United Stales
Environmental Protection
Agency
Office Of Air Quality
Planning And Standards
Research Triangle Park. NC 27711
EPA-453/R-98-010b
December 1998
Air
& EPA
NATIONAL EMISSION STANDARDS
FOR HAZARDOUS AIR POLLUTANTS
(NESHAP) FOR STEEL PICKLING - HC1
PROCESS FACILITIES AND
HYDROCHLORIC ACID
REGENERATION PLANTS -
BACKGROUND INFORMATION FOR
PROMULGATED STANDARDS
-------
Disclaimer
This report has been reviewed by the Emission Standards Division of the Office of Air Quality
Planning and Standards, EPA, and approved for publication. Mention of trade names or
commercial products is not intended to constitute endorsement or recommendation for use.
Copies of this report are available through the Library Services Office (MD-35), U.S.
Environmental Protection Agency, Research Triangle Park, NC 27711, or from National
Technical Information Services, 5285 Port Royal Road, Springfield, VA 22161.
-------
U.S. ENVIRONMENTAL PROTECTION AGENCY
National Emission Standards for Hazardous Air Pollutants for Steel
Pickling HC1 Process Facilities and Hydrochloric Acid Regeneration
Plants -- Background Information for Promulgated Standards
Prepared by:
April 29, 1999
James H. Maysilles (Date)
Metals Group
Emission Standards Division
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
1. The standards will reduce emissions of hydrochloric acid (HC1) from steel pickling
facilities that use the HC1 process and from acid regeneration plants that treat waste
pickling liquor to produce reusable HC1. Chlorine emissions from acid regeneration
plants will also be reduced. The rules establish final HC1 control requirements for steel
pickling lines, hydrochloric acid regeneration plants, and hydrochloric acid storage
vessels. The final standards are promulgated under the authority of Section 112(d) of
the Clean Air Act.
2. Copies of this document have been sent to the following Federal Departments:
Agriculture, Commerce, Defense, Energy, Health and Human Services, Interior,
Labor, and Transportation; the Office of Management and Budget; the National Science
Foundation; and other interested parties.
3. Copies of this document may be obtained from:
National Technical Information Services
5285 Port Royal Road
Springfield, Virginia 22161
4. For additional information contact:
Mr. James Maysilles
Emission Standards Division (MD-13)
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
111
-------
This page is intentionally left blank.
-------
Table of Contents
PAGE
1.0 INTRODUCTION 1-1
1.1 SUMMARY OF RULE CHANGES SINCE PROPOSAL 1-1
1.1.1 Applicability 1-1
1.1.2 Definitions 1-1
1.1.3 Emission Standards 1-2
1.1.4 Operations and Maintenance 1-3
1.1.5 Performance Testing 1-4
1.1.6 Monitoring Requirements 1-4
1.1.7 Delegation of Authority 1-5
1.2 SUMMARY OF FINAL RULE IMPACTS 1-5
2.0 COMMENTERS ON PROPOSED RULE 2-1
3.0 APPLICABILITY 3-1
3.1 CLARIFICATION OF THE MAJOR SOURCE PROVISION 3-1
3.2 CLARIFICATION OF THE 50 PERCENT HC1 PROVISION ... .3-1
3.3 MULTI-LINE AND MULTI-ACID OPERATIONS 3-2
3.4 HYDROCHLORIC ACID STORAGE VESSELS 3-3
3.5 DEFINITION OF NEW OR RECONSTRUCTED SOURCE 3-4
4.0 DEFINITIONS 4-1
4.1 BATCH AND CONTINUOUS PICKLING LINES 4-1
4.2 STEEL PICKLING 4-1
5.0 STANDARDS FOR PICKLING LINES 5-1
5.1 NUMERICAL STANDARDS 5-1
5.2 FUME CAPTURE CONSIDERATIONS 5-12
6.0 STANDARDS FOR ACID REGENERATION PLANTS 6-1
6.1 BASIS FOR THE HC1 NUMERICAL STANDARD 6-1
6.2 BASIS FOR THE C12 NUMERICAL STANDARD 6-5
7.0 STANDARDS FOR ACID STORAGE VESSELS 7-1
7.1 INCLUSION OF STANDARDS FOR ACID STORAGE VESSELS . . . 7-1
7.2 FLEXIBILITY IN CHOICE OF CONTROL SYSTEMS 7-1
8.0 MAINTENANCE REQUIREMENTS 8-1
8.1 ESTABLISHMENT OF NECESSITY FOR REQUIREMENTS 8-1
8.2 SPECIFICITY OF REQUIREMENTS 8-1
8.3 TIMING OF RESPONSE TO OPERATING PARAMETER
EXCURSIONS 8-3
-------
Table of Contents (Continued)
PAGE
9.0 PERFORMANCE TESTING 9-1
9.1 ESTABLISHMENT OF MONITORING PARAMETERS 9-1
9.2 OPERATION OF PROCESS DURING TESTING ... 9-2
10.0 TEST METHODS . . 10-1
10.1 APPLICABILITY OF PROPOSED METHODS . ..10-1
11.0 MONITORING REQUIREMENTS . . 11-1
11.1 CEMS REQUIREMENT . 11-1
11.2 SCRUBBER MONITORING PARAMETERS . 11-2
11.3 MONITORING DURING PROCESS STOPPAGE .11-4
11.4 ACID STORAGE VESSELS 11-5
11.5 PROCESS MONITORING - ACID REGENERATION PLANTS . 11-5
11.6 FREQUENCY OF PERFORMANCE TESTING 11-5
11.7 FREQUENCY OF MONITORING DEVICE CALIBRATION 11-6
11.8 VIOLATIONS 11-6
12.0 REPORTING REQUIREMENTS . .12-1
12.1 CERTIFICATION OF REPORTS . 12-1
12.2 COMBINING REPORTS . . . .... 12-1
13.0 RECORDKEEPING REQUIREMENTS .. . 13-1
13.1 USEFULNESS OF INFORMATION IN RECORDS . . 13-1
14.0 MISCELLANEOUS COMMENTS 14-1
14.1 POLLUTION PREVENTION MEASURES 14-1
14.2 COST OF CONTROLS . 14-1
VI
-------
Tables
PAGE
2-1 List of Public Commenters on the Proposed NESHAP
for Steel Pickling Facilities HC1 Process 2-3
5-1 Pickling line scrubber outlet HC1 concentration data
from one facility 5-3
5-2a Pickling line configuration and scrubber design data
from EPA's original data base 5-5
5-2b Pickling line scrubber inlet/outlet HC1 data
from EPA's original data base 5-6
5-3 Summary of emission data for new and existing source MACT 5-8
6-1 Test data for HC1 emissions from acid regeneration plants
using two-stage scrubbing, water in first stage 6-2
6-2 Test data for HC1 emissions from acid regeneration plant
using single-stage water scrubbing 6-4
6-3 Test data for C12 emissions from acid regeneration plant 6-6
vn
-------
This page is intentionally left blank.
-------
1.0 INTRODUCTION
1.1 SUMMARY OF RULE CHANGES SINCE PROPOSAL
The EPA proposed the rule National Emission Standards for Hazardous Air Pollutants
for Steel Pickling Facilities HC1 Process (40 CFR Part 63) on September 18, 1997 (62 FR
49051). Based on the public comments received by the EPA at the proposal as well as the
EPA's evaluation of additional information obtained after proposal, certain requirements of the
rule have been changed from those proposed. -The major changes affect provisions
establishing the rule applicability, definitions, emission standards, operations and maintenance,
performance testing, and monitoring. In addition, the EPA has made changes to the specific
regulatory text to clarify the EPA's intent in the application and implementation of the rule
requirements. The substantive changes to the rule since proposal are summarized as follows.
1.1.1 Applicability
The promulgated rule clarifies that subject steel pickling sources include only those
pickling hot-rolled or hot-formed carbon steel. Additionally, the acid solution used for
pickling must contain 6 percent or more by weight of HC1 and be at a temperature of 100 °F
or higher. For the purposes of this rule, steel pickling is limited to hydrochloric acid pickling
of carbon steel.
1.1.2 Definitions
The title acid regeneration plant is changed to hydrochloric acid regeneration plant to
clarify the applicability of the rule.
The title acid storage tank is changed to hydrochloric acid storage vessel to clarify the
applicability of the rule. The definition is changed to apply only to a stationary vessel, not a
temporary or mobile vessel, that is used for the bulk containment of virgin or regenerated
hydrochloric acid. The term "vessel" rather than "tank" is used for bulk storage containers in
order to be consistent with terminology used in other subparts of part 63 to define containers
1-1
-------
that are used for chemical storage. Similarly, the term "tank" is used for containers that are
integral parts of processes, such as acid baths used in pickling lines.
Carbon steel is defined as steel that contains approximately 2 percent or less carbon,
1.65 percent or less manganese, 0.6 percent or less silicon, and 0.6 percent or less copper.
The definition of closed-vent system is clarified to include a system containing any
device that is capable of reducing or collecting emissions.
A definition for hydrochloric acid regeneration plant production mode has been added
to help clarify that monitoring requirements apply only while the plant is operating in a
manner to produce usable regenerated acid and iron oxide.
To ensure that a knowledgeable plant official signs maintenance records and reports,
the term responsible maintenance official has been added and defined as a person designated
by the owner or operator as having the knowledge and authority to sign documents that are
required under the promulgated rule.
Specialty steel, which is excluded from the rule, is defined as a category of steel that
includes silicon electrical, alloy, tool, and stainless steels.
The revised definition of steel pickling emphasizes that it is the chemical removal of
iron oxide mill scale formed on steel surfaces during hot rolling and hot forming operations
prior to shaping or coating of finished steel products. Removal of light rust or scale from
finished steel products and activation of the metal surface prior to plating or coating are
specifically excluded from the definition.
1.1.3 Emission Standards
The proposed rule required existing pickling sources to meet either a minimum HC1
collection efficiency standard of 97.5 percent or, alternatively, a maximum process or control
device outlet HC1 concentration standard of 10 parts per million by volume (ppmv). Standards
for new or reconstructed sources were 99 percent or 3 ppmv, respectively. The standards for
existing pickling sources have been changed to 97 percent or 18 ppmv. For new or
reconstructed sources, the standards are 99 percent or 6 ppmv for continuous lines, 97 percent
or 18 ppmv for batch lines.
1-2
-------
The proposed rule required existing hydrochloric acid regeneration plants to meet
maximum outlet concentration standards of 8 ppmv for HC1 and 4 ppmv for chlorine (C12).
Standards for new or reconstructed sources were 3 ppmv for HC1 and 4 ppmv for C12. The
standards for existing sources are changed to 25 ppmv for HC1 and 6 ppmv for C12. The
standards for new or reconstructed sources are changed to 12 ppmv for HC1 and 6 ppmv for
C12. As in the proposed rule, owners or operators of existing acid regeneration facilities may
determine an alternate C12 emission standard specific for the source if they can show to the
Administrator's satisfaction that the 6 ppmv C12 standard cannot reasonably be met. The
provision in the proposed rule for owners or operators of new or reconstructed hydrochloric
acid regeneration plants to request approval for a source specific C12 concentration standard is
removed. Upon reconsideration, this provision is not consistent with the statutory requirement
that all new sources are to achieve the new source MACT numerical limit. The expectation is
that owners and operators are to design and construct new sources capable of meeting the
standard.
1.1.4 Operations and Maintenance
The revised rule requires an owner or operator of a pickling facility or acid
regeneration plant to write an emission control device operation and maintenance plan that is
consistent with good maintenance practices and, for scrubbers, contains at a minimum the
elements described in § 63.1160(b)(2)(i) through § 63.1160(b)(2)(iv) of the final rule. The
plan is no longer required to be submitted to the permitting authority, but it is required to be
incorporated by reference into the source's Title V permit. The rule also requires
hydrochloric acid regeneration plants to be operated in a manner consistent with good air
pollution control practices.
Monitoring the pressure drop across the scrubber is now required as an aid for
detecting changes that indicate a need for maintenance, not as an operating parameter.
When an operating problem is discovered, initiation of procedures to correct the
problem must be begun within one working day. Procedures to be initiated are the applicable
actions that are specified in the maintenance plan. Corrective actions must be completed as
soon as practicable, and all records of repairs and inspections must be signed by a responsible
maintenance official.
1-3
-------
1.1.5 Performance Testing
Under the revised rule, each scrubber control device must be monitored for makeup
water flow rate and, for scrubbers that operate with recirculation, recirculation water flow
rate. The minimum values for these rates must be established during initial or subsequent
performance testing. Compliance with emission standards may be determined by either the
average of three consecutive sampling runs or the average of any three of four consecutive
runs during a test. Compliant operating parameter values for wet scrubbers and acid
regeneration plants may be determined by the average values measured during any of the runs
used to demonstrate compliance rather than the average values measured during the entire
testing period. During testing under Method 1, no traverse point shall be within one inch of
the stack or duct wall. For existing acid regeneration plants wishing to petition for a new C12
emission standard because they are not able to meet the published standard, it must be shown
that a plant cannot meet a concentration standard of 6 ppmv.
