EPA-Emission of Coal

24854 Federal Register / Vol. 88, No.
78 / Monday, April 24, 2023 / Proposed Rules
ENVIRONMENTAL PROTECTION Docket ID No. EPA–HQ–OAR–2018– air-pollution/mercury-and-air-toxics-

AGENCY 0794, Mail Code 28221T, 1200 standards.
Pennsylvania Avenue NW, Washington, The EPA will make every effort to
40 CFR Part 63 DC 20460. follow the schedule as closely as
[EPA–HQ–OAR–2018–0794; FRL–6716.3– • Hand/Courier Delivery: EPA Docket possible on the day of the hearing;
01–OAR] Center, WJC West Building, Room 3334, however, please plan for the hearings to
1301 Constitution Avenue NW, run either ahead of schedule or behind
RIN 2060–AV53 Washington, DC 20004. The Docket schedule.
Center’s hours of operation are 8:30 Each commenter will have 4 minutes
National Emission Standards for to provide oral testimony. The EPA
a.m.–4:30 p.m., Monday–Friday (except
Hazardous Air Pollutants: Coal- and encourages commenters to provide the
federal holidays).
Oil-Fired Electric Utility Steam Instructions: All submissions received EPA with a copy of their oral testimony
Generating Units Review of the must include the Docket ID No. for this by submitting the text of your oral
Residual Risk and Technology Review rulemaking. Comments received may be testimony as written comments to the
AGENCY: Environmental Protection posted without change to https:// rulemaking docket.
Agency (EPA). www.regulations.gov/, including any The EPA may ask clarifying questions
personal information provided. For during the oral presentations but will
ACTION: Proposed rule.
detailed instructions on sending not respond to the presentations at that
SUMMARY: The EPA is proposing to comments and additional information time. Written statements and supporting
amend the National Emission Standards on the rulemaking process, see the information submitted during the
for Hazardous Air Pollutants (NESHAP) SUPPLEMENTARY INFORMATION section of comment period will be considered
for Coal- and Oil-Fired Electric Utility this document. with the same weight as oral testimony
Steam Generating Units (EGUs), FOR FURTHER INFORMATION CONTACT: For
and supporting information presented at
commonly known as the Mercury and questions about this proposed action, the public hearing.
Air Toxics Standards (MATS). Please note that any updates made to
contact Sarah Benish, Sector Policies
Specifically, the EPA is proposing to any aspect of the hearing will be posted
and Programs Division (D243–01),
amend the surrogate standard for non- online at https://www.epa.gov/
Office of Air Quality Planning and
mercury (Hg) metal HAP (filterable stationary-sources-air-pollution/
Standards, U.S. Environmental
particulate matter (fPM)) for existing mercury-and-air-toxics-standards.
Protection Agency, Research Triangle
coal-fired EGUs; the fPM compliance While the EPA expects the hearing to go
Park, North Carolina 27711; telephone
demonstration requirements; the Hg forward as described in this section,
number: (919) 541–5620; and email
standard for lignite-fired EGUs; and the please monitor our website or contact
address: benish.sarah@epa.gov.
definition of startup. These proposed the public hearing team at (888) 372–
SUPPLEMENTARY INFORMATION: 8699 or by email at
amendments are the result of the EPA’s Participation in virtual public
review of the May 22, 2020 residual risk SPPDpublichearing@epa.gov to
hearing. The public hearing will be held determine if there are any updates. The
and technology review (RTR) of MATS. via virtual platform on May 9, 2023 and EPA does not intend to publish a
DATES: will convene at 11 a.m. Eastern Time document in the Federal Register
Comments. Comments must be (ET) and conclude at 7 p.m. ET. If the announcing updates.
received on or before June 23, 2023. EPA receives a high volume of If you require the services of an
Under the Paperwork Reduction Act registrations for the public hearing, we interpreter or special accommodation
(PRA), comments on the information may continue the public hearing on May such as audio description, please pre-
collection provisions are best assured of 10, 2023. The EPA may close a session register for the hearing with the public
consideration if the Office of 15 minutes after the last pre-registered hearing team and describe your needs
Management and Budget (OMB) speaker has testified if there are no by May 1, 2023. The EPA may not be
receives a copy of your comments on or additional speakers. The EPA will able to arrange accommodations without
before May 24, 2023. announce further details at https:// advanced notice.
Public hearing. The EPA will hold a www.epa.gov/stationary-sources-air- Docket. The EPA has established a
virtual public hearing on May 9, 2023. pollution/mercury-and-air-toxics- docket for this rulemaking under Docket
See SUPPLEMENTARY INFORMATION for standards. ID No. EPA–HQ–OAR–2018–0794.1 All
information on requesting and The EPA will begin pre-registering documents in the docket are listed in
registering for a public hearing. speakers for the hearing no later than 1 https://www.regulations.gov/. Although
ADDRESSES: You may send comments, business day following publication of listed, some information is not publicly
identified by Docket ID No. EPA–HQ– this document in the Federal Register. available, e.g., Confidential Business
OAR–2018–0794, by any of the The EPA will accept registrations on an
following methods: individual basis. To register to speak at 1 As explained in a memorandum to the docket,
• Federal eRulemaking Portal: the virtual hearing, please use the the docket for this action includes the documents
https://www.regulations.gov/ (our online registration form available at and information, in whatever form, in Docket ID
Nos. EPA–HQ–OAR–2009–0234 (National Emission
preferred method). Follow the online https://www.epa.gov/stationary-sources- Standards for Hazardous Air Pollutants for Coal-
instructions for submitting comments. air-pollution/mercury-and-air-toxics- and Oil-fired Electric Utility Steam Generating
ddrumheller on DSK120RN23PROD with PROPOSALS2
• Email: a-and-r-docket@epa.gov. standards or contact the public hearing Units), EPA–HQ–OAR–2002–0056 (National
Include Docket ID No. EPA–HQ–OAR– team at (888) 372–8699 or by email at Emission Standards for Hazardous Air Pollutants
for Utility Air Toxics; Clean Air Mercury Rule
2018–0794 in the subject line of the SPPDpublichearing@epa.gov. The last (CAMR)), and Legacy Docket ID No. A–92–55
message. day to pre-register to speak at the (Electric Utility Hazardous Air Pollutant Emission
• Fax: (202) 566–9744. Attention hearing will be May 8, 2023. Prior to the Study). See memorandum titled Incorporation by
Docket ID No. EPA–HQ–OAR–2018– hearing, the EPA will post a general reference of Docket Number EPA–HQ–OAR–2009–
0234, Docket Number EPA–HQ–OAR–2002–0056,
0794. agenda that will list pre-registered and Docket Number A–92–55 into Docket Number
• Mail: U.S. Environmental speakers in approximate order at: EPA–HQ–OAR–2018–0794 (Docket ID Item No.
Protection Agency, EPA Docket Center, https://www.epa.gov/stationary-sources- EPA–HQ–OAR–2018–0794–0005).
VerDate Sep<11>2014 19:44 Apr 21, 2023 Jkt 259001 PO 00000 Frm 00002 Fmt 4701 Sfmt 4702 E:\FR\FM\24APP2.SGM 24APP2
Federal Register / Vol. 88, No. 78 / Monday, April 24, 2023 / Proposed Rules 24855
Information (CBI) or other information Electronic files should not include Preamble acronyms and
whose disclosure is restricted by statute. special characters or any form of abbreviations. Throughout this
Certain other material, such as encryption and be free of any defects or document the use of ‘‘we,’’ ‘‘us,’’ or
copyrighted material, is not placed on viruses. For additional information ‘‘our’’ is intended to refer to the EPA.
the internet and will be publicly about the EPA’s public docket, visit the We use multiple acronyms and terms in
available only in hard copy. With the EPA Docket Center homepage at https:// this preamble. While this list may not be
exception of such material, publicly www.epa.gov/dockets. exhaustive, to ease the reading of this
available docket materials are available Submitting CBI. Do not submit preamble and for reference purposes,
electronically in Regulations.gov. information containing CBI to the EPA the EPA defines the following terms and
Instructions. Direct your comments to through https://www.regulations.gov/. acronyms here:
Docket ID No. EPA–HQ–OAR–2018– Clearly mark the part or all of the Btu British Thermal Units
0794. The EPA’s policy is that all information that you claim to be CBI. CAA Clean Air Act
comments received will be included in For CBI information on any digital CBI Confidential Business Information
the public docket without change and storage media that you mail to the EPA, CEMS continuous emissions monitoring
may be made available online at https:// note the Docket ID No., mark the outside systems
www.regulations.gov/, including any CFR Code of Federal Regulations
of the digital storage media as CBI, and
personal information provided, unless CO2 carbon dioxide
identify electronically within the digital CPMS continuous parameter monitoring
the comment includes information storage media the specific information system
claimed to be CBI or other information that is claimed as CBI. In addition to EAV equivalent annualized value
whose disclosure is restricted by statute. one complete version of the comments ECMPS Emissions Collection and
Do not submit electronically to https:// that includes information claimed as Monitoring Plan System
www.regulations.gov/ any information CBI, you must submit a copy of the EGU electric utility steam generating unit
that you consider to be CBI or other comments that does not contain the EIA Energy Information Administration
information whose disclosure is EJ environmental justice
information claimed as CBI directly to EPA Environmental Protection Agency
restricted by statute. This type of the public docket through the
information should be submitted as ESP electrostatic precipitator
procedures outlined in Instructions FF fabric filter
discussed in the Submitting CBI section section of this document. If you submit FGD flue gas desulfurization
of this document. any digital storage media that does not fPM filterable particulate matter
The EPA may publish any comment GWh gigawatt-hour
contain CBI, mark the outside of the
received to its public docket. HAP hazardous air pollutant(s)
digital storage media clearly that it does
Multimedia submissions (audio, video, HCl hydrogen chloride
not contain CBI and note the Docket ID
etc.) must be accompanied by a written HF hydrogen fluoride
comment. The written comment is No. Information not marked as CBI will Hg mercury
considered the official comment and be included in the public docket and the Hg0 elemental Hg vapor
should include discussion of all points EPA’s electronic public docket without HQ hazard quotient
you wish to make. The EPA will prior notice. Information marked as CBI IGCC integrated gasification combined
will not be disclosed except in cycle
generally not consider comments or IPM Integrated Planning Model
comment contents located outside of the accordance with procedures set forth in
40 Code of Federal Regulations (CFR) lb Pounds
primary submission (i.e., on the Web, LEE low emitting EGU
cloud, or other file sharing system). For part 2.
MACT maximum achievable control
additional submission methods, the full Our preferred method to receive CBI technology
EPA public comment policy, is for it to be transmitted electronically MATS Mercury and Air Toxics Standards
information about CBI or multimedia using email attachments, File Transfer MM million
submissions, and general guidance on Protocol (FTP), or other online file MW megawatt
sharing services (e.g., Dropbox, NAICS North American Industry
making effective comments, please visit
OneDrive, Google Drive). Electronic Classification System
https://www.epa.gov/dockets/ NEEDS National Electric Energy Data
commenting-epa-dockets. submissions must be transmitted
System
The https://www.regulations.gov/ directly to the OAQPS CBI Office at the NESHAP National Emission Standards for
website allows you to submit your email address oaqpscbi@epa.gov, and as Hazardous Air Pollutants
comment anonymously, which means described above, should include clear OAQPS Office of Air Quality Planning and
the EPA will not know your identity or CBI markings and note the Docket ID Standards
contact information unless you provide No. If assistance is needed with OMB Office of Management and Budget
it in the body of your comment. If you submitting large electronic files that PDF Portable Document Format
send an email comment directly to the exceed the file size limit for email PM particulate matter
ppm parts per million
EPA without going through https:// attachments, or if you do not have your
PV present value
www.regulations.gov/, your email own file sharing service, please email RIA regulatory impact analysis
address will be automatically captured oaqpscbi@epa.gov to request a file RTR residual risk and technology review
and included as part of the comment transfer link. If sending CBI information SC–CO2 social cost of carbon
that is placed in the public docket and through the postal service, please send SO2 sulfur dioxide
made available on the internet. If you it to the following address: OAQPS tpy tons per year
submit an electronic comment, the EPA Document Control Officer (C404–02), TBtu trillion British thermal units
recommends that you include your OAQPS, U.S. Environmental Protection WebFIRE Web Factor Information Retrieval
name and other contact information in Agency, Research Triangle Park, North System
the body of your comment and with any Carolina 27711, Attention Docket ID No. Organization of this document. The
digital storage media you submit. If the EPA–HQ–OAR–2018–0794. The mailed information in this preamble is
EPA cannot read your comment due to CBI material should be double wrapped organized as follows:
technical difficulties and cannot contact and clearly marked. Any CBI markings I. Executive Summary
you for clarification, the EPA may not should not show through the outer A. Background and Purpose of the
be able to consider your comment. envelope. Regulatory Action
24856 Federal Register / Vol. 88, No. 78 / Monday, April 24, 2023 / Proposed Rules
B. Summary of the Major Provisions of the I. Executive Summary determined not to update the MACT
Regulatory Action standard.
II. General Information A. Background and Purpose of the
A. Does this action apply to me? Regulatory Action On January 20, 2021, President Biden
B. Where can I get a copy of this document signed Executive Order 13990,
Exposure to hazardous air pollution
and other related information? (‘‘HAP,’’ sometimes known as toxic air ‘‘Protecting Public Health and the
III. Background pollution, including Hg, chromium, Environment and Restoring Science to
A. What is the authority for this action? arsenic, and lead) can cause a range of Tackle the Climate Crisis’’ (86 FR 7037;
B. What is this source category and how
adverse health effects including January 25, 2021). The Executive order,
does the current NESHAP regulate its among other things, instructed the EPA
harming people’s central nervous
HAP emissions? to review the 2020 final rule titled,
C. What data collection activities were
system; damage to their kidneys; and
cancer. Recognizing the dangers posed ‘‘National Emission Standards for
conducted to support this proposed Hazardous Air Pollutants: Coal- and Oil-
action? by HAP, Congress enacted Clean Air Act
(CAA) section 112. Under CAA section Fired Electric Utility Steam Generating
D. What other relevant background
information and data are available? 112, the EPA is required to set standards Units—Reconsideration of
E. How does the EPA perform the (known as ‘‘MACT’’ (maximum Supplemental Finding and Residual
technology review? achievable control technology) Risk and Technology Review’’ (85 FR
IV. Review of 2020 Residual Risk and standards) for major sources of HAP that 31286; May 22, 2020) (2020 Final
Technology Review ‘‘require the maximum degree of Action) and to consider publishing a
A. Summary of the 2020 Residual Risk reduction in emissions of the hazardous notice of proposed rulemaking
Review air pollutants . . . (including a suspending, revising, or rescinding that
B. Summary of the 2020 Technology prohibition on such emissions, where action. The 2020 Final Action included
Review a finding that it is not appropriate and
achievable) that the Administrator,
V. Analytical Results and Proposed Decisions necessary to regulate coal- and oil-fired
A. Review of the 2020 Residual Risk
taking into consideration the cost of
Review achieving such emission reduction, and EGUs under CAA section 112 as well as
B. Review of the 2020 Technology Review any non-air quality health and the RTR for the MATS rule. The results
C. What are the results and proposed environmental impacts and energy of the EPA’s review of the 2020
decisions based on our technology requirements, determines is appropriate and necessary finding were
review, and what is the rationale for achievable.’’ 42 U.S.C. 7412(d)(2). To proposed on February 9, 2022 (87 FR
those decisions? ensure a minimum level (or ‘‘floor’’) of 7624) (2022 Proposal) and finalized on
D. What other actions are we proposing, emissions reductions, Congress required March 6, 2023 (88 FR 13956). In the
and what is the rationale for those that MACT standards for existing 2022 Proposal, the EPA also solicited
actions? sources ‘‘shall not be less stringent than information on the performance and
E. What compliance dates are we . . . the average emission limitation cost of new or improved technologies
proposing, and what is the rationale for achieved by the best performing 12 that control hazardous air pollutant
the proposed compliance dates? percent of existing sources’’; and MACT
VI. Summary of Cost, Environmental, and (HAP) emissions, improved methods of
standards for new sources ‘‘shall not be operation, and risk-related information
Economic Impacts
A. What are the affected sources? less stringent than the emission control to further inform the EPA’s review of
B. What are the air quality impacts? that is achieved in practice by the best the 2020 MATS RTR. This action
C. What are the cost impacts? controlled similar source[.]’’ 42 U.S.C. presents the proposed results of the
D. What are the economic impacts? 7412(d)(3). These requirements EPA’s review of the MATS RTR.
E. What are the benefits? effectively obligated all sources to
F. What analysis of environmental justice reduce emissions as well as the best In particular, with respect to the
did we conduct? sources in their category. Congress did standard for fPM (as a surrogate for non-
VII. Request for Comments not stop there, however. First, it Hg metals), and the standard for Hg
VIII. Statutory and Executive Order Reviews required the EPA, 8 years after setting from EGUs that burn lignite coal, the
A. Executive Order 12866: Regulatory the standard, to address any residual EPA proposes to conclude that
Planning and Review and Executive risks posed by the source category developments since 2012—and in
Order 13563: Improving Regulation and (called the ‘‘residual risk review’’). particular the fact that the majority of
Regulatory Review sources are vastly outperforming the
Second, and as explained in more detail
B. Paperwork Reduction Act (PRA) MACT standards with control
C. Regulatory Flexibility Act (RFA) below, it required the EPA, at least
every 8 years on an ongoing basis, to technologies that are cheaper and more
D. Unfunded Mandates Reform Act
review and revise as necessary the effective than the EPA forecast while a
(UMRA)
E. Executive Order 13132: Federalism MACT standard taking into account smaller number of sources’ performance
F. Executive Order 13175: Consultation developments in practices, processes lags behind—warrant strengthening
and Coordination With Indian Tribal and control technologies (called the these standards. While the 2012 MATS
Governments ‘‘technology review’’). For EGUs, drove critical HAP reductions at much
G. Executive Order 13045: Protection of Congress also required the EPA to make lower cost than estimated, coal-fired
Children From Environmental Health a one-time determination of whether it EGUs still emit a substantial amount of
Risks and Safety Risks is ‘‘appropriate and necessary’’ to HAP and developments since 2012
H. Executive Order 13211: Actions regulate this source category under CAA provide opportunities to address these
Concerning Regulations That section 112. The EPA found regulation emissions and ensure that all coal-fired
Significantly Affect Energy Supply,
of EGUs ‘‘appropriate and necessary’’ in EGUs are performing at levels
Distribution, or Use
I. National Technology Transfer and
2000 and reaffirmed that finding in 2012 achievable by the fleet. These proposed
Advancement Act (NTTAA) and 2016. MACT standards were revisions would ensure that the EPA’s
J. Executive Order 12898: Federal Actions originally set for EGUs in 2012, and standards continue to fulfill Congress’s
To Address Environmental Justice in those standards remain in place today. direction to require the maximum
Minority Populations and Low-Income In 2020, the EPA conducted the 8-year degree of reduction of HAP while taking
Populations residual risk and technology review and into account the statutory factors.
B. Summary of the Major Provisions of for the non-Hg metals. For existing coal- the EPA is proposing to require that all
the Regulatory Action fired EGUs, most units have chosen to coal-fired EGUs demonstrate
The 2012 MATS Final Rule demonstrate compliance with the non- compliance with the fPM emission
established emission standards to limit Hg metal HAP surrogate fPM emission standard by using PM CEMS.
emissions of HAP from coal- and oil- standard of 3.0E–02 pounds of fPM per Accordingly, because almost all
fired EGUs. The rule required that million British thermal units of heat regulated sources have chosen to
affected sources limit emissions of Hg, input (lb/MMBtu). demonstrate compliance with the non-
of non-Hg metal HAP (e.g., chromium, CAA section 112(d)(2) directs the EPA Hg HAP metal standards by
nickel, arsenic, lead), acid gas HAP (e.g., to require the maximum degree of HAP demonstrating compliance with the
hydrogen chloride (HCl), hydrogen emission reductions achievable, taking surrogate fPM standard and because of
fluoride (HF), selenium dioxide (SeO2)), into account certain considerations, and the benefits of PM CEMS use for
and organic HAP (e.g., formaldehyde, CAA section 112(d)(3) sets the floor for demonstrating compliance, the EPA is
dioxins/furans). Since MATS was emission standards based on the proposing to remove the total and
promulgated in 2012, power sector reductions achieved by the best individual non-Hg metals emission
emissions of Hg, acid gas HAP, and non- performing sources. The MATS was limits from MATS. Requiring the use of
Hg metal HAP have decreased by about based upon the EPA’s analysis under PM CEMS, if finalized, would also
86 percent, 96 percent, and 81 percent, CAA sections 112(d)(2) and (d)(3) in render the current compliance method
respectively, as compared to 2010 2012. CAA section 112(d)(6) further for the LEE program superfluous, since
emissions levels (See Table 4 at 84 FR requires the EPA, at least every 8 years, LEE is an optional stack testing program
2689, February 7, 2019). Still, coal- and to review and revise standards taking and the considered fPM limits are both
oil-fired EGUs remain the largest into account developments in practices, below the current fPM LEE program
domestic emitter of Hg and many other processes and control technologies. limit of 1.5E–02 lb/MMBtu (i.e., 50
HAP, including many of the non-Hg After reviewing developments in the percent of the current fPM standard).
HAP metals and HCl. Exposure to these current emission levels of fPM from Therefore, the EPA also proposes to
HAP, at certain levels and duration, is existing coal-fired EGUs, the costs of remove fPM, as well as the total and
associated with a variety of adverse control technologies, and the individual non-Hg HAP metals, from the
health effects. These adverse health effectiveness of those technologies, as LEE program.
well as the costs of meeting a standard The EPA is also proposing to establish
effects may include irritation of the
that is more stringent than 3.0E–02 lb/ a more protective Hg emission standard
lung, skin, and mucus membranes;
MMBtu and the other statutory factors, for existing lignite-fired EGUs.
detrimental effects on the central
the EPA is proposing to revise the non- Currently, existing lignite-fired EGUs
nervous system; damage to the kidneys;
Hg metal surrogate fPM emission must meet a Hg emission standard of
alimentary effects such as nausea and
standard for all existing coal-fired EGUs 4.0E–06 lb/MMBtu 6 or an alternative
vomiting; and cancer.2 See 77 FR 9310
to a more stringent fPM emission output-based emission standard of 4.0E–
for a fuller discussion of the health
standard of 1.0E–02 lb/MMBtu, which is 02 pounds of Hg per gigawatt-hour
effects associated with these pollutants.
comparable to the MATS new source output (lb/GWh). The EPA recently
Three of the key metal HAP emitted by
standard for fPM.5 The EPA is also collected information on current Hg
EGUs (inorganic arsenic (As),
soliciting comment on opportunities to emission levels and controls for lignite-
hexavalent chromium (Cr), and nickel
revise the MATS fPM emission standard fired EGUs from information provided
compounds (Ni)) have been classified as
to an even more stringent level of 6.0E– routinely to the EPA and to the Energy
human carcinogens, while two others
03 lb/MMBtu. Information Administration (EIA) and
(cadmium (Cd) and selenium (Se)) are
The EPA is also proposing a revision by using the information collection
classified as probable human
to the requirements for demonstrating authority provided under CAA section
carcinogens.3
To address emissions of these non-Hg compliance with the fPM emission 114. That information showed
metal HAP, MATS sets individual standard. Currently, EGUs that do not developments that demonstrate that
emission limits for each of the 10 non- qualify for the low emitting EGU (LEE) lignite-fired EGUs can achieve a Hg
Hg metals emitted from coal- and oil- program can demonstrate compliance emission rate that is much lower than
fired EGUs. Alternatively, affected with the fPM standard either by the current standard, and that there are
sources may meet an emission standard conducting quarterly performance cost-effective control technologies and
for ‘‘total non-Hg metals’’ by summing testing (i.e., quarterly stack testing) or by methods of operation that are available
the emission rates of each of the non-Hg using PM continuous emission to achieve a more stringent standard.
metals. The MATS rule also allows monitoring systems (PM CEMS). After Accordingly, the EPA is proposing that
affected sources to meet a filterable PM considering updated information on the lignite-fired EGUs must meet the same
(fPM) 4 emission standard as a surrogate costs for quarterly performance testing Hg emission standard as EGUs firing
compared to the costs of PM CEMS and other types of coal (i.e., bituminous, and
2 77 FR 9310. on the measurement capabilities of PM subbituminous) which is 1.2 lb/TBtu or
3 U.S. EPA. Table 1. Prioritized Chronic Dose- CEMS, as well as other benefits of using an alternative output-based standard of
Response Values for Screening Risk Assessments. PM CEMS, which include increased 1.3E–02 lb/GWh. The EPA is not
Available at: https://www.epa.gov/fera/dose- transparency and accelerated proposing to revise the current Hg
response-assessment-assessing-health-risks-
identification of anomalous emissions, emission standard for existing EGUs
associated-exposure-hazardous-air-pollutants.
4 Total PM is composed of the filterable PM firing non-lignite coal.
fraction (fPM) and the condensible PM fraction. In will allow the use of continuous PM monitoring Finally, the EPA is proposing to
establishing fPM as a surrogate for the non-Hg metal systems, which measure filterable (but not total) remove one of the two options for
HAP, the EPA explained that most of the non-Hg PM, thereby providing a more continuous measure
of compliance.
defining the startup period for MATS-
metal HAP are present overwhelmingly in the fPM
fraction. Selenium may be present in both the fPM 5 The fPM standard for new coal-fired EGU is affected EGUs. The first option defines
fraction and/or as the acid gas, SeO2, in the 9.0E–02 lb/MWh, which is an output-based
condensible PM fraction. SeO2 is an acid gas HAP emission standard. See 78 FR 24073. This emission 6 The emission standard of 4.0E–06 lb/MMBtu is
and is well controlled by the emission limit for acid is equivalent for a new coal-fired EGU with a heat more often written as 4.0 lb/TBtu (pounds of Hg per
gas HAP. In addition, using fPM as the surrogate rate of 9.0 MMBtu/MWh (9,000 Btu/kWh). trillion British thermal units).
startup as either the first-ever firing of rulemakings. For that reason, the EPA published under section 112(c) of the
fuel in a boiler for the purpose of has not reviewed and is not proposing CAA on December 20, 2000 (65 FR
producing electricity, or the firing of any revisions to the MATS new source 79825). CAA section 112(a)(8) defines
fuel in a boiler after a shutdown event emission standards. In some cases, an EGU as any fossil fuel-fired
for any purpose. Under the first option, however, proposed revisions to existing combustion unit of more than 25
startup ends when any of the steam source emission standards may be more megawatts (MW) that serves a generator
from the boiler is used to generate stringent than the corresponding new that produces electricity for sale. A unit
electricity for sale over the grid or for source emission standard. In those that cogenerates steam and electricity
any other purpose (including on-site instances, the EPA has addressed that and supplies more than one-third of its
use). In the second option, startup is illogical outcome by proposing to revise potential electric output capacity and
defined as the period in which the corresponding new source standard more than 25 MW electrical output to
operation of an EGU is initiated for any to be at least as stringent as the any utility power distribution system for
purpose, and startup begins with either proposed revision to the existing source sale is also considered an EGU.
the firing of any fuel in an EGU for the standard. B. Where can I get a copy of this
purpose of producing electricity or The EPA is also not proposing to document and other related
useful thermal energy (such as heat or revise MATS emission standards for information?
steam) for industrial, commercial, existing Integrated Gasification
heating, or cooling purposes (other than Combined Cycle (IGCC) EGUs, nor to In addition to being available in the
the first-ever firing of fuel in a boiler the MATS emission standards for any of docket, an electronic copy of this action
following construction of the boiler) or the subcategories of existing oil-fired is available on the internet. Following
for any other purpose after a shutdown EGUs. signature by the EPA Administrator, the
event. Under the second option, startup In addition to generally soliciting EPA will post a copy of this proposed
ends 4 hours after the EGU generates comments on all aspects of this action at https://www.epa.gov/
electricity that is sold or used for any proposed action, the EPA has identified stationary-sources-air-pollution/
other purpose (including on-site use), or several aspects of the proposal on which mercury-and-air-toxics-standards.
4 hours after the EGU makes useful comments are specifically requested. Following publication in the Federal
thermal energy (such as heat or steam) In selecting a proposed standard, as Register, the EPA will post the Federal
for industrial, commercial, heating, or discussed in detail below, the EPA Register version of the proposal and key
cooling purposes, whichever is earlier. considered the statutory direction and technical documents at this same
The EPA is proposing to remove the factors laid out by Congress in CAA website.
second option, which is currently being section 112. Separately, pursuant to E.O. A memorandum showing the rule
used by fewer than 10 EGUs as 12866, the EPA prepared an analysis of edits that would be necessary to
discussed in section V.D.1 of this the potential costs and benefits incorporate the changes proposed in
preamble. associated with this action. This this action to 40 CFR part 63, subpart
The EPA is not proposing to modify analysis, ‘‘Regulatory Impact Analysis UUUUU is available in the docket for
the HCl emission standard (nor the for the Proposed National Emission this action (Docket ID No. EPA–HQ–
alternative sulfur dioxide (SO2) Standards for Hazardous Air Pollutants: OAR–2018–0794). Following signature
emission standard), which serves as a Coal- and Oil-Fired Electric Utility by the EPA Administrator, the EPA also
surrogate for all acid gas HAP (HCl, HF, Steam Generating Units Review of the will post a copy of this document to
SeO2) for existing coal-fired EGUs. An Residual Risk and Technology Review’’ https://www.epa.gov/stationary-sources-
evaluation of recent compliance data for (Ref. EPA–452/R–23–002), is available air-pollution/mercury-and-air-toxics-
HCl and/or SO2 emissions revealed that in the docket, and is briefly summarized standards.
approximately two-thirds of coal-fired here and in section VI of this preamble. III. Background
EGUs operate at or below the alternative
SO2 emission standard of 2.0E–01 lb II. General Information A. What is the authority for this action?
SO2/MMBtu (SO2 may be used as an A. Does this action apply to me? 1. Statutory Authority
alternative surrogate for acid gas HAP at
coal-fired EGUs with operational flue The source category that is the subject The statutory authority for this action
gas desulfurization (FGD) systems and of this proposal is coal- and oil-fired is provided by sections 112 and 301 of
SO2 CEMS). Approximately one-third of EGUs regulated under 40 CFR part 63, the CAA, as amended (42 U.S.C. 7401 et
coal-fired EGUs have a SO2 emission subpart UUUUU. The North American seq.). Section 112 of the CAA
rate above the current SO2 standard, but Industry Classification System (NAICS) establishes a multi-stage regulatory
instead operate in compliance with the codes for the coal- and oil-fired EGU process to develop standards for
primary acid gas HAP limit for HCl of industry are 221112, 221122, and emissions of HAP from stationary
2.0E–03 lb HCl/MMBtu, with most 921150. This list of categories and sources. Generally, during the first stage
using an FGD system and/or by firing NAICS codes is not intended to be Congress directed the EPA to establish
coal with low chlorine content and high exhaustive, but rather provides a guide technology-based standards to ensure
alkalinity. The EPA did not identify any for readers regarding the entities that that all sources control pollution at the
new technologies or developments in this proposed action is likely to affect. level achieved by the best-performing
existing technologies that would The proposed standards, once sources, referred to as the maximum
achieve additional emission reductions. promulgated, will be directly applicable achievable control technology (MACT).
Based on this review, the EPA is not to the affected sources. Federal, state, After the first stage, Congress directed
proposing revisions to the acid gas HAP local, and tribal government entities that the EPA to review those standards
emission standards for coal-fired EGUs. own and/or operate EGUs subject to 40 periodically to determine whether they
The EPA is unaware of any new coal- CFR part 63, subpart UUUUU would be should be strengthened. Within 8 years
or oil-fired EGUs in development and affected by this proposed action. The after promulgation of the standards, the
has not projected any new coal- or oil- coal- and oil-fired EGU source category EPA must evaluate the MACT standards
fired EGUs in EPA modeling to support was added to the list of categories of to determine whether additional
various power sector-related major and area sources of HAP standards are needed to address any
remaining risk associated with HAP generally available control technologies emission regulations in the 1990 CAA
emissions. This second stage is or management practices (GACT Amendments under section 112(n)(1)(a)
commonly referred to as the ‘‘residual standards) in lieu of MACT standards.8 of the CAA that is not applicable to
risk review.’’ In addition, the CAA also For categories of major sources and other source categories. This provision
requires the EPA to review standards set any area source categories subject to required the EPA to conduct a study to
under CAA section 112 on an ongoing MACT standards, the next stage in evaluate the hazards to public health
basis no less than every 8 years and standard-setting focuses on identifying that are reasonably anticipated to occur
revise the standards as necessary taking and addressing any remaining (i.e., as a result of HAP emissions from EGUs,
into account any ‘‘developments in ‘‘residual’’) risk pursuant to CAA and to make a one-time finding of
practices, processes, and control section 112(f)(2). The residual risk whether to regulate EGUs under CAA
technologies.’’ This review is commonly review requires the EPA to update section 112 if the EPA found that doing
referred to as the ‘‘technology review,’’ standards if needed to provide an ample so was ‘‘appropriate and necessary.’’ 42
and is the subject of this proposal. The margin of safety to protect public health. U.S.C. 7412(n)(1)(A) (the ‘‘appropriate
discussion that follows identifies the Concurrent with that review, and then and necessary finding’’). Once this one-
most relevant statutory sections and at least every 8 years thereafter, CAA time finding was made, if the decision
briefly explains the contours of the section 112(d)(6) requires the EPA to was to regulate, Congress subjected
methodology used to implement these review standards promulgated under EGUs to the same standards and
statutory requirements. CAA section 112 and revise them ‘‘as procedures as other source categories.
In the first stage of the CAA section necessary (taking into account Id. (‘‘The Administrator shall regulate
112 standard-setting process, the EPA developments in practices, processes, electric utility steam generating units
promulgates technology-based standards and control technologies).’’ See Portland under this section’’ if he finds doing so
under CAA section 112(d) for categories Cement Ass’n v. EPA, 665 F.3d 177, 189 is ‘‘appropriate and necessary.’’); see
of sources identified as emitting one or (D.C. Cir. 2011) (‘‘Though EPA must also New Jersey v. EPA, 517 F.3d 574
more of the HAP listed in CAA section review and revise standards ‘no less (D.C. Cir. 2008) (establishing that, on the
112(b). Sources of HAP emissions are often than every eight years,’ 42 U.S.C. applicability of CAA section 112(c)(9)’s
either major sources or area sources, and 7412(d)(6), nothing prohibits EPA from delisting requirements, coal- and oil-
CAA section 112 establishes different reassessing its standards more often.’’). fired EGUs are treated similarly as other
requirements for major source standards In conducting this review, which we CAA section 112 regulated sources once
and area source standards. ‘‘Major call the ‘‘technology review,’’ the EPA is listed under CAA section 112(c)).
sources’’ are those that emit or have the not required to recalculate the MACT The EPA originally made the
potential to emit 10 tons per year (tpy) floors that were established in earlier appropriate and necessary finding in
or more of a single HAP or 25 tpy or rulemakings. Natural Resources Defense 2000. This was followed by a series of
more of any combination of HAP. All Council (NRDC) v. EPA, 529 F.3d 1077, affirmations and reversals of this
other sources are ‘‘area sources.’’ For 1084 (D.C. Cir. 2008); Association of finding, as well as a Supreme Court
major sources, CAA section 112(d)(2) Battery Recyclers, Inc. v. EPA, 716 F.3d decision that required the EPA to
provides that the technology-based 667 (D.C. Cir. 2013). The EPA may consider the costs of regulation in
NESHAP must reflect ‘‘the maximum consider cost in deciding whether to making this finding. See Michigan v.
degree of reduction in emissions of the revise the standards pursuant to CAA EPA, 576 U.S. 743 (2015). On February
[HAP] subject to this section (including section 112(d)(6). See e.g., Nat’l Ass’n 9, 2022, the EPA published a notice of
a prohibition on such emissions, where proposed rulemaking reaffirming that it
for Surface Finishing v. EPA, 795 F.3d
achievable) that the Administrator, remains appropriate and necessary to
1, 11 (D.C. Cir. 2015). The EPA is
taking into consideration the cost of regulate HAP, including Hg, from coal-
required to address regulatory gaps,
achieving such emission reduction, and and oil-fired EGUs after considering
such as missing MACT standards for
any non-air quality health and cost.9 The EPA’s consideration of costs
listed air toxics known to be emitted
environmental impacts and energy in its decision to reaffirm the
from the source category. Louisiana
requirements, determines is appropriate and necessary finding was
Environmental Action Network (LEAN)
achievable.’’ These standards are based on estimated and realized costs
v. EPA, 955 F.3d 1088 (D.C. Cir. 2020).
commonly referred to as MACT
In this action, the EPA is proposing to from the first stage of CAA section 112
standards. CAA section 112(d)(3) also
reconsider the 2020 Final Action’s risk regulation, i.e., establishing MACT-
establishes a minimum control level for
and technology review pursuant to the based standards and determining
MACT standards, known as the MACT
EPA’s inherent authority to reconsider whether additional ‘‘beyond-the-floor’’
‘‘floor.’’ 7 In certain instances, as
previous decisions and to revise, standards are needed to address
provided in CAA section 112(h), the
replace, or repeal a decision to the remaining risk.
EPA may set work practice standards in Consistent with Congress’s direction,
extent permitted by law and supported
lieu of numerical emission standards. after making the appropriate and
by a reasoned explanation. FCC v. Fox
The EPA must also consider control necessary finding, the EPA treated EGUs
Television Stations, Inc., 556 U.S. 502,
options that are more stringent than the like all other source categories. As
515 (2009); see also Motor Vehicle Mfrs.
floor. Standards more stringent than the required by CAA section 112(d)(2), the
Ass’n v. State Farm Mutual Auto. Ins.
floor are commonly referred to as EPA first set a floor based on the best
Co., 463 U.S. 29, 42 (1983).
‘‘beyond-the-floor’’ standards. For area
12 percent of performers, and then

