Proposed Rule: Air Pollution Standards of Performance For New Stationary Sources: Hospital, Medical, and Infectious Waste Incinerators

Tuesday,
February 6, 2007
Part II
Environmental
Protection Agency
40 CFR Part 60
Standards of Performance for New
Stationary Sources and Emission
Guidelines for Existing Sources: Hospital/
Medical/Infectious Waste Incinerators;
Proposed Rule
jlentini on PROD1PC65 with PROPOSAL2
VerDate Aug<31>2005 18:17 Feb 05, 2007 Jkt 211001 PO 00000 Frm 00001 Fmt 4717 Sfmt 4717 E:\FR\FM\06FEP2.SGM 06FEP2
5510 Federal Register / Vol. 72, No. 24 / Tuesday, February 6, 2007 / Proposed Rules
ENVIRONMENTAL PROTECTION Public Hearing. If anyone contacts made available on the Internet. If you
AGENCY EPA by February 26, 2007 requesting to submit an electronic comment, EPA
speak at a public hearing, EPA will hold recommends that you include your
40 CFR Part 60 a public hearing on March 8, 2007. If name and other contact information in
[EPA–HQ–OAR–2006–0534; FRL–8274–9] you are interested in attending the the body of your comment and with any
public hearing, contact Ms. Pamela disk or CD–ROM you submit. If EPA
RIN 2060–A004 Garrett at (919) 541–7966 to verify that cannot read your comment due to
a hearing will be held. technical difficulties and cannot contact
Standards of Performance for New you for clarification, EPA may not be
ADDRESSES: Submit your comments,
Stationary Sources and Emission able to consider your comment.
Guidelines for Existing Sources: identified by Docket ID No. EPA–HQ–
OAR–2006–0534, by one of the Electronic files should avoid the use of
Hospital/Medical/Infectious Waste special characters, any form of
Incinerators following methods:
www.regulations.gov: Follow the on- encryption, and be free of any defects or
AGENCY: Environmental Protection line instructions for submitting viruses.
Agency (EPA). comments. Public Hearing: If a public hearing is
E-mail: Send your comments via held, it will be held at EPA’s Campus
ACTION: Proposed rule.
electronic mail to located at 109 T.W. Alexander Drive in
SUMMARY: On September 15, 1997, EPA a-and-r-Docket@epa.gov, Attention Research Triangle Park, NC, or an
adopted new source performance Docket ID No. EPA–HQ–OAR–2006– alternate site nearby. Persons interested
standards (NSPS) and emission 0534. in presenting oral testimony must
guidelines for hospital/medical/ Facsimile: Fax your comments to contact Ms. Pamela Garrett at (919) 541–
infectious waste incinerators (HMIWI). (202) 566–1741, Attention Docket ID No. 7966 at least 2 days in advance of the
The NSPS and emission guidelines were EPA–HQ–OAR–2006–0534. hearing.
established under sections 111 and 129 Docket: EPA has established a docket
Mail: Send your comments to: EPA
of the Clean Air Act (CAA). On for this action under Docket ID No.
Docket Center (EPA/DC), Environmental
November 14, 1997, the Sierra Club and EPA–HQ–OAR–2006–0534 and Legacy
Protection Agency, Mailcode 6102T,
the Natural Resources Defense Council Docket ID No. A–91–61. All documents
1200 Pennsylvania Ave., NW.,
(Sierra Club) filed suit in the U.S. Court in the docket are listed in the
Washington, DC 20460, Attention
of Appeals for the District of Columbia www.regulations.gov index. Although
Docket ID No. EPA–HQ–OAR–2006–
Circuit (the Court) challenging EPA’s listed in the index, some information is
0534.
methodology for adopting the not publicly available, e.g., CBI or other
Hand Delivery: Deliver your
regulations. On March 2, 1999, the information whose disclosure is
comments to: EPA Docket Center (EPA/
Court issued its opinion. The Court restricted by statute. Certain other
DC), EPA West Building, Room 3334,
remanded the rule to EPA for further material, such as copyrighted material,
1301 Constitution Ave., NW.,
explanation of the Agency’s reasoning will be publicly available only in hard
Washington, DC 20460, Attention
in determining the minimum regulatory copy form. Publicly available docket
‘‘floors’’ for new and existing HMIWI. materials are available either
0534. Such deliveries are accepted only
The Court did not vacate the electronically at www.regulations.gov or
during the normal hours of operation
regulations, so the NSPS and emission in hard copy at the EPA Docket Center
(8:30 a.m. to 4:30 p.m., Monday through
guidelines remained in effect during the EPA/DC, EPA West, Room 3334, 1301
Friday, excluding legal holidays), and
remand and were fully implemented by Constitution Ave., NW., Washington,
special arrangements should be made
September 2002. This action provides DC. The Public Reading Room is open
for deliveries of boxed information.
EPA’s proposed response to the from 8:30 a.m. to 4:30 p.m., Monday
Instructions: Direct your comments to
questions raised in the Court’s remand. through Friday, excluding legal
Section 129(a)(5) of the CAA requires holidays. The telephone number for the
0534. The EPA’s policy is that all
EPA to review and, if appropriate, revise Public Reading Room is (202) 566–1744,
comments received will be included in
the NSPS and emission guidelines every and the telephone number for the EPA
the public docket and may be made
5 years. In this action, EPA also is Docket Center is (202) 566–1742.
available online at www.regulations.gov,
proposing our response to this 5-year including any personal information FOR FURTHER INFORMATION CONTACT: Ms.
review, which would revise the provided, unless the comment includes Mary Johnson, Energy Strategies Group,
emission limits in the NSPS and information claimed to be Confidential Sector Policies and Programs Division
emission guidelines to reflect the levels Business Information (CBI) or other (D243–01), Environmental Protection
of performance actually achieved by the information whose disclosure is Agency, Research Triangle Park, North
emission controls installed to meet the restricted by statute. Do not submit Carolina 27711; telephone number:
emission limits set forth in the information that you consider to be CBI (919) 541–5025; fax number: (919) 541–
September 15, 1997, NSPS and emission or otherwise protected through 5450; e-mail address:
guidelines. www.regulations.gov or e-mail. The johnson.mary@epa.gov.
DATES: Comments. Comments must be www.regulations.gov Web site is an SUPPLEMENTARY INFORMATION:
received on or before April 9, 2007. ‘‘anonymous access’’ system, which Organization of This Document. The
Under the Paperwork Reduction Act, means EPA will not know your identity following outline is provided to aid in
comments on the information collection or contact information unless you locating information in this preamble.
provisions must be received by the provide it in the body of your comment. I. General Information
Office of Management and Budget If you send an e-mail comment directly A. Does the proposed action apply to me?
(OMB) on or before March 8, 2007. to EPA without going through B. What should I consider as I prepare my
comments?
Because of the need to resolve the issues www.regulations.gov, your e-mail II. Background
raised in this action in a timely manner, address will be automatically captured III. Summary
EPA will not grant requests for and included as part of the comment A. Litigation and Proposed Remand
extensions beyond these dates. that is placed in the public docket and Response
Federal Register / Vol. 72, No. 24 / Tuesday, February 6, 2007 / Proposed Rules 5511
B. Proposed Amendments (CAA Section VI. Impacts of the Proposed Action for New H. Executive Order 13211: Actions That
129(a)(5) 5-Year Review) Units Significantly Affect Energy Supply,
IV. Rationale VII. Relationship of the Proposed Action to Distribution or Use
A. Rationale for the Proposed Response to Section 112(c)(6) of the CAA I. National Technology Transfer
the Remand VIII. Statutory and Executive Order Reviews Advancement Act
B. Rationale for the Proposed Amendments A. Executive Order 12866: Regulatory
Planning and Review I. General Information
(CAA Section 129(a)(5) 5-Year Review)
B. Paperwork Reduction Act
V. Impacts of the Proposed Action for
C. Regulatory Flexibility Act
A. Does the proposed action apply to
Existing Units D. Unfunded Mandates Reform Act me?
A. What are the primary air impacts? E. Executive Order 13132: Federalism
B. What are the water and solid waste Regulated Entities. Categories and
F. Executive Order 13175: Consultation
impacts? and Coordination with Indian Tribal
entities potentially affected by the
C. What are the energy impacts? Governments proposed action are those which operate
D. What are the secondary air impacts? G. Executive Order 13045: Protection of HMIWI. The NSPS and emission
E. What are the cost and economic Children from Environmental Health and guidelines for HMIWI affect the
impacts? Safety Risks following categories of sources:
Category NAICS Code Examples of potentially regulated entities
Industry ........................................ 622110, 622310, 325411, Private hospitals, other health care facilities, commercial research labora-
325412, 562213, 611310. tories, commercial waste disposal companies, private universities.
Federal Government .................... 622110, 541710, 928110 ............ Federal hospitals, other health care facilities, public health service,
armed services.
State/local/Tribal Government ..... 622110, 562213, 611310 ............ State/local hospitals, other health care facilities, State/local waste dis-
posal services, State universities.
This table is not intended to be CBI will not be disclosed except in 3. Docket
exhaustive, but rather provides a guide accordance with procedures set forth in The docket number for the proposed
for readers regarding entities likely to be 40 CFR part 2. action regarding the HMIWI NSPS (40
affected by the proposed action. To If you have any questions about CBI CFR part 60, subpart Ec) and emission
determine whether your facility would or the procedures for claiming CBI, guidelines (40 CFR part 60, subpart Ce)
be affected by the proposed action, you please consult the person identified in is Docket ID No. EPA–HQ–OAR–2006–
should examine the applicability the FOR FURTHER INFORMATION CONTACT 0534.
criteria in 40 CFR 60.50c of subpart Ec section.
and 40 CFR 60.32e of subpart Ce. If you 4. Worldwide Web (WWW)
have any questions regarding the 2. Tips for Preparing Your Comments In addition to being available in the
applicability of the proposed action to a When submitting comments, docket, an electronic copy of this
particular entity, contact the person remember to: proposed action is available on the
listed in the preceding FOR FURTHER WWW through the Technology Transfer
a. Identify the rulemaking by docket
INFORMATION CONTACT section. Network Web site (TTN Web).
number and other identifying
information (subject heading, Federal Following signature, EPA posted a copy
B. What should I consider as I prepare
Register date and page number). of the proposed action on the TTN’s
my comments?
policy and guidance page for newly
1. Submitting CBI b. Follow directions. The EPA may proposed or promulgated rules at
ask you to respond to specific questions http://www.epa.gov/ttn/oarpg. The TTN
Do not submit information that you or organize comments by referencing a provides information and technology
consider to be CBI electronically Code of Federal Regulations (CFR) part exchange in various areas of air
through www.regulations.gov or e-mail. or section number. pollution control.
Send or deliver information identified c. Explain why you agree or disagree;
as CBI to only the following address: suggest alternatives and substitute II. Background
Ms. Mary Johnson, c/o OAQPS language for your requested changes. Section 129 of the CAA, entitled
Document Control Officer (Room C404– d. Describe any assumptions and ‘‘Solid Waste Combustion,’’ requires
02), U.S. EPA, Research Triangle Park, provide any technical information and/ EPA to develop and adopt NSPS and
NC 27711, Attention Docket ID No. or data that you used. emission guidelines for solid waste
EPA–HQ–OAR–2006–0534. Clearly incineration units pursuant to CAA
e. If you estimate potential costs or
mark the part or all of the information sections 111 and 129. Sections 111(b)
burdens, explain how you arrived at
that you claim to be CBI. For CBI and 129(a) of the CAA (NSPS program)
your estimate in sufficient detail to
information in a disk or CD ROM that address emissions from new HMIWI
allow for it to be reproduced.
you mail to EPA, mark the outside of the units, and CAA sections 111(d) and
disk or CD ROM as CBI and then f. Provide specific examples to 129(b) (emission guidelines program)
identify electronically within the disk or illustrate your concerns, and suggest address emissions from existing HMIWI
CD ROM the specific information that is alternatives. units. The NSPS are directly enforceable
claimed as CBI. In addition to one g. Explain your views as clearly as Federal regulations. The emission
complete version of the comment that possible, avoiding the use of profanity guidelines are not directly enforceable
includes information claimed as CBI, a or personal threats. but, rather, are implemented by State air
copy of the comment that does not h. Make sure to submit your pollution control agencies through
contain the information claimed as CBI comments by the comment period sections 111(d)/129 State plans.
must be submitted for inclusion in the deadline identified in the preceding An HMIWI is defined as any device
public docket. Information marked as section titled DATES. used to burn hospital waste or medical/
infectious waste. Hospital waste means NOX, or metals (Pb, Cd, and Hg). The 2- environmental impacts and energy
discards generated at a hospital, and second combustion level includes a requirements, determines is achievable
medical/infectious waste means any minimum secondary chamber for new and existing units in each
waste generated in the diagnosis, temperature of 1800 °F and residence category.’’ This level of control is
treatment, or immunization of human time of 2 seconds. These combustion referred to as a maximum achievable
beings or animals, in research pertaining conditions will provide additional control technology, or MACT standard.
thereto, or in the production or testing control of CDD/CDF, CO, and PM, but In promulgating a MACT standard,
of biologicals (e.g., vaccines, cultures, will not reduce emissions of acid gases EPA must first calculate the minimum
blood or blood products, human (HCl and SO2), NOX, or metals (Pb, Cd, stringency levels for new and existing
pathological waste, sharps). Hospital/ and Hg). The 2-second combustion solid waste incineration units in a
medical/infectious waste does not conditions are considered to be the best category, generally based on levels of
include household waste, hazardous level of combustion control (i.e., good emissions control achieved or required
waste, or human and animal remains combustion) that is applied to HMIWI. to be achieved by the subject units. The
not generated as medical waste. An Wet scrubbers and dry scrubbers minimum level of stringency is called
HMIWI typically is a small, dual- provide control of PM, CDD/CDF, HCl, the MACT floor, and CAA section
chamber incinerator that burns about and metals, but do not influence CO, 129(a)(2) provides that the ‘‘degree of
800 pounds per hour (lb/hr) of waste. SO2 (at the low concentrations emitted reduction in emissions that is deemed
Smaller units burn as little as 13 lb/hr by HMIWI units), or NOX; in fact, there achievable for new units in a category
while larger units burn as much as 3,700 are no technologies currently used by shall not be less stringent than the
lb/hr. HMIWI that will consistently reduce emissions control that is achieved in
Incineration of hospital/medical/ SO2 or NOX emissions. (See Legacy practice by the best controlled similar
infectious waste causes the release of a Docket ID No. A–91–61, item II-A–111; unit, as determined by the
wide array of air pollutants, some of 60 FR 10669, 10671–10677; and 61 FR Administrator. Emissions standards for
which exist in the waste feed material 31742–31743.) existing units in a category may be less
and are released unchanged during On September 15, 1997, EPA adopted stringent than standards for new units
combustion, and some of which are NSPS (40 CFR part 60, subpart Ec) and in the same category but shall not be
generated as a result of the combustion emission guidelines (40 CFR part 60, less stringent than the average emissions
process itself. These pollutants include subpart Ce) for entities which operate limitation achieved by the best
particulate matter (PM); heavy metals, HMIWI. The NSPS and emission performing 12 percent of units in the
including lead (Pb), cadmium (Cd), and guidelines are designed to reduce air category.’’
mercury (Hg); toxic organics, including pollution emitted from new and existing The minimum stringency
chlorinated dibenzo-p-dioxins/ HMIWI, including HCl, CO, Pb, Cd, Hg, requirements form the first and least
dibenzofurans (CDD/CDF); carbon PM, CDD/CDF (total, or 2,3,7,8- stringent regulatory option EPA must
monoxide (CO); nitrogen oxides (NOX); tetrachlorinated dibenzo-p-dioxin toxic consider in the determination of MACT
and acid gases, including hydrogen equivalent (TEQ)), NOX, SO2, and for a source category. EPA must also
chloride (HCl) and sulfur dioxide (SO2). opacity. The NSPS apply to HMIWI for determine whether to control emissions
In addition to the use of good which construction began after June 20, ‘‘beyond the floor,’’ after considering the
combustion control practices, HMIWI 1996, or for which modification began costs, non-air quality health and
units are typically controlled by wet after March 16, 1998. The NSPS became environmental impacts, and energy
scrubbers or dry sorbent injection fabric effective on March 16, 1998, and its requirements of such more stringent
filters (dry scrubbers). requirements apply as of that date or at control. These are the two steps EPA
Combustion control includes the start-up of a HMIWI unit, whichever is took in the 1997 HMIWI rulemaking.
proper design, construction, operation, later. The emission guidelines apply to Finally, every 5 years after adopting a
and maintenance of HMIWI to destroy HMIWI for which construction began on MACT standard under section 129, CAA
or prevent the formation of air or before June 20, 1996, and required section 129(a)(5) requires EPA to review
pollutants prior to their release to the compliance by September 2002. and, if appropriate, revise the
atmosphere. Test data indicate that as CAA section 129 requires EPA to incinerator standards. In addition to
secondary chamber residence time and establish technology-based emission responding to the Court’s remand in
temperature increase, emissions standards that reflect levels of control Sierra Club v. EPA, 167 F.3d 658 (D.C.
decrease. Combustion control is most EPA determines are achievable for new Cir. 1999), this proposed action includes
effective in reducing CDD/CDF, PM, and and existing units, after considering our first set of proposed revisions to the
CO emissions. The 0.25-second costs, non-air quality health and HMIWI standards, also known as the 5-
combustion level includes a minimum environmental impacts, and energy year review.
secondary chamber temperature of requirements associated with the
1700 °F and a 0.25-second secondary implementation of the standards. III. Summary
chamber residence time. These In setting forth the methodology EPA A. Litigation and Proposed Remand
combustion conditions are typical of must use to establish the technology- Response
older HMIWI. The 1-second combustion based performance standards and
level includes a minimum secondary emissions guidelines, CAA section 1. What was EPA’s general methodology
chamber temperature of 1700 °F and 129(a)(2) provides that standards for determining MACT?
residence time of 1 second. These ‘‘applicable to solid waste incineration The methodology used to determine
combustion conditions are typical of units promulgated under section 111 MACT is similar for source categories
newer HMIWI. Compared to 0.25- and this section shall reflect the under sections 112 and 129 of the CAA.
second combustion, 1-second maximum degree of reduction in However, because each source category
combustion will achieve substantial emissions of [certain listed air is unique and the data available to
reductions in CDD/CDF and CO pollutants] that the Administrator, determine the performance capabilities
emissions, and will provide some taking into consideration the cost of of technology can vary from one source
control of PM, but will not reduce achieving such emission reduction, and category to another, the basic
emissions of acid gases (HCl and SO2), any non-air quality health and methodology must be adapted to fit the
source category in question. As the option. Sometimes, a direct calculation of the relatively small database (60 FR
Court pointed out in the HMIWI of the actual emissions values from the 10686).
litigation, it ‘‘generally defer[s] to an best performing 12 percent of sources a. EPA’s Methodology for New
agency’s decision to proceed on the provides the basis for this regulatory HMIWI. In determining the least
basis of imperfect scientific information, option. More often, EPA determines the stringent regulatory option allowed by
rather than to ‘invest the resources to technology used by the average source the CAA for new HMIWI, EPA first
conduct the perfect study.’ ’’ Sierra Club in the best performing 12 percent of examined the data available for various
v. EPA, 167 F.3d at 662. sources and establishes the floor based air pollution control technologies
In general, all MACT analyses involve on the technology assessment for that applied to HMIWI to determine the
an assessment of the air pollution average source. More stringent performance capabilities of the
control systems or technologies used by regulatory options reflect other technologies (i.e., the achievable
the better performing units in a source technologies capable of achieving better emission limitations) (60 FR 10671–73,
category. The technology assessment performance. 61 FR 31741–43). To determine the
can be based solely on actual emissions performance capabilities, EPA grouped
data, on knowledge of the air pollution 2. What was EPA’s methodology in the all of the test data by control technology
control in place in combination with 1997 HMIWI rulemaking? and established the numerical value for
actual emissions data, or on State On February 27, 1995, EPA published the achievable emission limitations
regulatory requirements, which give an a notice of proposed rulemaking somewhat higher than the highest test
indication of the actual performance of regarding emissions standards for data point for each particular control
the regulated units. For each source HMIWI units (60 FR 10654). The technology. (See Legacy Docket ID No.
category, the assessment of the proposal was the result of several years A–91–61, items IV–B–46, 47, 48, and
technology involves a review of actual of reviewing available information. 49.) Following the determination of
emissions data with an appropriate During the public comment period for performance capability, EPA identified
accounting for emissions variability. the proposal, EPA received over 700 the best control technology for each air
Where there is more than one method or letters, some of which contained new pollutant for each subcategory of
technology to control emissions, the information or indicated that the HMIWI, and established the numerical
analysis results in a series of potential commenters were in the process of values for the least stringent regulatory
regulations (called regulatory options), gathering more information for EPA to option at the achievable emission
one of which is selected as MACT. consider. The new information led EPA limitation associated with that
The first regulatory option considered to consider the need for numerous particular control technology. (See 60
by EPA must be at least as stringent as changes to the proposed rule, and on FR 10673; Legacy Docket ID No. A–91–
the CAA’s minimum stringency June 20, 1996, the Agency published a 61, item IV–B–38; 61 FR 31745–46.)
requirements. However, MACT is not Other, more stringent, regulatory
re-proposal (61 FR 31736). Following an
necessarily the least stringent regulatory options were developed reflecting the
additional public comment period, EPA
option. EPA must examine more actual performance of other, more
published the final rule on September
stringent regulatory options to effective, control technologies (61 FR
15, 1997 (62 FR 48348).
determine MACT. Unlike the minimum 31766–68).
stringency requirements, EPA must During the data-gathering phase of As stated in the 1996 re-proposal, the
consider various impacts of the more developing the 1995 proposal, EPA least stringent regulatory option for new
stringent regulatory options in found it difficult to obtain an accurate large HMIWI units (units with
determining MACT. Only if the more count of the thousands of HMIWI units maximum waste burning capacity of
stringent regulatory options are nationwide, or to find HMIWI units more than 500 lb/hr) was based on good
considered to have unreasonable with add-on air pollution control combustion (i.e., 2-second combustion
impacts does EPA select the first ‘‘floor- systems in place. A few HMIWI units level) and a combination of two control
based’’ regulatory option as MACT. with combustion control were tested to technologies, high-efficiency wet
As stated earlier, the CAA requires assess performance of combustion scrubbers and dry injection/fabric filter
that MACT for new sources be no less control in reducing emissions. One unit dry scrubbers with carbon (61 FR
stringent than the emissions control with a wet scrubber, and a few units 31746). New medium units (units with
achieved in practice by the best with dry scrubbing systems were tested maximum waste burning capacity of
controlled similar unit. After EPA’s to determine performance capabilities of more than 200 lb/hr but less than or
assessment of technology, EPA add-on controls. (See 61 FR 31738.) equal to 500 lb/hr) would need to use
determines the best control currently in Altogether, data were available from good combustion and a combination of
use for a given pollutant and establishes only 7 out of the estimated then- two control technologies, high-
one potential regulatory option at the operating 3,700 existing HMIWI units efficiency wet scrubbers and dry
emission level achievable by that (60 FR 10674). Because EPA was under injection/fabric filter dry scrubbers
control. More stringent potential a court-ordered deadline to propose and without carbon, to meet the least
regulatory options might reflect controls adopt standards for HMIWI that did not stringent regulatory option. Id. New
used on other sources that could be provide sufficient time to collect more small units (units with maximum waste
applied to the source category in actual emissions data (see consent burning capacity of less than or equal to
question. decree entered in Sierra Club v. EPA, 200 lb/hr) would need to use good
For existing sources, the CAA requires Nos. CV–92–2093 and CV–93–0284 combustion and a moderate-efficiency
that MACT be no less stringent than the (E.D.N.Y.)), EPA proceeded to develop wet scrubber to meet the least stringent
average emissions limitation achieved the regulations with the existing data, as regulatory option. Id.
by the best performing 12 percent of described below. However, EPA In EPA’s final standards promulgated
units in a source category. EPA must specifically requested comment on in 1997, EPA selected an overall more
determine some measure of the average EPA’s MACT determinations and on stringent regulatory option for new
emissions limitation achieved by the EPA’s conclusions about the HMIWI (62 FR 48365). The final
best performing 12 percent of units to performance capabilities of air pollution standards were based on emission limits
form the least stringent regulatory control technologies on HMIWI in light achievable with good combustion and a
moderate-efficiency wet scrubber for with ‘‘combustion control.’’ (See 60 FR could also be used with good
new small HMIWI, and good 10674; 61 FR 31745; Legacy Docket ID combustion at large and medium
combustion and a combined dry/wet No. A–91–61, item IV–B–24 at 2.) In existing units (61 FR 31745). EPA
control system with carbon for new previous Federal Register notices further stated that its inclination was to
medium and large HMIWI. Id. These regarding HMIWI (60 FR 10654, 61 FR establish emission limitations for large
standards reflected the MACT floor 31736, and 62 FR 48348), this level of and medium existing units based on
emissions levels for new small and large control was referred to as regulatory options representing the
HMIWI, but were more stringent than ‘‘uncontrolled,’’ which is misleading MACT floors (61 FR 31778). Small
the MACT floor for new medium because sources with combustion existing units would need only to use
HMIWI. Id. EPA estimated that the control emit lesser amounts of CDD/ good combustion practices to meet the
standards would reduce emissions from CDF, CO, and PM. In the latter situation regulatory option representing the
these units of HCl by up to 98 percent, described above, the average of the MACT floor (61 FR 31745). With respect
PM and Pb by up to 92 percent, Cd by regulatory limits plus enough to small existing units, EPA stated that
up to 91 percent, CDD/CDF by up to 87 combustion-controlled emission values it had no inclination with regard to
percent, Hg by up to 74 percent, and to account for 12 percent of units in the which regulatory option should be used
CO, SO2, and NOX by up to 52 percent subcategory was calculated. (See Legacy to establish emission limitations and
(62 FR 48366). Docket ID No. A–91–61, item IV–B–24 requested comment on requiring use of
at 2–4.) good combustion and a low-efficiency
b. EPA’s Methodology for Existing After calculating the averages of
HMIWI. For existing units, EPA did not wet scrubber (61 FR 31778–79).
regulatory limits and combustion-
have sufficient emissions data to fully controlled emission values, EPA In EPA’s final standards promulgated
characterize the actual emissions examined the resulting calculated in 1997, EPA selected an overall more
performance of the best performing 12 values to determine what level of air stringent regulatory option for existing
percent of existing HMIWI, and, based pollution control would be needed to HMIWI (62 FR 48371). The final
exclusively on such data, EPA did not meet the calculated average values. (See standards were based on emission limits
have a clear indication of the technology 60 FR 10675–78; 61 FR 31755–56.) For achievable with good combustion and a
used by the best 12 percent of units. As many pollutants, the calculated averages low-efficiency wet scrubber for most
a result, EPA used emission limits presented no clear indication of the type existing small HMIWI, good combustion
included in State regulations and State- of air pollution control used by the best and a moderate-efficiency wet scrubber
issued permits (hereinafter referred to as performing units. However, the for existing medium HMIWI, and good
regulatory limits) as surrogate calculated values for three key combustion and a high-efficiency wet
information to determine emissions pollutants, PM, CO, and HCl, did scrubber for existing large HMIWI (62
limitations achieved by the best provide a good indication of the type of FR 48371). The final standards allow
performing 12 percent of units in each air pollution control used on the best small HMIWI that meet certain rural
subcategory (60 FR 10674). EPA performing 12 percent of units. The criteria to meet emissions limits
believed this information could be level of air pollution control associated achievable with good combustion alone.
expected to reliably reflect levels of with the calculated average values for Id. These standards reflected the MACT
performance achieved on a continuous PM, CO, and HCl formed the technical floor emissions levels for existing small
basis by better-controlled units that basis of the least stringent regulatory HMIWI meeting rural criteria, medium
must meet these limits or risk violating option considered by EPA (61 FR 31756, HMIWI, and large HMIWI, but were
enforceable requirements. EPA assumed Table 13). The emission limitations more stringent than the MACT floor for
that all HMIWI were achieving their assigned to each pollutant reflected the most existing small HMIWI (i.e., non-
regulatory limits (60 FR 10674). Where actual performance of the technology on rural) (62 FR 48371–72). The final
there were regulatory limits for more which they were based. Finally, EPA standards for existing medium and large
than 12 percent of units in a developed a series of regulatory options HMIWI were structured so that either a
subcategory, the regulatory limits were based on progressively more stringent dry scrubber or a wet scrubber could be
ranked from the most stringent to least technologies and assigned emission used to achieve the emission limits.
stringent, and the average of the limitations to each regulatory option EPA estimated that the final emission
regulatory limits for the top 12 percent based on the actual performance guidelines would reduce emissions of
of units in the subcategory was capabilities of the technologies (61 FR CDD/CDF by up to 97 percent, Hg by up
calculated. Id.; 61 FR 31744–45. Where 31757, Table 14). to 95 percent, PM by up to 92 percent,
the number of units subject to specific As stated in the 1996 re-proposal, Pb by up to 87 percent, Cd by up to 84
emissions limitations did not comprise large existing units would need to use percent, CO by up to 82 percent, HCl by
12 percent of the population in a good combustion and a high-efficiency up to 98 percent, and SO2 and NOX by
subcategory, EPA assumed those units wet scrubber to meet the least stringent up to 30 percent (62 FR 48372). Table
with regulatory limits were the best regulatory option, while medium 1 of this preamble summarizes the
performing units, and the remaining existing units would need to use good emission limits for the NSPS and
units in the top 12 percent were combustion and a moderate-efficiency emission guidelines promulgated in
assigned an emission value associated wet scrubber, although dry scrubbers 1997.
TABLE 1.—SUMMARY OF PROMULGATED EMISSION LIMITS

Pollutant (units) Unit Size 1 Limit for existing HMIWI 2 Limit for new HMIWI 2
HCl (parts per million by volume (ppmv)) ........... L, M, S ........ 100 or 93% reduction .......................................... 15 or 99% reduction.
SR ............... 3,100 ................................................................... N/A.3
CO (ppmv) ........................................................... L, M, S ........ 40 ........................................................................ 40
SR ............... 40 ........................................................................ N/A.
Pb (milligrams per dry standard cubic meter L, M ............. 1.2 or 70% reduction ........................................... 0.07 or 98% reduction.3
(mg/dscm)).
TABLE 1.—SUMMARY OF PROMULGATED EMISSION LIMITS—Continued

Pollutant (units) Unit Size 1 Limit for existing HMIWI 2 Limit for new HMIWI 2
S .................. 1.2 or 70% reduction ........................................... 1.2 or 70% reduction.

