EPA-453/R-02-003
National Emission Standards for
Hazardous Air Pollutants (NESHAP) for
Primary Copper Smelters -
Background Information for
Promulgated Standards
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Metals Group, MD-13
Research Triangle Park, NC 27711
December 2001
-------
This report has been reviewed by the Emission Standards
Division of the Office of Air Quality Planning and Standards
of the United States Environmental Protection Agency 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 (MD-35), U.S. Environmental
Protection Agency, Research Triangle Park, NC 27711, or from
the National Technical Information Services 5285 Port Royal
Road, Springfield, VA 22161.
U.S. Environmental Protection Agency
Region 5, Library (PL-12J)
77 West Jackson Boulevard, 12th Floor
Chicago, 1L 60604-3590
11
-------
-------
Contents
Section
Chapter 1 - Introduction
1.1 Public Participation in Rule Development
1.2 Commenters on Proposed Rule and Supplemental
Proposal
Changes to Copper Industry Since Proposal .
EPA's Response to Comments
1.3
1.4
1-1
1-2
1-6
1-6
Chapter 2 - Particulate Matter Emission Limits
2.1 Use of Particulate Matter as Surrogate Pollutant
for Metal HAP Emissions 2-1
2.2 Consideration of Kennecott Utah Copper Smelter
in MACT Floor Determinations 2-3
2.3 Consideration of Beyond-the-floor Alternatives 2-7
2.4 Emissions from Copper Concentrate Dryers . . . 2-9
2.5 Process Off-gas Emissions from Smelting Furnaces,
Slag Cleaning Vessels, and Batch Converters . . 2-11
2.6 Process Fugitive Emissions from Smelting Furnaces,
Slag Cleaning Vessels, and Batch Converters . . 2-17
2.7 Alternative Emission Limit for Combined Exhaust
Gas Streams 2-22
Chapter 3 - Copper Converter Department Opacity Limits
3.1 Relationship of Opacity to Metal HAP Emissions from
Copper Converter Departments 3-1
3.2 Determination of MACT Floor for Copper Converter
Departments 3-2
3.3 Consideration of Beyond-the-floor Alternatives 3-9
3.4 Method 9 Measurement Errors and Interferences . 3-10
3.5 Use of Method 22 for New Source Standard . . . 3-11
3.6 Effect of Non-HAP Feedstock Impurities on
Method 9 Opacity Readings 3-12
3.7 Validity of Field Opacity Data 3-13
3.8 Achievability of Opacity Limit for Pierce-Smith
Converters 3-14
3.9 Achievability of Opacity Limit for Hoboken
Converters 3-15
3.10 Achievability of Opacity Limit for New Sources 3-16
3.11 "No Visible Emissions" Limit for All Batch
Converters 3-17
3.12 Expressing Opacity Limit as Whole Percent . . . 3-18
3.13 Applicability to Continuous "Bath" Converting
Technology 3-19
111
-------
Contents (continued)
Section
Chapter 4 - Fugitive Dust Work Practice Standards
4.1 Emission Limit Standards for Fugitive Dust
Sources 4-1
4.2 Fugitive Dust Control Plan Requirements . . , . . 4-2
4.3 Duplication of State Implementation
Plan Requirements 4-4
Chapter 5 - Rule Implementation Requirements
5.1 Compliance Dates 5-1
5.2 Performance Test Requirements 5-2
5.3 Control Device Continues Monitoring
Requirements 5-3
5.4 Converter Capture System Inspection
Requirements 5-4
5.5 Baghouse Monitoring and Inspection Requirements 5-5
5.6 Operating Limit for Baghouse Leak Detector
Alarms 5-7
5.7 Exemption for Fugitive Dust Control Baghouses . 5-9
5.8 ESP Monitoring Requirements 5-10
5.9 Operating Parameter Excursions 5-11
5.10 Use of Wet ESP 5-13
Chapter 6 - Control Costs and Economic Impacts
6.1 Capital Investment Cost Estimates 6-1
6.2 Monitoring, Recordkeeping, and Reporting Cost
Estimates 6-2
6.3 Economic Impact Analysis 6-4
Chapter 7 - Legislative Requirements
7.1 Pollution Prevention Act 7-1
7.2 Endangered Species Act 7-3
Chapter 8 - Other Comments on Proposal Preamble
8.1 Major Source Status of Kennecott Utah Copper
Smelter 8-1
8.2 Health Impact Characterization 8-3
8.3 Primary Copper Smelter Descriptions 8-5
Appendix A - Analysis of Eeyond-the-floor Alternatives for
Control of Process Fugitive HAP Emissions from Pierce-Smith
and Hoboken Converters A-l
Appendix B - Summary of Pierce-Smith Converter Opacity
Data for 1-minute Intervals When "Blowing Without
Interferences" B-l
IV
-------
Table
Table 1-1.
Table 1-2.
Table 2-1.
Table 2-2.
Table 3-1.
Tables
List of Public Commenters on Proposed Rule
List of Public Commenters on Supplemental
Proposal
Page
1-3
1-5
Summary of Method 5B Test Results for Sulfuric
Acid Plants at Primary Copper Smelters . . . 2-16
Summary of Method 5 Test Results for Baghouses
Used to Control Process Fugitive Emissions
from Pierce-Smith Converters 2-19
Summary of Data for Pierce-Smith Converter
Secondary Capture Systems
3-7
v
-------
Chapter 1
Introduction
Under the authority of Section 112 of the Clean Air Act
(CAA), the United States Environmental Protection Agency (EPA)
is developing National Emission Standards (NESHAP) for primary
copper smelters. This document presents our responses to
public comments on the proposed rule and the supplemental
proposal.
1.1 Public Participation in Rule Development
We proposed the NESHAP for primary copper smelters on
April 20, 1998 (see 63 FR 19582). A 90-day comment period
(April 20, 1998 to July 20, 1998) was provided to accept
written comments from the public on the proposed rule. (The
original 60-day comment period was extended an additional
30 days at the request of a commenter [see 63 FR 29963,
June 2, 1998].) Also, the opportunity for a public hearing
was provided by us to allow any interested persons to present
oral comments on the proposed rule. However, we did not
receive a request for a public hearing, so a public hearing
was not held.
After our review and evaluation of comments received on
the proposed rule, we concluded that a change to the proposed
standard for the control of process emissions from smelting
furnaces, slag cleaning vessels, and batch converters was
warranted. On June 26, 2000, a supplement to proposed rule
1-1
-------
and notice of public hearing was published in the Federal
Register (see 65 FR 39326). Specifically, instead of the
equipment standard specified in the original proposal, we
proposed a numerical emission standard that would limit the
maximum concentration of total particulate matter in the off-
gas discharged from these processes. We also proposed
additional bag leak detector requirements for baghouses used
to meet the particulate matter emission limits.