1.1.6 Monitoring Requirements
Each monitoring device used to monitor control device parameters must be calibrated
in accordance with the manufacturer's instructions but not less frequently than once per year.
Operation of a scrubber control device with makeup water and recirculation water flow rates
less than the values established during initial or subsequent performance testing requires
initiation of corrective action as specified in the facility's maintenance plan. The requirement
to install and operate a continuous emission monitoring system (CEMS) if excursions of the
control device operating parameters are excessive is removed.
The requirement for periodic performance tests is also revised. The final rule requires
that the owner or operator conduct performance tests for each air pollution control device
either annually or on an alternative schedule that is approved by the permitting authority, but
no less frequently than every 2l/2 years or twice per title V permit term.
The proposed rule inadvertently stated that exceedances of scrubber operating
parameters were violations of the emission limit. The intention was to state that exceedances
of acid regeneration plant operating parameters were violations of the emission limit. This
requirement has been changed so that exceedances of scrubber operating parameters only
require initiation of corrective action according to the maintenance plan, and exceedances of
acid regeneration plant operating parameters are not violations of the emission limit but instead
are violations of the operational standard.
1-4
-------
Monitoring requirements for acid storage vessels are revised. The definition of closed-
vent system now includes provisions to transport emissions back into any device that is capable
of reducing or collecting the emissions. Under the final rule, the owner or operator must
make semiannual instead of monthly inspections of each vessel to ensure proper operation of
the closed-vent system and either the air pollution control device or enclosed loading and
unloading line, whichever is applicable.
1.1.7 Delegation of Authority
The proposed rule specified that authority for approval of an alternative test method
and alternative nonopacity emission standards would be retained by the Administrator and not
transferred to a State. Authority for approval of monitoring parameters for hydrochloric acid
regeneration plants and alternative monitoring requirements for wet scrubbers is also retained
by the Administrator because these parameters are fundamental to effective monitoring and
cannot be delegated. The Administrator will also retain authority to waive recordkeeping
requirements. Authority to approve an alternative performance testing schedule is delegated to
the States.
1.2 SUMMARY OF FINAL RULE IMPACTS
The EPA estimates that implementation of the subpart CCC standards will reduce
nationwide HC1 emissions from steel pickling operations and acid regeneration plants by
approximately 2,200 Mg/yr. The total nationwide capital investment cost to pickling facility
and acid regeneration plant owners to implement the subpart CCC standards is estimated by
the EPA to be $20 million. The total nationwide annual cost for these standards is estimated
to be $4.9 million per year.
The EPA concludes that the promulgation of the final subpart CCC standards will not
have a significant economic impact on steel pickling facility or acid regeneration plant owners
and operators. Model plant costs represent the cost to individual facilities, and these costs
approximate 1.0 percent or are less than 1.0 percent of sales and of the cost of production for
all model plants. The magnitude of the costs relative to production costs of the industry and
sales revenues leads to a conclusion that this standard will not significantly adversely impact
firms in the steel pickling industry. No plant closures are anticipated nor significant
employment losses. Significant regional impacts are also not expected.
1-5
-------
2.0 COMMENTERS ON PROPOSED RULE
The U.S. Environmental Protection Agency proposed standards on September 18, 1997
under the authority of Section 112 of the Clean Air Act as amended in November 1990 that
would control air emissions from steel pickling lines that employ the HC1 process and
associated HC1 acid regeneration plants. The preamble to the proposed rule discussed the
availability of a background information document (BID) (EPA-453/R-97-012)1 that presents
information used in development of the proposed rule.
Comments from the public on the rulemaking were solicited at the time of proposal,
and copies of the Federal Register notice and the BID for the proposed rule were distributed to
interested parties. A 60-day comment period from September 18, 1997, to November 17,
1997, was provided to accept written comments from the public on the proposed rule and BID.
The opportunity for a public hearing was provided to allow interested persons to present oral
comments on the rulemaking. However, the EPA did not receive a request for a public
hearing, so a public hearing was not held.
A total of 15 letters presenting comments on the proposed rule and on the BID for the
proposed rule were received by the EPA. Copies of the comment letters are available for
public inspection in the docket for the rulemaking (Docket No. A-95-43). This docket is
located at the EPA's Air and Radiation docket and Information Center, Waterside Mall, Room
M-1500, First Floor, 401 M Street, SW, Washington, DC 20460.
The commenters, their affiliations, and the EPA docket number assigned to their
correspondence are presented in Table 2-1. The commenter affiliation distribution for these
letters is as follows: nine individual companies, five trade associations (including
STAPPA/ALAPCO), and one Federal agency.
1 National Emission Standards for Hazardous Air Pollutants (NESHAP) for Steel
Pickling HC1 Process — Background Information for Proposed Standards,
Publication Number EPA-453/R-97-012, June 1997.
2-1
-------
The primary purpose of this document is to present the EPA's responses to the
comments on the proposed rulemaking. Many of the comment letters contain multiple
comments regarding various aspects of the rulemaking. For the purpose of orderly
presentation, the comments are categorized by the following topics:
• Chapter 3.0 Applicability
• Chapter 4.0 Definitions
• Chapter 5.0 Standards for Pickling Lines
• Chapter 6.0 Standards for Acid Regeneration Plants
• Chapter 7.0 Standards for Acid Storage Vessels
• Chapter 8.0 Maintenance Requirements
• Chapter 9.0 Performance Testing
• Chapter 10.0 Test Methods
• Chapter 11.0 Monitoring Requirements
• Chapter 12.0 Reporting Requirements
• Chapter 13.0 Recordkeeping Requirements
• Chapter 14.0 Miscellaneous Comments
2-2
-------
Table 2-1. List of Public Commenters on the Proposed NESHAP for Steel Pickling Facilities - HC1 Process
Docket item
number
Commenter name and address
Docket item
number
Commenter name and address
to
IV-D-1 James R. Malloch
Vice President-Iron Oxide
Magnetics International Incorporated
1111 North State Road 149
Burns Harbor, Indiana 46304-9249
IV-D-2 Matthew F Johnston
Corporate Engineer, Environmental Affairs
Worthington Industries
1205 Dearborn Drive
Columbus, Ohio 43085
IV-D-3 David A. Sulc
Environmental Engineer
Nucor Steel
Division of Nucor Corporation
Route 2, Box 311
Crawfordsville, Indiana 47933-9450
IV-D-4 Michael J. Wax
Deputy Director
Institute of Clean Air Companies
1660 L Street, NW
Suite 1100
Washington, DC 20036-5603
IV-D-5 Sarah B. Peirce-Sandner
Kodak Park Environmental Services
Health, Safety, and Environment
Eastman Kodak Company
Rochester, New York 14652-6263
IV-D-6 Arthur E. Hall, P.E.
Supervisor, Chemical Processing
Wheatland Tube Company
1 Council Avenue
Wheatland, Pennsylvania 16161
IV-D-7 John L. Wittenborn
Chet M. Thompson
Counsel to the Specialty Steel Industry of North
America
Collier, Shannon, Rill, and Scott, PLLC
3050 K Street, NW
Suite 400
Washington, DC 20007
IV-D-8 Walter Sieckman
Chief Executive Officer
International Steel Services, Incorporated
Foster Plaza No. 7
661 Andersen Drive
Pittsburgh, Pennsylvania 15220
(Continued)
-------
Docket item
number
Commenter name and address
Docket item
number
Commenter name and address
to
IV-D-9 John L. Wittenborn
Chet M. Thompson
Counsel to the Steel Manufacturers Association
Collier, Shannon, Rill, and Scott, PLLC
3050 K Street, NW
Suite 400
Washington, DC 20007
IV-D-10 David J. Holmberg
Manager - Environmental Services
Acme Steel Company
13500 S. Perry Avenue
Riverdale, Illinois 60627-1182
IV-D-11 Bruce A. Steiner
Vice President, Environment and Energy
American Iron and Steel Institute
1101 17th Street, NW
Suite 1300
Washington, DC 20036-4700
IV-D-12 Gary R. Allie
Environmental Regulation Manager
Environmental Affairs
Inland Steel Company
Indiana Harbor Works
3210 Watling Street
East Chicago, Indiana 46312
IV-D-13 Felicia R. George
Assistant Commissioner
Indiana Department of Environmental Management
100 North Senate Avenue
P.O. Box 6015
Indianapolis, Indiana 46206-6015
IV-D-14 Neil Stone
ESCO Engineering
179 Lansdowne Avenue
Kmgsville, Ontario, CANADA N9Y 3J2
IV-D-15 Robert H. Colby, Chair
ALAPCO Air Toxics Committee
Bliss M. Higgins, Chair
STAPPA Air Toxics Committee
STAPPA/ALAPCO
444 North Capitol Street, NW
Washington, DC 20001
-------
3.0 APPLICABILITY
3.1 CLARIFICATION OF THE MAJOR SOURCE PROVISION
Comment: Four commenters (IV-D-1, IV-D-2, IV-D-11, IV-D-12) stated that the rule
should be clarified to reflect the fact that it applies only to facilities or parts of facilities that
are major sources of hazardous air pollutants (HAP). It does not apply to major sources of
criteria pollutants or to area sources of HAP.
Response: The suggested clarification has been made.
3.2 CLARIFICATION OF THE 50 PERCENT HC1 PROVISION
Comment: Four commenters (IV-D-2, IV-D-7, IV-D-9, IV-D-11) requested
clarification of the 50 percent acid threshold that defines pickling by the HC1 process. They
are uncertain if the 50 percent concentration refers to the acid in the bath or to acid used in a
mixture of acids. They believe the 50-percent threshold may be arbitrary and capricious and
that EPA should develop a de minimis value that excludes rinse tanks. They reason that EPA
excludes wastewaters, therefore a de minimis value must exist. The 50-percent threshold and
the methods for establishing it for specific facilities should be clarified. They ask if all tanks
in a line are included or just the tanks in which acid concentration stays above 50 percent, and
will EPA's rule prevent facilities from adding, e.g., more rinse tanks to bring the average line
concentration below 50 percent acid. The commenters recommend that EPA establish a de
minimis value on a tub-by-tub basis to impact only those tubs that contain 50 percent acid. One
commenter suggests alternative language: "The provisions of this subpart apply to all new and
existing steel pickling facilities using an acid solution in which hydrochloric acid (HC1) is the
only pickling acid, or using a mixture of acids in which HC1 constitutes at least 50 percent by
weight of all acids in the pickling solution." Two other commenters recommend that
applicability determinations be made on a tub-by-tub basis, and that only acid tubs that exceed
the applicable threshold be subject to regulation. Another commenter asserts that the
applicability provisions of the proposed rule can be interpreted to mean that HC1 at
concentrations greater than 50 percent are subject to the standard. This interpretation is
3-1
-------
contrary to EPA's expressed intent. The commenter recommends that the provision be
clarified.
Response: The EPA agrees with the commenters' observations that a great deal of
confusion exists regarding the 50 percent HC1 criterion. In searching for alternative criteria,
the Metals Handbook. Ninth Edition. Volume 5: Surface Cleaning. Finishing, and Coating.
published by the American Society for Metals, 1990, was consulted, and new criteria in terms
of HC1 concentration and bath temperature were identified. Based on these new criteria, the
rule requires control of all ventilated tanks (acid tubs) containing pickling solution that
contains 6 percent or more HC1 (by weight) and has a temperature of 100 °F or higher; these
conditions are the minimum HC1 concentration and temperature at which HC1 pickling is
conducted according to the referenced handbook.
3.3 MULTI-LINE AND MULTI-ACID OPERATIONS
Comment: Two commenters (IV-D-7, IV-D-9) stated that the proposed rule is based in
large part on models that do not depict the entire steel pickling universe. None of EPA's
models analyzed the impacts of the rule on (1) lines that utilize both HC1 acid tubs and other
acid tubs, (2) lines that utilize HC1 in conjunction with another acid in a single tub, (3) non-
HC1 dedicated scrubbers that also handle emissions from HC1 rinse tubs, or (4) facilities that
capture and control emissions from HC1 and non-HCl lines using a common scrubber.
The EPA does not discuss "multiple acid" or "mixed-acid" pickling lines that have
multiple acid tubs, some of which use HC1 and others that use either sulfuric or nitric acid.
The EPA has provided no data indicating whether mixed-acid emissions impact scrubber
efficiency or whether scrubbers that handle mixed-acid lines are capable of meeting the
proposed standards. The EPA should collect and analyze sufficient data to determine the
impact on and whether its proposed standards are appropriate and achievable for these types of
pickling lines.