SR ............... 10 ........................................................................ N/A.
Cd (mg/dscm) ...................................................... L, M ............. 0.16 or 65% reduction ......................................... 0.04 or 90% reduction.
S .................. 0.16 or 65% reduction ......................................... 0.16 or 65% reduction.
SR ............... 4 .......................................................................... N/A.
Hg (mg/dscm) ...................................................... L, M, S ........ 0.55 or 85% reduction ......................................... 0.55 or 85% reduction.
SR ............... 7.5 ....................................................................... N/A.
PM (grains per dry standard cubic foot (gr/dscf)) L .................. 0.015 ................................................................... 0.015
M ................. 0.03 ..................................................................... 0.015
S .................. 0.05. .................................................................... 0.03.
SR ............... 0.086 ................................................................... N/A.
CDD/CDF, total (nanograms per dry standard L, M ............. 125 ...................................................................... 25
cubic meter (ng/dscm)).
S .................. 125 ...................................................................... 125
SR ............... 800 ...................................................................... N/A.
CDD/CDF, TEQ (ng/dscm) .................................. L, M ............. 2.3 ....................................................................... 0.6
S .................. 2.3 ....................................................................... 2.3
SR ............... 15 ........................................................................ N/A.
NOX (ppmv) ......................................................... L, M, S ........ 250 ...................................................................... 250
SR ............... 250 ...................................................................... N/A.
SO2 (ppmv) .......................................................... L, M, S ........ 55 ........................................................................ 55
SR ............... 55 ........................................................................ N/A.
1L = Large; M = Medium; S = Small; SR = Small Rural
2 Allemission limits are measured at 7 percent oxygen.
3 Not applicable.
c. Compliance by HMIWI. At the time units and 10 new commercial units of which meet the rural criteria).
of promulgation (September 1997), EPA would be constructed. Twenty-one percent of the existing
estimated that there were approximately After shutdown of approximately 97 HMIWI are commercially owned. Of the
2,400 HMIWI operating in the United percent of the 2,400 HMIWI that were four new HMIWI, three are large units,
States. Those units combusted operating in 1997, there are currently 72 and one is a medium unit. Two of the
approximately 830 thousand tons of existing HMIWI at 67 facilities. new units are county-owned but accept
hospital/medical/infectious waste Additionally, only 4 new HMIWI at 3 waste from other sources, similar to
annually. Of those existing HMIWI, facilities began operation following the commercial units. The actual emissions
about 48 percent were small units, 29 1997 rulemaking. These 76 existing and reductions achieved as a result of
percent were medium units, and 20 new units are estimated to combust implementation of the standards
percent were large units. About 3 approximately 165 thousand tons of exceeded the 1997 projections for all
percent of the HMIWI were commercial waste annually. Of the 72 existing nine of the regulated pollutants. A
units. EPA projected that no new small HMIWI subject to the emission comparison of the estimated pollutant
or medium HMIWI would be guidelines, 44 are large units, 20 are reductions versus the actual reductions
constructed, and that up to 60 new large medium units, and 8 are small units (6 is presented in Table 2 of this preamble.
TABLE 2.—COMPARISON OF ESTIMATED POLLUTANT REDUCTIONS VERSUS ACTUAL POLLUTANT REDUCTIONS

Estimated emissions reduction, Actual emissions
Pollutant percent reduction, percent 1
HCl .......................................................................................... 98 ............................................................................................ 99.2

CO .......................................................................................... 75 to 82 .................................................................................. 98.1
Pb ........................................................................................... 80 to 87 .................................................................................. 98.7
Cd ........................................................................................... 75 to 84 .................................................................................. 99.0
Hg ........................................................................................... 93 to 95 .................................................................................. 99.0
PM .......................................................................................... 88 to 92 .................................................................................. 98.1
CDD/CDF, total ....................................................................... 96 to 97 .................................................................................. 99.5
CDD/CDF, TEQ ...................................................................... 95 to 97 .................................................................................. 99.6
NOX ........................................................................................ 0 to 30 .................................................................................... 70.6
SO2 ......................................................................................... 0 to 30 .................................................................................... 92.6
1 Reflects the effect of unit shutdowns as well as the effect of compliance with the promulgated standards.
3. What was the Sierra Club’s challenge? Club claimed that EPA had violated not including mandatory pollution
CAA section 129 by setting emission prevention or waste minimization

On November 14, 1997, the Sierra standards for HMIWI under CAA requirements in the HMIWI standards;
Club and the Natural Resources Defense sections 129 and 111 that are less and that EPA had not adequately
Council (Sierra Club) filed suit in the stringent than the statutory minimum considered the non-air quality health
U.S. Court of Appeals for the District of stringency required by section 129(a)(2); and environmental impacts of the
Columbia Circuit (the Court). The Sierra that EPA had violated section 129 by standards. For new units, the Sierra
Club argued that to satisfy the statutory performance of the ‘best controlled b. The Court’s Ruling on Existing
phrase ‘‘best controlled similar unit’’ in similar unit’ under the worst reasonably Units. With respect to existing units, the
CAA section 129(a)(2), EPA should have foreseeable circumstances [* * *]. It is Court first rejected the Sierra Club’s
identified the single best performing reasonable to suppose that if an ‘‘claim that EPA’s decision to base the
unit in each subcategory and based the emissions standard is as stringent as ‘the floors on regulatory data fails the first
MACT floor for that subcategory on that emissions control that is achieved in step of the Chevron test. None of the
particular unit’s performance, rather practice’ by a particular unit, then that Sierra Club’s arguments establish that
than consider the performance of other particular unit will not violate the Congress has ‘directly addressed’ and
units using the same technology. The standard. This only results if ‘achieved rejected the use of regulatory data.’’ Id.,
Sierra Club also argued that EPA in practice’ is interpreted to mean at 661. After noting that the Sierra
erroneously based the new unit floors ‘achieved under the worst foreseeable Club’s statutory objections to EPA’s
on the emissions of the worst circumstances.’ In National Lime Ass’n methodology appeared to be premised
performing unit using a particular v. EPA, 627 F.2d 416, 431 n. 46 (D.C. on ‘‘the counterintuitive proposition
technology. Regarding existing units, Cir. 1980), we said that where a statute that an ‘achieved’ level may not be
the Sierra Club claimed that the plain requires that a standard be ‘achievable,’ ‘achievable,’ or, as Sierra Club puts it,
meaning of CAA section 129(a)(2)’s it must be achievable ‘under most may be better than ‘EPA’s notions about
words, ‘‘average emissions limitation adverse circumstances which can what is achievable,’ ’’ id. at 662, the
achieved by the best performing 12 reasonably be expected to recur.’ The Court rejected the Sierra Club’s statutory
percent of units,’’ precludes the use of same principle should apply when a objections to using regulatory data and
regulatory data, and claimed that the standard is to be derived from the uncontrolled (i.e., combustion-
legislative history of section 129(a)(2) operating characteristics of a particular controlled) emissions values. In other
reflects congressional intent to prohibit unit.’’ Sierra Club v. EPA, 167 F.3d at words, the Court implicitly embraced
EPA from relying on regulatory data. 665. Thus, the Court refused to embrace EPA’s view, under the principle of
Moreover, the Sierra Club claimed that, the Sierra Club’s interpretation of CAA National Lime, that the MACT floor is
for HMIWI, using regulatory data was section 129(a)(2) as requiring EPA to premised on the fundamental concept
impossible because such data existed for base the MACT floor on only the lowest that it be ‘‘achievable,’’ and should not
fewer than 12 percent of units, and emissions data points observed (i.e., the
be set at a level that happens to be
because doing so would impermissibly level achieved by the best performing
reflected by the lowest observed data
import an achievability requirement unit for each pollutant).
Relating to the Sierra Club’s claim point without consideration of
into the unit floor determination.
that EPA erred in considering the variability in operating conditions.
Finally, the Sierra Club argued that EPA
emissions of units other than the best Then, after analyzing and rejecting the
failed to require HMIWI units to
controlled unit, the Court refused to rule Sierra Club’s arguments that the plain
undertake programs to reduce the Hg
that EPA’s approach was unlawful, and language of the CAA and its legislative
and chlorinated plastic in their waste
posited that ‘‘[p]erhaps considering all history forbid EPA’s methodology, the
streams, in violation of CAA section
units with the same technology is Court further ruled that it found
129(a)(3), and that EPA failed to
consider the fact that CDD/CDF and Hg justifiable because the best way to ‘‘nothing inherently impermissible
from HMIWI can contaminate water, predict the worst reasonably foreseeable about construing the statute to permit
sediment, and soil, and can performance of the best unit with the the use of regulatory data—if they allow
bioaccumulate in food, in violation of available data is to look at other units’ EPA to make a reasonable estimate of
the CAA’s requirement that EPA performance. Or perhaps EPA the performance of the top 12 percent of
consider non-air quality impacts of reasonably considered all units with the units. Indeed, the Sierra Club conceded
setting HMIWI emissions standards. same technology equally ‘well- at oral argument that ‘a reasonable
controlled,’ so that each unit with the sample’ may be used ‘to find out what
4. What was the Court’s ruling? best technology is a ‘best-controlled the best 12 percent are doing.’ Oral Arg.
On March 2, 1999, the Court issued its unit’ even if such units vary widely in Tr. at 11. To be sure, the Sierra Club did
opinion in Sierra Club v. EPA, 167 F.3d performance.’’ Sierra Club v. EPA, 167 not concede that permit data may be
658 (D.C. Cir. 1999). While the Court F.3d at 665. used. But neither has it provided any
rejected the Sierra Club’s claims However, the Court concluded that basis for believing that state and local
regarding pollution prevention and non- the possible rationale for this treatment limitations are such weak indicators of
air quality impacts, and rejected the of new units was not presented in the performance that using them is
Sierra Club’s statutory arguments under rulemaking record with enough clarity necessarily an impossible stretch of the
CAA section 129, the Court remanded for the Court to determine that EPA’s statutory terms. [* * *] We therefore
the rule to EPA for further explanation ‘‘path may reasonably be discerned.’’ Id. reject the Sierra Club’s argument that
regarding how EPA derived the MACT Moreover, the Court ruled that EPA had the CAA forbids the use of permit and
floors for new and existing HMIWI ‘‘not explained why the phrase best regulatory data, and hold that the use of
units. Furthermore, the Court did not controlled similar unit encompasses all such information is permissible as long
vacate the regulations, stating that ‘‘[i]t units using the same technology as the as it allows a reasonable inference as to
is possible that EPA may be able to unit with the best observed the performance of the top 12 percent of
explain [EPA’s basis for the standards]’’ performance, rather than just that unit units. Similarly, as long as there is a
in response to the concerns raised by itself[. * * * W]e do not know what reasonable basis for believing that some
the Court. Id., at 664. The regulations interpretation the agency chose, and of the best performing 12 percent of
remain in effect during the remand. thus cannot evaluate its choice.’’ Sierra units are uncontrolled [i.e., combustion
a. The Court’s Ruling on New Units. Club v. EPA, 167 F.3d at 665. The Court controlled], EPA may include data
In response to the Sierra Club’s claims further directed EPA to provide points giving a reasonable
regarding EPA’s treatment of new units, additional explanation regarding how representation of the performance of
the Court opined that ‘‘EPA would be the Agency had calculated the upper those units in its averaging.’’ Sierra Club
justified in setting the floors at a level bound of the best-controlled unit’s v. EPA, 167 F.3d at 662, 663. Thus, the
that is a reasonable estimate of the performance through rounding. Id. Court rejected all of the Sierra Club’s
arguments that the CAA prohibits EPA the * * * floor at the level of the worst best performing sources depends on
from basing MACT floor determinations performing plant in its databases using whether factors other than MACT
on permit or regulatory data, or on th[e same] technology [as the median control contribute to emissions[,]’’ id.,
uncontrolled (i.e., combustion- plant]’’ had not been shown by the and that ‘‘the relevant question here is
controlled) emissions values. Sierra Club to reflect a not reasonable not whether control technologies
However, in addressing the manner in estimate. NLA II, 233 F.3d at 633. experience variability at all, but whether
which EPA had specifically relied upon In addition, the Court partially the variability experienced by the best-
such data in the HMIWI rulemaking, the clarified its position regarding EPA’s performing sources can be estimated by
Court concluded that ‘‘[a]lthough EPA approach of accounting for emissions relying on emissions data from the
said that it believed the combination of performance variability by setting floors worst-performing sources using the
regulatory and uncontrolled [i.e., at a level that reasonably estimates ‘‘the MACT control.’’ Id., at 865.
combustion-controlled] data gave an performance of the ‘best controlled
accurate picture of the relevant similar unit’ under the worst reasonably In the specific case of the HWC rule,
[HMIWI]s‘ performance, it never foreseeable circumstances.’’ Sierra Club, the Court concluded that, since record
adequately said why it believes this. 167 F.3d at 665. In NLA II, the Court evidence showed that non-MACT
[* * *] First, EPA has said nothing stressed that EPA should not simply set factors influenced emissions
about the possibility that [HMIWI]s floors at levels reflecting the worst performance, EPA could not base floors
might be substantially overachieving the foreseeable circumstances faced by any simply on the worst-performing MACT
permit limits. If this were the case, the worst performing unit in a given source sources’ emissions. Id., at 866. However,
permit limits would be of little value in category, and that while considering all the Court also reiterated that ‘‘[i]f in the
estimating the top 12 percent of units with the same technology may be case of a particular source category or
[HMIWI]s’ performance. [* * *] justifiable because the best way to HAP, the Agency can demonstrate with
Second, EPA never gave any reason for predict the worst reasonably foreseeable substantial evidence—not mere
its apparent belief that [HMIWI]s that performance of the best unit with assertions—that MACT technology
were not subject to permit requirements available data is to look at other units’ significantly controls emissions, or that
did not deploy emission controls of any performance, such an approach would factors other than the control have a
sort. Unless there is some finding to this satisfy the CAA ‘‘if pollution control negligible effect, the MACT approach
effect, it is difficult to see the rationality technology were the only factor could be a reasonable means of
in using ‘uncontrolled’ [i.e., determining emission levels of that satisfying the statute’s requirements.’’
combustion-controlled] data for the HAP.’’ NLA II, 233 F.3d at 633. Id.
units that were not subject to regulatory In Cement Kiln Recycling Coalition v.
requirements.’’ Id., at 663–664. The EPA, 255 F.3d 855 (D.C. Cir. 2001) 5. Are revisions to the emission limits
Court further questioned the rationality (‘‘CKRC’’), the Court again refined its being proposed in response to the
of EPA using the highest of its test run view on when it is appropriate for EPA remand?
data in cases where the regulatory data to base MACT floors on the performance
Yes, the proposed response to the
did not alone comprise the necessary 12 of air pollution control technology. In
that case, the Sierra Club challenged remand would revise some of the
percent. Id., at 664.
c. Subsequent Court Rulings Relevant EPA’s MACT standards for hazardous emission limits in both the NSPS and
to the Remand. Following the Court’s waste combustors (HWC), and argued emission guidelines. Relative to the
remand of the HMIWI MACT floors in that factors other than MACT NSPS, the emission limits for CO, Pb,
Sierra Club v. EPA, the Court issued a technology influenced the emissions Cd, Hg, PM, and CDD/CDF would be
series of rulings in other cases performance of the best performing revised. Relative to the emission
addressing MACT rules that bear on sources. guidelines, the emission limits for HCl,
EPA’s proposed response regarding The Court agreed that since EPA’s Pb, Cd, and CDD/CDF would be revised.
HMIWI. The first of these was Nat’l record evidence in the HWC rulemaking EPA believes that the revised emission
Lime Ass’n v. EPA, 233 F.3d 625 (D.C. showed that factors besides MACT limits being proposed as a result of its
Cir. 2000) (‘‘NLA II’’), which involved controls significantly influenced HWC response to the remand can be achieved
challenges to EPA’s MACT standards emission rates, ’’emissions of the worst- with the same emission control
under CAA section 112(d) for portland performing MACT source may not technology currently used by HMIWI.
cement manufacturing facilities. In that reflect what the best-performers actually The proposed emission limits for the
case, the Sierra Club argued that EPA achieve.’’ CKRC, 255 F.3d at 864. EPA NSPS and emission guidelines
should have based its estimate of the top had claimed that MACT floors must be necessary to respond to the Court’s
performing 12 percent of sources on achievable by all sources using MACT remand are summarized in Table 3 of
actual emissions data, in order to technology, and that to account for the this preamble. Note that in several
‘‘reasonably estimate’’ such best-performing sources’’ operational cases, further amendments to the
performance. But the Court determined variability we had to base floors on the emission limits are being proposed as a
that EPA’s approach of selecting ‘‘the worst performers’’ emissions. But the result of our 5-year review under CAA
median [performing] plant out of the Court stressed that ‘‘whether variability section 129(a)(5). Those proposed
best twelve percent of the plants for in the MACT control accurately amendments are discussed in the
which it had information and set[ting] estimates variability associated with the following section of this preamble.
TABLE 3.—SUMMARY OF PROPOSED EMISSION LIMITS IN RESPONSE TO THE REMAND

Pollutant Proposed remand limit for Proposed remand limit for

Unit size 1
(units) existing HMIWI 2 new HMIWI 2
HCl (ppmv) ................................................................................... L, M, S ........ 78 or 93% reduction 3 ................. 153 or 99% reduction 3.
SR ............... 3,100 3 ......................................... N/A 4.
CO (ppmv) .................................................................................... L, M, S ........ 40 3 .............................................. 32
SR ............... 40 3 .............................................. N/A 4.
TABLE 3.—SUMMARY OF PROPOSED EMISSION LIMITS IN RESPONSE TO THE REMAND—Continued

Pollutant Proposed remand limit for Proposed remand limit for
Unit size 1
(units) existing HMIWI 2 new HMIWI 2
Pb (mg/dscm) ............................................................................... L, M ............. 0.78 or 71% reduction ................ 0.060 or 98% reduction 3.
S .................. 0.78 or 71% reduction ................ 0.78 or 71% reduction.
SR ............... 8.9 ............................................... N/A 4.
Cd (mg/dscm) ............................................................................... L, M ............. 0.11 or 66% reduction 3 .............. 0.030 or 93% reduction.
S .................. 0.11 or 66% reduction 3 .............. 0.11 or 66% reduction 3.
SR ............... 4 3 ................................................ N/A 4.
Hg (mg/dscm) ............................................................................... L, M ............. 0.55 3 or 87% reduction .............. 0.45 or 87% reduction.
S .................. 0.55 3 or 87% reduction .............. 0.47 or 87% reduction.
SR ............... 6.6 ............................................... N/A 4.
PM (gr/dscf) ................................................................................. L .................. 0.015 3 ......................................... 0.009
M ................. 0.030 3 ......................................... 0.009
S .................. 0.050 3 ......................................... 0.018
SR ............... 0.086 3 ......................................... N/A 4.
CDD/CDF, total (ng/dscm) ........................................................... L, M ............. 115 .............................................. 20
S .................. 115 .............................................. 111
SR ............... 800 3 ............................................ N/A 4.
CDD/CDF, TEQ (ng/dscm) .......................................................... L, M ............. 2.2 ............................................... 0.53
S .................. 2.2 ............................................... 2.1
SR ............... 15 3 .............................................. N/A 4.
NOX (ppmv) ................................................................................. L, M, S ........ 250 3 ............................................ 225
SR ............... 250 3 ............................................ N/A 4
SO2 (ppmv) .................................................................................. L, M, S ........ 55 3 .............................................. 46
SR ............... 55 3 .............................................. N/A 4.
2 Allemission limits are measured at 7 percent oxygen.
3 No change proposed.
4 Not applicable.
B. Proposed Amendments (CAA Section processes, but do reflect more efficient limits were based primarily on permit
129(a)(5) 5-Year Review) practices in operation of the control information and other regulatory
Section 129(a)(5) of the CAA requires technologies that sources used in order requirements, and not on actual
EPA to conduct a review of the NSPS to meet the 1997 MACT standards. performance or stack test data. To this
and emissions guidelines at 5 year Following year 2002 compliance with end, it was highly uncertain at
intervals and, if appropriate, revise the the emission guidelines, EPA gathered promulgation what the precise
NSPS and emission guidelines pursuant information on the performance levels performance efficiency and day-to-day
to the requirements under sections 111 actually being achieved by HMIWI that operational variability associated with
and 129 of the CAA. In conducting such were operating under the guidelines. the promulgated regulatory
reviews, EPA attempts to assess the After implementation of the guidelines requirements would yield. Thus, the
performance of and variability in 1997, approximately 94 percent of 2002 compliance test information
associated with the installed emissions HMIWI shut down, and 3 percent provided the first quantitative
control equipment (and developments demonstrated eligibility for exemptions assessment of the performance of the
in practices, processes and control from the HMIWI regulation. Those installed control equipment’s ability to
technologies) and to revise as necessary HMIWI that remained in operation attain the NSPS and emissions guideline
and appropriate the NSPS and emission either continued operation with their limits.
guidelines. In these reviews, EPA takes existing configuration or were retrofitted The goal of the current technology
into account the currently installed with add-on air pollution control review is to assess the performance
equipment and its performance and devices in order to meet the standards. efficiency of the installed equipment
operational variability. As appropriate, The retrofits were completed on time, and to ensure that the emission limits
we also consider new technologies that and the controls installed to meet the reflect the performance of the
have been demonstrated to reliably required emission limitations were technologies required by the MACT
control emissions from the source highly effective in reducing emissions of standards. In addition, the review
category. In setting numerical emission all of the CAA section 129 pollutants addresses whether new technologies
limits from single, ‘‘snap shot’’ stack test emitted by HMIWI. For those HMIWI, and processes and improvements in
data, EPA must exercise technical relative to a 1995 baseline, the emission practices have been demonstrated at
judgment to ensure the achievability of guidelines reduced organic emissions sources subject to the emissions
such limits over the course of (CDD/CDF) by about 90 percent, metals limitations. EPA’s intent for future
anticipated operating conditions. EPA emissions (Pb, Cd, and Hg) by more than technology reviews is to include similar
has completed the 5-year review, and 80 percent, and acid gas emissions (HCl analyses that also assess risk along with
the proposed amendments discussed and SO2) by more than 70 percent. new technologies. For the current
below reflect the changes that EPA has Including shutdowns and exemptions, review, while new technologies have
determined are appropriate in addition nationwide HMIWI emissions of not yet been demonstrated to reliably
to the amendments that are necessary to organics, metals, and acid gases each control emissions more efficiently at
respond to the Court’s remand. These decreased by about 99 percent or more reasonable cost at HMIWI units than
proposed amendments do not reflect relative to a 1995 baseline. It should be those used to meet MACT,
adoption of new control technologies or noted that the original HMIWI emission improvements in operational practices
do support some additional revision of levels currently being achieved by NSPS and emission guidelines. EPA’s
the standards, in order to better reflect HMIWI. The revisions discussed in the technology review demonstrates that the
the best operation of the MACT following text apply to both the NSPS proposed emission limits can be
controls. and the emission guidelines, unless achieved with the same emission
otherwise specified. control technology currently used by
These proposed amendments would
revise the NSPS and emission 1. Are revisions to the emission limits HMIWI. The proposed emission limits
guidelines, in some cases beyond the being proposed? for the NSPS and emission guidelines
point needed to respond to the Court’s are summarized in Tables 4 and 5 of this
Yes, the proposed amendments would
remand, based on the performance revise the emission limits in both the preamble.
TABLE 4.—SUMMARY OF PROPOSED 5-YEAR REVIEW EMISSION LIMITS FOR NEW HMIWI
Pollutant (units) Unit Size 1 Proposed Limit 2
HCl (ppmv) ....................................................................................................................................... L, M, S .............. 15 3 or 99% reduction 3.

CO (ppmv) ....................................................................................................................................... L, M, S .............. 25
Pb (mg/dscm) .................................................................................................................................. L, M .................. 0.060 or 99% reduction.
S ....................... 0.64 or 71% reduction.
Cd (mg/dscm) .................................................................................................................................. L, M .................. 0.0050 or 99% reduction.
S ....................... 0.060 or 74% reduction.
Hg (mg/dscm) .................................................................................................................................. L, M .................. 0.19 or 96% reduction.
S ....................... 0.33 or 96% reduction.
PM (gr/dscf) ..................................................................................................................................... L, M .................. 0.0090.
S ....................... 0.018.
CDD/CDF, total (ng/dscm) ............................................................................................................... L, M .................. 16
S ....................... 111
CDD/CDF, TEQ (ng/dscm) .............................................................................................................. L, M .................. 0.21
S ....................... 2.0
NOX (ppmv) ..................................................................................................................................... L, M, S .............. 212
SO2 (ppmv) ...................................................................................................................................... L, M .................. 21
S ....................... 28
1L = Large; M = Medium; S = Small
2 All
emission limits are measured at 7 percent oxygen.
TABLE 5.—SUMMARY OF PROPOSED 5-YEAR REVIEW EMISSION LIMITS FOR EXISTING HMIWI
Pollutant (units) Unit Size 1 Proposed Limit 2
HCl (ppm) ........................................................................................................................................ L, M, S .............. 51 or 94% reduction.