A 60-day comment period (June 26, 2000 to August 25,
2000) was provided to accept written comments from the public
on the supplemental proposal. Again, the opportunity for a
public hearing was offered to allow any interested persons to
present oral comments on the supplemental proposal. We did
not receive a request for a public hearing, so a public
hearing was not held.
1.2 Commenters on Proposed Rule and Supplemental Proposal
In response to our request for public comment, we • •
received a total of 11 comment letters regarding the proposed
primary copper smelter NESHAP (63 FR 19582). Two of the
commenters each submitted two separate and distinct comment
letters. In addition, we received a total of eight comment
letters regarding the supplemental proposal (65 FR 39326). A
copy of each comment letter is in the docket for this
rulemaking (Docket No. A-96-22). Table 1-1 lists the
commenter, commenter's affiliation, and docket index number
for each of the comment letters on the original proposal.
Table 1-2 lists this information for the comment letters on
the supplemental proposal.
1-2
-------
Table 1-1. List of Public Commenters on Proposed Rule
Docket A-96-22
Entry
IV-D-1
IV-D-2
IV-D-3
IV-D-4
IV-D-5
IV-D-06
IV-D-7
Commenter Name, Affiliation,
and Address
Krishna Parameswaran
Manager, Regulatory Development Government Affairs
ASARCO Incorporated
180 Maiden lane, New York, NY 10038
May 20, 1998
Jeffrey C. Smith
Executive Director
Institute of Clean Air Companies
1 660 L. Street NW
Suite 11 00
Washington DC 20036-5603
June 1, 1998
Jeffrey T. Smith, Alexis Perez, and Izmir Simarouba
Students, Business and Society, Florida International University
1 5029 SW 96th Terrace
Miami, Florida 33196
June 15, 1998
Wayne H. Leipold
Sr. Environmental Engineer
Cyprus Miami Mining Corporation
P.O. Box4444
Clay pool, Arizona 85532
July 16, 1998
Krishna Parameswaran
Manager, Regulatory Development Government
ASARCO Incorporated
180 Maiden Lane, New York, NY 10038
July 17, 1998
Affairs
Wayne H. Leipold
Sr Environmental Engineer
Cyprus Miami Mining Corporation
P. O. Box 4444
Claypool, Arizona 85532
July 17, 1998
Ursula K Trueman
Utah Division of Air Quality
Department of Environmental Quality
State of Utah
P.O. Box 144820
Salt Lake City, Utah 841 14-4820
July 17, 1998
1-3
-------
Table 1-1. (continued)
Docket A-92-1 6
Entry
IV-D-8
IV-D-9
IV-D-10
IV-D-11
Commenter Name, Affiliation, and Address
David S. Baron
Assistant Director
Arizona Center for Law in the Public Interest
1840 East River Road
Suite 207.
Tucson, Arizona 85718
July 20, 1998
Steven J. Burr
Lewis and Roca Lawyers (on behalf of the Arizona Mining
Association)
40 North Central Ave.
Phoenix, Arizona 85004-4429
July 20, 1998
William J. Adams
Director, Environmental Science
Kennecott Utah Copper Corporation
P.O. Box 6001
Magna, Utah 84044-6001
July 20, 1998
L M. Pruett
Director, Environmental Services Department
Phelps Dodge Corporation
2600 N. Central Avenue
Phoenix, Arizona 85004-3014
July 20, 1998
1-4
-------
Table 1-2. List of Public Commenters on Supplemental Proposal
Docket A-96-22
Entry
IV-D-13
IV-D-14
IV-D-15
IV-D-16
IV-D-17
IV-D-18
IV-D-19
IV-D-20
Commenter Name, Affiliation, and Address
Jeff Messere
School of Engineering
Indiana University
Bloomington, IN 47401
August 21 , 2000
Jeff J. Parker,
Environmental Manager
BMP Copper, Inc.
P.O. Box M
San Manual, Arizona 85004-4429
August 23, 2000
William J. Adams
Director, Environmental Science
Kennecott Utah Copper Corporation
P.O. Box 6001
Magna, Utah 84044-6001
August 24, 2000
Randy E. Brogdon
Gallagher & Kennedy, Attorneys at Law (on behalf of Phelps Dodge Corp.)
2575 East Camelback Road
Phoenix, Arizona 85016-9225
August 24, 2000
Neil Gambell
Environmental Services Manager - Ray Complex
ASARCO Incorporated
P.O. Box 8
Hayden Arizona 85235
August 25, 2000
John Shanahan
Director, Air Quality Government Affairs
National Mining Association
1 130 17th Street, NW
Washington, D.C. 20036
/August 25, 2000
Chuck Shipley
Arizona Mi.iing Association
40 North Central Ave.
Phoenix, Arizona 85004-4429
August 25, 2000
EarthJustice Legal Defense Fund
1625 Massachusetts Ave., NW
Suite 702
Washington, D.C. 20036
August 29, 2000
1-5
-------
1.3 Changes to Copper Industry Since Proposal
Since we proposed the primary copper smel-ter NESHAP,
several changes have occurred in the copper industry in the
United States. First, corporate ownership has changed for
three of the primary copper smelters potentially subject to
this NESHAP. The smelter near Miami, Arizona, owned and
operated by the Cyprus Miami Mining Corporation during the
time we were developing the proposed rule is now owned by the
Phelps Dodge Corporation. The name of this smelter is now the
Phelps Dodge Miami smelter. The smelters located in Hayden,
Arizona and El Paso, Texas were owned and operated by Asarco
Incorporated at the time of rule proposal. As a result of a
corporate merger, Asarco is now a subsidiary of Groupo Mexico,
S.A. de C.V., the third largest producer of copper in the
world.
Second, four of the smelters potentially subject to the
NESHAP have suspended operations and are not producing copper:
the Asarco smelter in El Paso, Texas; the BHP Copper smelter
near San Manuel, Arizona; and both of the Phelps Dodge
smelters in New Mexico. At this time, it is unknown when and,
even if, any or all of these smelters will resume copper
production.
1.4 EPA's Response to Comments
All of the comments we received regarding the primary
copper smelter NESHAP were reviewed and considered. To
clarify and obtain additional information about some specific
comments, we held follow-up discussions with individual
commenters regarding specific issues raised in their written
comments submitted to us during the comment periods. Copies
of correspondence and other information exchanged between us
and the- commenters during the post-comment period are
1-6
-------
available for public inspection in the docket for this
rulemaking. Changes to the proposed rule have been made in
response to specific comments where it was determined to be
appropriate. This document presents a summary of each
substantive comment and our response to the comment. The
comments are grouped by general topic in Chapters 2.through 8
as follows:
Chapter 2 - Part^culate matter emissions limits.
Chapter 3 - Copper converter department opacity limits.
Chapter 4 - Fugitive dust work practice standards.