Scrubbers that are not used primarily to control emissions from HC1 lines and that only
control rinse tub emissions should not be subject to the proposed rule. The EPA could rectify'
the situation by expressly providing in the rule or preamble that non-HCl dedicated scrubbers
that also handle emissions only from HC1 rinse tubs are not subject to the rule, or the EPA
could establish a de minimis HC1 concentration applicability threshold.
3-2
-------
Response: The intent of the rule was to address the segment of the industry that
pickles carbon steel using HC1. After the rule was proposed, the EPA received information on
another segment of the industry that pickles steel using other acids or combinations of HC1 and
other acids. Steel pickled by these processes are specialty steels that include silicon electrical,
alloy, tool, and stainless steels. Information received by the EPA included evidence that major
sources for HC1 may exist within the specialty steel pickling industry segment, but available
source test data did not verify that the emission control technology employed in carbon steel
pickling could be employed with the same effectiveness in specialty steel pickling.
Consequently the EPA decided to defer regulation of the specialty steel pickling industry and
proceed with its original intention to regulate only carbon steel pickling using the HC1 process
by this rulemaking. The applicability statement is revised and definitions for carbon steel and
specialty steel are added to clarify this intent. Carbon steel is defined as steel that contains
approximately 2 percent or less carbon, 1.65 percent or less manganese, 0.6 percent or less
silicon, and 0.6 percent or less copper. Specialty steel is defined as a category of steels that
includes silicon electrical, alloy, tool, and stainless steels. These definitions are taken from
the publication Everything You Always Wanted to Know About Steel - A Glossary of Terms
and Concepts, edited by M. G. Applebaum, Salomon Brothers Inc., Chicago, 1997. The EPA
has also established a de minimis HC1 concentration for operations that are subject to this rule
that will probably exclude most rinse tanks. Thus, the situations described by the commenter
will not be subject to this regulation. The EPA will determine at a later date if the specialty
steel pickling industry should be regulated under this part of the CFR, and, if so, whether it
will be regulated by amending subpart CCC or under a separate subpart.
3.4 HYDROCHLORIC ACID STORAGE VESSELS
Comment: Two commenters (IV-D-7, IV-D-9) recommended that mobile storage tanks
be excluded from the rule; for example, 55-gallon drums and up to 500-gallon plastic
containers on pallets should be excluded.
Response: The EPA agrees that small portable vessels should be excluded from the
rule. Requiring the installation of controls on these sources would be burdensome and was not
intended. The rule is revised to include only stationary storage vessels.
3-3
-------
3.5 DEFINITION OF NEW OR RECONSTRUCTED SOURCE
Comment: One commenter (IV-D-8) believes that the proposed rule will require
reconstruction of existing scrubber systems, forcing the process to become subject to new
source rules. The definition of reconstructed source should be eliminated.
Response: Changes or additions to air pollution control devices do not constitute
reconstruction of the source and are not included in the changes that would make a facility or
process subject to reconstruction and modification requirements.
3-4
-------
4.0 DEFINITIONS
4.1 BATCH AND CONTINUOUS PICKLING LINES
Comment: One commenter (IV-D-12) believes that the definition for batch and
continuous pickling lines should exclude the phrase "... drained, and rinsed by spraying or
immersion in one or more rinse tanks to remove residual acid." This term could be
misconstrued to require an air pollution control device. The commenter also believes that the
definition for continuous pickling line should exclude the phrase "... rinsed in another vessel
or series of vessels to remove residual acids." This term could be misconstrued to require an
air pollution control device.
Response: The rule is meant to include all ventilated tanks that are part of a steel
pickling process to which the rule applies, which may include some rinse tanks. The rule does
not require installation of ventilation systems not previously installed.
4.2 STEEL PICKLING
Comment: Five comments (IV-D-5, IV-D-7, IV-D-9, IV-D-11, IV-D-12) were
submitted regarding the definition of "steel pickling" proposed for the rule. Two commenters
believe that the proposed definition of pickling is too broad and can be interpreted to include
maintenance activities such as preparing surfaces prior to painting, cleaning equipment, and
cleaning parts. One commenter believes that the second sentence of the definition for steel
pickling contradicts the first sentence regarding "removal of rust" and is concerned that the
definition of steel pickling may capture non-manufacturing operations under the rule. This
commenter also proposed language for the definition that would exclude the operations named
above and would substitute "mill scale" for "scale" and "rust" for "light rust" on the basis
that the former term is more accurate in describing the formation of surface material during
the manufacture of steel and the latter term reduces the chance for subjective interpretation of
the thickness of rust layers. This commenter proposed the following definition for steel
pickling: "Chemical removal of iron oxide mill scale through contact of an aqueous solution
of hydrochloric acid with semi-finished steel products where such contact occurs prior to
shaping or coating of the finished steel product. This definition does not include operations
4-1
-------
for the removal of rust and scale from finished steel products and/or equipment or for the
activation of the metal surface prior to plating or painting." Another commenter
recommended that the definition of steel pickling should also exclude continuous annealing,
normalizing, and coating.
Two other commenters recommended that cold pickling lines (those pickling cold
rolled or reduced steel products) should be specifically excluded from the rule. Both
commenters suggested adding the following sentence to § 63.1155: "The provisions of this
subpart do not apply to HC1 pickling lines that remove scale from steel prior to plating or
coating (i.e.. "cold pickling"). "Cold pickling" should be defined as "HC1 pickling of cold
rolled or reduced steel products."
Another commenter stated that surface cleaning using HC1 should be specifically
excluded to avoid interpretation confusion. Include in the definition: " This definition
does not include operations for the removal of light rust, surface cleaning, or for activation of
the metal surface prior to plating, including processes such a annealing, normalizing, coating,
etc."
Response: The EPA agrees that the definition of steel pickling should be crafted to
avoid misinterpretation. The definition is revised so that steel pickling means the chemical
removal of iron oxide mill scale that is formed on steel surfaces during hot rolling or hot
forming of semi-finished steel products through contact with an aqueous solution of acid where
such contact occurs prior to shaping or coating of the finished steel product. This definition
does not include removal of light rust or scale from finished steel products or activation of the
metal surface prior to plating or coating.
4-2
-------
5.0 STANDARDS FOR PICKLING LINES
5.1 NUMERICAL STANDARDS
Comment: Five commenters (IV-D-6, IV-D-7, IV-D-9, IV-D-10, IV-D-11) stated that
the EPA did not base the standards on the best performing 12 percent of sources. The
language in the Act directs the EPA to derive numerical limits for new sources from the best
performing scrubbers for a given option, but EPA used this approach in deriving existing
source standards. The EPA only considered 10 of the 152 existing continuous pickling lines
(seven percent), then used only four of the 10 available data sets and determined the
concentration limit from only two data sets. The EPA has not justified not using all data sets.
The averages of all ten tests, 29.3 ppmv and 97.3 percent, are more representative of the
actual variation in the test data which could be expected for properly controlled sources and
should be the basis for the limits.
Response: As explained in the preamble to the proposed rule, the EPA based MACT
on technology, not on actual emissions. In determining MACT, the EPA considered
alternative approaches for establishing the MACT floor; these include (1) information on State
regulations and/or permit conditions, (2) source test data that characterize actual emissions
discharged by sources, and (3) use of a technology floor and an accompanying demonstrated
achievable emission level that accounts for process and air pollution control device variability.
No Federal air emission standard currently apply to steel pickling or acid regeneration sources,
and existing State standards cannot be directly related to the requirements of this rule.
Applicable test data are only available from 10 of 152 continuous pickling lines. These data
points are too few to establish 12 percent MACT floors based on actual releases. The best 18
points would be required for continuous lines, the best seven for batch lines, and substantially
more data points than 18 and seven, respectively, would required in order to identify those
best points.
By comparison with the limited utility of State regulations and source test data, a
substantial body of information is available on the types, configurations, and operating
conditions of air pollution control devices applied across the industry. This information was
collected through the comprehensive survey by the EPA of known steel pickling facilities that
5-1
-------
was conducted in 1992 through the Information Collection Request (ICR) approved by the
Office of Management and Budget for NESHAP information gathering. This survey produced
substantial information on the design and operation of emission control equipment but little
information on actual emissions. The EPA therefore used the technology floor approach to
establishing MACT for pickling lines. Details of this approach are discussed in the preamble
to the proposed rule.
The characteristics of the scrubbers constituting the existing source and new source
levels of control were determined by evaluating the results of emission tests conducted on units
currently employed in the industry. Ten valid sets of emission test data on scrubbers applied
to continuous strip and push-pull strip pickling lines were collected and evaluated. All tests
were conducted on sieve tray and vertical packed bed scrubbers. Fundamental design
measures of performance for units of these types include the number of trays and depth of
packing. Based on the data analysis, existing source MACT was represented by a sieve tray
scrubber with three trays and vertical packed bed scrubbers with five to 10 feet of packing.
Each scrubber was used in conjunction with either a mesh pad or chevron type droplet
eliminator. New source MACT was represented by a six-plate tray scrubber with a mesh pad
droplet eliminator and a packed bed scrubber with a two-stage droplet eliminator consisting of
a mesh pad unit followed by a chevron unit. Data from pickling lines controlled by devices of
these descriptions were used to represent the capabilities of MACT for this application. The
EPA determined the standards from these data as discussed in the comments and responses
below.
Comment: Two commenters (IV-D-3, IV-D-6) stated that the standards are
unnecessarily stringent in that they do not reflect what is achievable on a continuous basis
considering natural process and control device variations. One commenter states that based on
experience, controls of 97.5 percent efficiency or 10 ppmv can be achieved on long term
averages, but the process variability is typically plus or minus 50 percent in individual
sampling runs. A more reasonable limitation for existing lines would be an average reduction
of 95 percent or an average concentration of 15 ppmv, with a minimum reduction of 90
percent or a maximum concentration of 30 ppmv. New sources should be given similar
considerations. The commenter submitted no data or specific rationale, however, to support
these values. The other commenter stated that the EPA has no assurance that its data are
representative of long term performance. This commenter submitted data showing a wide
variation in HC1 emissions over a three year period from one facility using the same control
technology where no known malfunctions occurred to cause the variation (presented in table
5-1), and also stated that data presented in the EPA Background Information Document also
5-2
-------
illustrate a wide variation in HC1 emissions between and within facilities. Using a statistical
argument based on standard deviations in data, the standard should be at least 15 ppmv for
new sources and 35.8 ppmv for existing sources. One commenter believes that inaccuracies of
the sampling methods do not permit setting an emission standard as low as that proposed.
Table 5-1. Pickling line scrubber outlet HC1 concentration data
from one facility.
Date
(Month/Year)
06/94
10/94
02/95
09/95
03/96
05/96
10/96
03/97
08/97
Average (and range) HC1 concentration, ppmv
10.6
0.4
1.3
90.
4.5
2.9
5.6
2.4
178.
(7.3 12.9)
(0.41 0.48)
(0.95 1.5)
(88 90)
(2.2-6.2)
(2.3 3.1)
(3.7 7.4)
(1.7 2.8)
(72 - 245)
Response: The EPA is not required to use a specific statistical procedure in arriving at
values for emission standards. Regarding the submitted data, the commenter's facility's nine
tests are comprised of seven tests for which all data points, including individual sampling runs,
are within a 13 ppmv concentration limit. The remaining two tests have averages that are
about 19 and 37 times the average of the other seven tests. The EPA believes these two tests
cannot be the result of normal air pollution control device operation during normal process
operation.
Because the facility supplied only scrubber outlet concentration data, no inferences
about scrubber efficiency could be drawn. Test reports and accompanying process data were
5-3
-------
omitted from the comment, which allows for only an incomplete analysis. Although the
facility states that no known malfunctions occurred to cause the high values for two tests, the
EPA believes they represent abnormal conditions in the process or control system that are
atypical of the scrubber's performance. The EPA believes that this view is sustained by the
factors of 19 and 37 differences found from the average of the remaining tests. The two tests
are not representative of a well-performing and well-maintained scrubber system.
It seems clear that, assuming that the testing procedure was valid and barring
malfunctions or obvious errors in operations, the facility's scrubber can meet an annual
compliance test based a concentration standard of 13 ppmv or higher on a continuous and long
term basis and a lower standard most of the time.
Regarding accuracy of sampling, this issue is discussed in Sections 9.0 and 10.0,
PERFORMANCE TESTING and TEST METHODS. The EPA believes that the test method
is sufficiently accurate for the proposed emission standards for new and existing facilities.