SR ..................... 398
CO (ppm) ......................................................................................................................................... All ..................... 25
Pb (mg/dscm) .................................................................................................................................. L, M, S .............. 0.64 or 71% reduction.
SR ..................... 0.60
Cd (mg/dscm) .................................................................................................................................. L, M, S .............. 0.060 or 74% reduction.
SR ..................... 0.050
Hg (mg/dscm) .................................................................................................................................. L, M, S .............. 0.33 or 96% reduction.
SR ..................... 0.253
PM (gr/dscf) ..................................................................................................................................... L ....................... 0.015
M ...................... 0.030 3
S ....................... 0.030
SR ..................... 0.030
CDD/CDF, total (ng/dscm) ............................................................................................................... L, M, S .............. 115
SR ..................... 800 3
CDD/CDF, TEQ (ng/dscm) .............................................................................................................. L, M, S .............. 2.0
SR ..................... 15 3
NOX (ppmv) ..................................................................................................................................... All ..................... 212
SO2 (ppmv) ...................................................................................................................................... All ..................... 28
2 All emission limits are measured at 7 percent oxygen.
As indicated by Table 5 of this of the waste stream of a small rural 2. Are other amendments being
preamble, the proposed emission limits hospital not including certain materials proposed?
for Pb, Cd, and Hg for existing small that are in the waste stream of a non-
rural HMIWI are more stringent than rural hospital and that cause relatively The proposed amendments would
those being proposed for existing large, higher Pb, Cd and Hg emissions. also make the following changes based
medium, and small HMIWI. We believe on information received during
that this better emissions performance implementation of the HMIWI NSPS
by existing small rural HMIWI is a result and emission guidelines and would
apply equally to the NSPS and emission consistent with CAA section 129, compliance deadlines regarding any
guidelines, unless otherwise specified. revised State plans containing the amended standards. The applicability
a. Performance Testing and revised emission limits and other date for the NSPS units, with respect to
Monitoring Amendments. The proposed requirements in the proposed the standards as promulgated in 1997,
amendments would allow sources to use amendments would be due within 1 remains June 20, 1996; however, units
the results of previous emissions tests to year after promulgation of the for which construction is commenced
demonstrate compliance with the amendments. That is, revised State after the date of this proposal, or
revised emission limits as long as the plans would have to be submitted to modification is commenced on or after
sources certify that the previous test EPA 1 year after the date on which EPA the date 6 months after promulgation of
results are representative of current promulgates revised standards. the amended standards, would be
operations. Only those sources whose The proposed amendments to the subject to more stringent NSPS emission
previous emissions tests do not emission guidelines then would allow limits than units for which construction
demonstrate compliance with one or HMIWI units up to 3 years from the date or modification was completed prior to
more revised emission limits would be of approval of a State plan, but not later those dates. Under the proposed
required to conduct another emissions than 5 years after promulgation of the amendments, units that commenced
test for those pollutants (note that revised standards, to demonstrate construction after June 20, 1996, and on
sources are already required to test for compliance with the amended or before February 6, 2007, or that are
HCl, CO, and PM on an annual basis). standards. Consistent with CAA section modified before the date 6 months after
The proposed amendments would 129, EPA expects States to require the date of promulgation of any revised
require, for existing HMIWI, annual compliance as expeditiously as final standards, would continue to be or
inspections of scrubbers and fabric practicable. HMIWI units have already would become subject to the NSPS
filters, and a one-time Method 22 visible installed the emission control emission limits that were promulgated
emissions test of the ash handling equipment necessary to meet the in 1997 and that remain in the 40 CFR
operations to be conducted during the proposed revised limits, and EPA, part 60, subpart Ec NSPS, except where
next compliance test. For new HMIWI, therefore, anticipates that most State the revised emission guidelines would
the proposed amendments would plans will include compliance dates be more stringent. In that case, HMIWI
require CO continuous emissions sooner than 5 years following that are NSPS units under the 1997 rule
monitoring systems (CEMS), bag leak promulgation of the amendments. In would also need to comply with the
detection systems for fabric-filter most cases, the only changes necessary revised emission guidelines for existing
controlled units, annual inspections of are to review the revisions and adjust sources, by the applicable compliance
scrubbers and fabric filters, and Method the emission monitoring and reporting date for such existing sources. Similarly,
22 visible emissions testing of the ash accordingly. emission guidelines units under the
handling operations to be conducted In revising the emission limits in a
1997 rule would need to meet the
during each compliance test. For State plan, a State has two options.
revised emission guidelines by the
existing HMIWI, use of CO CEMS would First, it could include both the current
applicable compliance date for the
be an approved alternative, and specific and the new emission limits in its
revised guidelines. HMIWI that
language with requirements for CO revised State plan, which allows a
commence construction after February
CEMS is included in the proposed phased approach in applying the new
6, 2007 or that are modified 6 months
amendments. For new and existing limits. That is, the State plan would
or more after the date of promulgation
HMIWI, use of PM, HCl, multi-metals, make it clear that the current emission
of any revised standards would have to
and Hg CEMS, and semi-continuous limits remain in force and apply until
meet the revised NSPS emission limits
dioxin monitoring (continuous sampling the date the new emission limits are
effective (as defined in the State plan). being added to the subpart Ec NSPS and
with periodic sample analysis) also are any remaining NSPS limits from the
approved alternatives, and specific States whose HMIWI units do not find
it necessary to improve their 1997 rule, as applicable, within 6
language for these alternatives is months after the promulgation date of
included in the proposed amendments. performance in order to meet the new
emission limits may want to consider a the amendments or upon startup,
b. Other Amendments. The proposed
second approach where the State would whichever is later.
amendments would revise the definition
of ‘‘Minimum secondary chamber insert the new emission limits in place IV. Rationale
temperature’’ to read ‘‘Minimum of the current emission limits, follow
procedures in 40 CFR part 60, subpart A. Rationale for the Proposed Response
secondary chamber temperature means
B, and submit a revised State plan to to the Remand
90 percent of the highest 3-hour average
secondary chamber temperature (taken, EPA for approval. If the revised State This action responds to the Court’s
at a minimum, once every minute) plan contains only the new emission remand by (1) further explaining the
measured during the most recent limits (i.e., the current emission limits reasoning processes by which EPA
performance test demonstrating are not retained), then the new emission determined the MACT floors and the
compliance with the PM, CO, and limits must become effective MACT standards for new and existing
dioxin/furan emission limits.’’ immediately since the current limits HMIWI for the portions of those
The proposed amendments would would be removed from the State plan. processes that are being retained under
require sources to submit, along with our remand response, and (2) explaining
4. Has EPA changed the applicability
each test report, a description of how revisions to the processes, the MACT
date of the 1997 NSPS?
operating parameters are established floors, and the MACT standards for new
during the initial performance test and No; however, HMIWI may be treated and existing HMIWI that result from our
subsequent performance tests. differently under the amended response to the remand.
standards than they were under the
3. Is an implementation schedule being 1997 standards in terms of whether they 1. New HMIWI
proposed? are ‘‘existing’’ or ‘‘new’’ sources, and The Court raised three issues with
Yes; under the proposed amendments there will be new dates defining what regard to EPA’s treatment of the MACT
to the emission guidelines, and are ‘‘new’’ sources and imposing floor for new units and the achievable
emission limitations. First, the Court 116 (1985); American Petroleum Inst. v. that elimination of feasibility
asked EPA to explain why the floor was EPA, 661 F.2d 340, 347 n. 23 (5th Cir. requirements and specification of
based on the highest emissions levels of 1981); Bunker Hill Co. v. EPA, 572 F.2d particular control systems indicated that
the ‘‘worst-performing’’ unit employing 1286, 1302 (9th Cir. 1977); Marathon Oil congressional amendment of CAA
the MACT technology rather than on the Co. v. EPA, 564 F.2d 1253, 1266–67 (9th section 202(a)(6) resulted in an
lowest observed emissions levels of the Cir. 1977); FMC v. Train, 639 F.2d 973, ‘‘absolute’’ standard).) MACT standards
best performing unit using the MACT 985–86 (4th Cir. 1976).) As discussed under CAA sections 111 and 129 are
technology. (See Sierra Club v. EPA, 167 elsewhere in this preamble, in CKRC the ‘‘technology-based,’’ rather than
F.3d at 665.) Second, the Court Court stressed that where record ‘‘absolute’’ standards. The legislative
requested further explanation of why evidence suggests that factors other than history to the 1990 CAA Amendments
EPA considered multiple units application of control technology clearly shows that Congress intended
employing the MACT technology, rather influence emissions, EPA will not be the MACT standards to be technology-
than identify the single best-performing able to demonstrate ‘‘that floors based based. (See I A Legislative History, at
unit and basing the floor on that on the worst-performing MACT sources’ 863 (Senator Durenberger referring to
particular unit’s performance with that emissions represent ‘a reasonable ‘‘the MACT technology-based
technology. Id. Third, the Court estimate of the performance of the [best- standards’’ in debates on the bill
requested further explanation of EPA’s performing] units.’ ’’ CKRC, 255 F.3d at reported by the Conference Committee);
procedure for determining the 866, quoting Sierra Club, 167 F.3d at id., at 1128 (Senator Dole explaining
achievable emission limitation from the 662. However, the Court reiterated that that changes made to CAA section 129
available data, where EPA selected a where EPA’s record demonstrates that in the Conference Committee ‘‘make the
numerical value somewhat higher than MACT technology significantly controls technology test more closely
the highest observed data point. The emissions, or that factors other than the approximate the role of the NSPS’’); S.
Court stated that EPA’s procedure control have a negligible effect, the Rep. No. 101–228, at 133–134 (1989)
‘‘[m]ay be justifiable as a means of approach of accounting for variability (referring to CAA section 112 MACT
reasonably estimating the upper bound by basing the floor on the highest standards as ‘‘technology-based
of the best-controlled unit’s emissions resulting from a source using standards’’ and noting that technology-
performance, but in the absence of MACT technology ‘‘could be a based effluent standards under the
agency explanation of both the decision reasonable means of satisfying the Clean Water Act served as a model for
to increase the levels and the choice of statute’s requirements.’’ CKRC, at 866. the new MACT standards).)
method for determining the increases, a. Applicability of National Lime to
CAA Section 129. CAA section 129(a)(3) CAA section 129 does not specify a
we are in no position to decide.’’ Id. type of control technology for HMIWI,
states that ‘‘[s]tandards under section
As discussed in detail below, for the 111 and this section applicable to solid but instead requires EPA to develop
first two issues, the Court described waste incineration units shall be based floor levels already achieved in practice
potential rationale for EPA’s method. on methods and technologies for by one or more units, and then issue
However, because the Court concluded removal or destruction of pollutants standards that EPA determines are
that this rationale was not adequately before, during, or after combustion ‘‘achievable’’ for units in that source
presented in the rulemaking record, the [* * *].’’ This language requires that category. As the Court stated in National
Court asked for further clarification by such a standard be based on the degree Lime Ass’n v. EPA (627 F.2d 416, 431
EPA. In subsequent cases the Court of reduction in air pollutant emissions n. 46 (D.C. Cir. 1980)) (‘‘NLA I’’), and
further addressed these potential that can be achieved through restated in Sierra Club, ‘‘where a statute
rationales, and discussed under what application of a particular method of requires a standard to be achievable, it
circumstances they would and would pollution control, and any other factors must be achievable ‘under most adverse
not be persuasive. In fact, the Court’s that record evidence shows significantly circumstances which can reasonably be
potential rationale for EPA’s method affect emissions performance. Much like expected to recur.’ ’’ (See Sierra Club,
reflects the principles used by EPA in the language in CAA sections 111 and 167 F.3d at 665.) In other words, ‘‘EPA
determining the MACT floor for new 129 governing the HMIWI standards, would be justified in setting floors at a
units and the achievable emission Congress has used similar language in level that is a reasonable estimate of the
limitations for this source category, and other statutes to direct adoption of performance of the ‘best controlled
is the method that has been used by technology-based standards. (See, e.g., similar unit’’ under the worst
EPA throughout most of the Agency’s CAA section 169(3) defining ‘‘best reasonably foreseeable circumstances[.]’’
30-year history in developing achievable available control technology’’; Clean Id. This concept of ‘‘worst reasonably
technology-based emission limitations Water Act section 301(b)(2)(A), for ‘‘best foreseeable circumstances’’ is
for source categories in cases where the available technology economically fundamental in developing achievable
application of control technology has achievable’’ or ‘‘BAT’’ standards; Clean technology-based emission limitations,
been the only means by which sources Water Act section 304(b)(1) for ‘‘best since, once the standard is in force,
have limited emissions, and the practicable technology’’ or ‘‘BPT’’ sources will be expected to comply with
variability of technology performance is standards.) it at all times by relying on the
a critical factor in determining an As the Court has stated, technology that formed the basis for
emission limitation’s achievability. (See, ‘‘[t]echnology-based provisions [in the EPA’s determination that the
e.g., American Iron and Steel Inst. v. CAA] require EPA to promulgate promulgated emissions limitation is
EPA, 115 F.3d 979, 1000 (D.C. Cir. standards only after finding that the achievable. As the Court stated in Sierra
1997); BP Exploration & Oil, Inc., v. requisite technology exists or may be Club, ‘[i]t is reasonable to suppose that
EPA, 66 F.3d. 784, 794 (6th Cir. 1995); feasibly developed. Absolute standards, if an emissions standard is as stringent
NRDC v. EPA, 790 F.2d 289, 299 (3d Cir. on the other hand, require compliance as ‘the emissions control that is
1986); National Ass’n of Metal Finishers with statutorily prescribed standards achieved in practice’ by a particular
v. EPA, 719 F.2d 624, 659 (3d Cir. 1983); and time tables, irrespective of present unit, then that particular unit will not
rev’d on other grounds sub nom, technologies.’’ (See NRDC v. Reilly, 983 violate the standard. This only results if
Chemical Mfrs. Ass’n v. NRDC, 470 U.S. F.2d 259, 268 (D.C. Cir. 1993) (holding ‘achieved in practice’ is interpreted to
mean ‘achieved under the worst the minimum stringency level of such a systems.’’ Id. The Court then stated that
foreseeable circumstances.’ ’’ Id. standard. ‘‘where test results are relied upon, it
EPA agrees with the Court that, in In addition, Congress’ use of the should involve the selection or use of
order to satisfy the requirements of NLA phrases ‘‘as determined by the test results in a manner which provides
I, ‘‘[t]he same principle should apply Administrator’’ and ‘‘achieved in some assurance of the achievability of
when a standard is to be derived from practice’’ in CAA section 129(a)(2) in the standard for the industry as a whole,
the operating characteristics of a the directive to establish MACT floors given the range of variable factors found
particular unit[,]’’ as is the case under shows that Congress expected EPA to relevant to the standards’ achievability.’’
CAA section 129(a)(2). Id. CAA section consider variability in operating Id. This does not mean that EPA must
129(a)(2) requires that the new unit conditions and other relevant factors in test every plant, but it does mean that
MACT floor be ‘‘not less stringent than the Agency’s determinations. The term ‘‘due consideration must be given to the
the emissions control that is achieved in ‘‘practice’’ is defined as ‘‘[r]epeated or possible impact on emissions of
practice by the best controlled similar customary action; habitual performance; recognized variations in operations and
unit, as determined by the a succession of acts of a similar kind; some rationale offered for the
Administrator.’’ It would have been custom; usage.’’ (See Black’s Law achievability of the promulgated
unreasonable for EPA to base the MACT Dictionary 1172 (6th ed. 1990).) Thus, standards given the tests conducted and
floors solely on the lowest levels of achieved in ‘‘practice’’ means achieved the relevant variables identified.’’ Id., at
emissions observed without an on a repeated, customary, or habitual 434. Thus, applying NLA I to the HMIWI
assessment of whether those observed basis. Under the statutory mandate that rule adopted under CAA sections 111
levels could be met on a continuous the level ‘‘achieved in practice’’ be and 129, it is really a misnomer to
basis, and the CAA and its legislative ‘‘determined by the Administrator,’’ characterize EPA as basing the MACT
history provide no support in deviating EPA must exercise its judgment, based floor on the emissions of the ‘‘worst
from the general practice EPA has on an evaluation of the relevant factors performing’’ unit using the technology
followed in the wake of NLA I. In a and available data, to determine the in question, since that unit’s level of
report on H.R. 3030, the House level of emissions control that can be emissions necessarily more closely
Committee on Energy and Commerce customarily achieved using the relied- represents the level ‘‘achieved in
explained that ‘‘MACT is not intended upon technology under variable practice’’ by the given technology than
to require unsafe control measures, or to conditions. Merely locating the lowest would the lowest emissions level
drive sources to the brink of shutdown.’’ emissions data point and setting the observed at a source using that ‘‘best’’
(See H.R. Rep. No. 101–490, pt. 1, at 328 MACT standard at that level would not technology.
(1990).) This view is consistent with constitute a considered ‘‘determination
NLA I, which involved challenges to by the Administrator’’ as to what has b. Variability Between Facilities or
standards EPA promulgated under been ‘‘achieved in practice.’’ (See, e.g., Units. In remanding the NSPS at issue
section 111 of the CAA and is Senate Debate on Conference Report, in NLA I, the Court noted that its
particularly applicable to the HMIWI 10–26–90, reprinted in I A Legislative decisions under CAA section 111
rulemaking under CAA section 129, History of CAA Amendments of 1990, ‘‘evince a concern that variables be
since this rule has its basis in authority 103d Cong., 1st Sess. at 1128–1129 accounted for, that the
in both section 129 and section 111. (Comm. Print 1993) (exchange between representativeness of test conditions be
(See CAA section 129(a)(1)(A) and (C).) Senators Dole and Durenberger [sic] ascertained, that the validity of
Moreover, interpreting CAA section confirming that the phrase ‘‘achieved in tests be assured and the statistical
129 as subject to the principles of NLA practice’’ accounts for the distinction significance of results be determined.’’
I appropriately notes the critical between research-type pollution control (See NLA I, 627 F.2d at 452–53.) (See,
distinction between a level of emissions systems and systems that are also, Portland Cement Ass’n v.
that has been continuously achieved ‘‘economically viable for widespread Ruckelshaus, 486 F.2d 375, 396 (D.C.
through performance using control use,’’ and stressing that MACT floors Cir. 1973), cert. denied, 417 U.S. 921
technology, and one that has been should rely upon technologies that can (1974).) When floors and standards are
observed at a single point in time. A ‘‘stand the rigors of day to day developed based on emissions data,
level that has been continuously operations’’).) EPA accounts for several types of
achieved is capable of being met under Ultimately, NLA I is controlling variability to avoid adopting
most conditions which can reasonably because the case addressed how unachievable standards. The first type
be expected to recur because variability standards must be set in the face of of variability is that concerning
in operating conditions is taken into variable operating conditions, and operational distinctions between
account. Such a level best effectuates involved one of the same provisions of facilities or units. As the Sierra Club
Congress’ intent because it ensures that the CAA, section 111, under which the Court stated in reviewing the HMIWI
the MACT floor will result in reduced HMIWI rule was promulgated. NLA I rule, ‘‘[p]erhaps considering all units
emissions without forcing sources to held that EPA is required to use data with the same technology is justifiable
shut down. A lowest observed emission that is representative of emissions that because the best way to predict the
level, however, is not representative of could be achieved in the industry as a worst reasonably foreseeable
a unit’s performance under most whole. (See 627 F.2d at 433.) In performance of the best unit with the
conditions which can reasonably be developing the standards at issue in that available data is to look at other units’
expected, and may be impossible to case, EPA relied upon tests of the performance. Or perhaps EPA
achieve on a regular, let alone emissions from particular units to reasonably considered all units with the
continuous, basis. While an observed determine the level of emissions control same technology equally ‘well-
lowest emissions level may be that was achievable across the entire controlled,’ so that each unit with the
appropriate for use in determining industry. The Court directed EPA to best technology is a ‘best-controlled
whether a source is in compliance with identify ‘‘variable conditions that may unit’ even if such units vary widely in
an emission standard that must be contribute substantially to the amount performance.’’ (See 167 F.3d at 665.)
continuously met, it is not an of emissions, or otherwise affect the These are two ways of saying essentially
appropriate level upon which to base efficiency of the emissions control the same thing, and these concepts have
been used by EPA throughout most of unit’’ as being a single data point from tests, in order for ‘‘a uniform standard
the Agency’s history in determining a single source provides merely a [to] be capable of being met under most
achievable technology-based emission snapshot of emissions performance that adverse conditions which can
limitations, in cases where application may not be replicable by either that reasonably be expected to recur[.]’’ (See
of control technology significantly single source or by other sources using NLA I, 627 F.2d at 431, n. 46.) (See also
controls emissions and no record the same control technology, and, Portland Cement Ass’n, 486 F.2d at 396
evidence indicates that factors other therefore, does not provide a basis for (noting industry point that ‘‘a single test
than the control have more than a enforceably requiring all sources to offered a weak basis’’ for inferring that
negligible effect. Examining multiple perform to that level. plants could meet the standards).)
units using the same technology gives Thus, the most reasonable way to Without accounting for variation among
the best picture of the performance interpret the statutory phrase ‘‘best different emissions tests, it can be
capability of that particular technology, controlled similar unit’’ in CAA section determined with a significant degree of
since it provides EPA with a more 129 is as encompassing all units using statistical confidence that even a single
complete set of data by which to the same technology and emissions unit will not be able to meet the
evaluate what levels of emissions limiting means as the single unit with standard over a reasonable period of
control a technology can achieve as it is the best observed performance, rather time, when one can expect adverse
applied to varying sources. Such an than just that single best performing conditions to recur. The Courts have
analysis is necessary especially when unit itself. A contrary interpretation recognized this basic principle in
adopting standards that all sources in a would seem to directly conflict with the reviewing technology-based effluent
category will have to be able to meet by Court’s directive in NLA I, and is not standards under the Clean Water Act.
using the identified technology. Since compelled by the Court rulings in Sierra As the U.S. Court of Appeals for the 5th
MACT floors and standards are Club, NLA II, and CKRC. Applying this Circuit stressed regarding ‘‘best
generally expressed as numerical approach to evaluating ‘‘best practicable technology’’ or ‘‘BPT’’
emissions limits, it is necessary to technologies’’ at ‘‘best controlled similar standards under section 304(b)(1) of the
account for this variability in order to units,’’ where different design Clean Water Act, ‘‘[t]he same plant
adopt a regulation that is ‘achievable’ by characteristics are identified (e.g., low- using the same treatment method to
the industry as a whole.’’ (See NLA I, efficiency versus moderate-efficiency remove the same toxic does not always
627 F.2d at 437.) versus high-efficiency wet scrubbers), achieve the same result. Tests
Section 129(a)(2) of the CAA requires the data are grouped such that each data conducted one day may show a different
that EPA determine the emissions set reflects the performance of an concentration of the same toxic than are
control achieved by the ‘‘best controlled ‘‘identical’’ control device, providing shown by the same test on the next day.
similar unit’’ when establishing the the best indication of the true This variability may be due to the
MACT floors for new units. A solid performance of each control device and inherent inaccuracy of analytical
waste incineration ‘‘unit’’ is defined as enabling the Agency to adopt a testing, i.e., ‘analytical variability,’ or to
‘‘a distinct operating unit of any facility numerical standard that can be met with routine fluctuations in a plant’s
which combusts any solid waste the subject technology at all units treatment performance.’’ (See Chemical
material’’ (CAA section 129(g)(1)). To employing this technology, and can be Mf’rs Ass’n v. EPA, 870 F.2d 177, 228
achieve the best level of pollution enforced. Again, where the record (5th Cir. 1989).) (See also American
control, that unit will utilize a particular evidence indicates that the only means Petroleum Institute v. EPA, 540 F.2d
method of pollution control (and of control of emissions at units is 1023, 1035–36 (10th Cir. 1976) (‘‘Even
possibly use other means that affect its application of control technology, and in the best treatment systems, changes
emissions performance). The emissions there is no record evidence showing that occur in ability to treat wastes. [* * *]
control achieved by that method (and by other means of emissions limitation [V]ariability factors present[] a practical
any additional means) is the emissions significantly affect emissions effort to accommodate for variations in
control achieved by the ‘‘best controlled performance, basing the MACT floor on plant operations’’); FMC Corp. v. Train,
similar unit.’’ Thus, the MACT floor for this approach is fully consistent with 539 F.2d 973, 985 (4th Cir. 1976)
new units is based on the ‘‘emissions the Court’s rulings in the MACT cases. (variability factors account for ‘‘the fact
control’’ that is attained by the specific c. Variability Between and Within that even in the best treatment systems
method of pollution control and any Tests at Facilities. Another type of changes continually occur in the
other means used to limit emissions at variability that EPA accounts for in treatability of wastes’’).)
the best similar unit, rather than merely order to ensure the achievability of The same types of differences leading
on the emissions measured at a technology-based standards that rely to between-test variability also cause
particular unit. upon application of pollution controls variations in results between various
In this way, by basing the MACT floor concerns operational distinctions runs comprising a single test, or
on the capability of a particular method between and within tests at the same ‘‘within-test variability.’’ A single test at
of pollution control used at ‘‘similar’’ unit. Regarding ‘‘between-test a unit usually includes at least three
‘‘best’’ ‘‘units,’’ instead of on the variability,’’ even where conditions separate test runs. (See 40 CFR
emissions measured at a single unit, appear to be the same when two or more 63.7(e)(3) (for MACT standards under
EPA ensures that the floors would not tests are conducted, variations in section 112 of the CAA), and 40 CFR
only be achievable by the single best emissions are often caused by different 60.8(f) (for NSPS under CAA section
performing unit, but are also achievable settings for emissions testing equipment 111).) (See also Portland Cement Ass’n,
by other units using the same and differences in sample handling. 486 F.2d at 397 (noting differences in
technology and/or emissions limiting Varying results may also be caused by conditions among several test runs).)
means as the best similar unit, and that use of different field teams to conduct d. Application of NLA I, Sierra Club,
it is reasonable to require the best the testing, or different laboratories to NLA II, and CKRC Principles in HMIWI
similar unit and all future new units to analyze the results. All these variations Rulemaking. Based on the record for the
meet this floor on a continuous basis. In are typical. 1997 rulemaking, the best way to
contrast, identifying the ‘‘emissions An achievable standard needs to determine the worst reasonably
control’’ of the ‘‘best controlled similar account for these differences between foreseeable circumstances for the
particular technologies used to control operator’s power to adopt. The reference and air pollution controls were
emissions at HMIWI was to first to a ‘‘degree of reduction’’ supports the inspected thoroughly, and maintenance
examine the highest data point actually view that the words ‘‘control’’ and was performed where necessary to
observed from HMIWI equipped with ‘‘limitation’’ appearing in section ensure that the incinerators and
each particular technology. If an 129(a)(2) require a source to have pollution controls were functioning
emission value has been observed and reduced emissions from uncontrolled properly. (See Legacy Docket ID No. A–
there is no reason to believe it levels through some control technique. 91–61, items II–A–93, II–A–94, and II–
represents poor performance (i.e., there See NLA II, 233 F.3d at 631–32 A–85.) During testing, most test runs
is nothing that can be done to prevent (rejecting position that EPA is required were conducted under representative
its recurring), it is likely to occur again to set new source floors at the lowest conditions to minimize emissions. (See
in the future and, therefore, reflects a recorded emission level for which it has Legacy Docket ID No. A–91–61, items
foreseeable circumstance. It is incorrect data and to set existing source floors at II–A–111, IV–B–46, and IV–B–47.)
to characterize the highest data point as the average of the lowest 12 percent or However, some test runs were purposely
the ‘‘worst performance’’ of the best recorded emission level data points). conducted under conditions that would
performing unit, or to characterize one The Court has recognized that EPA represent poor operation (e.g.,
control device’s performance as ‘‘better’’ may consider variability in estimating overcharging waste to the incinerator) to
than another’s based solely on the the degree of emission reduction determine the effect of improper
results of a single emission test. This is achieved by best-performing sources operation on emissions. (See Legacy
because such focuses relate to and in setting MACT floors. See Docket ID No. A–91–61, items II–A–111
essentially random single data Mossville Envt’l Action Now v. EPA, 370 and IV–B–46.) These test runs
occurrences, rather than to estimating F.3d 1232, 1241–42 (D.C. Cir 2004) demonstrated that improper operation
what a particular technology can be (holding EPA may consider emission results in higher emissions. (See Legacy
expected to continuously achieve. variability in estimating performance Docket ID No. A–91–61, items II–A–111,
Rather, each data point, whether from achieved by best-performing sources IV–B–46, and II–A–81.) Of course, the
one unit or from several identical units and may set floor at level that best- test runs reflecting poor operation were
using the same technology, should be performing source can expect to meet not used in developing the achievable
viewed as a snapshot of the actual ‘‘every day and under all operating emission limitations. Id. It is important
performance of the technology in use. conditions’’). Since an emission to note that such poor operation is
Along with an understanding of the limitation must be complied with at all precluded by the good combustion
factors affecting the performance of the times, for it to be achievable it must be requirements and the parametric
technology, each of these snapshots set at a level that will not force sources monitoring requirements in the 1997
gives information about the normal, and to violate it when operating conditions final rule. In addition to data gathered
unavoidable, variation in emissions that are not ideal and higher emissions by EPA directly, vendors of air pollution
would be expected to recur over time levels might be observed. For example, control systems submitted test reports to
when using the identified technology. a car which has been observed to EPA. (See Legacy Docket ID No. A–91–
Conversely, when there is evidence that consume 0.02 gallons of gasoline in a 61, items II–I–230 through 237, II–I–243
an emission data point reflects poor one-mile downhill stretch of highway and 244, II–I–248, IV–B–48 and 49, IV–
performance (design, operation, or cannot be said to have ‘‘achieved’’ a J–11, IV–J–15 and 16, IV–J–20, IV–J–24,
minimum 50 miles per gallon fuel
both), such a data point should not be IV–J–27, IV–J–29 through 31, IV–J–33
efficiency rate when that same car is
considered in determining the and 34, IV–J–39 and 40, and IV–J–47.)
later certain to consume 0.04 gallons of
achievable emission limitation The test reports were submitted
gasoline in a one-mile uphill stretch of
associated with the technology. primarily by wet scrubber vendors to
highway (25 miles per gallon). Rather,
Furthermore, a distinction must be demonstrate to EPA that wet scrubbers
the minimum fuel efficiency of the car
made between an emission level that could achieve lower emissions than
will be that which the car can meet in
has been ‘‘observed’’ and an emission EPA had concluded from the EPA-
adverse circumstances, the uphill
limitation that can be continuously collected data. (EPA had conducted
stretch. So it is with emissions
‘‘achieved.’’ The purpose of the MACT testing on only one wet scrubber
limitations, which cannot reasonably be
program is to compel sources to system.) (See 61 FR 31742; Legacy
set at levels which would force sources
replicate emission reduction strategies to operate in violation even when Docket ID No. A–91–61, item IV–B–48.)
used by the best-performing sources. properly employing the control The test reports and the data collected
Thus, MACT floors are based on the technology upon which the standards by EPA reflect the best performance of
control strategies used by the best- are based. the air pollution controls that can
performing sources to reduce emissions, The emission data used to develop the reasonably be expected when
not based on a snapshot level of emission limitations in the HMIWI continuously applied on HMIWI.
emissions from sources without regard regulations reflect properly designed MACT and other technology-based
to whether this level reflects application and operated air pollution control standards are necessarily derived from
of any replicable emission control technology on properly designed and short-term emissions test data, but such
strategies. CAA section 129(a)(2) does operated HMIWI, and emission data that data are not representative of the range
not direct EPA to assess relative reflected poor operation of the HMIWI of operating conditions that facilities
emission ‘‘levels’’ in determining MACT unit or the air pollution control face on a day-to-day basis. In statistical
floors; it directs EPA to assess the technology were excluded. (See Legacy terms, each test produces a limited data
degree of emissions ‘‘control’’ or Docket ID No. A–91–61, items II–A–111 sample, not a complete enumeration of
‘‘reduction’’ or ‘‘limitation’’ ‘‘achieved’’ and IV–B–14.) The incinerators selected the available data for performance of the
by the best-controlled or best- by EPA for testing represented a range unit over a long period of time. (See
performing sources. The plain meaning of incinerator designs and air pollution Natrella, Experimental Statistics,
of these words implies that a source is control systems in use on this source National Bureau of Standards Handbook
utilizing some method or technique to category. (See Legacy Docket ID No. A– 91, chapter 1 (revised ed., 1966).) EPA,
reduce emissions that is within a source 91–61, item IV–B–46.) The incinerators therefore, often needs to adjust the
short-term data to account for these associated with employed control stream certain solid waste components
varying conditions, so facilities properly strategies. contributing to toxic emissions from the
employing optimal controls can remain In the CKRC case, the Court left open incinerator (62 FR 48380, 48387).)
in compliance with the standards on a the possibility that the approach of e. Development of the Proposed
continuous basis. basing floors on the ‘‘worst-performing Revised Emission Limits. While we are
MACT sources’’ emissions represent ‘a proposing to respond to the Court’s
With the relatively small data sets
reasonable estimate of the performance remand regarding new units by basing
EPA had to work with in the 1997 of the [best-performing] units,’ ’’ CKRC floors and standards on the same control
HMIWI rulemaking, it is possible that at 866, quoting Sierra Club at 662, technologies that formed the basis for
EPA has not recorded the highest provided that ‘‘in the case of a particular the 1997 standards, in some cases it is
emissions levels that would occur under source category or HAP, the Agency can necessary to adjust the emission limits
the worst reasonably foreseeable demonstrate with substantial in order to correct for the concerns
circumstances. As the Court noted, it evidence—not mere assertions—that regarding our 1997 methodology that
would ‘‘generally defer to an agency’s MACT technology significantly controls the Court raised. As at promulgation of
decision to proceed on the basis of emissions, or that factors other than the the 1997 rule, EPA examined the data
imperfect scientific information, rather control have a negligible effect[.] CKRC available for various air pollution
than to ‘invest the resources to conduct at 866, citing NLA II at 633. The Court control technologies applied to HMIWI
the perfect study.’ ’’ (See Sierra Club, in Sierra Club essentially already found to determine the performance
167 F.3d at 662.) ‘‘[S]ince EPA had data this to be the situation for the HMIWI capabilities of the technologies;
on only one percent of about 3,000 rulemaking, and it was, therefore, identified the best control technology
[HMIWI], the data gathering costs of any appropriate for EPA to base its MACT for each air pollutant for each
non-sampling method may well have floor review in the 1997 rule strictly on subcategory of HMIWI (i.e., MACT
been daunting.’’ Id., at 663. In fact, the the emissions reductions achieved by floor); considered control technologies
‘‘perfect study’’ cannot be conducted, use of control technologies. The Sierra more stringent than the MACT floor;
regardless of the resources expended to Club had claimed that EPA wrongly made a determination regarding the
conduct it. Every study ends with some failed to require HMIWI units to achievable emissions levels from using
uncertainty in the results. There is no undertake programs to reduce the Hg control technologies upon which the
‘‘cookbook’’ methodology for and chlorinated plastics in HMIWI emission standards would be based; and
determining achievable emission waste streams. Sierra Club, at 666. then established numerical emission
limitations from data. In every case, but While the petitioner raised this limits achievable with those
especially in cases where data are objection in its challenge to the technologies. The proposed revised
limited as with the 1997 HMIWI promulgated standards, rather than its standards are based on the same
rulemaking, EPA must make judgments objection to the floor methodology, the technologies upon which the 1997 final
about what constitutes the worst Court’s response to the Sierra Club’s standards were based—good
reasonably foreseeable circumstance claim shows that in the case of the 1997 combustion and a moderate-efficiency
and put those judgments out for public HMIWI rulemaking, EPA appropriately wet scrubber for new small HMIWI, and
comment. In the case of the HMIWI focused on the control technologies good combustion and a combined dry/
rulemaking, the ‘‘high’’ data points used at HMIWI units, and that, wet control system with carbon for new
simply reflected the normal, and therefore, under the CKRC ruling it was medium and large HMIWI—and reflect
unavoidable, variation in emissions that appropriate, in this instance, to base the MACT floor emissions levels for
would be expected to recur over time floors on the highest emissions levels new small and large HMIWI, but are
when properly using the best control achieved by units employing the MACT more stringent than the MACT floor for
technologies and strategies we technologies. new medium HMIWI. The rationale for
The Court observed that ‘‘EPA does these determinations regarding
determined were being used at HMIWI
not deny that the waste stream identification of MACT can be found at
units. In fact, while the highest observed reductions the Sierra Club calls for 62 FR 48365.
value is a ‘‘foreseeable circumstance,’’ it would reduce pollution. The less As explained earlier in this preamble,
may not reflect the worst reasonably mercury in, the less mercury out, and we are proposing emission limits for
foreseeable circumstance. In the less chlorinated plastic in, the less each air pollutant for each subcategory
determining the 1997 final MACT HCl out. But the EPA has consistently of new HMIWI based on the highest
standards, EPA chose to account for the argued in its response to comments and observed data points associated with the
‘‘worst reasonably foreseeable here that it does not have evidence that control technologies upon which the
circumstance’’ by adding 10 percent to allows quantification of the relevant emission standards are based, since we
the highest observed emissions levels in output reduction. For mercury, the only identified the ‘‘best controlled similar
the data, and then rounding up those quantitative evidence before EPA was unit’’ as one using the relevant control
figures. Upon review of this approach in that a pollution prevention program technologies for each subcategory of
responding to the Court’s remand, we aimed at mercury could reduce mercury new units. The proposed percent
have determined that although the emissions from very high levels to reduction limits for HCl, Pb, Cd, and Hg
highest observed data point may not typical levels. See RTC at 7–14 to 7–15. were established based on average
reflect the ‘‘worst reasonably foreseeable For chlorinated plastics, there was no combustion-controlled emissions
circumstance,’’ we do not have quantitative evidence before the agency. estimates and highest observed data
information to support accounting for See RTC at 7–16, 7–18. The Sierra Club points associated with the control
the ‘‘worst reasonably foreseeable does not contest the adequacy of EPA’s technologies upon which the emission
circumstance’’ by adding 10 percent to data-gathering with respect to these standards for each of these pollutants
the highest observed emissions levels, measures.’’ Id. (Note that the emission for each subcategory are based. This is
and then rounding up those figures. We, guidelines and NSPS require HMIWI to the same approach used at the time of
therefore, propose to base revised prepare a waste management plan under promulgation with two exceptions—the
MACT standards for new HMIWI units §§ 60.35e and 60.55c that would proposed percent reduction limits do
on the highest observed data points segregate from the health care waste not include the addition of 10 percent
to the highest observed emissions levels, HMIWI in response to the remand are controls have not been demonstrated on
nor does it include the rounding up of presented in Table 6 of this preamble. HMIWI; and SO2 emissions are not
those figures. A summary of the control Note that MACT for NOX and SO2 are reduced by combustion control, and
technologies upon which the proposed ‘‘combustion control,’’ although acid gas controls are not effective in
standards for new HMIWI are based, the combustion control results in no reducing SO2 emissions from HMIWI at
highest observed data points associated emission reductions for those pollutants the low SO2 levels associated with
with those control technologies, and the because NOX emissions are not reduced HMIWI.
proposed emission limits for new by combustion control, and NOX add-on
TABLE 6.—SUMMARY OF REMAND RESPONSE FOR NEW HMIWI