Chapter 5 - Rule implementation requirements.
Chapter 6 - Control costs and economic impacts.
Chapter 7 - Legislative requirements.
Chapter 8 - Other comments on proposal preamble.
1-7
-------
Chapter 2
Particulate Matter Emission Limits
2.1 Use of Particulate Matter as a Surrogate Pollutant for
Metal HAP Emissions
Comment. One commenter [docket entries IV-D-8 and
IV-D-20] asserted that the CAA section 112 provisions require
us to establish emission standards for control of specific
metal HAP species, and that our decision to use particulate
matter as a surrogate pollutant for the specific metal HAP
species emitted from primary copper smelters is neither
legally nor technically valid. The commenter also asserted
that our use of opacity as a surrogate measure of metal HAP
emissions is inappropriate because we provide no data to
support that the HAP emissions contribute measurably to
opacity, or that reductions in opacity are necessarily
indicative of HAP emission reductions. The commenter claimed
that because total particulate emissions include substantial
non-HAP components, it is possible that a standard to meet a
numerical total particulate matter limit will achieve little
if any reductions in HAP emissions from the source being
controlled. The commenter notes that lead is a HAP emitted
from primary copper smelters and that the EPA has established
lead-specific HAP limits for the secondary lead smelting
NESHAP. Therefore, the commenter concludes that there are
alternatives to relying on total particulate matter as the
sole HAP surrogate for primary copper smelters.
2-1
-------
Response. Our decision to use particulate matter as a
surrogate pollutant for the specific metal HAP species emitted
from primary copper smelters is legally and technically valid.
The CAA does not require us to establish specific emission
limits on each individual species of HAP emitted from a
source. Section 112 of the CAA requires that we promulgate
standards that provide the maximum degree of reduction in HAP
emissions through application of MACT. This CAA section does
not prohibit us from using an appropriate surrogate pollutant
for individual HAP species to confirm the proper use of MACT.
The HAP emissions from primary copper smelters originate
primarily from metal impurities (e.g., arsenic, lead, cadmium,
antimony, and other heavy metal species listed as HAP) that
naturally occur in copper ore concentrates. During the
smelting process of the copper ore concentrates and the
subsequent converting process to produce blister copper, these
HAP metal species either are eliminated in the molten slag
tapped from the process vessels or are vaporized and
discharged in the process vessel off-gas. Upon cooling of the
process off-gas, the volatilized HAP metal species condense,
form aerosols, and behave as particulate matter.
The composition and amounts of metal HAP in the copper
ore concentrates can vary from one smelter to another as well
as over time at individual smelters depending on the ore
deposit from which the copper ore concentrate is derived.
This inherent variability and unpredictability of the metal
HAP compositions and amounts in copper ore concentrates affect
the composition and amount of HAP metals in the process off-
gas emissions. As a result, prescribing individual numerical
emission limits for each HAP metal species (e.g., a specific
emission limit for arsenic, a specific emission limit for
lead, etc.) is impracticable, if not impossible, to do.
Given that prescribing individual numerical emission
2-2
-------
limits for HAP metal is not a practicable approach in this
case, an alternative approach is to use particulate matter as
a surrogate pollutant for the metal HAP emitted from primary
copper smelters. An emission characteristic common to all
primary copper smelters and similar source categories is the
fact that the metal HAP compounds are a component of the
particulate matter contained in the process off-gas discharged
from smelting and converting operations. Strong direct
correlations exist between the emissions of particulate matter
and metal HAP compounds. Emission limits established to
achieve good control of particulate matter will also achieve
good control of metal HAP.
2.2 Consideration of Kennecott Utah Copper Smelter in MACT
Floor Determinations
Comment. Two commenters objected to the exclusion of the
Kennecott Utah Copper smelter from the primary copper smelter
source category definition and from consideration as part of
the MACT floor determination for new and existing sources.
Both commenters argued for a broader definition than that
contained in the April 1998 proposal. They supported a
definition similar to that used in the NSPS and inorganic
arsenic NESHAP that would include smelters using continuous
flash converting technology, like that used at the rebuilt
Kennecott smelter. Both commenters also argued for the need
to include the Kennecott smelter and its continuous flash
converting technology in the MACT floor determination for the
six smelters that employ the more conventional batch
converting technologies (Pierce-Smith and Hoboken).
Response. At the time we initiated work on the NESHAP,
the primary copper smelting source category consisted of seven
smelters, all of which were engaged in the production of anode
copper from copper ore concentrates by first smelting the
2-3
-------
concentrates to obtain molten copper matte in a flash smelting
furnace, and then converting the molten matte to blister
copper using batch converters followed by fire refining and
anode casting. Consequently, every smelter that potentially
could be a major HAP source used either Fierce-Smith
converters (five smelters) or Hoboken converters (one
smelter).
In the intervening years, Kennecott shutdown its existing
smelter at Garfield, Utah, that had used batch converters.
The company built a new smelter at the same location that uses
a flash smelting furnace similar to that used at the other
smelters, and a new continuous flash converter. The Kennecott
smelter is the only domestic smelter that does not use batch
converters, either Fierce-Smith or Hoboken designs, to produce
blister copper.
From the perspective of raw materials processed and final
products shipped, a smelter using batch converting technology
and a smelter using continuous flash converting technology
appear to be similar, both process copper sulfide ore
concentrate and produce anode copper for shipment to an
electrolytic refining facility. We agree that, in general,
the overall function of both smelters is to produce anode
copper from copper ore concentrates. However, there are
significant dissimilarities between how the anode copper is
produced at the smelter using continuous flash converting
technology compared with the smelters using batch converting
technology.
Continuous flash converting allows blister copper to be
produced in a continuous process at the Kennecott smelter
instead of a batch process as is required at the other
smelters. At the Kennecott smelter, molten copper matte
tapped from the continuous flash smelting furnace is first
granulated by quenching with water to form -solid granules of
2-4
-------
copper matte. These matte granules are then ground to a fine
texture, and fed -to the continuous flash converting furnace.
Slag and blister copper produced are tapped from ports near
the bottom of the furnace. Molten slag is transferred from
the furnace to a slag hauler for subsequent disposal. Molten
blister copper is transferred in heated launders directly to
the anode furnace for further fire-refining into anode copper.
Due to its unique design and operation, most of the
process fugitive emission sources associated with smelters
using batch converting are eliminated at the Kennecott
smelter. There are no transfers of molten material in open
ladles between the smelting, converting, and anode refining
departments at the Kennecott smelter. In addition, there are
no fugitive emissions associated with the repeated rolling-out
of converters for charging, skimming, and pouring. Also, only
one continuous flash converting furnace is needed at the
Kennecott smelter compared with the need for three or more
copper converter vessels at the other smelters.