Relative to the broad issues of stringency and achievability of the proposed standards,
the EPA agrees with the commenters in that the data used to determine the numerical limits are
sparse and that judgement should be used in considering variations in operations and in test
results. The numerical limit was therefore reexamined. The EPA conducted a thorough
review of all scrubber design and source test data. The original data base used to develop the
proposed standard is reproduced in tables 5-2a and 5-2b. Table 5-2a gives the basic pickle
line configuration and scrubber design data, and table 5-2b gives results for the individual
sampling runs for all tests.
Data from facilities 105 and 132 were used to establish the proposed numerical limits
for new sources; data from facilities 101, 44, 114, and 17 were used to determine the
proposed existing source limits. Data from facilities 115, 40, and 49 (both pickling lines)
were based on fewer than three sampling runs and were therefore not included in these
calculations.
In reconsidering the numerical limit determinations, data from all tests, including those
with only one or two sampling runs, were examined primarily in regard to variability in
individual test run results. The data are summarized in table 5-3. The data are grouped
separately for new and existing source MACT and arranged within those groups in order of
most efficient operation in terms of outlet concentration on the assumption that this option
would the easiest to verify and therefore the most likely design target.
5-4
-------
Table 5-2a. Pickling line configuration and scrubber design data
from EPA's original data base.
Facility
ID no.
105
132
115
101
40
44
114
49
17
49
Type
of line
Continuous
strip
Continuous
strip
Push-pull
strip
Push-pull
strip
Continuous
strip
Continuous
strip,
vertical
Continuous
strip
Continuous
strip
Continuous
strip
Continuous
strip
Heating
mode
Steam
sparging
Internal
heat
exchange
(HX)
External
HX
External
HX
External
HX
External
HX
External
HX
Steam
sparging
Internal
HX
Steam
sparging
Scrubber
type
Sieve
tray
Packed
tower
Sieve
tray
Packed
tower
Packed
tower
Packed
tower
Packed
tower
Packed
tower
Sieve
tray
Packed
tower
Design
features
Six trays
5' of 2"
Tellerettes
Five trays
8.5' of 2"
polypropylene
pall rings
7' of 2. 75"
polypropylene
Tellerettes
8' of 3. 5"
Tri-packs
10' of 3. 5"
Tri-packs
6.4' of 1.5"
to 3" saddles
Three trays
6.4' of 1.5"
to 3" saddles
Demister
10" mesh
pad
Mistmaster®
mesh pad
plus
8.5" chevron
8" chevron
External
12" vane
4" mesh
pad
6" mesh
pad
External
mesh pad
4" mesh
pad
8" chevron
4" mesh
pad
5-5
-------
Table 5-2b. Pickling line scrubber inlet/outlet HC1 data
from EPA's original data base.
Facility Sampling
ID no. run no.
105 1
2
3
Average
132 1
2
3
4
5
6
7
8
9
10
11
Average
115 1
101 1
2
3
4
HC1 collection
efficiency, %
99.96
99.94
99.99
99.96
97.6
99.8
99.3
99.8
99.6
99.8
99.6
99.8
99.5
99.6
99.4
99.5
99.0
97.5
98.3
98.1
98.4
HC1 concentration, ppmv
Inlet
5,960.
5,860.
3,620.
5.150.
204.
190.
197.
275.
240.
323.
323.
318.
265.
249.
239.
257.
697.
1,880.
3,750.
3,600.
4,570.
Outlet
2.4
3.5
0.3
2A
5.9
0.4
1.5
0.7
1.1
0.6
1.6
0.8
1.5
1.2
1.9
L6
7.7
46.
65.
70.
69.
Average
98.1
62,
5-6
-------
Table 5-2b. Pickling line scrubber inlet/outlet HC1 data
from EPA's original data base (continued).
Facility
ID no.
40
44
114
49
17
49
Sampling
run no.
1
2
Average
1
2
3
Average
1
2
3
Average
1
2
Average
1
2
3
Average
1
2
HC1 collection
efficiency, %
96.8
98.8
97.8
98.8
95.9
97.8
97.6
96.8
96.6
97.0
96.7
96.8
96.8
94.0
96.0
94.1
94.7
93.3
92.1
L HC1 concentration, ppmv
Inlet
112.
83.
1,000.
2,190.
1,140.
1.440.
247.
442.
526.
405.
1,660.
2,170.
270.
97.
86.
151.
1,180.
1,340.
Outlet
2.7
0.77
1.7
12.
90.
25.
42.
6.1
14.
18.
12.7
54.
72.
63.
15.
4.0
5.0
78.
106.
Average
92,
5-7
-------
Table 5-3. Summary of emission data for new and existing source MACT.
Facility
ID no.
132
105
1151
40 2
17
114
44
101
492
492
New/existing
source MACT
New
New
New
Existing
Existing
Existing
Existing
Existing
Existing
Existing
Outlet HC1
concentration,
ppmv
Average
1.6
2.1
7.7
1.7
8.0
12.7
42.
62.
63.
92.
Highest
run
5.9
3.5
7.7
2.7
15.
18.
90.
70.
72.
106.
Inlet HC1
concentration,
ppmv;
average
257.
5,150.
697.
98.
151.
405.
1,440.
3,450.
1,920.
1,260.
HC1 collection
efficiency, %
Average
99.5
99.96
99.0
97.8
94.7
97.0
97.5
98.1
96.8
92.7
Lowest
run
97.6
99.94
99.0
96.8
94.1
96.6
95.9
97.5
96.7
92.1
1 Data are from one sampling run.
2 Data are from two sampling runs.
Based on performance and design for both outlet concentration and collection
efficiency, three scrubbers appear to be superior to the others: the six tray scrubber employed
at facility 105, the five tray scrubber at facility 115, and the packed tower scrubber with the
combination mesh pad-chevron droplet eliminator at facility 132. Performance of the
scrubbers used as the basis for new source MACT was considered on the basis of long term
performance and variability in individual sampling runs. All three scrubbers served
continuous pickling lines. On the basis of average performance, all three scrubbers meet the
proposed new source standard for collection efficiency of 99 percent, and two meet the
proposed new source standard for outlet concentration of 3 ppmv. On the basis of worst
results from individual sampling runs, two scrubbers meet the proposed collection efficiency
standard but no scrubber meets the proposed concentration standard. To accommodate the
uncertainty in sampling, particularly in determining outlet concentration at these low levels,
the EPA decided to consider a new source standard for outlet concentration that could be met
5-8
-------
by the new source MACT scrubbers that did not meet the collection efficiency standard. This
concentration is 6 ppmv HC1, which is 5.9 rounded up to the nearest whole number. Based on
the worst individual sampling run results, all three scrubbers meet at least one of the two
alternative standards; one scrubber meets both the concentration standard of 6 ppmv and the
collection efficiency standard of 99 percent, one meets the concentration standard, and one
meets the collection efficiency standard. New source standards of 6 ppmv maximum outlet
concentration and 99 percent minimum collection efficiency are therefore promulgated for
continuous pickling lines.
Performance of the seven scrubbers used for the basis of existing source MACT for
continuous pickling lines was also considered on the basis of individual sampling runs. All of
these scrubbers also served continuous pickling lines. As discussed in the preamble to the
proposed rule, the concentration and collection efficiency standards were derived from the
scrubbers that were the better performers in each respect. Three units produced outlet HC1
concentrations of 1.7, 8.0, and 13 on the averages, 2.7, 15, and 18 ppmv for the worst runs;
all the others produced HC1 outlet concentrations of 42 ppmv or higher on the averages, 70
ppmv or higher for the worst runs. The concentration standard was therefore determined to be
18 ppmv HC1 from the performance of these three scrubbers. On the basis of HC1 collection
efficiency, with worst run efficiencies rounded off to the nearest percent, four of the seven
scrubbers meet a standard of 97 percent. The HC1 collection efficiency standard of 97 percent
was determined from the performance of the best four scrubbers. Five of the seven scrubbers
meet at least one of the alternative standards. The remaining two scrubbers would be
considered marginal to poor performers.
Comment: Two commenters (IV-D-7, IV-D-9) questioned the rationale of using data
from the best performing scrubbers to establish separate collection efficiency and concentration
limits because each owner or operator would have two options. The logic ignores the
statistical ability of scrubbers to comply with the proposed standard continuously and the very
basis for proposing alternative standards in the first instance. EPA "proposed alternative
standards out of the recognition that facilities with high inlet concentrations could not meet the
low HC1 outlet concentration standard and vice versa. Deriving the standards from the best
scrubbers for each option disregards the fact that the MACT floor is supposed to represent the
average of the best 12 percent and those facilities that have HC1 inlet concentrations too low to
comply with the proposed collection efficiency impossible and too high to comply with the
proposed 10 ppmv standard."
5-9
-------
Response: The EPA disagrees with the commenter. The logic expressed above is not
clear. The fact that the standard is not based on a statistical average has been discussed
previously. The assumption of the final standards is that at least some devices will not be able
to meet both options but all would be able to meet one or the other. Therefore the numerical
limits for each option were developed separately.
Comment: Two commenters (IV-D-7, IV-D-9) stated that the EPA has not sufficiently
justified its MACT determination for batch pickling lines. The rulemaking record contains no
data specific to batch pickling. Batch pickling lines are significantly different from continuous
lines in terms of design, operation, air capture rates, inlet concentrations, hood design,
product handling, and volume throughput rates. In light of these differences, the absence of
test data from batch lines, and limited data from continuous lines, it may not be appropriate
for the EPA to simply borrow and apply its MACT determination for continuous lines to batch
operations.
If the EPA promulgates this rule prior to supporting its MACT determination, batch
picklers will be in the position of not knowing if they can meet the standards until they have
spent the money to install or upgrade their pollution control equipment. The EPA would be
prudent to delay implementation of the proposed rule until it can demonstrate, based on batch
pickling-specific data, that the proposed standards do in fact constitute MACT.
Response: The commenters stated that there are significant differences between batch
and continuous pickling lines but do not give details nor any indication of how air pollution
control requirements are different. The commenters do not express any technical
considerations that have not already been addressed. Differences in fume capture systems
between batch and continuous operations, for example, are discussed in detail in chapter 4 of
the proposal Background Information Document. However, the effectiveness of the air
pollution control system is based on the characteristics of the gas stream, not the capture
system. According to scrubber manufacturers and designers, scrubber design considerations
are the same for both types of operations; refer, for example, to the contact summaries in
docket items IV-E-9 and IV-E-10. The major difference between batch and continuous
operations is that the HC1 concentration in batch line offgases varies during different phases of
the operating cycle. For example, the concentration can increase when steel is raised out of
the tank and allowed to drain before it is rinsed. Scrubbers can be designed on the basis of the
maximum concentration experienced.
5-10
-------
Regarding the ability of batch operations to meet the same standards as continuous
operations, the EPA notes the view expressed below by two commenters, one with extensive
relevant experience, that the proposed standards are reasonable and can be attained with
available control equipment.
After the comment period, the EPA received emission data from a batch pickling
operation (refer to docket item IV-D-23). Tests using EPA Method 26A were conducted in
July 1995 on an operation in which fumes from an open tank were captured by a push-pull slot
hood system and drawn through a vertical packed bed scrubber equipped with a mist
eliminator. No details on pickling process conditions were given. The scrubber design is
indicative of effective HC1 control. Only the outlet gas was sampled. Concentrations of HC1
measured in three sampling runs of one hour each were 5.1, 4.2, and 3.6 ppmv.
The only other information available for batch operations is from a test at another
facility that was conducted in June 1992. Fumes from an open tank were captured by a side
draft hood and drawn through a vertical packed bed scrubber. The pickling bath contained 14
percent HC1 at 120 °F, conditions that are relatively extreme for batch operations. Only one
sampling run, of one hour duration, was conducted on the scrubber outlet using the equivalent
of EPA Method 26A. A concentration of 6.3 ppmv HC1 was measured.
Results of these two tests give some indication that HC1 emission control for these
processes at levels achieved for continuous pickling lines is possible.
Based on these considerations, the EPA believes that control of batch pickling lines at
the level of existing source standards is achievable. However, the EPA agrees with the
commenters to the extent that control of batch lines at the new source standard level is less
certain. Because no clear limitation for new batch pickling lines could be determined from the
available information, particularly in considering the variation in operating conditions and
ventilation system design, the rule is revised to make the new source standard for batch
pickling lines the same as the existing source standard.
Comment: Five commenters (TV-D-3, IV-D-7, IV-D-9, IV-D-10, IV-D-15) stated that
for pickling lines, the options of meeting either a collection efficiency or an outlet
concentration standard should be retained. At a minimum, an outlet concentration standard
should be offered.
Response: Both options are retained.
5-11
-------
Comment: Two commenters stated that the proposed emission standards and options
are reasonable and can be attained cost-effectively with available control equipment. One of
these commenters has extensive credentials, having been involved in the design, operation,
testing, and evaluation of HC1 scrubbers, both packed bed and plate, on pickle lines and acid
regeneration plants for over 35 years and the original developer of the use of plate scrubbers
for HC1 absorption on pickle lines.
