Browse by Year
/ 2003
/ October
/ Thursday, October 30, 2003
[Federal Register: October 30, 2003 (Volume 68, Number 210)]
[Rules and Regulations]
[Page 61867-61903]
From the Federal Register Online via GPO Access [wais.access.gpo.gov]
[DOCID:fr30oc03-18]
[[Page 61867]]
-----------------------------------------------------------------------
Part II
Environmental Protection Agency
-----------------------------------------------------------------------
40 CFR Part 63
National Emission Standards for Hazardous Air Pollutants: Taconite Iron
Ore Processing; Final Rule
[[Page 61868]]
-----------------------------------------------------------------------
ENVIRONMENTAL PROTECTION AGENCY
40 CFR Part 63
[OAR 2002-0039; FRL-7551-2]
RIN 2060-AJ02
National Emission Standards for Hazardous Air Pollutants:
Taconite Iron Ore Processing
AGENCY: Environmental Protection Agency (EPA).
ACTION: Final rule.
-----------------------------------------------------------------------
SUMMARY: This action promulgates national emission standards for
hazardous air pollutants (NESHAP) for taconite iron ore processing
facilities. The final standards establish emission limitations for
hazardous air pollutants (HAP) emitted from new and existing ore
crushing and handling operations, ore dryers, indurating furnaces, and
finished pellet handling operations. The final standards will implement
section 112(d) of the Clean Air Act (CAA) by requiring all major
sources to meet HAP emission standards reflecting application of the
maximum achievable control technology (MACT).
The HAP emitted by taconite iron ore processing facilities include
metal compounds (such as manganese, arsenic, lead, nickel, chromium,
and mercury), products of incomplete combustion (including
formaldehyde), and the acid gases hydrogen chloride (HCl) and hydrogen
fluoride (HF). Exposure to these substances has been demonstrated to
cause adverse health effects, including chronic and acute disorders of
the blood, heart, kidneys, reproductive system, respiratory system and
central nervous system. Some of these substances are considered
carcinogens. However, it should be noted that the extent and degree to
which the health effects may be experienced depend on:
Pollutant-specific characteristics (e.g., toxicity, half-life in
the environment, bioaccumulation, and persistence); The ambient
concentrations observed in the area (e.g., as influenced by emission
rates, meteorological conditions, and terrain); The frequency and
duration of exposures; and Characteristics of exposed individuals
(e.g., genetics, age, pre-existing health conditions, and lifestyle),
which vary significantly within the general population.
EFFECTIVE DATE: October 30, 2003.
ADDRESSES: Docket. The official public docket is the collection of
materials used in developing the final rule and is available for public
viewing at the EPA Docket Center (EPA/DC), EPA West, Room B102, 1301
Constitution Ave., NW, Washington, DC 20460.
FOR FURTHER INFORMATION CONTACT: Conrad Chin, Metals Group (C439-02),
Emission Standards Division, U.S. EPA, Research Triangle Park, NC
27711, telephone number (919) 541-1512, electronic mail (e-mail) address, chin.conrad@epa.gov.
SUPPLEMENTARY INFORMATION:
Regulated Entities. Categories and entities potentially regulated
by this action include:
----------------------------------------------------------------------------------------------------------------
NAICS code
Category \1\ Example of regulated entities
----------------------------------------------------------------------------------------------------------------
Industry.................................................. 21221 Taconite Iron Ore Processing
Facilities [taconite ore crushing and
handling operations, indurating
furnaces, finished pellet handling
operations, and ore dryers].
Federal government........................................ ............ Not affected.
State/local/tribal government............................. ............ Not affected.
----------------------------------------------------------------------------------------------------------------
\1\ North American Industry Classification System.
This table is not intended to be exhaustive, but rather provides a
guide for readers regarding entities likely to be regulated by this
action. To determine whether your facility is regulated by this action,
you should examine the applicability criteria in Sec. 63.9581 of the
final rule. If you have any questions regarding the applicability of
this action to a particular entity, consult the person listed in the
preceding FOR FURTHER INFORMATION CONTACT section.
Docket. The EPA has established an official public docket for this
action including both Docket ID No. OAR-2002-0039 and Docket ID No. A-
2001-14. The official public docket consists of the documents
specifically referenced in this action, any public comments received,
and other information related to this action. All items may not be
listed under both docket numbers, so interested parties should inspect
both docket numbers to ensure that they have received all materials
relevant to the final rule. Although a part of the official docket, the
public docket does not include Confidential Business Information or
other information whose disclosure is restricted by statute. The
official public docket is available for public viewing at the EPA
Docket Center (Air Docket), EPA West, Room B102, 1301 Constitution
Ave., NW., Washington, DC. The EPA Docket Center Public Reading Room is
open from 8:30 a.m. to 4:30 p.m., Monday through Friday, excluding
legal holidays. The telephone number for the Reading Room is (202) 566-
1744, and the telephone number for the Air Docket is (202) 566-1742.
Electronic Docket Access. You may access the final rule
electronically through the EPA Internet under the ``Federal Register''
listings at http://www.epa.gov/fedrgstr/.
An electronic version of the public docket is available through
EPA's electronic public docket and comment system, EPA Dockets. You may
use EPA Dockets at http://www.epa.gov/edocket/ to view public comments,
access the index listing of the contents of the official public docket,
and to access those documents in the public docket that are available
electronically. Once in the system, select ``search,'' then key in the
appropriate docket identification number. Although not all docket
materials may be available electronically, you may still access any of
the publicly available docket materials through the docket facility in
the above paragraph entitled ``Docket.''
Worldwide Web (WWW). In addition to being available in the docket,
an electronic copy of the final rule will also be available on the WWW
through the Technology Transfer Network (TTN). Following signature, a
copy of the final rule will be placed on the TTN's policy and guidance
page for newly proposed or promulgated rules at http://www.epa.gov/ttn/oarpg.
The TTN provides information and technology exchange in various
areas of air pollution control. If more information regarding the TTN
is needed, call the TTN HELP line at (919) 541-5384.
Judicial Review. This action constitutes final administrative
action on the proposed NESHAP for taconite iron ore processing
facilities (67 FR 77562, December 18, 2002). Under CAA section
307(b)(1), judicial review of the final rule is available only by
filing a petition for review in the U.S. Court of Appeals for the
District of Columbia Circuit by December 29, 2003. Under
[[Page 61869]]
CAA section 307(b)(2), the requirements that are the subject of this
document may not be challenged later in civil or criminal proceedings
brought by the EPA to enforce these requirements.
Outline. The information presented in this preamble is organized as
follows:
I. Background
II. Summary of Final Rule
A. Who must comply with the final rule?
B. What are the affected sources and emission points?
C. What are the emission limitations?
D. What are the operation and maintenance requirements?
E. What are the general compliance requirements?
F. What are the initial compliance requirements?
G. What are the continuous compliance requirements?
H. What are the notification, recordkeeping, and reporting
requirements?
I. What are the compliance deadlines?
III. Summary of Responses to Major Comments
A. How did we revise the cost estimates and economic analysis?
B. How did we revise the performance testing requirements?
C. How did we revise the emission limitations?
D. How did we revise the continuous compliance requirements?
E. How did we revise the baseline emissions?
F. How did we select the pollutants?
IV. Summary of Environmental, Energy, and Economic Impacts
A. What are the air emission impacts?
B. What are the cost impacts?
C. What are the economic impacts?
D. What are the non-air health, environmental and energy
impacts?
V. Statutory and Executive Order Reviews
A. Executive Order 12866: Regulatory Planning and Review
B. Paperwork Reduction Act
C. Regulatory Flexibility Act
D. Unfunded Mandates Reform Act
E. Executive Order 13132: Federalism
F. Executive Order 13175: Consultation and Coordination with
Indian Tribal Governments
G. Executive Order 13045: Protection of Children from
Environmental Health & Safety Risks
H. Executive Order 13211: Actions that Significantly Affect
Energy Supply, Distribution, or Use
I. National Technology Transfer Advancement Act
J. Congressional Review Act
I. Background
Section 112(d) of the CAA requires us (the EPA) to establish
national emission standards for all categories and subcategories of
major sources of HAP and for area sources listed for regulation under
section 112(c). Major sources are those that emit or have the potential
to emit at least 10 tons per year (tpy) of any single HAP or at least
25 tpy of any combination of HAP. Area sources are stationary sources
of HAP that are not major sources. Additional information on the NESHAP
development process can be found in the preamble to the proposed rule
(67 FR 77562).
We received a total of 29 comment letters on the proposed NESHAP
from industry, State agencies, Federal agencies, environmental groups,
and private citizens. We offered to provide interested individuals the
opportunity for oral presentations of data, views, or arguments
concerning the proposed rule, but a public hearing was not requested.
Today's final rule reflects our full consideration of all the
comments we received. Major public comments on the proposed rule along
with our responses to these comments are summarized in section III of
this document. A detailed response to all the comments is included in
the Background Information Document (BID) for the promulgated standards
(Docket ID No. OAR-2002-0039).
II. Summary of Final Rule
A. Who Must Comply With the Final Rule?
Each owner or operator of an affected source at a taconite iron ore
processing plant that is (or is part of) a major source of HAP
emissions must comply with the final rule. A taconite iron ore
processing plant is a major source of HAP if it emits or has the
potential to emit any single HAP at a rate of 10 tons or more per year
or any combination of HAP at a rate of 25 tons or more per year.
B. What Are the Affected Sources and Emission Points?
The affected sources are each new or existing ore crushing and
handling operation, ore dryer, indurating furnace, and finished pellet
handling operation at a taconite iron ore processing facility that is
(or is part of) a major source of HAP emissions. Emission limitations
apply to each ore crushing and handling operation, each ore dryer, each
indurating furnace, and each finished pellet handling operation. These
processes, as well as their emissions and controls, are described in
the preamble to the proposed rule (67 FR 77564-77566).
C. What Are the Emission Limitations?
The final rule includes particulate matter (PM) emission limits,
operating limits for control devices, and work practice standards.
Particulate matter emissions serve as a surrogate measure of HAP
emissions.
Ore Crushing and Handling
The PM emissions limits for ore crushing and handling are 0.008
grains per dry standard cubic foot (gr/dscf) for existing sources and
0.005 gr/dscf for new sources. Compliance with the PM emissions limits
for ore crushing and handling is determined based on the flow-weighted
mean concentration of emissions for all ore crushing and handling units
at the plant.
Ore Dryers
The PM emission limits for each individual ore dryer are 0.052 gr/
dscf for existing dryers and 0.025 gr/dscf for new dryers. Ore dryers
with multiple stacks calculate their PM emissions as a flow-weighted
mean concentration of PM emissions from all stacks.
Indurating Furnaces
For each straight grate indurating furnace processing magnetite,
the PM emissions limits are 0.01 gr/dscf for existing straight grate
furnaces and 0.006 gr/dscf for new straight grate furnaces. For each
grate kiln indurating furnace processing magnetite, the PM emissions
limits are 0.01 gr/dscf for existing grate kiln furnaces and 0.006 gr/
dscf for new grate kiln furnaces. For each grate kiln indurating
furnace processing hematite, the PM emissions limits are 0.03 gr/dscf
for existing grate kiln furnaces and 0.018 gr/dscf for new grate kiln
furnaces. Indurating furnaces with multiple stacks calculate their PM
emissions as a flow-weighted mean concentration of PM emissions from
all stacks.
Finished Pellet Handling
The PM emissions limits for finished pellet handling operations are
0.008 gr/dscf for existing sources and 0.005 gr/dscf for new sources.
Compliance with the PM emissions limits for finished pellet handling is
determined based on the flow-weighted mean concentration of PM
emissions for all pellet handling units at the plant.
Operating Limits
For bag leak detection systems, we require that corrective actions
be initiated within 1 hour of a bag leak detection system alarm. For
dynamic wet scrubbers, the daily average scrubber water flow rate and
either the daily average fan amperage or the daily average pressure
drop must remain at or above the minimum levels established during the
initial performance test. For all other wet scrubbers, the daily
average pressure drop and daily average scrubber water flow rate must
remain at or above the level established during the initial performance
test. Plants using a
[[Page 61870]]
dry electrostatic precipitator (ESP) must either install and operate a
continuous opacity monitoring system (COMS) or maintain the daily
average secondary voltage and daily average secondary current for each
field at or above the minimum levels established during the initial
performance test. If demonstrating compliance using COMS, the average
opacity for each 6-minute period must remain at or below the level
established during the initial performance test. Plants using a wet ESP
must maintain the daily average secondary voltage for each field at or
above the minimum levels established during the initial performance
test; maintain the daily average stack outlet temperature at or below
the maximum levels established during the initial performance test; and
maintain the daily average water flow rate at or above the minimum
levels established during the initial performance test.