Pollutant Highest observed
Unit Size 1 MACT data point 2 Proposed emission limit 2
(units)
HCl (ppmv) ........................................... L, M, S ........ Wet scrubber ....................................... 9.3 15 3 or 99% reduction 3.
CO (ppmv) ........................................... L, M, S ........ Good combustion ................................ 32 32.
Pb (mg/dscm) ...................................... L, M ............. Dry scrubber w/carbon ........................ 0.06 0.060 or 98% reduction 3.
S .................. Wet scrubber ....................................... 1.1 0.78 4 or 71% reduction.
Cd (mg/dscm) ...................................... L, M ............. Dry scrubber w/carbon ........................ 0.03 0.030 or 93% reduction.
S .................. Wet scrubber ....................................... 0.14 0.11 4 or 66% reduction 3.
Hg (mg/dscm) ...................................... L, M ............. Dry scrubber w/carbon ........................ 0.45 0.45 or 87% reduction.
S .................. Wet scrubber ....................................... 0.47 0.47 or 87% reduction.
PM (gr/dscf) ......................................... L, M ............. Dry scrubber w/carbon ........................ 0.009 0.0090.
S .................. Moderate-efficiency wet scrubber ....... 0.018 0.018.
CDD/CDF, total (ng/dscm) ................... L, M ............. Dry scrubber w/carbon ........................ 20 20.
S .................. Wet scrubber ....................................... 111 111.
CDD/CDF, TEQ (ng/dscm) .................. L, M ............. Dry scrubber w/carbon ........................ 0.53 0.53.
S .................. Wet scrubber ....................................... 2.1 2.1.
NOX (ppmv) ......................................... L, M, S ........ Combustion Control 5 .......................... 225 225.
SO2 (ppmv) .......................................... L, M, S ........ Combustion Control 5 .......................... 46 46.
1L = Large; M = Medium; S = Small.
2 All
values are measured at 7 percent oxygen.
3 Nochange proposed.
4 Remand standards for existing small non-rural HMIWI are proposed.
5 Combustion control results in no emissions reduction.
Note that no change is proposed for Method 26A to measure HCl emissions combination of regulatory and
the emission limit for HCl for new large, below 20 ppmv and did not take uncontrolled (i.e., combustion-
medium, and small HMIWI. In this precautions to ensure no negative bias, controlled) data provided a ‘‘reasonable
situation, the highest observed emission in this action we are proposing to retain estimate’’ of HMIWI performance:
point (i.e., 9.3 ppmv) is not used as a the emission limit of 15 ppmv and also ‘‘First, EPA has said nothing about the
basis for the proposed emission limits. are including provisions that require possibility that [HMIWI] might be
Public comments concerning use of EPA sources to condition the filter before substantially overachieving the
Method 26A when testing for HCl testing, and use a cyclone and post test [regulatory] limits. [Footnote:] Although
emissions at sources with wet scrubbers purge if water droplets may be present. the agency conceded in its response to
were submitted with respect to the In the cases of Pb and Cd for new small comments that ‘actual emission data
recently promulgated standards for HMIWI, using the highest observed data routinely fall below the State permit
other solid waste incineration units (70 points would result in emission limits emission limits,’ [* * *] the context
FR 74870, December 16, 2005). The less stringent (i.e., higher) than the makes reasonably clear that the EPA
commenter asserted that EPA Method proposed emission limits for existing was referring to data on ‘actual
26A is not adequate for demonstrating small non-rural HMIWI. Because the emissions’ during tests; EPA implied
compliance with an HCl standard below existing source analysis provides limits that ‘these levels are not routinely
20 ppmv when sampling sources with that can be achieved by existing HMIWI, achieved in practice.’ [* * *] [End
wet scrubbers. Although EPA did not there is no reason to believe that new Footnote] If this were the case, the
concede that there is an outright HMIWI could not also meet the more permit limits would be of little value in
problem, we acknowledged that a tester stringent limits. This unanticipated estimating the top 12 percent of
may need to take certain precautions to result may be due to the small amount [HMIWI]s’ performance’’ (167 F.3d at
ensure that there is no bias when of Pb and Cd emissions data available 663, and at n. 3). According to the
sampling streams with low HCl for wet scrubbers at promulgation. Court, ‘‘[d]ata in the record suggest that
concentrations in certain environments Regardless, we are proposing emission the regulatory limits are in fact much
and promulgated an HCl emission limit limits for Pb and Cd for new small higher than emissions that units achieve
of 15 ppmv (versus the proposed limit HMIWI that are the same as those in practice.’’ Id., at 663.
of 3.7 ppmv). Method 26A also notes proposed for existing small non-rural ‘‘Second, EPA never gave any reason
that there is a possible measurable HMIWI. for its apparent belief that [HMIWI]s
negative bias below 20 ppmv HCl 2. Existing Units that were not subject to [regulatory
perhaps due to reaction with small limits] did not employ emission
amounts of moisture in the probe and The Court raised three specific controls of any sort. Unless there is
filter (40 CFR part 60, appendix A). concerns regarding EPA’s approach for some finding to this effect, it is difficult
Accordingly, because many of the wet- existing units in concluding that EPA to see the rationality in using
scrubber controlled HMIWI used had not adequately explained why the ‘uncontrolled’ data for the units that
were not subject to regulatory performing’’ units, and that EPA’s stated were substantially overachieving or
requirements’’ (167 F.3d at 664). The ‘‘belief’’ did not rise to the level of a underachieving their regulatory limits.
Court pointed out that ‘‘[d]ata submitted ‘‘reasonable estimate.’’ Id. However, in Second, there was no information before
by the American Hospital Association Mossville Envt’l Action Now v. EPA (370 the Agency suggesting that the State
[AHA] in 1995 indicate that over 55% F.3d 1232 (D.C. Cir. 2004)), the Court regulatory agencies erred in establishing
of [HMIWI]s in each category were concluded that ‘‘instead of simply the regulatory limits or that the States’
controlled by wet scrubbers.’’ Id., claiming that it believes its [relied upon] regulatory limits were outdated. It was
footnote omitted. As a result, the Court standards estimate what the best five thus reasonable for EPA to expect that
found it ‘‘difficult to see how it was plants actually achieve, EPA points to the State regulatory limits provided a
rational to include any uncontrolled some evidence. In its response to reasonable estimate of the actual
[i.e., combustion-controlled] units in the comments, EPA cited its analysis of performance of HMIWI units. Third, it
top 12 percent, at least with respect to three years of data, and * * * met its was reasonable for EPA to expect that
pollutants that wet scrubbing controls.’’ burden of establishing that its standards regulated entities take their regulatory
Id. reasonably estimate the performance of limits into account when designing their
Third, the Court held that ‘‘assuming the best five performing sources. Having control equipment. To some extent,
the regulatory data was a good proxy for cited the great variability of emission control equipment can be designed to
the better controlled units and that there levels, even within the same plants, and meet various levels of emissions, and
were shortfalls in reaching the necessary the inherent difficulty in other regulated entities do not normally spend
12 percent, EPA has never explained standards it considered, the EPA’s more money than necessary to meet a
why it made sense to use the highest of selection of the [relevant] standards as regulatory limit. As noted above, the
its test run data to make up the gap.’’ Id. the MACT floor is reasonable because it Court observed that ‘‘there seems no
Subsequent court decisions also has supported its decision with record reason to think that underachieving
addressed the type of information EPA data that shows the connection between firms would be overrepresented’’ by
may use to estimate emissions its MACT floor and the top performing regulatory limits (167 F.3d at 662).
performance and establish MACT floors plants.’’ 370 F.3d at 1242. Conversely, there is no reason to
for existing units. In NLA II, the Court a. The Possibility that HMIWI Sources generally assume that substantially
rejected the Sierra Club’s claim that it are Substantially Overachieving their overachieving firms would be
was unreasonable for EPA to select ‘‘the Regulatory Limits. With regard to the overrepresented in jurisdictions
median [performing] plant out of the Sierra Club Court’s first concern, the imposing regulatory limits. Rather, what
best twelve percent of the plants for Court itself noted early in its opinion is most likely is that sources in
which it had information and set the that ‘‘the necessary relationship [of regulated jurisdictions will have
* * * floor at the level of the worst regulatory data serving as a reasonable assessed whether steps to control
performing plant in its databases using proxy to indicate HMIWI performance] emissions are needed to comply with
th[e same] technology [as the median seems quite reasonable here. Indeed, it the regulatory limits, and that, in order
plant].’’ 233 F.3d at 630. As long as seems likely that any jurisdiction to account for emissions variability
EPA’s estimate of the performance of the bothering to impose limits would not when applying control technologies,
top 12 percent was reasonable, the Court knowingly set them below what it found they will be targeting their emissions
held, EPA was not required to use actual firms to be achieving in practice. And levels at some safe point below the
emissions data. Id. While in CKRC the there seems no reason to think that regulatory limits. Hence, with no
Court held that EPA had not justified in underachieving firms would be
information in the 1997 rulemaking
the HWC rulemaking basing the floor on overrepresented in jurisdictions making
record to indicate otherwise, EPA
emissions levels of the worst performing this effort.’’ 167 F.3d at 662. The Court
plant utilizing MACT control generally expected that regulatory limits
also expressed support for the notion
technology, when record evidence were being achieved, through
that, when faced with limited actual
indicated other factors beyond MACT application of emissions control
emissions information, a substitute
technology affected emissions methods, at emissions levels that
‘‘ ‘reasonable sample’ may be used ‘to
performance, the Court reiterated that sources deem necessary in order to
find out what the best 12 percent are
EPA could use estimates, as long as they minimize the risk of violating the
doing[,]’ ’’ (id., citing Oral Arg. Tr. at
reflected a ‘‘reasonable[] estimate [of] relevant limit, and were neither
11), and that ‘‘EPA typically has wide
the performance of the * * * best- latitude in determining the extent of substantially overachieving the limits
performing plants.’’ 255 F.3d at 862. data-gathering necessary to solve a nor underachieving them.
Specifically regarding the use of State problem.’’ Id. Specifically, the Court The Court noted that the
permits to determine MACT floors, the noted ‘‘that since EPA had data on only administrative record indicated that, in
Court in Northeast Maryland Waste one percent of about 3000 [HMIWI]s, some cases, sources were overachieving
Disposal Authority v. EPA, 358 F.3d 936 [* * *] the data-gathering costs of any their regulatory limits, where the floors
(D.C. Cir. 2004) (‘‘NMWDA’’), rejected non-sampling method may well have based on the weighted average of the
EPA’s approach for small municipal been daunting.’’ Id., at 663. regulatory limits and the ‘‘uncontrolled’’
waste combustion units because ‘‘as in There are three reasons why EPA (i.e., combustion-controlled) data were
Sierra Club, EPA stated only that it chose to use the regulatory limits at significantly higher than the values used
‘believes’ state permit limits reasonably their face value in calculating the for combustion-controlled data. (See 167
reflect the actual performance of the existing source MACT floor for the 1997 F.3d at 663, citing A–91–61, IV–B–024
best performing units without rule. First, regulatory data were used at 2–3). Here, the Court was referring to
explaining why this is so.’’ 358 F.3d at because there was very little actual some regulatory limits that, in fact,
954. There, EPA had asserted that the emissions data available and very little reflected higher emissions levels than
inherent variability of emission levels data available indicating the type of air did EPA’s uncontrolled (i.e.,
made other data inaccurate, but the pollution control used by the best combustion-controlled) emission
Court concluded that EPA gave ‘‘no performing units. (See 61 FR at 31738.) estimates, and suggested that in these
evidence that the [State] permit levels None of the available information cases it would be unreasonable for EPA
reflect the emission levels of the best- indicated that the regulated entities to view the best performing 12 percent
of sources as actually polluting at levels at least with respect to pollutants that summaries, HMIWI emissions test
so much higher than the test units for wet scrubbing controls.’’ Id., at 664. reports, and various inventories. (See
which EPA assumed no emissions With regard to the AHA ‘‘data’’ Legacy Docket ID No. A–91–61, items
controls were in place. Id., at 663–664. identified by the Court as indicating 55 IV–J–82, IV–B–07, II–B–94, II–D–175
EPA agrees that a regulatory limit percent of HMIWI use wet scrubbers, through 178, II–I–151, IV–J–89, IV–E–
does not reflect ‘‘actual performance’’ EPA believes that the Court was led by 65, IV–E–74, IV–E–86, and II–B–61;
when that limit is higher than the level this information into assuming that Docket EPA–HQ–OAR–2006–0534,
attributed to the worst reasonably unregulated HMIWI were in fact document titled ‘‘List of Test Reports
foreseeable performance of an applying add-on emissions controls, Used to Identify HMIWI Control
uncontrolled (i.e., combustion- when the record does not actually Devices’’). Our assessment that few
controlled) source. Since the data substantiate such an assumption, HMIWI were equipped with add-on
forming the basis for the existing source especially for small HMIWI. The AHA controls is also supported by economics
MACT floor must provide a reasonable asserts ‘‘almost all properly designed, in that it would not have made sense for
estimation of the ‘‘actual performance’’ operated, and controlled [HMIWI] can an HMIWI to be voluntarily equipped
of the best performing 12 percent of readily meet a particulate emission limit with an air pollution control device that
HMIWI, such high regulatory limits of 0.10 gr/dscf without an [add-on air costs one to three or more times as
should not have been included in the pollution control] system’’ (IV–D–637, much as the entire HMIWI. Further
best-performing 12 percent. Therefore, Exhibit 2, emphasis added). The AHA supporting our assessment is the fact
in our re-visiting the MACT floor for then concludes ‘‘[t]herefore, it is that the expected outcome of the
existing HMIWI based on the 1997 reasonable that as many as 50 percent of regulation (which was not refuted by
record, in situations for which there is those [HMIWI] having such an emission any commenters), that 50 to 80 percent
no information in the 1997 record limit would be uncontrolled.’’ Id. The of existing incinerators (including 100
indicating the presence of an add-on AHA goes on to assume that 50 percent percent of the small units) would shut
pollution control device (‘‘APCD’’) or of all HMIWI with particulate emission down rather than meet the regulations
other use of air pollution control limits of 0.10 gr/dscf or higher are because those that chose to meet the
methods but there are regulatory limits, controlled with wet scrubbers, while an regulations would have to install air
we propose the substitution of even higher percentage of units with pollution control to comply, was, in
combustion-controlled data for more stringent particulate emission fact, more than realized. (See 60 FR
limits are assumed to be controlled. Id. 10665, 61 FR 31768, and 62 FR 48372.)
regulatory limits where those data
This is akin to saying that, because In fact, all but 8 small units, 6 of which
reflect lower emissions levels than do
homeowners are generally not required meet the rural criteria and did not have
regulatory limits that appear to be
to install wet scrubbers on fireplaces, it to install air pollution control to
unrelated to actual controls. We propose
is reasonable to assume that as many as comply, 20 medium units, and 44 large
to continue to use combustion-
50 percent of homes with fireplaces do units have shut down, rather than meet
controlled data in situations for which
not have wet scrubbers, while the other the standards that would have been
there is no information indicating air
50 percent of home fireplaces are achieved by use of the very controls
pollution controls are in use and there
equipped with wet scrubbers. The AHA AHA appeared to assume were in place.
are no regulatory limits.
makes a basic assumption that at least Consequently, EPA concludes that the
b. Emission Control on HMIWI Not 50 percent of all HMIWI have wet
Subject to Regulatory Limits. The 1997 record, as confirmed by recent data
scrubbers, no matter what requirements
Court’s second concern was that EPA showing the vast reduction in sources
they are subject to. With no other
had not made a finding that HMIWI that (as opposed to sources installing
information to support its assumption,
were not subject to regulatory controls), shows that most HMIWI were
AHA’s ‘‘data’’ indicating 55 percent of
requirements did not use emissions not equipped with add-on air pollution
HMIWI are equipped with wet scrubbers
controls of any kind. The Court viewed control and that the use of uncontrolled
is altogether unreliable. In addition,
such a finding as a necessary (i.e., combustion-controlled) emission
EPA’s documented difficulty in
prerequisite to using uncontrolled (i.e., estimates where there was no indication
identifying sources with add-on controls
combustion-controlled) data for units of air pollution control (and where any
during the development of the HMIWI
not subject to regulatory requirements. applicable regulatory limits allowed
emission testing program is in direct
This issue can be partly resolved by conflict with the large number of higher levels of emissions than our
correcting a misunderstanding that may controlled sources suggested by the combustion-controlled emissions values
have resulted from our 1997 AHA ‘‘data.’’ reflected) was warranted. Based on the
administrative record. The Court Based on information from various number of HMIWI shutdowns, it
focused on information submitted in sources in the docket from the 1997 appears very likely that there were even
1995 by the AHA suggesting that ‘‘over rulemaking, including an AHA HMIWI fewer HMIWI with air pollution controls
55% of [HMIWI]s in each category were inventory, we now estimate that about than we estimated based on the
controlled by wet scrubbers.’’ (See 167 32 percent of large, 4 percent of information discussed above (i.e., that
F.3d at 664, citing AHA Comments, medium, and 1 percent of small HMIWI about 32 percent of large, 4 percent of
Exhibit 3.) Based on its review of the at the time of the 1997 rulemaking were medium, and 1 percent of small HMIWI
AHA comments, the Court assumed that equipped with add-on control systems. were equipped with add-on control
under EPA’s estimation of the HMIWI Other sources in the 1997 record that systems).
population, more than 12 percent in provided an indication of whether or c. EPA’s Use of the Highest Emissions
each category ‘‘would as a matter of not HMIWI were equipped with add-on Data to Reflect Uncontrolled (i.e.,
mathematical necessity have to be air pollution control and upon which Combustion-Controlled) Emissions. The
controlled.’’ Id., at 664, n. 8. The Court these estimates are based include a Court’s third concern regarded our use
then observed that ‘‘it is difficult to see survey of HMIWI in California and New of the highest of the test run data to
how it was rational to include any York, air permits from State regulatory reflect uncontrolled (i.e., combustion-
uncontrolled [i.e., combustion- agencies, responses to information controlled) emissions in cases where
controlled] units in the top 12 percent, collection requests, telephone contact regulatory data did not comprise the
necessary 12 percent of best performing TABLE 7.—UNCONTROLLED (I.E., COM- size category. Actual test data, where
sources. Our reason for this approach is BUSTION-CONTROLLED) PERFORM- available in the 1997 record, were the
the same as the reason described earlier ANCE VALUES initial type of pollutant-specific values
regarding new units for using the considered. Next, where the 1997 record
highest data point from MACT- Pollutant Performance has information indicating that a source
particular technology to reflect the (units) value 1 employed some type of add-on control
performance of that technology and but there are no test data or regulatory
HCl (ppmv) ........................... 2,770 limits for that source, an average of the
identify the ‘‘best controlled similar CO (ppmv) - .......................... 1 584.9
unit.’’ As the Court stated in NLA I, maximum dry and wet control system
Pb (micrograms per dry
‘‘where test results are relied upon, it standard cubic meter µg/
performance was determined for each
should involve the selection or use of dscm) ................................ 8,629 pollutant, and those values were added
test results in a manner which provides Cd (µg/dscm) ........................ 3,520 to the data set towards comprising the
Hg (µg/dscm) ........................ 6,543.4 best performing 12 percent. We believe
some assurance of the achievability of
PM (gr/dscf) .......................... 2 0.278 that use of these averages is an
the standard for the industry as a whole, CDD/CDF, total (ng/dscm) ... 2 8,102
appropriate method of estimating the
given the range of variable factors found CDD/CDF, TEQ (ng/dscm) ... 2 236
performance of HMIWI (1) where the
relevant to the standard’s achievability.’’ NOX (ppmv) .......................... 224.5 1997 record has limited information
(See 627 F.2d at 433). EPA reads the SO2 (ppmv) ........................... 46.39 indicating the presence of some type of
Court’s opinion in Sierra Club as at least 1 All performance values are measured at 7 add-on control but no test data for the
endorsing the principles of NLA I with percent oxygen. unit, and (2) where we are unsure if the
respect to existing units, as the Court 2 Based on 1-second combustion level
control is similar to, or is as efficient as,
described as ‘‘counterintuitive’’ the d. Determining the MACT Floor and those for which we have data, or if the
Sierra Club’s ‘‘proposition that an MACT for Existing Units. As discussed unit even employed a true control
‘achieved’ level may not be above, the Sierra Club Court identified device. As previously stated, we believe
‘achievable[.]’ ’’ (See 167 F.3d at 662). In some potential errors in EPA’s it very likely that there were fewer
addition, we also read CKRC as allowing methodology for determining the HMIWI with air pollution controls than
this approach, where no evidence in the existing source MACT floors for HMIWI. we estimated in 1997, and to which we
record contradicts the assumption that After reviewing the 1997 HMIWI record have assigned pollutant-specific average
‘‘factors other than the control have a in the context of the Court’s opinion, control device values. If, in fact, those
negligible effect [on emissions EPA agrees that, in determining the sources were employing true control
performance],’’ 255 F.3d at 866, and, MACT floor, the Agency should not devices, common sense dictates that
therefore, the presence or absence of have used regulatory limits that there wouldn’t have been the large
known effective MACT controls is the reflected higher emissions levels (and number of unit shut downs that
prime determinant of emissions that did not appear to be related to any occurred in response to the promulgated
performance. air pollution controls) than those standards. However, because we had
corresponding to EPA’s combustion- some indication that an add-on control
Where regulatory data indicating use device was in place on those sources,
controlled emission estimates.
of emissions control was absent in the Furthermore, as we examined the 1997 we recognize that the use of
1997 rulemaking record, EPA needed to record and our estimates of the uncontrolled (i.e., combustion-
find a surrogate emission limitation that performance of HMIWI where we had controlled) emission estimates (at
reflected uncontrolled (i.e., combustion- some indication that add-on controls promulgation) did not provide a
controlled) emissions, expecting, when may have been used, we determined reasonable estimate of their
not faced with data indicating that we should not have used performance. Similarly, use of
otherwise, that facilities with no combustion-controlled emission performance values associated with a
regulatory limits would not be estimates in the floor calculations to specific type of add-on control device
controlling their emissions with add-on represent the performance of those seems inappropriate when no details are
controls or other control methods sources. Additionally, for this available on the control device and
(beyond combustion control). In this rulemaking we propose that where there is, in fact, some doubt as to the
situation, EPA used the highest test data actual emissions test data reflecting presence of a true control device at all.
point from a well-operated HMIWI as a emissions performance was available in Despite the doubts of the presence of a
surrogate for the worst reasonably the 1997 record for use in determining true control device, the approach we
foreseeable circumstances. The highest the MACT floor, that data should take have selected assumes that the 1997
test data points reflect the normal, and precedence over other types of data (i.e., record is correct and assigns ‘‘default’’
unavoidable, variation in emissions that regulatory limits or performance performance values to the units that are
would be expected to recur over time. values). based on the expected performance of
Table 7 of this preamble summarizes the EPA’s reassessment of the 1997 the types of control devices used in the
performance values used for units for MACT floors and MACT decisions, industry in 1997. These default
based on an adjusted methodology that performance values, based on the
which there is no information indicating
addresses the Court’s issues discussed average of the maximum dry and wet
an APCD is present and there are no
above, results in proposed emission control system performance, also are
regulatory limits, or where regulatory used where regulatory limits exist but
limits do exist but reflect emissions limits that in many cases are more
stringent than the limits promulgated in are higher than the default performance
levels that are higher than the values for
1997. EPA’s first step in redoing the values.