Response: The EPA acknowledges these comments.
5.2 FUME CAPTURE CONSIDERATIONS
Comment: One commenter (IV-D-13) believes that requirements for ventilation system
hoods should be included in the standard, or should be characterized in the preamble as to
expected capture efficiency. The commenter requests guidance as to verification of air flows
to assist in determinations regarding non-conforming sources.
Response: The EPA is responsible for regulating emissions discharged to the ambient
air, not the workplace. If sources can meet and maintain workplace standards without hooding
or ventilation systems, it is not EPA's responsibility to require such systems.
Comment: Two commenters (IV-D-7, IV-D-9) stated that the EPA should clarify that
the proposed standard for efficiency is based on control efficiency, not capture efficiency
Response: The rule specifies that control device inlet/outlet measurements are to be
made to establish compliance with a collection efficiency standard.
5-12
-------
6.0 STANDARDS FOR ACID REGENERATION PLANTS
6.1 BASIS FOR THE HC1 NUMERICAL STANDARD
Comment: One commenter disagreed that sufficient source test data were available to
pursue an actual emissions approach for determining the MACT floor. The EPA evaluated
five measured scrubber outlet concentration values, then noted that one value was far out of
line with the others and did not consider this value is establishing the floor. No attempt to
review the next appropriate value was made by the EPA. Constructing a fifth data point in
lieu of actual data has no technical or regulatory basis under section 112 of the Clean Air Act.
The EPA should have used another facility's actual test data or conducted additional tests to
establish a fifth point.
A second commenter observed that the MACT floor on which the EPA bases its
standard is not representative of single stage water scrubbing. Caustic scrubbing technology,
contrary to EPA's belief, has been shown to be more effective in reducing HC1 emissions than
scrubbing with unneutralized water. The EPA notes in the proposed rule that no single stage
scrubber has demonstrated the capability of meeting the proposed existing source standard of 8
ppmv HC1. The EPA should consider the cost impacts to the industry for waste water
treatment and sludge disposal if the standard is to be based on caustic scrubbing.
A third commenter provided additional data, presented in table 6-1, which are from the
two acid plants that use two stage scrubbing. The data include outlet concentration data for
the first stage water scrubbers. All tests except for two consisted of three sampling runs of
three hours each using EPA Method 26A; the remaining two tests consisted of two sampling
runs.
Response: The EPA agrees with the first commenter in that the method used at
proposal to determine the floor was not appropriate, specifically, the manufacturing of a fifth
data point in lieu of having actual data followed by averaging. Furthermore, the EPA agrees
with the suggestion of the second commenter that the proposed existing source standard of 8
ppmv HC1 is not demonstrated to be achievable with single stage water scrubbing, the
predominant control technology used in the industry.
6-1
-------
Table 6-1. Test data for HCI emissions from acid regeneration plants
using two-stage scrubbing, water in first stage.
Date Plant
(Month/Year) (A or B)
03/93 A
B
04/94 A
B
11/95 A
B
03/96 A
B
11/96 A
B
Average (and range) HCI concentration. Dt>mv
First stage
outlet/
second stage
inlet
—
—
23 (17 31)
20 (14 25)
—
—
11.2 (9.1 14.6)
5.6 (5.5 5.6)2
19.4 (18.3 20.3)
13.7 (11.6 17.6)
Second
outlet
49
19.5
3.1
0.9
7.4
11.1
3.5
1.8
12.4
11.5
stage
(38
(9.0
(2.8
(0.9
(7.3
(10.7
(2.9-
(1.4
(9.9
(7.2
58) '
26.7) l
3.7) '
1.0) ^
7.5) '
11.9) '
4.5) '
2.1) '
15. 7) 3
17. 1)3
1 Caustic solution was used in the second stage scrubber.
2 Two sampling runs were made instead of three.
3 Water was used in the second stage scrubber.
The floor was therefore reexamined on the basis of the median of the best five
controlled sources on a technology basis. The best two controlled sources employ either two
stage acid recovery or two stage scrubbing, with neutralized water used in the last scrubbing
stages in both cases. The third best controlled source employs single stage scrubbing with
unneutralized water; this technology is also used by all of the remaining sources in this
subcategory. The standard for existing sources is therefore developed based on the
performance of single stage water scrubbing, which addresses the main concern of the second
commenter.
6-2
-------
With the inclusion of the information in table 6-1, long term data from two acid
regeneration plants are now available. Data from the plant for which the measured HC1
concentration was 16 ppmv were still restricted to the one test, which consisted of two
sampling runs with measured HC1 concentrations of 15.6 and 15.8 ppmv. The final data point
available was 137 ppmv HC1, which is so far out of line with the other data that the plant
tested could not be considered well controlled; data from this plant could therefore not be
used to establish an emission standard.
In order to determine a numerical concentration standard from all of the available
information, process and control system variability over time were taken into account by
considering HC1 concentration averages and also values for individual sampling runs. On the
basis of average outlet concentrations, it seems clear that the three plants being considered
meet a limit of 25 ppmv HC1. Considering all 19 individual runs from the three plants, except
for one run of 31 ppmv, all others are 25 ppmv or less. A maximum outlet concentration of
25 ppmv HC1 therefore seems reasonable for a standard based on single stage water scrubbing.
Regarding the new source standard for HC1, the data in table 6-1 include outlet
concentration data from four tests from each plant on second stage scrubbers that use
neutralized water. In these tests, three sampling runs of two or three hours were made using
Method 26A. Results of the first tests are much higher than those from the more recent three
tests and apparently do not reflect current operations. Results of the last three tests are
average HC1 concentrations ranging from 0.9 to 11.1 ppmv, with results of individual runs
ranging up to 11.9 ppmv.
The only other HC1 concentration data that have not already been discussed are from
the plant that employs two stage acid recovery plus a venturi scrubber that uses neutralized
water. Results from only one test are available; the average HC1 outlet concentration was 1.0
ppmv.
Considering the capability of a scrubber to meet a long term standard, results from the
first two plants seem more meaningful. These plants clearly meet an outlet concentration HC1
standard of 12 ppmv over the most recent three tests. A new source maximum outlet
concentration standard of 12 ppmv HC1 has therefore been demonstrated. Consequently, the
final standard is a maximum outlet HC1 concentration of 25 ppmv for existing sources, 12
ppmv for new sources.
6-3
-------
Comment: Two commenters (IV-D-1, IV-D-3) stated that the EPA did not demonstrate
that its standards for existing and new sources are based on a sustainable level of performance.
One commenter stated that there is a wide variation in HC1 emissions at different times using
the same control technology. This commenter provided the data given in table 6-2 at the
EPA's request to support the statement. Emissions range from 26 to 542 ppmv HC1 with,
according to the commenter, no obvious anomalies in the acid regeneration data. The EPA's
data illustrate that there is a wide variation between and within facilities. The standard
deviation for all data from which EPA determined its standard is 7.2 ppmv, which is far out of
range of the proposed limit.
Table 6-2. Test data for HC1 emissions from acid regeneration plant
using single stage water scrubbing.
Date Average (and range) HC1 concentration, ppmv
(Month/ Year)
06/94
10/94
02/95
09/95
11/95
06/96
10/96
03/97
06/97
08/97
69
308
64
298
59
31
49
470
90
34
(26
(52
(50
(245
(43
(28
(45
(336
(37
(30-
152)
467)
87)
387)
82)
35)
63)
542)
164)
35)
Response: By comparison with data from other facilities, the plant for which the data
in table 6-2 were taken cannot be well controlled, particularly considering the extreme range in
values between the lowest and highest measurements. Data from this facility are not relevant
in determining a standard based on the best performing plants. The issue of sustainable
performance has been addressed in the previous comment and response.
6-4
-------
Comment: One commenter (IV-D-8) recommended that the new source standard for
HC1 be set no lower than 5 ppmv and the existing source standard be set no lower than the
current permitting levels or 10 ppmv, whichever is lower. No rationale is given for these
values except that interference in testing due to ammonium chloride in the spent liquor may
complicate the measurement of HC1 at 3 ppmv.
Response: The revised HC1 standard accommodates the commenter's recommendation,
although in response to other considerations. The issue of interference in testing due to
ammonium chloride is discussed in section 10.0, TEST METHODS.
Comment: One commenter (IV-D-15) stated that the option of meeting a performance
standard such as either a collection efficiency or an outlet concentration standard for HC1
should be offered for acid regeneration plants as well as for pickling lines.
Response: The HC1 outlet concentration standard offered is in fact a performance
standard. In view of the fact that the concentration standard is relaxed substantially for both
existing and new sources from the proposed standard, the EPA believes that no alternative is
necessary.
Comment: Two commenters (IV-D-4, IV-D-14) believe that the proposed standard is
reasonable and can be attained cost-effectively with available control equipment.
Response: The EPA acknowledges these comments.
6.2 BASIS FOR THE C12 NUMERICAL STANDARD
Comment: Two commenters (IV-D-1, IV-D-3) stated that the C12 limit should be based
on five sources instead of three. The small sample size probably does not reflect variability at
each source. The 4 ppmv limit has not been shown to be continuously achievable. One
commenter stated that the existing source emission limits should be determined from the
average of five facilities plus two standard deviations; the standard should be at least 74.3
ppmv. For new sources, the standard should be 60 ppmv based on two standard deviations
from the mean of the EPA's data. The other commenter did not recommend specific standards
but provided additional data at the EPA's request (table 6-3).
6-5
-------
Table 6-3. Test data for CI2 emissions from acid regeneration plant.
Date
(Month/Year)
03/93
04/94
03/96
11/96
04/94
03/96
11/96
Plant
(A or B)
A
B
A
B
A
B
A
B
A
B
A
B
A
B
Average
0
1
2,
0.
0.
0,
5.
0.
5,
9.
0.
0.
4.
(and range) C12 concentration, ppmv
.7
.0
.1
.3
.5
,4
,3
,4
,1
,9
.4
,4
2
0.4
(0.
(0.
(1.
(0.
(0.
(0.
(3.
(0.
(3.
(7.
(0.
(0.
(2.
(0.
5
9
1
2
4
1
4
4
6
7
3
4
3
4
-0
1
3,
0,
0,
0.
7.
0.
7,
.9)'
• I)1
•4)1
.4)1-2
•6)1
•6)1
.I)1
.4)'
.3)3
14)2,3
0.
0.
5.
0.
5)3
4)3
6)3
5)3
1 Outlet of caustic scrubber preceded by water scrubber.
2 Two sampling runs were made instead of three.
3 Outlet of water scrubber that precedes caustic scrubber.
Response: As discussed under the basis for the HC1 numerical standard, the standards
for acid regeneration plants are now being based on technology, which is single stage water
scrubbing. As in the case of the HC1 standard, the C12 numerical standard was reconsidered
based on the body of data available for this technology.
6-6
-------
In addition to the data in the second half of table 6-3, C12 outlet concentration data from
other facilities are 3.3 and 60 ppmv, each based on one test. The 60 ppmv value is so far out
of line with the others that it cannot be considered representative of effective operation and
therefore cannot be used in determining the standard. Considering the data in the second half
of table 6-3, except for the 9.9 ppmv value measured for one test on acid plant B in 1994, all
other results are 5.1 ppmv or less. The 9.9 ppmv value from the 1994 test appears to not be
representative of current operations because the results of the 1996 tests are much lower, 0.4
ppmv in each case. It appears that a limit of 6 ppmv C12 can be met by these operations,
considering the variability in measurements (except for the one nonrepresentative value)
because only one sampling run gives a higher result (7.3 ppmv). The concentration standard
for C12 is therefore revised to 6 ppmv for existing sources.
Regarding the standard for new sources, the EPA is required to set the standard
according to the capabilities of the best controlled facility. The data in table 6-3 include
results of the four tests on the outlets of second stage scrubbers that use neutralized water.
Results are similar to those for the first stage water scrubbers. Average C12 concentrations
range from 0.4 to 5.3 ppmv, with results of individual tests ranging from 0.1 to 7.1 ppmv.
An individual plant cannot be identified that provides better performance than existing source
MACT. The new source standard for C12 is therefore the same as the existing source
standard, 6 ppmv.
Comment: One commenter (IV-D-8) recommended that the C12 limit be based on water
scrubbing in order to avoid generating caustic scrubbing solution for treatment and disposal.
A limit of 5 ppmv is recommended based on three tests on the same plant in which the
chlorine levels in the stack ranged from "not detectable" to 5.5 ppmv.
Response: The revised standard accommodates the commenter's recommendations.
The data cited are consistent with the data presented above.
Comment: One commenter (IV-D-3) supported the optional C12 concentration standard
to be established for each source.
