 63.10(b)(2)(ii) ...... No ...... See § 63.1549(b)(9) and (10) for recordkeeping of occurrence and duration of malfunctions and recordkeeping of actions taken during malfunction. § 63.10(b)(2)(iii) ...... Yes. § 63.10(b)(2)(iv)–(b)(2)(v) ...... No. § 63.10(b)(2)(vi)–(b)(2)(xiv) ...... Yes. § 63.(10)(b)(3) ...... Yes. § 63.10(c)(1)–(9) ...... Yes. § 63.10(c)(10)–(11) ...... No ...... See § 63.1549(b)(9) and (10) for recordkeeping of malfunctions. § 63.10(c)(12)–(c)(14) ...... Yes. § 63.10(c)(15) ...... No. § 63.10(d)(1)–(4) ...... Yes.

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00038 Fmt 4701 Sfmt 4702 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2 Federal Register / Vol. 76, No. 33 / Thursday, February 17, 2011 / Proposed Rules 9447

TABLE 1 TO SUBPART TTT OF PART 63—GENERAL PROVISIONS APPLICABILITY TO SUBPART TTT—Continued

Applies to subpart Reference TTT Explanation

§ 63.10(d)(5) ...... No ...... See § 63.1549(e)(9) for reporting of malfunctions. § 63.10(e)–((f) ...... Yes. § 63.11 ...... No ...... Flares will not be used to comply with the emission limits. § 63.12 through 63.15 ...... Yes.

[FR Doc. 2011–2866 Filed 2–16–11; 8:45 am] BILLING CODE 6560–50–P

VerDate Mar<15>2010 16:48 Feb 16, 2011 Jkt 223001 PO 00000 Frm 00039 Fmt 4701 Sfmt 9990 E:\FR\FM\17FEP2.SGM 17FEP2 jlentini on DSKJ8SOYB1PROD with PROPOSALS2