You must submit information on monitoring parameters if another
type of control device is used or if alternative monitoring parameters
are desired.
Work Practices
All plants subject to the final rule are required to prepare and
implement a written fugitive dust emissions control plan. The plan
describes in detail the measures that will be put in place to control
fugitive dust emissions from the following sources at a plant, as
applicable: stockpiles, material transfer points, plant roadways,
tailings basin, pellet loading areas, and yard areas. Existing fugitive
dust emission control plans that describe current measures to control
fugitive dust emission sources that have been approved as part of a
State implementation plan or title V permit would be acceptable,
provided they address the prior-listed fugitive dust emission sources.
D. What Are the Operation and Maintenance Requirements?
All plants subject to the final rule must prepare and implement a
written startup, shutdown, and malfunction plan according to the
requirements in 40 CFR 63.6(e). A written operation and maintenance
plan is also required for control devices subject to an operating limit
and indurating furnaces subject to good combustion practices (GCP).
This plan must describe the following: procedures for preventative
maintenance requirements for control devices, corrective action
requirements for baghouses and continuous parameter monitoring systems
(CPMS), and GCP for indurating furnaces. In the event of a bag leak
detection system alarm, the plan must include specific requirements for
initiating corrective action to determine the cause of the problem
within 1 hour, initiating corrective action to fix the problem within
24 hours, and completing all corrective actions needed to fix the
problem as soon as practicable. In the event you exceed an established
operating limit for an air pollution control device other than a
baghouse, you must initiate corrective action to determine the cause of
the operating limit exceedance and complete the corrective action
within 10 calendar days. Corrective action procedures you take must be
consistent with the installation, operation, and maintenance procedures
listed in your site-specific CPMS monitoring plan. For indurating
furnaces, you must maintain a proper and efficient combustion process
through the implementation of GCP.
E. What Are the General Compliance Requirements?
The final rule requires compliance with the emission limitations,
work practice standards, and operation and maintenance requirements at
all times, except during periods of startup, shutdown, and malfunction
as defined in 40 CFR 63.2. The owner or operator must develop and
implement a written startup, shutdown, and malfunction plan according
to the requirements in 40 CFR 63.6(e)(3).
The final rule also requires keeping a log detailing the operation
and maintenance of the process and emission control equipment. This
requirement applies during the period between the compliance date and
the date that continuous monitoring systems are installed and any
operating limits set.
F. What Are the Initial Compliance Requirements?
The final rule requires performance tests to demonstrate that each
affected source meets all applicable PM emission limits. The PM
concentration (front-half filterable catch only) is to be measured
using EPA Method 5, 5D, or 17 in 40 CFR part 60, appendix A. All
initial compliance tests must be completed no later than 180 days
following the compliance date.
To demonstrate initial compliance with the PM emission limit for
the ore crushing and handling affected source, the flow-weighted mean
concentration of PM emissions of all units within the affected source
must not exceed the applicable PM emission limit. Similarly, for the
finished pellet handling affected source, the flow-weighted mean
concentration of PM emissions of all units within the affected source
must not exceed the applicable PM emission limit. In lieu of conducting
performance tests for all ore crushing and handling and finished pellet
handling emission units, the plant may elect to form groups of up to
six similar emission units and conduct initial performance tests on a
representative unit within each group. Each plant must submit a testing
plan to the permitting authority for approval. The testing plan must
identify the emission units that will be grouped as similar, identify
the representative unit that will be tested for each group, and present
the proposed schedule for testing.
To demonstrate initial compliance with the PM emission limit for
each indurating furnace and each ore dryer, the flow-weighted mean
concentration of PM emissions of all stacks associated with each
furnace or each ore dryer must not exceed the applicable PM emission
limit.
The final rule also includes procedures for establishing site-
specific operating limits for control devices during the initial
performance test. To demonstrate initial compliance with the work
practice standards, plants must prepare, submit, and implement a
fugitive dust emission control plan on or before the compliance date.
To demonstrate initial compliance with the operation and maintenance
requirements, plants must prepare the operation and maintenance plan
and certify in their notification of compliance status that they have
prepared the written plans and will operate control devices and
indurating furnaces according to the procedures in the plan.
G. What Are the Continuous Compliance Requirements?
For ore crushing and handling, ore dryers, and finished pellet
handling units, you must conduct subsequent performance tests to
demonstrate continued compliance with the PM emission limits following
the schedule established in the title V permit for each plant. If a
title V permit has not been issued, you must submit a testing plan and
schedule to the permitting authority for approval.
For each indurating furnace, you must conduct subsequent
performance testing of all stacks based on the schedule established in
each plant's title V operating permit, but no less frequently than
twice per 5-year permit term. If a title V permit has not been issued,
then you must submit a testing plan and schedule to the permitting
authority for approval. The testing frequency in the testing plan must
provide for tests to be
[[Page 61871]]
conducted at least twice per 5-year period.
You are required to monitor operating parameters for control
devices subject to operating limits and carry out the procedures in
their fugitive dust emissions control plan and their operation and
maintenance plan. To demonstrate continuous compliance, you must keep
records documenting compliance with the rule requirements for
monitoring, the fugitive dust emissions control plan, the operation and
maintenance plan, and installation, operation, and maintenance of a
CPMS.
For baghouses, owners or operators are required to monitor the
relative change in PM loading using a bag leak detection system and to
make inspections at specified intervals. The bag leak detection system
must be installed and operated according to the EPA guidance document
``Fabric Filter Bag Leak Detection Guidance,'' EPA 454/R-98-015,
September 1997. The document is available on the TTN at http://www.epa.gov/ttnemc01/cem/tribo.pdf.
If the system does not work based
on the triboelectric effect, it must be installed and operated
consistent with the manufacturer's written specifications and
recommendations. The basic inspection requirements include daily,
weekly, monthly, or quarterly inspections of specified parameters or
mechanisms with monitoring of bag cleaning cycles by an appropriate
method. To demonstrate continuous compliance, the final rule requires
records documenting conformance with the operation and maintenance
plan, as well as the inspection and maintenance procedures.
For dynamic wet scrubbers, you must use CPMS to measure and record
the daily average scrubber water flow rate and either the daily average
fan amperage or the daily average pressure drop. For all other wet
scrubbers, you must use CPMS to measure and record the daily average
pressure drop and daily average scrubber water flow rate.
For dry ESP, you must either use a COMS to measure and record the
average opacity of emissions exiting each stack of the control device
for each 6-minute period, or use CPMS to measure and record the daily
average secondary voltage and daily average secondary current for each
field. You must operate and maintain the COMS according to the
requirements in 40 CFR 63.8 and Performance Specification 1 in 40 CFR
part 60, appendix B. These requirements include a quality control
program including a daily calibration drift assessment, quarterly
performance audit, and annual zero alignment.
For wet ESP, you must use CPMS to measure and record the daily
average secondary voltage for each field, the daily average stack
outlet temperature, and the daily average water flow rate.
The final rule requires you to prepare a site-specific monitoring
plan for CPMS that addresses installation, performance, operation and
maintenance, quality assurance, and recordkeeping and reporting
procedures. These requirements replace the more detailed performance
specifications contained in the proposed rule.
To demonstrate continuous compliance, you must keep records
documenting compliance with the monitoring requirements (including
installation, operation, and maintenance requirements for monitoring
systems) and the operation and maintenance plan.
H. What Are the Notification, Recordkeeping, and Reporting
Requirements?
The notification, recordkeeping, and reporting requirements are
based on the NESHAP General Provisions in 40 CFR part 63, subpart A.
Table 2 to subpart RRRRR of 40 CFR part 63 lists each of the
requirements in the General Provisions (Sec. Sec. 63.2 through 63.15)
with an indication of whether they apply.
You are required to submit each initial notification required in
the NESHAP General Provisions that applies to your plant. These include
an initial notification of applicability with general information about
the plant and notifications of performance tests and compliance status.
You are required to maintain the records required by the NESHAP
General Provisions that are necessary to document compliance, such as
performance test results; copies of startup, shutdown, and malfunction
plans and associated corrective action records; monitoring data; and
inspection records. Except for the operation and maintenance plan, the
fugitive dust emissions control plan, and the testing plan, all records
must be kept for a total of 5 years, with the records from the most
recent 2 years kept onsite. The final rule requires that the operation
and maintenance plan, the fugitive dust emissions control plan, and the
testing plan, be kept onsite and available for inspection upon request
for the life of the affected source or until the affected source is no
longer subject to the final rule requirements.
Semiannual reports are required for any deviation from an emission
limitation (including an operating limit), or operation and maintenance
requirement. Each report is due no later than 30 days after the end of
the reporting period. If no deviation occurred, only a summary report
is required. If a deviation did occur, more detailed information is
required.
An immediate report is required if actions taken during a startup,
shutdown, or malfunction are not consistent with the startup, shutdown,
and malfunction plan. Deviations that occur during a period of startup,
shutdown, or malfunction are not violations if you demonstrate to the
authority with delegation for enforcement that the source was operating
in accordance with the startup, shutdown, and malfunction plan.
An immediate report is required after the third consecutive and
unsuccessful attempt at corrective action for determining the cause of
exceedance of an operating limit for an air pollution control device
except for baghouses. The report must be submitted within 5 calendar
days after the third unsuccessful attempt at corrective action. This
report must notify the Administrator that a deviation has occurred and
document the types of corrective measures taken to address the problem
that resulted in the deviation of established operating parameters and
the resulting operating limits.
You must also submit the fugitive dust emissions control plan,
testing plan, and all operation and maintenance plans to the
Administrator on or before the applicable compliance date.
I. What Are the Compliance Deadlines?
The owner or operator of an existing affected source must comply by
October 30, 2006. An existing affected source is one constructed or
reconstructed before December 18, 2002. New or reconstructed sources
that startup on or before October 30, 2003 must comply by October 30,
2003. New or reconstructed sources that startup after October 30, 2003
must comply upon initial startup.
III. Summary of Responses to Major Comments
A. How Did We Revise the Cost Estimates and Economic Analysis?
Comment: Three commenters stated that the estimated total capital
cost impact of $47.3 million underestimates the cost to the industry.
One of the commenters stated that the costs for their plant were
underestimated.
Response: The capital equipment costs used in the cost analysis
conducted prior to proposal were based largely on historical industry
costs provided by industry and vendor estimates obtained by the EPA.
All of
[[Page 61872]]
the indurating furnace capital equipment replacement costs were based
on equipment and installation costs incurred by Minntac in 1991 to
install two new venturi scrubbers for furnace lines 4 and 5. For ore
crushing and handling and pellet handling units, the capital equipment
replacement costs were based on equipment costs obtained from two wet
scrubber vendors.
In follow-up discussions with the industry, industry
representatives indicated that the costs of purchasing and installing a
new wet scrubber were underestimated. For example, based on the cost
estimates provided by one plant, the installation of two new wet
scrubbers on their furnace would cost $18 million, not the $9.4 million
estimated by EPA. We asked each plant to provide an estimate of the
cost impact the limits in the final rule will have on their plant.
Overall, industry estimated a capital equipment and installation cost
of $57 million. The costs provided by industry are based on a
combination of costs estimated by plant engineers, previous equipment
replacement costs, and vendor cost estimates.
The EPA asserts that the impact estimate of $57 million provided by
the industry is a conservatively high estimate based on the fact that
some plants did not account for the averaging of the emissions for
those units within the ore crushing and handling and finished pellet
handling affected sources. However, in order to ensure that we fully
account for the cost impact to the industry, we used the conservatively
high estimates provided by the industry. Therefore, the capital cost
impact of the emission limits in the final rule was estimated to be
approximately $57 million, including emission control capital costs and
monitoring, recordkeeping, and reporting (MRR) capital costs. The
annual costs of the final rule are estimated to be $9 million per year,
including annualized capital and annual operational and MRR costs. For
more information on the industry provided costs and the revised cost
analysis, see the revised cost analysis memorandum in the docket.