sources, CAA section 112(d)(5) allows 2. EGU Regulation Under CAA Section conducted a beyond-the-floor analysis.
the EPA to set standards based on 112 That inquiry led to the current MATS,
7 Specifically, for existing sources, the MACT
Congress enacted a special provision established in 2012. As explained
‘‘floor’’ shall not be less stringent than the average concerning coal- and oil-fired EGU HAP above, the CAA then required the EPA,
emission reduction achieved by the best performing
12 percent of existing sources. For new sources 8 For categories of area sources subject to GACT 9 For further discussion on the history of the CAA
MACT shall not be less stringent than the emission standards, there is no requirement to address section 112(n)(1)(A) appropriate and necessary
control that is achieved in practice by the best residual risk, but, similar to the major source finding, please refer to the EPA’s February 9, 2022
controlled similar source. categories, the technology review is required. proposal (87 FR 7624).
within 8 years of promulgating the correct conflicts between preamble and set forth in the definitions for coal-fired
standards, to conduct the residual risk regulatory text and to clarify regulatory and oil-fired EGUs in 40 CFR 63.10042;
and technology reviews. Congress thus text. In addition, the electronic and any electric steam generating unit
contemplated that well after the EPA reporting requirements of the rule were combusting solid waste (i.e., a solid
determined the regulation of EGUs was amended on March 24, 2015 (80 FR waste incineration unit) subject to
appropriate and necessary and well after 15510), to allow for the electronic standards established under sections
the EPA set initial standards in submission of Portable Document 129 and 111 of the CAA.
accordance with the floor and beyond- Format (PDF) versions of certain reports For coal-fired EGUs, the rule
the-floor requirements in CAA section until April 16, 2017, while the EPA’s established standards to limit emissions
112(d)(2), that at least every 8 years Emissions Collection and Monitoring of Hg, acid gas HAP (e.g., HCl, HF), non-
thereafter on a continuing basis, the Plan System (ECMPS) is revised to Hg HAP metals (e.g., nickel, lead,
EPA would review and revise those accept all reporting that is required by chromium), and organic HAP (e.g.,
standards as necessary taking into the rule, and on April 6, 2017 (82 FR formaldehyde, dioxin/furan). Emission
account developments in practices, 16736), and on July 2, 2018 (83 FR standards for HCl serve as a surrogate
processes, and control technologies. The 30879), to extend the interim for the acid gas HAP, with an alternate
EPA has conducted over 100 technology submission of PDF versions of reports standard for SO2 that may be used as a
reviews and has regularly updated through June 30, 2018, and July 1, 2020,
emissions standards for HAP based surrogate for acid gas HAP for those
respectively. coal-fired EGUs with FGD systems and
upon the technology review. The MATS rule applies to coal- and SO2 CEMS installed and operational.
3. Executive Order 13990 oil-fired EGUs located at both major and Standards for fPM serve as a surrogate
area sources of HAP emissions. An for the non-Hg HAP metals, with
On January 20, 2021, President Biden existing affected source is the collection
signed Executive Order 13990, standards for total non-Hg HAP metals
of coal- or oil-fired EGUs in a and individual non-Hg HAP metals
‘‘Protecting Public Health and the subcategory within a single contiguous
Environment and Restoring Science to provided as alternative equivalent
area and under common control. A new standards. Work practice standards limit
Tackle the Climate Crisis.’’ The affected source is each coal- or oil-fired
Executive order, among other things, formation and emissions of organic
EGU for which construction or HAP.
instructs the EPA to review the 2020
reconstruction began after May 3, 2011.
Final Action titled, ‘‘National Emission For oil-fired EGUs, the rule
As previously stated in section II of this
Standards for Hazardous Air Pollutants: established standards to limit emissions
preamble, an EGU is a fossil fuel-fired
Coal- and Oil-Fired Electric Utility of HCl and HF, total HAP metals (e.g.,
combustion unit of more than 25 MW
Steam Generating Units— Hg, nickel, lead), and organic HAP (e.g.,
that serves a generator that produces
Reconsideration of Supplemental formaldehyde, dioxin/furan). Standards
electricity for sale. A unit that
Finding and Residual Risk and for fPM serve as a surrogate for total
cogenerates steam and electricity and
Technology Review’’ (85 FR 31286; May HAP metals, with standards for total
supplies more than one-third of its
22, 2020) and consider publishing a HAP metals and individual HAP metals
notice of proposed rulemaking potential electric output capacity and
more than 25 MW electric output to any provided as alternative equivalent
suspending, revising, or rescinding that standards. Work practice standards limit
action. utility power distribution system for
sale is also considered an EGU. The formation and emissions of organic
B. What is this source category and how MATS rule defines additional terms for HAP.
does the current NESHAP regulate its determining rule applicability, The MATS rule includes standards for
HAP emissions? including, but not limited to, definitions existing and new EGUs for seven
The NESHAP for the coal- and oil- for ‘‘coal-fired electric utility steam subcategories: two for coal-fired EGUs,
fired EGU source category (commonly generating unit,’’ ‘‘oil-fired electric one for IGCC EGUs, one for solid oil-
referred to as MATS) were initially utility steam generating unit,’’ and derived fuel-fired EGUs (i.e., petroleum
promulgated on February 16, 2012 (77 ‘‘fossil fuel-fired.’’ Certain types of coke-fired), and three for liquid oil-fired
FR 9304) (2012 MATS Final Rule), electric generating units are not subject EGUs. EGUs in six of the subcategories
under title 40 part 63, subpart UUUUU. to 40 CFR part 63, subpart UUUUU: any are subject to numeric emission limits
The MATS rule was amended on April unit designated as a major source for all the pollutants described above
19, 2012 (77 FR 23399), to correct stationary combustion turbine subject to except for organic HAP. Emissions of
typographical errors and certain subpart YYYY of 40 CFR part 63 and organic HAP are regulated by a work
preamble text that was inconsistent with any unit designated as an area source practice standard that requires periodic
regulatory text; on April 24, 2013 (78 FR stationary combustion turbine, other combustion process tune-ups. EGUs in
24073), to update certain emission than an IGCC unit; any EGU that is not the subcategory of limited-use liquid
limits and monitoring and testing a coal- or oil-fired EGU and that meets oil-fired EGUs with an annual capacity
requirements applicable to new sources; the definition of a natural gas-fired EGU factor of less than 8 percent of its
on November 19, 2014 (79 FR 68777), to in 40 CFR 63.10042; any EGU greater maximum or nameplate heat input are
revise definitions for startup and than 25 MW that has the capability of also subject to a work practice standard
shutdown and to finalize work practice combusting either coal or oil, but does consisting of periodic combustion
standards and certain monitoring and not meet the definition of a coal- or oil- process tune-ups but are not subject to
testing requirements applicable during fired EGU because it did not fire any numeric emission limits. Emission
periods of startup and shutdown; and sufficient coal or oil to satisfy the limits for existing EGUs are summarized
on April 6, 2016 (81 FR 20172), to average annual heat input requirement in Table 1.
TABLE 1—EMISSION LIMITS FOR EXISTING AFFECTED EGUS

Subcategory Pollutant Emission limit 1
Any coal-fired unit firing any rank of coal ............................... a. fPM .................................................... 3.0E–2 lb/MMBtu or 3.0E–1 lb/MWh.
OR OR
Total non-Hg HAP metals ..................... 5.0E–5 lb/MMBtu or 5.0E–1 lb/GWh.
OR OR
Individual HAP metals:
Antimony, Sb ......................................... 8.0E–1 lb/TBtu or 8.0E–3 lb/GWh.
Arsenic, As ............................................ 1.1 lb/TBtu or 2.0E–2 lb/GWh.
Beryllium, Be ......................................... 2.0E–1 lb/TBtu or 2.0E–3 lb/GWh.
Cadmium, Cd ......................................... 3.0E–1 lb/TBtu or 3.0E–3 lb/GWh.
Chromium, Cr ........................................ 2.8 lb/TBtu or 3.0E–2 lb/GWh.
Cobalt, Co .............................................. 8.0E–1 lb/TBtu or 8.0E–3 lb/GWh.
Lead, Pb ................................................ 1.2 lb/TBtu or 2.0E–2 lb/GWh.
Manganese, Mn ..................................... 4.0 lb/TBtu or 5.0E–2 lb/GWh.
Nickel, Ni ............................................... 3.5 lb/TBtu or 4.0E–2 lb/GWh.
Selenium, Se ......................................... 5.0 lb/TBtu or 6.0E–2 lb/GWh.
b. HCl ..................................................... 2.0E–3 lb/MMBtu or 2.0E–2 lb/MWh.
OR OR
SO2 2 ...................................................... 2.0E–1 lb/MMBtu or 1.5 lb/MWh.
Coal-fired unit low rank virgin coal ......................................... c. Hg ...................................................... 1.2 lb/TBtu or 1.3E–2 lb/GWh.
Coal-fired unit low rank virgin coal ......................................... c. Hg ...................................................... 4.0 lb/TBtu or 4.0E–2 lb/GWh.
IGCC unit ................................................................................ a. fPM .................................................... 4.0E–2 lb/MMBtu or 4.0E–1 lb/MWh.
OR OR
OR OR
Antimony, Sb ......................................... 1.4 lb/TBtu or 2.0E–2 lb/GWh.
Cobalt, Co .............................................. 1.2 lb/TBtu or 2.0E–2 lb/GWh.
Lead, Pb ................................................ 1.9E+2 lb/MMBtu or 1.8 lb/MWh.
Selenium, Se ......................................... 2.2E+1 lb/TBtu or 3.0E–1 lb/GWh.
c. Hg ...................................................... 2.5 lb/TBtu or 3.0E–2 lb/GWh.
Liquid oil-fired unit—continental (excluding limited-use liquid a. fPM .................................................... 3.0E–2 lb/MMBtu or 3.0E–1 lb/MWh.
oil-fired subcategory units).
OR OR
Total HAP metals .................................. 8.0E–4 lb/MMBtu or 8.0E–3 lb/MWh.
OR OR
Antimony, Sb ......................................... 1.3E+1 lb/TBtu or 2.0E–1 lb/GWh.
Cobalt, Co .............................................. 2.1E+1 lb/TBtu or 3.0E–1 lb/GWh.
Manganese, Mn ..................................... 2.2E+1 lb/TBtu or 3.0E–1 lb/GWh.
Nickel, Ni ............................................... 1.1E+2 lb/TBtu or 1.1 lb/GWh.
Hg .......................................................... 2.0E–1 lb/TBtu or 2.0E–3 lb/GWh.
c. HF ...................................................... 4.0E–4 lb/MMBtu or 4.0E–3 lb/MWh.
Liquid oil-fired unit—non-continental (excluding limited-use a. fPM .................................................... 3.0E–2 lb/MMBtu or 3.0E–1 lb/MWh.
liquid oil-fired subcategory units).
OR OR
Total HAP metals .................................. 6.0E–4 lb/MMBtu or 7.0E–3 lb/MWh.
OR OR
Antimony, Sb ......................................... 2.2 lb/TBtu or 2.0E–2 lb/GWh.

Chromium, Cr ........................................ 3.1E+1 lb/TBtu or 3.0E–1 lb/GWh.
Cobalt, Co .............................................. 1.1E+2 lb/TBtu or 1.4 lb/GWh.
Manganese, Mn ..................................... 2.0E+1 lb/TBtu or 3.0E–1 lb/GWh.
Nickel, Ni ............................................... 4.7E+2 lb/TBtu or 4.1 lb/GWh.
TABLE 1—EMISSION LIMITS FOR EXISTING AFFECTED EGUS—Continued

Subcategory Pollutant Emission limit 1
Hg .......................................................... 4.0E–2 lb/TBtu or 4.0E–4 lb/GWh.

c. HF ...................................................... 6.0E–5 lb/MMBtu or 5.0E–4 lb/MWh.
Solid oil-derived fuel-fired unit ................................................ a. fPM .................................................... 8.0E–3 lb/MMBtu or 9.0E–2 lb/MWh.
OR OR
OR OR
Individual HAP metals
Antimony, Sb ......................................... 8.0E–1 lb/TBtu or 7.0E–3 lb/GWh.
Arsenic, As ............................................ 3.0E–1 lb/TBtu or 5.0E–3 lb/GWh.
Chromium, Cr ........................................ 8.0E–1 lb/TBtu or 2.0E–2 lb/GWh.
Cobalt, Co .............................................. 1.1 lb/TBtu or 2.0E–2 lb/GWh.
Lead, Pb ................................................ 8.0E–1 lb/TBtu or 2.0E–2 lb/GWh.
Selenium, Se ......................................... 1.2 lb/TBtu 2.0E–2 lb/GWh.
OR OR
SO2 2 ...................................................... 3.0E–1 lb/MMBtu or 2.0 lb/MWh.
c. Hg ...................................................... 2.0E–1 lb/TBtu or 2.0E–3 lb/GWh.
1 Units of emission limits:
lb/MMBtu = pounds pollutant per million British thermal units fuel input;
lb/TBtu = pounds pollutant per trillion British thermal units fuel input;
lb/MWh = pounds pollutant per megawatt-hour electric output (gross); and
lb/GWh = pounds pollutant per gigawatt-hour electric output (gross).
2 Alternate SO limit may be used if the EGU has some form of FGD system and SO CEMS installed.
2 2
C. What data collection activities were survey and responses are available in power sector modeling. Other sources
conducted to support this proposed the docket for this action. used include the U.S. Department of
action? Energy’s EIA list of fuel consumption
D. What other relevant background
reported for 2021 under Form EIA–
On February 9, 2022, the EPA information and data are available?
923 13 and emissions test data collected
published a notice of proposed The EPA used multiple sources of from an ICR in 2010 (2010 ICR) when
rulemaking reaffirming that it remains information to support this proposed promulgating the 2011 Proposal.14
appropriate and necessary to regulate action. A comprehensive list of facilities In conducting the technology review,
coal- and oil-fired EGUs under CAA and EGUs that are subject to the MATS the EPA examined information
section 112 after considering the cost of rule was compiled primarily using the submitted to the EPA’s ECMPS as well
regulation. In that same action, the EPA list from the 2020 Final Action and as information that supports previous 40
solicitated information on the cost and publicly available information reported CFR part 63, subpart UUUUU actions to
performance of new or improved to the EPA and information contained in identify technologies currently being
technologies that control HAP the EPA’s National Electric Energy Data used by affected EGUs and to determine
emissions, on improved methods of System (NEEDS) database.10 Affected if there have been developments in
operation, and on risk-related sources are required to use the 40 CFR practices, processes, or control
information to further inform the EPA’s part 75-based ECMPS 11 for reporting technologies. In addition to the ECMPS
assessment of the MATS RTR. emissions and related data either data, we reviewed regulatory actions for
Generally, commenters were unaware of directly for EGUs that use Hg, HCl, HF, similar combustion sources and
new technologies, but indicated that or SO2 CEMS or Hg sorbent traps for conducted a review of literature
current technologies are more widely compliance purposes or indirectly as published by industry organizations,
used, more effective, and cheaper than PDF files for EGUs that use performance technical journals, and government
at the time of the adoption of MATS. test results, PM continuous parameter organizations.
Specific data or information used to monitoring system (CPMS) data, or PM
support this action are discussed in E. How does the EPA perform the
CEMS for compliance purposes. Directly technology review?
more detail in section V of this submitted data are maintained in
preamble. ECMPS; indirectly submitted data are Our technology review primarily
The EPA also issued a limited request maintained in Web Factor Information focuses on the identification and
evaluation of developments in practices,
for information pursuant to section 114 Retrieval System (WebFIRE).12 The