uncontrolled (i.e., combustion-
MACT analysis based on the 1997 Table 8 of this preamble summarizes
controlled) units. record for existing HMIWI was to the performance values for HCl, Pb, Cd,
determine the pollutant-specific values Hg, CDD/CDF, and PM for units for
that make up the best performing 12 which the 1997 record has information
percent of existing units within each indicating that they employed some
type of add-on control but has no test (46.39 ppmv) are the same as those ‘‘combustion control’’ although, as
data or regulatory limits corresponding presented in Table 7 of this preamble. previously explained in this preamble,
to specific controls, or where regulatory The next step in the MACT analysis combustion control results in no
limits exist but are higher than the for existing HMIWI was to determine emission reductions for those
values based on an average of the the average emission limitation pollutants. ‘‘Good combustion’’ (i.e., 2-
maximum dry and wet control system achieved by the best-performing 12 second combustion) was identified as
performance. percent of existing sources where there the MACT floor technology for all size
are 30 or more sources in the category units for CO. ‘‘Combustion control’’
TABLE 8.—PERFORMANCE VALUES or subcategory. Our general approach to floors were identified for medium
BASED ON AVERAGE OF MAXIMUM identifying the average emission HMIWI for Pb, Cd, Hg, and CDD/CDF
DRY AND WET CONTROL SYSTEM limitation has been to use a measure of and for small HMIWI for HCl, Pb, Cd,
PERFORMANCE central tendency, such as the arithmetic Hg, and CDD/CDF. However, for these
mean or the median. If the median is pollutants for all medium and most
Pollutant Performance used when there are at least 30 sources, small HMIWI, we have decided to
(units) value 1
then the emission level that is at the propose limits that are more stringent
bottom of the best performing 6 percent than the ‘‘combustion control’’ floors
HCl (ppmv) ........................... 53.165
of sources (i.e., the 94th percentile) and are consistent with the control
Pb (µg/dscm) ........................ 568.5
Cd (µg/dscm) ........................ 83.65 represents the MACT floor control level. technology-based MACT floors that
Hg (µg/dscm) ........................ 459.5 We based our MACT floors for each were identified for large HMIWI for
PM (gr/dscf) .......................... 0.0195 pollutant within each size category on these pollutants (i.e., Pb, Cd, Hg, and
CDD/CDF, total (ng/dscm) ... 65.35 this approach. We then determined the CDD/CDF for medium HMIWI and HCl,
CDD/CDF, TEQ (ng/dscm) ... 1.296 technology associated with each Pb, Cd, Hg, and CDD/CDF for small
1 All performance values are measured at 7 ‘‘average of the best-performing 12 HMIWI). The control technologies
percent oxygen. percent’’ value by comparing the identified as the MACT floors for HCl
average values to average performance and PM for medium HMIWI, and for PM
The values for CO, NOX and SO2 are
based on the performance of data for wet scrubbers, dry injection for small HMIWI, provide an indication
combustion-controlled HMIWI because, fabric filters (also known as dry of the level of control of the other
as stated at proposal and promulgation scrubbers), and combustion controls (no pollutants—a level of technology that is
of the 1997 HMIWI standards, as well as add-on air pollution controls). Those consistent with those technologies
earlier in this preamble, CO emission pollutants with average values that were identified for large HMIWI. The
levels are affected by combustion higher than the relevant combustion- rationale for not basing the proposed
practices rather than the control systems controlled emission estimate were emission limits on other technologies
used by HMIWI; NOX control had not identified as having a ‘‘combustion that would result in even more stringent
been demonstrated on HMIWI; and the control’’ floor, even if the pollutant is limits can be found at 62 FR 48371–72.
acid gas controls used by HMIWI were not reduced by combustion control. The As at the 1997 promulgation, MACT for
not effective in reducing SO2 emissions technology needed to meet the small HMIWI that meet certain ‘‘rural
from HMIWI due to the low inlet levels remaining average values reflects the criteria’’ was determined to be at the
of SO2 associated with hospital/ technology used by the average unit in MACT floor level for each pollutant
medical/infectious waste. Therefore, for the top 12 percent and serves as the (i.e., no ‘‘beyond-the-floor’’-based
units (1) where the 1997 record contains basis for the MACT floor. EPA then emission limits).
information indicating that they considered, on a pollutant-specific Table 9 of this preamble shows the
employed some type of add-on control basis, technologies that were more average emission value, based on the
but for which there was no test data or stringent than the MACT floor ranking of emissions data, regulatory
regulatory limits, or (2) where regulatory technologies. data, and performance data, of each
limits existed but were higher than the Add-on control technology-based pollutant for the top 12 percent of
values for CO, NOX, or SO2 based on MACT floors were identified for large HMIWI in each subcategory. The values
combustion-controlled HMIWI, the HMIWI for HCl, Pb, Cd, Hg, PM, and in Table 9 allow EPA to identify the
performance values for CO (584.9 CDD/CDF. The MACT floor technology technology associated with the average
ppmv), NOX (224.5 ppmv), and SO2 for all size units for NOX and SO2 is unit in the top 12 percent of HMIWI.
TABLE 9.— AVERAGE EMISSION VALUES FOR TOP 12 PERCENT OF HMIWI 1

HMIWI size
Pollutant (units)
Small Medium Large
HCl (ppmv) ..................................................................................................................................................... 2,770 53 50

CO (ppmv) ..................................................................................................................................................... 100 100 100
Pb (mg/dscm) ................................................................................................................................................ 8.63 8.63 0.569
Cd (mg/dscm) ................................................................................................................................................ 3.52 3.52 0.084
Hg (mg/dscm) ................................................................................................................................................ 6.54 4.27 0.460
PM (gr/dscf) ................................................................................................................................................... 0.080 0.030 0.020
CDD/CDF, total (ng/dscm) ............................................................................................................................. 8,102 8,102 65.4
CDD/CDF, TEQ (ng/dscm) ............................................................................................................................ 236 236 1.30
NOX (ppmv) ................................................................................................................................................... 225 225 225

SO2 (ppmv) .................................................................................................................................................... 46.4 46.4 46.4
1 All emission values are measured at 7 percent oxygen.
Table 10 of this preamble shows the

technology associated with each average
emission value.
TABLE 10.—MACT FLOOR TECHNOLOGY

HMIWI Size
Pollutant (units)
Small Medium Large
HCl (ppmv) ..................................... combustion control ....................... dry scrubber .................................. dry scrubber.
CO (ppmv) ..................................... good combustion .......................... good combustion .......................... good combustion.
Pb (mg/dscm) ................................ combustion control ....................... combustion control ....................... wet scrubber.
Cd (mg/dscm) ................................ combustion control ....................... combustion control ....................... wet scrubber.
Hg (mg/dscm) ................................ combustion control ....................... combustion control ....................... dry scrubber.
PM (gr/dscf) ................................... low-efficiency wet scrubbber ........ moderate-efficiency wet scrubber moderate-efficiency wet scrubber.
CDD/CDF, total (ng/dscm) ............. combustion control ....................... combustion control ....................... wet scrubber.
CDD/CDF, TEQ (ng/dscm) ............ combustion control ....................... combustion control ....................... wet scrubber.
NOX (ppmv) ................................... combustion control ....................... combustion control ....................... combustion control.
SO2 (ppmv) .................................... combustion control ....................... combustion control ....................... combustion control.
For small units, the CO and PM given the limited amount of actual data t * standard deviation. This general
values indicate that good combustion available in the 1997 record and the equation has been used by EPA in
control (i.e., 2-second combustion) and resulting need for use of various, and similar analyses. (See, e.g., 68 FR 27650;
a low-efficiency wet scrubber reflect the often presumptive, types of information 69 FR 55235–7; 70 FR 28615.) We
CO and PM MACT floors. For medium to formulate the best performing 12 selected the 90th percentile confidence
units, as well as large units, the CO, percent of HMIWI. At promulgation, we level for this one-sided t-statistics test.
HCl, and PM values indicate that good recognized the need to account for The 90th percentile provides a
combustion control used in conjunction variability and did so as described variability factor appropriate for well-
with either a dry scrubber or moderate- earlier in this preamble. Although we controlled sources that is based on data
efficiency wet scrubber reflects the CO, maintain that the methodology we used from well-controlled sources (i.e., the
HCl, and PM MACT floors. As was not unreasonable given the only sources that are still in operation).
previously stated, EPA concluded that available information at promulgation, Table 11 of this preamble presents the
emission limits for small units that meet we now have additional information
the rural criteria should reflect the values determined by adding the
(the 2002 compliance test data for all of variability factors to the average
MACT floor level of control for all the currently operating units) for use in
pollutants. The average emission value emission values for each pollutant for
calculating pollutant-specific variability existing large and medium HMIWI. The
and MACT floor level of control for PM
factors. While these data were not table also presents the proposed revised
vary by unit size, and we are proposing
available at promulgation, they are the emission limits for existing large and
emission limits based on those levels of
only data available for providing a medium HMIWI necessary to respond to
control. The average emission values,
and associated MACT floor levels of quantitative assessment of variability of the Court’s remand and the percent
control, for CO, NOX, and SO2 are the emissions from well-controlled HMIWI. reduction limits for HCl, Pb, Cd, and Hg.
same for all size units. For most small To determine the pollutant-specific The percent reduction limits are based
units and all medium units, however, variability factors, a statistical analysis on average combustion-controlled
we concluded that emission limits for was conducted. Specifically, the emissions estimates and maximum
HCl, Pb, Cd, Hg, and CDD/CDF should emission limit achievable for each performance values for the MACT
reflect the MACT floor level of control pollutant was determined based on the identified for each pollutant for each
for large units for those pollutants. combination of actual emissions test subcategory. This is the same approach
The resulting numerical emission data, regulatory data, and estimated used at the time of promulgation of the
limits were determined by combining performance levels (as described earlier) 1997 rule, except that the proposed
the appropriate average emission value and a statistics-based variability factor percent reduction limits do not include
for each pollutant for each size HMIWI calculated for each pollutant. To the addition of 10 percent to the
with a variability factor. We believe it is calculate the variability factors, we used maximum performance values or the
necessary to account for variability the general equation: variability factor = rounding up of those figures.
TABLE 11.—AVERAGE EMISSION VALUES, CONSIDERING VARIABILITY, AND EMISSION LIMITS 1—EXISTING LARGE AND
MEDIUM HMIWI
Large Medium
Pollutant
(units) Average + Emission Average + Emission
variability limit variability limit
HCl (ppmv) ....................................... 78 78 or 93% reduction 2 ...................... 57.9 78 3 or 93% reduction 2.

CO (ppmv) ....................................... 110 40 2 .................................................. 113 40 2
Pb (mg/dscm) .................................. 0.78 0.78 or 71% reduction ..................... 9.02 0.78 3 or 71% reduction 3.
Cd (mg/dscm) .................................. 0.11 0.11 or 66% reduction ..................... 3.56 0.11 3 or 66% reduction 2.
Hg (mg/dscm) .................................. 0.64 0.55 2 or 87% reduction ................... 4.34 0.55 2 or 87% reduction 3.
PM (gr/dscf) ..................................... 0.025 0.015 2 ............................................. 0.043 0.030 2
TABLE 11.—AVERAGE EMISSION VALUES, CONSIDERING VARIABILITY, AND EMISSION LIMITS 1—EXISTING LARGE AND
MEDIUM HMIWI—Continued
Large Medium
Pollutant
(units) Average + Emission Average + Emission
variability limit variability limit
CDD/CDF, total (ng/dscm) ............... 115 115 ................................................... 8,150 115 3

CDD/CDF, TEQ (ng/dscm) .............. 2.16 2.2 .................................................... 237 2.2 3
NOX (ppmv) ..................................... 284 250 2 ................................................ 273 250 2
SO2 (ppmv) ...................................... 61 55 2 .................................................. 51.8 55 2
1 All emission values are measured at 7 percent oxygen.
2 No change from current emission limit.
3 Emission limit is the same as that for large HMIWI.
Table 12 of this preamble presents the rural HMIWI and for existing small
same information for existing small non- HMIWI meeting the rural criteria.
TABLE 12.— AVERAGE EMISSION VALUES, CONSIDERING VARIABILITY, AND EMISSION LIMITS 1—EXISTING SMALL AND
SMALL RURAL HMIWI
Large Medium
Pollutant
(units) Average + Average +
Emission limit Emission limit
variability variability
HCl (ppmv) ......................................................... 2,772 78 3 or 93% reduction 2. .................................... 3,125 2 3,100
CO (ppmv) ......................................................... 103 40 2 .................................................................... 109 2 40
Pb (mg/dscm) .................................................... 8.85 0.78 3 or 71% reduction 3 .................................. 8.88 8.9

Cd (mg/dscm) .................................................... 3.54 0.11 3 or 66% reduction 2 .................................. 3.54 4
Hg (mg/dscm) .................................................... 6.55 0.55 2 or 87% reduction 3 .................................. 6.56 6.6
PM (gr/dscf) ....................................................... 0.095 0.050 2 ............................................................... 0.089 2 0.086
CDD/CDF, total (ng/dscm) ................................. 8,335 115 3 .................................................................. 8,518 2 800
CDD/CDF, TEQ (ng/dscm) ................................ 239 2.2 3 ................................................................... 244 2 15
NOX (ppmv) ....................................................... 225 250 2 .................................................................. 273 2 250
SO2 (ppmv) ........................................................ 46.4 55 2 .................................................................... 51.8 2 55
1 All
emission values are measured at 7 percent oxygen.
2 No change from current emission limit.
3 Emission limit is the same as that for large HMIWI.
For pollutants where this remand best controlled HMIWI units could including risk. EPA does not interpret
analysis (based on the average of the achieve when using MACT controls. such technology review requirements to
best performing 12 percent of HMIWI The resulting emission limits being require another analysis of MACT floors
plus the variability factor) resulted in proposed for medium HMIWI for HCl for existing and new units, but rather
emission limits less stringent (i.e., and SO2; for small HMIWI for NOX and requires us to consider developments in
higher) than the current emission limits, SO2; and for small rural HMIWI for SO2 pollution control in the industry and
we retained the current emission limits. are the same as those being proposed for assess the costs of potentially stricter
This is because we see no reason to large HMIWI because, in these standards reflecting those
instances, the medium, small, and small developments. (See, id., at 34436–47.)
upwardly revise standards that the
rural HMIWI are expected to achieve Moreover, as a general matter, EPA has
regulated industry has already
reductions similar to large HMIWI. stated that where we determine that
demonstrated are achievable based on
compliance data. In fact, now that we B. Rationale for the Proposed existing standards are adequate to
have received the 2002 compliance data Amendments (CAA Section 129(a)(5) 5- protect public health with an ample
for HMIWI units, it is apparent that Year Review) margin of safety and prevent adverse
EPA’s estimate of the achievable In recent rulemakings (see, e.g., 71 FR environmental effects, it is unlikely that
emissions performance levels from use 34422, 34436–38 (June 14, 2006) EPA would revise MACT standards
of the identified MACT technology was (proposed amendments to the NESHAP merely to reflect advances in air
for Hazardous Air Pollutants for Organic pollution control technology. Id., at
reasonably accurate. While we are not in
Hazardous Air Pollutants from the 34437–38.
this proposal attempting to justify our
prior existing unit MACT floor Synthetic Organic Chemical Under CAA section 112(d)(6), the first
decisions post hoc based on new data manufacturing Industry)) EPA has round of technology review for MACT
that we could not have relied upon in addressed the similar technology review standards is subject to the same
the 1997 rulemaking itself, we note that, requirement under CAA section statutory timeframe as EPA’s residual
similar to the Mossville case, we 112(d)(6). EPA stated that the statute risk review under CAA section 112(f)(2),
currently find ourselves in a situation provides the Agency with broad with both reviews occurring 8 years
where actual emissions data fairly discretion to revise MACT standards as following initial promulgation of
confirms our prior estimates of what the we determine necessary, and to account MACT. We interpret CAA section
for a wide range of relevant factors, 129(a)(5)’s technology review
requirement as providing us the same limits for existing units, 1997 NSPS We believe this approach reflects the
degree of discretion in terms of whether units, and new units were developed most reasonable interpretation of the
to revise MACT standards, for the following the procedures discussed review requirement of CAA section
reasons discussed in those prior below. 129(a)(5), and is consistent with how we
rulemakings. (See, id., at 34436–38.) As background, with one exception have interpreted the similar review
However, the deadline for the first resulting from the analyses associated requirement of CAA section 112(d)(6)
round of technology review under with our response to the Court remand, regarding MACT standards promulgated
section 129(a)(5) (5 years following the proposed emission limits for new under section 112. (See 71 FR 27327–28;
MACT promulgation) does not coincide and existing units are based on the 69 FR 48350–51; and 70 FR 20008.) The
with the deadline for residual risk application of the same control language of section 129(a)(5) directs
review under section 112(f)(2) (9 years, technologies upon which the 1997 EPA to ‘‘review’’ our promulgated
in the case of HMIWI standards). MACT standards were based. For new standards under CAA section 111/129,
Therefore, this first section 129(a)(5) large and medium units, both the and to ‘‘revise such standards and
technology review for HMIWI does not current and proposed emission limits requirements’’ ‘‘in accordance with this
account for or reflect our residual risk are based upon good combustion and section and section 111.’’ It does not,
analysis. In future rounds of review the application of combined control however, direct EPA to conduct, at 5-
under section 129(a)(5) for the HMIWI systems that include both dry scrubbers year-intervals, new MACT floor and
standards, we intend to follow our (i.e., dry injection fabric filters or spray beyond-floor analyses based on each 5-
general policy, and for our technology dryer fabric filters) with carbon years’ changing information as to what
reviews and conclusions to be informed injection and wet scrubbers. The current might comprise the top 12 percent of
by our residual risk analysis, which we and proposed emission limits for new sources or constitute the best controlled
will have performed by that point. small units are based on good similar unit. There is no indication that
In exercising its discretion under CAA combustion and the application of a Congress intended for section 129(a)(5)
section 129(a)(5), EPA is proposing in moderate-efficiency wet scrubber. For to inexorably force existing source
this technology review to adopt large, medium, and most small existing standards progressively lower and lower
emission limits based on the 2002 data units, the current and proposed in each successive review cycle, the
because it believes that these limits emission limits are based on good likely result of requiring successive
represent the cost-effective operation of combustion control for CO; combustion floor determinations.
the MACT control technology. EPA is controls (i.e., no add-on controls) for
aware of the possibility that regulated NOX and SO2; and the application of Following MACT compliance in
units are likely to operate at a level either dry scrubbers or wet scrubbers September 2002, EPA obtained
somewhat below emission standards in (with various ‘‘efficiencies’’ depending compliance test reports from all
order to account for operational on the size of the unit) for the remaining operating HMIWI (76 units at 70
variability. It is not our intent to pollutants. The current emission limits facilities) and used those data to
preclude this practice through for one additional subcategory, existing evaluate MACT performance. When the
successive rounds of the section small rural units, are based solely on HMIWI regulations were first proposed
129(a)(5) technology review. EPA good combustion (i.e., the MACT floor in 1995, re-proposed in 1996, and
requests comment on its proposal (as identified in the 1997 analysis was not promulgated in 1997, only limited
outlined below) to adopt more stringent based on add-on control technology). information was available about HMIWI
emission limits in this instance through With the exception of PM, the proposed emission controls, and significant
its section 129(a)(5) technology review. emission limits for existing small rural engineering judgment was necessary in
units also are based solely on good selecting the emission limits. The year
1. How were the proposed emission 2002 compliance data show that the
combustion. In our remand analysis, we
limits developed? control technologies that were installed
identified a low-efficiency wet scrubber
The proposed revised emission limits as being the MACT floor for PM for and the practices that were
resulting from our 5-year review of the these units. Although all small rural implemented to meet the 1997 NSPS
HMIWI standards under section units currently use only good and emission guidelines achieved
129(a)(5) of the CAA are based on the combustion, to address this difference reductions somewhat superior to what
performance of units within the in the MACT floors (i.e., 1997 analysis we expected under the 1997 limits for
industry that currently are subject to the versus remand analysis), we are many of the pollutants. EPA used the
MACT standards. One set of emission proposing a PM emission limit for compliance test data to develop the
limits is proposed for existing HMIWI existing small rural units based on the emission limits contained in the
regulated under CAA section 111(d)/ application of low-efficiency wet amendments we are proposing under
129(b) emission guidelines, and another scrubbers to existing small non-rural the 5-year review. EPA believes that the
set of emission limits is proposed for units (i.e., MACT floor for small non- proposed emission limits more
new HMIWI (units commencing rural units in the 1997 analysis as well accurately reflect actual real-world
construction after February 6, 2007) as the remand analysis). While this HMIWI MACT performance than what
regulated under CAA section 111(b)/ performance level is associated with the we had estimated in 1997 and what we
129(a) NSPS. Units that were subject to expected performance of a low- re-estimated based on the 1997 record in
the 1997 NSPS as new units (referred to efficiency wet scrubber, the combustion response to the Court’s remand
as ‘‘1997 NSPS units’’ for the remainder controls in place on these six existing (discussed previously in this preamble).
of this preamble) will remain subject to small rural units achieve this We believe that it is necessary, as well
the 1997 NSPS (including revisions performance level, based on the initial as appropriate, to update the 1997
resulting from EPA’s response to the compliance tests for these units. promulgated standards based on the
Court remand), but will also be subject In performing this 5-year review, we actual performance of MACT
to any requirements of the revised have not recalculated new MACT floors, technologies in situations where
emission guidelines that are more but have proposed to revise the compliance test data indicate that the
stringent than the 1997 NSPS emission limits to reflect the actual technologies achieve better performance
requirements. The proposed emission performance of the MACT technologies. levels than those we previously
estimated based on the information used to calculate the percentile values. efficiency wet scrubbers to existing
available at the time of promulgation. In many cases, this visual comparison small non-rural units (i.e., we are
a. Existing Units. The first step in the revealed that the 99 percent UTL value proposing the same PM emission limit
analysis was to assess the performance or remand limit fell within a break in for small rural and non-rural units).
of the HMIWI currently subject to the the data that indicated a level of While many of the resulting proposed
emission guidelines with respect to each performance that the technologies, emission limits for small rural units are
regulated pollutant. We first examined considering variability, could readily significantly more stringent than the
the data separately for each unit size, achieve but that the ‘‘worst performing’’ 1997 promulgated limits, the proposed
and the data showed, for all pollutants units were not achieving during their limits more accurately reflect the actual
except PM, that the performance of compliance tests. Thus, our analyses performance of these units.
units with add-on controls, regardless of indicate that the emission limits that we Finally, we examined the available
size, (excluding small rural units, which selected reflect the actual performance data for calculating percent reduction
do not employ add-on controls), is of the MACT control technologies while requirements for HCl, Pb, Cd, and Hg.
similar. Therefore, we combined the also serving to require modest Percent reduction standards were
data, regardless of unit size, for all of the improvements in performance from included in the 1997 promulgated
pollutants except PM, and conducted units that are not achieving the standards for these pollutants, and we
analyses on the combined data sets. In performance levels demonstrated in are proposing to update these
addition, for the pollutants with practice by the control technologies requirements to reflect the now-known
emission limits based on good currently being used in the industry. actual performance of HMIWI utilizing
For small non-rural HMIWI, we used MACT controls. For HCl, we obtained
combustion and combustion control
a different methodology for assessing percent reduction data from five large
(i.e., no add-on controls), namely CO,
PM performance because there are only HMIWI using dry scrubbers (i.e., the
NOX, and SO2, the data for small rural
two units and, therefore, statistics are control technology upon which the
units also were combined with the data
not a useful tool. Both of the small non- emission limits for existing large,
for all of the other subcategories of
rural units are equipped with wet medium, and small non-rural units are
units. Analyses were performed on each
scrubbers. Because existing medium based), and these data showed percent
data set, and we calculated the 99
units are predominantly equipped with reductions from 94.2 percent to greater
percent upper tolerance limit (UTL),
wet scrubbers, the PM emission limit than 99 percent. To account for
which is the emission level that 99 developed using the 99 percent UTL variability, we based the proposed
percent of the HMIWI would be value of the data set for existing percent reduction requirement of 94
expected to achieve. A similar medium units also is being proposed for percent on the lowest percent reduction
methodology was used for stack test- small non-rural units. recorded during the individual test runs
based emission limits in the 5-year A different methodology also was (i.e., 94.2 percent). The three-run test
review recently conducted for large used for assessing performance of the that included the 94.2 percent value
municipal waste combustors (MWC). In six small rural HMIWI. To determine showed significant variability and
the preamble to that final action, EPA the actual performance of these small demonstrates the need to account for
indicated that analysis of data to combustion-controlled units while variability. The percent reduction
estimate emission limits to be enforced considering the inherent variability in values for the three runs ranged from
by stack test methods must be done emissions, we obtained test data for all 94.2 percent to 97.8 percent while there
using a different approach (i.e., lower six units (although, as allowed in the was no identifiable change in the
percent UTL) than where enforcement is emission guidelines, not all of the operation of the unit or the dry
to be based on CEMS and that the pollutants were tested at every unit) and scrubber. For Pb and Cd from existing
percentile must also reflect a reasonable selected as the emission limit the large, medium, and small non-rural
consideration of emissions variability highest individual test run from the HMIWI, we used the same methodology
and compliance limitations of stack compliance testing for HCl, Pb, Cd, Hg, as for HCl, and the data sets showed
testing (See 71 FR 27329). EPA further and CDD/CDF. This methodology uses even greater variability. For Hg, we used
indicated that for this type of actual test data to provide a reasonable the only available estimate of percent
technology review, the 99 percent UTL estimate of the performance of the small reduction. The proposed percent
was appropriate to use as a tool for rural units for these pollutants, where reduction standards are 71 percent for
estimating achievable emission levels statistics are not a useful tool, while Pb, 74 percent for Cd, and 96 percent for
for emission limits enforced by stack accounting for variability. There are Hg. The 5-year review methodology
testing. Id. In this proposed rulemaking, exceptions to this methodology for CO, used to assess performance of existing
the 99 percent UTL was used as the NOX, and SO2. As previously mentioned HMIWI resulted in no change to the PM
starting point for selecting the revised in this preamble, the CO, NOX, and SO2 standards for existing large and medium
emission limits. We compared the 99 data for small rural units were units, and CDD/CDF standards for
percent UTL values to several other combined with the CO, NOX, and SO2 existing small rural units. All of the
values, including the 1997 promulgated data for the other subcategories of units. other standards for existing HMIWI
emission limits and the revised limits The 99 percent UTL methodology was were adjusted based on either the 5-year
that we are proposing in response to the then used as the starting point, as review or the remand analyses.
Court’s remand (‘‘remand limits’’). For previously described in this preamble, Table 13 of this preamble summarizes
several pollutants, the value associated to determine proposed emission limits the emission limits promulgated in
with the 99 percent UTL was higher that would apply to all of the 1997, the emission limits resulting from
than the remand limit. In these cases, subcategories of existing HMIWI. the proposed response to the Court
we selected the remand limit, rather Another exception to this methodology remand, and the emission limits being
than the 99 percent UTL value, as the is the proposed emission limit for PM. proposed as a result of the 5-year review
proposed emission limit. We also As previously explained in this for existing HMIWI. Note that these
graphically compared the 99 percent preamble, we are proposing a PM proposed limits for existing HMIWI only
values and remand limits, where emission limit for existing small rural apply to units for which construction
applicable, to all of the data that were units based on the application of low- was commenced on or before June 20,
1996, or for which modification was

commenced before March 16, 1998.
TABLE 13.—SUMMARY OF 1997 PROMULGATED EMISSION LIMITS, PROPOSED REMAND RESPONSE EMISSION LIMITS, AND
PROPOSED 5-YEAR REVIEW LIMITS FOR EXISTING HMIWI
Pollutant Unit Promulgated Proposed remand response Proposed 5-year review
(units) size1 limit2 limit2 limit2
HCl (ppmv) ............................... L, M, S ........ 100 or 93% reduction .............. 78 or 93% reduction ................ 51 or 94% reduction.
SR ............... 3,100 ........................................ 3,100 ........................................ 398
CO (ppmv) ................................ All ................ 40 ............................................. 40 ............................................. 25
Pb (mg/dscm) ........................... L, M, S ........ 1.2 or 70% reduction ............... 0.78 or 71% reduction ............. 0.64 or 71% reduction.
SR ............... 10 ............................................. 8.9 ............................................ 0.60
Cd (mg/dscm) ........................... L, M, S ........ 0.16 or 65% reduction ............. 0.11 or 66% reduction ............. 0.060 or 74% reduction.
SR ............... 4 ............................................... 4 ............................................... 0.050
Hg (mg/dscm) ........................... L, M, S ........ 0.55 or 85% reduction ............. 0.55 or 87% reduction ............. 0.33 or 96% reduction.
SR ............... 7.5 ............................................ 6.6 ............................................ 0.25
PM (gr/dscf) .............................. L .................. 0.015 ........................................ 0.015 ........................................ 0.015
M ................. 0.03 .......................................... 0.030 ........................................ 0.030
S .................. 0.05 .......................................... 0.050 ........................................ 0.030
SR ............... 0.086 ........................................ 0.086 ........................................ 0.030
CDD/CDF, total (ng/dscm) ....... L, M, S ........ 125 ........................................... 115 ........................................... 115
SR ............... 800 ........................................... 800 ........................................... 800
CDD/CDF, TEQ (ng/dscm) ....... L, M, S ........ 2.3 ............................................ 2.2 ............................................ 2.0
SR ............... 15 ............................................. 15 ............................................. 15
NOX (ppmv) .............................. All ................ 250 ........................................... 250 ........................................... 212
SO2 (ppmv) ............................... All ................ 55 ............................................. 55 ............................................. 28
Table 14 of this preamble summarizes before the date 6 months after limits, as presented in Table 14 of this
the emission limits promulgated in 1997 promulgation of the proposed limits. preamble, in subparts Ec or Ce of 40
and the emission limits being proposed Also note that where the proposed 5- CFR part 60. Instead, they must consider
as a result of EPA’s response to the year review limits for existing HMIWI the proposed revisions to subpart Ec of
Court remand for the 1997 NSPS are more stringent than those resulting 40 CFR part 60 regarding existing
HMIWI. Note that these proposed limits from the remand response for 1997 HMIWI, as well as in the proposed
for 1997 NSPS HMIWI only apply to NSPS HMIWI, the more stringent limits revisions to subpart Ce of 40 CFR part
units for which construction was for existing HMIWI are included in the 60 regarding 1997 NSPS HMIWI, and
commenced after June 20, 1996, and on table as the limits being proposed. comply with the more stringent
or before the date of this proposal, or for HMIWI subject to the 1997 NSPS, emission limit.
which modification is commenced however, will not find these proposed
TABLE 14.—SUMMARY OF 1997 PROMULGATED EMISSION LIMITS AND PROPOSED LIMITS IN RESPONSE TO THE REMAND
FOR 1997 NSPS HMIWI
Pollutant Proposed remand response

Unit size1 Promulgated limit2
(units) limit2
HCl (ppmv) ........................................................ L, M, S .............. 15 or 99% reduction ......................................... 15 or 99% reduction.