Comment: Two commenters stated that the costs of the rule as
proposed are disproportionate to the reduction in HAP.
Response: The revised estimate of annual compliance costs for the
final rule is $9 million per year, and this expenditure is estimated to
result in the reduction of 270 tpy of HAP and 10,538 tpy of PM. The
corresponding cost per ton of HAP reduced is $33,333; the corresponding
cost per ton of PM reduced is $854. These values are similar to or
lower than those in other MACT standards. In addition, the emission
limits in the final rule are based on the MACT floor level of control.
The CAA does not give the EPA the discretion to consider costs for the
MACT floor level of control.
Comment: One commenter stated that the costs and resources
associated with the administrative requirements (e.g., continuous
monitoring, stack testing) of the final rule will pose a significant
additional burden on their operations. The commenter cited estimated
costs of $515,000 for the installation of additional instrumentation
and monitoring equipment, an additional cost of $100,000 for dust
collector monitoring maintenance, and an additional cost of $45,000 for
stack testing. The commenter stated that their plant is already
operating under a title V permit and already has a well-controlled dust
control system in place. The commenter asserted that the increased
continuous monitoring and increased stack testing is not necessary to
protect human health or the environment and adds unnecessary costs.
Response: In the proposed rule, we included only those monitoring
and testing requirements that were necessary to ensure the continued
compliance with the PM emission limits. However, following a review of
the public comments and follow-up discussions with the industry and
States, we have written the final rule to reduce the monitoring and
testing burden:
[sbull] To reduce the monitoring burden, we have deleted the
requirements to conduct monthly transducer checks, quarterly gauge
calibration checks, semiannual flow sensor calibration checks, daily
pressure tap pluggage checks, and monthly electrical connection
continuity checks.
[sbull] We have reduced the indurating furnace stack testing burden
by removing the requirement to conduct simultaneous tests of all the
stacks on one furnace. The final rule allows plants to conduct
sequential testing of the stacks for a furnace, provided the tests are
completed ``within a reasonable period of time, such that the
indurating furnace operating characteristics remain representative for
the duration of the stack tests.''
[sbull] We have removed the volumetric flow rate and process
throughput rate criteria for grouping similar ore crushing and handling
and pellet handling emission units. This will allow more of these
emission units to be grouped together, and thus, will result in fewer
initial compliance tests being required for them.
[sbull] For dry ESP, we have allowed plants to monitor daily
average secondary voltage and daily average secondary current in lieu
of using a COMS.
Comment: According to one commenter, it is confusing that in one
section of the Economic Impact Assessment (EIA), the Agency concludes
that the final rule alone is unlikely to lead to mine closure, but
clearly states that it's possible that two or three firms may close or
sell some or all of their operations. The only consistent statement in
the EIA, according to the commenter, is that the proposed rule will add
to existing financial stresses in the industry.
Response: The empirical literature on steel mill capacity and
closure suggests that import and mini-mill competition are more
important explanatory variables for capacity and closure decisions than
are pollution abatement cost expenditures. The EPA's market and
facility impact analysis did not explicitly model mine closure
decisions because of limited mine-level data and because the costs of
compliance are relatively small. The EPA's data indicate that the
compliance costs alone are generally too low to result in facility
closure. However, we recognized that several companies that owned
taconite mines in 2000 were already under significant financial
hardship; four firms experienced operating losses in 2000, and several
were also operating under Chapter 11 protection. As a result, EPA
collected financial data and considered several criteria to determine
whether companies would be able to obtain financing for capital
investments associated with compliance, or might have to close or sell
individual mine operations. The EPA examined the following company
financial data:
[sbull] Change in profits projected by the economic model;
[sbull] Altman Z-scores;
[sbull] Current ratios; and
[sbull] Recent environmental compliance expenditures.
Based on our review, EPA concluded that two or three companies may
close or sell operations. A review of recent data from the U.S.
Geological Survey (USGS) and company financial reports confirms this
pattern. In 2001, financially-strapped steel companies sold assets.
Cleveland-Cliffs raised its total ownership of Tilden mine to 85
percent by acquiring an additional 45 percent share from Algoma Steel
Inc. Cleveland-Cliffs and Minnesota Power purchased LTV Steel Co. in
late 2001. Cleveland-Cliffs then acquired all the mining and processing
facilities, including 25 percent share of the
[[Page 61873]]
Empire mine. In the face of continuing financial pressures from mini-
mills and imports, steel companies may close or sell taconite
facilities if they cannot obtain financing for compliance. A USGS iron
ore expert contacted by EPA, however, stated that 2002 financial and
market conditions were somewhat better than 2001. This was confirmed by
reviewing financial statements for these firms; while still
experiencing difficult conditions, in 2002 conditions improved somewhat
compared to 2001.
Comment: One commenter from National Steel stated that it will
likely be forced to shut down because it will be unable to make the
upgrades necessary to comply with the rule as proposed. National
currently employs nearly 500 people. The rule as proposed is
anticipated to put these people out of work for a reduction of less
than 5 tons of HAP. In addition to the anticipated closure of
National's operations, the EPA analysis concluded that another one or
two taconite ore processing plants may also close.
Response: As noted in the previous response, EPA's analysis
suggests that the costs of achieving compliance are not sufficient
alone to result in taconite plants becoming unprofitable. However, EPA
recognizes that there are long-standing trends in the industry, such as
increased imports of iron and steel and increasing use of mini-mill
technology, that have resulted in decreasing demand for U.S.-produced
taconite pellets over time. Due to these trends, four companies owning
taconite facilities were unprofitable in 2000, and three of them
(including National Steel) were operating under the protection of
Chapter 11 of the bankruptcy code. The EPA's analysis recognizes that
firms that are unprofitable or in bankruptcy may have difficulty
obtaining financing for the capital investments needed to comply. Such
firms may choose to sell or shut down their taconite plants. The EPA
does not feel that such a decision should be entirely attributed to the
final rule. However, note that recent industry data seem to show that
in 2002, prices and profits improved somewhat due in part to the
decrease in taconite supply (due in part to LTV's closing of the Hoyt
Lakes facility) and in part due to tariff protection of several steel
products.
Comment: According to one commenter, the statement in the EIA that
two or three mines may close implies that Minnesota would see an
additional loss of approximately 900 direct employees and $20 million
in local taxes. The loss of 900 jobs equates to $67.5 million in wages
and benefits. These figures represent a realistic social impact and
create a different scenario than the one represented by the EPA in the
EIA. These economic impacts will be ``devastating'' to an area heavily
dependent on the mining industry.
Response: Chapter 4 of the EIA contains a regional impact analysis
carried out by EPA. The analysis is carried out using IMPLAN, a
regional-level input-output model. The total direct impact on each
region (a State in this analysis) is defined in the EIA as the change
in local expenditures resulting from final rule implementation. The
direct impact of the final rule is estimated based on the results of
the market model, and includes expenditures for compliance (in this
case, positive) and adjustments in outputs in response to price changes
(in this case, negative or positive). Generally, the direct impact
includes the net effect of reduction in local spending because of
output declines and the increase in local spending to implement the
controls. For the State of Minnesota, the EIA shows a net reduction in
local spending of $2.7 million. This is due to a loss of government
revenues since a portion of state revenues comes from taxes on the
total production from taconite iron ore. With the value of changes in
total output included, the total impact to Minnesota is a reduction of
$3.9 million in local spending.
Minnesota is estimated to experience a reduction of 30 full-time
employees as a result of the reduction in taconite production. Thus,
EPA estimates do show a reduction in local spending and employment in
Minnesota from final rule implementation, but not anywhere close to the
amounts asserted by the commenter.
A separate financial assessment examined the financial condition of
companies that own taconite facilities. Because of long-standing trends
in the iron and steel industry (including increasing use of electric
arc furnace mini-mill technology and increasing imports of iron and
steel), several of the owner companies have experienced financial
stress, and three are operating under Chapter 11 protection. For these
reasons, EPA concluded that at least those three firms may have some
difficulty obtaining the financing needed to make capital equipment
investments at their plants, including investments associated with
environmental compliance. The EPA stated that as many as two or three
additional taconite facilities were in danger of closing or selling
their taconite plants at the time of the analysis, due mainly to
factors unrelated to the rule as proposed. However, the additional
costs associated with the final rule will put additional stress on
these already stressed companies. Recent USGS data indicate that in
2001, financially-strapped taconite firms did sell assets to Cleveland
Cliffs. Since the original EIA, however, conditions have improved
somewhat in the industry. The reduced output due to the closure of Hoyt
Lakes, and the tariff, which has increased the effective price of
imported iron and steel commodities, have resulted in increased prices
and profits for iron and steel companies over the past year. Thus, the
companies are somewhat less vulnerable than they were at the time of
EPA's earlier analysis.
B. How Did We Revise the Performance Testing Requirements?
Comment: Two commenters stated that language should be included in
the final rule either authorizing some discretion on behalf of State
agencies or otherwise allowing testing completed between the
promulgation date and the compliance date to be counted as initial
compliance testing. The commenters stated that this will allow
additional time to spread out the compliance testing requirements.
Response: At proposal, plants were given 2 years after the
compliance date to conduct their initial compliance tests for ore
crushing and handling and pellet handling units, and 180 days after the
compliance date to conduct their initial compliance tests for
indurating furnaces. However, since the time of proposal, EPA has
determined that allowing more than 180 days for initial compliance is
not consistent with the 40 CFR part 63 General Provisions. Therefore,
we have written the initial compliance testing deadline for ore
crushing and handling and pellet handling units at 180 days after the
compliance date.
More than 180 days are needed to conduct compliance testing and to
reduce the burden of the final rule on the industry. Therefore, the EPA
has written the final rule to allow source tests conducted between the
promulgation date and the compliance date to be used for compliance
demonstration, as long as the tests are performed in accordance with
the requirements of the final rule. Since the compliance period is 3
years, plants will have a total of 3\1/2\ years to conduct the initial
compliance tests for all of their units.
Comment: Two commenters supported the part of the proposed standard
that allows plants to conduct initial performance tests by testing a
representative sample of units within a group of similar units.
However, in a
[[Page 61874]]
redline/strike-out version of the proposed rule submitted by the
commenters, they removed the specific criteria defining similar units
in Sec. 63.9620(f) and the criteria indicating the number of units
that must be tested per similar group in Sec. 63.9620(g). In the place
of these specific criteria, the commenters inserted a statement that
refers to criteria established by the State agency or in the title V
permit.
Response: In follow-up discussions with the commenters, EPA asked
the commenters to clarify their specific concerns regarding the
criteria for the testing of representative units. The commenters
indicated that their primary concern was with the criteria in
paragraphs (3) and (4) of Sec. 63.9620(f), which require the
volumetric flow rates of the emission units to be within plus or minus
10 percent of the representative emission unit, and the actual process
throughput rate to be within plus or minus 10 percent of the
representative emission unit. The commenters stated that these criteria
were so restrictive that they would not be able to group very many
units.
The EPA also conducted follow-up discussions with the Minnesota
Pollution Control Agency (MPCA) regarding the criteria they use for
grouping similar units. The MPCA staff indicated that the primary
reason they group emission units is to reduce the number of permitted
emission units, although the same groupings are used for testing
purposes. The grouping of emission units by MPCA was conducted
primarily on the basis of control type, installation date, and, to a
certain degree, process type. However, in some cases they do group
emission units from different processes. They do not group emission
units on the basis of flow rate or process throughput.
Based on these discussions with the commenter and MPCA, EPA has
determined that the criteria in Sec. 63.9620(f)(3) and (4) are too
restrictive and, therefore, do not achieve EPA's true intent--the
reduction of the initial compliance test burden for ore crushing and
handling and pellet handling emission units. As a result, EPA has not
included the criteria in Sec. 63.9620(f)(3) and (4) as proposed. The
criteria in Sec. 63.9620(f)(1) and (2) as proposed have been retained
in the final rule. In addition, we have included the following new
criteria: The representative unit must have parametric monitoring
values that encompass the characteristics of all the emission units
within the group.
Comment: Three commenters stated that the simultaneous testing of
multiple indurating furnace stacks is costly. Two of the commenters
stated that simultaneous testing is also impractical and possibly not
even feasible.