of the CAA to obtain information related NEEDS database contains generation processes, and control technologies that
to HAP emissions from coal- and oil- unit information used in the EPA’s have occurred since the MACT
fired EGUs to inform the technology standards were promulgated. Where we
review under CAA section 112(d)(6). 10 See https://www.epa.gov/airmarkets/power- identify such developments, we analyze
sector-modeling-platform-v515.
Specifically, the EPA collected 11 See https://ampd.epa.gov/ampd/. 13 See https://www.eia.gov/electricity/data/
information and data related to Hg 12 See https://cfpub.epa.gov/webfire; https:// eia923/.
emissions and control technologies for www.epa.gov/electronic-reporting-air-emissions/ 14 See https://www3.epa.gov/airtoxics/utility/
lignite-fired EGUs. The CAA section 114 webfire. utilitypg.html.
the technical feasibility, estimated costs, broadly applied to the industry and that under CAA sections 112(d)(2) and (3),
energy implications, non-air was not identified or considered during or, in specific circumstances, CAA
environmental impacts, and potential development of the original MACT sections 112(d)(4) or (h). Louisiana
emissions reductions of more stringent standards; and Environmental Action Network (LEAN)
standards, to ensure that the MACT • Any significant changes in the cost v. EPA, 955 F.3d 1088 (D.C. Cir. 2020).
standards continue to fulfill Congress’s (including cost effectiveness) of See sections III.C and III.D of this
direction to require the maximum applying controls (including controls preamble for information on the specific
degree of reduction of HAP taking into the EPA considered during the data sources that were reviewed as part
account the statutory factors. This development of the original MACT of the technology review.
analysis informs our decision of standards).
whether it is ‘‘necessary’’ to revise the • Any operational changes or other IV. Review of 2020 Residual Risk and
emissions standards. In addition, we factors that were not considered during Technology Review
typically consider the appropriateness the development of the original MACT
of applying controls to new sources standards. A. Summary of the 2020 Residual Risk
versus retrofitting existing sources. For In addition to reviewing the practices, Review
this exercise, we consider any of the processes, and control technologies that
Pursuant to CAA section 112(f)(2), the
following to be a ‘‘development’’: were considered at the time we
EPA conducted a residual risk review
• Any add-on control technology or originally developed (or last updated)
the NESHAP, we review a variety of (2020 Residual Risk Review) and
other equipment that was not identified
data sources in our investigation of presented the results of this review,
and considered during development of
potential practices, processes, or along with our decisions regarding risk
the original MACT standards;
• Any improvements in add-on controls to consider. We also review the acceptability, ample margin of safety,
control technology or other equipment NESHAP and the available data to and adverse environmental effects, in
(that were identified and considered determine if there are any unregulated the 2020 Final Action. The results of the
during development of the original emissions of HAP within the source risk assessment are presented briefly in
MACT standards) that could result in category and evaluate this data for use Table 2, and in more detail in the
additional emission reductions; 15 in developing new emission standards. document titled Residual Risk
• Any work practice or operational When reviewing MACT standards, the Assessment for the Coal- and Oil-Fired
procedure that was not identified or EPA is required to address regulatory EGU Source Category in Support of the
considered during development of the gaps, such as missing standards for 2020 Risk and Technology Review Final
original MACT standards; listed air toxics known to be emitted Rule (risk document for the final rule),
• Any process change or pollution from the source category, and any new available in the docket (Docket ID No.
prevention alternative that could be MACT standards must be established EPA–HQ–OAR–2018–0794–4553).
TABLE 2—COAL- AND OIL-FIRED EGU INHALATION RISK ASSESSMENT RESULTS IN THE 2020 FINAL ACTION
[85 FR 31286; May 22, 2020]
Maximum individual cancer risk Population at increased risk of Annual cancer incidence Maximum chronic noncancer Maximum
(in 1 million) 2 cancer ≥1-in-1 million (cases per year) TOSHI 3 screening
acute non-
Number Based on . . . Based on . . . Based on . . . Based on . . . cancer HQ 4
of
facilities 1 Actual Allowable Actual Allowable Actual Allowable Actual Allowable Based on
emissions emissions emissions emissions emissions emissions emissions emissions actual emis-
level level level level level level level level sions level
332 ........ 9 10 193,000 636,000 0.04 0.1 0.2 0.4 HQREL = 0.09
(arsenic)
1 Number of facilities evaluated in the risk analysis. At the time of the risk analysis there were an estimated 323 facilities in the Coal- and Oil-Fired EGU source cat-
egory; however, one facility is located in Guam, which was beyond the geographic range of the model used to estimate risks. Therefore, the Guam facility was not
modeled and the emissions for that facility were not included in the assessment.
2 Maximum individual excess lifetime cancer risk due to HAP emissions from the source category.
3 Maximum target organ-specific hazard index (TOSHI). The target organ systems with the highest TOSHI for the source category are respiratory and
immunological.
4 The maximum estimated acute exposure concentration was divided by available short-term threshold values to develop an array of hazard quotient (HQ) values.
HQ values shown use the lowest available acute threshold value, which in most cases is the reference exposure level (REL). When an HQ exceeds 1, we also show
the HQ using the next lowest available acute dose-response value.
1. Chronic Inhalation Risk Assessment the risk. The total estimated cancer was driven by emissions of nickel and
Results incidence from this source category was cobalt from oil-fired EGUs. No one was
0.04 excess cancer cases per year, or one exposed to TOSHI levels above 1 based
The results of the chronic inhalation excess case in every 25 years. on actual emissions from sources
cancer risk assessment based on actual Approximately 193,000 people were regulated under this source category.
emissions, as shown in Table 2 of this estimated to have cancer risks at or The EPA also evaluated the cancer
preamble, indicated that the estimated
above 1-in-1 million from HAP emitted risk at the maximum emissions allowed
maximum individual lifetime cancer from the facilities in this source by the MACT standard (i.e., ‘‘allowable
risk (cancer MIR) was 9-in-1 million, category.16 The estimated maximum emissions’’). As shown in Table 2 of this
with nickel emissions from certain oil- chronic noncancer TOSHI for the source preamble, based on allowable
fired EGUs as the major contributor to category was 0.2 (respiratory), which emissions, the estimated cancer MIR
15 This may include getting new or better technology performs better than anticipated during of them were facilities with oil-fired EGUs located
information about the performance of an add-on or development of the rule). in Puerto Rico.
existing control technology (e.g., emissions data 16 There were four facilities in the source category
from affected sources showing an add-on control with cancer risk at or above 1-in-1 million, and all
was 10-in-1 million, and, as before, bound Tier 1 noncancer screening value potential for Hg hazards through fish
nickel emissions from oil-fired EGUs of 1, which corresponds to a HQ of 1 For consumption for the source category
were the major contributor to the risk. facilities that exceeded a Tier 1 based upon an upper-end fish ingestion
The total estimated cancer incidence multipathway screening value of 1, we rate of 373 grams/day.
from this source category, considering used additional facility site-specific In evaluating the potential
allowable emissions, was 0.1 excess information to perform a refined multipathway risk from emissions of
cancer cases per year, or one excess case screening assessment through Tiers 2 lead compounds, rather than developing
in every 10 years. Based on allowable and 3, as necessary, to determine the a screening threshold emission rate, we
emissions, approximately 636,000 maximum chronic cancer and compared maximum estimated chronic
people were estimated to have cancer noncancer impacts for the source inhalation exposure concentrations to
risks at or above 1-in-1 million from category. For cancer, the highest Tier 2 the level of the current National
HAP emitted from the facilities in this screening value for the gardener Ambient Air Quality Standards
source category. The estimated scenario (rural) was 200 driven by (NAAQS) for lead (0.15 micrograms per
maximum chronic noncancer TOSHI for arsenic emissions. This screening value cubic meter). Values below the level of
the source category was 0.4 (respiratory) was reduced to 50 after accounting for the primary (health-based) lead NAAQS
based on allowable emissions, driven by plume rise in our Tier 3 screen. Because were considered to have a low potential
emissions of nickel and cobalt from oil- this screening value was much lower for multipathway risk. We did not
fired EGUs. No one was exposed to than 100-in-1 million, and because we estimate any exceedances of the lead
TOSHI levels above 1 based on expected the actual risk from a site- NAAQS in this source category, the
allowable emissions. specific assessment to further lower the maximum predicted Pb screen
2. Screening Level Acute Risk Tier 2 screening value by a factor of 50, concentration over a 3-month period for
Assessment Results we decided not to perform a site- this source category was equal to 0.005
specific assessment for cancer. For micrograms per cubic meter,
Because of the conservative nature of noncancer, the highest Tier 2 screening significantly below the Pb NAAQS.
the acute inhalation screening value was 30 (for Hg) for the fisher
assessment and the variable nature of scenario, with four facilities having 4. Environmental Risk Screening Results
emissions and potential exposures, screening values greater than 20. These An environmental risk screening
acute impacts are screened on an screening values were reduced to 9 or assessment for the coal- and oil-fired
individual pollutant basis, not using the lower after the plume rise stage of Tier EGU source category was conducted for
TOSHI approach. Table 2 of this
3. the following pollutants: arsenic,
preamble provides the worst-case acute
HQ (based on the REL) of 0.09, driven Because the final stage of Tier 3 (time- cadmium, dioxins/furans, HCl, HF, lead,
by emissions of arsenic. There were no series) was unlikely to reduce the Hg (methylmercury and mercuric
facilities that have acute HQs (based on highest Hg screening values to 1, we chloride), and POMs. In the Tier 1
the REL or any other reference values) conducted a site-specific multipathway screening analysis for PB–HAP (other
greater than 1. For more detailed acute assessment of Hg emissions for this than lead, which was evaluated
risk results, refer to the risk document source category. Analysis of the differently), POM emissions had no
available in the docket (Docket ID No. facilities with the highest Tier 2 and exceedances of any of the ecological
EPA–HQ–OAR–2018–0794–4553). Tier 3 screening values helped identify benchmarks evaluated. Arsenic and
the location for the site-specific dioxin/furan emissions had Tier 1
3. Multipathway Risk Screening and assessment and the facilities to model exceedances for surface soil
Site-Specific Assessment Results with TRIM.FaTE. The assessment benchmarks. Cadmium and
Potential multipathway health risks considered the effect that multiple methylmercury emissions had Tier 1
under a fisher and gardener scenario facilities within the source category may exceedances for surface soil and fish
were evaluated using a three-tier have on common lakes. The three benchmarks. Divalent Hg emissions had
screening assessment of the HAP known facilities selected were located near Tier 1 exceedances for sediment and
to be persistent and bio-accumulative in Underwood, North Dakota. All three surface soil benchmarks.
the environment (PB–HAP) emitted by facilities had Tier 2 screening values A Tier 2 screening analysis was
facilities in the coal- and oil-fired EGU greater than or equal to 20. Two of the performed for arsenic, cadmium,
source category. This evaluation facilities were near each other (16 dioxins/furans, divalent Hg, and
resulted in a site-specific assessment of kilometers (km) apart). The third facility methylmercury emissions. In the Tier 2
Hg using the EPA’s Total Risk Integrated was more distant, about 20 to 30 km screening analysis, arsenic, cadmium,
Methodology.Fate, Transport, and from the other facilities, but it was and dioxin/furan emissions had no
Ecological Exposure (TRIM.FaTE) model included in the analysis because it is exceedances of any of the ecological
for one location (three facilities located within the 50-km modeling domain of benchmarks evaluated. Divalent Hg
in North Dakota) as further described the other facilities and because it had an emissions from two facilities exceeded
below. Of the 322 MATS-affected elevated Tier 2 screening value. We the Tier 2 screen for a sediment
facilities modeled, 307 facilities had expected that the exposure scenarios we threshold level benchmark by a
reported emissions of carcinogenic PB– assessed for these facilities are among maximum screening value of 2.
HAP (arsenic, dioxins, and polycyclic the highest, if not the highest, that might Methylmercury emissions from the
organic matter (POM)) that exceeded a be encountered for other facilities in same two facilities exceeded the Tier 2
Tier 1 cancer screening value of 1, this source category based upon their Hg screen for a fish (avian/piscivores) no-
which corresponds to an upper bound emissions and their respective Tier 2 observed-adverse-effect-level (NOAEL)
maximum excess lifetime cancer risk screening values and aggregate impacts (merganser) benchmark by a maximum
that may be greater than 1-in-1 million. to common lakes. The refined site- screening value of 2. A Tier 3 screening
This source category also had 235 specific multipathway assessment assessment was performed to verify the
facilities reporting emissions of non- estimated an HQ of 0.06 for Hg for the existence of the lake associated with
carcinogenic PB–HAP (lead, Hg, and three facilities assessed. We believed the these screening values, and it was found
cadmium) that exceeded an upper assessment represented the highest to be located on-site and is a man-made
industrial pond, and, therefore, was emissions from the source category Based on the results of our
removed from the assessment. because there was not significant environmental risk screening
Methylmercury emissions from two collocation of other sources with EGUs. assessment, we also determined in 2020
facilities exceeded the Tier 2 screen for that more stringent standards were not
a surface soil NOAEL for avian ground 6. Decisions Regarding Risk
necessary to prevent an adverse
insectivores (woodcock) benchmark by a Acceptability, Ample Margin of Safety, environmental effect.
maximum screening value of 2. Other and Adverse Environmental Effect
surface soil benchmarks for B. Summary of the 2020 Technology
In determining whether residual risks Review
methylmercury, such as the NOAEL for are acceptable for this source category in
mammalian insectivores and the accordance with CAA section 112, the Pursuant to CAA section 112(d)(6),
threshold level for the invertebrate EPA considered all available health the EPA conducted a technology review
community, were not exceeded. Given information and risk estimation (2020 Technology Review) in the 2020
the low Tier 2 maximum screening uncertainty. The results of the risk Final Action, which focused on
value of 2 for methylmercury, and the analysis indicated that both the actual identifying and evaluating
fact that only the most protective and allowable inhalation cancer risks to developments in practices, processes,
benchmark was exceeded, a Tier 3 the individual most exposed were below and control technologies for the
environmental risk screen was not 100-in-1 million, which is the emission sources in the source category
conducted for methylmercury. presumptive limit of acceptability. Also, that occurred since the MATS rule was
For lead, we did not estimate any the highest chronic noncancer TOSHI promulgated. Control technologies
exceedances of the secondary lead and the highest acute noncancer HQ typically used to minimize emissions of
NAAQS. For HCl and HF, the average were below 1, indicating low likelihood pollutants that have numeric emission
modeled concentration around each of adverse noncancer effects from limits under the MATS rule include
facility (i.e., the average concentration inhalation exposures. There were also electrostatic precipitators (ESPs) and
of all off-site data points in the fabric filters (FFs) for control of non-Hg
low risks associated with ingestion,
modeling domain) did not exceed any HAP metals; wet scrubbers and dry
with the highest cancer risk being less
ecological benchmark. In addition, each scrubbers for control of acid gases (SO2,
than 50-in-1 million based on a
individual modeled concentration of HCl, and HF); and activated carbon
conservative screening assessment, and
HCl and HF (i.e., each off-site data point injection (ACI) for control of Hg. The
the highest noncancer hazard being less
in the modeling domain) was below the EPA determined that existing air
than 1 based on a site-specific
ecological benchmarks for all facilities. pollution control technologies that were
Based on the results of the multipathway assessment. Considering
in use were well-established and
environmental risk screening analysis, this information, the EPA determined in
provided the capture efficiencies
we did not expect an adverse 2020 that the residual risks of HAP
necessary for compliance with the
environmental effect as a result of HAP emissions from the coal- and oil-fired
MATS emission limits. Based on the
emissions from the coal- and oil-fired EGU source category were acceptable. effectiveness and proven reliability of
EGU source category. We then considered whether the these control technologies, and the
current standards provided an ample relatively short period of time since the
5. Facility-Wide Risk Results margin of safety to protect public health promulgation of the MATS rule, the
An assessment of risk from facility- and whether more stringent standards EPA did not identify any developments
wide emissions was performed to were necessary to prevent an adverse in practices, processes, or control
provide context for the source category environmental effect by taking into technologies, nor any new technologies
risks. Based on facility-wide emissions consideration costs, energy, safety, and or practices, for the control of non-Hg
estimates developed using the same other relevant factors. In determining HAP metals, acid gas HAP, or Hg.
estimates of actual emissions for whether the standards provided an However, in the 2020 Technology
emissions sources in the source ample margin of safety to protect public Review, the EPA did not consider
category, and emissions data from the health, we examined the same risk developments in the cost and
2014 National Emissions Inventory factors that we investigated for our effectiveness of these proven
(NEI) (version 2) for the sources outside acceptability determination and we also technologies, nor did the EPA evaluate
the source category, the estimated considered the costs, technological the current performance of emission
cancer MIR was 9-in-1 million, and feasibility, and other relevant factors reduction control equipment and
nickel emissions from oil-fired EGUs related to emissions control options that strategies at existing MATS-affected
were the major contributor to the risk. might reduce risk associated with EGUs, to determine whether revising the
The total estimated cancer incidence emissions from the source category. In standards was warranted. Organic HAP,
based on facility-wide emissions was our analysis, we considered the results including emissions of dioxins and
0.04 excess cancer cases per year, or one of the technology review, risk furans, are regulated by a work practice
excess case in every 25 years. assessment, and other aspects of our standard that requires periodic burner
Approximately 203,000 people were MACT rule review to determine tune-ups to ensure good combustion.
estimated to have cancer risks at or whether there were any cost-effective The EPA found that this work practice
above 1-in-1 million from HAP emitted controls or other measures that would continued to be a practical approach to
from all sources at the facilities in this reduce emissions further to provide an ensuring that combustion equipment
source category. The estimated ample margin of safety. The risk was maintained and optimized to run to
maximum chronic noncancer TOSHI analysis indicated that the risks from reduce emissions of organic HAP and
posed by facility-wide emissions was the source category are low for both continued to be more effective than
0.2 (respiratory), driven by emissions of cancer and noncancer health effects. establishing a numeric standard that
nickel and cobalt from oil-fired EGUs. Thus, we determined in 2020 that the cannot reliably be measured or
No one was exposed to TOSHI levels current MATS requirements provided monitored. Based on the effectiveness
above 1 based on facility-wide an ample margin of safety to protect and proven reliability of the work
emissions. These results were very public health in accordance with CAA practice standard, and the relatively
similar to those based on actual section 112. short amount of time since the
promulgation of the MATS rule, the described in Docket ID No. EPA–HQ– the EPA finds it is necessary taking into
EPA did not identify any developments OAR–2018–0794–0014 and account developments in practices,
in work practices nor any new work consideration of comments received in processes, and control technologies. The
practices or operational procedures for response to the 2022 Proposal. The EPA EPA also acknowledges that it received
this source category regarding the did not receive any new information in a petition for reconsideration from
additional control of organic HAP. response to the 2022 Proposal that environmental organizations that, in
After conducting the 2020 Technology would affect the EPA’s 2020 residual relevant part, sought the EPA’s
Review, the EPA did not identify risk analysis or the decisions emanating reconsideration of certain aspects of the
developments in practices, processes, or from that analysis. In reviewing the 2020 Residual Risk Review, which the
control technologies and, thus, did not 2020 residual risk analysis, the EPA has EPA continues to review and will
propose changes to emission standards determined that the risk analysis was a respond to in a separate action.18
or other requirements. More information rigorous and robust analytical review
concerning that technology review is in B. Review of the 2020 Technology
using approaches and methodologies
the memorandum titled Technology Review
that are consistent with those that have
Review for the Coal- and Oil-Fired EGU been utilized in residual risk analyses The EPA’s review of the 2020
Source Category, available in the docket and reviews for other industrial sectors. Technology Review included evaluating
(Docket ID No. EPA–HQ–OAR–2018– In addition, the results of the 2020 the technology review described in
0794–0015), and in the February 7, residual risk assessment, as summarized Docket ID No. EPA–HQ–OAR–2018–
2019, proposed rule. 84 FR 2700. On in section IV.A of this preamble, 0794–0015 and comments related to
May 20, 2020, the EPA finalized the first indicated low residual risk from the potential practices, processes, or
technology review required by CAA coal- and oil-fired EGU source category. controls received as part of the 2022
section 112(d)(6) for the coal- and oil- For these reasons, we are not proposing Proposal. The review also focused on
fired EGU source category regulated any revisions to the 2020 Residual Risk the identification and evaluation of any
under MATS. Based on the results of Review. Although we are not reopening developments in practices, processes,
that technology review, the EPA found the 2020 determination of whether and control technologies that have
that no revisions to MATS were residual risks would alone be sufficient occurred since finalization of the MATS
warranted. See 85 FR 31314 (May 22, under the CAA to necessitate new rule in 2012 and since publishing the
2020). standards, the EPA acknowledges that 2020 Technology Review. As explained
the revised standards being proposed in detail herein, based on this
V. Analytical Results and Proposed information, the EPA now concludes
Decisions under this technology review, as
explored below, will likely reduce HAP that developments in the costs and
As described in section IV, the EPA exposures to affected populations. In effectiveness of control technologies and
conducted a residual risk review under recognition of the hazardous nature of the related fact that emissions
CAA section 112(f) and presented these HAP, Congress intentionally performance still varies significantly,
results of the review in the 2020 Final created a two-pronged structure for warrant revising certain MACT
Action. Executive Order 13990, updating standards for toxic air standards.
‘‘Protecting Public Health and the pollutants that requires the EPA to Technology reviews can, and often
Environment and Restoring Science to continue assessing opportunities to do, include obtaining better information
Tackle the Climate Crisis’’ required the strengthen the standards under CAA about the performance of a control
EPA to review the 2020 Final Action section 112(d)(6) even after residual technology (e.g., emissions data from
and consider publishing a notice of risks have been addressed under CAA affected sources) showing that an add-
proposed rulemaking suspending, section 112(f)(2).17 Under this structure, on technology that was identified and
revising, or rescinding the 2020 Final considered during the development of
recognizing the value of reducing any
Action. As part of this effort, the EPA the original MACT standards works
exposure to HAP where feasible, the
solicited information to inform a review better (e.g., gets more emissions
EPA is obligated to update standards
of the MATS RTR in the 2022 Proposal reductions or costs less) than
where either the EPA finds it is
affirming it is appropriate and necessary anticipated. In fact, considering data on
necessary to provide an ample margin of
to regulate coal- and oil-fired EGUs outperforming sources and cost and
safety to protect public health or where
under CAA section 112. The EPA effectiveness of existing controls is well
summarizes the results of the review of 17 The EPA has long considered these two established. See, e.g., Coke Oven
the RTR and proposed decisions inquiries independent. See, e.g., Mineral Wool Batteries, 69 FR 48338, 48351 (August 9,
consequent of the review below and Production and Wool Fiberglass Manufacturing, 80 2014) (‘‘[A]lthough no new control
requests comment on specific FR 45280, 45292 (July 29, 2015) (explaining CAA technologies have been developed since
section 112(d)(6) and 112(f)(2) ‘‘standards rest on
considerations. In addition to generally independent statutory authorities and independent the original standards were
soliciting comments on all aspects of rationales.’’); see also Ass’n of Battery Recyclers, promulgated, our review of emissions
this proposed action, the EPA is Inc. v. EPA, 716 F.3d 667, 672 (D.C. Cir. 2013) (CAA data revealed that existing MACT track
requesting public comment on specific section 112(d)(6) ‘‘directs EPA to take into account batteries can achieve a level of control
developments in practices, processes, and control
issues as described below. In addition, technologies, . . . not risk reduction achieved by for door leaks and topside leaks more
the EPA is granting in part certain the additional controls.’’) (internal quotation stringent than that required by the 1993
petitions for reconsideration on the omitted). Indeed, the EPA has strengthened national emission standards . . .
Agency’s prior rulemakings, which are standards based upon its technology review while through diligent work practices to
finding residual risks acceptable numerous times.
discussed in further detail below. See, e.g., Site Remediation, 85 FR 41680 (July 10, identify and stop leaks.’’); Site
A. Review of the 2020 Residual Risk 2020); Organic Liquids Distribution, 85 FR 40740 Remediation, 85 FR 41680, 41690 (July
Review
(July 7, 2020); Ethylene Production, 85 FR 40386 10, 2020) (noting that commenters had
(July 6, 2020); Pulp Mills, 82 FR 47328 (Oct. 11, not identified developments like a
The EPA has reviewed the 2020 2017); Acrylic and Modacrylic Fibers Production,
79 FR 60898 (Oct. 8, 2014); Natural Gas Processing reduction in costs); Petroleum
Residual Risk Review as directed by Plants, 77 FR 49400 (Aug. 16, 2012); Wood
E.O. 13990. This included a review of Furniture Manufacturing Operations, 76 FR 72052 18 See Docket ID No. EPA–HQ–OAR–2018–0794–
the 2020 residual risk assessment (Nov. 21, 2011). 4565 at www.regulations.gov.
Refineries, 80 FR 75178, 75201 commenters noted that the large number of non-Hg metal HAP, but most
(December 1, 2015); Mineral Wood of EGUs that are outperforming the generally emit fPM that includes some
Production and Fiberglass current Hg and fPM standards would amount and combination of all the non-
Manufacturing, 80 FR 45280, 45284–85 support a decision to revise the Hg metal HAP. Lastly, the use of fPM as
(July 29, 2015); see also Nat’l Ass’n for standards (Docket ID No. EPA–HQ– a surrogate eliminates the cost of
Surface Finishing v. EPA, 795 F3d 1, OAR–2018–0794–4962). Specific performance testing to demonstrate
11–12 (D.C. Cir 2015). comments leading to our proposed compliance with numerous standards
For example, in the 2014 technology decisions are detailed below, and a for individual non-Hg metal HAP
review for Ferroalloys Production, the summary of this technology review is (Docket ID No. EPA–HQ–OAR–2009–
EPA found that PM emission levels provided in the memorandum ‘‘2023 0234). For these reasons, the EPA
were well below the MACT standards Technology Review for the Coal- and focused its review on the fPM emissions
established in the original 1999 Oil-Fired EGU Source Category,’’ which of coal-fired EGUs as a surrogate for
NESHAP. These findings can be found in Docket ID No. EPA– non-Hg metal HAP.
‘‘demonstrate[d] that the add-on HQ–OAR–2018–0794. Based on our In the 2020 Technology Review, the
emission control technology (venturi review of the 2020 Technology Review, EPA did not identify any developments
scrubber, positive pressure FF, negative the EPA is proposing to revise the in practices, processes, or control
pressure FF) used to control emissions current standards as discussed below. technologies for non-Hg metal HAP or
from the furnaces are quite effective in fPM. The assessment of implementation
reducing PM (used as a surrogate for C. What are the results and proposed and developments in non-Hg metal HAP
metal HAP) and that all of the facilities decisions based on our technology metal is summarized in the
have emissions well below the current review, and what is the rationale for memorandum, ‘‘Technology Review for
limits.’’ See 79 FR 60271 (October 6, those decisions? the Coal- and Oil-Fired EGU Source
2014). Therefore, the EPA determined This section summarizes the EPA’s Category,’’ which is included in Docket
that it was appropriate to revise the PM changes to the 2020 technology review ID No. EPA–HQ–OAR–2018–0794–
limits for furnaces. Similarly, in the and proposed decisions. Where the EPA 0015. The 2020 review simply presented
2017 technology review for Wool has identified developments in a list of PM control technologies used by
Fiberglass Manufacturing, the EPA practices, processes, or controls, we coal-fired EGUs in operation, finding
found that formaldehyde emissions had analyzed the technical feasibility, that the units primarily employ ESPs
decreased by approximately 95 percent estimated costs, energy implications, and FFs, and did not identify any new
since promulgation of the MACT and non-air environmental impacts, as control technologies to reduce non-Hg
Standards in the original 1999 NESHAP well as the potential emission metal HAP. That review did not
due to ‘‘(1) Improvements in control reductions associated with each consider or discuss the costs or
technology (e.g., improved bag development. In addition, we reviewed performance of already-installed
materials, replacement of older a variety of data sources in our controls nor discuss or analyze
baghouses) and (2) the use of investigation of developments in opportunities for improved
electrostatic precipitators,’’ as well as practices, processes, or controls. See performance. In the 2020 Technology
upgraded pollution prevention practices section III of this preamble for Review, the EPA concluded that ‘‘[t]he
(i.e., development and use non-phenol- information on the specific data sources PM air pollution control device
formaldehyde binders). See 82 FR 40975 that were reviewed as part of the technologies that are currently in use
(August 29, 2017). Although the EPA technology review. are well-established and provide the
declined to lower the formaldehyde 1. Filterable Particulate Matter (fPM) capture efficiencies necessary for
limit in this case, it was only because Emission Limit (as a Surrogate for Non- compliance with the subpart UUUUU
the source category had already Hg HAP Metals) [MATS] filterable PM limits.’’ In the
upgraded the technology (i.e., non- 2022 Proposal, the EPA solicited
phenol-formaldehyde binders), resulting As described in section III of this information on the cost and
in major sources becoming area sources preamble, EGUs in six subcategories are performance of new or improved control
that were no longer subject to the subject to numeric emission limits for technologies that control HAP emissions
NESHAP. each of the individual non-Hg metal and improved methods of operation.
As in those cases, here many HAP. Alternatively, certain affected In this review of the RTR, and
commenters provided data showing that EGUs can choose to demonstrate consistent with some past technology
control technologies are more widely compliance with an alternative total reviews, the EPA assessed the
used, more effective, and cheaper than non-Hg metal HAP emission limit. performance of the sources in the source
at the time EPA promulgated MATS. For Finally, affected EGUs can demonstrate category compared to current standards,
example, commenters explained that, compliance with an alternative fPM and the EPA accordingly expanded
due to the many options that are emission limit that serves as a surrogate upon the 2020 Final Action’s
available to control Hg emissions (e.g., for total non-Hg metal HAP. The EPA technology review to assess the fPM
control equipment, activated carbon, chose fPM as a surrogate for non-Hg emission performance of the fleet. This
reagents and sorbents, as well as fuel metal HAP in the original MATS review included evaluating the control
blending, non-carbon or improvements rulemaking because non-Hg metal HAP efficiency and costs of common control
to carbon-based solvents, wet and dry are predominantly a component of the systems used for fPM control, primarily
scrubber additives, oxidizing coal filterable fraction of total PM (which is ESPs and FFs, detailed in the
additives, and existing control comprised of a filterable fraction and a memorandum (Technical Memo), ‘‘2023
optimization) and a ‘‘robust industry of condensable fraction), and control of Technology Review for the Coal- and
technology suppliers that drive fPM results in co-reduction of non-Hg Oil-Fired EGU Source Category,’’ which
innovation through internal research metal HAP (with the exception of Se, is included in Docket ID No. EPA–HQ–
and development,’’ the costs of which may be present in the filterable OAR–2018–0794. As part of this effort,
compliance for end users has decreased fraction or in the condensable fraction the EPA reviewed more recent fPM
over time (Docket ID No. EPA–HQ– as the acid gas, SeO2). Additionally, not compliance data that was not available
OAR–2018–0794–4940). Similarly, all fuels emit the same type and amount during the 2020 Final Action. Although
our review of fPM compliance data for studies finding the actual costs of costs to improve controls to achieve
coal-fired EGUs indicated no new complying with air pollution additional fPM emission reductions
practices, processes, or control regulations are often substantially lower than assumed during promulgation of
technologies for non-Hg metal HAP, it than pre-compliance estimates assumed the original MATS fPM emission limit,
revealed two important developments in the 2012 MATS Final Rule. the EPA concluded that there were
that inform the EPA’s decision to Figure 1 shows that all coal-fired developments that warranted an
propose revisions to the standard. First, EGUs are reporting fPM emissions well examination of whether to revise the
it revealed that most existing coal-fired below the current MATS limit of 3.0E– standard.
EGUs are reporting fPM well below the 02 lb/MMBtu, and that 91 percent of To examine potential revisions, the
current fPM emission limit of 3.0E–02 EGUs are reporting fPM emissions at EPA used representative fPM emissions
lb/MMBtu. Information we received in levels lower than a third of the current as a surrogate for total non-Hg metal
response to the 2022 Proposal similarly limit. In fact, the average reported fPM HAP to evaluate three more stringent
noted that the fleet is reporting much rate of the EGUs assessed in Figure 1 is emission limits. The fPM emission
lower fPM rates than what is currently 4.8E–03 lb/MMBtu, which is 84 percent limits that were evaluated are (1) 1.5E–
allowed. Second, it revealed that the below the MATS current limit (the 02 lb/MMBtu, which is 50 percent of the
fleet is achieving these performance median is 4.0E–03 lb/MMBtu, or 87 current limit and the qualifying
levels at lower costs than assumed percent below the MATS current limit). emission rate for the LEE program (2)
during promulgation of the original The EPA evaluated the fPM emission 1.0E–02 lb/MMBtu, which is
MATS fPM emission limit. More performance of EGUs and binned them comparable to the MATS new source
specifically, one commenter presented by quartiles. The average fPM emission fPM emission limit; and (3) 6.0E–03 lb/
its fleetwide evaluation using data from rate reported by the best performing 25 MMBtu, which is the average fPM
100 coal units in the PJM percent was 1.4E–03 lb/MMBtu. Of the emission rate from the 2010 ICR.
Interconnection and in the Electric best performing 50 percent of EGUs Currently, 96 percent of existing coal-
Reliability Council of Texas (ERCOT) assessed, the average fPM emission rate fired capacity without known retirement
markets. The commenter’s analysis was 2.4E–03 lb/MMBtu and the average plans before the proposed compliance
suggested that only 42 EGUs would fPM rate reported by the best 75 percent period 20 already have demonstrated an
require additional capital or operating was 3.1E–03 lb/MMBtu. Of the best emission rate of 1.5E–02 lb/MMBtu or
costs to meet a more stringent fPM limit performing 95 percent, the average fPM lower, 91 percent of existing coal-fired
of 7.0E–03 lb/MMBtu, while 79 EGUs emission rate was 4.2E–03 lb/MMBtu. capacity have demonstrated an emission
would incur those costs to meet a limit Even the higher emitting units, with rate of 1.0E–02 lb/MMBtu or lower, and
of 3.75E–03 lb/MMBtu. The reported rates above the current fPM 72 percent of existing coal-fired capacity
commenter’s analysis suggested that LEE standard, are performing 30 percent have demonstrated an emission rate of
most units would incur costs in the to 43 percent below the current 6.0E–03 lb/MMBtu or lower. As
range of $0/kW to $75/kW (Docket ID standard. Even so, the handful of the mentioned above, the average fPM rate
No. EPA–HQ–OAR–2018–0794–5121). worst performing EGUs are reporting of the best performing 95 percent of
Other commenters pointed to an fPM at rates approximately three to four EGUs was 4.2E–03 lb/MMBtu, below the
independent report finding that units times the fleet average. most stringent option analyzed of 6.0E–
are doing ‘‘just enough’’ to satisfy the Because an evaluation of compliance 03 lb/MMBtu. The EPA evaluated
MATS limits and that EGUs can achieve data showed that a significant portion of reductions of the 10 individual non-Hg
fPM emission rates at or below 7.0E–03 coal-fired EGUs are performing well metal HAP, total non-Hg metal HAP,
lb/MMBtu with relatively low capital below the allowed emission limit and fPM and the associated costs for
cost upgrades to pollution control (Figure 1), and because the EPA each unit to achieve each of the three
systems.19 Commenters also cited obtained information indicating lower fPM emission limits listed above.
19 See https://www.andovertechnology.com/wp- 20 If the proposed revised emission limits are demonstrate compliance with the revised emission
content/uploads/2021/08/PM-and-Hg-Controls_ finalized, affected EGUs will have up to 3 years limits.
EP24AP23.000</GPH>
CAELP_20210819.pdf. after the effective date of the rule amendments to
Figure 1—fPM rate distribution for should finalize a more stringent emissions for each individual EGU.
affected coal-fired EGUs in the standard. The EPA’s decision to propose More specifically, according to the
continental U.S. in reference to the a standard of 1.0E–02 lb/MMBtu is commenter, the lower fPM emissions
three considered fPM limit (horizontal based on several factors. First, this level and thus lower UPL were attributed to:
dashed lines): 1.5E–02 lb/MMBtu, of control would ensure that the very (1) greater attention to fPM emissions
1.0E–02 lb/MMBtu, and 6.0E–03 lb/ worst performers bring their due to the monitoring and reporting
MMBtu. Percentages represent the performance level up to where the vast requirements of MATS; (2) efforts to
amount of existing capacity achieving majority of the fleet is performing. The restore ESPs and other equipment to
each of the limits. More information EPA notes that Figure 1 shows a ‘‘knee original designed performance levels;
available in the Technical Memo in the curve’’ that starts before 1.0E–02 (3) modest improvements to ESPs when
supporting this action. lb/MMBtu, with coal-fired EGUs above needed, such as addition of high
The EPA discussed the opportunity that rate emitting substantially more frequency transformer rectifier (TR) sets;
for improved performance of existing pollution than those below it. Bringing and (4) efforts to minimize the wear and
fPM control technologies in the 2012 this small number of sources (9 percent tear on filter bags and increased
MATS Final Rule. In the regulatory of coal-fired EGU capacity) to the attention to FF operation. Developments
impact analysis (RIA) supporting the performance of the rest of the fleet in the technology, including better
2012 MATS Final Rule, the EPA serves Congress’s mandate to the EPA to performance at lower costs, combined
estimated that 34 gigawatts (GW) of continually consider developments and with improved variability assumptions
coal-fired EGU capacity would perform to ensure that standards account for updated since promulgation of the 2012
ESP upgrades as part of their fPM developments ‘‘that create opportunities MATS Final Rule, presents an
emission limit compliance strategy.21 to do even better.’’ See LEAN, 955 F.3d opportunity to strengthen the MACT
EPA’s methodology was based on at 1093. As discussed above in section standard for fPM.
historic PM emission rates and reported V.B. of this document, the EPA has a Second, the EPA believes that a fPM
control efficiencies and is explained in number of times in the past updated its emission limit of 1.0E–02 lb/MMBtu
the IPM 4.10 Supplemental MACT standards to reflect appropriately takes into account the
Documentation for MATS.22 Depending developments where the majority of costs of control. The EPA evaluated the
on the incremental fPM reduction sources is vastly outperforming the costs to improve current PM control
needed to bring a unit into compliance, original MACT standards. systems and the cost to install better
units with existing ESPs for PM control According to comments received in performing PM controls (i.e., a new FF)
were assigned either a FF retrofit or one response to the solicitation in the 2022 to achieve a more stringent emission
of three tiered ESP upgrades to bring Proposal, since the MATS Final Rule limit. As noted above, data received
them into compliance. In response to was promulgated in 2012, since 2012 demonstrates that the costs
the solicitation in the 2022 Proposal, improvements to existing PM controls to of PM control upgrades are likely much
commenters provided detailed comply with the MATS fPM standard lower than the EPA estimated in 2012.
information on updated costs for similar were achieved at lower costs than had Table 3 summarizes the estimated cost-
upgrades for improved ESP been projected by the EPA. The effectiveness of the three emission
performance. Using that data and commenter also noted that industry limits evaluated for the existing fleet.
additional information from one of the installed far fewer FFs than the EPA For the purpose of estimating cost-
EPA’s engineering consultants, the EPA projected and that there were a smaller effectiveness, the analysis presented in
evaluated revised costs to upgrade number of ESP upgrades than projected. this table is based on the observed
existing PM controls. The cost The 2012 MATS Final Rule used the emissions rates of all existing coal-fired
effectiveness estimates presented in this Upper Predictive Limit (UPL) to EGUs except for those that have
section are based on an assumption that establish the fPM emission limit of announced plans to retire by the end of
eight units would need to upgrade 3.0E–02 lb/MMBtu for existing coal- 2028. Note that, unlike the cost and
existing ESPs to comply with a revised fired EGUs. The UPL considers the benefit projections presented in the RIA
fPM emission standard of 1.5E–02 lb/ average of the best performing EGUs, for this proposed rule, the estimates in
MMBtu, that 20 units would need to but also includes an allowance for this table do not account for any future
implement similar ESP upgrades to variation that is determined by a changes in the composition of the
comply with a revised fPM emission confidence level that the UPL will not operational coal-fired EGU fleet that are
standard of 1.0E–02 lb/MMBtu, and that be exceeded. A report 23 submitted to likely to occur by 2028 as a result of
65 units would need to install a new FF the EPA in response to the 2020 other factors affecting the power sector,
or modify an existing FF to meet a Proposal presented an updated UPL such as the Inflation Reduction Act
revised fPM emission limit of 6.0E–03 (using 2019 data compiled by Natural (IRA), future regulatory actions, or
lb/MMBtu. Resources Defense Council (NRDC) 24) changes in economic conditions. Of the
In this proposal, the EPA proposes to of 5.0E–03 lb/MMBtu, about one-sixth over 9 GW of coal-fired capacity that the
set an fPM emission limit of 1.0E–02 lb/ of the EPA’s 2011 estimate of 3.0E–02 EPA estimates would require control
MMBtu (0.010 lb/MMBtu) and seeks lb/MMBtu. The updated 5.0E–03 lb/ improvements to achieve the proposed
comment on whether its control MMBtu UPL value was attributed to fPM rate, less than 5 GW is projected to
technology effectiveness and cost updated fPM rates that were lower on be operational in 2028 (see section 3 of
assumptions are correct, and whether it average and reflected less variability in the RIA for this proposal).
21 Regulatory Impact Analysis for the Final 22 See Table 5–25 in Documentation Supplement 23 See https://www.andovertechnology.com/wp-
Mercury and Air Toxics Standards, available for EPA Base Case v.4.10_MATS—Updates for Final content/uploads/2021/08/PM-and-Hg-Controls_
https://www.epa.gov/sites/default/files/2015-11/ Mercury and Air Toxics Standards (MATS) Rule CAELP_20210819.pdf.
documents/matsriafinal.pdf and in the rulemaking available at https://www.epa.gov/sites/default/files/ 24 https://www.nrdc.org/resources/coal-fired-
docket. 2015-07/documents/suppdoc410mats.pdf and in power-plant-hazardous-air-pollution-emissions-

the rulemaking docket. and-pollution-control-data.
TABLE 3—SUMMARY OF COST EFFECTIVENESS ANALYSIS FOR THREE POTENTIAL fPM EMISSION LIMITS 1
Potential fPM emission limit
(lb/MMBtu)
1.5E–02 1.0E–02 6.0E–03
Affected Units (Capacity, GW) .................................................................................................... 8 (4.02) 20 (9.34) 65 (32.9)