Another difference between continuous flash converting
versus batch converting technology is that blister copper
produced by the continuous flash converter at the Kennecott
smelter contains higher levels of residual sulfur and metal
HAP impurities than levels seen in blister copper produced by
batch converters. As a result, the anode furnace and casting
departments at the Kennecott smelter use controls for sulfur
dioxide and metal HAP emissions that are not needed at
smelters using batch converters.
These differences aside, we have reconsidered whether the
source category definition included in the April 1998 prop.osal
should' be broadened to include smelters using continuous flash
converting technology like the Kennecott smelter. We have
concluded that the definition should be broadened and made
consistent with that used to define primary copper smelters
2-5
-------
pursuant to both the primary copper smelter NSPS and inorganic
arsenic NESHAP. We are changing the definition of primary
copper smelters to mean "any installation or any intermediate
process engaged in the production of copper from copper
sulfide ore concentrates through the use of pyrometallurgical
techniques."
Relative to the inclusion of the Kennecott smelter in the
MACT floor determination, we disagree with the commenters that
primary copper smelters using continuous flash converting
technology should be grouped with primary copper smelters
using batch converting processes for the existing source MACT
floor determination. Section 112 of the CAA provides the
Administrator the discretion to divide categories of sources
into subcategories where appropriate. In establishing such
subcategories for other source categories in the NESHAP
program, we have considered factors such as differences in
process operations (including differences between batch and
continuous operation), emission characteristics, control
device applicability, and opportunities for pollution
prevention.
We' believe that the design and operating differences
between these two classes of copper converting technologies
make these sources so dissimilar with respect to HAP emission
sources, level of HAP emissions, and the subsequent control
measures required to control HAP emissions from these sources
as to warrant the creation of two separate subcategories of
primary copper smelters; primary copper smelters using batch
converting technology and primary copper smelters using
continuous flash converting technology. We thus conclude that
consideration of the Kennecott smelter in the MACT floor
determinations for existing sources within the subcategory of
primary copper smelters using batch converting technology is
inappropriate since it is not among the pool of sources that
2-6
-------
comprises the subcategory.
2.3 Consideration of Beyond-the-floor Alternatives
Comment. One commenter [docket entry IV-D-08] stated
that there is no evidence in the record that we considered any
beyond-the-floor alternatives for reducing HAP emissions from
primary copper smelters. The commenter claimed that we did
not consider alternatives which would provide actual HAP
emission reductions beyond the current levels emitted by the
existing smelters. The commenter also stated that we did not
consider available process changes for refining copper ores
(such as the solvent extraction process) as possible MACT
alternatives.
Response. Section 112(d)(2) allows us to select as MACT
an alternative more stringent than the MACT floor provided
that the HAP control level selected is achievable taking into
consideration cost and any nonair quality health and
environmental impacts and energy requirements. The objective
is to achieve the maximum degree of HAP emissions reduction
without imposing .unreasonable economic or other impacts. We
reviewed and reconsidered our conclusions regarding beyond-
the-floor alternatives for the emission limitation standards.
Section 112(d)(2)(A) does allow us to establish standards
which reduce or eliminate HAP emissions from an affected
source through process changes, substitution of materials, or
other modifications. We are aware that a number of process
modifications and changes for refining copper ores do exist.
However, application of these modifications and processes
either are not applicable to or are not commercially viable
for the existing primary copper smelters using batch copper
converting.
As discussed in our response presented in section 2.1 of
this chapter, most of the process fugitive emission sources
2-7
-------
associated with smelters using batch converting are eliminated
by the continuous flash copper converting technology used at
the Kennecott smelter. However, it is our judgement that even
though a beyond-the-floor alterative requiring the existing
batch converters to be dismantled and replaced with continuous
flash copper technology may be technically feasible to
implement at some or all of the existing smelters potentially
subject to the rule, it is not an economically viable
alternative. The total cost paid for building the new
Kennecott smelter using continuous flash copper converting
technology is on the order of one billion dollars. Even using
as much of the existing smelter equipment as possible, the
total capital investment of replacing the existing batch
copper converting process at a smelter with the new continuous
flash copper converting process would be in hundreds of
millions of dollars. Given the current economic condition of
the copper industry in the United States and the fact that
none of the companies operating primary copper smelters using
batch copper converting plans to change to flash copper
converting, a regulatory requirement to do so would impose an
enormous economic burden on these smelters.
The commenter specifically suggests as a possible beyond-
the-floor alternative replacing the existing copper smelting
process with a solvent extraction process. The majority of
copper ore deposits in the mines that supply the copper ore
concentrates to these smelters are composed of copper sulfide
ores. The solvent extraction process is suitable only for
processing copper oxide ores. The process cannot be used for
the copper sulfide ore concentrates processed at the smelters
potentially subject to the NESHAP.
Material substitution is not an option for controlling
HAP emissions from primary copper smelters. The HAP emissions
from primary copper smelters originate primarily from metal
-------
impurities (e.g., arsenic, lead, cadmium, antimony, and other
heavy metal species listed as HAP) that naturally occur in
copper ore concentrates processed at these smelters. A
beyond-the-floor alternative based on material substitution
would require limiting the levels of metal impurities in the
copper ore concentrate feedstock that could be processed at a
primary copper smelter. Given the priority nature of these
feedstocks, such a limitation is infeasible.
In regards to beyond-the-floor alternatives specifically
for reducing HAP emissions from batch converters, we have
evaluated two beyond-the-floor'alternatives for copper
converter departments based on the control technologies used
at the Asarco El Paso smelter to control air emissions from
the copper converter building. Our analysis of these beyond-
the-f loor alternatives is discussed further in Section 2.5 of
this chapter.
2.4 Emissions from Copper Concentrate Dryers
Comment. One commenter [docket entry IV-D-8] stated that
the proposed particulate matter emission value of 50 mg/dscm
for existing copper concentrate dryers is incorrectly based on
the median limit for existing sources instead of the average
limit for theses sources as required by the CAA. Furthermore,
the commenter claimed we have not provided economic or
technical analyses to demonstrate why a lower level of 23
mg/dscm, as required in the State operating permit for one of
the existing smelters, is not achievable at the other existing
smelters. The commenter stated that the CAA requires us to
select this lower limit.
Response. Section 112(d)(3) of the CAA defines the
minimum or baseline level of HAP emission control that we can
select to be MACT for a particular source, and we refer to
this minimum level as the "MACT floor." For existing sources,
2-9
-------
we are directed by section 112 to define the MACT floor for a
source category with fewer than 30 existing sources to be the
average emission limitation achieved by the best performing
five existing sources (for which we have or reasonably can
obtain emission data). We are not required by the CAA to
select the best performing source as MACT for existing
sources. Nor does the CAA require that "average emission
limitation" must be determined by averaging the emissions data
for all five best performing sources.