Response: In follow-up discussions with the commenters, they
stressed that some furnaces have as many as five stacks. In order to
test these stacks simultaneously, they would need to have five source
testing teams on site at the same time. The commenters stated that this
would be very expensive. The commenters stated that for their current
title V permits, they are not required to conduct simultaneous tests of
all stacks for a furnace. In our discussions with MPCA, they confirmed
that, although they require all plants with permits to test all furnace
stacks, they do not require that the plants test all the stacks on a
furnace simultaneously. Also, in these discussions, it was noted that
the operating conditions are consistent enough that emissions should
not vary significantly over a short period of time. Based on these
discussions, EPA agrees that the simultaneous testing of indurating
furnace stacks would be costly and would provide no additional
compliance assurance. Therefore, in order to reduce the source testing
burden of the final rule on the industry and to maintain consistency
with current testing requirements, EPA has not included the requirement
for simultaneous testing in the final rule.
Comment: Two commenters stated that any requirements for sample
volume or sample time should be removed from the initial and continuous
compliance testing requirements. The commenters stated that the final
rule should not include provisions that are different from already
established EPA test methods.
Response: In the proposed rule, we specified a minimum sample
volume of 60 dscf for EPA Method 5 (40 CFR part 60, Appendix A) tests
to ensure that enough PM is collected to provide accurate results. The
EPA Method 5 does not contain specifications for sample volume or
sample time (i.e., sampling duration). Therefore, it is not uncommon
for the EPA to specify a minimum sample volume or sample time
corresponding to emission characteristics of an industry for EPA Method
5 tests. For example, the Integrated Iron and Steel NESHAP specifies a
minimum sample volume (60 dscf) for EPA Method 5 tests.
Based on historical Method 5 tests from taconite plants, most 1-
hour tests sampled about 30 to 50 dscf and obtained a dry catch of 2 to
20 milligrams (mg). The EPA's Emissions Measurement and Assessment
Division recommends a dry particulate catch of approximately 20 mg for
an accurate Method 5 test. At the same historical particulate
concentrations, a sample volume of 60 dscf or a test of 2 hours in
duration will obtain a dry catch of approximately 20 to 30 mg. In the
proposed rule, we specified a minimum sample volume of at least 60 dscf
for each run of a Method 5 test to ensure that an adequate amount of
dry catch is obtained. However, since proposal we have determined that
specifying a 2-hour sampling time will provide a greater assurance that
an adequate catch is obtained. For example, with a sample volume of 60
dscf, a 20-mg dry catch is obtained for units with emissions of 0.005
gr/dscf or greater. By comparison, given the typical sampling rates of
0.75 to 1 dscf per minute from the historical tests, specifying a 2-
hour test provides a 20-mg dry catch for units with emissions as low as
0.003 gr/dscf. In addition, specifying the sampling time is consistent
with other recently published rules, such as the Portland Cement
NESHAP. Therefore, we have modified the testing requirements in the
final rule by removing the requirement for a sample volume of 60 dscf
and adding the requirement that the duration of each test run be at
least 2 hours.
C. How Did We Revise the Emission Limitations?
Comment: Two commenters stated that the emission limits should be
set at two significant figures and not three significant figures. The
commenters asserted that using three significant figures implies more
precision than exists in reality and establishes limits that are
unrealistically stringent and that do not allow for natural variations.
Response: In the proposed rule, we numerically expressed the
emission limits for all affected sources, new and existing, to three
digits (e.g., 0.011 gr/dscf, 0.025 gr/dscf, and 0.008 gr/dscf). Thus,
the proposed emission limits were already expressed as one or two
significant figures. However, the intent of the commenters is for the
EPA to consider rounding the proposed emission limits to two digits to
account for normal variability in the taconite iron ore processing
operations, performance of air pollution control equipment, and source
testing procedures.
We have reevaluated how natural variations were accounted for in
the proposed emission limits for existing sources. The PM emission
limits for existing sources in the ore crushing and handling affected
source and the finished pellet handling affected source remain at 0.008
gr/dscf. In the final rule, you have the option to determine an
[[Page 61875]]
overall, flow-weighted average PM concentration for all emission units
within each of these two affected sources. One purpose for the flow-
weighted average PM concentration procedure is to account for natural
variability in the various types of emission units within each affected
source, the processing operations, the performance of air pollution
control equipment, and source testing procedures.
The PM emission limits for existing sources in the indurating
furnace affected source will be rounded to two digits. For both
existing straight grate and grate kiln indurating furnaces processing
magnetite, the PM emission limit is 0.01 gr/dscf. For existing grate
kiln indurating furnaces processing hematite, the PM emission limit is
0.03 gr/dscf. After we considered the amount of PM source test data
available in establishing the MACT floor, observed variability in
measured PM concentrations from the furnace exhaust stacks, and noted
fluctuations in the taconite iron ore process, we determined that it is
appropriate to round the PM emission limits for existing indurating
furnaces to two decimal places in order to fully account for natural
variability. Even after rounding the PM emission limits for existing
indurating furnaces, we will still achieve nearly the same level of
emission reduction, while offering increased flexibility to the
industry to comply with the emission standards of the final rule.
The PM emission limit for existing ore dryers was determined to be
the level of control indicated by the existing State limit of 0.052 gr/
dscf. Therefore, it is not appropriate to round the PM emission limit
for existing ore dryers. The PM emission limit for existing ore dryers
is 0.052 gr/dscf in the final rule.
The PM emission limits for all new affected sources represent an
actual performance level achieved by the best performing source in each
affected source. Thus, the new source emission limits can be achieved
through the proper design and construction/reconstruction of a new
affected source.
Comment: Three commenters stated that the final rule should more
clearly describe how to calculate the flow-weighted mean PM emissions
concentration for the material handling operations.
Response: We agree with the commenters and have written Sec. Sec.
63.9621 and 63.9622 to provide additional clarification for calculating
the flow-weighted mean PM emissions concentration for ore crushing and
handling and finished pellet handling. Specifically, the final rule
clarifies that when calculating the flow-weighted mean PM emissions for
ore crushing and handling and finished pellet handling, the ``average''
PM concentration corresponding to each emission unit in an affected
source is multiplied by the maximum design volumetric flow rate of the
corresponding emission unit. The ``average'' PM concentration from an
emission unit is derived as the arithmetic mean of a PM source test
comprised of three valid sampling runs on the emission unit. If the
affected source elects to conduct representative compliance testing for
a group of similar emission units, the PM concentration determined for
the tested emission unit will be assigned to the other emission units
identified as similar within the group.
D. How Did We Revise the Continuous Compliance Requirements?
Operating Limits
Comment: Two commenters objected to using operating limits
established during the performance test to determine continuous
compliance. The commenters stated that a performance test is only a
snapshot of an operation at a point in time and may not encompass the
full operational variability that occurs. The commenters stated that
this approach effectively sets a new more stringent NESHAP emission
limit at the emissions level actually emitted during the performance
test. Therefore, the commenters stated that any operation outside of
the operating parameter range should not be classified as a deviation.
The commenters stated that the D.C. Circuit Court has made it clear
that MACT standards are to represent the best performing source on its
worst day (see National Lime v. EPA, 233 F.3d 625, 51 ERC 1737 (D.C.
Cir. 2000), and Cement Kiln Recycling Coalition v. EPA, 255 F.3d 855,
52 ERC 1865 (D.C. Cir 2001)). The commenters asserted that as long as a
source is operating properly, follows procedures in the malfunction
plan, and proceeds appropriately to corrective action, then variations
within the range of proper operation should not constitute deviations.
The commenters stated that the EPA may require plants to log such
information and even report it, but not necessarily as a deviation
under title V.
Response: In follow-up discussions with the industry, we were able
to determine that the taconite industry's primary concern regarding the
operating limits was being able to maintain the equipment so that they
did not exceed the established operating limit. Specifically, their
concerns included their ability to maintain the pressure drop above the
operating limit for venturi-rod deck units with a fixed throat and/or a
volumetric flow dependent of process conditions; and, their ability to
operate and obtain meaningful readings of opacity from dry ESP using a
COMS in conditions of high moisture and low opacity.
Regarding the measurement of the pressure drop, we have increased
the averaging time from hourly to daily. The daily averaging period
addresses industry's concerns about their ability to control pressure
drop during short periods of time when the scrubber may experience a
pressure drop lower than the operating limit. In addition, for dynamic
wet scrubbers, we have provided the flexibility of monitoring either
the daily average pressure drop or the daily average fan amperage, in
addition to the daily average scrubber water flow rate. This addresses
industry's concern that for dynamic wet scrubbers, both pressure drop
and fan amperage are good indicators of proper performance.
Regarding the measurement of opacity using COMS, we have verified
with equipment vendors that COMS are available that will provide
accurate readings under the moisture and low opacity conditions present
at taconite facilities. However, we understand that currently there are
no COMS in operation at taconite plants and that due to costs or site-
specific operating conditions a COMS may not be the best option.
Therefore, in the final rule have provided plants the flexibility to
establish their operating limit either as the 6-minute average opacity
or as the daily average secondary voltage and the daily average
secondary current for each field.
In addition, we have included language in the final rule to clarify
when not meeting an operating limit becomes an exceedance.
Specifically, after the first two times that you do not meet the
operating limit, you must take corrective action. After the third time
that you do not meet the operating limit, you must submit a written
report within 5 calendar days and report the third unsuccessful attempt
of corrective action as a deviation and continue corrective action.
Bag Leak Detection Systems
Comment: Two commenters stated that the requirement in Sec.
63.9634(d)(1) of the proposed rule that requires that the bag leak
detection system not alarm for more than 5 percent of the time should
be deleted from the final rule.
[[Page 61876]]
Two commenters pointed out that Sec. 63.7833(d)(1)(iii) of the
proposed rule specifies that 1 hour of alarm be logged even if
procedures are implemented to determine the cause of the alarm and
corrective action is taken in less than 1 hour. The commenters
contended that the requirement artificially and unfairly inflates the
semiannual percentage of alarm time and does not provide an incentive
for sources to initiate procedures as quickly as may be possible. The
commenters suggested that the final rule should require the plant to
``count the actual amount of time it took to initiate procedures to
determine the cause of the alarm.''
Three commenters stated that in the requirement in Sec.
63.9634(d)(1)(v) that the bag leak detection system not alarm for more
than 5 percent of the ``total operating time,'' it is unclear if the
``total operating time'' refers to the operating time of the affected
source or the time the baghouse is actually evacuating emissions
generated by the affected source. The commenters pointed out that some
baghouses, by design, evacuate emissions for only a few minutes each
hour. The commenters recommended that EPA clarify its intent that the
``total operating time'' refers to the total operating time of the
affected source.
Response: We agree with the commenters and have not included the 5
percent operating limit requirement for baghouse leak detectors in
Sec. 63.9634(d)(1) of the final rule. As a result, the requirements to
log alarm time and to determine the ratio of the sum of the alarm times
to the total operating time have also not been included. However, it is
important that corrective action be initiated promptly, so we are
retaining the requirement in Sec. 63.9600(b)(2) that you ``initiate
corrective action to determine the cause of the alarm within 1 hour of
the alarm, initiate corrective action to correct the cause of the
problem within 24 hours of the alarm, and complete the corrective
action as soon as practicable.''
Wet Scrubber CPMS
Comment: Three commenters stated that the labor hours required for
the monthly transducer checks and the quarterly gauge calibration
checks for the pressure drop sensor (Sec. 63.9632(b)(1)(iv)), and the
semiannual flow sensor calibration checks (Sec. 63.9632(b)(2)(iii))
are excessive compared to the potential emissions control improvement.
Two of the commenters suggested that rather than mandatory monthly,
quarterly, or semiannual calibration checks, any control unit which
emits less than 5 percent of the total annual PM emissions at the plant
should be allowed to reduce the periodic checks required by each of the
cited provisions to once annually. The other commenter suggested that
the EPA should allow each source to propose an alternative method to
the proposed calibration checks to the appropriate permitting agency.
Three commenters stated that the daily pressure tap pluggage check
(Sec. 63.9632(b)(1)(iii)) and monthly electrical connection continuity
checks (Sec. 63.9632(b)(1)(vi)) are overly burdensome and costly to
implement. The commenters argued that the manual labor and clock hours
required for such continuity checks would be so large that the
monitoring systems would have to be shut down so frequently and for
such a length of time that they would have virtually no operating time.