Annual Cost ($M) ......................................................................................................................... 13.9–19.3 77.3–93.2 633
fPM Reductions (tons/year) ......................................................................................................... 463 2,074 6,163
Total non-Hg metal HAP Reductions (tons/year) ........................................................................ 1.41 6.34 24.7
Total non-Hg metal HAP Cost Effectiveness ($k/ton) ................................................................. 9,860–13,700 12,200–14,700 25,600
Total non-Hg metal HAP Cost Effectiveness—Allowable ($k/ton) .............................................. 35.4–49.1 197–238 1,610
1 Note that these values represent annual cost and projected emission reductions assuming the affected coal-fired EGUs operate consistent
with their operation in their lowest quarter (see Technical Memo accompanying this action for more information).
The cost estimates presented in this of the fleet is already reporting fPM its peak during this period) in 2007
table could be overestimated for a rates below 6.0E–03 lb/MMBtu, both the dollars. The annual control cost
number of reasons, and the EPA seeks total costs and the total fPM and non- estimates for this proposal based on the
comment on these cost and cost- Hg metal HAP reductions for the three cost-effectiveness analysis in Table 3
effectiveness estimates and how they potential emission limits are modest in constitute at most about 0.2 percent of
may change over time. Additionally, the the context of the total control costs and sector sales at their lowest over the 2000
information in Table 3 shows that coal- emissions of the coal fleet. The cost- to 2019 period. Making similar
fired EGUs have demonstrated an ability effectiveness estimates for EGUs comparisons of the estimated capital
to meet these limits with existing reporting fPM rates above 6.0E–03 lb/ and total compliance costs to historical
control technology. It is possible that MMBtu to achieve similar performance trends in sector-level capital and
some EGUs with the same or similar as the rest of the fleet range from production costs, respectively, would
technologies may be able to achieve a $9,860,000 to $25,600,000 per ton of yield similarly small values. Because
lower fPM rate at significantly lower non-Hg metal HAP for the three this cost-effectiveness evaluation only
cost than assumed here, and possibly potential emission limits. considers improved fPM control needed
without any additional capital For this proposal, the costs—either at a few units and not the entire fleet,
investments. Furthermore, since the the annual control cost estimates we also evaluated an alternative cost-
EGU-specific fPM emissions rate is presented above in Table 3 or the effectiveness approach that considers
calculated using the largest 1 percent of projected total annual system-wide allowable emissions, assuming emission
fPM rates for the quarter with the lowest compliance costs presented in Table 3– reductions achieved if all evaluated
emissions, some EGUs may readily 4 in the RIA—represent a very small EGUs emit the maximum allowable
achieve lower fPM rates with improved fraction of typical capital and total amount of fPM (i.e., at the current
operation. While such factors could expenditures for the power sector. In the standard of 3.0E–02 lb/MMBtu), and the
likely lower the overall cost estimates 2022 Proposal (reaffirming the associated costs for EGUs to comply
and improve cost-effectiveness, this appropriate and necessary finding), the with the three potential fPM standards.
table presents estimates based on the EPA evaluated the compliance costs that Using this approach, the EPA estimates
best information available to the EPA at were projected in the 2012 MATS rule the cost-effectiveness (based on
this time. relative to the typical annual revenues, allowable rather than actual emissions)
The EPA considers costs in various capital expenditures, and total (capital of control of non-Hg HAP metals to
ways, depending on the rule and and production) expenditures.26 range from $35,400/ton to $49,100/ton
affected sector. For example, the EPA (January 11, 2022); 80 FR 37381 (June for a 1.5E–02 lb/MMBtu emission limit,
has considered, in previous CAA 30, 2015). Using electricity sales data from $197,000/ton to $238,000/ton for a
section 112 rulemakings, cost- from the U.S. EIA, the analysis in the 1.0E–02 lb/MMBtu emission limit, and
effectiveness, the total capital costs of 2022 Proposal demonstrated that $1,610,000/ton for a 6.0E–03 lb/MMBtu
proposed measures, annual costs, and revenues from retail electricity sales emission limit.
costs compared to total revenues (e.g., increased from $276.2 billion in 2000 to The EPA strives to minimize the
cost to revenue ratios).25 Because much a peak of $356.6 billion in 2008 (an uncertainty and the costs associated
increase of about 29 percent during this with the measurements used to
25 See, e.g., Mercury Cell Chlor-Alkali Plants
period) and have slowly declined since demonstrate compliance with emission
Residual, 87 FR 27002, 27008 (May 6, 2022) limits. For fPM measurements, the EPA
(considered annual costs and average capital costs to a post-2011 low of $331.0 billion in
per facility in technology review and beyond-the- 2019 (a decrease of about 7 percent from believes that appropriate approaches to
floor analysis); Primary Copper Smelting, 87 FR minimizing both uncertainty and costs
1616, 1635 (proposed Jan. 11, 2022) (considered would include limiting sampling times
total annual costs and capital costs, annual costs, 14254 (March 18, 2015) (considered total annual
costs and capital costs, and average annual costs to 3 hours per run and maintaining the
and costs compared to total revenues in proposed
beyond-the-floor analysis); Phosphoric Acid and capital costs and annualized costs per facility random error contribution to the
Manufacturing and Phosphate Fertilizer Production in technology review); Chromium Electroplating, 77 tolerance given to PM CEMS—which is
Phosphate Fertilizer Production Plants and FR 58225, 58226 (Sept. 19, 2012) (considered total one component of uncertainty—
Phosphoric Acid Manufacturing Plants, 80 FR annual costs and capital costs in technology
50386, 50398 (Aug. 19, 2015) (considered total review); Oil and Natural Gas, 77 FR 49490, 49523 consistent with that of existing fPM
annual costs and capital costs compliance costs and (Aug. 16, 2012) (considered total capital costs and emission limits. The impact of sampling
annualized costs for technology review and beyond annualized costs and capital costs in technology times and random errors on measurable
the floor analysis); Ferroalloys Production, 80 FR review). C.f. NRDC v. EPA, 749 F.3d 1055, 1060 emission limits is described in the ‘‘PM
37366, 37381 (June 30, 2015) (considered total (D.C. Cir. 2014 . . .
annual costs and capital costs, annual costs, and 26 See Cost TSD for 2022 Proposal at Docket ID CEMS Random Error Contribution by
costs compared to total revenues in technology No. EPA–HQ–OAR–2018–0794–4620 at Emission Limit’’ memorandum,
review); Off-site Waste Recovery, 80 FR 14251, regulations.gov. available in the rulemaking docket. The
EPA believes that available PM CEMS pollutants. This proposal is consistent allowed by the 2012 MATS Final Rule
will be able to accurately measure the with the EPA’s authority pursuant to and help prevent any emissions
proposed fPM emission limit of 1.0E–02 CAA section 112(d)(6) to take increases. The EPA does not anticipate
lb/MMBtu, as the average random error developments in practices, processes, any significant non-air health,
contribution is under that of existing and control technologies into account to environmental, or energy impacts as a
emission limits. Although sources have determine if more stringent standards result of these proposed amendments.
reported fPM values as low as 2.0E–04 are achievable than those initially set by Our assessment of control options,
lb/MMBtu, given the 3-hour sampling the EPA in establishing MACT floors, costs, and emission reductions is
duration and the current fPM detection based on developments that occurred in summarized in the memorandum ‘‘2023
limit, the EPA currently believes, as the interim. See LEAN v. EPA, 955 F.3d Technology Review for the Coal- and
described in the memorandum, that 1088, 1097–98 (D.C. Cir. 2020). As Oil-Fired EGU Source Category’’ in
some PM CEMS may struggle to meet discussed above in this section, the EPA Docket ID No. EPA–HQ–OAR–2018–
the EPA’s guideline for average random finds that the vast majority of existing 0794.
error contribution to the PM CEMS coal-fired EGUs are performing well The EPA is not proposing the highest
tolerance to demonstrate compliance below the 2012 MATS fPM emission limit examined (1.5E–02 lb/MMBtu)
with a fPM emission limit of 6.0E–03 lb/ requirements, and that they are because it would largely leave in place
MMBtu or lower. The EPA solicits achieving these levels at lower costs the status quo, in which, despite the
comment on the implications for the than the EPA assumed in the 2012 proven feasibility and effectiveness of
costs of measuring emissions to rulemaking. While this proposal in no control technologies, a number of
demonstrate compliance—whether way refutes that the EPA’s initial MACT sources are lagging far behind. The EPA
through stack testing or PM CEMS—of standards were set at correct levels does not consider a proposed revision to
alternate emission limits set at or below based on the available information at this standard to be consistent with its
6.0E–03 lb/MMBtu as compared to the the time, consistent with CAA section statutory charge.
proposed fPM emission limit of 1.0E–02 112’s statutory scheme requiring the While the EPA is not proposing the
lb/MMBtu, including run durations, EPA to regularly revisit those standards, most stringent limit examined (6.0E–03
fPM detection levels, and random error the EPA now proposes to find that more lb/MMBtu) or an even more stringent
calculations. stringent standards are achievable, as limit, the EPA is taking comment on
The EPA seeks comment broadly on chiefly evidenced by the large majority whether it should consider finalizing
how we should consider costs in the of facilities that are reporting fPM at such a standard. Such a standard would
context of this rule. Taking all of the emission rates well below the current achieve far more emissions reductions
foregoing discussion into account, the standard. than the emission standards that the
EPA believes that the middle option, a This proposed emission limit is EPA is proposing in this action. It
limit of 1.0E–02 lb/MMBtu best comparable to the new source standard would also ensure that the bottom
balances the critical importance of for fPM in MATS. This proposed lowest performing quarter of the fleet
reducing hazardous emissions pursuant emission limit is estimated to reduce would have to improve their
to the EPA’s statutory obligations under non-Hg metal HAP by 6.34 tons per year performance to the level already
CAA section 112(d)(6) and ensuring that (and fPM emissions by 2,074 tons/year) demonstrated by the remaining three-
the worst performers are required to at annual costs between $77.3 and $93.2 quarters of the fleet. The EPA declines
perform at the level of the remainder of million. While the 2020 Residual Risk to propose 6.0E–03 lb/MMBtu as the
the fleet with the costs of doing so in the Review concluded that the residual primary policy option here in light of
context of this industry. Considering all risks are at an acceptable level, Congress the above presentation of potential
the cost metrics, the EPA believes that required the EPA to conduct technology costs, including the EPA’s current
the cost of the proposed standards is reviews on an ongoing basis, at least assessment of measurement uncertainty,
reasonable, and modest in the context of every 8 years, independent of the when considering the current fleet.
this industry. Based on the foregoing residual risk review.27 Moreover, These cost estimates are based on the
discussion and these analyses, the EPA Congress required the EPA to set the assumption that existing ESP-controlled
is proposing to revise the fPM emission standards at the maximum degree of units would need to install a new FF in
limit, as a surrogate for the total non-Hg emissions reductions (including order to meet the lower limit, or if
metal HAP, to 1.0E–02 lb/MMBtu as prohibition on emissions) that is existing FF-controlled units do not meet
supported by our analyses of technical achievable taking into account the the more stringent limit, those units
feasibility, control costs, cost- statutory factors. The technological would need to upgrade their FF bags. If
effectiveness, and economics. The EPA standard approach of CAA section 112 these assumptions are unnecessarily
believes this standard appropriately is based on the premise that, to the conservative, the total costs and
balances CAA section 112’s direction to extent there are controls available to associated cost-effectiveness values may
achieve the maximum degree of reduce HAP emissions, sources should be considerably lower than estimated.
emissions reductions while taking into be required to use them. Since 91 The EPA seeks comment on whether
account the statutory factors, including percent of the anticipated capacity of there are lower cost compliance options
cost. The EPA is further seeking the fleet is already achieving a limit for units with existing ESPs.
comment on whether a standard of below 1.0E–02 lb/MMBtu, the EPA An additional factor affecting the total
6.0E–03 lb/MMBtu or lower (for proposes that this emissions limit level estimated compliance cost is the size
example 2.4E–03 lb/MMBtu, which is is technologically feasible and and composition of the generating fleet.
the average emission of the best demonstrated for a range of control As noted above, the cost estimates in
performing 50 percent of units configurations. Additionally, this Table 3 do not account for market and
evaluated) would represent a better revised limit would result in policy developments that are likely to
balancing of the statutory factors. significantly lower allowable fPM further change the universe of regulated
Indeed, Congress designed CAA emissions from the source category sources and reduce the expected costs of
section 112 to achieve significant compared to the level of emissions meeting more protective fPM standards.
reductions in HAP emissions, which it In the likely case that the power sector’s
recognized are particularly harmful 27 See discussion in section V.A, above. transition to lower-emitting generation
is accelerated by the IRA, for example, where EPA has overestimated costs, measurement of the pollutant of
the total costs and emissions reductions including some of the generation and concern. Such data supply real-time,
achieved by each of the three alternative storage technologies discussed above as quality-assured feedback that can lead
fPM standards shown in Table 3 would well as the cost of PM controls to improved control device and power
also be an overestimate, and the EPA’s themselves. plant operation, which, in turn, can lead
judgment could change about which 2. PM Emission Monitoring to fPM emission reductions. Moreover,
standard most appropriately balances quick detection of potential problems
CAA section 112’s direction to achieve Under the current rule, EGU owners with PM emissions as provided by PM
the maximum degree of emissions or operators may choose among CEMS, coupled with appropriate
reductions while taking into account quarterly testing, PM CEMS, and PM corrective measures, can prevent
cost and other the statutory factors. The CPMS to demonstrate compliance with instances of non-compliance, which
EPA seeks comment on how the IRA the alternate fPM emission limit in otherwise could go undetected and
and other market and policy MATS. The initial MATS ICR, available uncorrected until the next quarterly PM
developments should inform the at www.reginfo.gov,28 anticipated that test. This quicker identification and
Agency’s determination. all EGU owners or operators would use correction of high emitting EGUs will
PM CEMS for compliance purposes and lead to less pollution emitted and lower
Additionally, the EPA notes that other
estimated Equivalent Uniform Annual pollutant exposure for local
future state and federal policies could
Cost (EUAC) for the beta gauge PM communities. In addition to significant
affect the size, composition, and fPM
CEMS to be $65,388. As mentioned in value of more efficient pollution
emissions rate of the future coal-fired
the 2012 proposed Portland Cement abatement, transparency of EGU
EGU fleet. The EPA seeks comment on
NESHAP,29 beta gauge technology, also emissions as provided by PM CEMS,
the extent to which, and how, to take
referred to as beta attenuation, allows along with real-time assurance of
these future policies into account when
PM CEMS to be much less sensitive to compliance has intrinsic value to the
considering the total cost and cost
changes in particle characteristics than public and communities as well as
effectiveness of a more stringent fPM light-based PM CEMS technologies such
emission limit. instrumental value in holding sources
as light-scatter or scintillation. Beta accountable.
The EPA requests public comment on attenuation PM CEMS extracts a sample
all aspects of this proposed rule, from the stack gas and collects the fPM Since promulgation of MATS, two
including our evaluation of the costs on filter tape. The device periodically important developments in the PM
and efficacy of control option advances the tape from the sampling CEMS industry have occurred, which
assumptions. Among other issues, the mode to an area where the sample is the EPA identified as part of this
EPA requests comment on whether we exposed to beta radiation. The detector technology review: cessation of beta
have accurately assessed the variability measures the amount of beta radiation gauge PM CEMS manufacturing and
of fPM emissions and requests emitted by the sample and that amount reduced overall costs for non-beta gauge
information on the costs, pollution can be directly related to the mass of the PM CEMS instruments and installation.
reduction benefits, and cost- filter. The unannualized purchase cost These two occurrences have reduced the
effectiveness of applying lower emission for a beta gauge PM CEMS and its current one-time costs for PM CEMS,
limits to sources subject to MATS; and installation were estimated to be making their use even more cost-
whether there are other factors the EPA $115,267 in the initial MATS ICR; and effective. As shown in Table 4 below,
should consider that would support a the EUAC for beta gauge PM CEMS was average non-beta gauge instrument and
lower emission limit, including the estimated to be less expensive than installation costs obtained from
contribution that HAP from these quarterly EPA Method 5 (M5) testing for representatives of the Institute of Clean
sources make to the overall pollution fPM. Even so, not all EGU owners or Air Companies (ICAC), a trade
burden. The EPA seeks comment on operators chose the most cost-effective association consisting of air pollution
requiring existing coal-fired EGUs to means of demonstrating compliance control and measurement and
meet a fPM standard of 6.0E–03 lb/ with the fPM emission limits. Review of monitoring system manufacturers and of
MMBtu or a more stringent standard reports submitted to WebFIRE and environmental equipment and service
considering the higher emission ongoing ICR renewals shows PM CEMS providers, and from Envea/Altech, a PM
reductions as well as the larger total are used for compliance purposes by CEMS manufacturer and vendor, show
costs such a standard would entail to about one-third of EGU owners or about a 48 percent reduction (from
inform our consideration of whether the operators. In addition to being more $109,420 to $57,095) from average
more stringent standard would reduce cost-effective for compliance purposes, comparable costs determined from the
the overall pollution burden in these PM CEMS provide regulators and the EPA’s CEMS Cost Model and
communities. The EPA also seeks public, as well as the EGU owners or Monitoring Cost/Benefit Analysis Tool
comment on whether there are any areas operators, direct and continuous (MCAT).
TABLE 4—NON-BETA GAUGE PM CEMS COST ESTIMATES USING M5I FOR PS 11
One time costs, Annual costs,
$ $ EUAC,
Data source PM CEMS type $
Instrument and Other initial Capital Operation and Other annual

Audits
installation costs recovery maintenance costs
EPA MCAT .................... In situ ............................ 119,295 81,220 22,016 1,558 54,877 11,219 89,670
Extractive ...................... 152,850 81,220 25,700 2,579 54,877 12,241 95,397
EPA CEMS Cost Model In situ ............................ 65,107 79,813 15,912 2,689 54,392 6,525 79,518
28 See the supporting statement 2137ss06.docx in 29 See 77 FR 42375, July 18, 2012.
ICR reference number 201202–2060–005 at OMB
Control Number 2060–0567.
TABLE 4—NON-BETA GAUGE PM CEMS COST ESTIMATES USING M5I FOR PS 11—Continued
One time costs, Annual costs,
$ $ EUAC,
Data source PM CEMS type $
Instrument and Other initial Capital Operation and Other annual
Audits
installation costs recovery maintenance costs
Extractive ...................... 100,427 84,458 20,300 3,689 54,392 7,525 85,906
Average .................. ....................................... 109,420 81,678 20,982 2,629 54,635 9,378 87,623
ICAC .............................. Low ................................ 35,000 ........................ 3,843 12,000 14,290 ........................ 30,133
High ............................... 40,000 ........................ 4,392 12,000 14,290 ........................ 30,682
Envea/Altech .................. Dry ................................. 34,743 ........................ 3,821 ........................ 14,290 ........................ 18,111
Wet ................................ 118,585 ........................ 13,020 ........................ 14,290 ........................ 27,310
Average .................. ....................................... 57,095 ........................ 6,269 12,000 14,290 ........................ 32,559
Generally, EPA models include other maintenance as a line item in its annual runs used to develop the correlation of
initial costs associated with PM CEMS costs, but neither ICAC nor Envea/ the instrument with the emissions limit
installation, including those associated Altech include audits or other items in remained unchanged, at about 1 hour
with planning, selecting equipment, and their annual costs estimates. Because per run. Such short run durations led to
conducting correlation testing, in its EPA believes some EGUs may require inherent measurement uncertainty
models; such one-time costs are PM spiking—an approach that involves accounting for more than half the
annualized along with instrument and introducing known amounts of fPM to emission limit at the expected portland
installation costs. The proposed lower increase fPM concentration without cement plant operating condition,
fPM emission limit will require longer altering control device equipment—the leading some to question whether
duration runs for M5 testing and may EPA added $14,290 (the annualized cost values provided by instrumentation
require the use of M5I, which was of conducting $35,000 p.m. spiking were appropriately related to emissions.
designed for PM CEMS correlation every 3 years at an interest rate of 7 The conditions experienced by
testing at low fPM levels. Initial costs in percent) to the audit portion of all portland cement facilities that required
Table 4 for M5I emission testing are entries. As mentioned earlier, omission revisions to emission limits and
$58,000; such testing includes 18 runs of specifically named costs does not compliance determination method are
of 3-hour duration spread over 9 total necessarily mean that those costs have not similar to those expected to be faced
days. PM CEMS correlation testing for been excluded; rather these costs may by EGU owners or operators subject to
the proposed lower fPM levels using M5 be included in other listed costs. Using MATS. First, the fuel used by coal-fired
is estimated to be $41,000. Of course, the data provided and explained above, EGUs is more uniform and its
the quarterly testing run durations the average EUAC for PM CEMS that characteristics are more consistent than
would need to increase if PM CEMS rely on M5I correlation testing is about those of the fuel and additive mixtures
were not used; annual cost for M5 63 percent lower than the average EUAC used by portland cement kilns. Such
testing with 3 hour run duration is from EPA models (from $87,623 to fuel combustion particle consistency
estimated to be $85,127 ($82,000 for $32,559). Given that the annual cost of allows technologies such as light
testing, and $3,127 for 24 hours of site quarterly M5 testing for fPM is now scattering and scintillation, in addition
technical support); quarterly testing estimated to be $85,127, annualized to beta gauges, to be used by PM CEMS
using M5I with runs of similar duration other one-time costs and operation and for compliance determination purposes.
is estimated to be $107,127. However, maintenance, audits, and other Moreover, consistent fPM particle
neither ICAC nor Envea/Altech annualized costs—if omitted by the characteristics for EGUs provide stable
explicitly included those costs as line manufacturers—would have to be more correlations for those EGUs with
items in their estimates. This does not than $52,568 for PM CEMS to be less existing PM CEMS; while the fPM
necessarily mean that such costs have cost-effective than quarterly testing. particle characteristics provide
been excluded; if such costs have been As mentioned in the proposed correlations that remain within
included, then the estimates do not Portland Cement NESHAP from 10 years specifications, as evidenced by ongoing
change, but if such costs have not been ago (see 77 FR 42374, July 18, 2012), the relative correlation audits, the existing
included, the estimates may increase. EPA was aware of the potential correlations do not change and can
Their average capital recovery cost, difficulty use of PM CEMS might have continue to be used now and in the
determined from the sum of the created in determining compliance for future without having to develop a new
instrument, installation, and other that rulemaking due to the low end of correlation. Second, the proposed
initial costs amortized over 15 years at emission limits (0.04 lb/ton clinker, MATS emission limit of 1.0E–02 lb/
a 7 percent interest rate, is about 70 which translates to a range of about 5 to MMBtu, which translates to about 7.3
percent lower than that obtained from 8 mg/dscm, depending on particle mg/dscm, coupled with a minimum
the average capital recovery cost characteristics) and to the short duration sampling collection time of 3 hours per
obtained from the EPA models. As of emission test runs. The EPA run, based on a typical sampling rate of
shown in the table, EPA models also addressed those concerns for that 3⁄4 cubic feet per minute, avoids the
include annual costs for operation and rulemaking by proposing to raise the measurement problems described by the
maintenance, relative response and emission limit to 0.07 lb/ton clinker, Portland Cement NESHAP by reducing
correlation audits, and other items such which translated to a range of about 7 the average inherent measurement
as reporting and recordkeeping. The to 14 mg/dscm, and to no longer require uncertainty for half of the proposed
sum of those items plus the capital PM CEMS use; instead, owners or emission limit (where the EGU is
recovery cost yields EUAC of PM CEMS. operators would use their PM CEMS as expected to operate) from more than 50
ICAC includes operation and PM CPMS. Even so, the durations of test to 80 percent. In addition, use of 3 hour
run durations would allow for a 6.0E– Aerosol Generator (QAG), would allow levels proposed in this rule. Moreover,
03 lb/MMBtu (or about 4.4 mg/dscm) direct PM CEMS calibration, as opposed the EPA seeks comment on additional or
MATS emission limit, which the EPA is to the development of a curve that other approaches that could be
seeking comment on, to have an average provides a correlation for the PM employed to facilitate PM CEMS use for
inherent measurement uncertainty due CEMS.31 That study relied on six the proposed emission levels. Specific
to random error of 14 percent at the emission rates, four of which were at or comments on direct PM CEMS
target PM CEMS operational limit of under 5 mg/dscm, and reported calibration methods, such as the QAG,
3.0E–03 lb/MMBtu. As shown, inherent successful sample collection and as well as limitations, are welcome.
measurement uncertainty does not transport. EPRI continued this work and
The EPA solicits comment on the
appear to be problematic for the primary provided a technical update in 2014,32
availability of beta gauge instruments,
proposed emission limit, but, as but the EPA is unaware of specific
on the current average costs of non-beta
mentioned earlier, some PM CEMS may recommendations or suggestions
gauge PM CEMS instruments and
have difficulty meeting the inherent regarding QAG application to PM
measurement uncertainty—specifically, CEMS. While we believe the use of the installation, on ICAC’s annual costs, and
the average random error component— QAG could lower fPM monitoring costs on Envea/Altech’s annual costs. When
of the alternative proposed emission for PM CEMS use, we seek more commenting on EPA model estimates or
limit. Note that the primary proposed information on its application for lower ICAC’s or Envea/Altech’s estimates,
MATS emission limit is just above the fPM limits as measured by PM CEMS; please provide specific PM CEMS
fPM limit for new EGUs, as 9.0E–02 lb/ specifically, we solicit comment on instrument type, manufacturer, and
MWh on an electrical output basis whether implementation of the QAG is model; cost information broken down
translates to about 9.0E–03 lb/MMBtu another reason that PM CEMS costs by initial cost including instrument type
on a heat input basis. MATS requires have decreased. and installation cost, and annual cost,
use of PM CEMS for new EGUs, along For these reasons, we propose to including operation and maintenance,
with minimum sampling collection time require the use of PM CEMS as the audit, and other costs in your
of 3 hours per run.30 Proposed use of method to demonstrate compliance with comments. Moreover, please identify in
runs of at least 3 hour durations and the fPM emissions limit for coal-fired your comments specific items included
emission limits of 1.0E–02 lb/MMBtu and IGCC EGUs pursuant to the EPA’s in your cost information, such as
would be consistent with run durations authority under CAA section 112(d)(6). installation, operation, and maintenance
and limits already in MATS. Third, If our proposal is finalized, EGU owners provisions. The EPA also solicits
Performance Specification 11 (PS 11), or operators currently relying on comment on the cost-effectiveness of
which provides procedures and quarterly PM emissions testing would PM CEMS as compared to quarterly PM
acceptance criteria for validating PM need to install, operate, and maintain emissions testing. Also, the EPA solicits
CEMS technologies, already anticipates PM CEMS. Such a switch is projected to comment on the availability of PM
and includes approaches for developing be more cost-effective, more CEMS and their use for compliance
low-level emission correlations for PM informative, and more effective in purposes, especially when compared to
CEMS. Those techniques include assuring compliance than use of less frequent, more expensive measures.
varying process operations; varying fPM quarterly testing. Those EGU owners or The EPA is aware that some EGUs
control device conditions; PM spiking operators already using PM CEMS as may be on enforceable schedules to
zero point methods when the previous their means of compliance cease operations, which may be just
techniques are not able to provide the 3 determination would maintain their beyond the three-year compliance date
distinct fPM concentration levels. As current approach; while some may have the EPA proposes for PM CEMS
mentioned earlier, average costs for fPM no need for additional expenditures, the monitoring requirements in section V.E,
spiking are about $35,000 every 3 years, proposal includes the costs associated below, and that owners or operators of
or $14,290 annually at an interest rate with revised and ongoing correlation EGUs may be unable to recoup
of 7 percent, and not every EGU will testing and spiking for all EGUs. Since investments in PM CEMS if the
need to adjust its existing correlation in a proposed requirement for use of PM instruments are not in operation for at
order to continue to use its existing PM CEMS renders the current compliance least a certain period of time beyond
CEMS to demonstrate compliance with option for the LEE program superfluous, their installation date. Therefore, the
the proposed limits; however, for the EPA proposes to remove the EPA seeks comment on whether EGUs
purposes of this proposal, costs for individual and total non-Hg metal HAP
should be able to continue to use
spiking will be included in annual PM and the surrogate fPM from the LEE
quarterly emissions testing past the
CEMS cost estimates. In addition to program for all MATS-affected EGUs
proposed compliance date for a certain
these techniques to aid PM CEMS use and solicits comments on removing
period of time or until EGU retirement,
for rules with low level emissions, the these limits.
The EPA seeks comment on whichever occurs first, provided the
EPA is aware that the Electric Power EGU is on an enforceable schedule for
Research Institute (EPRI) began working distinctions between portland cement
plants and EGUs that would facilitate ceasing coal- or oil-fired operation. In
with an instrument manufacturer in addition, the EPA seeks comment on
2009, prior to MATS promulgation, to PM CEMS use at EGUs. Specifically, the
EPA seeks comment on the ability, type, what would qualify as an enforceable
develop a National Institute of schedule, such as that contained in the
and capabilities of PM CEMS to
Standards and Technology (NIST) Agency’s ‘‘EGUs Permanently Ceasing