We believe that the average emission limitation is an
expression of the central tendency. This central tendency can
be the average (i.e., mean), the median, the mode, or some
other appropriate statistical measure. The mean is determined
by averaging the emissions data for all five best performing
sources. The median is the emission level indicated for the
third best performing source. The mode is the emission level
that occurs most often among the five best performing sources.
Section 112 does not require us to use one of these
expressions over another. Nor does section 112 require us to
use the same method for all affected sources within a source
category. Instead, we determine, for each case, what measure
of central tendency best fits the circumstances considering
the type, quantity, and quality of the available information
and the approach selected for the MACT floor determination
(i.e., information on actual emissions, allowable emissions,
or application of control technoloy).
At the six primary copper smelters potentially subject to
this NESHAP, each existing copper concentrate dryer is vented
to either a baghouse or ESP to meet a particulate matter
emission limit established in the smelter's operating permit.
Four of the dryers must comply with the particulate emission
limit of 50 mg/dscm (0.022 gr/dscf) required by the primary
copper smelter NSPS (40 CFR 60 subpart P). The other two
2-10
-------
dryers are subject to particulate emission limits established
by the State in which the sources are operated. One dryer must
meet a particulate emission limit of 0.01 gr/dscf
(approximately 23 mg/dscm). The second dryer must meet a
particulate emission limit of 0.03 gr/dscf limit
(approximately 69 mg/dscm). For the top five controlled
sources, the average emission level is 45 mg/dscm, the median
is 50 mg/dscm, and the mode is also 50 mg/dscm. The average
value is essentially the same as the median and mode value.
We are selecting a particulate matter emission level of
50 mg/dscm as the MACT floor for existing copper concentrate
dryers. We conclude that there are no reasonable alternatives
beyond the MACT floor for control of process particulate
emissions from existing copper concentrate dryers. Therefore,
we have reaffirmed our original selection of 50 mg/dscm as
MACT and the standard for existing copper concentrate dryers.
2.5 Process Off-gas Emissions from Smelting Furnaces, Slag
Cleaning Vessels, and Batch Converters
Comment. One commenter [docket entry IV-D-8] disagreed
with our original decision to propose an equipment standard
instead of an emission standard for control of metal HAP
emissions in process off-gas from smelting furnaces, slag
cleaning vessels, and batch converters at the affected primary
copper smelters. The commenter argued that we are required by
the CAA to establish an emission standard for these sources
unless it can be demonstrated that prescribing and enforcing
of a numerical limit is not feasible. In the case of the
proposed NESHAP for primary copper smelters, the commenter'
stated that we provided no documentation to support a
determination that it is not feasible to prescribe a numerical
limit for the metal HAP emissions from sulfuric acid plants
operated at primary copper smelters.
2-11
-------
Response. Based on this comment and new information
received after the original proposal, we reconsidered our
selection of the equipment standard for the process off-gas
streams vented from smelting furnaces, slag cleaning vessels,
and batch converters. On June 26, 2000, we published a
supplement to the proposed rule (see 65 FR 39326) that
proposed adding to the equipment standard we originally
proposed for these sources, namely the treatment of all
process off-gas in a oy-product sulfuric acid plant or
equivalent, a numerical emission limit for total particulate
matter in the plant tail gas. A 60-day comment period was
provided on the supplemental proposal. We received comments
regarding the proposed emission limit for the by-product
sulfuric acid plant tail gas. Our response to the comment on
the supplemental proposal is presented below.
Comment. Seven commenters [docket entries IV-D-13,
IV-D-14, IV-D-15, IV-D-16, IV-D-17, IV-D-18, and IV-D-19]
disagreed with our proposal to establish a particulate
emission limit for the tail gas exhaust from the by-product
sulfuric acid plants used to treat the process off-gases
discharged from smelting furnaces, slag cleaning vessels, and
batch converters. Reasons cited include: 1) Method 5 is an
inappropriate test method for measuring HAP concentrations in
acid plant tail gas because Method 5 measures as particulate
matter material that is not HAP (i.e., sulfuric acid mist and
waters of hydration); and 2) the proposed numerical limit is
based on data for only four sources not the five best
performing sources as is required by CAA section 112 for
establishing MACT.
JResponse. For the process off-gases discharged from
smelting furnaces, slag cleaning vessels, and batch
converters, we originally proposed an equipment standard that
2-12
-------
would require these sulfur dioxide'rich process off-gases to
be vented to a by-product sulfuric acid plant with its
ancillary particulate matter precleaning and conditioning
systems, or other type of sulfur recovery process unit capable
of achieving comparable levels of particulate matter removal.
At the time of proposal, all six smelters in the source
category operated by-product sulfuric acid plants.
After careful review and evaluation of 1) comments
received objecting to our use of an equipment standard rather
than a numerical emission limit and 2) new emissions data
obtained since proposal, we concluded that a change in the
proposed standards for process off-gas emissions was
warranted. As a result, we issued a supplement to the
proposed rule (65 FR 39326, June 26, 2000) in which we
proposed a numerical emission standard that would limit the
concentration of total particulate matter in the off-gases
discharged. Specifically, we proposed to set a total
particulate matter emission limit for acid plant tail gas.of
23 milligrams per dry standard cubic meter (mg/dscm) based on
Method 5 measurements.
In response to the commenters concerns regarding the use
of total particulate matter as the surrogate for HAP and the
use of Method 5 for determining compliance, we examined more
closely the suitability of Method 5 for measuring particulate
matter in tail gas from sulfuric acid plants at primary copper
smelters. Method 5 is the basic reference test method used
for determining particulate matter emissions from stationary
sources. The sampling probe and filter temperature specified
for Method 5 (250°F) is below the acid dewpoint for sulfuric
acid. Consequently, when sampling sulfuric acid plant tail
gas by Method 5, condensed sulfuric acid mist and waters of
hydration not driven off at the sampling temperature are
included in the probe wash and filter catch, along with any
2-13
-------
metal HAP contained in the tail gas. Thus, we agree that
establishing and determining compliance with a total
particulate matter emission limit based on Method 5 may
include sulfuric acid mist condensables not related' to the
control or emissions of metal HAP. Based on some limited test
data obtained using Arizona Method Al (a test method adopted
by the State of Arizona for measuring particulate matter in
sulfur containing gas streams that excludes acid condensate),
the condensate may account for as much as 88 percent of the
total particulate catch.