According to the commenters, these provisions should be modified so as
to provide ``a program within the CPMS to alarm the process unit
operator and to record the alarm for a zero value indication and for a
static value indication that satisfies the requirement of this
provision.'' In addition, one commenter stated that, if no change is
made, the labor costs for the continuity checks must be factored into
the economic analysis.
Response: The specific installation, operation, and maintenance
requirements for wet scrubber CPMS have not been included in the final
rule. Therefore, the requirements for monthly transducer checks,
quarterly gauge calibration checks, semiannual flow sensor calibration
checks, daily pressure tap pluggage checks, and monthly electrical
connector continuity checks have not been included in the final rule.
In place of the specific requirements, we have included the requirement
that, for each CPMS, you must develop and make available a site-
specific monitoring plan that addresses the following:
[sbull] Installation of CPMS sampling probe so that measurement is
representative of control of the exhaust emissions.
[sbull] Performance and equipment specifications for the sample
interface, the parametric signal analyzer, and the data collection and
reduction system.
[sbull] Performance evaluation procedures and acceptance criteria
(e.g., calibrations).
[sbull] Ongoing operation and maintenance procedures in accordance
with the general requirements of Sec. 63.8(c)(1), (3), (4)(ii), (7),
and (8).
[sbull] Ongoing data quality assurance procedures in accordance
with the general requirements of Sec. 63.8(d).
[sbull] Ongoing recordkeeping and reporting procedures in
accordance with the general requirements of Sec. 63.10(c), (e)(1), and
(e)(2)(i).
Comment: Three commenters stated that it is inappropriate to set a
single (pressure drop) point for operating wet scrubbers and
recommended that EPA remove the pressure drop requirement and rely on
the operation and maintenance plan for compliance. The commenters
pointed out that venturi-rod deck scrubbers operate over a range of
pressure drop that is affected by scrubbing water flow rate, scrubber
water flow distribution, water temperature, gas temperature, and the
square of the process gas flow rate. The commenters stated that
operators cannot directly control the pressure drop in a venturi-rod
deck scrubber. By setting the average pressure drop at the minimum
level established during the performance test, the commenters stated
that the rule effectively forces a source to operate well below the
emission limit.
Response: In follow-up discussions with the commenters, it was
clarified that their comments referred only to venturi-rod deck
scrubbers installed on indurating furnaces. These venturi-rod deck
scrubbers are fixed-throat scrubbers for which the pressure drop can be
measured, but not directly controlled. Two commenters stated that they
cannot directly control the pressure drop across the venturi-rod deck
scrubbers because of the following factors:
[sbull] The scrubbers are of a fixed-throat design;
[sbull] The fan drawing or pushing air through the scrubber
operates at a fixed speed and fixed diameter; and
[sbull] The damper prior to the scrubber is used to control the
overall flow of air through the system; therefore, it cannot be used to
control the pressure drop to the scrubber without affecting the entire
process. The damper is opened more or closed more, as necessary, to
modulate the air flow as changes occur in the process. As production
rate increases, the damper is opened more and, therefore, the pressure
drop across the scrubber increases. Due to these factors, the pressure
drop across the venturi-rod deck scrubbers on the furnaces is more
variable than other controls and is difficult to regulate.
The commenters presented data showing the variability of the
pressure drop for their venturi-rod deck scrubbers. One commenter
presented pressure drop readings taken every 20 minutes that ranged
from 12 to 4 inches of pressure drop, with very few points below 4
inches of pressure drop.
[[Page 61877]]
However, after excluding periods of malfunction and looking at the
daily average pressure drop instead of instantaneous readings, the data
showed that the daily average pressure drop for each scrubber fell
within a narrow range. The difference between the lowest daily average
pressure drop and the highest daily average pressure drop was only
about 2 or 3 inches of pressure drop. Based on these data, the
commenter stated that they were confident that they could maintain a
pressure drop at or above the operating limit based on a daily average.
The other commenter provided daily average pressure drop for their
venturi-rod deck scrubbers. The data showed that on a daily average
basis, the pressure drop for each venturi-rod deck scrubber varied by 1
to 3.6 inches over a period of 2 months. The commenter requested that
they be allowed to use historical pressure drop data to establish the
pressure drop operating limit for venturi-rod deck scrubbers on
indurating furnaces. In addition, the commenter requested that
compliance with the pressure drop operating limit for venturi-rod deck
scrubbers on indurating furnaces be determined on a daily average
basis.
To address the technical issues raised by the commenters, we have
written the final rule to allow the use of pressure drop data from PM
tests conducted on or after December 18, 2002 (the proposal date) to
establish the operating limit for venturi-rod deck scrubbers
controlling emissions from indurating furnaces. The historical pressure
drop data must be from a certified test for which the PM emission
concentration was at or below the applicable indurating furnace limit
in Table 1 to the final rule. In addition, the basis for compliance
with the pressure drop operating limit for venturi-rod deck scrubbers
on indurating furnaces has been written as an hourly average not a
daily average.
COMS
Comment: Two commenters stated that there should not be any
requirement to install or operate a COMS. The commenters do not support
setting an opacity limit on a case-by-case and site-by-site basis. In
addition, the commenters asserted that the opacity will be low enough
to be outside of the range of error for the test method (the COMS), and
sources could create a reportable deviation without truly exceeding the
actual opacity limit. Instead, the commenters stated that there should
be a requirement for a visible emission check, as is required in the
Portland Cement NESHAP.
Response: We have verified with equipment vendors that COMS are
available that will provide accurate readings at low opacity
conditions. Certain models of COMS can measure opacity as low as 0.1
percent with an accuracy of +/- 0.3 percent. In addition, the COMS
vendors indicated that the COMS will provide accurate readings under
the moisture conditions present at taconite facilities (typically 9
percent moisture). However, we understand that currently there are no
COMS in operation at taconite plants (one facility has scheduled a
trial installation for later this year) and that due to equipment and
installation costs or site-specific operating conditions, a COMS may
not be the best option for each plant. Therefore, in the final rule we
have provided two options for the operating limits for dry ESP: the 6-
minute average opacity, as monitored using a COMS; or the daily average
secondary voltage and the daily average secondary current for each
field, as monitored using a CPMS.
During our dry ESP discussions with industry, it was requested that
we add specific monitoring requirements for wet ESP. After discussion
with the industry and State agencies, we established the following
monitoring parameters for wet ESP:
[sbull] Daily average secondary voltage for each field;
[sbull] Daily average stack outlet temperature; and
[sbull] Daily average water flow rate.
Therefore, the final rule contains requirements to establish
operating limits for these parameters during the initial performance
test. Plants must also monitor these parameters such that they are
maintained at or above the operating limits (for secondary voltage and
water flow rate), or below the operating limits (for stack outlet
temperature).
E. How Did We Revise the Baseline Emissions?
Comment: Two commenters stated that the HAP emission values in the
preamble need to be updated to accurately reflect what is currently
being emitted. Specifically, one of the commenters stated that U.S.
Steel has more recent testing data that can be used to update the
estimates. Another one of the commenters asserted that HAP emissions
from taconite ore plants are inaccurately characterized. The commenter
stated that several companies have more recent test data and EPA can
revise the HAP emissions accordingly. The commenter stated that a more
accurate depiction of the emissions will alter the economic analysis.
Response: In follow-up discussions with the industry, we asked them
to submit any test data that were not reflected in the proposal
analyses. We received the following additional emission tests:
[sbull] Engineering Emissions Test Report for Tilden conducted the
week of November 4, 1999. Tested PM, nitrogen oxides (NOX),
HCl, HF, benzene, hexane, toluene, formaldehyde, metals, and asbestos.
[sbull] Particulate and Metals Emission Study for Tilden conducted
May 7 to 11, 2002. Tested total PM and metals.
[sbull] MPCA spreadsheet incorporating Minntac emissions tests for
December 2002 and August 2001. Tested formaldehyde, HCl, HF, chlorine,
and fluorine.
[sbull] Northshore formaldehyde emissions tests conducted on March
6, 2003. We have reviewed the test data listed above and have revised
the baseline HAP emissions as appropriate. The baseline HAP emissions
have been modified as follows:
[sbull] Baseline formaldehyde emissions were updated for Minntac,
Northshore, and Tilden. The baseline formaldehyde emissions for EVTAC
and Inland were also updated, since their formaldehyde emission factors
were based on Northshore estimates. This resulted in a decrease in
baseline formaldehyde emissions from 180.7 to 30.1 tpy. This had no
effect on the HAP emission reduction estimate since we assumed that
there would be no formaldehyde emission reductions.
[sbull] Baseline HCl and HF emissions were updated for Minntac and
Tilden. This resulted in a decrease in baseline HCl emissions from
349.1 to 274 tpy and a decrease in baseline HF emissions from 308 to
229 tpy. As a result, the emission reduction from acid gases decreased
from 356.1 to 256 tpy.
F. How Did We Select the Pollutants?
Mercury
Comment: Seventeen commenters stated that EPA has a statutory
obligation to set emission standards for mercury. Several commenters
specifically cited National Lime. One commenter stated that the fact
that no specific type of control technology has yet proven effective
and affordable for taconite processing cannot legally excuse the
industry from regulation. Thirteen commenters asserted that EPA's
practice of not setting standards for industries that do not yet
control their emissions is illegal and encourages the industry to do as
little as possible to control mercury.
[[Page 61878]]
One commenter encouraged EPA to consult with the Minnesota
Department of Natural Resources, Division of Lands and Minerals, to get
the most up-to-date information on potential mercury control strategies
for taconite facilities before promulgation. The commenter stated that
viable mercury control technologies or strategies may be identified in
the very near future. The commenter asserted that the best strategies
to control mercury may be operational modifications such as different
handling practices for captured dust from indurating furnaces.
Two commenters stated that the EPA must set an emission standard
for mercury based on the statute's ``minimum stringency requirement''
(i.e., the MACT floor) even if specific technologies or operating
practices to achieve it have not been identified. One commenter stated
that if no such controls or practices are being used, EPA must find
some other factor on which to base the standard. Three commenters
suggested that EPA determine the floor based on the average mercury
emission level of the five plants (or furnaces) with the lowest
emissions, and then set the mercury emission limit there. One commenter
stated that if certain plants will not be able to meet such a standard
within 4 years, the statute provides relief through a Presidential
exemption for a period of not more than 2 years. The commenter also
contends that the CAA allows relief for a company that makes a
significant effort to identify and implement effective controls but is
still unable to meet the standard by the 4-year deadline. The commenter
stated that EPA included a similar provision in the Portland Cement
NESHAP. The commenter believes that setting a standard would induce the
industry to invest in research and development to meet it. The
commenter stated that promising mercury control technologies for the
taconite industry are on the horizon. The commenter stated that the EPA
should investigate the COHPAC-TOXECON system, corona discharge, and
catalytic oxidation, as well as an iron oxide sorbent system being
tested in Minnesota.
One commenter stated that EPA recognized in the proposed rule that
the mercury content of the taconite ore is the ``key factor'' affecting
mercury emissions. The commenter reasoned that by setting a mercury
standard, plants that use ore with high mercury content will have to
find ways to reduce mercury emissions, including switching to cleaner
raw materials or installing pollution controls.
One commenter stated that the final rule should consider precluding
the use of coal, even as a secondary fuel, to control mercury
emissions.
Thirteen commenters recommended that EPA establish a reasonable
limit for mercury and allow relief for a company that is unable to meet
the limit after making appropriate technological or research
investments.
Two commenters requested more information supporting EPA's finding
that ``we were unable to find any viable control technologies or
operating procedures for achieving reduction in mercury emissions from
indurating furnaces at taconite iron ore plants.'' One of the
commenters requested the cost of control per ton of mercury control
that was estimated in EPA's analysis. Both commenters stated that
control technologies being developed for coal-fired power plants could
be used to control mercury emissions from taconite facilities. Two
commenters mentioned activated carbon injection as a potential mercury
control for taconite plants.
One commenter stated that, both within the binational program and
in national policy documents, the EPA insinuates that the NESHAP
program is the means by which the Agency will achieve mercury reduction
goals. The commenter asserted that an emission limit for mercury should
be set that pushes the industry to research and develop control
technology but also allows for relief if a company is unable to meet
the standard after diligently pursuing such technology. The standard
should also include mercury monitoring requirements.