CO (ppmv) ........................................................ L, M, S .............. 40 ...................................................................... 25 3.
Pb (mg/dscm) ................................................... L, M .................. 0.07 or 98% reduction ...................................... 0.060 or 98% reduction.
S ....................... 1.2 or 70% reduction ........................................ 0.64 3 or 71% reduction.
Cd (mg/dscm) ................................................... L, M .................. 0.04 or 90% reduction ...................................... 0.030 or 93% reduction.
S ....................... 0.16 or 65% reduction ...................................... 0.0603 or 74% reduction 3.
Hg (mg/dscm) ................................................... L, M, S .............. 0.55 or 85% reduction ...................................... 0.333 or 96% reduction 3.
PM (gr/dscf) ...................................................... L, M .................. 0.015 ................................................................. 0.0090
S ....................... 0.03 ................................................................... 0.018
CDD/CDF, total (ng/dscm) ................................ L, M .................. 25 ...................................................................... 20
S ....................... 125 .................................................................... 111
CDD/CDF, TEQ (ng/dscm) ............................... L, M .................. 0.6 ..................................................................... 0.53
S ....................... 2.3 ..................................................................... 2.0 3.
NOX (ppmv) ...................................................... L, M, S .............. 250 .................................................................... 212 3.
SO2 (ppmv) ....................................................... L, M, S .............. 55 ...................................................................... 28 3.
1L
= Large; M = Medium; S = Small
2 All
emission limits are measured at 7 percent oxygen.
3 Because the proposed 5-year review limit for existing HMIWI is more stringent than the one resulting from the remand response for 1997
NSPS HMIWI, the more stringent limit for existing HMIWI is being proposed.
b. New Units. The first step in the HMIWI was to assess the performance of combined dry/wet control system,
analysis for new large and medium the units currently operating a which is the control technology upon
which the 1997 NSPS for large and initial step in selection of these small HMIWI, we used the same
medium HMIWI was based. Four units emission limits, we used the methodology as for new large and
currently are operating such controls, performance values representative of medium units because the MACT
and we obtained compliance test data control with a moderate-efficiency wet control technology upon which the
for each unit for use in assessing scrubber as determined for the existing reductions are based is the same (wet
performance. We selected as the medium HMIWI. We then compared scrubbers). For Pb and Cd from new
proposed emission limit the highest these values to the values for new small small HMIWI, we used the same
individual test run from the compliance units developed in response to the methodology as for new large and
testing for each pollutant. This remand and, in each case, we selected medium units, except that the MACT
methodology uses actual test data from the more stringent value as the proposed control technology upon which the
the best-controlled sources in the emission limit. reductions are based is a wet scrubber.
industry to provide a reasonable To determine proposed percent For Hg, we used the only available
estimate of the performance of these reduction requirements for new units estimate of percent reduction. The
units, while accounting for variability. for HCl, Pb, Cd, and Hg, we followed a proposed percent reduction standards
In several instances, the emission limit methodology similar to that used for for new small units are 99 percent for
suggested by the highest run from the existing units. For HCl, we obtained HCl, 71 percent for Pb, 74 percent for
four combined-control sources was percent reduction data from two units Cd, and 96 percent for Hg. The 5-year
higher than either the emission limit for controlled with the MACT control review methodology used to assess
new sources that we are proposing in technology for HCl for new large and performance of new units resulted in no
response to the Court remand or the 5- medium units (wet scrubbers), and these change to the HCl standards for all new
year review emission limit that we are data showed percent reductions greater units. All of the other standards for new
proposing for existing sources. This was than 99 percent. To account for units were adjusted based on either the
likely a result of the small amount of variability, we based the percent 5-year review or the remand analyses.
data that we used to establish the limits, reduction requirement of 99 percent on Table 15 of this preamble summarizes
and, in these instances, we are the lowest percent reduction recorded the emission limits promulgated in 1997
proposing the most stringent among during the individual test runs (i.e., 99.1 and the emission limits being proposed
these three limits for new sources. percent). We used the same as a result of the 5-year review for new
Although there are no small HMIWI methodology for each of the three HMIWI. Note that these proposed limits
subject to the current NSPS, we are metals for new large and medium units, for new HMIWI only apply to units for
proposing emission limits based on the and the corresponding percent which construction is commenced after
performance of moderate-efficiency wet reduction standards based on the MACT the date of this proposal, or for which
scrubbers, which is the control control technology (dry scrubbers) are modification is commenced on or after
technology upon which the 1997 limits 99 percent for Pb, 99 percent for Cd, and the date 6 months after promulgation of
for new small units was based. As an 96 percent for Hg. For HCl from new the proposed limits.
TABLE 15.—SUMMARY OF 1997 PROMULGATED EMISSION LIMITS AND PROPOSED 5-YEAR REVIEW LIMITS FOR NEW
HMIWI
Pollutant Proposed 5-year review
Unit size 1 Promulgated limit 2
(units) limit 2
HCl (ppmv) ........................................................ L, M, S .............. 15 or 99% reduction ......................................... 15 or 99% reduction.

CO (ppmv) ........................................................ L, M, S .............. 40 ...................................................................... 25
Pb (mg/dscm) ................................................... L, M .................. 0.07 or 98% reduction ...................................... 0.060 or 99% reduction.
S ....................... 1.2 or 70% reduction ........................................ 0.64 or 71% reduction.
Cd (mg/dscm) ................................................... L, M .................. 0.04 or 90% reduction ...................................... 0.0050 or 99% reduction.
S ....................... 0.16 or 65% reduction ...................................... 0.060 or 74% reduction.
Hg (mg/dscm) ................................................... L, M .................. 0.55 or 85% reduction ...................................... 0.19 or 96% reduction.
S ....................... 0.55 or 85% reduction ...................................... 0.33 or 96% reduction.
PM (gr/dscf) ...................................................... L, M .................. 0.015 ................................................................. 0.0090
S ....................... 0.03 ................................................................... 0.018
CDD/CDF, total (ng/dscm) ................................ L, M .................. 25 ...................................................................... 16
S ....................... 125 .................................................................... 111
CDD/CDF, TEQ (ng/dscm) ............................... L, M .................. 0.6 ..................................................................... 0.21
S ....................... 2.3 ..................................................................... 2.0
NOX (ppmv) ...................................................... L, M, S .............. 250 .................................................................... 212
SO2 (ppmv) ....................................................... L, M .................. 55 ...................................................................... 21
S ....................... 55 ...................................................................... 28
1L = Large; M = Medium; S = Small.
2. How did EPA determine the proposed proposing retaining the current These proposed requirements were
performance testing and monitoring requirements of the rule and adding the selected to provide additional assurance
requirements? following requirements: Annual that sources continue to operate at the
We are proposing minor adjustments inspections of scrubbers and fabric levels established during their initial
to the performance testing and filters; and one-time testing of the ash performance test. The proposed
monitoring requirements that were handling operations at the time of the amendments would allow sources to use
promulgated in 1997. For existing next compliance test using EPA Method the results of previous emissions tests to
HMIWI and 1997 NSPS HMIWI, we are 22 of appendix A of 40 CFR part 60. demonstrate compliance with the
revised emission limits as long as the combustion unit is operating properly. and enforcement of the same emission
sources certify that the previous test The proposed amendments incorporate limits must be carefully considered in
results are representative of current the use of performance specification relation to an appropriate averaging
operations. Only those sources whose (PS)–4B (Specifications and Test period for data reduction. EPA
previous emissions tests do not Procedures for Carbon Monoxide and considered this issue and concluded the
demonstrate compliance with one or Oxygen Continuous Monitoring Systems use of a 24-hour block average was
more revised emission limits would be in Stationary Sources) of appendix B of appropriate to address PM emissions
required to conduct another emissions 40 CFR part 60. variability, and EPA has included the
test for those pollutants (note that The proposed CO emission limits are use of a 24-hour block average in the
sources are already required to test for based on data from infrequent (normally proposed rule. The 24-hour block
HCl, CO, and PM on an annual basis). annual) stack tests and compliance average would be calculated following
Additional requirements also are would be demonstrated by stack tests. procedures in EPA Method 19 of
proposed for new HMIWI. For new The change to use of CO CEMS for appendix A of 40 CFR part 60. An
sources, we are proposing retaining the measurement and enforcement of the owner or operator of an HMIWI unit
current requirements and adding the same emission limits must be carefully who wishes to use PM CEMS would be
following requirements: Use of CO considered in relation to an appropriate required to notify EPA 1 month before
CEMS; annual inspections of scrubbers averaging period for data reduction. starting use of PM CEMS and 1 month
and fabric filters; use of bag leak EPA considered this issue and before stopping use of the PM CEMS.
detection systems on fabric filter-based concluded the use of a 24-hour block d. Other CEMS and Monitoring
control systems; and annual testing of average was appropriate to address CO Systems. EPA also is proposing the
the ash handling operations using EPA emissions variability, and EPA has optional use of HCl CEMS, multi-metals
Method 22 of appendix A of 40 CFR part included the use of a 24-hour block CEMS, Hg CEMS, and semi-continuous
60. For existing sources, in addition to average in the proposed rule. The 24- dioxin monitoring as alternatives to the
the proposed changes in monitoring hour block average would be calculated existing methods for demonstrating
requirements, we also are proposing to following procedures in EPA Method 19 compliance with the HCl, metals (Pb,
allow for the optional use of bag leak of appendix A of 40 CFR part 60. Cd, and Hg), and CDD/CDF emissions
detection systems. We also are clarifying Facilities electing to use CO CEMS as an limits. For the reasons explained above
that the rule allows for the following optional method would be required to for CO CEMS and PM CEMS, EPA has
optional CEMS use: CO CEMS for notify EPA 1 month before starting use concluded that the use of 24-hour block
existing sources and 1997 NSPS of CO CEMS and 1 month before averages would be appropriate to
sources; and PM CEMS, HCl CEMS, stopping use of the CO CEMS. In address emissions variability, and EPA
multi-metals CEMS, Hg CEMS, and addition, EPA specifically requests has included the use of 24-hour block
semi-continuous dioxin monitoring for comment on whether continuous averages in the proposed rule. The 24-
existing, 1997 NSPS, and new sources. monitoring of CO emissions should be hour block averages would be calculated
The optional use of HCl CEMS, multi- required for all existing HMIWI and all following procedures in EPA Method 19
metals CEMS, and semi-continuous 1997 NSPS HMIWI. of appendix A of 40 CFR part 60.
dioxin monitoring will be available on c. PM CEMS. The proposed Although final performance
the date a final performance amendments would allow the use of PM specifications are not yet available for
specification for these monitoring CEMS as an alternative testing and HCl CEMS and multi-metals CEMS, EPA
systems is published in the Federal monitoring method. Owners or is considering development of
Register or the date of approval of a site- operators who choose to rely on PM performance specifications. The
specific monitoring plan. The proposed CEMS would be able to discontinue proposed rule specifies that these
testing and monitoring provisions are their annual PM compliance test. In options will be available to a facility on
discussed below. addition, because units that demonstrate the date a final performance
a. Bag Leak Detection Systems. The compliance with the PM emission limits specification is published in the Federal
proposed amendments would provide, with a PM CEMS would clearly be Register or the date of approval of a site-
as an alternative PM monitoring meeting the opacity standard, specific monitoring plan.
technique for existing sources and 1997 compliance demonstration with PM The use of HCl CEMS would allow
NSPS sources and a requirement for CEMS would be considered a substitute the discontinuation of HCl sorbent flow
new sources, the use of bag leak for opacity testing. Owners and rate monitoring, scrubber liquor pH
detection systems on HMIWI controlled operators that use PM CEMS also would monitoring, and the annual testing
with fabric filters. Bag leak detection be able to discontinue their monitoring requirements for HCl. EPA has proposed
systems have been applied successfully of minimum wet scrubber pressure PS–13 (Specifications and Test
at many industrial sources. EPA is drop, horsepower, or amperage. The Procedures for Hydrochloric Acid
proposing to drop the opacity testing proposed amendments incorporate the Continuous Monitoring Systems in
requirements for HMIWI that use bag use of PS–11 (Specifications and Test Stationary Sources) of appendix B of 40
leak detection systems. Procedures for Particulate Matter CFR part 60 and believes that
b. CO CEMS. The proposed Continuous Emission Monitoring performance specification can serve as
amendments would require the use of Systems at Stationary Sources) of the basis for a performance specification
CO CEMS for new sources, and allow appendix B of 40 CFR part 60 for PM for HCl CEMS use at HMIWI. In
the use of CO CEMS on existing sources CEMS, and PS–11 QA Procedure 2 to addition to the procedures used in
and 1997 NSPS sources. Owners and ensure that PM CEMS are installed and proposed PS–13 for initial accuracy
operators that use CO CEMS would be operated properly and produce good determination using the relative
able to discontinue their annual CO quality monitoring data. accuracy test, a comparison against a
compliance test as well as their The proposed PM emission limits are reference method, EPA is taking
monitoring of the secondary chamber based on data from infrequent (normally comment on an alternate initial
temperature. The continuous annual) stack tests and compliance accuracy determination procedure,
monitoring of CO emissions is an would be demonstrated by stack tests. similar to the one in section 11 of PS–
effective way of ensuring that the The use of PM CEMS for measurement 15 (Performance Specification for
Extractive FTIR Continuous Emissions 15 using the dynamic or analyte spiking continuous automated sampling system
Monitor Systems in Stationary Sources) procedure. would take effect on the date a final
of appendix B of 40 CFR part 60 using Relative to the use of Hg CEMS, EPA performance specification is published
the dynamic or analyte spiking believes that PS–12A (Specifications in the Federal Register or the date of
procedure. and Test Procedures for Total Vapor approval of a site-specific monitoring
Phase Mercury Continuous Emission plan. Semi-continuous monitoring of
EPA believes multi-metals CEMS can
Monitoring Systems in Stationary dioxin would allow the discontinuation
be used in many applications, including
Sources) of appendix B of 40 CFR part of fabric filter inlet temperature
HMIWI. EPA has monitored side-by-side
60 can provide the basis for using Hg monitoring. Dioxin/furan sorbent flow
evaluations of multi-metals CEMS with CEMS at HMIWI. An owner or operator
EPA Method 29 of appendix A of 40 rate monitoring could not be eliminated
of an HMIWI unit who wishes to use Hg in favor of semi-continuous monitoring
CFR part 60 at industrial waste CEMS would be required to notify EPA
incinerators and found good correlation. of dioxin because it also is an indicator
1 month before starting use of Hg CEMS of Hg control. Additionally, there is no
EPA also approved the use of multi- and 1 month before stopping use of the
metals CEMS as an alternative annual CDD/CDF test that could be
Hg CEMS. The use of multi-metals
monitoring method at a hazardous waste eliminated. If semi-continuous
CEMS or Hg CEMS would allow the
combustor. EPA believes it is possible to monitoring of dioxin as well as multi-
discontinuation of wet scrubber outlet
adapt proposed PS–10 (Specifications flue gas temperature monitoring. metals CEMS or Hg CEMS are used, Hg
and Test Procedures for Multi-metals Mercury sorbent flow rate monitoring sorbent flow rate monitoring and CDD/
Continuous Monitoring Systems in could not be eliminated in favor of a CDF sorbent flow rate monitoring (in
Stationary Sources) of appendix B of 40 multi-metals CEMS or Hg CEMS both cases activated carbon is the
CFR part 60 or other EPA performance because it also is an indicator of CDD/ sorbent) could be eliminated. EPA
specifications to allow the use of multi- CDF control. Additionally, there is no requests comment on other parameter
metals CEMS at HMIWI. In addition to annual metals test that could be monitoring requirements that could be
the procedures used in proposed PS–10 eliminated. eliminated upon use of any or all of the
for initial accuracy determination using The semi-continuous monitoring of optional CEMS discussed above. Table
the relative accuracy test, a comparison dioxin would entail use of a continuous 16 of this preamble presents a summary
against a reference method, EPA is automated sampling system and of the HMIWI operating parameters, the
taking comment on an alternate initial analysis of the sample using EPA pollutants influenced by each
accuracy determination procedure, Reference Method 23 of appendix A of parameter, and alternative monitoring
similar to the one in section 11 of PS– 40 CFR part 60. The option to use a options for each parameter.
TABLE 16.—SUMMARY OF HMIWI OPERATING PARAMETERS, POLLUTANTS INFLUENCED BY EACH PARAMETER, AND
ALTERNATIVE MONITORING OPTIONS FOR EACH PARAMETER
Pollutants Influenced by Operating Parameter (by Con-
trol Device Type)
Operating parameter/monitoring require- Alternative monitoring options
ment Combined sys-
Dry scrubber Wet scrubber tem
Maximum charge rate ................................ All 1 ................... All 1 ................... All 1 ................... None.
Minimum secondary chamber temperature PM, CO, CDD/ PM, CO, CDD/ PM, CO, CDD/ CO CEMS 2.
CDF. CDF. CDF.
Maximum fabric filter inlet temperature ..... CDD/CDF ......... ........................... CDD/CDF ......... Semi-continuous dioxin monitoring system
(SCDMS).
Minimum CDD/CDF sorbent flow rate ....... CDD/CDF ......... ........................... CDD/CDF ......... SCDMS and multi-metals CEMS or Hg
CEMS.
Minimum Hg sorbent flow rate ................... Hg ..................... ........................... Hg.
Minimum HCl sorbent flow rate ................. HCl ................... ........................... HCl .................... HCl CEMS.
Minimum scrubber pressure drop/ horse- ........................... PM .................... PM .................... PM CEMS.
power amperage.
Minimum scrubber liquor flow rate ............. ........................... HCl, PM, Cd, HCl, PM, Cd, HCl CEMS, PM CEMS, multi-metals
Pb, Hg, CDD/ Pb, Hg, CDD/ CEMS, and SCDMS.
CDF. CDF.
Minimum scrubber liquor pH ...................... ........................... HCl ................... HCl .................... HCl CEMS.
Maximum flue gas temperature (wet ........................... Hg ..................... ........................... Hg CEMS or multi-metals CEMS.
scrubber outlet).
Do not use bypass stack (except during All 1 ................... All 1 ................... All 1 ................... None.
startup, shutdown, and malfunction).
Air pollution control device inspections ...... All 1 ................... All 1 ................... All 1 ................... None.
1 ‘‘All’’
pollutants designation does not include SO2 and NOX, which are regulated at combustion-controlled levels (no add-on controls) and
have no associated parameter monitoring.
2 Optional method for existing and 1997 NSPS sources; required for new sources.
Table 17 of this preamble presents a and approved alternative compliance

summary of the HMIWI test methods methods.
TABLE 17.—SUMMARY OF HMIWI TEST METHODS AND APPROVED ALTERNATIVE METHODS