Response: The optional standard is retained for existing sources but removed for new
sources, as discussed in Section 1.1, SUMMARY OF RULE CHANGES SINCE PROPOSAL.
6-7
-------
7.0 STANDARDS FOR ACID STORAGE VESSELS
7.1 INCLUSION OF STANDARDS FOR ACID STORAGE VESSELS
Comment: One commenter (IV-D-11) believes that storage vessels are not significant
sources of HC1 and may be much more costly to control than indicated in the EPA's models if
remotely located from the scrubber. The commenter suggests that costs for installing vent
piping may be considerably higher than estimated by the models. The commenter also
suggests that storage vessels be excluded from the rule.
Response: The rule will be reworded to make clear that alternative means for
controlling storage vessels are allowed. If the vessels are distant from the pickling or acid
regeneration scrubber, a bubbler type device, for example, could be used that does not require
forced ventilation. Vent lines can be routed to the bubbler such that a water (or caustic
solution) seal is maintained and the scrubbing medium is kept below its HC1 saturation
concentration.
7.2 FLEXIBILITY IN CHOICE OF CONTROL SYSTEMS
Comment: Two commenters (IV-D-7, IV-D-9) believe the EPA should clarify that
"control devices" for storage vessels are not a specific control technology, and that facilities
can use any method that is demonstrated to minimize emissions to the atmosphere (e.g.,
bubbling through a drum or small tank of caustic solution or water).
Response: The EPA agrees with the commenter. No specific control device is
required for storage tanks. The definition of closed-vent system has been reworded to make
the EPA's intention clear. Examples of devices that might be used include systems that bubble
emissions through a small tank of water or caustic without the aid of a fan. However, larger
facilities may find it advantageous to route emissions from a tank farm or acid regeneration
plant to a pickling line scrubber or to build a separate scrubber system for control.
Comment: One commenter (IV-D-6) described a fumeless pickling system in which the
pickling tanks are not ventilated to a scrubber. Because the pickling line has no control
7-1
-------
system, acid storage vessels serving this line could not use the same system serving the
pickling line and, therefore, would require a separate control system. The commenter
recommended that allowances be made for storage vessels serving fumeless pickling lines.
Response: Referring to the previous response, requirements for acid storage vessel
emission controls are flexible enough to accommodate this situation. In addition, unless the
subject facility is part of or collocated with a major source, it is likely not a major source and
not subject to regulation.
7-2
-------
8.0 MAINTENANCE REQUIREMENTS
8.1 ESTABLISHMENT OF NECESSITY FOR REQUIREMENTS
Comment: One commenter (IV-D-12) stated that the maintenance activities should be
guidelines and not requirements. They do not further the rule (beyond required monitoring) to
limit emissions and assure compliance with the limits. The EPA did not show authority to
require specific maintenance activities and did not demonstrate that the required activities are
correlated to emissions.
Response: The EPA has the authority to establish operation and maintenance
requirements pursuant to section 112 of the Clean Air Act. These requirements are
enforceable independently of emissions limitations or other requirements in relevant standards.
Maintenance requirements are necessary to help ensure that emission control equipment
continues to operate at a level consistent with its operation at the time of compliance testing.
The EPA's statement of these requirements is in 40 CFR 63.6(e)(l)(iii), Operation and
Maintenance Requirements.
8.2 SPECIFICITY OF REQUIREMENTS
Comment: Three commenters (IV-D-3, IV-D-7, IV-D-9) stated the following. The
EPA's maintenance plan should not establish specific elements of the required maintenance
plan, i.e., following manufacturer's recommended maintenance, cleaning scrubber internals
and mist eliminators at intervals sufficient to prevent fouling, having set intervals for
inspecting system components to identify, repair, or replace as needed. The "Any Credible
Evidence Rule" obviates the need for the maintenance plan with specific actions. Two of the
commenters recommend that the EPA amend proposed § 63.1159 by eliminating the
requirement that maintenance plans must include the elements set forth at
§ 63.1159(b)(2)(i)-(iv); these elements should be included only as potential elements that may
be included in the plan. Another commenter believes that the operation and maintenance plan
should not require strict adherence to the manufacturer's operating manual. Many
manufacturer's manuals contain steps that are determined not be necessary and/or that only the
manufacturer's proprietary products should be used. The EPA should change the wording to,
8-1
-------
for example, "substantially include" the elements set forth in the manufacturer's operating
manual.
Response: The EPA has reviewed the proposed maintenance plan requirements and
decided that revisions are appropriate. Manufacturer's instructions for older equipment may
require materials no longer available. Manufacturers may no longer be in business so that
required parts or materials cannot be purchased except by substitution from a source other than
the original manufacturer. Therefore, the EPA has revised the rule so that it no longer
requires adherence to the manufacturer's manual. The facility must write an operation and
maintenance plan that is consistent with good maintenance practices and includes, at a
minimum, the list of items described in the rule. The EPA believes that inclusion of these
items is reasonable. Additionally, pressure drop must be monitored once per shift as a means
of discovering scrubber operational anomalies that may require maintenance. No specific
pressure drop deviation limit is required, but the monitoring records are required to be kept
along with the recycle and makeup water flow rates as discussed in Section 11.0,
MONITORING REQUIREMENTS.
Comment: Three commenters stated that the operation and maintenance plan should
not be part of the source's Title V operating permit. Plan approval places a substantial burden
on permitting authorities. The details of these plans are frequently changed as operational
problems are addressed. Such a requirement could cause administrative nightmares if a source
is required to go through the Title V permit modification process every time it modifies a plan,
especially during the early stages of the rule. Approval of plans by informal action would
encourage timely revision.
Response: The EPA agrees and has revised the rule so that the maintenance plan is
only required to be incorporated by reference into the permit. Also, it is not required to be
submitted to the permitting authority
Comment: One commenter (IV-D-3) believes the requirement that the "responsible
plant official" sign records of inspections is overly burdensome. The requirement is
acceptable if "responsible plant official" means that an employee delegated the responsibility
by the "responsible official" must sign.
Response: The EPA agrees with the commenter and has added the definition
"Responsible Maintenance Official," who is a person having signature authority for signing
reports required under the rule.
8-2
-------
8.3 TIMING OF RESPONSE TO OPERATING PARAMETER EXCURSIONS
Comment: One commenter (IV-D-3) stated that the requirement to initiate repairs
within one day is excessive and unworkable. It is unclear what "initiate corrective action"
means. In some cases, corrective action may require engineering analysis to determine the
source of the problem and effective corrective action. If this provision is retained, the
commenter recommended that it be written as a requirement that repairs begin promptly and
provide a "safe harbor" that repairs commenced within one day are considered to be prompt.
Two commenters (IV-D-7, IV-D-9) stated that the proposed requirement that
maintenance plans be implemented within one working day is too stringent. There may be
situations when initiating the plan within 24 hours would be impractical or impossible. In
some cases, a facility may have to rely on an outside contractor to conduct necessary action.
Instead of establishing a time-specific deadline, the EPA should provide that "facilities must
initiate corrective action as soon as practically possible, but no later than three working days."
One commenter (IV-D-11) stated that the requirement for corrective action within one
day of detection of an operating problem with a control device is neither reasonable or in
keeping with the notification and repair requirements of other NESHAP rules. The
commenter recommended that the requirement be changed to include a first attempt at repair
within five working days of detection.
Response: The EPA believes that it is reasonable to expect operators to initiate
procedures towards corrective action within one day and complete repairs or maintenance as
soon as practicable. Initiation of procedures may consist of notification of a contractor or
service group that corrective action is necessary. The rule is revised to clarify that the
procedures to be initiated are the actions that are specified in the maintenance plan.
8-3
-------
9.0 PERFORMANCE TESTING
9.1 ESTABLISHMENT OF MONITORING PARAMETERS
Comment: One commenter (IV-D-1) stated that establishment of site specific scrubber
operating parameters as a measure of compliance without first establishing the relationship
between the parameters and the emissions in question is not appropriate. The EPA has made
not established any relationship between the proposed mandated parameters and actual
emissions. This information was not evaluated during the MACT development; therefore,
site specific parameters should not become mandated compliance parameters.
Response: Without implementation of continuous emissions monitoring systems,
monitoring of relevant operating parameters in combination with routine and preventative
maintenance is essential to enhanced compliance assurance. The requirement for operating
parameter monitoring is retained in the rule.
Comment: One commenter (IV-D-11) stated that in setting parameter operating limits,
the full range of values observed during a compliance test should be used, not the average.
Because an average is being established, at least one of the tests must necessarily be above the
average if all three tests are not identical. Alternatively, operation within the manufacturer's
recommendation should be considered compliant unless other limits are demonstrated to be
appropriate. Another commenter (IV-D-3) believes that owners and operators should be able
to establish compliant operating parameters using individual runs from compliance tests and
not be restricted to multiple tests. Using multiple runs during a test will greatly diminish costs
and repetitive sampling without substantially diminishing the assurance of compliance.
Response: The EPA agrees that some flexibility is establishing operating parameter
compliant values is appropriate. The rule is revised to allow an average parameter value
measured during any of the runs used to demonstrate compliance to be used as the compliant
value rather than the average value measured over the entire testing period.
Comment: Two commenters (IV-D-11, IV-D-3) believe operators should have the
option of conducting compliance demonstration tests as needed to show appropriate ranges of
9-1
-------
scrubber parameters. Establishment of parameters should not be limited to the initial
performance test.
Response: A facility always has the opportunity to conduct additional compliance tests.
The rule allows facilities to conduct multiple performance tests to establish alternative
compliant operating parameter values and to reestablish compliant values during any
performance test conducted after the initial performance test.
Comment: One commenter (IV-D-14) stated that requirements for adhering to
operating parameter limits should be set so that a facility is not penalized for improving
scrubber operation. For example, gas flow rate might be lowered if a more effective capture
system were installed without reducing scrubber efficiency, but could lead to
out-of-permit-range operating conditions for pressure drop. This type of situation is
counterproductive. Two other commenters (IV-D-1, IV-D-10) discussed mist eliminators and
associated them with performance. They described a scenario in which a company installs a
mist eliminator after initial compliance testing that causes increased pressure drop. The
increased pressure drop violates the standard.
Response: It is not the intent of the rule to prevent improvements to emission control
systems. The emission limits, however, are not to be exceeded. Documentation must be
provided that a change in system hardware or operation is beneficial in terms of emission
reduction, or does not degrade emission control below the permitted limits. As previously
discussed, the operator is always free to establish new operating parameter averages or limits.
The issue will be less significant with the choice of makeup and recirculating water flow rates
as the scrubber operating parameters instead of pressure drop.
9.2 OPERATION OF PROCESS DURING TESTING
Comment: One commenter (IV-D-2) stated that testing should be performed at
representative conditions rather than worst-case conditions. There is no relationship between
worst-case conditions and maximum rated capacity.
Response: The proposed standards require testing ". . .under conditions representative
of normal process operations." This requirement, typical of many standards, is retained for
steel pickling and acid regeneration standards and is consistent with the commenter's request.
9-2
-------
10.0 TEST METHODS
10.1 APPLICABILITY OF PROPOSED METHODS
Comment: Two commenters (IV-D-11, IV-D-14) had the following statement.
Method 26A is not validated for steel pickling, only for municipal waste incinerators (MWIs).
MWIs have higher temperatures, less moisture (and no liquid droplets), and no ferric chloride
content, which could interfere with test results. The EPA's tests also show variations of as
much as 700 percent for the same pickling line. Test bias may have resulted in an improperly
low standard. Inexplicable negative biases are reported in the EPA municipal waste
incinerator validation report2 for Method 26A. These biases are such that validation for
pickling sources is required. The practical level of quantification (PLQ) for Method 26A has
not been established for pickling sources, and should be developed using Method 301. Also,
ferric chloride might cause a positive bias for the HC1 measurements. One commenter
believes that conditions encountered with HC1 pickling tests include high humidity in the gas
stream, extremely high solubility of HC1 gas in water, condensation in the gas stream,
refluxing in the stack, and the use of stack tip entrainment eliminators. These conditions lead
to several measurement problems, all of which tend to bias results towards improperly high
HC1 concentration because of enriched droplet capture in the sampling probe or
maldistribution of HC1 with regard to sampling probe location. Sampling data show six cases
in which the range of measured maximum concentrations varies from 1.3 to 9.3 times the
minimum concentration for heated pickling lines or acid regeneration plants. They
recommend that the testing protocol include provisions for testing after all control devices
(including stack-tip mist eliminators) and allow for discard of test results more than 50 percent
above the average.
Response: The comments do not bring up any technical concerns regarding
measurements at pickling or acid regeneration sites. A well designed and conscientiously run
field validation of Method 26A specifically at these source categories would not likely uncover
2 Laboratory and Field Evaluation of a Methodology for Determination of
Hydrogen Chloride Emissions from Municipal and Hazardous Waste Incinerators,
EPA:AREAL, Publication No. EPA-600/3-89-064, April 1989.