Three commenters stated that if mercury emissions from the taconite
industry are not reduced, the goals of the binational program to
protect the Lake Superior Basin cannot be met. One commenter stated
that, if EPA does not intend to set standards for mercury emissions
from industries that currently do nothing to control their emissions
and that do not develop control technology on a voluntary basis, its
regulations (if not its authority) are inadequate to protect the Great
Lakes and other Great Waters from mercury deposition. The commenter
stated that EPA's refusal to take action under CAA section 112(m)
because authority is available under CAA section 112(d), and then
failing to use the CAA section 112(d) authority is unacceptable.
Furthermore, the commenter stated that Congress directed the EPA to
take action to protect the Great Waters by 1995. The commenter stated
that postponing regulations until residual risk standards are required
violates the spirit (if not the letter) of the congressional mandate.
One commenter stated that beyond-the-floor standards are warranted
for mercury. The commenter stated that a mercury standard based on
developing technologies is ``achievable.'' The commenter stated that
EPA could base beyond-the-floor mercury standards on the reductions
that could be achieved through raw material change (low-mercury ore),
fuel change (natural gas), or control technologies (wet scrubbers,
carbon beds, or activated carbon injection). The commenter recommended
that EPA investigate the COHPAC-TOXECON system, whereby a pulse-jet
baghouse is installed downstream from existing ESP controls, and a
sorbent injection system is installed between the existing ESP and the
baghouse. The commenter also suggested that EPA look at developing
multipollutant technologies, such as corona discharge, catalytic
oxidation, and iron oxide sorbent systems being tested in Minnesota.
One commenter cited estimated costs for activated carbon systems
that were developed for coal-fired boilers that ranged from $4,940 to
$70,000 per pound ($9.9 to $140 million/ton) of mercury removed at 90
percent control (USDOE, September 2002; NESCAUM, June 2000). The
commenter also provided costs for carbon filter beds used in European
waste incinerators of $513 to $1,083 per pound ($1.0 to $2.2 million/
ton) of mercury removed at 99 percent control. The commenter stated
that the control costs for indurating furnaces should lie somewhere
between the two cost ranges. The commenter also provided estimated
costs for enhanced wet scrubbing systems for coal-fired boilers of
$76,000 to $174,000 per pound ($152 to $348 million/ton) of mercury
removed (NESCAUM, June 2000).
Response: There is no way to set a floor standard for mercury that
is ``achievable,'' as required by CAA section 112(d)(2), because there
is no standard that can be duplicated by different sources or
replicable by the same source. The opinion in National Lime did not
deal with a situation where an emission standard was unachievable for
these reasons. Mercury emitted from taconite iron ore processing plants
originates primarily from the ore itself and to a much lesser extent
the fuels powering the process. None of the taconite iron ore
processing plants control mercury emissions by using at-the-stack
controls. Thus, any differences in mercury emissions from existing
indurating furnaces reflect different mercury levels in raw materials
or fossil
[[Page 61879]]
fuels used at the individual plants. Attempting to base a mercury
standard (either a floor standard, or a beyond-the-floor standard) on
raw material substitution (i.e., ore substitution), however, would lead
to unachievable standards for all sources, because this means of
control is not duplicable or even replicable.
A study by the Coleraine Minerals Research Laboratory in 1997
stated that ``the mercury volatilized during pellet induration is not
the same for every taconite operation. There is a correlation between
the amount of mercury volatilized during induration and the location of
the taconite operation. The taconite operations that are located on the
west end of the Mesabi Iron Range volatilize more mercury during pellet
induration than those on the east end of the range.'' This correlation
was confirmed in a report by the Minnesota Department of Natural
Resources (Berndt, 2002) with the mercury concentrations present in the
ore varying from 21 parts per billion (ppb) at the west end of the
range to 0.6 ppb for facilities located on the east end of the range.
Each taconite iron ore processing plant is located directly proximate
to its own mining source. Transportation costs of procuring raw
materials from other locations are prohibitive. A plant has no access
to the raw ore used by another plant and, consequently, could not
duplicate the mercury emissions performance of the other plant. The ore
processing operations at a given plant are dependent on the type of ore
mined. The east range ores are typically finer and harder requiring
different processing steps in crushing, grinding, and flotation.
Because of the differences in processing for each type of ore, it is
not feasible for any one facility to process different ores mined from
multiple locations in the range. Moreover, because iron ore deposits
are variable in mercury content, there is no way to assure that even a
source processing its own ore could replicate its own performance,
since the next ore batch could contain higher concentrations of
mercury. Based on the above justifications, we have determined that it
is infeasible for taconite plants to reduce mercury emissions by
switching to ``cleaner'' ores.
Natural gas is the primary fuel used by the taconite industry to
fuel the process. From the period of 1995 to 1997, the burning of coal
constituted only between 9 and 18 percent of the overall energy input
for taconite indurating furnaces. During the same period, natural gas
constituted between 73 and 83 percent of the overall energy input for
taconite indurating furnaces. Although very little coal is used overall
by the industry, it is critical for certain plants to have coal
available to them as a backup fuel when natural gas may not be
available or when seasonal fluctuations in the price of natural gas
make its use uneconomical. Therefore, based on the negligible impact of
coal on mercury emissions in the industry and the importance of
maintaining backup fuel options, fuel switching is not a feasible means
of controlling HAP metal emissions (including mercury) for the taconite
industry.
Based on these facts, EPA cannot accept the comment that it must
establish a floor standard by averaging the lowest mercury emission
values of the so-called best-performing 12 percent of sources. In the
next performance test, all of these mercury values could be higher (no
matter what method would be used to establish ``best performing''),
because there are no means of controlling ore concentrations or
feasibly using fuel substitution. Such a standard simply could not be
achieved by any source. Not only is this not the intent of a
technology-based standard, but would result in sources being out-of-
compliance and, thus, possibly shutting them down. This is not how MACT
was intended to function. ``MACT is not intended * * * to drive sources
to the brink of shutdown * * *'' (H.R. Rep. No. 101-490, 101st Cong. 2d
sess. 328).
We note further that the mercury in the ore and the fuel is present
in trace amounts. The Minnesota Department of Natural Resources stated
that ``mercury present in taconite occurs as a trace element, and
cannot be eliminated by simply using a different fuel source or by
eliminating mercury-bearing components from material to be combusted.''
(Berndt, 2002) This supports the Agency's technical determinations that
control via substitutions of feed or fuel is neither feasible nor
likely to be effective since random variability in the feed will likely
result in equal amounts of mercury being emitted in any case. Indeed,
as stated above, it is not clear that even a single source could
reliably duplicate its own performance for mercury emissions due to the
small amounts emitted and random variabilities in the mercury content
of the iron ore.
The commenters themselves acknowledge that viable controls for
mercury are not currently available for the taconite industry:
[sbull] One commenter stated that ``viable mercury control
technologies or strategies may be identified in the very near future.''
[sbull] One commenter stated that ``setting a standard would induce
the industry to invest in research and development to meet it.'' The
commenter also stated that ``promising mercury control technologies for
the taconite industry are on the horizon.''
[sbull] Two commenters stated that ``control technologies being
developed for coal-fired plants could be used to control mercury
emissions from taconite facilities.'' Section 112(d) of the CAA
requires that the EPA establish emission standards that are
``achievable for new or existing sources.'' Since we have not been able
to identify any currently employed operating practices that effectively
reduce mercury emissions which are duplicable or replicable, we cannot
develop an achievable floor standard.
Some commenters also suggested extended compliance periods (beyond
the 3 years provided by section 112(i)(3) of the CAA). The problem,
however, is not one of time but of the lack of existence of any means
of floor control. Control of emissions via raw material or fuel
substitution will not be available regardless of time allowed for
compliance.
Several commenters also noted that EPA's action here could
undermine efforts to control mercury deposition in the Great Lakes and
questioned the adequacy of EPA's action in light of the Agency's
obligation under section 112(m)(6) of the CAA to ``determine whether
the other provisions of this section 112 are adequate to prevent
serious adverse effect to public health and serious or widespread
environmental effects'' in the Great Lakes. The EPA, however, is not
reopening its existing determination that the section 112(d) and (f)
standards are adequate for this purpose. See generally 63 FR 14090
(March 24, 1998); ``Deposition of Air Pollutants to the Great Waters:
First Report to Congress (EPA-453/R-93-055, 1994); ``Deposition of Air
Pollutants to the Great Waters: Second Report to Congress'' (EPA-453/R-
97-011, 1997). The EPA notes further that the section 112(f) residual
risk process must evaluate (among other things) whether a more
stringent standard for mercury is needed to prevent an adverse
environmental effect (taking into consideration costs, energy, safety
and other relevant factors).
The commenters' statements regarding potential at-the-stack control
options are legitimate considerations for beyond-the-floor standards,
but after evaluating the possibility of such
[[Page 61880]]
controls against technical considerations and the section 112(d)(2)
factors, we do not feel that a beyond-the-floor standard for mercury is
warranted.
One commenter indicated that different handling practices for
captured dust from indurating furnaces, as discussed in a report by the
Minnesota Department of Natural Resources (Berndt, 2002), would be a
good method for controlling mercury. The control option investigated in
the report involves placing magnetite dust collected by the wet
scrubbers, which was found to be high in mercury, into the waste stream
rather than recycling the dust back to the indurating furnace. A review
of the report cited by the commenter reveals that, for the two taconite
plants studied, the costs of this approach ranged from $28 to $254
million per ton of mercury removed ($14,000 to $127,000 per pound of
mercury removed). This high cost results from the loss of over $1
million of magnetite dust product ($25 per long ton) to prevent
approximately 30 pounds of mercury emissions. The study concludes that
``due to the high cost of this emission control method, the large
uncertainty in the cost estimates, and the limited amount of emission
reduction, it appears that more research is needed before mercury
emission control methods can be put into practice in taconite
processing facilities.'' We believe that the high cost, the small
reduction in HAP emissions, and increased waste disposal do not justify
this beyond-the-floor alternative at this time.
Other potential mercury controls cited by the commenters include:
wet flue gas desulfurization (FGD), baghouses, activated carbon
injection, activated carbon/baghouse system (COHPAC), corona discharge,
electro-catalytic oxidation, and injection of copper-coated magnetic
taconite concentrate.
Ninety seven percent of the mercury emitted from taconite plants is
emitted from the indurating furnaces. The mercury emitted from the
taconite indurating furnaces is primarily elemental mercury. Wet
scrubbing systems, such as wet FGD, ``are very effective at removing
soluble ionic mercury, but are not very effective at removing insoluble
elemental mercury'' (NESCAUM, 2000). Therefore, wet FGD systems were
not considered to be a technically viable beyond-the-floor option.
Baghouses and control systems that utilize them, such as the COHPAC
system, cannot be used on taconite indurating furnace stacks due to the
high moisture content of the exhaust gas. The high moisture content of
the exhaust gas causes plugging problems that make the baghouses
ineffective. Therefore, baghouses and control systems based on baghouse
technology were not considered to be a technically viable beyond-the-
floor option.
In pilot scale studies at several electricity generating boilers,
carbon injection has provided up to a 90 percent reduction in mercury
emissions. Estimated costs for installing activated carbon injection
systems on electricity generating boilers range from $10 to $140
million per ton of mercury removed ($5,000 to $70,000 per pound of
mercury removed) (NESCAUM, 2000; USDOE, 2002). Activated carbon
injection has been demonstrated to provide 95 percent control of
mercury emissions for municipal waste combustors (NESCAUM, 2000). Costs
for installing activated carbon injection for municipal waste
combustors range from $0.4 to $1.74 million per ton of mercury reduced
($211 to $870 per pound of mercury reduced). However, NESCAUM points
out that ``this working experience with small sources is not directly
transferable to large coal-fired boilers because of their different
flue gas characteristics'' (NESCAUM, 2000). The cost per pound of
mercury removed for this industry with activated carbon injection would
be considerably higher than the estimated cost for a utility boiler
because the capital and fixed operating costs would be similar while
these plants have very low mercury emissions. The high cost, small
reduction in HAP emissions, increased energy usage, and additional
waste generation do not justify this beyond-the-floor alternative at
this time.