Method 5B was developed specifically to measure
nonsulfuric acid particulate matter in circumstances when
appreciable quantities of condensable sulfuric acid are
present in the stack exhaust to be tested. The procedure is
identical to Method 5 except that the front-half of the Method
5 sampling train is maintained at 320°F instead of 250°F, and
the probe and filter samples are to be heated in an oven to
320°F for 6 hours before weighing. At the higher sampling
temperature, most of the sulfuric acid mist and waters of
hydration present pass through the probe and filter without
condensing. Heating the probe wash residues and sample filter
in an oven before weighing volatilizes any condensed sulfuric
acid that may have collected in the front-half. Because
sulfuric acid mist and waters of hydration are not counted as
part of the total particulate catch, the total particulate
matter concentration value measured in the front-half by
Method 5B will be lower than the concentration value that
would have been measured on the filter using Method 5. Given
the gas stream characteristics of sulfuric acid plant tail
gas, it is our conclusion that Method 5B is the appropriate
test method to use for setting a particulate matter
concentration limit that serves as a surrogate for metal HAP
emissions contained in the tail gas from sulfuric acid plants.
2-14
-------
Lacking any available Method 5B emissions test data to
set an emission limit, we convened a meeting with company
representatives of each of the six smelters potentially
subject to this NESHAP. Two options were considered:
1) derive an emission limit based on the available Method 5
test data and a conversion factor inferred from the limited
Arizona Method 1A test data; or 2) gather actual Method 5B
test data by testing each of the operating by-product sulfuric
acid plants. The consensus view was that Method 5B testing
was needed to establish a credible emission limit.
A test program was planned and implemented jointly by us
and the companies owning the three copper smelters currently
producing copper. The source tests were conducted by an
independent consultant hired by the smelter companies. Four
individual test runs were conducted at each of the three
smelters. To our best knowledge, all of the tests were
conducted at normal smelter production levels and under normal
acid plant operating conditions. The nonsulfuric acid
particulate matter test results expressed in units of
concentration are summarized in the Table 2-1.
We considered two approaches in selecting the level of
the standard: 1) base the emission limit on the highest
credible individual run measured at the three smelters; or 2)
base the limit on the highest three-run average measured at
the highest emitting smelter. If we base the emission limit
on the highest individual run, the standard expressed in
concentration units would be 6.2 mg/dscm. If we base the
emission limit using the highest three-run average (highest
single performance test), the standard would be 5.0 mg/dscm.
In selecting the appropriate level for the emission
limit, consideration must be given to the full range of
smelter process and acid plant operating conditions, which can
reasonably be foreseen to recur, under which the standard is
2-15
-------
Table 2-1. Summary of Method SB Test Results for
Sulfuric Acid Plants at Primary Copper Smelters
Primary
Copper
Smelter
ASARCO
Hayden
Phelps Dodge
Chino
Phelps Dodge
Miami
Test Run
Run 1
Run 2
Run 3
Run 4
Average
Run 1
Run 2
Run 3
Run 5 (a)
Average
Run 1
Run 2
Run3
Run 4
Average
Nonsulfuric Acid
Particulate Matter Concentration
grains per
dry standard cubic
foot
0.00075
0.00104
0.00063
0.00147
0.00097
0.00254
0.00070
0.00269
0.00138
0.00183
0.00036
0.00041
0.00036
0.00039
0.00038
milligrams per
dry standard cubic
meter
1.7
2.4
1.4
3.4
2.2
5.8
1.6
6.2
3.2
4.3
0.9
0.9
0.8
0.9
0.9
(a) A fifth test was required at this smelter because the results for Run 4 were invalidated when the
sampling probe hit the inside of the stack.
2-16
-------
to be achieved. This is especially important where the
emission limit is applied to a gas stream in which the outlet
loading will typically fluctuate within a range of values
during the course of normal operations. After examining the
design and operating conditions of the three acid plants
tested, we can find no discernible differences among the three
plants which would lead us to conclude that one is superior or
inferior to another. In addition, we believe that each test
run was conducted under conditions representative of
acceptable sulfuric acid plant performance.
Based on the above considerations, we believe that the
performance of the sulfuric acid plant under a reasonable
worst case circumstance is best represented by the single
highest individual run, and that selecting this highest value
will ensure that the standard will be met under all
foreseeable acceptable operating conditions. Therefore, we
are selecting 6.2 mg/dscm of nonsulfuric acid particulate
matter based on measurements using Method 5B as the emission
limit for the sulfuric acid plant tail gas.
2.6 Process Fugitive Emissions from Smelting Furnaces, Slag
Cleaning Vessels, and Batch Converters
Comment. Four commenters [docket entries IV-D-5,
IV-D-6, IV-D-9, IV-D-11] stated that the proposed emission
limit of 16 mg/dscm for the process fugitive emissions from
smelting furnaces, slag cleaning vessels, and batch converters
is overly stringent and is not representative of the MACT
floor. The commenters claimed that the source test data we
used to select the value consisted of only a few source tests,
and that these tests do not account for the range of
variability in emissions associated with normal operating
conditions. The commenters recommended that the value of the
standard be increased to 50 mg/dscm to be consistent with the
2-17
-------
particulate matter emission limit that we proposed for
existing copper concentrate dryers.
Response. We selected the application of baghouses as
MACT for controlling process fugitive HAP emissions based on
the control devices used to control fugitive emissions (i.e.,
secondary emissions) from batch converters (see 63 FR 19595
and 19597). Four of the five smelters using secondary hoods
to capture the converter fugitive emissions, vent the captured
gas stream to a baghoase for control (the fifth smelter
employs an ESP). Because the common practice at the smelters
is to vent the emissions captured by the hoods over the
smelting and slag cleaning vessel tapping ports to the same
control device used to control converter secondary emissions,
we also selected use of baghouses as the MACT floor for
controlling process fugitive emissions from the matte and slag
tapping operations at the smelting furnaces and slag cleaning
vessels. Consistent with other NESHAP based on application of
baghouses as MACT for control of particulate matter emissions,
we selected concentration units as the format of the standard.
The data used to select the proposed emission limit
consists of results from four performance tests, one test for
each of the four smelters employing baghouses for the control
of converter secondary emissions. Each test is comprised of
three test runs conducted at the baghouse outlets using
Method 5. The test results are summarized in Table 2-2.
For the proposed emission limit, we selected the highest
average concentration (16 mg/dscm) measured among the four
performance tests. Since proposal we have reexamined the data
and our approach to setting the standard. A close review of
each of the performance tests shows a high degree of
variability and imprecision among individual test runs within
a performance test with the highest measured values ranging
from m to 4^ times the lowest measured values. Given the
2-1!