Pollutant/parameter Test method(s) 1 Approved alternative method(s) Comments
PM ......................................................... Method 5, Method PM CEMS ............................................. PM CEMS are optional for all sources
29. in lieu of annual PM test.
CO ......................................................... Method 10 ............. CO CEMS ............................................. CO CEMS are optional for existing and
1997 NSPS sources in lieu of annual
CO test; CO CEMS are required for
new sources.
HCl ......................................................... Method 26 or HCl CEMS ............................................ HCl CEMS are optional for all sources
Method 26A. in lieu of annual HCl test.
Cd .......................................................... Method 29 ............. Multi-metals CEMS.
Pb .......................................................... Method 29 ............. Multi-metals CEMS.
Hg .......................................................... Method 29 ............. ASTM D6784–02, multi-metals CEMS
or Hg CEMS.
CDD/CDF .............................................. Method 23 ............. Semi-continuous dioxin monitoring sys-
tem.
Opacity .................................................. Method 22 ............. Bag leak detection system or PM Bag leak detection systems are op-
CEMS. tional for existing and 1997 NSPS
sources; and are required for new
sources.
Flue and exhaust gas analysis ............. Method 3, 3A, or ASME PTC 19–10–1981 Part 10.
3B.
Opacity from ash handling .................... Method 22 ............. None.
1 EPA Reference Methods in appendix A of 40 CFR part 60.
V. Impacts of the Proposed Action for A. What are the primary air impacts? that we estimate may be required to
Existing Units As a result of the modest enable some facilities to achieve the
improvements estimated to be required proposed emission limits. We estimate
The emission limits for existing that the adjustments could result in
at 22 HMIWI such that they would
HMIWI that we are proposing as part of emissions of 211 lb/yr of PM; 1,880 lb/
achieve the proposed emission limits,
this action are based on the actual yr of CO; 1,230 lb/yr of NOX; and 1,450
EPA estimates that a total of
performance of the MACT control lb/yr of SO2 from the increased
approximately 24,700 pounds per year
technologies. This proposed action is (lb/yr) of the regulated pollutants would electricity and natural gas usage.
expected to result in modest be reduced. Approximate reductions by E. What are the cost and economic
improvements in performance being pollutant follow: impacts?
required by HMIWI that are not • HCl—20,600 lb/yr
achieving the performance levels • CO—400 lb/yr EPA estimates that the national total
demonstrated in practice by the control • Pb—35 lb/yr costs for the 72 existing HMIWI and 4
technologies currently being used in the • Cd—3 lb/yr 1997 NSPS HMIWI to comply with this
industry. Based on compliance test • Hg—30 lb/yr proposed action would be
reports from all existing operating • PM—2,700 lb/yr approximately $488,000 in the first year
HMIWI (72 units at 67 facilities) • CDD/CDF—0.0007 lb/yr of compliance. This estimate includes
• NOX—200 lb/yr the costs that would be incurred by the
following MACT compliance in • SO2—700 lb/yr
September 2002, 18 existing large 22 HMIWI that we anticipate needing to
HMIWI and 4 existing medium HMIWI B. What are the water and solid waste improve performance (i.e., costs of
are likely to find it necessary to improve impacts? improvements in emissions control and
performance of their units in order to emissions tests for pollutants for which
EPA estimates that approximately 80
achieve the proposed emission limits the improvements are made), and the
tpy of additional solid waste and
additional monitoring (i.e., annual
which their compliance test data 267,000 gallons per year of additional
control device inspections), testing (i.e.
indicates they would not meet. The wastewater would be generated as a
initial Method 22 test), and
modest improvements anticipated result of the increase of lime use by
recordkeeping and reporting costs that
include adding lime (for SO2), some facilities.
would be incurred by all 76 HMIWI as
increasing lime use (for HCl and SO2), C. What are the energy impacts? a result of this proposed action.
increasing natural gas use (for CO and Approximately 50 percent of the
CDD/CDF), and increasing scrubber EPA estimates that approximately
3,600 megawatt-hours per year of estimated total cost in the first year is
horsepower (for Pb, Cd, and Hg). for emissions control, 11 percent is for
additional electricity would be required
Facilities may resubmit previous monitoring, 32 percent is for testing,
to support the increase in scrubber
compliance test data that indicates that and 7 percent is for recordkeeping and
horsepower that we estimate would be
their HMIWI meets the proposed reporting. National total costs for
required to enable some facilities to
emission limits if the facility certifies achieve the proposed emission limits. subsequent years are estimated to be
that the test results are representative of approximately $308,000 per year, with
current operations. Those facilities D. What are the secondary air impacts? approximately 78 percent of the total
would then not be required to test for Secondary air impacts associated with cost for emissions control, 18 percent
those pollutants to prove compliance this proposed action are direct impacts for monitoring, and 3 percent for testing.
with the emission limits. that result from the increase in natural Economic impact analyses focus on
gas use and/or wet scrubber horsepower changes in market prices and output
levels. If changes in market prices and VII. Relationship of the Proposed of the CAA section 112(c)(6) pollutants:
output levels in the primary markets are Action to Section 112(c)(6) of the CAA Dioxins, furans, and Hg. As explained
significant enough, impacts on other Section 112(c)(6) of the CAA requires below, the air pollution controls
markets are also examined. EPA’s EPA to identify categories of sources of necessary to comply with the
economic impact analysis for this seven specified pollutants to assure that requirements of the HMIWI NSPS and
proposed action assessed the magnitude sources accounting for not less than 90 emission guidelines also effectively
of the cost of market changes resulting percent of the aggregate emissions of reduce emissions of the following CAA
from the proposed amendments by each such pollutant are subject to section 112(c)(6) pollutants that are
comparing annualized costs to annual standards under CAA section 112(d)(2) emitted from HMIWI units: POM and
sales. We were able to assess the cost of or 112(d)(4). EPA has identified medical PCBs. Although the CAA section 129
waste incinerators as a source category HMIWI standards do not have separate,
market changes for 70 HMIWI (sales
that emits five of the seven CAA section specific emissions standards for PCBs
information was unavailable for the
112(c)(6) pollutants: Polycyclic organic and POM, emissions of these two CAA
other 6 units). For purposes of assessing section 112(c)(6) pollutants are
economic impacts of the proposed matter (POM), dioxins, furans, Hg, and
polychlorinated biphenyls (PCBs). (The effectively controlled by the same
action, the total annualized cost of this control measures used to comply with
proposed action is estimated to be POM emitted by HMIWI is composed of
16 polyaromatic hydrocarbons (PAH) the numerical emissions limits for the
$328,000 and was determined by first enumerated CAA section 129 pollutants.
annualizing at 7 percent over 15 years and extractable organic matter (EOM).)
In the Federal Register notice Source Specifically, as byproducts of
the difference between the first year combustion, the formation of PCBs and
Category Listing for Section 112(d)(2)
costs and subsequent year costs for each POM is effectively reduced by the
Rulemaking Pursuant to Section
of the 76 HMIWI, and adding to that combustion and post-combustion
112(c)(6) Requirements, 63 FR 17838,
value the subsequent year costs for each 17849, Table 2 (1998), EPA identified practices required to comply with the
HMIWI; followed by then combining the medical waste incinerators (now CAA section 129 standards. Any PCBs
annualized costs for the 76 HMIWI. The referred to as HMIWI) as a source and POM that do form during
$328,000 was distributed among the 76 category ‘‘subject to regulation’’ for combustion are further controlled by the
HMIWI, resulting in cost-to-sales ratios purposes of CAA section 112(c)(6) with various post-combustion HMIWI
ranging from 0.0006 percent to 0.06 respect to the CAA section 112(c)(6) controls. The add-on PM control
percent, with an average cost-to-sales pollutants that HMIWI emit. HMIWI are systems (either fabric filter or wet
ratio of 0.003 percent. Because of the solid waste incineration units currently scrubber) and activated carbon injection
small size of these regulatory costs and regulated under CAA section 129. For in the fabric filter-based systems further
estimated impacts, no additional market purposes of CAA section 112(c)(6), EPA reduce emissions of these organic
analysis is needed. Neither the modest has determined that standards pollutants, as well as reducing Hg
national costs nor the facility level costs promulgated under CAA section 129 are emissions. The post-MACT compliance
are anticipated to significantly impact substantively equivalent to those tests at currently operating HMIWI show
any market. promulgated under CAA section 112(d). that the HMIWI MACT regulations
(See Id. at 17845; see also 62 FR 33625, reduced Hg emissions by greater than 80
VI. Impacts of the Proposed Action for 33632 (1997).) As discussed in more percent and CDD/CDF emissions by
New Units detail below, the CAA section 129 about 90 percent from pre-MACT levels.
standards effectively control emissions In light of the fact that similar controls
The current NSPS apply to HMIWI for have been demonstrated to effectively
of the five identified CAA section
which construction began after June 20, reduce emissions of POM and PCBs
112(c)(6) pollutants. Further, since CAA
1996, or for which modification began from another incineration source
section 129(h)(2) precludes EPA from
after March 16, 1998. There are three category (municipal solid waste
regulating these substantial sources of
new HMIWI and one modified HMIWI the five identified CAA section 112(c)(6) combustors), it is, therefore, reasonable
that are subject to the current NSPS. No pollutants under CAA section 112(d), to conclude that POM and PCB
additional units have become subject to EPA cannot further regulate these emissions are substantially reduced at
the NSPS since 2002. Considering this emissions under that CAA section. As a all 76 HMIWI. Thus, while the proposed
information, EPA does not anticipate result, EPA considers emissions of these rule does not identify specific limits for
any new HMIWI, and, therefore, no five pollutants from HMIWI units POM and PCB, they are, for the reasons
impacts of the proposed standards for ‘‘subject to standards’’ for purposes of noted above, nonetheless ‘‘subject to
new units. However, in the unlikely CAA section 112(c)(6). regulation’’ for purposes of section
event that a new HMIWI is constructed, As required by the statute, the CAA 112(c)(6) of the CAA.
we are proposing new emission limits section 129 HMIWI standards include VIII. Statutory and Executive Order
for those units based on performance of numeric emission limitations for the Reviews
the control technology upon which nine pollutants specified in that section.
The combination of good combustion A. Executive Order 12866: Regulatory
current NSPS limits are based, as well
practices and add-on air pollution Planning and Review
as additional monitoring requirements,
including use of CO CEMS and use of control equipment (dry sorbent injection Under Executive Order 12866 (58 FR
bag leak detection systems for fabric fabric filters, wet scrubbers, or 51735; October 4, 1993), this proposed
filters. Because EPA does not anticipate combined fabric filter and wet scrubber action is a ‘‘significant regulatory
any new HMIWI, we, therefore, do not systems) effectively reduces emissions action’’ because it is likely to raise novel
expect there to be any air impacts, water of the pollutants for which emission legal or policy issues arising out of legal
or solid waste impacts, energy impacts, limits are required under CAA section mandates, the President’s priorities, or
129: Hg, CDD/CDF, Cd, Pb, PM, SO2, the principles set forth in the Executive
or cost or economic impacts associated
HCl, CO, and NOX. Thus, the NSPS and Order. Accordingly, EPA submitted this
with the proposed standards for new
emissions guidelines specifically proposed action to the Office of
sources.
require reduction in emissions of three Management and Budget (OMB) for
review under Executive Order 12866, Federal agency. This includes the time entities, small entity is defined as
and any changes made in response to needed to review instructions; develop, follows: (1) A small business as defined
OMB recommendations have been acquire, install, and utilize technology by the Small Business Administration’s
documented in the docket for this and systems for the purposes of (SBA) regulations at 13 CFR 121.201; (2)
action. collecting, validating, and verifying a small governmental jurisdiction that is
information, processing and a government of a city, county, town,
B. Paperwork Reduction Act
maintaining information, and disclosing school district or special district with a
The information collection and providing information; adjust the population of less than 50,000; or (3) a
requirements associated with this existing ways to comply with any small organization that is any not-for-
proposed action are included in the previously applicable instructions and profit enterprise that is independently
information collection requirements requirements; train personnel to be able owned and operated and is not
addressing the HMIWI standards in to respond to a collection of dominant in its field.
their entirety, which have been information; search data sources; After considering the economic
submitted for approval to the OMB complete and review the collection of impacts of this proposed action on small
under the Paperwork Reduction Act, 44 information; and transmit or otherwise entities, I certify that this action will not
U.S.C. 3501 et seq. The Information disclose the information. have a significant economic impact on
Collection Request (ICR) documents An Agency may not conduct or a substantial number of small entities.
prepared by EPA have been assigned sponsor, and a person is not required to Because none of the HMIWI facilities
EPA ICR number 1899.04 for subpart Ce respond to a collection of information are expected to be significantly
and 1730.05 for subpart Ec. unless it displays a currently valid OMB impacted by this proposed action, that
The requirements in this proposed control number. The OMB control also means that none of the four small
action result in industry recordkeeping numbers for EPA’s regulations are listed entity-owned facilities would be
and reporting burden associated with in 40 CFR part 9. expected to be significantly impacted.
review of the amendments for all To comment on the Agency’s need for None of the 22 HMIWI that we estimate
HMIWI, initial EPA Method 22 testing this information, the accuracy of the would need to make improvements in
for all HMIWI, annual inspections of provided burden estimates, and any order to meet the proposed emission
scrubbers and fabric filters for all suggested methods for minimizing limits are owned by small entities. The
HMIWI, and stack testing and respondent burden, including the use of only estimated economic impacts on
development of new parameter limits automated collection techniques, EPA small entities would result from the
for HMIWI that need to make has established a public docket for this additional monitoring requirements
performance improvements. The total action, which includes these ICR (annual control device inspections),
nationwide recordkeeping and reporting documents, under Docket ID No. EPA– testing requirements (one-time EPA
burden of this proposed action is HQ–OAR–2006–0534. Submit any Method 22 testing), and associated
estimated at 722 hours at a cost of comments related to the ICR documents recordkeeping and reporting
approximately $32,800. This burden for this proposed action to EPA and requirements of this proposed action.
and cost would only be applicable once. OMB. See ADDRESSES section at the We continue to be interested in the
After that, the total nationwide beginning of this notice for where to potential impacts of this proposed
recordkeeping and reporting burden and submit comments to EPA. Send action on small entities and welcome
costs would be $0 (above and beyond comments to OMB at the Office of comments on issues related to such
current burden and costs). Information and Regulatory Affairs, impacts.
The annual average burden associated Office of Management and Budget, 725
with the emission guidelines over the D. Unfunded Mandates Reform Act
17th Street, NW., Washington, DC
first 3 years following promulgation of 20503, Attention: Desk Office for EPA. Title II of the Unfunded Mandates
this proposed action is estimated to be Since OMB is required to make a Reform Act (UMRA) of 1995, Public
49,878 hours at a total annual labor cost decision concerning the ICR between 30 Law 104–4, establishes requirements for
of $2,433,045. The total annualized and 60 days after February 6, 2007, a Federal agencies to assess the effects of
capital/startup costs and operation and comment to OMB is best assured of their regulatory actions on State, local,
maintenance (O&M) costs associated having its full effect if OMB receives it and Tribal governments and the private
with the monitoring requirements, EPA by March 8, 2007. The final action will sector. Under section 202 of the UMRA,
Method 22 testing, storage of data and respond to any OMB or public EPA generally must prepare a written
reports, and photocopying and postage comments on the information collection statement, including a cost-benefit
over the 3-year period of the ICR are requirements contained in this proposal. analysis, for proposed and final rules
estimated at $407,953 and $333,258 per with ‘‘Federal mandates’’ that may
year, respectively. (The annual C. Regulatory Flexibility Act result in expenditures by State, local,
inspection costs are included under the The Regulatory Flexibility Act (RFA) and Tribal governments, in the
recordkeeping and reporting labor generally requires an agency to prepare aggregate, or by the private sector, of
costs.) The annual average burden a regulatory flexibility analysis of any $100 million or more in any 1 year.
associated with the NSPS over the first rule subject to notice and comment Before promulgating an EPA rule for
3 years following promulgation of this rulemaking requirements under the which a written statement is needed,
proposed action is estimated to be 2,004 Administrative Procedures Act or any section 205 of the UMRA generally
hours at a total annual labor cost of other statute unless the agency certifies requires EPA to identify and consider a
$91,011. The total annualized capital/ that the proposed action will not have reasonable number of regulatory
startup costs are estimated at $13,046, a significant economic impact on a alternatives and adopt the least costly,
with total operation and maintenance substantial number of small entities. most cost-effective, or least burdensome
costs of $36,310 per year. Small entities include small businesses, alternative that achieves the objectives
Burden means the total time, effort, or small government organizations, and of the proposed rule. The provisions of
financial resources expended by persons small government jurisdictions. section 205 do not apply when they are
to generate, maintain, retain, or disclose For purposes of assessing the impacts inconsistent with applicable law.
or provide information to or for a of this proposed action on small Moreover, section 205 allows EPA to
adopt an alternative other than the least State law. Thus, Executive Order 13132 H. Executive Order 13211: Actions That
costly, most cost-effective, or least does not apply to this proposed action. Significantly Affect Energy Supply,
burdensome alternative if EPA In the spirit of Executive Order 13132, Distribution or Use
publishes with the final rule an and consistent with EPA policy to This proposed action is not a
explanation why that alternative was promote communications between EPA ‘‘significant energy action’’ as defined in
not adopted. Executive Order 13211, ‘‘Actions
Before EPA establishes any regulatory and State and local governments, EPA
specifically solicits comment on this Concerning Regulations That
requirements that may significantly or
proposed action from State and local Significantly Affect Energy Supply,
uniquely affect small governments,
officials. Distribution, or Use’’ (66 FR 28355; May
including Tribal governments, EPA
22, 2001) because it is not likely to have
must develop a small government F. Executive Order 13175: Consultation
agency plan under section 203 of the a significant adverse effect on the
and Coordination With Indian Tribal supply, distribution, or use of energy.
UMRA. The plan must provide for Governments
notifying potentially affected small EPA estimates that the requirements in
governments, enabling officials of this proposed action would cause some
Executive Order 13175, (65 FR 67249;
affected small governments to have HMIWI to increase the horsepower of
November 9, 2000), requires EPA to
meaningful and timely input in the their wet scrubbers, resulting in
develop an accountable process to approximately 3,600 megawatt-hours
development of EPA’s regulatory ensure ‘‘meaningful and timely input by
proposals with significant Federal per year of additional electricity being
Tribal officials in the development of used.
intergovernmental mandates, and regulatory policies that have Tribal
informing, educating, and advising Given the negligible change in energy
implications.’’ consumption resulting from this
small governments on compliance with
the regulatory requirements. This proposed action does not have proposed action, EPA does not expect
EPA has determined that this Tribal implications, as specified in any price increase for any energy type.
proposed action does not contain a Executive Order 13175. It will not have The cost of energy distribution should
Federal mandate that may result in substantial direct effects on Tribal not be affected by this proposed action
expenditures of $100 million or more governments, on the relationship at all since the action would not affect
for State, local, and Tribal governments, between the Federal government and energy distribution facilities. We also
in the aggregate, or the private sector in Indian tribes, or on the distribution of expect that there would be no impact on
any 1 year. Thus, this proposed action power and responsibilities between the the import of foreign energy supplies,
is not subject to the requirements of Federal government and Indian tribes, and no other adverse outcomes are
section 202 and 205 of the UMRA. In as specified in Executive Order 13175. expected to occur with regard to energy
addition, EPA has determined that this EPA is not aware of any HMIWI owned supplies.
proposed action contains no regulatory or operated by Indian Tribal I. National Technology Transfer
requirements that might significantly or governments. Thus, Executive Order Advancement Act
uniquely affect small governments. 13175 does not apply to this proposed
Therefore, this proposed action is not Section 12(d) of the National
action. Technology Transfer and Advancement
subject to the requirements of section
203 of the UMRA. G. Executive Order 13045: Protection of Act (NTTAA) of 1995 (Pub. L. 104–113,
Children From Environmental Health Section 12(d), 15 U.S.C. 272 note)
E. Executive Order 13132: Federalism directs EPA to use voluntary consensus
and Safety Risks
Executive Order 13132 (64 FR 43255; standards (VCS) in its regulatory
August 10, 1999), requires EPA to Executive Order 13045 (62 FR 19885; activities, unless to do so would be
develop an accountable process to April 23, 1997), applies to any rule that: inconsistent with applicable law or
ensure ‘‘meaningful and timely input by (1) Is determined to be ‘‘economically otherwise impractical. The VCS are
State and local officials in the significant’’ as defined under Executive technical standards (e.g., materials
development of regulatory policies that Order 12866, and (2) concerns an specifications, test methods, sampling
have federalism implications.’’ ‘‘Policies environmental health or safety risk that procedures, and business practices) that
that have federalism implications’’ are EPA has reason to believe may have a are developed or adopted by VCS
defined in the Executive Order to disproportionate effect on children. If bodies. The NTTAA directs EPA to
include regulations that have the regulatory action meets both criteria, provide Congress, through OMB,
‘‘substantial direct effects on the States, EPA must evaluate the environmental explanations when the Agency does not
on the relationship between the national health or safety effects of the planned use available and applicable VCS.
government and the States, or on the rule on children, and explain why the This proposed action involves
distribution of power and planned regulation is preferable to other technical standards. EPA cites the
responsibilities among the various potentially effective and reasonably following standards: EPA Methods 1, 3,
levels of government.’’ This proposed feasible alternatives EPA considered. 3A, 3B, 5, 9, 10, 10B, 22, 23, 26, 26A,
action does not have federalism and 29 in 40 CFR part 60, appendix A.
implications. It will not have substantial EPA interprets Executive Order 13045 Consistent with the NTTAA, EPA
direct effects on the States, on the as applying only to those regulatory conducted searches to identify
relationship between the national actions that are based on health or safety voluntary consensus standards in
government and the States, or on the risks, such that the analysis required addition to these EPA methods. No
distribution of power and under section 5–501 of the Executive applicable voluntary consensus
responsibilities among the various Order has the potential to influence the standards were identified for EPA
levels of government, as specified in regulation. This proposed action is not Methods 9 and 22. The search and
Executive Order 13132. This proposed subject to Executive Order 13045 review results are in the docket for this
action will not impose substantial direct because it is based on technology proposed action.
compliance costs on State or local performance and not on health and Two voluntary consensus standards
governments, and will not preempt safety risks. were identified as acceptable
alternatives to EPA test methods for the Subpart Ce—[Amended] (1) At a minimum, an inspection shall
purposes of this proposed action. The include the following:
voluntary consensus standard ASME 2. Section 60.32e is amended by (i) Inspect air pollution control
PTC 19–10–1981–Part 10, ‘‘Flue and revising paragraphs (a) and (i) to read as device(s) for proper operation, if
Exhaust Gas Analyses,’’ is cited in the follows: applicable;
proposed action for its manual method (ii) Ensure proper calibration of
for measuring the oxygen content of § 60.32e Designated facilities. thermocouples, sorbent feed systems,
exhaust gas. This part of ASME PTC 19– (a) Except as provided in paragraphs and any other monitoring equipment;
10–1981–Part 10 is an acceptable (b) through (h) of this section, the and
alternative to EPA Method 3B. designated facility to which the (iii) Generally observe that the
guidelines apply is each individual equipment is maintained in good
The voluntary consensus standard HMIWI for which construction was operating condition.
ASTM D6784–02, ‘‘Standard Test commenced on or before June 20, 1996 (2) Within 10 operating days
Method for Elemental, Oxidized, and each individual HMIWI currently following an air pollution control device
Particle-Bound and Total Mercury Gas subject to subpart Ec as promulgated in inspection, all necessary repairs shall be
Generated from Coal-Fired Stationary 1997 (for which construction was completed unless the owner or operator
Sources (Ontario Hydro Method),’’ is an commenced after June 20, 1996 but no obtains written approval from the State
acceptable alternative to EPA Method 29 later than February 6, 2007 or for which agency establishing a date whereby all
(portion for mercury only) as a method modification commenced after March necessary repairs of the designated
for measuring Hg. 16, 1998 but no later than 6 months after facility shall be completed.
The search for emissions the date of promulgation of this (d) For approval, a State plan shall
measurement procedures identified 16 subpart). require that each HMIWI subject to the
other voluntary consensus standards. * * * * * emission limits under § 60.33e(a)
EPA determined that these 16 standards (i) Beginning 3 years after the date of undergo an air pollution control device
identified for measuring emissions of promulgation of this subpart, or on the inspection annually (no more than 12
the pollutants subject to emission effective date of an EPA approved months following the previous annual
standards in this proposed action were operating permit program under Clean air pollution control device inspection),
impractical alternatives to EPA test Air Act title V and the implementing as outlined in paragraphs (c)(1) and (2)
methods for the purposes of this action. regulations under 40 CFR part 70 in the of this section.
Therefore, EPA does not intend to adopt State in which the unit is located, 5. Section 60.37e is amended by
these standards for this purpose. A whichever date is later, designated revising paragraphs (a) and (b)(1) and
document that discusses the facilities subject to this subpart shall adding paragraph (e) to read as follows:
determinations for these 16 methods is operate pursuant to a permit issued § 60.37e Compliance, performance testing,
located in the docket to this proposed under the EPA-approved operating and monitoring guidelines.
action. permit program. (a) Except as provided in paragraph
Section 60.56c of subpart Ec of 40 3. Section 60.33e is amended by
(b) of this section, for approval, a State
CFR part 60 and § 60.37e of subpart Ce revising paragraph (b) to read as follows:
plan shall include the requirements for
of 40 CFR part 60 list the testing § 60.33e Emission guidelines. compliance and performance testing
methods included in the proposed listed in § 60.56c of subpart Ec of this
* * * * *
action. Under 40 CFR 60.8(b) and part, excluding the fugitive emissions
(b) For approval, a State plan shall
60.13(i) of subpart A (General annual testing requirement under
include the requirements for emission
Provisions), a source may apply to EPA § 60.56c(c)(3), the CO CEMS
limits at least as protective as those
for permission to use alternative test requirements under § 60.56c(c)(5), and
requirements listed in Table 2 of this
methods or alternative monitoring the bag leak detection system
subpart for any small HMIWI
requirements in place of any required requirements under § 60.57c(g). Sources
constructed on or before June 20, 1996
testing methods, performance may, however, elect to use CO CEMS as
which is located more than 50 miles
specifications, or procedures. specified under § 60.56c(c)(5) or bag
from the boundary of the nearest
List of Subjects in 40 CFR Part 60 Standard Metropolitan Statistical Area leak detection systems as specified
(defined in § 60.31e) and which burns under § 60.57c(g).
Environmental protection, less than 2,000 pounds per week of (b) * * *
Administrative practice and procedure, hospital waste and medical/infectious (1) Conduct the performance testing
Air pollution control, Intergovernmental waste. The 2,000 lb/week limitation requirements in § 60.56c(a), (b)(1)
relations, Reporting and recordkeeping does not apply during performance through (b)(9), (b)(11) (Hg only), (b)(12),
requirements. tests. and (c)(1) of subpart Ec of this part. The
Dated: January 26, 2007.
2,000 lb/week limitation under
* * * * *
4. Section 60.36e is amended by § 60.33e(b) does not apply during
Stephen L. Johnson,
adding paragraphs (c) and (d) to read as performance tests.
Administrator.
follows: * * * * *
For the reasons stated in the (e) The owner or operator of a
preamble, title 40, chapter I, part 60 of § 60.36e Inspection guidelines. designated facility may use the results
the Code of Federal Regulations is * * * * * of previous emissions tests to
proposed to be amended as follows: (c) For approval, a State plan shall demonstrate compliance with the
require that each HMIWI subject to the emission limits, provided that the
PART 60—[AMENDED] emission limits under § 60.33e(a) conditions in paragraphs (e)(1) through
undergo an initial air pollution control (e)(3) of this section are met:
1. The authority citation for part 60 device inspection that is at least as (1) The previous emissions tests must
continues to read as follows: protective as the following within 1 year have been conducted using the
Authority: 42 U.S.C. 7401, et seq. following approval of the State plan: applicable procedures and test methods
listed in § 60.56c(b)(1) through (b)(9), (1) Maintain records of the annual air (c) State plans that specify measurable
(b)(11) (Hg only), and (b)(12). Previous pollution control device inspections, and enforceable incremental steps of
emissions test results obtained using any required maintenance, and any progress towards compliance for
EPA-accepted voluntary consensus repairs not completed within 10 days of designated facilities planning to install
standards are also acceptable. an inspection or the timeframe the necessary air pollution control
(2) The HMIWI at the affected facility established by the State regulatory equipment may allow compliance on or
shall be operated in a manner (e.g., with agency; and before the date 3 years after EPA
charge rate, secondary chamber (2) Submit an annual report approval of the State plan (but not later
temperature, etc.) that would be containing information recorded under than 5 years after the date of
expected to result in the same or lower paragraph (c)(1) of this section no later promulgation of this subpart). Suggested
emissions than observed during the than 60 days following the year in measurable and enforceable activities to
previous emissions test(s), and the which data were collected. Subsequent be included in State plans are:
HMIWI may not have been modified reports shall be sent no later than 12 * * * * *
such that emissions would be expected calendar months following the previous
(d) * * *
to exceed (notwithstanding normal test- report (once the unit is subject to
to-test variability) the results from permitting requirements under title V of (3) If an extension is granted, require
previous emissions test(s). the Act, the owner or operator shall compliance with the emission
(3) The previous emissions test(s) submit these reports semiannually). The guidelines on or before the date 3 years
must have been conducted in 1997 or report shall be signed by the facilities after EPA approval of the State plan (but
later. manager. not later than 5 years after the date of
6. Section 60.38e is amended by 7. Section 60.39e is amended as promulgation of this subpart).
revising paragraph (a) and adding follows: * * * * *
paragraph (c) to read as follows: a. By revising paragraph (a); (f) The Administrator shall develop,
b. By revising paragraph (c)
§ 60.38e Reporting and recordkeeping implement, and enforce a plan for
introductory text;
guidelines. c. By revising paragraph (d)(3); and existing HMIWI located in any State that
(a) For approval, a State plan shall d. By revising paragraph (f). has not submitted an approvable plan
include the reporting and recordkeeping within 2 years after the date of
requirements listed in § 60.58c(b), (c), § 60.39e Compliance times. promulgation of this subpart. Such
(d), (e), and (f) of subpart Ec of this part, (a) Not later than 1 year after the date plans shall ensure that each designated
excluding § 60.58c(b)(7) (siting). of promulgation of this subpart, each facility is in compliance with the
* * * * * State in which a designated facility is provisions of this subpart no later than
(c) For approval, a State plan shall operating shall submit to the 5 years after the date of promulgation of
require the owner or operator of each Administrator a plan to implement and this subpart.
HMIWI subject to the emission limits enforce the emission guidelines. 8. Table 1 to subpart Ce is revised to
under § 60.33e(a) to: * * * * * read as follows:
TABLE 1 TO SUBPART CE.—EMISSION LIMITS FOR SMALL, MEDIUM, AND LARGE HMIWI
Emission limits
Units HMIWI size
Pollutant (7 percent oxygen, dry basis)
Small Medium Large
Particulate matter ....... Milligrams per dry standard cubic meter (mg/ 69 (0.030) .................. 69 (0.030) .................. 34 (0.015).
dscm) (grains per dry standard cubic foot
(gr/dscf)).
Carbon monoxide ...... Parts per million by volume (ppmv) ............... 25 ............................... 25 ............................... 25.
Dioxins/furans ............ Nanograms per dry standard cubic meter 115 (50) or 2.0 (0.87) 115 (50) or 2.0 (0.87) 115 (50) or 2.0 (0.87).
total dioxins/furans (ng/dscm) (grains per
billion dry standard cubic feet (gr/109
dscf)) or ng/dscm TEQ (gr/109 dscf).
Hydrogen chloride ...... ppmv or percent reduction ............................. 51 or 94% .................. 51 or 94% .................. 51 or 94%
Sulfur dioxide ............. Ppmv ................................................................ 28 ............................... 28 ............................... 28.
Nitrogen oxides .......... Ppmv ................................................................ 212 ............................. 212 ............................. 212.
Lead ........................... mg/dscm (grains per thousand dry standard 0.64 (0.28) or 71% ..... 0.64 (0.28) or 71% ..... 0.64 (0.28) or 71%.
cubic feet (gr/103 dscf)) or percent reduc-
tion.
Cadmium .................... mg/dscm (gr/103 dscf) or percent reduction .. 0.060 (0.026) or 74% 0.060 (0.026) or 74% 0.060 (0.026) or 74%.
Mercury ...................... mg/dscm (gr/103 dscf) or percent reduction .. 0.33 (0.14) or 96% ..... 0.33 (0.14) or 96% ..... 0.33 (0.14) or 96%.
9. Table 2 of subpart Ce is revised to

read as follows:
TABLE 2 TO SUBPART CE.—EMISSION LIMITS FOR SMALL HMIWI WHICH MEET THE CRITERIA UNDER § 60.33E(B)
Units
Pollutant HMIWI emission limits
(7 percent oxygen, dry basis)
Particulate matter ................................ mg/dscm (gr/dscf) ............................................................................................. 69 (0.030).