traceable aerosol generator that injects accurately measure fPM emissions at the
Coal Combustion by 2028’’ included in
known particle size distribution and the 2020 Steam Electric ELG
31 See A Qualitative Aerosol Generator Designed
mass into PM CEMS. Such an for Particulate Matter (PM) Continuous Emissions Reconsideration Rule (85 FR 64640,
instrument, known as a Quantitative Monitoring Systems (CEMS) Calibration, available 64679, and 64710; 10/13/2020), as well
at www.epri.com/research/products/1017574. as what the maximum duration of
30 See Table 1 to subpart UUUUU of 40 CFR part 32 See Quantitative Aerosol Generator (QAG) for
63. At a typical sampling rate of 3⁄4 cubic foot per Calibration of Particulate Monitors: 2014 Technical
operation using quarterly emissions
minute, a run would require 3 hours to collect at Update, available at www.epri.com/research/ testing for compliance purposes should
least 4 cubic meters of sample. products/3002003343. be.
3. Review of the Hg Emission Standards the limited free halogen in the flue gas on the use of ACI for Hg control. See 77
a. Overview of Hg Emissions From and prevent oxidation of the Hg0. FR 9304, February 16, 2012. The EPA
Some coal-fired power plants— established a final Hg emission standard
Combustion of Coal
especially those firing bituminous of 4.0 pounds of Hg per trillion British
Mercury is a naturally occurring coal—achieve some level of Hg thermal units of heat input (lb Hg/TBtu)
element found in small and varying emissions control using existing for lignite-fired utility boilers. The EPA
quantities in coal. During combustion of equipment that was installed to remove promulgated a final Hg emission
coal, Hg is volatilized and converted to other pollutants, including PM, SO2, standard for EGUs firing non-lignite
elemental Hg vapor (Hg0) in the high and nitrogen oxides (NOX). Particulate- coals, including bituminous and
temperature regions of the boiler. Hg0 bound Hg (Hgp) is effectively removed subbituminous coal, of 1.2 lb Hg/TBtu.
vapor is difficult to capture because it along with PM in PM control equipment Under CAA section 112(d)(1), the
is typically nonreactive and insoluble in such as FFs and ESPs. Soluble Hg2+ Administrator has the discretion to
aqueous solutions. However, under compounds (such as HgCl2) can be ‘‘distinguish among classes, types, and
certain conditions, the Hg0 vapor in the effectively captured in wet FGD sizes of sources within a category or
flue gas can be oxidized to divalent Hg systems. And, while a selective catalytic subcategory’’ in establishing standards.
(Hg2+). The Hg2+ can bind to the surface reduction (SCR) system that has been Any basis for subcategorization must be
of solid particles (e.g., fly ash) in the installed for NOX control does not itself related to an effect on HAP emissions
flue gas stream, often referred to as capture Hg, it can under the right that is due to the difference in class,
‘‘particulate bound Hg’’ (Hgp), and be conditions enhance the oxidation of Hg0 type, or size of the units. See 76 FR
removed in a downstream PM control in the flue gas for increased Hg removal 25036–25037.
device. Oxidized Hg compounds can in a downstream PM control device or When developing the MATS rule, the
also be soluble and can be removed in in a wet FGD scrubber. EPA examined available Hg emissions
a wet scrubber. The presence of chlorine However, because the Hg in their flue data from coal-fired EGUs and found
in gas-phase equilibrium favors the gas tends to be present in the non- that there were no lignite-fired EGUs
formation of mercuric chloride (HgCl2) reactive Hg0 phase, EGUs firing among the top performing 12 percent.
at flue gas cleaning temperatures. subbituminous coal or lignite often get The EPA then determined that the
However, Hg0 oxidation reactions are little to no control from equipment difference in the emissions from the
kinetically limited as the flue gas cools designed and installed for other lignite-fired EGUs was due to a
and, as a result, Hg often enters the flue pollutants. While some bituminous difference in the class, type, or size of
gas cleaning device(s) as a mixture of coal-fired EGUs require use of those units and finalized two
Hg0, Hg2+ compounds, and Hgp. This additional Hg-specific control subcategories of coal-fired EGUs for Hg
technology, such as injection of a emissions. See 76 FR 25036–67. The
partitioning into various species of Hg
sorbent or chemical additive, to EPA considered basing the subcategory
has considerable influence on selection
supplement the control that these units definition solely on an EGU (1) being
of Hg control approaches. In general,
already achieve from criteria pollutant designed to burn lignite and (2) burning
because of the presence of higher
control equipment, these Hg-specific lignite. However, the EPA decided not
amounts of halogen (especially chlorine)
control technologies are often required to do so because of the concern that
in bituminous coals, most of the Hg in
as part of the Hg emission reduction such a definition would allow sources
the flue gas from bituminous coal-fired
strategy at EGUs that are firing to potentially meet the definition by
boilers is in the form of Hg2+
subbituminous coal or lignite. As combusting very small amounts of low
compounds, typically HgCl2 and is more
mentioned, the Hg in the flue gas for rank virgin lignite. In the preamble of
easily captured in downstream control the 2012 MATS Final Rule, the EPA
those EGUs tends to be in the non-
equipment. Conversely, both reactive Hg0 phase due to lack of free suggested a scenario where an EGU that
subbituminous coal and lignite have halogen to promote the oxidation was not designed to burn lignite and did
lower halogen content, compared to that reaction. To alleviate this challenge, not routinely burn lignite could import
of bituminous coals, and the Hg in the activated carbon and other sorbent one truck full of low rank virgin coal
flue gas from boilers firing those fuels providers and control technology and burn a very small quantity of it
tends to be in the form of Hg0 and is vendors developed methods to periodically to meet the subcategory
more challenging to control in introduce halogen into the flue gas to definition. To avoid creating this
downstream control equipment. improve the control of Hg emissions potential loophole, the EPA also
Fly ash is typically classified as acidic from EGUs firing subbituminous coal finalized a requirement that the unit be
(pH less than 7.0), mildly alkaline (pH and lignite. This was primarily through constructed and operated at or near a
greater than 7.0 to 9.0), or strongly the injection of pre-halogenated (often mine containing the low rank virgin
alkaline (pH greater than 9.0). The pH pre-brominated) activated carbon coal it burns. The EPA indicated that
of the fly ash is usually determined by sorbents or through the injections of the final definition would prevent other
the calcium/sulfur ratio and the amount halogen-containing chemical additives EGUs that are not firing lignite from
of halogen. The ash from bituminous along with conventional sorbents. This complying with the less stringent Hg
coals tends to be acidic due to the challenge to controlling Hg emissions emission standard. The final definition,
relatively higher sulfur and halogen was a challenge for EGUs firing as specified in the 2012 MATS Final
content and the glassy (nonreactive) subbituminous coal and for EGUs firing Rule (77 FR 9369, February 16, 2012),
nature of the calcium present in the ash. lignite. was: ‘‘Unit designed for low rank virgin
Conversely, the ash from subbituminous coal subcategory means any coal-fired
and lignite coals tends to be more b. Hg Emission Standards in the 2012 EGU that is designed to burn and that
alkaline due to the lower amounts of MATS Final Rule is burning non-agglomerating virgin coal
sulfur and halogen and a more alkaline In the 2012 MATS Final Rule, the having a calorific value (moist, mineral
and reactive (non-glassy) form of EPA promulgated a beyond-the-floor matter-free basis) of less than 19,305 kJ/
calcium in the ash. The natural standard for Hg for the subcategory of kg (8,300 Btu/lb) that is constructed and
alkalinity of the subbituminous and existing coal-fired units designed for operates at or near the mine that
lignite fly ash can effectively neutralize low rank virgin coal (i.e., lignite) based produces such coal.’’
c. Beyond-the-Floor Analysis for the MMacf injection rate for units with MATS rule, the EPA estimated Hg
2012 MATS Final Rule installed FF and using treated (i.e., emissions had been reduced to 4 tons,
For the 2012 MATS Final Rule, the brominated) activated carbon or at an an 86 percent decrease.34 This decline
EPA calculated beyond-the-floor costs injection rate of 3.0 lb/MMacf for units was due to the installation and use of
for Hg controls by assuming injection of using treated activated carbon with Hg controls as well as other significant
brominated activated carbon at a rate of installed ESPs. changes in the power sector (e.g., coal
3.0 pounds of sorbent per million actual Petitioners challenged the beyond- plant retirements, increase use of
cubic feet of flue gas (lb/MMacf) for the-floor standard for lignite-fired EGUs, natural gas and renewable energy, etc.)
lignite-fired EGUs with an ESP for PM claiming that the final standard is not in the same time period.
control and at an injection rate of 2.0 lb/ achievable because they asserted that i. Hg Emissions From Coal-Fired EGUs
MMacf for lignite-fired units with a the standard would require in 2021
baghouse (also known as a fabric filter, unrealistically high levels of Hg
FF). The sorbent injection rate of 2.0 lb/ reduction. In White Stallion v. EPA, the Hg emission reductions have
MMacf for lignite-fire units with FFs is Court of Appeals of the District of continued to decline since 2017 as more
consistent with the rate assumed for all Columbia Circuit rejected petitioners’ coal-fired EGUs have retired or reduced
other coal types. The EPA assumed a challenge to the final beyond-the-floor utilization. The EPA estimated that 2021
sorbent injection rate of 3.0 lb/MMacf standard on the basis that the EPA had Hg emissions from coal-fired EGUs were
for lignite-fired units with ESPs, which adequately concluded during the 3 tons (a 90 percent decrease compared
is lower than the sorbent injection rate rulemaking process that the standard for to pre-MATS levels).35 However, units
of 5.0 lb/MMacf that the EPA assumed lignite units were achievable if sources burning lignite coal (or permitted to
for EGUs firing using other (non-lignite) increased their use of a particular burn lignite) accounted for a
coal types. In the Beyond-the-Floor control technology, ACI. See White disproportionate amount of the total Hg
Memo (see Docket ID No. EPA–HQ– Stallion Energy Center, LLC v. EPA, 748 emissions in 2021. As shown in Table
OAR–2009–0234–20130), the EPA F.3d 1222, 1251 (D.C. Cir. 2014). 5 below, 16 of the top 20 Hg-emitting
indicated that this lower sorbent EGUs were lignite-fired EGUs. Overall,
d. Hg Emission Reductions Since
injection rate was appropriate, because lignite-fired EGUs were responsible for
Promulgation of the 2012 MATS Final
a higher rate would likely result in Hg almost 30 percent of all Hg emitted from
Rule
emission reductions greater than those coal-fired EGUs in 2021, while
needed to meet the beyond-the-floor The EPA estimated annual Hg generating about 7 percent of total 2021
standard of 4.0 lb/TBtu noting that emissions from coal-fired power plants megawatt-hours. Lignite accounted for 8
greater than 90 percent control can be in 2010 (pre-MATS) to be 29 tons.33 In percent of total U.S. coal production in
achieved at lignite-fired units at a 2.0 lb/ 2017, after full implementation of the 2021.
TABLE 5—TOP HG-EMITTING EGUS IN 2021

2021 Hg
Rank EGU Fuel emissions State
(lb)
1 .................... Coal Creek 2 ...................................................... Lignite ................................................................. 181.8 ND

2 .................... Coal Creek 1 ...................................................... Lignite ................................................................. 175.6 ND
3 .................... Oak Grove 2 ....................................................... Lignite ................................................................. 149.8 TX
4 .................... Martin Lake 3 ...................................................... Lignite/Subbituminous ........................................ 134.4 TX
5 .................... Oak Grove 1 ....................................................... Lignite ................................................................. 112.7 TX
7 .................... Milton R Young B2 ............................................. Lignite ................................................................. 103.1 ND
9 .................... Antelope Valley B2 ............................................. Lignite ................................................................. 89.8 ND
10 .................. Coyote B1 ........................................................... Lignite ................................................................. 79.9 ND
11 .................. H W Pirkey Power Plant 1 * ............................... Lignite/Subbituminous ........................................ 71.1 TX
12 .................. Antelope Valley B1 ............................................. Lignite ................................................................. 69.6 ND
13 .................. San Miguel SM–1 ............................................... Lignite ................................................................. 64.6 TX
14 .................. Sandy Creek Energy Station S01 ...................... Subbituminous .................................................... 53.5 TX
15 .................. Limestone LIM2 .................................................. Lignite/Subbituminous ........................................ 52.5 TX
16 .................. Milton R Young B1 ............................................. Lignite ................................................................. 52.4 ND
17 .................. Comanche 3 ....................................................... Subbituminous .................................................... 50.3 CO
18 .................. Leland Olds 2 ..................................................... Lignite ................................................................. 50.1 ND
19 .................. James H Miller Jr 3 ............................................ Subbituminous .................................................... 42.9 AL
20 .................. Labadie 2 ............................................................ Subbituminous .................................................... 42.5 MO
* This unit has announced its intention to retire in 2023.
ii. Limited CAA Section 114 Request 7414(a), the EPA solicited information EGUs to inform this CAA section
related to Hg emissions and Hg control 112(d)(6) technology review. The
In May 2021, pursuant to authority in technologies from certain lignite-fired selected lignite-fired EGUs were asked
section 114 of the CAA, 42 U.S.C.
33 Memorandum: Emissions Overview: Hazardous 34 2017 Power Sector Programs Progress Report; 35 2021 Power Sector Programs Progress Report;
Air Pollutants in Support of the Final Mercury and available at https://www.epa.gov/sites/default/files/ available at https://www3.epa.gov/airmarkets/
Air Toxics Standard. EPA–454/R–11–014. 2019-12/documents/2017_full_report.pdf and in the progress/reports/pdfs/2021_full_report.pdf and in
November 2011; Docket ID No. EPA–HQ–OAR– rulemaking docket. the rulemaking docket.
2009–0234–19914.
to provide information on their control the CAA section 114 information included in the CAA section 114
configuration for Hg and for other air request is shown below in Table 6. information request) is also included.
pollutants (e.g., criteria pollutants such Additional information on the location, The additional information was
as PM, NOX, SO2). Selected information size (capacity), firing configuration, and obtained from the EPA’s NEEDS
on lignite-fired EGU control control configuration of lignite-fired database.36
configurations that was obtained from EGUs (including those few that were not
TABLE 6—CONTROL CONFIGURATIONS FOR LIGNITE-FIRED EGUS
Capacity
Plant name State Firing Control configuration Hg control description Hg control
(MW)
Antelope Valley #1 ....... ND 450 tangent ..... ACI + SDA + FF ............ Does not use activated carbon as its Nalco non-carbon, non-halogenated
Antelope Valley #2 ....... ND 450 tangent ..... ACI + SDA + FF. sorbent, instead injects a liquid liquid sorbent added to dry scrub-
sorbent to the scrubber. The facil- ber; M-Sorb additive (bromide).
ity stopped using refined coal in
December 2021.
Coal Creek #1 .............. ND 574 tangent ..... ACI + ESPC + WFGD ... Information not collected in the CAA 114 request.
Coal Creek #2 .............. ND 573 tangent ..... ACI + ESPC + WFGD.
Coyote .......................... ND 429 cyclone ..... ACI + SDA + FF ............ Information not collected in the CAA 114 request.
Leland Olds #1 ............. ND 222 wall ........... SNCR + ACI + ESPC + Activated carbon and oxidizer injec- ME2C SEA SF10 Oxidizer and SB24
WFGD. tions for Hg control. Activated Carbon.
Leland Olds #2 ............. ND 445 cyclone ..... SNCR + ACI + ESPC +
WFGD.
Milton R Young #1 ....... ND 237 cyclone ..... SNCR + ACI + ESPC + Hg controlled by Powdered Activated DARCO Hg-H non-halogenated Pow-
Milton R Young #2 ....... ND 447 cyclone ..... WFGD. Carbon Injection plus Oxidizing dered Activated Carbon + ADA M-
SNCR + ACI + ESPC + Agent/Halogen Injection System. Prove additive.
WFGD.
Spiritwood Station ........ ND 92 FBC .......... SNCR + ACI + SDA + Hg emissions are controlled by acti- Activated Carbon sorbent (not speci-
FF. vated carbon injection system and fied).
a CEMS. The activated carbon in-
jection feed rate is adjusted to
maintain emissions below the 4.0
lb/TBtu standard.
Limestone #1 ............... TX 831 tangent ..... SNCR + ACI + ESPC + Information not collected in the CAA 114 request.
WFGD.
Limestone #2 ............... TX 858 tangent ..... SNCR + ACI + ESPC +
WFGD.
Major Oak #1 ............... TX 152 FBC .......... Reagent Injection + Hg is controlled by the introduction of Cabot DARCO Hg-H non-Brominated
Major Oak #2 ............... TX 153 FBC .......... SNCR + ACI + FF. activated carbon into each boiler AC + ADA–ES M-Prove additive.
Reagent Injection + duct directly in front of the
SNCR + ACI + FF. baghouse. A halogen fuel additive
is also applied to the lignite before
it enters the day silos.
Martin Lake #1 ............. TX 800 tangent ..... ACI + ESPC + WFGD ... Brominated additive injected into the ME2C SEA process (non-Brominated
Martin Lake #2 ............. TX 805 tangent ..... ACI + ESPC + WFGD. furnace and activated carbon in- AC + chemical additive).
Martin Lake #3 ............. TX 805 tangent ..... ACI + ESPC + WFGD. jected upstream of the air heater.
In 2020 and 2021 Refined Coal
System applied an aqueous bro-
mine salt solution to the coal.
Oak Grove #1 .............. TX 855 tangent ..... SCR + ACI + FF + Brominated activated carbon injected ADA–CS Br–AC.
Oak Grove #2 .............. TX 855 wall ........... WFGD. downstream of the air heater.
SCR + ACI + FF + From 2018 to 2021, the unit was
WFGD. equipped with a Refined Coal Sys-
tem for Hg control. This system
applied an aqueous bromine salt
solution to the coal downstream of
the crusher. The refined coal sys-
tem is no longer in service.
Red Hills #1 ................. MS 220 FBC .......... Reagent Injection + ACI Hg is controlled by injection of acti- ADA–CS non-Br AC + ADA–ES M45
Red Hills #2 ................. MS 220 FBC .......... + FF. vated carbon into each boiler duct liquid additive.
Reagent Injection + ACI directly in front of the baghouse. A
+ FF. fuel additive is also applied to the

lignite before it enters the day
silos. The application of fuel addi-
tives ended in December 2021.
36 National Electric Energy Data System (NEEDS) www.epa.gov/power-sector-modeling/national-

v621 rev: 10–14–22, available at: https:// electric-energy-data-system-needs-v6.
TABLE 6—CONTROL CONFIGURATIONS FOR LIGNITE-FIRED EGUS—Continued

Capacity
Plant name State Firing Control configuration Hg control description Hg control
(MW)
San Miguel ................... TX 391 wall ........... SNCR + ACI + ESPC + Hg is captured using a sorbent en- ME2C SEA process (non-Br AC +
WFGD. hanced additive (SEA) injected powder-based chemical additive).
onto the lignite at the pulverizer
feeders or directly into the furnace
to promote the oxidation and cap-
ture of Hg. This is followed by an
ACI system located in the boiler
exit duct work upstream of the air
heaters. The scrubber system also
reduces Hg emissions.
Note: ACI = activated carbon injection; SDA = spray dryer absorber (dry scrubber); FF = fabric filter; ESPC = cold side electrostatic precipitator; WFGD = wet flue
gas desulfurization scrubber; SNCR = selective non-catalytic reduction (NOX control); reagent injection = sorbent injection into fluidized bed combustor.
Most, but not all, of the EGUs utilized capture in downstream control EGUs that were designed to burn lignite
a combination of the use of a chemical equipment (e.g., FGD scrubbers, PM utilized refined coal to some extent in
additive and injection of a sorbent as control devices). Several of the facilities 2021, as summarized in Table 7. EIA
their Hg control strategy. One facility in noted that use of refined coal as a part form 923 does not specify the type of
North Dakota (Antelope Valley) uses a of their Hg control strategy was coal that is ‘‘refined’’ when reporting
liquid sorbent that is injected to the SO2 discontinued at the end of 2021 when boiler or generator fuel use. For this
scrubber (spray dryer absorber, SDA). the refined coal production tax credit technology review, the EPA has
Many of the EGUs used ‘‘refined coal.’’ (created by the American Jobs Creation assumed that the facilities have utilized
Refined coal is typically produced by Act of 2004) expired. According to a ‘‘refined lignite,’’ as reported in fuel
U.S. Government Accountability Office
mixing proprietary additives to receipts on EIA form 923. However,
audit report, refined coal producers
feedstock coal to help capture emissions several ‘‘lignite-fired EGUs’’ located in
claimed approximately $8.9 billion in
when the coal is burned. For example, tax credits between 2010 and 2020. Texas reported very high use of
these additives may promote the According to fuel use information subbituminous coal in 2021 (ranging
oxidation of Hg to Hg2+ compounds for supplied to EIA (on form 923), 13 of 22 from 76 percent up to > 99 percent).
TABLE 7—2021 FUEL USE AT LIGNITE-FIRED EGUS

Distillate fuel Subbituminous
Natural gas Lignite coal Refined coal
Plant name oil coal
(%) (%) (%)
(%) (%)
Antelope Valley 1 ................................................................. 0.0 0.6 5.8 93.5 0.0

Antelope Valley 2 ................................................................. 0.0 0.6 5.8 93.5 0.0
Coal Creek 1 ........................................................................ 0.1 0.0 0.0 99.9 0.0
Coal Creek 2 ........................................................................ 0.1 0.0 0.0 99.9 0.0
Coyote 1 ............................................................................... 0.3 0.0 99.7 0.0 0.0
Leland Olds 1 ....................................................................... 0.3 0.0 37.6 62.1 0.0
Leland Olds 2 ....................................................................... 0.3 0.0 6.2 93.6 0.0
Milton R Young 1 ................................................................. 0.4 0.0 17.0 82.6 0.0
Milton R Young 2 ................................................................. 0.2 0.0 12.1 87.6 0.0
Spiritwood Station 1 ............................................................. 0.0 35.6 0.0 64.4 0.0
Limestone 1 ......................................................................... 0.0 0.2 0.0 0.0 99.8
Limestone 2 ......................................................................... 0.0 0.8 0.0 0.0 99.2
Major Oak Power 1 .............................................................. 0.0 0.2 99.8 0.0 0.0
Major Oak Power 2 .............................................................. 0.0 0.0 100.0 0.0 0.0
Martin Lake 1 ....................................................................... 0.1 0.0 23.5 0.0 76.4
Martin Lake 2 ....................................................................... 0.1 0.0 22.4 0.0 77.5
Martin Lake 3 ....................................................................... 0.1 0.0 19.2 0.0 80.6
Oak Grove 1 ........................................................................ 0.0 1.9 3.4 94.7 0.0
Oak Grove 2 ........................................................................ 0.0 0.0 3.7 96.3 0.0
Red Hills Generating Facility 1 ............................................ 0.0 0.3 0.0 99.7 0.0
Red Hills Generating Facility 2 ............................................ 0.0 0.3 0.0 99.7 0.0
San Miguel 1 ........................................................................ 0.2 0.0 99.8 0.0 0.0
e. CAA Section 112(d)(6) Technology standard, the EPA evaluated the 2021 to burn lignite but burned a significant
Review of the Hg Standards performance of EGUs firing non-lignite amount of subbituminous coal) emitted
i. Review of the Hg Emission Standard coals and found that EGUs firing Hg at an average annual rate of 0.6 lb
for Non-Lignite-Fired EGUs primarily bituminous coal emitted Hg at Hg/TBtu (with a range of 0.1 to 1.2 lb/
an average annual rate of 0.4 lb Hg/TBtu TBtu). This represents a control range of
The final MATS Hg emission limit for (with a range of roughly 0.2 to 1.2 lb Hg/ 98 to 77 percent (assuming an average
EGUs firing non-lignite coals (i.e., TBtu). EGUs firing primarily inlet concentration of 5.5 lb/TBtu). The
bituminous and subbituminous coals) is subbituminous coal in 2021 (not EPA has information on the control
1.2 lb Hg/TBtu. To review that emission including those EGUs that are permitted configurations of these non-lignite
EGUs. However, because the non- additives are injected, the type and rate ii. Review of the Hg Emission Standard
lignite-fired EGUs were not included in of chemical additive injection. For those for Lignite-Fired EGUs
the limited CAA section 114 reasons, the EPA is not proposing to
information collection, the EPA does adjust the Hg emission standard for non- The final MATS Hg emission limit for
not have detailed information on the lignite-fired EGUs at this time. However, EGUs firing lignite coal is 4.0 lb Hg/
type of sorbent injected (e.g., activated the EPA solicits comment on the TBtu—more than three times the
carbon or non-carbonaceous; pre- performance of Hg controls for non- standard for non-lignite coal. To review
halogenated, etc.). The EPA also does lignite-fired EGUs, including that emission standard, the EPA
not have detailed information on the information on the type and injection evaluated the data obtained in the 2022
injection rate of sorbents used for Hg rate of sorbents used for Hg control, as CAA section 114 data survey along with
control (if any). Similarly, the EPA does well as the possibility of additional cost- the emissions data reported to the EPA
not have information on the type of effective measures to further reduce Hg and the fuel use data submitted to EIA.
quantity of chemical additives used (if from equipment installed for criteria The 2021 performance of lignite-fired
any). However, the bituminous coal- pollutants. The EPA also seeks comment EGUs (including those permitted to
fired EGUs are already achieving an on whether there would be a reasonably burn lignite but that utilized significant
average annual rate of 0.4 lb/TBtu and efficient way to more thoroughly survey amounts of subbituminous coal in 2021)
the subbituminous coal-fired EGUs are the types of controls—including the is shown in Table 8 below. The table
types of sorbents used and their shows a ‘‘Hg Inlet’’ level which reflects
already achieving an average annual rate
injection rates—used to limit Hg the maximum Hg content of the range of
of 0.6 lb/TBtu. The typical Hg control
emissions at non-lignite-fired EGUs, and feedstock coals that the EPA assumes is
performance curves for sorbent injection
whether conducting such additional available to each of the plants in the
show a leveling off such that increasing
information collection would be Integrated Planning Model, IPM,37 the
the amount of sorbent results in worthwhile.
diminishing improvement in Hg control. estimated control (percentage) needed to
In addition, the EPA notes that several
Based on full-scale demonstration meet an emission standard of 4.0 lb Hg/
states have adopted Hg reduction
testing of Hg sorbents, this leveling off TBtu (the current standard for lignite-
standards that go beyond the 2012
typically takes place somewhere greater fired EGUs) and the estimated control
MATS Final Rule in their reduction
than 90 percent capture. Without target. For instance, Connecticut, (percentage) to meet an emission
knowing the type of sorbent being Minnesota, Montana, New York, standard of 1.2 lb Hg/TBtu (the current
injected or the rate of the sorbent Oregon, and Utah all established input- standard for non-lignite-fired EGUs).
injection, it is difficult to determine based Hg limits below 1.2 lb/TBtu. For The table also shows the estimated 2021
whether additional emission reductions further detail on all 18 states with Hg inlet concentration from actual 2021
could be achieved in a cost-effective existing Hg emissions limits, see fuel usage (as mentioned earlier, some
manner. For bituminous coal-fired EGUs Chapter 3 of EPA’s IPM documentation, units utilized significant quantities of
that do not utilize sorbent injection but available in the docket. The EPA solicits non-lignite fuel, e.g., subbituminous
rely on co-benefit control from information about the cost and coal, natural gas, etc.) and the 2021 Hg
equipment installed for criteria effectiveness of control strategies that emissions reported to the EPA. The EPA
pollutants, it is difficult to determine EGUs in these states utilize to meet then estimated the apparent level of Hg
whether additional Hg emission more stringent Hg emission standards control for 2021 and the level of control
reduction could be obtained in a cost- than those promulgated in the 2012 that would been needed to achieve the
effective manner with knowledge of the MATS Final Rule, as well as any other emission standard applicable to the
levels of Hg control achieved in each of available control strategies that the EPA non-lignite-firing EGUs (1.2 lb Hg/
the installed controls and, if chemical should consider and their costs. TBtu).
TABLE 8—HG EMISSIONS AND CONTROL PERFORMANCE OF LIGNITE-FIRED EGUS IN 2021

Est 2021 Hg
Est Hg control Est Hg control Est 2021 Hg Est 2021 Hg
Hg inlet 2021 Hg outlet control at 1.2
Plant name at 4.0 lb/TBtu at 1.2 lb/TBtu inlet control
(lb/TBtu) (lb/TBtu) lb/TBtu
(%) (%) (lb/TBtu) (%) (%)
Antelope Valley #1 ....... 7.81 48.8 84.6 7.76 2.87 63.0 84.5
Antelope Valley #2 ....... 7.81 48.8 84.6 7.76 2.74 64.6 84.5
Coal Creek #1 .............. 7.81 48.8 84.6 7.80 3.62 53.6 84.6
Coal Creek #2 .............. 7.81 48.8 84.6 7.80 3.89 50.2 84.6
Coyote .......................... 7.81 48.8 84.6 7.79 3.17 59.2 84.6
Leland Olds #1 ............. 7.81 48.8 84.6 7.79 2.51 67.8 84.6
Leland Olds #2 ............. 7.81 48.8 84.6 7.79 3.02 61.3 84.6
Milton R Young #1 ....... 7.81 48.8 84.6 7.78 3.23 58.4 84.6
Milton R Young #2 ....... 7.81 48.8 84.6 7.79 3.20 58.9 84.6
Spiritwood Station ........ 7.81 48.8 84.6 5.03 1.86 63.1 76.1
Limestone #1 ............... 14.88 73.1 91.9 6.24 0.94 84.9 80.8
Limestone #2 ............... 14.88 73.1 91.9 6.20 1.59 74.4 80.7

Major Oak #1 ............... 14.65 72.7 91.8 14.62 1.24 91.5 91.8
Major Oak #2 ............... 14.65 72.7 91.8 14.65 1.31 91.1 91.8
Martin Lake #1 ............. 14.65 72.7 91.8 8.22 2.32 71.8 85.4
Martin Lake #2 ............. 14.65 72.7 91.8 8.13 2.99 63.2 85.2
Martin Lake #3 ............. 14.65 72.7 91.8 7.85 3.04 61.3 84.7
37 Discussion of how these assumptions were
developed for use in the EPA’s IPM modeling is

available in Chapter 7 of the IPM Documentation.
TABLE 8—HG EMISSIONS AND CONTROL PERFORMANCE OF LIGNITE-FIRED EGUS IN 2021—Continued