-------
Table 2-2. Summary of Method 5 Test Results for Baghouses Used to Control
Process Fugitive Emissions from Fierce-Smith Converters
Primary
Copper
Smelter
ASARCO
El Paso
ASARCO
Hayden
Phelps Dodge
Chino
Phelps Dodge
Hidalgo
Test Run
Run 1
Run 2
Run3
Average
Run 1
Run 2
Run 3
Average
Run 1
Run 2
Run 3
Average
Run 1
Run 2
Run3
Average
Total Particulate Matter Concentration
at Baghouse Outlet
grains per
dry standard
cubic foot
0.0082
0.0073
0.0054
0.0070
0.0053
0.0038
0.0027
0.0039
0.009
0.008
0.002
0.006
0.0099
0.0047
0.0028
0.0058
milligrams per
dry standard
cubic meter
18.9
16.8
12.5
16.1
12.2
8.8
6.2
9.0
20.7
18.4
4.6
13.8
22.8
10.8
6.5
13.4
2-19
-------
lack of precision among the test results, we reconsidered
whether relying on the highest three-run average measured at
one smelter truly accounts for the full range of acceptable
process and control device operating conditions which can be
reasonably foreseen to recur. Upon reflection, we believe
that a more conservative and, perhaps, better approach in this
case is to set the standard based on the highest single
credible test run. This will provide better assurance that
the standard is achievable under reasonable worst case
circumstances. Of the 12 individual test'runs, the value of
the highest run and the value selected for the final standard
is 23 mg/dscm.
Comment. One commenter [docket entry IV-D-8] stated that
the existing NESHAP for arsenic emissions from primary copper
smelters (40 CFR 61 subpart 0) establishes a total particulate
emission limit of 11.6 mg/dscm (0.006 gr/dscf) for captured
fugitive process emissions from batch converters, and we do
not explain why a similar value or more stringent value is not
being adopted for this NESHAP established under 40 CFR
part 63.
Response. There are two reasons why the value for the
emission limit for captured fugitive process emissions from
batch converters under this current rulemaking (subpart QQQ in
40 CFR part 63) is different from the limit for these same
source types that we promulgated on August 4, 1986 under the
NESHAP for arsenic emissions from primary copper smelters
under subpart 0 of 40 CFR part 61. First, the CAA statutory
directives that we must follow in developing a NESHAP were
changed by the 1990 Amendments to the CAA. At the time we
were developing 40 CFR 61 subpart 0, section 112 of the CAA
required us to establish standards in a risk management
framework. The 1990 CAA amendments revised section 112 to
2-20
-------
require that we establish standards for HAP to reflect
application of MACT to the sources regulated by the standard.
In other words, the level of the standard established in the
earlier rule for primary copper smelters under part 61 was
risk-based while the level of the standard we selected for the
rule for primary copper smelters under part 63 is control
technology-based. Using the two different approaches resulted
in the selection of different standards. For the NESHAP under
part 61, our selection of the emission limit for captured
fugitive process emissions from batch converters was based on
a computer modeling analysis of health risks to people living
in the vicinity of specific primary copper smelters. For the
NESHAP under part 63, our selection of the emission limit for
process fugitive emissions from copper converters was based on
our assessment of performance test data for the baghouses in
use at smelters to control these emission sources.
A second reason for the different emission limit values
is that the group and operating practices of primary copper
smelters for which we developed the two standards are
different. The standards in 40 CFR 61 subpart 0 were
developed for those primary copper smelters at which the total
annual average arsenic charging rate to the copper converters
at the smelter is equal to or greater than 75 kilograms per
hour (kg/hr) . At the time we developed the NESHAP under
part 61, the only smelter in the United States operating at or
above an annual average total arsenic charging rate of
75 kg/hr was the ASARCO smelter in El Paso, Texas. This
smelter no longer operates at these levels. Today, none of
the primary copper smelters in the United States is subject to
the standards under 40 CFR 61 subpart 0.
2-21
-------
2.7 Alternative Emission Limit for Combined Exhaust Gas
Streams
Comment. Two commenters [docket entries IV-D-5,IV-D-9]
stated that the equation in the proposed rule that would be
used to calculate the alternative emission limit for a
combined exhaust gas stream should be modified to account for
the exhaust gas stream from a slag cleaning vessel.
Response. In developing the primary copper smelter
NESHAP, we recognized that at some smelters the exhaust gas
streams from several affected sources are combined upstream of
the control device, and consequently treated in the same
downstream control device. Also, for new control device
installations, some smelter owners and operators may prefer to
install a single control device to handle a combination of gas
streams from several affected sources based on site-specific
considerations. We provided for these situations in the
proposed rule by including an equation by which the smelter
owner or operator could demonstrate compliance of several
affected sources subject to different total particulate matter
emission limits with an alternative single total particulate
matter emission limit. We intend that this equation provides
for any possible combination of affected sources exhaust gas
streams that are subject to a total particulate emission limit
under the rule.
The version of the equation in the proposed rule did not
provide for the exhaust gas from a slag cleaning vessel that
is not treated in a sulfuric acid plant and is subject to a
separate total particulate emission limit. This is an
oversight. The equation contained in the final rule is
revised to include slag cleaning vessels in the alternative
emission limit calculation.
2-22
-------
Chapter 3
Copper Converter Department Opacity Limits
3.1 Relationship of Opacity to Metal HAP Emissions from
Copper Converter Departments
Comment. One commenter [docket entry IV-D-8] asserted
that our decision to use opacity as a surrogate measure of
metal HAP emissions is inappropriate because we provide no
'data to support that the HAP emissions contribute measurably
to opacity, or that reductions in opacity are necessarily
indicative of HAP emission reductions.
.Response. As discussed in our response in Section 2.1,
we are using particulate matter as a surrogate measure of
metal HAP emissions from primary copper smelters. We did not
state or imply in our proposal that we are using opacity as a
surrogate measure of metal HAP emissions. In lieu of having
specific capture efficiency values, we are using the opacity
of the visible emissions exiting the converter building as an
indicator of converter capture system performance.
During the converting process to produce blister copper,
HAP metal in the copper matte are eliminated in the molten
slag tapped from the converter vessels or are vaporized and
discharged in the process off-gas. Upon cooling of the off-
gas, the volatilized HAP metal species condense, form
aerosols, and behave as particulate matter. Opacity is a
measure of the degree to which transmitted light is obscured.
The opacity of visible emissions that escape capture by the
copper converter primary and secondary hoods is a function of
3-1
-------
the particulate matter being emitted (as well as other factors
such as lighting conditions and observer position).
We conducted field tests at each of the primary copper
smelters that operate Fierce-Smith or Hoboken converters.
Based on these test results, we concluded that the proposed
opacity observation protocol using Method 9 is a reasonable
indicator of the particulate matter emissions which escape
capture by the converter primary and secondary hood systems
when the converters are operating in the blowing mode. Given
that opacity is an indicator of the level of particulate
matter emitted, designing and operating a copper converter
capture system to minimize the visible emissions from the
building will increase the amount of particulate matter
captured and vented to a control device. Given that metal HAP
emissions from copper converters behave as particulate matter,
increasing the level of particulate matter emissions control
will increase the level of metal HAP emissions control.