10-1
-------
any evidence that there is a problem in this application. The EPA knows from its studies that
the method is capable of measuring to fractional ppmv levels. Review of data from a 1997
study3 at a light weight aggregate kiln burning hazardous waste provides a minimum detection
limit estimate of about 0.04 ppmv. The EPA estimated the method precision (reported as the
standard deviation of individual runs) to be 0.42 ppmv at 3 ppmv. This value would lead to
the precision estimate of the mean of a 3 run test of 0.24 ppmv (0.42^3°-5). If water droplets
are routinely present, then the method has to be followed carefully to avoid gathering poor
quality data. The EPA has not knowingly field validated the method in the presence of water
droplets, but isokinetic sampling is the accepted way to address this problem.
The commenters contended that the EPA provides no justification to the preamble
statement "the EPA considers the method is equally valid for measuring emissions for pickling
and acid regeneration sources." They go on to say that HC1 pickling emissions are generally
100 to 200 °F and contain water droplets. The presence of water droplets increases the
potential for negative bias.
The EPA responds that the method is validated at a municipal waste combustor (MWC)
where the sample matrix is a more severe test of the method in terms of potential chemical
interferents, and the stack is at a higher temperature. The higher stack temperature at MWCs
is a more severe test of the method in that the probe and filter temperatures are less than the
stack temperature, which, in theory, could lead to condensation of HC1 in the probe. An
effective control system would be expected to include a mist eliminator, thus minimizing the
potential for excessive water droplet effect. In addition, the test method has provisions to
overcome the potential negative bias encountered if water droplets are present. These
provisions include maintaining the probe and filter temperature during sampling, the use of a
cyclone separator ahead of the filter to collect the water prior to its reaching the filter, and a
documented end-of-test purging provision to evaporate this collected moisture and HC1 prior to
sample recovery to assure that it is accounted for. The EPA knows that if these procedures
are not followed, then there is a potential for loss of HC1 in the probe and (optional) cyclone.
Field validation of the method at a pickling operation to assure that these procedures, when
conscientiously applied, overcome the potential negative bias caused by excess moisture
droplets would be expensive, difficult to design, do little to improve the design of the method,
3 Dioxins/Furans, HC1, C12, and Related Testing at a Hazardous Waste Burning
Light-Weight Aggregate Kiln, EPA:OSW, Draft Report, Prepared by Energy and
Environmental Research Corporation, October 10, 1997.
10-2
-------
or assure that future data are collected conscientiously following the documented method's
procedures.
One commenter also remarked on the MWC validation being done with midget
impingers rather than the large impingers. The EPA report No. 600/3-89-0644 concludes that
there is an inexplicable negative bias compared to those using midget impinges. The most
likely cause of the low bias at low (3 to 4 ppmv) concentrations is absorption of HC1 on
alkaline particulate matter collected on the filter. This condition is not expected at steel
pickling plants and, hence, field validation would not be of value. For added clarification, see
pages 48 and 49 of the previously cited report.5
The commenter also stated that proper field validation of Method 26A would provide
the true PLQ that would take into account the normal variations resulting solely from the test
procedures. Determining the actual PLQ of Method 26A on HC1 pickling emissions is
essential to ensure that the final NESHAP limitations are not set lower than the level that can
be consistently quantified by the required testing. The recommendation already discussed in
this response should also apply to HC1 regeneration plants since the limit of 3 ppmv HC1 is at
the lower limit of the range tested.
The EPA notes that the commenter provided the Method 301 definition of PLQ. There
is general agreement that the intent of the Method 301 calculation procedure of 10 tunes the
standard deviation should use the standard deviation at or near the limit of detection. (The
actual Method 301 language adds ". . .at the blank level.") The EPA believes the commenter
cites an erroneous conclusion from a Rigo and Rigo Associates, Incorporated, document,6 that
a recent quad-train study at an MWC had a PLQ of at least 125 ppmv at 7 percent oxygen for
Method 26A. The study was done in a concentration range of 105 to 636 ppmv at 7 percent
oxygen, instead of near the acceptable blank limit of the method. These conditions lead to an
inflated standard deviation estimate and a subsequent over estimate of the PLQ. Draft results
from the 1997 EPA study7 using a quad-train arrangement at a light weight aggregate kiln
4 See footnote 2
5 See footnote 2
6 Quantitation Limits for Reference Methods 23, 26, and 29. H. Rigo, Rigo and
Rigo Associates, Inc., and A. Chandler, A. S. Chandler and Associates Ltd.
Undated document.
7 See Footnote 3
10-3
-------
where the actual (uncorrected for dilution) stack concentration of HC1 ranged from 0.22 to
1.29 ppmv (more closely approaching the theoretical lower limit of the method) results in an
estimated method standard deviation of 0.12 ppmv at zero. The EPA used these data to
extrapolate an estimated method standard deviation of 0.42 ppmv at 3 ppmv as described
above. This value compares favorably with the original MWC validation report's estimate of
standard deviations of 0.24 ppmv and 0.49 ppmv at concentration of 3.9 ppmv and 15.3 ppmv,
respectively.
Regarding positive bias caused by ferric chloride, it would have to have a significant
vapor pressure at the filter temperature to pass through the Teflon matte filter in the test
equipment. This is not the case.
The EPA believes the test method is appropriate for steel pickling and acid
regeneration operations and will continue to require its use (or an approved substitute) for the
standard. However, in order to reduce the possibility of collecting water droplets from the
stack walls that may be present because of refluxing in the stack or high humidity, the EPA
believes that Reference Method 1 should be modified for this application to specify that no
sampling point be closer to the stack wall than one inch.
Comment: One commenter (IV-D-8) stated that ammonia is commonly used as a
precipitating agent in waste HC1, resulting in ammonium chloride formation. The commenter
believes that some ammonium chloride will be decomposed in the acid regeneration plant
roaster, but significant amounts may exit in the waste gas and will be recovered along with
HC1 in gas cleaning. The commenter is currently investigating the possibility of direct
measurement of ammonium chloride in the acid plant scrubbers but does not at present have
data to offer. The commenter understands that ammonium chloride can interfere in the
measurement of HC1 at low levels.
Response: Ammonium chloride is identified as a possible interferent in EPA Reference
Method 26A that would be expected to appear as chloride ion and thus be measured as HC1.
If an acid regeneration plant cannot meet the standard for HC1, it would have the option of
demonstrating that ammonium chloride is present in the waste pickle liquor fed to the plant
and seeking relief in the HC1 emission limit on that basis. However, the need for relief seems
unlikely. Ammonium chloride would not be expected to pass the filter that is required for this
method at the filter temperature. Ammonium chloride decomposes from the solid state at
339 °C, which is far above the probe temperature of 248 °F (120 °C) used for sampling acid
regeneration plant emissions.
10-4
-------
11.0 MONITORING REQUIREMENTS
11.1 GEMS REQUIREMENT
Comment: Four commenters (IV-D-1, IV-D-7, IV-D-9, IV-D-10) stated that excessive
excursions of operating parameters should not trigger implementation of CEMS. Process
parameter variations are a normal part of a well run operation. The relationship between site
specific parameters and emissions is not demonstrated. Installation of a CEMS would not
affect the number of excursions. Two other commenters (IV-D 3, IV-D-11) stated that more
than six excursions of operating parameters should not be the criterion for CEMS installation.
Either a five percent criterion or a statistical representation of variations from established
levels would be more appropriate. In addition, seven commenters (IV-D-7. IV-D-8, IV-D-9,
IV-D-10, IV-D-11, IV-D-14, IV-D-15) stated that the use of CEMS should not be required.
No systems have been demonstrated to have the capability to accurately measure and record
compliance for this application. Commercially available systems for monitoring at the
proposed levels are expensive, difficult to calibrate and maintain, and not reliable to the level
of operation required. Manufacturers have cautioned that using such devices in an acidic
application containing water droplets would interfere with test methodology and be corrosive
to the testing apparatus. Conditions of high humidity and acidity make it unlikely that an in
situ sensor will ever work.
Response: After reviewing the comments, the EPA agrees that reliable operation of
currently available CEMS cannot be assured for this application. At best, inordinately
burdensome maintenance and operating procedures would be required. The CEMS
requirement is therefore deleted.
Comment: Four commenters (IV-D-4, IV-D-7, IV-D-8, IV-D-12) suggested
alternatives to the CEMS requirement. The rule should allow discretion by the permitting
authority to consider good cause or correction. Implementation of a quality improvement plan
(QIP) would be a better alternative that would have the effect of reducing emissions. Stack
testing could be allowed to see if the source was really in noncompliance.
11-1
-------
Response: The CEMS requirement has been deleted. The requirement to initiate
corrective action upon any operating parameter excursion is retained, which would have a the
effect of reducing emissions, similar to the intent of a QIP A source still has the option of
conducting another performance test to reestablish compliant parameter values.
Comment: One commenter (IV-D-4) applauded giving sources the option of
continuously monitoring emissions as an alternative to performing annual tests and monitoring
control device parameters.
Response: Sources may voluntarily use continuous monitors if they comply with all
applicable provisions of subpart A to part 63. The requirement to use a CEMS under certain
circumstances is deleted.
11.2 SCRUBBER MONITORING PARAMETERS
Comment: Five commenters (IV-D-1, IV-D-7, IV-D-9, IV-D-10, IV-D-11) stated that
the allowed variations in pressure drop and acidity are too restrictive. Pressure drop can
change due to factors unrelated to HAP emissions. Limiting the variation to one inch of water
column is arbitrary Operating requirements should encompass the range recommended by the
manufacturer, or facilities should be allowed to develop appropriate ranges using a variety of
methods such as compliance testing, engineering principles, and manufacturer warrantees.
Minimum pH should not be restricted to the maximum recorded during testing. Requirements
for acidity or pH should encompass the full range of conditions recommended by the
manufacturer, or acidity should be monitored only to assure that it is not too high.
Two other commenters (IV-D 3, IV-D-12) stated that the frequency of monitoring
pressure drop and acidity is excessive. If acidity is monitored at all, once a day is sufficient,
which would be in line with pressure drop readings.
Response: The EPA agrees that some flexibility is justified. Some guidelines in
operating parameter variations, however, are appropriate. Manufacturers' warrantees or
recommendations are not always available. The option of conducting additional performance
tests addresses some of the commenters7 concerns, including the one inch water column
variation in pressure drop. However, referring to the discussion below, pressure drop and
acidity will no longer be the operating parameters required to be monitored.
11-2
-------
Comment: Five commenters (IV-D-3, IV-D-6, IV-D-10, IV-D-14, IV-D-15) stated
that pressure drop and acidity are not appropriate monitoring parameters. A relationship
between these parameters and scrubber efficiency has not been demonstrated. Given the lack
of variation of scrubbing efficiency between caustic solution and clear water, monitoring
acidity is questionable. Also, the requirement to measure acidity is vague.
Three commenters (IV-D-6, IV-D-11, IV-D-14) suggested that other parameters than
pressure drop and acidity would be better indicators of scrubber performance. Scrubber water
flow rate is a more valid indicator of efficient scrubbing. For packed bed scrubbers, betters
parameters are pressure drop, air flow rate, and water flow rate to the top of the packing. For
plate scrubbers, pressure drop and visual observation provide assurance of correct operation.
Other parameters suggested are fan amps and liquid conductivity.
Response: In considering all of these comments, the EPA concludes that scrubber
makeup water and recycle water flow rates are better indicators of scrubber performance than
pressure drop and acidity, on the basis that the mechanism for HC1 collection is absorption in
water, which can be done effectively even with slightly acidic water. The rule is revised,
eliminating the requirements for monitoring scrubber pressure drop and scrubbing effluent
acidity and replacing them with the requirements to monitor scrubber makeup water flow rate
and, for scrubbers that operate with recirculation, recirculation water flow rate. The
monitoring requirements are as follows:
• Scrubber makeup water and recirculation water flow rates must be monitored
continuously and recorded at least once per shift while the scrubber is operating,
• the compliant values for flow rates are established during initial performance testing or
any subsequent performance testing that is done to determine compliance with emission
standards or to establish new operating conditions,
• corrective action must be taken if either of the flow rates falls below the compliant
value that is established during performance testing,
• initiation of corrective action procedures must begin within 1 working day, and
• the monitoring devices must be calibrated according to the manufacturer's schedule, or
not less frequently than once per year in the absence of a manufacturer's schedule.
11-3
-------
Monitoring of pressure drop is moved from the monitoring requirements section to the
maintenance requirements section. Pressure drop must be monitored as a means of
discovering scrubber operational anomalies that may require maintenance. No specific
pressure drop deviation limit is required, but the monitoring records are required to be kept in
addition to the recycle and makeup water flow rates. Flow rate increases large enough to
cause flooding would be considered malfunctions.