Est 2021 Hg
Est Hg control Est Hg control Est 2021 Hg Est 2021 Hg
Hg inlet 2021 Hg outlet control at 1.2
Plant name at 4.0 lb/TBtu at 1.2 lb/TBtu inlet control
(lb/TBtu) (lb/TBtu) lb/TBtu
(%) (%) (lb/TBtu) (%) (%)
Oak Grove #1 .............. 14.88 73.1 91.9 14.60 2.01 86.2 91.8
Oak Grove #2 .............. 14.88 73.1 91.9 14.88 2.59 82.6 91.9
Red Hills #1 ................. 12.44 67.8 90.4 12.40 1.33 89.3 90.3
Red Hills #2 ................. 12.44 67.8 90.4 12.40 1.35 89.1 90.3
San Miguel ................... 14.65 72.7 91.8 14.62 2.81 80.8 91.8
As can be seen in the table, all lignite- proposing to revise the Hg emission Low-rank coals such as lignite and
fired EGUs are estimated to meet the standard for lignite-fired EGUs to 1.2E– subbituminous generally have lower
current standard by achieving a level of 06 lb/MMBtu. chlorine contents than higher rank coals
control of less than 75 percent. The such as bituminous coal.43
i. Both Lignite and Subbituminous Coal
average reported 2021 Hg emission rate
Are Low Rank Coals With Low Halogen As mentioned earlier, the halogen
for lignite-fired EGUs located in North
Content content of the coal—especially
Dakota was 3.0 lb Hg/TBtu with an
average control of 83.7 percent. The Coal is classified into four main types, chlorine—largely influences the
average reported 2021 Hg emission rate or ranks: 40 anthracite, bituminous, oxidation state of Hg in the flue gas
for lignite-fired EGUs located in Texas subbituminous, and lignite. The ranking stream. As a result, the halogen content
and Mississippi was 2.0 lb Hg/TBtu depends on heating value of the coal. of the coal directly influences the ability
(with an average control of 88.2 Anthracite has the highest heating value to capture and contain the Hg before it
percent). of all ranks of coal and is mostly used is emitted into the atmosphere. As
by the metals industry (it is rarely using explained earlier, ash from lignite and
f. Proposed Revision of the Hg Emission for power production). Anthracite subbituminous coals tends to be more
Standard for Lignite-Fired EGUs accounted for less than 1 percent of the alkaline (relative to that from
Several commenters have provided coal mined in the U.S. in 2021. bituminous coal) due to the lower
information on new developments in Hg Bituminous coal is also considered a amounts of sulfur and halogen and the
control technology. One commenter 38 ‘‘high rank coal’’ because of its higher presence of a more alkaline and reactive
indicated that improvements in halogen heating value. It is the most abundant (non-glassy) form of calcium in the ash.
and ACI technologies have significantly rank of domestic coal and accounted for The natural alkalinity of the
lowered the costs of those pollution about 45 percent of total U.S. coal
subbituminous and lignite fly ash can
control systems. The use of production in 2021. Bituminous coal is
effectively neutralize the limited free
computational fluid dynamics and used to generate electricity and in other
physical modeling has also improved industries. halogen in the flue gas and prevent
pollutant capture and reduced sorbent Subbituminous coal and lignite are oxidation of the Hg0. This makes control
consumption. The commenter further referred to as ‘‘low rank coals.’’ They of Hg from both subbituminous coal-
noted that ACI systems operate more both have lower heating values than fired EGUs and lignite-fired EGUs more
reliably, and many users utilize bituminous coal. Subbituminous coal challenging than the control of Hg from
technology to improve the dispersion of accounted for about 46 percent of total bituminous coal-fired EGUs. However,
sorbents in flue gas for better U.S. coal production in 2021, with the because control strategies and
performance. After reviewing the vast majority produced in the Powder technologies were developed to
available literature and other studies River Basin (PRB) of Wyoming and introduce halogens to the flue gas
and available information, the Montana. Lignite has the lowest energy stream, EGUs firing subbituminous
assumptions made regarding Hg control content of all coal ranks. Lignite coals have been able to meet the 1.2 lb/
in the 2012 MATS Final Rule, and the accounted for about 8 percent of total TBtu emission standard in the 2012
information obtained from compliance U.S. coal production in 2021.41 About MATS Final Rule. As mentioned earlier,
reports and the 2022 CAA section 114 56 percent was mined in North Dakota EGUs firing subbituminous coal in 2021
information collection, the EPA has (Fort Union lignite) and about 36 emitted Hg at an average annual rate of
determined that there are developments percent was mined in Texas (Gulf Coast 0.6 lb Hg/TBtu with measured values as
in practices, processes, and control lignite). low as 0.1 lb/TBtu. Clearly EGUs firing
technologies since 2012 that warrant Chlorine is the most abundant
halogen in coal. Bromine may also be subbituminous coal have found control
consideration of revising the Hg options to meet—and exceed—the 1.2
standards for lignite-fired EGUs. As present in coal but is typically in much
lower concentrations than chlorine.42 lb/TBtu emission standard despite the
explained below, the EPA has further challenges presented by the low natural
determined that available controls and halogen content of the coal and
the EPA establish that obtaining better information
methods of operation that will allow
on performance of controls can provide the basis for production of difficult-to-control

lignite-fired EGUs to meet the same Hg updates to standards under a technology review. elemental Hg vapor in the flue gas
emission standard that is being met by 40 ‘‘Coal Explained, Types of Coal’’ Energy
EGUs firing on non-lignite coals, and Information Administration, available at stream.

that the costs of doing so are www.eia.gov/energyexplained/coal and in the
rulemaking docket.
reasonable.39 Therefore, the EPA is 41 EIA Annual Coal Report 2021, October 2022,
https://www.eia.gov/coal/annual/pdf/acr.pdf.
38 See EPA–HQ–OAR–2018–0794–1171. 42 See Figure 5 in the U.S. Geological Survey Combustion’’ available at https://pubs.usgs.gov/fs/
39 As discussed in section V.B above, prior CAA publication ‘‘Mercury and Halogens in Coal—Their 2012/3122/pdf/FS2012-3122_Web.pdf.
section 112(d)(2) technology reviews conducted by Role in Determining Mercury Emissions From Coal 43 Id.
ii. The Hg Content of Fort Union Lignite in 2021, firing significant amounts of needed to control their Hg emissions to
and PRB Subbituminous Coal Are subbituminous coal. Firing high levels less than 92 percent to meet an emission
Similar of non-lignite coal (in some cases greater standard of 1.2 lb/TBtu. Based on this,
As can be seen in Table 8 above, for than 99 percent non-lignite coal), while we expect that the units could meet the
the 2012 MATS Final Rule, the EPA remaining subject to the less stringent proposed, more stringent, emission
estimated the Fort Union lignite-fired Hg emission standard for the standard of 1.2 lb/TBtu by utilizing
EGUs inlet Hg concentration at up to 7.8 subcategory of lignite-fired EGUs seems brominated activated carbon at the
lb/TBtu and estimated the inlet Hg to fit the scenario that the EPA injection rates suggested in the beyond-
concentration of subbituminous coal- expressed concern about in the 2012 the-floor memo 46 from the 2012 MATS
fired EGUs at up to 8.65 lb/TBtu. These MATS Final Rule preamble—that Final Rule.
values are very similar to results from a ‘‘sources to potentially meet the To determine the cost-effectiveness of
published study that found the average definition by combusting very small that strategy, we calculated the
Hg concentration of Fort Union lignite amounts of low rank virgin coal incremental cost-effectiveness (cost per
and PRB subbituminous coals to be very [lignite].’’ See 77 FR 9379. lb of Hg controlled) for a model 800 MW
similar. The study found that the Fort lignite-fired EGU. We calculated the
iv. The Proposed More Stringent Hg incremental cost of injecting non-
Union lignite samples contained an Emission Standard Can Be Achieved,
average of 8.5 lb/TBtu and the PRB brominated activated carbon sorbent at
Cost-Effectively, Using Available a sufficiently large injection rate of 5.0
subbituminous coal samples contained Control Technology
an average of 7.5 lb/TBtu.44 Despite the lb/MMacf to achieve an emission rate of
For the 2012 MATS Final Rule, the 1.2 lb/TBtu versus the cost to meet an
similarities in Hg content, halogen
EPA calculated beyond-the-floor costs emission rate of 4.0 lb/TBtu using non-
content, and alkalinity between Fort
for Hg controls by assuming injection of brominated activated carbon sorbent at
Union lignite and PRB subbituminous
brominated activated carbon at a rate of an emission rate of 2.5 lb/MMacf. For an
coal, EGUs firing subbituminous coal in
3.0 lb/MMacf for units with ESPs and 800 MW lignite-fired EGU, the
2021 emitted Hg at an average annual
injection rates of 2.0 lb/MMacf for units incremental cost effectiveness was
rate of 0.6 lb Hg/TBtu while those firing
with baghouses (also known as FF). Yet, $8,703 per incremental lb of Hg
on Fort Union lignite emitted Hg at an
in responses to the CAA section 114 removed. The actual cost-effectiveness
average annual rate of 3.0 lb Hg/TBtu.
information survey, only one facility is likely lower than this value as it is
While the EGUs firing Fort Union lignite
(Oak Grove) explicitly indicated use of unlikely that sources will need to inject
at an average emission rate of 3.0 lb Hg/
brominated activated carbon. Oak Grove brominated activated carbon sorbent at
TBtu are complying with the 2012
units #1 and #2 (both using FF for PM rates as high as 5.0 lb/MMacf (the Oak
MATS Final Rule emission standard of
control) reported use of brominated Grove units were injecting less than 0.5
4.0 lb Hg/TBtu, it is difficult to justify
activated carbon at an average injection lb/MMacf) and is well below the cost
why those units should not meet a
rate of less than 0.5 lb/MMacf for that the EPA has found to be acceptable
similar level of Hg control as that of the
operation at capacity factor greater than in previous rulemakings (e.g., $27,500/
EGUs firing PRB subbituminous coal
70 percent. The Oak Grove units fired, lb Hg was proposed to be cost-effective
given the similarities between the two
in 2021, using mostly refined coal.45 for the Primary Copper RTR (87 FR
fuels—especially the similarities in Hg
That injection rate is considerably less 1616); approximately $27,000/lb Hg was
content, halogen content, and alkalinity.
than the 2.0 lb/MMacf assumed. found to be cost-effective in the beyond-
iii. The Hg Content of Gulf Coast Lignite From the CAA 114 information the-floor analysis supporting the 2012
Is Greater Than That of Fort Union survey, the average injection rate MATS Final Rule 47).
Lignite; and Several Lignite-Fired EGUs reported for non-halogenated sorbents In summary, the EPA is proposing to
in Texas Have Co-Fired Significant was 2.5 lb/MMacf. The average sorbent revise the Hg emission standard for
Quantities of Subbituminous Coal injection rate ranged from 10–65 percent lignite-fired EGUs from 4.0E–06 lb/
The Hg content of Gulf Coast lignite of the maximum design sorbent MMBtu to 1.2E–06 lb/MMBtu, which is
tends to be higher than that of the Fort injection rate (the average was 36 the same Hg emission limit that non-
Union lignite. As can be seen in Table percent of the maximum design rate). As lignite-fired EGUs must meet. We are
8 above, for the 2012 MATS Final Rule, mentioned earlier, most sources utilized proposing to revise this emission
the EPA estimated the inlet Hg a control strategy of sorbent injection standard while recognizing that Hg from
concentration for Gulf Coast lignite-fired coupled with chemical (usually the combustion of lignite is challenging
EGUs at an average inlet Hg halogenated) additives. In the beyond- to capture because of the lack of
concentration of up to 14.9 lb/TBtu (as the-floor analysis in the 2012 MATS naturally occurring halogen in the fuel
compared to average inlet Hg Final Rule, we noted that the results and because of the natural alkalinity of
concentrations of up to 7.8 lb/TBtu for from various demonstration projects the resulting fly ash. However, Hg from
Fort Union lignite). Despite the higher suggests that greater than 90 percent Hg the combustion of subbituminous coal is
Hg content in Gulf Coast lignite, EGUs control can be achieved at lignite-fired similarly challenging to capture for the
permitted as lignite-fired had, in 2021, units using brominated activated carbon same reasons. Yet, EGUs firing
an average Hg emission rate of 2.0 lb/ sorbent at an injection rate of 2.0 lb/ subbituminous coal in 2021 emitted Hg
TBtu—which was lower than the 2021 MMacf for units with installed FFs for at an average rate of 0.6 lb/TBtu and
average emission rate of EGUs firing PM control and at an injection rate of some as low as 0.1 lb/TBtu. From the
Fort Union lignite (at 3.0 lb/TBtu). This 3.0 lb/MMacf for units with installed CAA section 114 information survey,
is due, in part, because some EGUs in ESPs for PM control. As shown in Table very few lignite-fired EGUs are using the
Texas that are permitted as lignite-fired 8 above, all units (in 2021) would have control technology that the EPA
units (and thus subject to the Hg identified as the most effective for Hg
45 EIA form 923 does not specify the rank of coal
emission standard of 4.0 lb/TBtu) were, that is ‘‘refined’’ in boiler or generator fuel data. For
control in the 2012 MATS Final Rule,
this technology review, the EPA has assumed that
44 ‘‘Mercury in North Dakota lignite’’, Katrinak, 46 See Docket ID No. EPA–HQ–OAR–2009–0234–
facilities reporting the use of refined coal have
K.A.; Benson, S.A.; Henke, K.R.; Hassett, D.J.; Fuel utilized ‘‘refined lignite,’’ which was confirmed in 20130 at regulations.gov.
Processing Technology, 39, 35, 1994. EIA form 923 fuel receipts and costs. 47 Ibid.
brominated ACI, which many to use of refined coal and that any When the EPA finalized the 2012
demonstration projects have shown can negative health, ecological, and MATS Final Rule, the primary air
achieve Hg control of greater than 90 productivity effects associated with pollution control devices installed at
percent. Although we are not proposing bromine transfer to water effluent will EGUs for the control of acid gases were
to mandate the use of any particular be minimized or avoided, especially wet scrubbers (wet FGD), dry scrubbers
control technology, we have shown that given the EPA’s proposed zero- (dry FGD or spray dryer absorber, SDA),
use of brominated activated carbon discharge requirements under the Clean and reagent injection (at fluidized bed
sorbent injection can be used to cost- Water Act (88 FR 18824; March 29, combustors). These technologies are still
effectively meet the more stringent 2023). in wide use for acid gas HAP control.
emission. An additional acid gas control
4. No Revisions to Work Practice technology—dry sorbent injection
We also considered the energy Standards for Organic HAP
implications and non-air environmental (DSI)—was in limited use in the power
impacts of this proposed revision of the Following promulgation of the 2020 sector at the time the MATS rule was
Hg emission standard for lignite-fired Final Action, in which the EPA found finalized but has seen increased use
EGUs. We do not anticipate any energy no developments in new technology or since (approximately 20 percent of
implications from this proposed methods of operation that would result EGUs operating in 2021 utilized DSI for
revision as most units are already using in cost-effective emission reductions of acid gas control for one reason or
sorbent injection technology as part of organic HAP and thus did not revise the another).
the Hg control strategy and we do not work practice standards for organic A wet FGD scrubber uses an alkaline
HAP, the EPA received a petition for liquid slurry (usually a limestone or
project significant changes in unit
reconsideration that, in relevant part, lime slurry) to remove acidic gases from
operations as a result of the proposed
requested the EPA to reconsider work an exhaust stream. The acid gases react
revision. Regarding the non-air
practice standards for organic HAP.48 with the alkaline compounds in the
environmental impact, we anticipate
Our review of new technology and of slurry and are removed as scrubber
that there may be positive non-air
methods of operation conducted as part solids (e.g., CaSO3 or CaSO4) or may be
environmental impacts. The current
of this technology review proposal also captured due to their solubility in the
strategies employed by most lignite-
found no developments that would scrubber slurry. Most wet FGD
fired EGUs involve the injection of
result in cost-effective emission scrubbers have SO2 removal efficiencies
oxidizing halogen additives and,
reductions of organic HAP. Likewise, exceeding 90 percent and perform even
separately, injection of sorbent better for HCl and HF. Dry FGD
(typically non-brominated activated we are not proposing revisions to the
organic HAP work practice standards scrubbers (SDA) are an acid gas
carbon). Because homogeneous (gas- pollution control system where an
phase) oxidation of Hg0 is kinetically finalized in the 2012 MATS Final
Rule.49 The EPA acknowledges that it alkaline sorbent slurry is injected into
limited, most of the Hg0 oxidation is the flue gas stream to react with and
thought to occur as heterogeneous received a petition for reconsideration
from environmental organizations that, neutralize acid gases in the exhaust
(solid-phase) reactions resulting from stream forming a dry powder material
halogens or other oxidants attached to in relevant part, sought the EPA’s
reconsideration of organic HAP work which is then captured in a downstream
flue gas solids (e.g., unburned carbon, PM control device (usually an FF).
other). This is essentially a two-step practice standards, which the EPA
continues to review and will respond to Alkaline sorbent injection systems
process where the injected (or natural) (reagent injection) are also used in
halogen (chloride or bromide) must first in a separate action.50
fluidized bed combustors (FBC) and
attach to a flue gas solid and then 5. No Proposed Revisions to the Acid circulating fluidized bed (CFB) boilers
contact and react with gas-phase Hg0. Gas Standards for Coal-Fired EGUs for control of acid gases. In that use, the
The addition of sorbent that has already alkaline sorbent (usually powdered
been pre-halogenated (most often The EPA evaluated the use of control
technologies and strategies that are limestone) is injected into the
brominated) is more efficient as the first combustion chamber with the primary
step occurs prior to injection. This commonly used for control of acid gas
HAP (e.g., HCl, HF). These control fuel. Dry sorbent injection (DSI) is an
means that less bromine will be add-on air pollution control system in
unutilized and captured in a technologies and strategies include the
use of wet FGD scrubbers, spray drier which a dry alkaline powdered sorbent
downstream control device or (typically sodium- or calcium-based) is
potentially included in the plant water absorber (SDA) scrubbers, reagent
injection (for fluidized combustors), dry injected into the flue gas steam
effluent discharge. The EPA requests upstream of a PM control device to react
comment on its expectation that most sorbent injection (DSI), and use of low
sulfur or low halogen fuels. As with and neutralize acid gases in the
EGUs (including lignite-fired EGUs) will exhaust stream forming a dry powder
no longer use ‘‘refined coal’’ due to the described in section III of this preamble,
EGUs in six subcategories are subject to material that may be removed in a
expiration of the refined coal tax credit. primary or secondary PM control
The amount of Br on brominated numeric emission limits for acid gas
HAP (e.g., HCl, HF). Emission standards device. The EPA evaluated the use of
activated carbon is much less than that these control technologies (wet FGD
used to produce refine coal, and Br is for HCl serve as a surrogate for all acid
gas HAP, with an alternate standard for scrubbers, SDA, reagent injection, and
retained on the activated carbon sorbent DSI), and the strategic use of low sulfur
where it reacts with gas phase Hg and SO2 that may be used as a surrogate for
or low halogen fuels.
is captured by downstream control the acid gas HAP at coal-fired EGUs The EPA reviewed compliance data
devices. Thus, the EPA believes that with operational FGD systems and SO2 for SO2 and/or HCl, as shown in Figure
cross-media transfers of bromine to CEMS. 3 of the Technical Memo, showing
receiving waterbodies and emitted to 48 See Docket ID No. EPA–HQ–OAR–2018–0794–
EGUs with highest SO2 emissions in
the atmosphere, especially when wet 4565 at www.regulations.gov.
2021 to those with the lowest SO2
FGD is not employed, are not expected 49 See 40 CFR 63.9991, Table 3. emissions in 2021. Approximately two-
(or would certainly be lower) with the 50 See Docket ID No. EPA–HQ–OAR–2018–0794– thirds of coal-fired EGUs have
use of brominated sorbents as compared 4565 at www.regulations.gov. demonstrated compliance with the
alternative SO2 emission standard rather or dry FGD scrubbers, with units an independent comprehensive report
than the HCl emission limit. About one- utilizing sorbent injection emitting at to show acid gas emission controls had
third of EGUs have demonstrated slightly higher rates. The units that better performance and lower capital
compliance with the primary acid gas utilize DSI with an FF tend to have costs than the EPA assumed in the 2011
emission limit for HCl. And some lower HCl emissions than those that modeling (Docket ID No. EPA–HQ–
sources have reported emissions data utilize DSI with an ESP. This is an OAR–2018–0794–4962). That report
that demonstrates compliance with expected outcome as the filter cake on suggested control technology
either of the standards. The emission the FF provides great opportunity for improvements to acid gas controls to
rates for HCl that are shown in Figure contact with the gas phase acid gases. achieve revised HCl emission standards
3 of the Technical Memo distinguish Overall, the EPA has evaluated acid of 1.0E–03 lb HCl/MMBtu, 6.0E–04 lb
between EGUs that utilize some sort of gas emissions data from MATS-affected HCl/MMBtu, and 1.0E–05 lb HCl/
acid gas control system—which would EGUs and have determined that some MMBtu through addition of new DSI
be a wet FGD scrubber, a dry scrubber units have demonstrated compliance systems, upgrades to existing DSI
(an SDA), reagent injection or DSI—and with the primary HCl emission standard systems, upgrades to existing wet and
EGUs that do not have a wet FGD using acid gas control technologies (wet dry scrubbers, and, for the most
scrubber or an SDA and do not utilize FGD scrubbers, SDA, reagent injection, stringent options, installation of new
either reagent injection or DSI. All of the and DSI) and through the strategic use FFs. However, as mentioned earlier—
EGUs with no acid gas controls are units of low-halogen, high-alkalinity fuels. and as detailed further in the Technical
that were firing subbituminous coal and Other units have demonstrated Memo—it is not clear that such
were likely able to demonstrate compliance with acid gas emission improvements targeting acid gases
compliance with the HCl emission limits by meeting or exceeding the would result in corresponding
standard due to the low natural chlorine alternative surrogate SO2 emission reductions in HCl or HF emissions, as
content and high alkalinity of most standard. The average HCl emission emissions of HCl and HF are already
subbituminous coals. rates for units with add-on acid gas much easier to control than emissions of
All sources submit SO2 emissions controls was 4.0E–04 lb HCl/MMBtu SO2.
data to comply with other CAA which is approximately 80 percent
In summary, the EPA has not
requirements (e.g., the Acid Rain below the MATS HCl emission limit.
identified any new control technologies
Program). As mentioned earlier, some The average HCl emission rates for units
or any improvements to existing acid
sources submitted emissions data that with no add-on acid gas controls was
gas controls that would result in
demonstrates compliance with either 8.0E–04 lb HCl/MMBtu (approximately
additional cost-effective acid gas HAP
the HCl standard or the alternative SO2 60 percent below the MATS HCl
emission reductions from coal-fired
standard. The average SO2 emission rate emission limit). It is not clear that
EGUs and is, therefore, not proposing
for units at or below the alternative SO2 improvements in a wet or dry FGD
revisions to the acid gas emission
emission limit was 9.0E–02 lb SO2/ scrubber would result in additional HCl
MMBtu, which is approximately 55 emission reductions since HCl standards or for the surrogate SO2
percent below the SO2 emission limit of emissions are already much easier to emission standard. However, the EPA
2.0E–01 lb SO2/MMBtu. The average control than SO2 emissions. The EPA solicits comment on any new practices,
HCl emission rate for units does not have information on the processes, or technologies for control of
demonstrating compliance with the SO2 sorbent injection rates for DSI systems; acid gas HAP emissions, including any
standard but also reporting HCl so, we cannot assess whether increased information on whether increased
emissions was 4.0E–04 lb HCl/MMBtu, sorbent injection would result in sorbent injection rates (for sources using
which is approximately 80 percent additional HCl emission reductions. DSI or SDA controls) would result in
below the HCl emission limit of 2.0E– Units using DSI in combination with an additional HCl emission reductions, that
03 lb HCl/MMBtu. This result is ESP would almost certainly see could inform the potential for additional
consistent with the EPA’s rationale for improved performance if they were to cost-effective acid gas HAP emission
establishing the alternative SO2 replace the ESP with a FF. However, reductions from coal-fired EGUs.
emission limit—because HCl emissions that small incremental reduction in HCl 6. No Proposed Revisions to Standards
are much more easily controlled than emissions would come at a high cost for Continental Liquid Oil-Fired EGUs
SO2 emissions (HCl and HF are much and would certainly not be a cost-
more reactive and much more water effective option. The annual capacity factors of most
soluble than SO2), controlling emissions In the 2020 Technology Review, the continental liquid oil-fired units are
of SO2 using FGD controls very EPA concluded that ‘‘the existing acid low. Based on available data reported to
effectively controls emissions of HCl. gas pollution control technologies that the EIA and the EPA’s Clean Air
Note that an EGU may demonstrate are currently in use are well-established Markets Program Data (CAMPD), in
compliance with the acid gas surrogate and provide the capture efficiencies 2021 the average annual capacity factor
SO2 standard only if the unit has some necessary for compliance with the for liquid oil-fired units was 3 percent.
type of installed acid gas control and an promulgated MATS rule limits.’’ Additionally, there were only two
operational SO2 CEMS. Comments received during the 2020 continental liquid oil-fired units
The EPA looked further at the HCl Proposal did not provide any new identified with 2-year capacity factors
emissions of the EGUs operating in 2021 practices, processes, or control greater than 8 percent. Those two units
with and without acid gas controls. The technologies for acid gas control. One primarily fire natural gas but had heat
average emission rate of EGUs with no commenter noted that ‘‘in the short time input-based percentages of fuel oil firing
add-on acid gas control was 8.0E–04 lb since the RTR was finalized, there have that were about 16 percent in at least
HCl/MMBtu, which is 60 percent below been no developments in practices, one of the years from 2019 through 2021
the SO2 emission limit. processes, or control technologies, nor (i.e., slightly above the 15 percent that
The EPA looked closer at the relative any new technologies or practices for would qualify them as oil-fired units).
performance of acid gas controls for HCl the control of . . . acid gas HAP’’ Therefore, it is likely that there are very
emissions. The best performing EGUs (Docket ID No. EPA–HQ–OAR–2018– few continental liquid oil-fired units
tend to be those that utilize either wet 5121). Another commenter pointed to that would be outside of the definition
of the limited-use liquid oil-fired emissions during periods of limited Because of the low capacity factors of
subcategory. operation. the Hawaii oil-fired EGUs and the near-
Furthermore, for the continental term retirement dates of most of the
7. No Proposed Revisions to Standards
liquid oil-fired units with available data Puerto Rico liquid oil-fired EGUs and
for Non-Continental Liquid Oil-Fired
that are likely limited-use units, the plans for a transition to greater use of
EGUs
cumulative percentage of heat input natural gas for the remaining boilers, the
from residual fuel oil in 2021 was 32 Hawaiian Electric Company (HECO) EPA is not proposing to revise emission
percent, the heat input of distillate fuel operates 12 liquid oil-fired boilers at its standards for non-continental oil-fired
oil was 4 percent, and the heat input Waiau Generating Station (Pearl City, EGUs.
from natural gas was 64 percent. HI) and at its Kahe Generating Station However, the EPA seeks comment on
Because the capacity factors of most (Kapolei, HI). Their average capacity whether the fPM surrogate emission
continental liquid oil-fired units are factor in 2021 was 29.6 percent (on a net standard is appropriate for these non-
low, and most combustion by those basis) and they fire on residual fuel oil. continental liquid oil-fired EGUs. As
units is using fuel (i.e., natural gas) with HECO has, in compliance reports, mentioned, the largest risks identified in
low metallic HAP emission rates, the reported fPM emission rates to the EPA the 2020 RTR were associated with
EPA is not proposing changes to the that are below the fPM emission rate of nickel emissions from residual oil-fired
total HAP metals (which includes Hg), 3.0E–02 lb/MMBtu. EGUs located in Puerto Rico. The EPA
In Puerto Rico, there are 14 liquid oil- solicits comment on eliminating or
nor to the standards for the individual
fired MATS-affected EGUs (3,552 MW revising the fPM standard for existing
HAP metals, nor to the HAP metal
total capacity) at four separate facilities non-continental sources, and, instead,
surrogate fPM emission standard for
operated by the Puerto Rico Electric requiring these EGUs to comply with
continental liquid oil-fired electricity
Power Authority (PREPA). The EGUs the existing emission limits for the
generating units.
operate using residual fuel oil and do individual metals, including nickel. In
However, given there have been not currently have any emission
several recent temporary and localized addition, the EPA also seeks comment
controls for NOX, PM or SO2. At least on the appropriateness of including new
increases in oil combustion at two of the units have dual fuel HAP standards for EGUs in Puerto Rico
continental liquid oil-fired EGUs during capabilities and have operated on high and Hawaii, as well as other non-
periods of extreme weather conditions, levels of natural gas. There is limited continental U.S. areas, such as Guam
such as the 2023 polar vortex in New stack testing data available, but testing and the Virgin Islands, and the means
England, the EPA seeks comment on done in 2021 and 2022 indicated fPM of demonstrating compliance with the
whether the current definition of the emission rates ranging from 2.6E–02 lb/ new HAP standards. For example, the
limited-use liquid oil-fired subcategory MMBtu to 2.9E–02 lb/MMBtu, a range EPA seeks input on whether, in order to
remains appropriate or if, given the that is just below the fPM emission rate reduce HAP emissions and associated
increased reliance on oil-fired of 3.0E–02 lb/MMBtu. risks in these places, oil-fired EGUs
generation during periods of extreme As mentioned earlier in section IV.A should be required to switch from
weather, a period other than the current of this preamble summarizing the 2020 residual oil to cleaner fuels, or to switch
24-month period or a different threshold Residual Risk Review, the results of the to cleaner fuels after a certain number
would be more appropriate for the chronic inhalation cancer risk of hours of operation, or should be
current definition. The EPA also seeks assessment based on actual emissions subject to an annual limit of residual oil
comment on the appropriateness of indicated that the estimated maximum firing. The EPA solicits comment on
including new HAP standards for EGUs individual lifetime cancer risk (cancer whether compliance with a HAP metal
subject to the limited use liquid oil-fired MIR) was 9-in-1 million, with nickel emission limit could be demonstrated
subcategory, as well as on the means of emissions from oil-fired EGUs at these by fuel sampling and analysis. The EPA
demonstrating compliance with the new four facilities in Puerto Rico as the solicits comment on the need for
HAP standards. For example, in order to major contributor to the risk. The total additional, cost-effective methods to
reduce HAP emissions during periods of estimated cancer incidence from this minimize HAP emissions in non-
extreme weather conditions, it may be source category was 0.04 excess cancer continental states and territories—
appropriate for limited-use liquid oil- cases per year, or one excess case in including Hawaii, Puerto Rico, the U.S.
fired EGUs to use distillate fuel oil every 25 years. Approximately 193,000 Virgin Islands, and Guam. We solicit
instead of residual oil, or to switch from people were estimated to have cancer comment on any special
residual oil to cleaner fuels after a risks at or above 1-in-1 million from considerations—including the
certain number of hours of operation, or HAP emitted from the facilities in this availability of clean fuels such as
to be subject to an annual or seasonal source category. The estimated distillate fuel oil and natural gas—in
limit of residual oil firing. The EPA maximum chronic noncancer TOSHI for non-continental areas.
solicits comment on each of these the source category was 0.2
options. (respiratory), which was driven by 8. No Proposed Revisions to Standards
The EPA also solicits comment on emissions of nickel and cobalt from the for IGCC EGUs
establishing a HAP emission limit on oil-fired EGUs. The EPA is aware of two existing
liquid oil-fired EGUs (including those in Since these oil-fired EGUs do not IGCC facilities that meet the definition
the limited-use subcategory and those have installed control devices for HAP of an IGCC EGU. The Edwardsport
located in non-continental areas) where metals (PM controls), there is no Power Station, located in Knox County,
compliance would be demonstrated opportunity to improve their Indiana, includes two IGCC EGUs that
through fuel sampling and analysis. The performance in the same ways the EPA had 2021 average capacity factors of
EPA seeks comment from the regulated found available to some coal-fired approximately 85 percent and 67
community, citizens, and regulatory EGUs. PREPA has recently proposed percent. The Polk Power Station,
authorities on the need for a revision to near-term retirement dates (by 2026) for located in Polk County, Florida, had a
the limited-use oil-fired subcategory 10 of the 14 oil-fired EGUs with two of 2021 average capacity factor of
definition and on additional, cost- the other four remaining boilers burning approximately 70 percent, but burned
effective methods to minimize HAP mostly natural gas. only natural gas in 2021.
While this subcategory has a less Electric Utility Steam Generating Units promulgated startup work practices.51
stringent fPM standard of 4.0E–02 lb/ and Standards of Performance for In Chesapeake Climate Action Network
MMBtu (as compared to that of coal- Fossil-Fuel-Fired Electric Utility, v. EPA, the D.C. Circuit remanded the
fired EGUs), recent compliance data Industrial-Commercial-Institutional and alternative work practice standard for
indicates fPM emissions well below the Small Industrial-Commercial- startup and shutdown to the EPA for
most stringent standard option of 6.0E– Institutional Steam Generating Units (79 reconsideration based on a petition for
03 lb/MMBtu that was evaluated for FR 68777; November 19, 2014), the EPA reconsideration from environmental
coal-fired EGUs. Since there are only took final action on its reconsideration groups. 952 F.3d 310 (D.C. Cir. 2020). In
two IGCC EGU facilities, and the EPA is of the startup and shutdown provisions this action, and in conjunction with the
unaware of any developments in the by adding an alternative work practice EPA’s authority pursuant to CAA
HAP emission controls used at IGCC standard for startup periods. That section 112(d)(6), the EPA is granting in
units, the EPA is not proposing to revise alternative work practice standard, part petitions for reconsideration which
any of the emission standards for this referred to as paragraph (2) of the sought the EPA’s review of startup and
subcategory. However, the EPA is definition of ‘‘startup’’, required clean shutdown provisions.52 As part of our
proposing that the affected facilities fuel use to the maximum extent obligation to address the remand on this
must install a PM CEMS to demonstrate possible, operation of PM control issue, we reviewed the information
compliance with the existing fPM limit. devices within 1 hour of introduction of available to us. As discussed below, that
Further, the EPA solicits comment on primary fuel (i.e., coal, residual oil, or information shows that the conditions
cost-effective methods to achieve solid oil-derived fuel) to the EGU, contained in the alternative work
additional HAP emission reductions practice standard do not represent what
collection and submission of records of
from this subcategory. the best performers are able to do;
clean fuel use and emissions control
D. What other actions are we proposing, device capabilities and operation, as moreover, as a practical matter, few
and what is the rationale for those well as adherence to applicable EGUs have chosen to use the alternative
actions? numerical standards within 4 hours of work practice standard.
In addition to the proposed actions the generation of electricity or thermal The EPA was able to identify 14 EGUs
described above, we are proposing energy for use either on site or for sale with the ability to generate up to 8.4 GW
additional revisions to the NESHAP. over the grid (i.e., the end of startup) that chose to use the alternative work
and to continue to maximize clean fuel practice for startup periods. As shown
1. Startup Requirements use throughout that period. The EPA in Table 9 below, six of those EGUs with
In the Reconsideration of Certain provided this alternative work practice the ability to generate up to 3.2 GW
Startup/Shutdown Issues: National because many commenters asserted it have retired and one of those EGUs with
Emission Standards for Hazardous Air would be difficult, if not impossible, for the ability to generate up to 0.7 GW will
Pollutants From Coal- and Oil-Fired their EGUs to meet the already- retire by 2025.
TABLE 9—EGUS RELYING ON PARAGRAPH (2) OF THE DEFINITION OF ‘‘STARTUP’’

EGU name Unit ORIS code MW Notes Fuel
Prairie State Generating ................ 1 ........................ 55856 877 .................................... Bituminous.