The corona discharge, electro-catalytic oxidation, and copper-
coated magnetic taconite concentrate injection control technologies are
describe by the commenter as ``emerging technologies * * * that could
potentially be applied to the taconite sector as they mature and become
more cost-effective.'' Based on the commenter's own description, these
technologies are not currently ready for application to the taconite
industry. Therefore, these technologies were not considered in the
beyond-the-floor analysis.
In evaluating these potential beyond-the-floor options, we were
unable to identify any viable control technologies or operating
practices for achieving reductions in mercury emissions from taconite
iron ore plants. Consequently, we chose the floor level of no emissions
reduction as MACT.
Since specific controls for mercury are not currently present in
the industry and operating practices that effectively reduce mercury
emissions have not been identified, we are selecting no emissions
reduction as new source MACT.
Asbestos
Comment: Seventeen commenters stated that EPA should set a limit
for asbestos emissions from taconite plants as is required by the CAA.
One commenter stated that asbestos is designated as a HAP by the CAA.
The commenter reasoned that if asbestos is emitted by the taconite
industry, the statute requires that EPA set a standard for asbestos
fibers. Based on the decision in Reserve Mining Co. v. EPA, 514 f.2d
492, 526 (1975), the commenter contends that the EPA must consider
asbestos to be a HAP emitted by the taconite industry. One commenter
contended that ``lack of information'' about asbestos emissions is an
invalid reason for not setting standards.
Two commenters asserted that 30 years ago, EPA stated that it
intended to regulate asbestos emissions from the taconite industry. The
same commenter stated that the 1973 asbestos NESHAP had excluded
``mineral processing operations that may contain asbestos as a
contaminant.'' The commenter further pointed out the Congress rejected
this approach when it passed the CAA Amendments of 1990.
One of the commenters pointed out that in a 1975 Reserve Mining
decision, the U.S. Court of Appeals for the Eighth Circuit stated in
regard to emissions from the Co. plant (now operated by Northshore)
that ``Reserve discharges fibers substantially identical and in some
instances identical to fibers of amosite asbestos.'' The trial court
heard extensive evidence as to the chemistry, crystallography, and
morphology of the cummingtonite-grunerite present in the mined ore.
This evidence demonstrated that, at the level of the individual fiber,
a portion of Reserve's cummingtonite-grunerite cannot be meaningfully
distinguished from amosite asbestos. Reserve attempted to rebut this
testimony by showing that the gross morphology of the two minerals
differed and the characteristics of the two minerals varied when
considered in crystal aggregations. Since, according to the opinions of
some experts, the individual fiber probably serves as a carcinogenic
agent, the district court viewed the variations in mineralogy as
irrelevant and determined that Reserve discharges fibers substantially
identical and in some instances identical to amosite asbestos.
One commenter stated that it should be noted in the proposal
preamble that only one mine remains operating at the
[[Page 61881]]
eastern end of the Mesabi Range where acicular (needle-like) minerals
may be present in the ore. The commenter also stated that the proposal
preamble overstated the efforts of EPA's work group investigation of
asbestos in taconite ore. The commenter asserted that the work group is
focused mainly on vermiculite and is unlikely to study or recommend
``solutions'' for the taconite industry.
One commenter stated that EPA's refusal to set beyond-the-floor
standards for asbestos is unlawful.
Response: Although we are compelled to develop MACT standards for
HAP from major sources, and ``asbestos'' is listed as a HAP in section
112(b) of the CAA, ``asbestos'' is not a single chemical substance or
an easily identified group of chemicals or substances. Our previous
regulatory experience with asbestos as an air pollutant has been
limited to those substances commercially used for their properties,
such as a high resistance to heat and most chemicals. More recently,
the Agency has become concerned with those and similar substances that
may occur as a contaminant in other mined materials and then be
released into the air during processing activities.
When Congress listed ``asbestos'' as a HAP in section 112(b)(1), it
did not further explain the term in the statute, and EPA is not aware
of any legislative history addressing the term asbestos. Currently, EPA
regulatory definitions for ``asbestos'' are provided in the Asbestos
NESHAP, as revised in 1990 (40 CFR 61.141, subpart M), and the
regulations for addressing asbestos-containing materials in schools (40
CFR 763.83). Both rulemakings, which focus on commercial asbestos,
define asbestos as the asbestiform varieties of six different minerals:
chrysotile (serpentinite), crocidolite (riebeckite), amosite
(cummingtonite-grunerite), anthophyllite, actinolite, and tremolite. As
some commenters have indicated, it is correct that the ore from the
eastern end of the Mesabi Range is comprised to some extent of
cummingtonite-grunerite and ferroactinolite (an iron-based form of
actinolite), two of the above listed asbestos-like minerals.
Similarly, other Federal agencies' standards for ``asbestos,'' for
example, the Occupational Safety and Health Administration (OSHA), were
developed for commercial asbestos products and not asbestos as a
contaminant in another material (29 CFR parts 1910, 1915, and 1926).
Current OSHA workplace air regulations apply only to chrysotile,
crocidolite, amosite, and the asbestiform varieties of anthophyllite,
tremolite, and actinolite. The word asbestos is often added after the
mineral (e.g., tremolite asbestos) to signify that the asbestiform
variety of the mineral is being referred to. This is not necessary for
chrysotile, crocidolite, or amosite because these are terms specific to
the asbestiform varieties of the minerals (which are serpentine,
riebeckite, and cummintonite-grunerite, respectively).
Since the EPA first regulated asbestos as a HAP, a distinction has
been made on applying the term asbestos to commercially manufactured
products and not as a contaminant in other materials. When the Asbestos
NESHAP was promulgated in 1973, the EPA Administrator made explicit in
accompanying comments that the NESHAP only apply to asbestos mines and
asbestos mills. Approximately 1 year after the rule was promulgated,
EPA further clarified the rule by stating it does not apply to asbestos
occurring as a contaminant as distinguished from asbestos as a product
(39 FR 15397, May 3, 1974). In a 1974 revision to the Asbestos NESHAP,
the Administrator added a definition of ``commercial asbestos'' to
distinguish asbestos which is produced as a product from asbestos which
occurs as a contaminant in other materials.
Furthermore, when the CAA was amended in 1990, EPA's approach in
developing NESHAP was significantly altered through the use of the HAP
list under section 112(b) and the application of technology-based
standards under section 112(d) instead of a strict risk-based approach.
However, the CAA amendments in 1990 did not provide any further
guidance on how the definition of asbestos could be applied beyond its
use in the Asbestos NESHAP to address asbestos as a contaminant in
other materials.\1\ Based on EPA's historical use of the term
``asbestos,'' it has been used in the context for commercially produced
products and not, as yet, as a contaminant in other products. In
summary, there is no technical or regulatory consensus on the set of
minerals pertinent to contaminant asbestos.
---------------------------------------------------------------------------
\1\ We thus disagree with the commenter who stated, without
citation, that the 1990 amendments to the CAA were intended to
compel section 112(d) standards to control the fibers emitted from
non-commercial sources. The commenter is correct in that section 112
is not limited to commercial asbestos emissions, but nothing in the
statute or its legislative history of which EPA is aware indicate
that Congress intended a particular meaning of ``asbestos'' or that
particular fiber-emitting sources be regulated under section 112 by
virtue of the inclusion of ``asbestos'' in the list of HAP.
---------------------------------------------------------------------------
Notwithstanding the real technical uncertainties as to how to
classify the fibers in the Northshore emissions, commenters argued that
the issue had already been decided by virtue of the Eighth Circuit's
Reserve Mining decision, which found that Reserve Mining (now
Northshore) emitted asbestos for purposes of ordering injunctive
relief. First, any suggestion that EPA is now precluded from making a
different factual determination is not correct. The issue decided in
Reserve Mining is different from the one involved here: whether the
Northshore fibers are ``asbestos'' for purposes of section 112 (b) of
the CAA, a provision not at issue in Reserve Mining since it did not
even exist at the time of the decision.
Second, EPA is not acting in the context of a plea for general
injunctive relief (as in Reserve Mining), but rather to implement a
limited grant of statutory authority to regulate the HAP ``asbestos.''
We have looked for existing, objective means of determining if
Northshore's fibers are ``asbestos'' and currently find the situation
uncertain. In light of this uncertainty, we are not establishing MACT
standards for the fibers emitted by Northshore. Rather, the issue of
which non-commercial fibers are ``asbestos'' for purposes of section
112(b) is one that must first be decided in a broader context.
In response to the events surrounding exposures of residents to
asbestos that occurred as a contaminant in a vermiculite mine in Libby,
Montana, EPA is currently studying the complex issues involved with
asbestos emissions from beneficiation and subsequent processing of
minerals where asbestos may be present as a contaminant. One component
of this activity is a comprehensive update to the asbestos entry in the
Agency's Integrated Risk Information System (IRIS). In the hazard and
dose-response assessment pieces of the update, the current information
on mineralogy, size, bioactivity and chemistry of different asbestos
fibers is being considered. Within the past 3 years, the Agency has
sponsored or co-sponsored several technical meetings aimed at bringing
together the current knowledge on asbestos, its characteristics and
related health effects. These include, but are not limited to:
[sbull] May 24-25, 2001, ``Asbestos Health Effects Conference'' in
Oakland, California;
[sbull] February 25-27, 2003, ``Asbestos Cancer Risk Peer
Consultation'' in San Francisco, California; and
[sbull] June 12-13, 2003, ``Asbestos Mechanisms of Toxicity
Workshop'' in Chicago, Illinois. Integration of the information
gathered through these and other mechanisms will compose the
[[Page 61882]]
support documents for the new IRIS file and will assist us in
decisionmaking regarding contaminant asbestos.
As part of the response to the findings in Libby, the Agency has
developed an action plan which identifies steps necessary to gather the
information needed to decide whether regulations for sources of
contaminant asbestos emissions are warranted. The action plan specifies
vermiculite mining and processing operations as the first area of
focus. Contrary to one commenter's assertion, the action plan also
includes plans to assess emissions, exposure and risk associated with
asbestos that occurs as a contaminant from other mining and processing
operations, including taconite ore mining and processing. That
assessment will inform decisions on specific risk-based regulation of
asbestos that occurs as a contaminant in taconite ore mining and
processing. Specific risk-based emission limitations for asbestos are
not included in the technology-based final rule.
In addition, an International Fiber Symposium was held in St. Paul,
MN in April 2003. The papers presented at the symposium are in a peer-
review process and will then be published. Once the proceedings are
published, the Minnesota Department of Health (MDH) will determine if
they can conduct a risk assessment for fibers or if they can draw any
conclusions about the potential health impacts from fibers. Based on
MDH's findings, the MPCA and Minnesota Department of Natural Resources
may make policy changes with respect to fibers. Until then, MPCA will
continue to regulate airborne fibers from Northshore as required by the
court who deemed the fibers a health concern.
Finally, we note that Northshore is in fact controlling emissions
of its fibers in part with baghouses, which are the optimum control
technology for air emission of fibers (a point made, among other
places, in the Reserve Mining decision itself). Since the Reserve
Mining decision, ambient air monitoring around the plant has
demonstrated a significant reduction in fiber emissions through the
installation of high efficiency baghouses on ore crushing and handling
emission units and wet ESP on the indurating furnace exhaust stacks.
Baghouses are not a control option for indurating furnaces due to the
high moisture content (10 to 15 percent) in the exhaust gases. The high
moisture content causes PM to cake and plug the filtering material
causing filters to be ineffective. In addition, further reductions in
fiber emissions are expected through compliance with the PM emission
standards in the final rule. Representatives at Northshore have
indicated that existing emission units equipped with multiclones are
likely to be replaced with more efficient PM control devices in order
to comply with the PM emission standards in the final rule. Northshore
representatives provided us with the estimated costs for such an
equipment upgrade, and these control costs are reflected in our revised
cost impacts for the final rule.
Formaldehyde
Comment: One commenter stated that EPA has a statutory obligation
to set emission standards for formaldehyde. The commenter asserted that
the standard for formaldehyde must be at least as stringent as the
average formaldehyde emission level of the five best performing plants.
The commenter stated that whether or not there are feasible control
technologies for formaldehyde is irrelevant.