Comment: Four commenters (IV-D-7, IV-D 9, IV-D-11, IV-D-12) stated that facilities
should be allowed to develop their own monitoring protocols. The EPA should set forth
minimum monitoring requirements and allow facilities to develop site specific protocols that
they can justify. Allowing facilities to have the opportunity to propose and have approved
alternative parameters to measure scrubber performance would be consistent with the CAM
rule.
Response: Alternative monitoring options can be approved under §63.8(b) of the
general provisions to this part. This provision is clarified in the final rule.
11.3 MONITORING DURING PROCESS STOPPAGE
Comment: Six commenters (IV-D-1, IV-D-6, IV-D-10, IV-D-11, IV-D-12, IV-D-15)
believe that monitoring should not be required during nonoperating periods such as stoppage
for maintenance and repair. One commenter suggests that a simple process should be used
whereby the facility provides a preventive maintenance plan that incorporates the downturns
and maximizes efforts to reduce acid mist emissions during that period.
Response: Periods of stoppage for maintenance and repairs would be covered under
the Startup, Shutdown, and Malfunction Plan (SSMP). The rule is revised to state that
monitoring scrubber operation is required only while the scrubber is operating. Scrubber
operation is determined by the operation and maintenance requirements.
The rule is also revised to clarify that monitoring acid plant operations is required only
while the plant is operating in production mode. Discussions with plant operators after
proposal have revealed that plants often operate in modes that are designed, for example, to
maintain temperature while acid and iron oxide production are temporarily suspended. These
operations are conducted under conditions that are not predicted to produce byproduct
chlorine.
11-4
-------
11.4 ACID STORAGE VESSELS
Comment: Two commenters (IV-D-3, IV-D-11) stated that storage vessel inspections
should be changed from a monthly to a semiannual basis to be consistent with the requirement
under other Subpart L NESHAP rules. Inspection of control devices on storage vessels should
be conducted at the same frequency as compliance testing on the scrubber.
Response: The reference is to Subpart L of part 61, National Emission Standard for
Benzene Emissions from Coke Byproduct Recovery Plants. The requirement in subpart L is to
monitor connections and seals on each control system that recovers or destroys emissions from
process vessels, tar storage tanks, and tar-intercepting sumps. The EPA believes that the
requirements for this subpart should not be more stringent than those for rules with similar
monitoring requirements and has revised the rule to require semiannual rather than monthly
acid storage vessel inspections.
11.5 PROCESS MONITORING ACID REGENERATION PLANTS
Comment: One commenter (IV-D-3) stated that monitoring excess air feed rate every
eight hours is excessive. Once a day or once per shift is sufficient to reflect performance.
Response: The EPA agrees that once per shift, which was the original intent of the
requirement, is reasonable and is requiring that parameters that determine excess air feed rate
be monitored and recorded at least once per shift.
11.6 FREQUENCY OF PERFORMANCE TESTING
Comment: Three commenters (IV-D 1, IV-D 11, IV-D-12) stated that annual stack
testing is excessive when coupled with parametric monitoring. One commenter recommended
that stack testing only be required if the control device is out of range. The other commenters
recommended testing no more frequently than every 2-1/2 years or every five years.
Response: In lieu of continuous emissions monitoring or other means for determining
continuous compliance, enhanced compliance assurance is established in this rule by
monitoring of relevant operating parameters in combination with routine and preventive
maintenance plus periodic performance testing. Annual testing is typically required hi such
situations. The EPA believes, however, that some flexibility can be allowed hi view of the
requirement to also monitor parameters. The rule is revised to allow facilities to conduct
11-5
-------
performance testing on an alternative schedule that is approved by the applicable permitting
authority but no less frequently than every 2-1/2 years or twice per Title V permit term.
11.7 FREQUENCY OF MONITORING DEVICE CALIBRATION
Comment: One commenter (IV-D-12) believes that parametric monitoring devices
should not have to be calibrated semiannually unless drift is seen. Also, manufacturers may
certify calibration for periods longer than six months.
Response: The EPA disagrees that instruments need not be calibrated if no drift is
seen. While drift may be apparent in some case, conditions may exist that mask drift, such as
a change in parameter value opposite drift. However, the EPA will accept the manufacturer's
stated calibration interval up to one year. If the manufacturer's stated interval is less than one
year, the facility must use that stated interval.
11.8 VIOLATIONS
Comment: One commenter (IV-D-10) believes that there should be no violation if the
operators monitor the required parameters and decline to take corrective action if they note
excursions of the parameters that they determine are not affecting emissions or normal
operations.
Response: Operating within the range of parameter values found at the time of
compliance indicates that the scrubber is operating so that emission limits are being met. The
purpose of monitoring scrubber parameters is to determine when the scrubber is operating
outside the allowed range of values determined at the time of the compliance test. If operating
outside the allowed range, the scrubber may not be meeting its emission limit. Judgement by
the operator as to emission levels from the scrubber, based solely on observation of the
system, is contrary to the intent of the rule.
Comment: Four commenters (IV-D-3, IV-D-10, IV-D-11, IV-D-12) stated that
excursions of control device or acid plant operating parameters should not be considered
violations. Out of range measurements should be treated as indicators of potential problems
requiring further investigation or corrective action, which would be consistent with the CAM
rule. A strong enough relationship between variations in pressure drop or acidity and HC1
emissions has not been demonstrated.
11-6
-------
Response: The proposed rule inadvertently stated that exceedances of scrubber
operating parameters were violations of the emission limit. The intention was to state that
exceedances of acid regeneration plant operating parameters were violations of the emission
limit. The rule is revised to state that excursions of scrubber monitoring parameters only
require corrective action as specified by the maintenance requirements and are not violations
of the emission limit.
Regarding acid plant monitoring parameters, the EPA's policy is that linking
excursions of operating parameters to violations of the emissions limit is preferred but is only
defensible where a strong correlation between the parameters values and emissions can be
demonstrated. The EPA reexamined the appropriateness of the linkage of acid regeneration
plant operating parameters with emissions and agrees with the commenters that a strong
enough correlation has not been demonstrated. The rule is revised so that excursions of acid
regeneration plant operating parameters are a violation of the operational standard and not the
emission limit.
11-7
-------
12.0 REPORTING REQUIREMENTS
12.1 CERTIFICATION OF REPORTS
Comment: One commenter (IV-D-12) stated that reports should only require
certification by the inspector who has intimate knowledge of the operation of the system. The
reporting requirements should not be specified in the rules, but in the Title V reporting
requirements as the individual States deem adequate.
Response: The duty to report is required under §63.10 of the General Provisions of
Title III regulations. The rules for steel pickling and acid regeneration extend the requirement
to a specific category of sources.
One reason for specifying a Responsible Plant Official (see definitions in 40 CFR
parts 70 and 71) is to identify an individual who has sufficient responsibility and authority to
ensure that required actions are properly taken and to understand the consequences of
improper action. However, the EPA recognizes that knowledge of facility systems is
important to understanding how actions should be implemented.
The action taken to accommodate this recognition is the definition of a new category of
official (see responsible maintenance official in Section 1.1.2, Definitions) who has the
knowledge and the authority to sign reports required under the rule.
12.2 COMBINING REPORTS
Comment: One commenter (IV-D-3) believes that the regulations should allow the
facility to combine its malfunction and shutdown plan with its operations and maintenance
plan, if so desired.
Response: The commenter provided no rationale for the statement. For environmental
purposes, the various procedures described above have separate purposes and separate
requirements, and therefore they are retained as separately required plans.
12-1
-------
13.0 RECORDKEEPING REQUIREMENTS
13.1 USEFULNESS OF INFORMATION IN RECORDS
Comment: One commenter (IV-D-3) believes that the requirement for maintaining
startup and shutdown records is ambiguous, burdensome, and of no environmental benefit.
No guidance is provided on what constitutes a startup or shutdown. If required, startup and
shutdown should be defined to exclude the normal stopping and starting of the pickling line
during its daily operation.
Response: The EPA disagrees that no environmental benefit is gained from keeping
startup and shutdown records. These records can be used as an enforcement tool to ensure
continued compliance with environmental rules or to show periods of inactivity when, for
example, emissions would not be expected to occur.
The EPA agrees that maintaining records of normal daily interruptions in line
operations is onerous if not routinely practiced. This is not the intent of the recordkeeping
requirement. Each facility writes its own SSMP and therefore can provide specific definitions
of normal startup and shutdown versus intermittent stops and starts characteristic of daily
operation. However, as part of the SSMP, these definitions are subject to approval by the
facility's permitting authority.
Comment: One commenter (IV-D-3) suggested that for the air pollution control device
recordkeeping, startup and shutdown should be defined to include only "abnormal" cases,
perhaps periods of a day or more. Startup and shutdown recordkeeping of the pickle line is
overly burdensome and would require recordkeeping for routine operations and maintenance,
and should not be required.
Response: As described in the previous response, each facility writes its own SSMP
and can define normal startup and shutdown. It is not the intent of the rule to require
recording of normal line stoppages.
13-1
-------
Comment: One commenter (IV-D-3) stated that the requirement to record "all"
maintenance is overly burdensome. The requirement should be limited to specific, named
items.
Response: Maintenance records provide the facility with a means of ensuring that
required preventive maintenance is performed in a timely manner and of showing compliance
with required actions.
Each facility writes its own maintenance plan. Preventive maintenance performed on
the air pollution control equipment is defined in the plan, in addition to steps that would be
appropriate for on-demand maintenance. The facility is required to record all maintenance
activities specified in the maintenance plan.
13-2
-------
14.0 MISCELLANEOUS COMMENTS
14.1 POLLUTION PREVENTION MEASURES
Comment: One commenter (IV-D-15) stated that pollution prevention and waste
minimization are very important routes to good waste management and eventual compliance
with the standard. Although pollution prevention is not the complete answer, it is imperative
that the final standard clearly define the avenues by which the affected sources may comply
with the standard through pollution prevention.
Response: Two pollution prevention measures have been identified and are already in
common use: using rinse water for scrubbing, and regenerating waste pickle liquor. The
EPA is not aware of any other obvious pollution prevention measures and is therefore reluctant
to be more prescriptive than it is in the final rule. For example, the use of acids other than
HC1 may produce additional waste streams and thus cannot be verified as pollution prevention
options.
14.2 COST OF CONTROLS
Comment: One commenter (IV-D-8) stated that in the preamble to the proposed rule,
on page 49063, last paragraph, the incremental cost of reducing regeneration plant HC1
emissions is given as $2.9 million for capital costs and $1.0 for annual cost. The latter figure
should be $1.0 million.
Response: The commenter is correct.
14-1
-------
!. REPORT NO.
EPA-453/R-98-010b
4. TITLE AND SUBTITLE
National Emission Stai
(NESHAP) for Steel P
Hydrochloric Acid Rej
Background Informatic
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
2.
idards for Hazardous Air Pollutants
ickling HC1 Process Facilities and
generation Plants —
)n for Promulgated Standards
7. AUTHOR(S)
9. PERFORMING ORGANIZATION NAME AND ADDRESS
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Research Triangle Park, NC 27711
12. SPONSORING AGENCY NAME AND ADDRESS
Director, Office of Air Quality Planning and Standards
Office of Air and Radiation
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
3. RECIPIENT'S ACCESSION NO.
5. REPORT DATE
December 1998
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-D6-0014
13. TYPE OF REPORT AND
Final
PERIOD COVERED
14. SPONSORING AGENCY CODE
EPA/200/04
15. SUPPLEMENTARY NOTES
16. ABSTRACT
A rule regulating hazardous air pollutant emissions from the steel pickling (HC1 process) industry
is being promulgated under the authority of sections 101, 112, 114, 116, and 301 of the Clean Air
Act. This document presents background information for the final rule that includes a summary of
rule changes since proposal, a summary of final rule impacts, a list of commenters on the proposed
rule, and detailed comments on the proposed rule and responses by the EPA. The rule is
promulgated under 40 CFR 63, subpart CCC.
17.
a. DESCRIPTORS
KEY WORDS AND DOCUMENT ANALYSIS
b. IDENTIFIERS/OPEN ENDED TERMS
Air Pollution Air Pollution Control
Hazardous Air Pollutants Scrubbers
Steel Pickling
National Emission Standards
Hydrochloric Acid
Hydrogen Chloride
Chlorine
18. DISTRIBUTION STATEMENT
Release Unlimited
19. SECURITY CLASS (This Report)
Unclassified
20. SECURITY CLASS (This Page)
Unclassified
c. COSATI Field/Group
21. NO. OF PAGES
72
22. PRICE
EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDITION IS OBSOLETE
------- |