Prairie State Generating ................ 2 ........................ 55856 877 .................................... Bituminous.
Brame Energy Center .................... Rodemacher 2 .. 6190 552 .................................... Subbituminous.
Brame Energy Center .................... Madison 3–1 ..... 6190 600 .................................... Petroleum coke, coal.
Brame Energy Center .................... Madison 3–2 ..... 6190 600 .................................... Petroleum coke, coal.
Dolet Hills ....................................... 1 ........................ 51 720 Retired 2021 .............. Lignite.
Sherburne ...................................... 3 ........................ 6090 938.7 Retires 2034 ............... Subbituminous.
Westwood ...................................... 1 ........................ 50611 36 .................................... Waste coal.
Centralia ......................................... BW21 ................ 3845 729.9 Retired 2020 .............. Subbituminous.
Centralia ......................................... BW22 ................ 3845 729.9 Retires 2025 ............... Subbituminous.
St Johns River ............................... 1 ........................ 207 679 Retired 2018 .............. Bituminous.
St Johns River ............................... 2 ........................ 207 679 Retired 2018 .............. Bituminous.
HMP&L Station 2 ........................... H1 ..................... 1382 200 Retired 2019 .............. Bituminous.
HMP&L Station 2 ........................... H2 ..................... 1382 200 Retired 2019 .............. Bituminous.
After the planned retirements in 2025, of the definition of ‘‘startup’’, as well as coal- and oil-fired EGUs that generated
just seven EGUs with the ability to their associated retirement dates as over 98 percent of electricity in 2022
generate up to 4.5 GW will remain; this reported to the Department of Energy’s have made the requisite adjustments,
represents less than 0.4 percent of EIA. Commenters, particularly owners whether through greater clean fuel
electrical generation from all affected or operators of affected EGUs, should capacity, better tuned equipment, better
sources and less than 1.7 percent of the provide us with corrected information trained staff, a more efficient or better
278 GW of coal-fired and other, non- as, or if, necessary. Despite comments
design structure, or a combination of

natural gas fossil-fired electrical from EGU owners or operators and their factors, to be able to meet the
generation available in 2022. We solicit industry representatives opposing use of requirements of paragraph (1) of the
comment on whether we have identified paragraph (1) of the definition of definition of ‘‘startup.’’
all of the EGUs relying on paragraph (2) ‘‘startup’’, the owners or operators of
51 See Assessment of Startup Period at Coal-Fired 52 See Docket ID No. EPA–HQ–OAR–2018–0794–
Electric Generating Units, available at Docket ID No. 4565 at www.regulations.gov; see also Chesapeake
EPA–HQ–OAR–2009–0234–20378. Climate Action Network v. EPA, 952 F.3d 310 (D.C.
Cir. 2020).
Consistent with the MACT emission emission limits renders the LEE option suitable or could be adjusted to account
standard setting requirement for using for non-Hg metals (individual and total) for appropriate detection levels for
the average of the best performing 12 obsolete and the EPA is proposing to ongoing compliance purposes. In
percent of sources to establish emission remove those standards as well. addition, were our proposal to remove
standards, we propose to remove the Removal of the non-Hg metals emissions non-Hg metals from the rule not
alternative work practice standards, i.e., limits simplifies the compliance finalized, very frequent emissions
those contained in paragraph (2) of the determination path for EGU owners or testing, perhaps on the order of weekly,
definition of ‘‘startup’’, from the rule. As operators and reduces the amount of might be able to provide more
demonstrated by the majority of EGUs regulatory text, making the rule clearer information on compliance status.
currently relying on the work practice yet continuing to ensure that non-Hg While not continuous, as provided by
standards in paragraph (1) of the metals emissions remain below limits CEMS, such information would be more
definition of ‘‘startup’’, we believe such on an ongoing basis, particularly when frequent than provided by the quarterly
a change is achievable by all EGUs; the fPM is measured as proposed with emissions testing required by the rule.
further, we expect such a change would PM CEMS, given that non-Hg metals We solicit comment on appropriate
result in little to no additional emissions provided for one EGU are means to determine compliance with
expenditure since the additional obtained via quarterly stack testing. We non-Hg metals emission limits,
recordkeeping and reporting provisions solicit comment on the number of EGUs provided our proposed approach—
associated with the work practice that currently rely on non-Hg metals removal of non-Hg metals emission
standards of paragraph (2) of the emissions measurement for MATS limits—is not finalized. Please include
definition of ‘‘startup’’ were more compliance purposes; to the extent that in your comments information related to
expensive than the requirements of other EGU owners or operators rely on the frequency of collected data, the
paragraph (1) of the definition of non-Hg metals emissions for compliance continuity of data supplied by your
‘‘startup.’’ We solicit comment on our purposes, please be sure to identify each suggested means of compliance, and
proposal to remove the work practice EGU, its nameplate generating capacity, initial and ongoing annual costs of your
standards of paragraph (2) of the its anticipated or announced retirement suggested means of compliance.
definition of ‘‘startup.’’ date (if applicable), and its Office of
3. Removing Use of PM CPMS for
Regulatory Information Systems (ORIS)
2. Removing Non-Hg Metals Limits Compliance Determinations
Code. We solicit comment on our
The current MATS rule contains proposal to remove the non-Hg metals Use of PM CPMS for compliance
individual and total non-Hg metals emission limits from all existing MATS- purposes appears to be limited to four
emissions limits, as well as fPM affected EGUs. EGUs at one site in South Carolina, and
emission limits. Those fPM emission If we were to change our position by these EGUs account for less than 0.5
limits serve as alternative emission deciding against removing the non-Hg percent of all EGUs in operation.
limits because fPM was found to be a metals emission limits from MATS and According to submitted reports, each of
surrogate for either individual or total if our proposal to revise the fPM the EGUs relies on an instrument (Sick
non-Hg metals emissions. As explained emission limits was accepted, we would Maihak RWE–200) which provides a
and used above to quantify individual develop non-Hg emission limits by milliamp signal that is used to develop
and total non-Hg metals reductions from multiplying the revised fPM emission an ongoing operating limit; this
our proposed fPM emission limit limit by each individual (or total) non- instrument is advertised by its maker to
revision, the relationship between Hg PM ratio identified in the be able to serve as a PM CEMS with
individual and total non-Hg metals and aforementioned Emission Factor little to no modification, meaning that
fPM was determined by EGU fuel type Development for RTR Risk Modeling the instrument can provide direct
and control device using data collected Dataset for Coal- and Oil-fired EGUs measurement of fPM in terms of the
by the 2010 ICR.53 While EGU owners memorandum.55 The resulting values emission standard—pounds per million
or operators have the ability to use would become the individual non-Hg BTU. Given that PM CPMS use costs
individual or total non-Hg metals metals emission limits; their sum would more than PM CEMS use, that PM
emissions as the compliance method for become the total non-Hg metals CPMS does not provide continuous
the 358 EGUs when this action takes emission limit. We solicit comment on values in terms of the emission
effect and with generation of at least 25 our proposed approach to develop non- standard, that PM CPMS is rarely in use
MW,54 we are aware of just one owner Hg metals emission limits in the event among EGUs, and that the existing PM
or operator who provides non-Hg metals that our preferred approach—removing CPMS can be used as PM CEMS, we
data—both individual and total—along the non-Hg metals emission limits—is propose to remove the ability to use PM
with fPM data for compliance purposes not selected. Note that should our CPMS for compliance purposes in
for one waste coal-fired EGU with proposed approach to remove non-Hg MATS. The EPA solicits comment on
generating capacity of 46.1 MW. Given metals emission limits from MATS not the use of PM CPMS for compliance
that owners or operators of the other be finalized, we would need to adjust purposes; to the extent there are other
EGUs applicable to MATS have chosen the compliance determination method EGU owners or operators using PM
to demonstrate compliance with only because the current quarterly emissions CPMS, commenters should identify
the fPM emission limit, we propose to testing would not be consistent with the each EGU, along with its ORIS code and
remove the non-Hg metals emission continuous monitoring and compliance MW nameplate capacity, as well as the
limits—both individual and total—from determination method afforded by PM CPMS manufacturer and model in
MATS. Removal of the non-Hg metals acceptance of our proposal to require use. The EPA also solicits comment on
use of PM CEMS for compliance with the proposal to replace PM CPMS with
53 See Emission Factor Development for RTR Risk
the fPM emission limit. At least one PM CEMS for compliance use in MATS;
Modeling Dataset for Coal- and Oil-fired EGUs,
CEMS manufacturer offers a multi- when providing comments, please
available at https://www.regulations.gov at Docket provide detailed costs—including initial
ID No. EPA–HQ–OAR–2018–0794–0010. metals instrument that would be
54 Data obtained from the Emissions and instrument cost, installation cost, and
Generation Resource Integrated Database (eGRID), 55 See https://www.regulations.gov at Docket ID operating and maintenance costs—as
available at https://www.epa.gov/egrid. No. EPA–HQ–OAR–2018–0794–0010. well as a description of ongoing
operating activities from those EGUs the promulgation date. The EPA HAP, the proposed revision to the Hg
with existing PM CPMS used for considers 3 years to be as expedient as standard for lignite-fired EGUs to 1.2 lb/
compliance purposes. can be required considering the TBtu, the proposal to require PM CEMS
potential need to upgrade or replace to demonstrate compliance, and the
E. What compliance dates are we
monitoring systems. The EPA solicits removal of the startup definition
proposing, and what is the rationale for
comment on whether 3 years is an number two. The more stringent
the proposed compliance dates?
appropriate amount of time for EGUs to regulatory option examined in the RIA
The EPA is proposing to revise the upgrade or replace monitoring systems, tightens the proposed revision to the
fPM emission limit for existing coal- and whether quarterly stack testing fPM standard to 0.006 lb/MMBtu. The
fired EGUs and the Hg emission limit should continue to apply for EGUs that other three proposed amendments are
for lignite-fired EGUs. The EPA is have a binding commitment to not changed in the more stringent
proposing up to 3 years after the permanently cease operations in the regulatory option examined in the RIA.
effective date for EGUs subject to MATS near term. Additionally, the EPA Finally, the less stringent regulatory
to meet these new emission limits. proposes to remove fPM and the total option examined in the RIA assumed
However, the EPA solicits comment on and individual non-Hg HAP metals from the fPM and Hg limits remain
whether more than 1 year is needed to the LEE program no later than 3 years unchanged and examines just the
comply considering the potential need after the promulgation date to align with proposed PM CEMS requirement and
to upgrade control systems. In addition, the proposed compliance method of PM removal of startup definition number
the EPA is proposing that affected EGUs CEMS. Lastly, the EPA is proposing to two.
demonstrate compliance with the fPM remove the alternative work practice
emission limit using PM CEMS, A. What are the affected sources?
standard in paragraph (2) of the
removing the alternative compliance definition of ‘‘startup.’’ The EPA The EPA estimates that there are 302
options. Sources must demonstrate that proposes that affected sources must coal- and 56 oil-fired EGUs that will be
compliance has been achieved, by utilize paragraph (1) of the definition of subject to the MATS rule by the
conducting the required performance ‘‘startup’’ as specified in 40 CFR part 63, compliance date.
tests, and other activities as specified in subpart UUUUU, no later than 180 days B. What are the air quality impacts?
40 CFR part 63, subpart UUUUU, after the effective date.
including a minimum sampling The EPA estimated emissions
collection time of 3 hours per run, no VI. Summary of Cost, Environmental, reductions under the proposed rule for
later than 3 years after the promulgation and Economic Impacts the years 2028, 2030, and 2035 based
date. To demonstrate initial compliance In accordance with E.O. 12866 and upon IPM projections. The EPA also
using PM CEMS, the initial performance 13563, the guidelines of OMB Circular used IPM to estimate emissions
test consists of 30-boiler operating days. A–4, and EPA’s Guidelines for reductions for the more stringent
If the PM CEMS is certified prior to the Preparing Economic Analyses,56 the regulatory option examined in the RIA.
compliance date, the test begins with EPA prepared an RIA for this proposal. The less stringent regulatory option
the first operating day on or after that The RIA analyzes the benefits and costs presented in the RIA has no quantified
date. If the PM CEMS is not certified associated with the projected emissions emissions reductions associated with
prior to the compliance date, the test reductions under the proposed the proposed requirements for PM
begins with the first operating day after requirements, a less stringent set of CEMS and the removal of startup
certification testing is successfully requirements, and a more stringent set definition number two that constitute
completed. Continuous compliance of requirements to inform the EPA and the less stringent regulatory option
with the revised fPM emission limit is the public about these projected presented in the RIA.
required to be demonstrated on a 30- impacts. The emissions reduction estimates
boiler operating day rolling average We start this section of the preamble presented in the RIA include reductions
basis, defined in 40 CFR 63.10021(b), as describing how the RIA for this in pollutants directly targeted by this
the arithmetic average emissions rates proposed rule structured the proposed rule, such as Hg, and changes in other
over the last continuous 30 days and less and more stringent regulatory pollutants emitted from the power
provided the boiler was operating. The options in the RIA. The proposed sector as a result of the compliance
EPA proposes to remove the use of PM regulatory option in the RIA includes actions projected under this proposed
CPMS for compliance determinations the proposed revision to the fPM rule. Table 10 presents the projected
and the non-Hg metal emission limits— standard to 0.010 lb/MMBtu, in which emissions reductions under the
both individual and total—3 years after fPM is a surrogate for non-Hg metal proposed rule.
TABLE 10—PROJECTED EGU EMISSIONS IN THE BASELINE AND UNDER THE PROPOSED RULE: 2028, 2030, AND 2035
Emissions reductions
Year Less stringent More stringent

Proposed rule regulatory regulatory
option option
Hg (lbs.)
2028 ............................................................................................................................................. 62.0 0.0 208.0

2030 ............................................................................................................................................. 67.0 0.0 169.0
2035 ............................................................................................................................................. 82.0 0.0 168.0
56 U.S. EPA (2014). Guidelines for Preparing U.S. Environmental Protection Agency, Office of Policy, National Center for Environmental
Economic Analyses. U.S. EPA. Washington, DC, Economics.
TABLE 10—PROJECTED EGU EMISSIONS IN THE BASELINE AND UNDER THE PROPOSED RULE: 2028, 2030, AND 2035—
Continued
Emissions reductions
Year Less stringent More stringent

Proposed rule regulatory regulatory
option option
PM2.5 (thousand tons)
2028 ............................................................................................................................................. 0.4 0.0 2.6

2030 ............................................................................................................................................. 0.4 0.0 1.5
2035 ............................................................................................................................................. 0.8 0.0 1.3
SO2 (thousand tons)
2028 ............................................................................................................................................. 0.9 0.0 11.6

2030 ............................................................................................................................................. 0.5 0.0 0.3
2035 ............................................................................................................................................. 1.5 0.0 8.8
Ozone-season NOX (thousand tons)
2028 ............................................................................................................................................. 0.2 0.0 7.2

2030 ............................................................................................................................................. 0.4 0.0 5.1
2035 ............................................................................................................................................. 3.2 0.0 5.6
Annual NOX (thousand tons)
2028 ............................................................................................................................................. 0.4 0.0 18.1

2030 ............................................................................................................................................. 0.8 0.0 9.5
2035 ............................................................................................................................................. 3.4 0.0 8.7
HCl (thousand tons)
2028 ............................................................................................................................................. 0.0 0.0 0.2

2030 ............................................................................................................................................. 0.0 0.0 0.1
2035 ............................................................................................................................................. 0.0 0.0 0.1
CO2 (million metric tons)
2028 ............................................................................................................................................. 0.2 0.0 21.9

2030 ............................................................................................................................................. 0.8 0.0 8.7
2035 ............................................................................................................................................. 4.6 0.0 2.9
Section 3 of the RIA presents a implement the proposed requirements. current and planned transmission
detailed discussion of the emissions The compliance cost estimates were capacity, by setting limits to the ability
projections under the regulatory options developed with EPA’s Power Sector to transfer power between regions using
as described in the RIA. Section 3 also Modeling Platform v6 using IPM, a the bulk power transmission system.
describes the compliance actions that state-of-the-art, peer-reviewed dynamic, The model includes state-of-the-art
are projected to produce the emissions deterministic linear programming model estimates of the cost and performance of
reductions in Table 10. Please see of the contiguous U.S. electric power air pollution control technologies with
section VI.E of this preamble and sector. IPM provides forecasts of least respect to Hg and other HAP controls.
section 4 of the RIA for detailed cost capacity expansion, electricity
dispatch, and emission control We estimate the present value (PV) of
discussions of the projected health,
welfare, and climate benefits of these strategies while meeting electricity the projected compliance costs over the
emissions reductions. demand and various environmental, 2028 to 2037 period, as well as estimate
transmission, dispatch, and reliability the equivalent annual value (EAV) of
C. What are the cost impacts? constraints. IPM’s least-cost dispatch the flow of the compliance costs over
The power industry’s compliance solution is designed to ensure this period. All dollars are in 2019
costs are represented in this analysis as generation resource adequacy, either by dollars. Consistent with Executive Order
the change in electric power generation using existing resources or through the 12866 guidance, we estimate the PV and
costs between the baseline and policy construction of new resources. IPM EAV using 3 and 7 percent discount
scenarios. In simple terms, these costs addresses reliable delivery of generation rates. Table 11 presents the estimates of
are an estimate of the increased power resources for the delivery of electricity compliance costs across the regulatory
industry expenditures required to between the 78 IPM regions, based on options examined in the RIA.
TABLE 11—PROJECTED COMPLIANCE COSTS OF THE PROPOSED RULE, LESS STRINGENT ALTERNATIVE, AND MORE
STRINGENT ALTERNATIVE, 2028 THROUGH 2037
[Millions 2019$, discounted to 2023] a
3% Discount rate 7% Discount rate
Proposed Less stringent More stringent Proposed Less stringent More stringent
Present Value (PV) .................................. 330 ¥45 4,600 230 ¥31 3,400
Equivalent Annualized Value (EAV) ........ 38 ¥5.2 540 33 ¥4.5 490
a Values have been rounded to two significant figures.
The PV of the compliance costs for ongoing are estimated using IPM should help the EPA maintain an ample
the proposal, discounted at the 3 outputs and labor intensities, as margin of safety. Furthermore, there is
percent rate, is estimated to be about reported in section 5 of the RIA. the potential for reductions in Hg and
$330 million, with an EAV of about $38 non-Hg HAP emissions to enhance
E. What are the benefits?
million. At the 7 percent discount rate, ecosystem services and improve
the PV of the compliance costs of the Pursuant to E.O. 12866, the RIA for ecological outcomes, both of which can
proposal is estimated to be about $230 this action analyzes the benefits have positive economic effects although
million, with an EAV of about $33 associated with the projected emissions it is difficult to estimate these benefits
million. For a detailed description of reductions under this proposal to and consequently they have not been
these compliance cost projections, inform the EPA and the public about included in the set of quantified
please see section 3 of the RIA, which these projected impacts. This proposed benefits.
is available in the docket for this action. rule is projected to reduce emissions of The proposed rule is expected to
Hg and non-Hg metal HAP, PM2.5, SO2, reduce emissions of direct PM2.5, NOX,
D. What are the economic impacts? and SO2 nationally throughout the year.
NOX, and CO2 nationwide. The
This proposed action has energy potential impacts of these emissions Because NOX and SO2 are also
market implications. The power sector reductions are discussed in detail in precursors to secondary formation of
analysis supporting this action indicates section 4 of the RIA. ambient PM2.5, reducing these emissions
that there are important power sector The projected reductions in Hg would reduce human exposure to
impacts that are worth noting, although emissions should reduce the ambient PM2.5 throughout the year and
they are small relative to recent market- bioconcentration of methylmercury in would reduce the incidence of PM2.5-
driven changes in the sector and fish in nearby waterbodies. Subsistence attributable health effects. This
compared to some other EPA air fishing is associated with vulnerable proposed rule is also expected to reduce
regulatory actions for EGUs. populations, including minorities and ozone-season NOX emissions nationally.
There are several small national those of low socioeconomic status. In the presence of sunlight, NOX and
changes in energy prices projected to Methylmercury exposure to subsistence volatile organic compounds (VOCs) can
result from the proposed revisions to the fishers from lignite-fired units is below undergo a chemical reaction in the
MATS rule. Retail electricity prices are the current reference dose (RfD) for atmosphere to form ozone. Reducing
projected to increase in the contiguous methylmercury neurodevelopmental NOX emissions in most locations
U.S. by an average of less than 0.1 toxicity. The EPA considers exposures reduces human exposure to ozone and
percent in 2028, 2030, and 2035. In at or below the RfD are unlikely to be the incidence of ozone-related health
2035, the delivered natural gas price is associated with appreciable risk of effects, though the degree to which
anticipated to increase by less than 0.1 deleterious effects across the ozone is reduced will depend in part on
percent in response to the proposed population. However, no RfD defines an local concentration levels of VOCs.
rule. There are several other types of exposure level corresponding to zero The health effect endpoints, effect
energy impacts associated with the risk; moreover, the RfD does not estimates, benefit unit-values, and how
proposed revisions to MATS. Some represent a bright line above which they were selected, are described in the
coal-fired capacity, about 500 MW (less individuals are at risk of adverse effects. TSD titled Estimating PM2.5- and Ozone-
than 1 percent of operational coal In addition, there was no evidence of a Attributable Health Benefits, which is
capacity), is projected to become threshold for methylmercury-related referenced in the RIA for this action.
uneconomic to maintain by 2028. Coal neurotoxicity within the range of Our approach for updating the
production for use in the power sector exposures in the Faroe Islands study endpoints and to identify suitable
is not projected to change significantly which served as the primary basis for epidemiologic studies, baseline
by 2028. the RfD.57 Reductions in Hg emissions incidence rates, population
The short-term estimates for from lignite-fired facilities should demographics, and valuation estimates
employment needed to design, further reduce exposure to is summarized in section 4 of the RIA.
construct, and install the control methylmercury for subsistence fisher This proposed rule is projected to
equipment in the 3-year period before sub-populations located in the vicinity reduce PM2.5 and ozone concentrations,
the compliance date are also provided of these facilities. The projected producing a projected PV of monetized
using an approach that estimates reductions in non-Hg metal HAP may health benefits of about $1.9 billion,
employment impacts for the lead to reduced exposure to with an EAV of about $220 million
environmental protection sector based carcinogenic metal HAP for residential discounted at 3 percent.
on projected changes from IPM on the populations near these facilities, which Because of projected changes in
number and scale of pollution controls dispatch under the proposed
and labor intensities in relevant sectors. 57 U.S. EPA. 2001. IRIS Summary for requirements, the proposed rule is also
Finally, some of the other types of Methylmercury. U.S. Environmental Protection projected to reduce CO2 emissions. The
employment impacts that will be Agency, Washington, DC. (USEPA, 2001). EPA estimated the climate benefits from
this proposed rule using estimates of the We estimate the global social benefits Group on the Social Cost of Greenhouse
social cost of greenhouse gases (SC– of CO2 emission reductions expected Gases (IWG) finds that, taken together,
GHG), specifically the social cost of from the proposed rule using the SC– the limitations suggest that these SC–
carbon (SC–CO2). The SC–CO2 is the GHG estimates presented in the CO2 estimates likely underestimate the
monetary value of the net harm to February 2021 TSD: Social Cost of damages from CO2 emissions. The IWG
society associated with a marginal Carbon, Methane, and Nitrous Oxide is currently working on a
increase in CO2 emissions in a given Interim Estimates under E.O. 13990. comprehensive update of the SC–GHG
year, or the benefit of avoiding that These SC–GHG estimates are interim estimates (under E.O. 13990) taking into
increase. In principle, SC–CO2 includes values developed under E.O. 13990 for consideration recommendations from
the value of all climate change impacts use in benefit-cost analyses until the National Academies of Sciences,
(both negative and positive), including updated estimates of the impacts of Engineering and Medicine, recent
(but not limited to) changes in net climate change can be developed based scientific literature, public comments
agricultural productivity, human health on the best available climate science received on the February 2021 TSD and
effects, property damage from increased and economics. We have evaluated the other input from experts and diverse
flood risk natural disasters, disruption SC–GHG estimates in the TSD and have stakeholder groups. The EPA is
of energy systems, risk of conflict, determined that these estimates are participating in the IWG’s work. In
environmental migration, and the value appropriate for use in estimating the addition, while that process continues,
of ecosystem services. The SC–CO2, global social benefits of CO2 emission the EPA is continuously reviewing
therefore, reflects the societal value of reductions expected from this proposed developments in the scientific literature
reducing emissions of the gas in rule. After considering the TSD, and the on the SC–GHG, including more robust
question by one metric ton and is the issues and studies discussed therein, the methodologies for estimating damages
theoretically appropriate value to use in EPA finds that these estimates, while from emissions, and looking for
conducting benefit-cost analyses of likely an underestimate, are the best opportunities to further improve SC–
policies that affect CO2 emissions. In currently available SC–GHG estimates. GHG estimation going forward. Most
practice, data and modeling limitations These SC–GHG estimates were recently, the EPA has developed a draft
naturally restrain the ability of SC–CO2 developed over many years using a updated SC–GHG methodology within a
estimates to include all the important transparent process, peer-reviewed sensitivity analysis in the RIA of the
physical, ecological, and economic methodologies, the best science EPA’s November 2022 supplemental
impacts of climate change, such that the available at the time of that process, and proposal for oil and gas standards that
estimates are a partial accounting of with input from the public. As is currently undergoing external peer
climate change impacts and will discussed in section 4.4 of the RIA, review and a public comment process.
therefore, tend to be underestimates of these interim SC–CO2 estimates have a See section 4.4 of the RIA for more
the marginal benefits of abatement. The number of limitations, including that discussion of this effort.
EPA and other Federal agencies began the models used to produce them do not Table 12 presents the estimated PV
regularly incorporating SC–GHG include all of the important physical, and EAV of the projected health and
estimates in their benefit-cost analyses ecological, and economic impacts of climate benefits across the regulatory
conducted under E.O. 12866 58 since climate change recognized in the options examined in the RIA in 2019
2008, following a Ninth Circuit Court of climate-change literature and that dollars discounted to 2023. The table
Appeals remand of a rule for failing to several modeling input assumptions are includes benefit estimates for the less
monetize the benefits of reducing CO2 outdated. As discussed in the February and more stringent regulatory options
emissions in a rulemaking process. 2021 TSD, the Interagency Working examined in the RIA for this proposal.
TABLE 12—PROJECTED BENEFITS OF THE PROPOSED RULE, LESS STRINGENT ALTERNATIVE, AND MORE STRINGENT
ALTERNATIVE, 2028 THROUGH 2037
Present value (PV)
3% Discount rate 7% Discount rate d
Health Benefits c ....................................... 1,900 0.0 11,000 1,200 0.0 7,100

Climate Benefits d ..................................... 1,400 0.0 3,200 d 1,400 d 0.0 d 3,200
Benefits e ........................................... 3,300 0.0 14,000 2,600 0.0 10,000
Equal annualized value (EAV) b

Health Benefits c ....................................... 220 0.0 1,300 170 0.0 1,000