Carbon monoxide ................................ Ppmv .................................................................................................................. 25.
TABLE 2 TO SUBPART CE.—EMISSION LIMITS FOR SMALL HMIWI WHICH MEET THE CRITERIA UNDER § 60.33E(B)—
Continued
Units
Pollutant HMIWI emission limits
(7 percent oxygen, dry basis)
Dioxins/furans ...................................... ng/dscm total dioxins/furans (gr/109 dscf) or ng/dscm TEQ (gr/109 dscf) ....... 800 (350) or 15 (6.6).
Hydrogen chloride ............................... ppmv or percent reduction ............................................................................... 398.
Sulfur dioxide ...................................... Ppmv .................................................................................................................. 28.
Nitrogen oxides ................................... Ppmv .................................................................................................................. 212.
Lead .................................................... mg/dscm (gr/103 dscf) or percent reduction .................................................... 0.60 (0.26).
Cadmium ............................................. mg/dscm (gr/103 dscf) or percent reduction .................................................... 0.050 (0.022).
Mercury ............................................... mg/dscm (gr/103 dscf) or percent reduction .................................................... 0.25 (0.11).
Subpart Ec—[Amended] is not limited to, an instrument that j. By adding paragraph (k).
operates on triboelectric, light-
10. Section 60.50c is amended by scattering, light-transmittance, or other § 60.56c Compliance and performance
revising paragraphs (a), (k) and (l) to testing.
effects to monitor relative PM loadings.
read as follows: * * * * *
* * * * *
(b) Except as provided in paragraph
§ 60.50c Applicability and delegation of Minimum secondary chamber
(k) of this section, the owner or operator
authority. temperature means 90 percent of the
of an affected facility shall conduct an
(a) Except as provided in paragraphs highest 3-hour average secondary
initial performance test as required
(b) through (h) of this section, the chamber temperature (taken, at a
under § 60.8 to determine compliance
affected facility to which this subpart minimum, once every minute) measured
with the emission limits using the
applies is each individual hospital/ during the most recent performance test
procedures and test methods listed in
medical/infectious waste incinerator demonstrating compliance with the PM,
paragraphs (b)(1) through (b)(12) of this
(HMIWI): CO, and dioxin/furan emission limits.
section. The use of the bypass stack
(1) For which construction is * * * * * during a performance test shall
commenced after June 20, 1996 but no 12. Section 60.52c is amended by invalidate the performance test.
later than February 6, 2007; revising paragraph (c) to read as follows:
(2) For which modification is * * * * *
commenced after March 16, 1998 but no § 60.52c Emission limits. (4) EPA Reference Method 3, 3A, or
later than 6 months after the date of * * * * * 3B of appendix A of this part shall be
promulgation of this subpart; (c) On and after the date on which the used for gas composition analysis,
(3) For which construction is initial performance test is completed or including measurement of oxygen
commenced after February 6, 2007; or is required to be completed under concentration. EPA Reference Method 3,
(4) For which modification is § 60.8, whichever date comes first, no 3A, or 3B of appendix A of this part
commenced after 6 months after the owner or operator of an affected facility shall be used simultaneously with each
date of promulgation of this subpart. shall cause to be discharged into the of the other EPA reference methods. As
atmosphere visible emissions of an alternative, ASME PTC–19–10–1981-
* * * * *
(k) The requirements of this subpart combustion ash from an ash conveying Part 10 may be used.
shall become effective 6 months after system (including conveyor transfer * * * * *
the date of promulgation of this subpart. points) in excess of 5 percent of the (6) EPA Reference Method 5 or 29 of
(l) Beginning 3 years after the date of observation period (i.e., 9 minutes per 3- appendix A of this part shall be used to
promulgation of this subpart, or on the hour period), as determined by EPA measure the particulate matter
effective date of an EPA-approved Reference Method 22 of appendix A of emissions. As an alternative, PM CEMS
operating permit program under Clean this part, except as provided in may be used as specified in paragraph
Air Act title V and the implementing paragraphs (d) and (e) of this section. (c)(4) of this section.
regulations under 40 CFR part 70 in the * * * * * (7) EPA Reference Method 9 of
State in which the unit is located, 13. Section 60.56c is amended as appendix A of this part shall be used to
whichever date is later, affected follows: measure stack opacity. As an
facilities subject to this subpart shall a. By revising paragraph (b) alternative, demonstration of
operate pursuant to a permit issued introductory text; compliance with the PM standards
under the EPA approved State operating b. By revising paragraphs (b)(4) and using bag leak detection systems as
permit program. (b)(6) through (b)(8), (b)(9) introductory specified in § 60.57c(g) or PM CEMS as
11. Section 60.51c is amended by text, and (b)(10); specified in paragraph (c)(4) of this
adding a definition for ‘‘Bag leak c. By revising paragraph (b)(11); section is considered demonstrative of
detection system’’ in alphabetical order d. By revising paragraphs (c)(2) compliance with the opacity
and revising the definition for through (4); requirements.
‘‘Minimum secondary chamber e. By adding paragraphs (c)(5), and (8) For affected facilities under
temperature’’ to read as follows: (c)(6); § 60.50c(a)(1) and (a)(2), EPA Reference
f. By revising paragraph (d) Method 10 or 10B of appendix A of this
§ 60.51c Definitions. introductory text; part shall be used to measure the CO

Bag leak detection system means an g. By adding paragraphs (e)(6) and (7); emissions. As an alternative, CO CEMS
instrument that is capable of monitoring h. By adding paragraphs (f)(7) through may be used as specified in paragraph
PM loadings in the exhaust of a fabric (9); (c)(4) of this section.
filter in order to detect bag failures. A i. By adding paragraphs (g)(6) through (9) EPA Reference Method 23 of
bag leak detection system includes, but (9); and appendix A of this part shall be used to
measure total dioxin/furan emissions. and shall not be recovered or cleaned forego a performance test for that
As an alternative, an owner or operator until the end of testing. pollutant for an additional 2 years. If
may elect to sample dioxins/furans by (ii) For the duration of sampling, a any performance test indicates
installing, calibrating, maintaining, and temperature around the probe and filter noncompliance with the respective
operating a continuous automated (and cyclone, if used) between 120 °C emission limit, a performance test for
sampling system for monitoring dioxin/ (248 °F) and 134 °C (273 °F) shall be that pollutant shall be conducted
furan emissions as specified in maintained. annually until all annual performance
paragraph (c)(6) of this section. For (iii) If water droplets are present in tests over a 3-year period indicate
Method 23 sampling, the minimum the sample gas stream, the requirements compliance with the emission limit. The
sample time shall be 4 hours per test specified in paragraphs (b)(10)(iii)(A) use of the bypass stack during a
run. If the affected facility has selected and (B) of this section shall be met. performance test shall invalidate the
the toxic equivalency standards for (A) The cyclone described in section performance test.
dioxins/furans, under § 60.52c, the 6.1.4 of EPA Reference Method 26A of (3) For large HMIWI under
following procedures shall be used to appendix A of this part shall be used. § 60.50c(a)(1) and (a)(2) and for all
determine compliance: (B) The post-test moisture removal HMIWI under § 60.50c(a)(3) and (a)(4),
procedure described in section 8.1.6 of determine compliance with the visible
* * * * *
EPA Reference Method 26A of appendix emission limits for fugitive emissions
(10) EPA Reference Method 26 or 26A A of this part shall be used.
of appendix A of this part shall be used from flyash/bottom ash storage and
(11) EPA Reference Method 29 of handling by conducting a performance
to measure HCl emissions, with the appendix A of this part shall be used to
additional requirements for Method 26A test using EPA Reference Method 22 on
measure Pb, Cd, and Hg emissions. As an annual basis (no more than 12
specified in paragraphs (b)(10)(i) an alternative, Hg emissions may be
through (iii) of this section. As an months following the previous
measured using ASTM D6784–02. As an performance test).
alternative, HCl CEMS may be used as alternative for Pb, Cd, and Hg, multi-
specified in paragraph (c)(4) of this (4) Facilities using optional CEMS to
metals CEMS, or Hg CEMS, may be used
section. If the affected facility has demonstrate compliance with the PM,
as specified in paragraph (c)(4) of this
selected the percentage reduction CO, HCl, Pb, Cd, and/or Hg emission
section. If the affected facility has
standards for HCl under § 60.52c, the limits under § 60.52c shall:
selected the percentage reduction
percentage reduction in HCl emission (i) Determine compliance with the
standards for metals under § 60.52c, the
(%RHCl) is computed using the appropriate emission limit(s) using a 24-
percentage reduction in emissions
following formula: hour block average, calculated as
(%Rmetal) is computed using the
specified in section 12.4.1 of EPA
(%RHCl) = (Ei¥Eo)/Ei × 100 following formula:
Reference Method 19 of appendix A of
Where: (%Rmetal) = (Ei¥Eo)/Ei × 100 this part.
%RHCl=percentage reduction of HCl Where: (ii) Operate all CEMS in accordance
emissions achieved; %Rmetal=percentage reduction of metal with the applicable procedures under
Ei=HCl emission concentration measured at emission (Pb, Cd, or Hg) achieved; appendices B and F of this part. For
the control device inlet, corrected to 7 Ei=metal emission concentration (Pb, Cd, or those CEMS for which performance
percent oxygen (dry basis); and Hg) measured at the control device inlet,
Eo=HCl emission concentration measured at
specifications have not yet been
corrected to 7 percent oxygen (dry basis);
the control device outlet, corrected to 7 and
promulgated (HCl, multi-metals), this
percent oxygen (dry basis). Eo=metal emission concentration (Pb, Cd, or option takes effect on the date a final
Hg) measured at the control device performance specification is published
(i) The probe and filter shall be outlet, corrected to 7 percent oxygen (dry in the Federal Register or the date of
conditioned prior to sampling using the basis). approval of a site-specific monitoring
procedure described in paragraphs plan.
* * * * *
(b)(10)(i)(A) through (C) of this section. (c) * * * (iii) Be allowed to substitute use of an
(A) Assemble the sampling train(s) (2) Except as provided in paragraphs HCl CEMS for the HCl annual
and conduct a conditioning run by (c)(4) and (c)(5) of this section, performance test, minimum HCl sorbent
collecting between 14 liters per minute determine compliance with the PM, CO, flow rate, and minimum scrubber liquor
(L/min)_(0.5 cubic feet per minute (ft 3/ and HCl emission limits by conducting pH to demonstrate compliance with the
min)) and 30 L/min (1.0 ft 3/min) of gas an annual performance test (no more HCl emission limit.
over a 1-hour period. Follow the than 12 months following the previous (iv) Be allowed to substitute use of a
sampling procedures outlined in section performance test) using the applicable PM CEMS for the PM annual
8.1.5 of Method 26A of appendix A of procedures and test methods listed in performance test and minimum pressure
this part. For the conditioning run, paragraph (b) of this section. If all three drop across the wet scrubber, if
water may be used as the impinger performance tests over a 3-year period applicable, to demonstrate compliance
solution. indicate compliance with the emission with the PM emission limit.
(B) Remove the impingers from the limit for a pollutant (PM, CO, or HCl), (v) Be allowed to substitute use of a
sampling train and replace with a fresh the owner or operator may forego a CO CEMS for the CO annual
impinger train for the sampling run, performance test for that pollutant for performance test and minimum
leaving the probe and filter (and the subsequent 2 years. At a minimum, secondary chamber temperature to
cyclone, if used) in position. Do not a performance test for PM, CO, and HCl demonstrate compliance with the CO
recover the filter or rinse the probe shall be conducted every third year (no emission limit.
before the first run. Thoroughly rinse more than 36 months following the (5) For affected facilities under
the impingers used in the previous performance test). If a § 60.50c(a)(3) and (a)(4), determine
preconditioning run with deionized performance test conducted every third compliance with the CO emission limit
water and discard these rinses. year indicates compliance with the using a CO CEMS according to
(C) The probe and filter assembly emission limit for a pollutant (PM, CO, paragraphs (c)(5)(i) and (c)(5)(ii) of this
shall be conditioned by the stack gas or HCl), the owner or operator may section:
(i) Determine compliance with the CO corrective action is required, each alarm (9) For all HMIWI, operation of the
emission limit using a 24-hour block is counted as a minimum of 1 hour. If affected facility above the CDD/CDF
average, calculated as specified in it takes longer than 1 hour to initiate emission limit as measured by the
section 12.4.1 of EPA Reference Method corrective action, the alarm time is continuous automated sampling system
19 of appendix A of this part. counted as the actual amount of time specified in paragraph (c)(6) of this
(ii) Operate the CO CEMS in taken to initiate corrective action. If the section shall constitute a violation of the
accordance with the applicable bag leak detection system is used to CDD/CDF emission limit.
procedures under appendices B and F of demonstrate compliance with the * * * * *
this part. opacity limit, this would also constitute (k) The owner or operator of an
(iii) Use of a CO CEMS may be a violation of the opacity emission limit. affected facility may use the results of
substituted for the CO annual (f) * * * previous emissions tests to demonstrate
performance test and minimum (7) For HMIWI under § 60.50c(a)(3) compliance with the emission limits,
secondary chamber temperature to and (a)(4), operation of the affected provided that the conditions in
demonstrate compliance with the CO facility above the CO emission limit as paragraphs (k)(1) through (k)(3) of this
emission limit. measured by the CO CEMS shall section are met:
(6) Facilities using a continuous constitute a violation of the CO (1) The previous emissions tests shall
automated sampling system to emission limit. have been conducted using the
demonstrate compliance with the (8) For all HMIWI, operation of the applicable procedures and test methods
dioxin/furan emission limits under affected facility above the PM, CO, HCl, listed in paragraph (b) of this section.
§ 60.52c shall record the output of the Pb, Cd, and/or Hg emission limit as Previous emissions test results obtained
system and analyze the sample using measured by the CEMS specified in using EPA-accepted voluntary
EPA Reference Method 23 of appendix paragraph (c)(4) of this section shall consensus standards are also acceptable.
A of this part. This option to use a constitute a violation of the applicable (2) The HMIWI at the affected facility
continuous automated sampling system emission limit. shall be operated in a manner (e.g., with
takes effect on the date a final (9) For all HMIWI, operation of the charge rate, secondary chamber
performance specification applicable to affected facility above the CDD/CDF temperature, etc.) that would be
dioxin/furan from monitors is published emission limit as measured by the expected to result in the same or lower
in the Federal Register or the date of continuous automated sampling system emissions than observed during the
approval of a site-specific monitoring specified in paragraph (c)(6) of this previous emissions test(s) and the
plan. The owner or operator of an section shall constitute a violation of the HMIWI may not have been modified
affected facility who elects to CDD/CDF emission limit. such that emissions would be expected
continuously sample dioxin/furan (g) * * * to exceed (notwithstanding normal test-
emissions instead of sampling and (6) For HMIWI under § 60.50c(a)(3) to-test variability) the results from
testing using EPA Reference Method 23 and (a)(4), operation of the affected previous emissions test(s).
shall install, calibrate, maintain, and facility above the CO emission limit as (3) The previous emissions test(s)
operate a continuous automated measured by the CO CEMS shall shall have been conducted in 1997 or
sampling system and shall comply with constitute a violation of the CO later.
the requirements specified in emission limit. 14. Section 60.57c is amended as
§ 60.58b(p) and (q) of subpart Eb of this (7) For HMIWI under § 60.50c(a)(3) follows:
part. and (a)(4), failure to initiate corrective a. By revising paragraph (a);
(d) Except as provided in paragraphs action within 1 hour of a bag leak b. By adding paragraph (e);
(c)(4), (c)(5), and (c)(6) of this section, detection system alarm; or failure to c. By adding paragraph (f); and
the owner or operator of an affected operate and maintain the fabric filter d. By adding paragraph (g).
facility equipped with a dry scrubber such that the alarm is not engaged for
followed by a fabric filter, a wet more than 5 percent of the total § 60.57c Monitoring requirements
scrubber, or a dry scrubber followed by operating time in a 6-month block (a) Except as provided in
a fabric filter and wet scrubber shall: reporting period shall constitute a § 60.56c(c)(4) through (c)(6), the owner
* * * * * violation of the PM emission limit. If or operator of an affected facility shall
(e) * * * inspection of the fabric filter install, calibrate (to manufacturers’
(6) For HMIWI under § 60.50c(a)(3) demonstrates that no corrective action is specifications), maintain, and operate
and (a)(4), operation of the affected required, no alarm time is counted. If devices (or establish methods) for
facility above the CO emission limit as corrective action is required, each alarm monitoring the applicable maximum
measured by the CO CEMS shall is counted as a minimum of 1 hour. If and minimum operating parameters
constitute a violation of the CO it takes longer than 1 hour to initiate listed in Table 3 to this subpart (unless
emission limit. corrective action, the alarm time is optional CEMS are used as a substitute
(7) For HMIWI under § 60.50c(a)(3) counted as the actual amount of time for certain parameters as specified) such
and (a)(4), failure to initiate corrective taken to initiate corrective action. If the that these devices (or methods) measure
action within 1 hour of a bag leak bag leak detection system is used to and record values for these operating
detection system alarm; or failure to demonstrate compliance with the parameters at the frequencies indicated
operate and maintain the fabric filter opacity limit, this would also constitute in Table 3 at all times except during
such that the alarm is not engaged for a violation of the opacity emission limit. periods of startup and shutdown.
more than 5 percent of the total (8) For all HMIWI, operation of the * * * * *
operating time in a 6-month block affected facility above the PM, CO, HCl, (e) The owner or operator of an
reporting period shall constitute a Pb, Cd, and/or Hg emission limit as affected facility shall ensure that each
violation of the PM emission limit. If measured by the CEMS specified in HMIWI subject to the emission limits in
inspection of the fabric filter paragraph (c)(4) of this section shall § 60.52c undergoes an initial air
demonstrates that no corrective action is constitute a violation of the applicable pollution control device inspection that
required, no alarm time is counted. If emission limit. is at least as protective as the following:
(1) At a minimum, an inspection shall be capable of detecting PM emissions at amount of time taken to initiate
include the following: concentrations of 10 milligrams per corrective action.
(i) Inspect air pollution control actual cubic meter (0.0044 grains per 15. Section 60.58c is amended as
device(s) for proper operation, if actual cubic foot) or less. follows:
applicable; (3) The bag leak detection system a. By adding paragraphs (b)(2)(xvi)
(ii) Ensure proper calibration of sensor shall provide an output of through (xviii);
thermocouples, sorbent feed systems, relative PM loadings. b. By revising paragraph (b)(6);
and any other monitoring equipment; (4) The bag leak detection system c. By revising paragraph (c)
and shall be equipped with a device to introductory text;
(iii) Generally observe that the continuously record the output signal d. By revising paragraph (c)(2);
equipment is maintained in good from the sensor. e. By adding paragraph (c)(4);
operating condition. (5) The bag leak detection system f. By revising paragraph (d)
(2) Within 10 operating days introductory text;
shall be equipped with an audible alarm
following an air pollution control device g. By adding paragraphs (d)(9)
system that will sound automatically
inspection, all necessary repairs shall be through (11); and
when an increase in relative PM
completed unless the owner or operator h. By adding paragraph (g).
emissions over a preset level is detected.
obtains written approval from the
The alarm shall be located where it is § 60.58c Reporting and recordkeeping
Administrator establishing a date
easily heard by plant operating requirements.
whereby all necessary repairs of the
personnel.
designated facility shall be completed. * * * * *
(f) The owner or operator of an (6) For positive pressure fabric filter (b) * * *
affected facility shall ensure that each systems, a bag leak detector shall be (2) * * *
HMIWI subject to the emission limits installed in each baghouse compartment (xvi) Records of the annual air
under § 60.52c undergoes an air or cell. pollution control device inspections,
pollution control device inspection (7) For negative pressure or induced any required maintenance, and any
annually (no more than 12 months air fabric filters, the bag leak detector repairs not completed within 10 days of
following the previous annual air shall be installed downstream of the an inspection or the timeframe
pollution control device inspection), as fabric filter. established by the Administrator.
outlined in paragraphs (e)(1) and (e)(2) (8) Where multiple detectors are (xvii) For affected facilities using a
of this section. required, the system’s instrumentation bag leak detection system, records of
(g) For affected facilities under and alarm may be shared among each alarm, the time of the alarm, the
§ 60.50c(a)(3) and (a)(4) using an air detectors. time corrective action was initiated and
pollution control device that includes a (9) The baseline output shall be completed, and a brief description of the
fabric filter and not using PM CEMS, established by adjusting the range and cause of the alarm and the corrective
determine compliance with the PM the averaging period of the device and action taken.
emission limit using a bag leak establishing the alarm set points and the (xviii) For affected facilities under
detection system and meet the alarm delay time according to section § 60.50c(a)(3) and (a)(4), concentrations
requirements in paragraphs (g)(1) 5.0 of the ‘‘Fabric Filter Bag Leak of CO as determined by the continuous
through (g)(12) of this section for each Detection Guidance.’’ emission monitoring system.
bag leak detection system. Affected (10) Following initial adjustment of
* * * * *
facilities under § 60.50c(a)(1) and (a)(2) the system, the sensitivity or range,
(6) The results of the initial, annual,
may elect to demonstrate continuous averaging period, alarm set points, or
and any subsequent performance tests
compliance with the PM emission limit alarm delay time may not be adjusted.
conducted to determine compliance
using a bag leak detection system and In no case may the sensitivity be
with the emission limits and/or to
meet the requirements in paragraphs increased by more than 100 percent or
establish or re-establish operating
(g)(1) through (g)(12) of this section. decreased more than 50 percent over a
parameters, as applicable, and a
(1) Each triboelectric bag leak 365-day period unless such adjustment
description of how the operating
detection system shall be installed, follows a complete fabric filter
parameters were established or re-
calibrated, operated, and maintained inspection that demonstrates that the
established, if applicable.
according to the ‘‘Fabric Filter Bag Leak fabric filter is in good operating
Detection Guidance,’’ (EPA 454/R–98– condition. Each adjustment shall be * * * * *
recorded. (c) The owner or operator of an
015, September 1997). This document is
(11) Record the results of each affected facility shall submit the
available from the U.S. Environmental
inspection, calibration, and validation information specified in paragraphs
Protection Agency (U.S. EPA); Office of
check. (c)(1) through (c)(4) of this section no
Air Quality Planning and Standards;
(12) Initiate corrective action within 1 later than 60 days following the initial
Sector Policies and Programs Division;
hour of a bag leak detection system performance test. All reports shall be
Measurement Policy Group (D–243–02),
alarm; operate and maintain the fabric signed by the facilities manager.
Research Triangle Park, NC 27711. This
document is also available on the filter such that the alarm is not engaged * * * * *
Technology Transfer Network (TTN) for more than 5 percent of the total (2) The values for the site-specific
under Emission Measurement Center operating time in a 6-month block operating parameters established
Continuous Emission Monitoring. Other reporting period. If inspection of the pursuant to § 60.56c(d) or § 60.56c(i), as
types of bag leak detection systems shall fabric filter demonstrates that no applicable, and a description of how the
be installed, operated, calibrated, and corrective action is required, no alarm operating parameters were established
maintained in a manner consistent with time is counted. If corrective action is during the initial performance test.
the manufacturer’s written required, each alarm is counted as a * * * * *
specifications and recommendations. minimum of 1 hour. If it takes longer (4) For each affected facility that uses
(2) The bag leak detection system than 1 hour to initiate corrective action, a bag leak detection system, analysis
shall be certified by the manufacturer to the alarm time is counted as the actual and supporting documentation
demonstrating conformance with EPA maintenance, and any repairs not section no later than [DATE 30 DAYS
guidance and specifications for bag leak completed within 10 days of an AFTER DATE OF PUBLICATION OF
detection systems in § 60.57c(g). inspection or the timeframe established FINAL RULE]. All reports shall have
(d) An annual report shall be by the Administrator. been signed by the facility’s manager.
submitted 1 year following the (10) For affected facilities using a bag (1) The previous emissions test results
submission of the information in leak detection system, records of each as recorded using the methods and
paragraph (c) of this section and alarm, the time of the alarm, the time procedures in § 60.56c(b)(1) through
subsequent reports shall be submitted corrective action was initiated and (12), as applicable. Previous emissions
no more than 12 months following the completed, and a brief description of the test results recorded using EPA-
previous report (once the unit is subject cause of the alarm and the corrective accepted voluntary consensus standards
to permitting requirements under title V action taken. are also acceptable.
of the Clean Air Act, the owner or (11) For affected facilities under
§ 60.50c(a)(3) and (a)(4), concentrations (2) Certification that the test results
operator of an affected facility must
of CO as determined by the continuous are representative of current operations.
submit these reports semiannually). The
annual report shall include the emission monitoring system. (3) The values for the site-specific
information specified in paragraphs * * * * * operating parameters established
(d)(1) through (9) of this section. All (g) The owner or operator of an pursuant to § 60.56c(d) or (i), as
reports shall be signed by the facilities affected facility that uses the results of applicable.
manager. previous emissions tests to demonstrate (4) The waste management plan as
* * * * * compliance with the emission limits specified in § 60.55c.
(9) Records of the annual air pollution shall submit the information specified 16. Table 1 to subpart Ec is revised to
control device inspection, any required in paragraphs (g)(1) through (g)(4) of this read as follows:
TABLE 1 TO SUBPART EC OF PART 60.—EMISSION LIMITS FOR SMALL, MEDIUM, AND LARGE HMIWI
Emission limits
Units HMIWI size
Pollutant (7 percent oxygen dry basis)
Small Medium Large
1. Units for which construction is commenced after June 20, 1996 but no later than February 6, 2007 or for which modification is commenced on
or after March 16, 1998 but no later than [THE DATE 6 MONTHS AFTER PROMULGATION OF THE FINAL RULE]
Particulate matter ....... Milligrams per dry standard cubic meter 41 (0.018) .................. 21 (0.0090) ................ 21 (0.0090).
(grains per dry standard cubic foot).
Carbon monoxide ...... Parts per million by volume ........................... 32 1 ............................. 32 1 ............................. 32 1.
Dioxins/furans ............ Nanograms per dry standard cubic meter 111 (49) or 2.1 (0.92) 20 (8.7) or 0.53 (0.23) 20 (8.7) or 0.53
total dioxins/furans (grains per billion dry (0.23).
standard cubic feet) or nanograms per dry
standard cubic meter TEQ (grains per bil-
lion dry standard cubic feet).
Hydrogen chloride ...... Parts per million by volume or percent reduc- 15 or 99% .................. 15 or 99% .................. 15 or 99%.
tion.
Sulfur dioxide ............. Parts per million by volume ........................... 46 1 ............................. 46 1 ............................. 46 1.
Nitrogen oxides .......... Parts per million by volume ........................... 225 1 ........................... 225 1 ........................... 225 1.
Lead ........................... Milligrams per dry standard cubic meter 0.78 1 (0.34) or 71% .. 0.060 (0.026) or 98% 0.060 (0.026) or 98%.
(grains per thousand dry standard cubic
feet) or percent reduction.
Cadmium .................... Milligrams per dry standard cubic meter 0.11 1 (0.048) or 66% 0.030 (0.013) or 93% 0.030 (0.013) or 93%.
Mercury ...................... Milligrams per dry standard cubic meter 0.47 1 (0.21) or 87% .. 0.45 1 (0.20) or 87% .. 0.45 1 (0.20) or 87%.
2. Units for which construction is commenced after February 6, 2007 or for which modification is commenced after [THE DATE 6 MONTHS
AFTER PROMULGATION OF THE FINAL RULE]
Particulate matter ....... Milligrams per dry standard cubic meter 41 (0.018) .................. 21 (0.0090) ................ 21 (0.0090).
(grains per dry standard cubic foot).
Carbon monoxide ...... Parts per million by volume ........................... 25 ............................... 25 ............................... 25.
Dioxins/furans ............ Nanograms per dry standard cubic meter 111 (49) or 2.0 (0.87) 16 (7.0) or 0.21 16 (7.0) or 0.21
total dioxins/furans (grains per billion dry (0.092). (0.092).
standard cubic feet) or nanograms per dry
standard cubic meter TEQ (grains per bil-
lion dry standard cubic feet).
Hydrogen chloride ...... Parts per million by volume or percent reduc- 15 or 99% .................. 15 or 99% .................. 15 or 99%.
tion.
Sulfur dioxide ............. Parts per million by volume ........................... 28 ............................... 21 ............................... 21.
Nitrogen oxides .......... Parts per million by volume ........................... 212 ............................. 212 ............................. 212.
Lead ........................... Milligrams per dry standard cubic meter 0.64 (0.28) or 71% ..... 0.060 (0.026) or 99% 0.060 (0.026) or 99%.
TABLE 1 TO SUBPART EC OF PART 60.—EMISSION LIMITS FOR SMALL, MEDIUM, AND LARGE HMIWI—Continued
Emission limits
Units HMIWI size
Pollutant (7 percent oxygen dry basis)
Small Medium Large
Cadmium .................... Milligrams per dry standard cubic meter 0.060 (0.026) or 74% 0.0050 (0.0022) or 0.0050 (0.0022) or
(grains per thousand dry standard cubic 99%. 99%.
Mercury ...................... Milligrams per dry standard cubic meter 0.33 (0.14) or 96% ..... 0.19 (0.083) or 96% ... 0.19 (0.083) or 96%.
1 Emission limit is less stringent than the corresponding limit for existing sources contained in subpart Ce. Sources that would be subject to the
emission limits in this table also would be subject to regulation under State plans or Federal plans that would implement subpart Ce and would
be subject to limits at least as stringent as those in subpart Ce.
[FR Doc. E7–1617 Filed 2–5–07; 8:45 am]

BILLING CODE 6560–50–P

Proposed Rule: Air Pollution Standards of Performance For New Stationary Sources: Hospital, Medical, and Infectious Waste Incinerators

Uploaded by

Document Information

Original Title

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

Proposed Rule: Air Pollution Standards of Performance For New Stationary Sources: Hospital, Medical, and Infectious Waste Incinerators

Uploaded by

Copyright:

Available Formats

Tuesday,

Category NAICS Code Examples of potentially regulated entities

TABLE 1.—SUMMARY OF PROMULGATED EMISSION LIMITS

TABLE 1.—SUMMARY OF PROMULGATED EMISSION LIMITS—Continued

S .................. 1.2 or 70% reduction ........................................... 1.2 or 70% reduction.

TABLE 2.—COMPARISON OF ESTIMATED POLLUTANT REDUCTIONS VERSUS ACTUAL POLLUTANT REDUCTIONS

HCl .......................................................................................... 98 ............................................................................................ 99.2

CAA section 129 by setting emission prevention or waste minimization

TABLE 3.—SUMMARY OF PROPOSED EMISSION LIMITS IN RESPONSE TO THE REMAND

Pollutant Proposed remand limit for Proposed remand limit for

TABLE 3.—SUMMARY OF PROPOSED EMISSION LIMITS IN RESPONSE TO THE REMAND—Continued

HCl (ppmv) ....................................................................................................................................... L, M, S .............. 15 3 or 99% reduction 3.

HCl (ppm) ........................................................................................................................................ L, M, S .............. 51 or 94% reduction.

TABLE 6.—SUMMARY OF REMAND RESPONSE FOR NEW HMIWI

1997. EPA’s first step in redoing the values.

TABLE 9.— AVERAGE EMISSION VALUES FOR TOP 12 PERCENT OF HMIWI 1

HCl (ppmv) ..................................................................................................................................................... 2,770 53 50

NOX (ppmv) ................................................................................................................................................... 225 225 225

Table 10 of this preamble shows the

TABLE 10.—MACT FLOOR TECHNOLOGY

HCl (ppmv) ....................................... 78 78 or 93% reduction 2 ...................... 57.9 78 3 or 93% reduction 2.

CDD/CDF, total (ng/dscm) ............... 115 115 ................................................... 8,150 115 3

HCl (ppmv) ......................................................... 2,772 78 3 or 93% reduction 2. .................................... 3,125 2 3,100

CO (ppmv) ......................................................... 103 40 2 .................................................................... 109 2 40

Pb (mg/dscm) .................................................... 8.85 0.78 3 or 71% reduction 3 .................................. 8.88 8.9

CDD/CDF, total (ng/dscm) ................................. 8,335 115 3 .................................................................. 8,518 2 800

CDD/CDF, TEQ (ng/dscm) ................................ 239 2.2 3 ................................................................... 244 2 15

NOX (ppmv) ....................................................... 225 250 2 .................................................................. 273 2 250

SO2 (ppmv) ........................................................ 46.4 55 2 .................................................................... 51.8 2 55

1996, or for which modification was

Pollutant Proposed remand response

HCl (ppmv) ........................................................ L, M, S .............. 15 or 99% reduction ......................................... 15 or 99% reduction.

HCl (ppmv) ........................................................ L, M, S .............. 15 or 99% reduction ......................................... 15 or 99% reduction.

Table 17 of this preamble presents a and approved alternative compliance

TABLE 17.—SUMMARY OF HMIWI TEST METHODS AND APPROVED ALTERNATIVE METHODS

9. Table 2 of subpart Ce is revised to

Particulate matter ................................ mg/dscm (gr/dscf) ............................................................................................. 69 (0.030).

§ 60.51c Definitions. introductory text; part shall be used to measure the CO

[FR Doc. E7–1617 Filed 2–5–07; 8:45 am]

You might also like