Response: As EPA stated at proposal, formaldehyde (and other
organic HAP) are emitted in very low concentrations by taconite
processing indurating furnaces, not because these organic HAP are
contained in feed or fuel input to the process, but rather as products
of incomplete combustion (PIC) necessarily generated when fossil fuels
are burned (in any type of process, not just in indurating furnaces)
(67 FR 77570). Formaldehyde from indurating furnace emissions has been
measured through stack testing at concentrations that are typically
less than 1 part per million (ppm).
The EPA stated somewhat inaccurately at proposal that formaldehyde
emissions from indurating furnaces are currently uncontrolled. It is
clear from context that we meant that there are no current ``at-the-
stack'' controls for formaldehyde (and other PIC) emissions from these
furnaces, although control of the combustion process minimizes PIC
(including formaldehyde) formation and hence PIC emissions. We
reiterate that at-the-stack controls in place to control PM emissions
have no effect on PIC emissions. We also know of no feasible at-the-
stack control technology for reducing formaldehyde emissions at these
extremely low concentrations and at the exhaust gas temperatures
typically encountered at indurating furnaces.
The only known technology for the control of formaldehyde emissions
at concentrations of less than 1 ppm is thermal catalytic oxidation, in
which formaldehyde is contacted with a precious metal catalyst in the
presence of oxygen and high temperature (650 to 1,350 [deg]F) to yield
carbon dioxide and water. Destruction efficiencies of 85 to 90 percent
have been demonstrated on formaldehyde emissions contained in the
exhaust gas from stationary combustion turbines at concentrations in
the parts per billion range and temperatures of 1,000 [deg]F or higher.
Destruction efficiencies, however, decrease exponentially at reaction
temperatures below 650 [deg]F, reaching less than 10 percent at exhaust
gas temperatures of 300 [deg]F or lower, which is typical of most
indurating furnaces. Burning large quantities of additional fuel, such
as natural gas, to heat the exhaust gases to the desired temperature
would generate large additional quantities of carbon dioxide (a gas
potentially connected to global climate change) and NOX
(ozone precursors). As at proposal, given the significant issues of
technical feasibility and adverse environmental impacts associated with
use of this technology, it is not the proper basis for MACT standards
(67 FR 77571).
We also reiterate that fuel switching is not a justifiable means of
control. Most indurating furnaces currently utilize natural gas as a
fuel, and PIC emissions are higher for natural gas than for coal, but
switching to coal would increase emissions of HAP metals in much larger
amounts than the minimal PIC emissions attributable to natural gas
burning. See S. Rep. 101-228, 101st Cong. 1st sess. at 168 (``In cases
where control strategies for two or more different pollutants are in
actual conflict, the Administrator shall apply the same principle--
maximum protection of human health shall be the objective test.'')
Consequently, the only form of control currently used and feasible
to minimize formaldehyde emissions is the proper and efficient
operation of an indurating furnace with GCP. It is clear from the low
measured levels of formaldehyde emitted from these furnaces that this
means of control is highly effective.
In general, good efficiency of a combustion device is governed by
time, temperature, and turbulence, the three ``T's'' of combustion.
Efficient combustion is achieved when a selected fuel reaches an
optimum temperature for a minimum residence time with sufficient
turbulence to allow oxidation of all organic compounds to completely
react to the products of combustion--water and carbon dioxide. However,
there are many phenomena associated with combustion that lead to the
formation of PIC. Examples of possible phenomena include: Unburned
fuel, quenches or cool zones in the combustion area, fuel rich zones,
low
[[Page 61883]]
combustion temperatures, insufficient air (oxygen) contact with fuel
due to limited turbulence, and changes to the combustion process due to
load swings or feed changes.
Good combustion practices typically encompass several elements such
as the proper operation of the combustion process, routine inspection
and performance analysis of the process, and preventative maintenance.
More specific examples of GCP indicating the range of existing
practices are listed below:
[sbull] Maintain operator logs;
[sbull] Develop procedures for startup, shutdown, and malfunction;
[sbull] Perform periodic evaluations or inspections;
[sbull] Perform burner or control adjustments/tune-ups;
[sbull] Monitor and maintain concentrations of carbon monoxide
(CO), oxygen (O2), or carbon dioxide (CO2) in
compliance with site-specific concentration limits in the combustion
exhaust;
[sbull] Monitor and maintain combustion temperatures above a site-
specific minimum value;
[sbull] Monitor fuel/air metering;
[sbull] Comply with a CO or total organic carbon (TOC) emission
limit;
[sbull] Maintain proper liquid fuel atomization;
[sbull] Monitor fuel quality and handling procedures;
[sbull] Maintain combustion air distribution; and
[sbull] Maintain fuel dispersion.
Although all indurating furnaces need to use GCP to minimize PIC
emissions, determining what precisely is GCP involves site-specific
determinations for each furnace. For example, some indurating furnaces
have been required to install NOX emission controls such as
low NOX burners. The basic method used in reducing
NOX emissions is a reduction in combustion temperature,
which is the opposite strategy needed for minimizing PIC (i.e.,
increasing combustion temperature). Thus, due to differences in furnace
design, operation, firing fuel, process controls, and air pollution
control equipment, one set of GCP established for one type of
indurating furnace may be different from those needed for another type
of indurating furnace.
In addition, State operating permits for the taconite indurating
furnaces do not require any specific set of GCP. However, based on
discussions held with industry representatives, all sources already use
a wide variety of work practices (e.g., existing Standard Operating
Procedures) to maintain proper and efficient operation of each
indurating furnace. See the July 11, 2003 memorandum, ``Meeting Minutes
on Good Combustion Practices with Taconite Industry Representatives.''
Sources have a strong and inherent economic incentive to ensure that
fuel is not wasted, and that the combustion device operates properly
and is appropriately maintained. The lack of a uniform approach to
assuring combustion efficiency is not surprising given the differences
of indurating furnace designs, and the fact that existing Federal/State
standards do not include GCP requirements for indurating furnaces.
Thus, we have determined that site-specific GCP are the MACT floor
for formaldehyde emissions from existing sources. In evaluating
potential beyond-the-floor options, we considered the only known at-
the-stack technology for the control of formaldehyde emissions at
concentrations of less than 1 ppm--thermal catalytic oxidation, which
was described earlier. However, as discussed previously, given the
significant issues of technical feasibility (e.g., low exhaust gas
temperatures, high volumetric flow rates of exhaust gas, and low
concentrations of formaldehyde), adverse environmental impacts in the
form of increased energy use, and the tremendous additional cost
associated with use of this technology, we determined that a standard
based on use of thermal catalytic oxidation was not a viable beyond-
the-floor option. Since there is no other form of emission control or
work practice to control formaldehyde emissions from indurating
furnaces, the site-specific GCP documented in the operation and
maintenance plan were also determined as the MACT floor for
formaldehyde emissions from new indurating furnace sources.
We further find that under CAA section 112(h)(1), it is not
feasible to prescribe or enforce an emission standard for HAP because
at-the-stack controls are not feasible (as explained earlier), and
monitoring parameters related to GCP can only meaningfully result in
minimization of PIC emissions if such monitoring parameters are
quantified on a site-specific basis.
Since it is not possible to identify any uniform requirements or
set of work practices that would meaningfully reflect the use of GCP,
the final rule requires each source to identify site-specific work
practices for each indurating furnace and to document these GCP in an
operation and maintenance plan in accordance with Sec. 63.9600 of the
final rule. A GCP control strategy could include a number of combustion
conditions and work practices which, applied collectively, promote good
combustion performance and minimize the formation of formaldehyde/PIC
emissions. Thus, the MACT requirement for these sources is to use GCP,
and for each source to develop an operation and maintenance plan that
details appropriate operating parameters for each of the following
elements of GCP, or explains why such operating parameters are either
inappropriate or unnecessary for the source (``inappropriate'' or
``unnecessary'' to be determined by the degree to which PIC formation
from fuel combustion in the furnace is minimized):
[sbull] Proper operating conditions for each indurating furnace
(e.g., minimum combustion temperature, maximum CO concentration in the
furnace exhaust gases, burner alignment, or proper fuel-air
distribution/mixing).
[sbull] Routine inspection and preventative maintenance and
corresponding schedules of each indurating furnace.
[sbull] Performance analyses of each indurating furnace.
[sbull] Keeping applicable operator logs.
[sbull] Keeping applicable records to document compliance with each
element.
A source's compliance with its startup, shutdown, and malfunction
plan also will contribute to GCP.
A final determination that the values established in the operation
and maintenance plan are appropriate GCP for the source would then be
achieved by submitting the plan to the Administrator on or before the
compliance date that is specified in Sec. 63.9583 of the final rule
for the affected source. The operation and maintenance plan must
explain why the chosen elements and work practices are considered GCP
for the affected source. The quantified parameters (e.g., furnace
operating temperature) contained in the plan become enforceable
operating conditions unless and until the Administrator acts to
establish new parameters.
The Administrator will evaluate the demonstration and determine
whether the chosen elements and work practices minimize the formation
of formaldehyde (and other PIC) and so constitute GCP for the furnace.
The Administrator will review the adequacy of the site-specific
procedures and the records to demonstrate that the plan constitutes
GCP. If the Administrator determines that any portion of the plan is
not adequate, we can reject those portions of the plan and request
additional information addressing the relevant issues.
[[Page 61884]]
Finally, with respect to the commenter's point that EPA is
obligated to establish MACT standards for formaldehyde, EPA has
established such standards, based on GCP implemented by means of an
operation and maintenance plan and site-specific determinations through
the permitting process, as explained above.
HCl and HF
Comment: One commenter stated that EPA has a clear statutory
obligation to set emission standards for each listed HAP, including HCl
and HF. The commenter asserted that, just because plants are achieving
some incidental control of acid gases, it does not free EPA of its
statutory obligation to set a specific emission limit for HCl and HF.
Two commenters stated that EPA must set a standard for HCl and HF that
reflects, at a minimum, the average emission level achieved by the five
best performing plants. One commenter cited the National Lime opinion
which states ``The CAA requires EPA to set MACT floors upon the average
emission limitation achieved; it nowhere suggests that this achievement
must be the product of specific intent.''
One commenter stated that EPA's rejection of beyond-the-floor
standards for HCl and HF is not logical when a technology is available
and substantially reduces HAP. The commenter contended that available
acid gas control technology would yield a far greater degree of
reduction than is required by EPA's proposed standards, which require
no reduction at all.
Response: Acid gases (HCl and HF) are formed in the indurating
furnace due to the presence of chlorides and fluorides in pellet
additives, such as dolomite and limestone, as well as in the ore
bodies. The taconite industry has not installed equipment specifically
for the purpose of controlling acid gases from indurating furnace
stacks, but, as the commenters correctly note, intent is irrelevant in
determining HAP control (National Lime). What matters is the extent of
control, where control in fact occurs. Test data for HCl and HF
emissions were available from seven indurating furnaces at six taconite
plants. Since most of the furnaces have multiple stacks, these tests
represent emissions from fifteen control devices: 8 venturi scrubbers,
2 multiclones, 3 dry ESP, and 2 wet ESP. These data show that, except
for emissions from stacks controlled with multiclones, HCl and HF are
emitted from indurating furnaces at very low concentrations, typically
less than 3 ppm.
Of the six plants for which HCl and HF test data were available,
three plants conducted PM emissions tests concurrently with the HCl and
HF tests. These tests represent emissions from 3 furnaces and 8
emission control devices: 4 venturi scrubbers, 2 multiclones, and a dry
ESP/wet ESP ducted together. An analysis of the HCl and HF emissions
data and the corresponding PM emissions data indicates that, for this
industry, there is a correlation between acid gas and PM emissions from
control devices on indurating furnaces. Specifically, the data indicate
that stacks with higher PM emissions also have higher acid gas
emissions, and likewise, stacks with lower PM emissions have lower acid
gas emissions (``Correlation of Acid Gas Emissions to PM Emissions for
Taconite Indurating Furnaces,'' July 2003). Consistent with this
correlation, the best performing sources for PM are also the best
performing for acid gas emissions.
There is an engineering basis for this correlation. Due to the
strong affinity of acid gases for water, PM control equipment that uses
water, such as wet scrubbers and wet ESP, has the capability of
reducing HCl and HF emissions substantially. Therefore, wet scrubbers
and wet ESP control technologies used for the reduction of PM emissions
from taconite indurating furnaces to achieve the MACT level of control
for HAP metals are expected to achieve a reduction of acid gas
emissions as well. St |