Climate Benefits d ..................................... 170 0.0 380 d 170 d 0.0 d 380
58 Benefit-cost analyses have been an integral part economic consequences of regulations as part of the actions, an agency provide an assessment of the
of executive branch rulemaking for decades. rulemaking development process. E.O. 12866, potential costs and benefits of the regulatory action,
Presidents since the 1970s have issued executive released in 1993 and still in effect today, requires and that this assessment include a quantification of
orders requiring agencies to conduct analysis of the that for all economically significant regulatory benefits and costs to the extent feasible.
Equal annualized value (EAV) b
Benefits e ........................................... 390 0.0 1,700 330 0.0 1,400

a Values have been rounded to two significant figures. Rows may not appear to sum correctly due to rounding.
b The EAV of benefits are calculated over the 10-year period from 2028 to 2037.
c The projected monetized benefits include those related to public health associated with reductions in PM2.5 and ozone concentrations. The
projected health benefits are associated with several point estimates and are presented at real discount rates of 3 and 7 percent.
d Climate benefits are based on reductions in CO emissions and are calculated using four different estimates of the social cost of carbon diox-
2
ide (SC–CO2): model average at 2.5 percent, 3 percent, and 5 percent discount rates; 95th percentile at 3 percent discount rate. For the presen-
tational purposes of this table, we show the climate benefits associated with the average SC–CO2 at a 3 percent discount rate, but the Agency
does not have a single central SC–CO2 point estimate. Climate benefits in this table are discounted using a 3 percent discount rate to obtain the
PV and EAV estimates in the table. We emphasize the importance and value of considering the benefits calculated using all four SC–CO2 esti-
mates. Section 4.4 of the RIA presents estimates of the projected climate benefits of this proposal using all four rates. We note that consider-
ation of climate benefits calculated using discount rates below 3 percent, including 2 percent and lower, is warranted when discounting intergen-
erational impacts.
e Several categories of benefits remain unmonetized and are thus not directly reflected in the quantified benefit estimates in the table. Non-
monetized benefits include benefits from reductions in Hg and non-Hg metal HAP emissions and from the increased transparency and acceler-
ated identification of anomalous emission anticipated from requiring CEMS.
This proposed rule is projected to F. What analysis of environmental 1. Are there potential EJ concerns
reduce PM2.5 and ozone concentrations, justice did we conduct? associated with environmental stressors
producing a projected PV of monetized Executive Order 12898 directs the affected by the regulatory action for
health benefits of about $1.9 billion, EPA to identify the populations of population groups of concern in the
with an EAV of about $220 million concern who are most likely to baseline?
discounted at 3 percent. The projected experience unequal burdens from 2. Are there potential EJ concerns
PV of monetized climate benefits of the environmental harms; specifically, associated with environmental stressors
proposal are estimated to be about $1.4 minority populations, low-income affected by the regulatory action for
billion, with an EAV of about $170 populations, and Indigenous peoples.59 population groups of concern for the
million using the SC–CO2 discounted at Additionally, Executive Order 13985 is regulatory option(s) under
3 percent. Thus, this proposed rule intended to advance racial equity and
would generate a PV of monetized consideration?
support underserved communities
benefits of $3.3 billion, with an EAV of through federal government actions.60 3. For the regulatory option(s) under
$390 million discounted at a 3 percent The EPA defines environmental justice consideration, are potential EJ concerns
rate. (EJ) as the fair treatment and meaningful created or mitigated compared to the
At a 7 percent discount rate, this involvement of all people regardless of baseline?’’
proposed rule is expected to generate race, color, national origin, or income, To address these questions in the
projected PV of monetized health with respect to the development, EPA’s first quantitative EJ analysis in
benefits of $1.2 billion, with an EAV of implementation, and enforcement of the context of a MATS rule, the EPA
about $170 million discounted at 7 environmental laws, regulations, and developed a unique analytical approach
percent. Climate benefits remain policies. The EPA further defines the that considers the purpose and specifics
discounted at 3 percent in this benefits term fair treatment to mean that ‘‘no of the proposed rulemaking, as well as
analysis and are estimated to be about group of people should bear a the nature of known and potential
$1.4 billion, with an EAV of about $170 disproportionate burden of disproportionate and adverse exposures
million using the SC–CO2. Thus, this environmental harms and risks, and impacts. However, due to data
proposed rule would generate a PV of including those resulting from the limitations, it is possible that our
monetized benefits of $2.6 billion, with negative environmental consequences of analysis failed to identify disparities
an EAV of $330 million discounted at a industrial, governmental, and that may exist, such as potential EJ
7 percent rate. The potential benefits commercial operations or programs and
characteristics (e.g., residence of
from reducing Hg and non-Hg metal policies.’’ 61 In recognizing that minority
historically red lined areas),
HAP were not monetized and are and low-income populations often bear
an unequal burden of environmental environmental impacts (e.g., other
therefore not directly reflected in the ozone metrics), and more granular
monetized benefit-cost estimates harms and risks, the EPA continues to
consider ways of protecting them from spatial resolutions (e.g., neighborhood
associated with this proposal. Potential
adverse public health and scale) that were not evaluated. Also due
benefits from the increased transparency
environmental effects of air pollution. to data and resource limitations, we
and accelerated identification of
The EPA’s EJ technical guidance 62 discuss HAP and climate EJ impacts of
anomalous emission anticipated from
requiring CEMS were also not states that ‘‘[t]he analysis of potential EJ this action qualitatively (sections 6.3
monetized in this analysis and are concerns for regulatory actions should and 6.6 of the RIA).
therefore also not directly reflected in address three questions: For this proposed rule, we employ
the monetized benefit-cost comparisons. two types of analysis to respond to the
We nonetheless consider these impacts 59 59 FR 7629, February 16, 1994. previous three questions: proximity
60 86 FR 7009, January 20, 2021.
in our evaluation of the net benefits of analyses and exposure analyses. Both
61 https://www.epa.gov/environmentaljustice.
the rule and find, if we were able to 62 U.S. Environmental Protection Agency (EPA),
types of analyses can inform whether
monetize these beneficial impacts, the 2015. Guidance on Considering Environmental there are potential EJ concerns for
proposal would have greater net benefits Justice During the Development of Regulatory population groups of concern in the
than shown in Table 12. Actions.
baseline (question 1).63 In contrast, only local environmental stressors, such as Guidance document more directly than
the exposure analyses, which are based local NO2 and SO2 emitted from affected the proximity analyses, as it evaluates a
on future air quality modeling, can sources in this proposed rule, traffic, or form of the environmental stressor
inform whether there will be potential noise. The baseline analysis indicates targeted by the regulatory action.
EJ concerns after implementation of the that on average the populations living Baseline ozone and PM2.5 analyses show
regulatory options under consideration within 10 km of coal plants potentially that certain populations, such as
(question 2) and whether potential EJ subject to the proposed or alternate Hispanics, Asians, those linguistically
concerns will be created or mitigated filterable PM standards have a higher isolated, those less educated, and
compared to the baseline (question 3). percentage of people living below two children may experience somewhat
While the exposure analysis can times the poverty level than the national higher ozone and PM2.5 concentrations
respond to all three questions, several average. In addition, on average the compared to the national average.
caveats should be noted. For example, percentage of the Native American Therefore, also in response to question
the air pollutant exposure metrics are population living within 10 km of 1, there likely are potential EJ concerns
limited to those used in the benefits lignite plants potentially subject to associated with ozone and PM2.5
assessment. For ozone, that is the proposed Hg standard is higher than the exposures affected by the regulatory
maximum daily 8-hour average, national average. Relating these results action for population groups of concern
averaged across the April through to EJ question 1, we conclude that there
in the baseline. However, these baseline
September warm season (AS–MO3) and may be potential EJ concerns associated
exposure results have not been fully
for PM2.5 that is the annual average. This with directly emitted pollutants that are
explored and additional analyses are
ozone metric likely smooths potential affected by the regulatory action (e.g.,
daily ozone gradients and is not directly SO2) for certain population groups of likely needed to understand potential
relatable to the NAAQS, whereas the concern in the baseline (question 1). implications. Due to the small
PM2.5 metric is more similar to the long However, as proximity to affected magnitude of the exposure changes
term PM2.5 standard. The air quality facilities does not capture variation in across population demographics
modeling estimates are also based on baseline exposure across communities, associated with the rulemaking relative
state level emission data paired with nor does it indicate that any exposures to the magnitude of the baseline
facility-level baseline emissions and or impacts will occur, these results disparities, we infer that post-policy EJ
provided at a resolution of 12 km2. should not be interpreted as a direct ozone and PM2.5 concentration burdens
Additionally, here we focus on air measure of exposure or impact. are likely to remain after
quality changes due to this proposed As HAP exposure results generated as implementation of the regulatory action
rulemaking and infer post-policy part of the 2020 Residual Risk analysis or alternative under consideration
exposure burden impacts. were below both the presumptive (question 2).
Exposure analysis results are acceptable cancer risk threshold and Question 3 asks whether potential EJ
provided in two formats: aggregated and noncancer health benchmarks and this concerns will be created or mitigated as
distributional. The aggregated results proposed regulation should further compared to the baseline. Due to the
provide an overview of potential ozone reduce exposure to HAP, there are no very small magnitude of differences
exposure differences across populations ‘disproportionate and adverse effects’ of across demographic population post-
at the national- and state-levels, while potential EJ concern. Therefore, we did policy ozone and PM2.5 exposure
the distributional results show detailed not perform a quantitative EJ assessment impacts, we do not find evidence that
information about ozone concentration of HAP risk.
potential EJ concerns related to ozone
changes experienced by everyone This proposed rule is also expected to
reduce emissions of direct PM2.5, NOX, and PM2.5 concentrations will be created
within each population.
In section 6 of the RIA we utilize the and SO2 nationally throughout the year. or mitigated as compared to the
two types of analysis to address the Because NOX and SO2 are also baseline.64
three EJ questions by quantitatively precursors to secondary formation of Prior to this proposed rule, the EPA
evaluating: (1) the proximity of affected ambient PM2.5 and NOX is a precursor initiated a public outreach effort to
facilities to populations of potential EJ to ozone formation, reducing these gather input from stakeholder groups
concern (section 6.4); and (2) the emissions would impact human likely to be interested in this proposed
potential for disproportionate ozone and exposure. Quantitative ozone and PM2.5 rule. Specifically, the EPA presented on
PM2.5 concentrations in the baseline and exposure analyses can provide insight a National EJ call on September 20,
concentration changes after rule into all three EJ questions, so they are 2022, to share information about the
implementation across different performed to evaluate potential proposed rule and solicit feedback about
demographic groups (section 6.5). Each disproportionate impacts of this potential EJ considerations. The webinar
of these analyses depends on mutually rulemaking. Even though both the was attended by individuals
exclusive assumptions, was performed proximity and exposure analyses can representing state governments,
to answer separate questions, and is potentially improve understanding of federally recognized tribes,
associated with unique limitations and baseline EJ concerns (question 1), the environmental non-governmental
uncertainties. two should not be directly compared. organizations, higher education
Baseline demographic proximity This is because the demographic institutions, industry, and the EPA.65
analyses can be relevant for identifying proximity analysis does not include air
populations that may be exposed to quality information and is based on 64 Please note, exposure results should not be
current, not future, population extrapolated to other air pollutant. Detailed EJ

63 The baseline for proximity analyses is current analytical results can be found in Section 6 of the
information.
population information, whereas the baseline for RIA.
ozone exposure analyses are the future years in
The baseline analysis of ozone and
65 This does not constitute the EPA’s tribal
which the regulatory options will be implemented PM2.5 concentration burden responds to
consultation under E.O. 13175, which is described
(e.g., 2023 and 2026). question 1 from EPA’s EJ Technical
in section VIII.F of this proposed rule.
In addition to the engagement found at https://www.epa.gov/laws- presents results for the less stringent
conducted prior to this proposed rule, regulations/laws-and-executive-orders. and more stringent alternatives that are
the EPA is providing the public, examined in the RIA for this proposal.
A. Executive Order 12866: Regulatory
including those communities Under E.O. 12866, the EPA is directed
Planning and Review and Executive
disproportionately impacted by the to consider all of the costs and benefits
Order 13563: Improving Regulation and
burdens of pollution, opportunities to of its actions, not just those that stem
Regulatory Review
engage in the EPA’s public comment from the regulated pollutant.
period for this proposed rule, including This action was submitted to the OMB Accordingly, the projected monetized
by hosting a public hearing. This public for review under section 3(f)(1) of benefits of the proposal include health
hearing will occur according to the Executive Order 12866. Any changes benefits associated with projected
schedule identified in the made in response to recommendations reductions in fine particulate matter
SUPPLEMENTARY INFORMATION under the received as part of review under (PM2.5) and ozone concentration. The
heading entitled Participation in virtual Executive Order 12866 have been projected monetized benefits also
public hearing of this proposed rule. documented in the docket. The EPA include climate benefits due to
prepared an analysis of the potential reductions in CO2 emissions. The
VII. Request for Comments costs and benefits associated with this projected health benefits are associated
action. This analysis, ‘‘Regulatory with several point estimates and are
We solicit comments on this proposed Impact Analysis for the Proposed presented at real discount rates of 3 and
action. In addition to general comments National Emission Standards for 7 percent. The projected climate
on this proposed action, we are also Hazardous Air Pollutants: Coal- and Oil- benefits in this table are based on
interested in additional data that may Fired Electric Utility Steam Generating estimates of the SC–CO2 at a 3 percent
improve the analyses. We are Units Review of the Residual Risk and discount rate and are discounted using
specifically interested in receiving any Technology Review’’ (Ref. EPA–452/R– a 3 percent discount rate to obtain the
information regarding developments in 23–002), is available in the docket and PV and EAV estimates in the table. The
practices, processes, and control is briefly summarized in section VI of power industry’s compliance costs are
technologies that reduce HAP this preamble and here. represented in this analysis as the
emissions. We are also interested in Table 13 presents the estimated PV change in electric power generation
comments on any reliance interests and EAV of the projected health costs between the baseline and policy
stakeholders may have that would be benefits, climate benefits, compliance scenarios. In simple terms, these costs
affected by this proposed action. costs, and net benefits of the proposed are an estimate of the increased power
VIII. Statutory and Executive Order rule in 2019 dollars discounted to 2023. industry expenditures required to
Reviews The estimated monetized net benefits implement the proposed requirements
are the projected monetized benefits and represent the EPA’s best estimate of
Additional information about these minus the projected monetized costs of the social cost of the proposed
statutes and Executive orders can be the proposed rule. Table 13 also rulemaking.
TABLE 13—PROJECTED MONETIZED BENEFITS, COMPLIANCE COSTS, AND NET BENEFITS OF THE PROPOSED RULE, LESS
STRINGENT ALTERNATIVE, AND MORE STRINGENT ALTERNATIVE, 2028 THROUGH 2037
Present value
(PV)
Less More Less More

Proposed Proposed
stringent stringent stringent stringent
Health Benefits c ....................................... 1,900 0.0 11,000 1,200 0.0 7,100

Climate Benefits d ..................................... 1,400 0.0 3,200 d 1,400 d 0.0 d 3,200
Compliance Costs .................................... 330 ¥45 4,600 230 ¥31 3,400
Net Benefits e .................................... 3,000 45 9,800 2,400 31 6,900
Equal Annualized Value (EAV) b
Less More Less More

Proposed Proposed
stringent stringent stringent stringent
Health Benefits c ....................................... 220 0.0 1,300 170 0.0 1,000

Climate Benefits d ..................................... 170 0.0 380 d 170 d 0.0 d 380
Compliance Costs .................................... 38 ¥5.2 540 33 ¥4.5 490
Net Benefits e .................................... 350 5.2 1,100 300 4.5 900

a Valueshave been rounded to two significant figures. Rows may not appear to sum correctly due to rounding.
b The EAV of costs and benefits are calculated over the 10-year period from 2028 to 2037.
c The projected monetized benefits include those related to public health associated with reductions in PM
2.5 and ozone concentrations. The
projected health benefits are associated with several point estimates and are presented at real discount rates of 3 and 7 percent.
d Climate benefits are based on reductions in CO emissions and are calculated using four different estimates of the social cost of carbon diox-
2
ide (SC–CO2): model average at 2.5 percent, 3 percent, and 5 percent discount rates; 95th percentile at 3 percent discount rate. For the presen-
tational purposes of this table, we show the climate benefits associated with the average SC–CO2 at a 3 percent discount rate, but the Agency
does not have a single central SC–CO2 point estimate. Climate benefits in this table are discounted using a 3 percent discount rate to obtain the
PV and EAV estimates in the table. We emphasize the importance and value of considering the benefits calculated using all four SC–CO2 esti-
mates. Section 4.4 of the RIA presents estimates of the projected climate benefits of this proposal using all four rates. We note that consider-
ation of climate benefits calculated using discount rates below 3 percent, including 2 percent and lower, is warranted when discounting intergen-
erational impacts.
e Several categories of benefits remain unmonetized and are thus not directly reflected in the quantified benefit estimates in the table. Non-
monetized benefits include benefits from reductions in Hg and non-Hg metal HAP emissions and from the increased transparency and acceler-
ated identification of anomalous emission anticipated from requiring CEMS.
As shown in Table 13, this proposed B. Paperwork Reduction Act (PRA) Total estimated cost: $100,100,000
rule is projected to reduce PM2.5 and OMB has previously approved the (per year), includes $49,600,000
ozone concentrations, producing a information collection activities annualized capital or operation &
projected PV of monetized health contained in the existing regulations maintenance costs.
benefits of about $1.9 billion, with an An agency may not conduct or
and has assigned OMB control number
EAV of about $220 million discounted sponsor, and a person is not required to
2060–0567. The information collection
at 3 percent. The proposed rule is also respond to, a collection of information
activities in this proposed rule, which
projected to reduce greenhouse gas unless it displays a currently valid OMB
are a revision to the existing approved
emissions in the form of CO2, producing control number. The OMB control
information collection activities, have
a projected PV of monetized climate numbers for the EPA’s regulations in 40
been submitted for approval to the OMB
benefits of about $1.4 billion, with an CFR are listed in 40 CFR part 9.
under the PRA. The ICR document that Submit your comments on the EPA’s
EAV of about $170 million using the the EPA prepared has been assigned
SC–CO2 discounted at 3 percent. The PV need for this information, the accuracy
EPA ICR number 2137–12. You can find of the provided burden estimates and
of the projected compliance costs are a copy of the ICR in the docket for this any suggested methods for minimizing
$330 million, with an EAV of about $38 rule, and it is briefly summarized here. respondent burden to the EPA using the
million discounted at 3 percent. The information collection activities docket identified at the beginning of this
Combining the projected benefits with in this proposed rule include rule. The EPA will respond to any ICR-
the projected compliance costs yields a continuous emission monitoring, related comments in the final rule. You
net benefit PV estimate of $3 billion and performance testing, notifications and may also send your ICR-related
EAV of $350 million. periodic reports, recording information, comments to OMB’s Office of
At a 7 percent discount rate, this monitoring and the maintenance of Information and Regulatory Affairs
proposed rule is expected to generate records. The information generated by using the interface at https://
projected PV of monetized health these activities will be used by the EPA www.reginfo.gov/public/do/PRAMain.
benefits of $1.2 billion, with an EAV of to ensure that affected facilities comply Find this particular information
about $170 million. Climate benefits with the emission limits and other collection by selecting ‘‘Currently under
remain discounted at 3 percent in this requirements. Records and reports are Review—Open for Public Comments’’ or
net benefits analysis. Thus, this necessary to enable delegated by using the search function. OMB must
proposed rule would generate a PV of authorities to identify affected facilities receive comments no later than June 23,
monetized benefits of $2.6 billion, with that may not be in compliance with the 2023.
an EAV of $340 million discounted at a requirements. Based on reported
information, delegated authorities will C. Regulatory Flexibility Act (RFA)
7 percent rate. The PV of the projected
decide which units and what records or The EPA certifies that this proposed
compliance costs are $230 million, with
processes should be inspected. The action will not have a significant
an EAV of $33 million discounted at 7
recordkeeping requirements require economic impact on a substantial
percent. Combining the projected
only the specific information needed to number of small entities under the
benefits with the projected compliance
determine compliance. These Regulatory Flexibility Act (RFA). The
costs yields a net benefit PV estimate of
recordkeeping and reporting EPA chose to examine the projected
$2.4 billion and an EAV of $300 million. requirements are specifically authorized impacts of a more stringent regulatory
The potential benefits from reducing by CAA section 114 (42 U.S.C. 7414). option than proposed on small entities
Hg and non-Hg metal HAP were not Respondents/affected entities: The in order to present a scenario of
monetized and are therefore not directly respondents are owners or operators of ‘‘maximum cost impact.’’ As projected
reflected in the monetized benefit-cost coal- and oil-fired EGUs. The NAICS cost impacts of the proposed rule is
estimates associated with this proposal. codes for the coal- and oil-fired EGU dominated by cost impacts of the more
Potential benefits from the increased industry are 221112, 221122, and stringent alternative also examined in
transparency and accelerated 921150. the RIA, a no SISNOSE conclusion for
identification of anomalous emission Respondent’s obligation to respond: the more stringent option can be
anticipated from requiring CEMS Mandatory per 42 U.S.C. 7414 et seq. extended to the proposed rule and less
requiring were also not monetized in Estimated number of respondents: stringent option.
this analysis and are therefore also not 187 per year. In 2028, the EPA identified 26
directly reflected in the monetized Frequency of response: The frequency potentially affected small entities
benefit-cost comparisons. We of responses varies depending on the operating 41 units at 27 facilities, and of
nonetheless consider these impacts in burden item. Responses include daily these 26, only two small entities may
our evaluation of the net benefits of the calibrations, quarterly inspections, and experience compliance cost increases
rule and find, if we were able to semiannual compliance reports. greater than 1 percent of revenue under
monetize these beneficial impacts, the Total estimated burden: 443,000 the proposed rule, and three small
proposal would have greater net benefits hours (per year). Burden is defined at 5 entities may experience such increases
than shown in Table 13. CFR 1320.3(b). under the more stringent alternative.
Details of this analysis are presented in during startup and shutdown periods. Although this action does not have
section 5 of the RIA, which is in the Also, APPA stated that they support the tribal implications as specified in
public docket. continuation of the 30-day rolling Executive Order 13175, the EPA
average to assure compliance with consulted with tribal officials during the
D. Unfunded Mandates Reform Act of
MATS emission requirements to allow development of this action. On
1995 (UMRA)
for hourly variability caused by unit September 1, 2022, the EPA sent a letter
This action does not contain an operation and load requirements, to all federally recognized Indian tribes
unfunded mandate of $100 million or including startup and shutdown events. initiating consultation to obtain input
more as described in UMRA, 2 U.S.C. on this proposal. The EPA did not
1531–1538, and does not significantly or E. Executive Order 13132: Federalism receive any requests from consultation
uniquely affect small governments. This This action does not have federalism from Indian tribes. The EPA also
action imposes no enforceable duty on implications. It will not have substantial participated in the September 2022
any state, local, or tribal governments or direct effects on the states, on the National Tribal Air Association EPA Air
the private sector. In light of the interest relationship between the National Policy Update Call to solicit input on
in this rule among governmental Government and the states, or on the this proposed action.
entities, the EPA initiated consultation distribution of power and
with governmental entities. The EPA G. Executive Order 13045: Protection of
responsibilities among the various Children From Environmental Health
invited the following 10 national levels of government.
organizations representing state and Risks and Safety Risks
The EPA believes, however, that this
local elected officials to a virtual This proposed rule is a ‘‘[c]overed
action may be of interest to state and/
meeting on September 22, 2022: (1) regulatory action’’ under Executive
or local governments. Consistent with Order 13045 because it is a significant
National Governors Association, (2)
National Conference of State the EPA’s policy to promote regulatory action as described in section
Legislatures, (3) Council of State communication between the EPA and 3(f)(1) of Executive Order 12866, and
Governments, (4) National League of state and local governments, the EPA the EPA believes that, even though the
Cities, (5) U.S. Conference of Mayors, (6) consulted with representatives of state residual risk assessment showed all
National Association of Counties, (7) and local governments in the process of modeled exposures to HAP to be below
International City/County Management developing the proposed amendments thresholds for public health concern,
Association, (8) National Association of to permit them to have meaningful and the rule should reduce HAP exposure by
Towns and Townships, (9) County timely input into its development. The reducing emissions of Hg and non-Hg
Executives of America, and (10) EPA’s consultation regarded planned HAP with the potential to reduce HAP
Environmental Council of States. These actions for the review of the MATS RTR. exposure to vulnerable populations
10 organizations representing elected The EPA met with 10 national including children. Accordingly, we
state and local officials have been organizations representing state and have evaluated the potential for
identified by the EPA as the ‘‘Big 10’’ local elected officials to provide general environmental health or safety effects
organizations appropriate to contact for background on the review of the MATS from exposure to HAP on children. The
purpose of consultation with elected RTR, answer questions, and solicit input results of this evaluation are contained
officials. Also, the EPA invited air and from state and local governments. The in the RIA and are available in the
utility professional groups who may UMRA discussion in this preamble docket for this action. The EPA believes
have state and local government includes a description of the that the PM2.5-related, ozone-related,
members, such as the Association of Air consultation. In the spirit of E.O. 13132, and CO2-related benefits projected
Pollution Control Agencies, National and consistent with EPA policy to under this proposed rule will further
Association of Clean Air Agencies, and promote communications between state improve children’s health. Specifically,
others to participate in the meeting. The and local governments, the EPA the PM2.5 and ozone EJ exposure
purpose of the consultation was to specifically solicits comment on this analyses in section 6 of the RIA suggests
provide general background on the proposed action from state and local that nationally, children (ages 0–17) will
review of the MATS RTR, answer officials. experience at least as great a reduction
questions, and solicit input from state F. Executive Order 13175: Consultation in annual PM2.5 and ozone exposures as
and local governments. Subsequent to and Coordination With Indian Tribal adults (ages 18–64) will experience in
the September 22, 2022, meeting, the Governments 2028, 2030 and 2035 under all
EPA received a letter from the American regulatory alternatives of this
Public Power Association (APPA). The This action does not have tribal rulemaking.
EPA opened a non-rulemaking docket implications as specified in Executive
for public input on the EPA’s efforts to Order 13175. The Executive order H. Executive Order 13211: Actions
reduce greenhouse gas emissions from defines tribal implications as ‘‘actions Concerning Regulations That
new and existing fossil fuel-fired EGUs. that have substantial direct effects on Significantly Affect Energy Supply,
The APPA letter was submitted to the one or more Indian tribes, on the Distribution, or Use
non-rulemaking docket. See Docket ID relationship between the Federal This action is not a ‘‘significant
No. EPA–HQ–OAR–2022–0723–0016. In Government and Indian tribes.’’ The energy action’’ because it is not likely to
that letter, APPA stated that they were amendments proposed in this action have a significant adverse effect on the
not able to identify any new cost- would not have a substantial direct supply, distribution, or use of energy.
effective technologies to reduce HAP effect on one or more tribes, change the For 2028, the compliance year for the
emissions and that many of the current relationship between the Federal proposed standards, the EPA projects a
technologies used are state-of-the-art Government and tribes, or affect the less than 0.1 percent change in retail
controls that continue to reduce HAP distribution of power and electricity prices on average across the
emissions. In addition, APPA stated responsibilities between the Federal contiguous U.S., a less than 0.1 percent
there have been no developments in the Government and Indian tribes. Thus, reduction in coal-fired electricity
emission control practices or processes Executive Order 13175 does not apply generation, and a less than 0.1 percent
available to control HAP emissions to this action. increase in natural gas-fired electricity
generation. The EPA does not project a threshold and the RfD, and this be meaningfully exacerbated or
significant change in utility power proposed regulation will likely further mitigated in the regulatory alternatives
sector delivered natural gas prices in reduce exposure to HAP. As such, the under consideration as compared to the
2028. Details of the projected energy EPA believes that this action does not baseline due to the small magnitude of
effects are presented in section 3 of the result in disproportionate and adverse ozone and PM2.5 concentration changes
RIA, which is in the public docket. effects on people of color, low-income associated with this rule relative to
populations, and/or Indigenous peoples. baseline disparities and the very small
I. National Technology Transfer and The EPA believes that PM2.5 and differences in the distributional
Advancement Act (NTTAA) ozone exposures that exist prior to this analyses of post-policy ozone and PM2.5
This rulemaking does not involve action result in disproportionate and exposure impacts. Importantly, the
technical standards. adverse human health or environmental action described in this rule is expected
effects on people of color, low-income to lower ozone and PM2.5 in certain
J. Executive Order 12898: Federal populations and/or Indigenous peoples.
Actions To Address Environmental areas, and thus mitigate some pre-
Specifically, baseline PM2.5 and ozone existing health risks across all
Justice in Minority Populations and and exposure analyses show that certain
Low-Income Populations populations evaluated.
populations, such as Hispanics, Asians, The documentation for these analyses
Executive Order 12898 (59 FR 7629, those linguistically isolated, those less is contained in section VI.F of this this
February 16, 1994) directs federal educated, and children may experience proposed rule and in section 6,
agencies, to the greatest extent disproportionately higher ozone and Environmental Justice Impacts of the
practicable and permitted by law, to PM2.5 exposures as compared to the RIA, which is in the public docket.
make EJ part of their mission by national average. The EPA believes that
identifying and addressing, as this action is not likely to change List of Subjects in 40 CFR Part 63
appropriate, disproportionately high existing disproportionate PM2.5 and Environmental protection, Air
and adverse human health or ozone exposure impacts on people of pollution control, Hazardous
environmental effects of their programs, color, low-income populations and/or substances, Reporting and
policies, and activities on minority Indigenous peoples. American Indians recordkeeping requirements.
populations (people of color and/or may also experience disproportionately
Indigenous peoples) and low-income higher ozone concentrations than the Michael S. Regan,
populations. reference group. We do not find Administrator.
HAP risks were below both the evidence that potential EJ concerns [FR Doc. 2023–07383 Filed 4–21–23; 8:45 am]
presumptive acceptable cancer risk related to ozone or PM2.5 exposures will BILLING CODE 6560–50–P

EPA-Emission of Coal

Uploaded by

Copyright:

Available Formats

You might also like

EPA-Emission of Coal

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

EPA-Emission of Coal

Uploaded by

Copyright:

Available Formats

24854 Federal Register / Vol. 88, No.

78 / Monday, April 24, 2023 / Proposed Rules

ENVIRONMENTAL PROTECTION Docket ID No. EPA–HQ–OAR–2018– air-pollution/mercury-and-air-toxics-

12 percent of performers, and then

TABLE 1—EMISSION LIMITS FOR EXISTING AFFECTED EGUS

Antimony, Sb ......................................... 2.2 lb/TBtu or 2.0E–2 lb/GWh.

TABLE 1—EMISSION LIMITS FOR EXISTING AFFECTED EGUS—Continued

Hg .......................................................... 4.0E–2 lb/TBtu or 4.0E–4 lb/GWh.

for information pursuant to section 114 Retrieval System (WebFIRE).12 The

lignite-fired EGUs. The CAA section 114 webfire. utilitypg.html.

CAELP_20210819.pdf. after the effective date of the rule amendments to

docket. 2015-07/documents/suppdoc410mats.pdf and in power-plant-hazardous-air-pollution-emissions-

1.5E–02 1.0E–02 6.0E–03

Affected Units (Capacity, GW) .................................................................................................... 8 (4.02) 20 (9.34) 65 (32.9)

Instrument and Other initial Capital Operation and Other annual

Extractive ...................... 100,427 84,458 20,300 3,689 54,392 7,525 85,906

Standards and Technology (NIST) Agency’s ‘‘EGUs Permanently Ceasing

TABLE 5—TOP HG-EMITTING EGUS IN 2021

1 .................... Coal Creek 2 ...................................................... Lignite ................................................................. 181.8 ND

+ FF. fuel additive is also applied to the

36 National Electric Energy Data System (NEEDS) www.epa.gov/power-sector-modeling/national-

TABLE 6—CONTROL CONFIGURATIONS FOR LIGNITE-FIRED EGUS—Continued

TABLE 7—2021 FUEL USE AT LIGNITE-FIRED EGUS

Antelope Valley 1 ................................................................. 0.0 0.6 5.8 93.5 0.0

TABLE 8—HG EMISSIONS AND CONTROL PERFORMANCE OF LIGNITE-FIRED EGUS IN 2021

Limestone #2 ............... 14.88 73.1 91.9 6.20 1.59 74.4 80.7

37 Discussion of how these assumptions were

developed for use in the EPA’s IPM modeling is

TABLE 8—HG EMISSIONS AND CONTROL PERFORMANCE OF LIGNITE-FIRED EGUS IN 2021—Continued

on performance of controls can provide the basis for production of difficult-to-control

EGUs firing on non-lignite coals, and Information Administration, available at stream.

TABLE 9—EGUS RELYING ON PARAGRAPH (2) OF THE DEFINITION OF ‘‘STARTUP’’

Prairie State Generating ................ 1 ........................ 55856 877 .................................... Bituminous.

design structure, or a combination of

Year Less stringent More stringent

2028 ............................................................................................................................................. 62.0 0.0 208.0

Year Less stringent More stringent

PM2.5 (thousand tons)

2028 ............................................................................................................................................. 0.4 0.0 2.6

SO2 (thousand tons)

2028 ............................................................................................................................................. 0.9 0.0 11.6

Ozone-season NOX (thousand tons)

2028 ............................................................................................................................................. 0.2 0.0 7.2

Annual NOX (thousand tons)

2028 ............................................................................................................................................. 0.4 0.0 18.1

HCl (thousand tons)

2028 ............................................................................................................................................. 0.0 0.0 0.2

CO2 (million metric tons)

2028 ............................................................................................................................................. 0.2 0.0 21.9

3% Discount rate 7% Discount rate

Present value (PV)

3% Discount rate 7% Discount rate d

Health Benefits c ....................................... 1,900 0.0 11,000 1,200 0.0 7,100

Benefits e ........................................... 3,300 0.0 14,000 2,600 0.0 10,000

Equal annualized value (EAV) b

3% Discount rate 7% Discount rate d

Health Benefits c ....................................... 220 0.0 1,300 170 0.0 1,000

Equal annualized value (EAV) b