3.2 Determination of MACT Floor for Copper Converter
Departments
Comment. Several commenters [docket entries IV-D-5,
IV-D-9, IV-D-11] disagreed with our MACT floor determination
for existing Pierce-Smith converters. The commenters claimed
that CAA section 112(d)(3) requires us to determine the MACT
floor for existing sources based on applicable "emissions
limitations" rather than relying on actual emissions data as
we did for the proposed rule. Using an emissions limitations
approach based on application of existing State regulations,
the commenters concluded that the opacity limit for existing
Pierce-Smith converters should be established at a value of
40 percent opacity.
The same commenters stated that if test data on actual
emissions is used for determining the MACT floor for Pierce-
3-2
-------
Smith converters, then the average 'emission limitation should
be represented by the emissions data for the median performing
source of the five best performing sources rather than the
average of the emissions data for all five sources as was done
for the proposed standard. In this case, the commenters
claimed that the median technology for Fierce-Smith converters
is the use of primary and secondary ventilation systems for
the prevention and capture of emissions coupled with air
pollution control devices for sulfur dioxide and particulate
matter control. The commenters identified the controls used
at the Hayden and Hidalgo smelters as the median technology
for Fierce-Smith converters.
Response. We disagree with the commenters' assertion
that CAA section 112(d)(3) requires us to establish MACT
floors for existing sources based on applicable "emissions
limitations." We have and continue to use several approaches
to establishing MACT floors depending on the type and quality
of the available information. Typically, we examine several
approaches and rely on the one best suited for each particular
circumstance. The approaches include: 1) reliance on
information such as test data on actual emissions from the
pool of sources (the best five sources or best 12 percent)
that comprise the best performers; 2) information on
applicable emissions limitations or standards specified in
State and local regulations and/or operating permits; or 3) a
technology approach based on the application of a specific
control technology and accompanying performance data. We
believe that each of these approaches has merit, and we have
relied on using each to various degrees throughout the MACT
program.
The emissions limitations approach to establish the MACT
floor for Fierce-Smith converters was examined at proposal and
dismissed. Of the five smelters in the source category that
3-3
-------
operate Fierce-Smith converters, only three are subject to an
emissions limitation. The converter building at one smelter
is subject to a zero percent opacity limit specified in the
facility's operating permit. The converter buildings at the
two smelters located in Arizona are arguably subject to the
State's general 40 percent opacity limit applicable to process
fugitive emissions from any source. The converter buildings
at the remaining two smelters, both located in New Mexico, are
not subject to an opacity limit. Then and now, the commenters
supported establishing the MACT floor based on the median or
third most stringent emission limitation. Using this
approach, the MACT floor would be 40 percent opacity.
The emissions limitation approach advanced by the
commenters is workable only when the outcome produces a
realistic inference of actual performance of the best
performing sources. This has been affirmed unequivocally by
the DC Circuit Court in Sierra Club vs. EPA, 167F.3d. in which
the court opined that to comply with the statute, the EPA's
method of setting emissions floors must reasonably estimate
the performance of the relevant best performing sources.
Observations made by us and the industry at all five of the
smelters operating Fierce-Smith converters indicate that
actual visible emissions from the converter buildings are
typically in the range of zero percent to 10 percent opacity,
well below the 40 percent opacity value supported by the
commenters. Consequently, we believe that the use of the
emissions limitation approach in this case is not appropriate.
We assessed how using the median technology approach
would affect the selection of the MACT floor for Fierce-Smith
converters. To do so, we evaluated each of the five smelters
operating Fierce-Smith converters to determine the median
performing source based on both performance data and
engineering design. Using either approach, our assessment
3-4
-------
shows that the Chino Mines smelter is the median performing
source of the five smelters that operate Fierce-Smith
converters, not the Hayden or Hidalgo smelters as suggested by
the commenters. In addition, the opacity value prescribed to
the Chino Mines smelter is 3 percent, the same as the value we
proposed for the opacity limit for Fierce-Smith converters
based on averaging opacity data for all five sources.
To select the median technology based on source
performance data, we ranked the converter capture systems used
at the five smelters in order of decreasing performance using
the average overall opacity value for each smelter. This
ranking assumes that the average opacity value is indicative
of the overall capture efficiency of the control system (i.e.,
the lower the opacity, the higher the capture efficiency).
For our assessment, we used the overall average opacity values
rounded to the next highest whole percent for the five
smelters used for the MACT floor determination at proposal.
The results of this ranking show that the best performing
source is the El Paso smelter (zero percent opacity) followed
by, in decreasing order, the San Manual smelter (1 percent
opacity), the Chino Mines smelter (3 percent), the Hidalgo
smelter (5 percent), and the Hayden smelter (8 percent
opacity). The median performing smelter of the five smelters
that operate Fierce-Smith converters is the third best
performer, the Chino Mines smelter.
For the engineering design-based assessment, we first
assembled pertinent information on the primary and secondary
capture systems used at each of the five affected smelters.
The information included hood ventilation rates (both primary
and secondary), converter blowing rates (amount of air blown
through the tuyeres into the molten bath), and detailed
information on the design and physical configurations of each
secondary hood.
3-5
-------
Each of five smelters uses the same basic approach to
capturing emissions from their Fierce-Smith converter during
slag and copper blows: specifically, a retractable primary
hood for capturing the voluminous process emissions generated
during blowing, and a fixed or sliding secondary hood(for
capturing the secondary or fugitive emissions that escape
capture by the primary hood. Although the basic approach used
at each smelter is fundamentally the same, there are, however,
differences among the smelters in both the design and
operation of their primary and secondary capture systems that
affect performance. Table 3-1 presents a summary of salient
formation on the physical and operational differences between
these converter capture systems operated at the five smelters.
The El Paso smelter uses a converter capture system
design that is unique compared with the designs used at any of
the other smelters. Instead of the fixed or sliding secondary
hood designs used by other four smelters, each converter at
the El Paso smelter is equipped with an air curtain secondary
hood. The air curtain hood encloses the sides and back area
around the converter mouth. During converter blowing
operations, a horizontal jet of air flows across the open top
of the enclosure to provide a continuous sheet or curtain of
air that sweeps the process fugitive emissions into an exhaust
hood, and subsequently a particulate control device. Capture
efficiencies greater than 90 percent are achieved using air
curtain hood systems. Also at the El Paso smelter, any
process fugitive emissions that escape capture by the air
curtain hoods are further controlled by evacuating the entire
converter building to a particulate control device. Thus,
effectively 100 percent of the process fugitive emissions from
converter operations at the El Paso smelter are captured.
Clearly, the use of air curtain secondary hoods in combination
with a tertiary building evacuation system represents the best
3-6
-------
Table 3-1. Summary of Data for Fierce-Smith Converter
Secondary Capture Systems
Primary
Copper
Smelter
ASARCO
El Paso
BHP Copper
San Manuel
Phelps Dodge
Chino Mines
Phelps Dodge
Hidalgo
ASARCO
Hayden
Converter
Secondary Capture
System
Air curtain hood +
building evacuation system
2-piece sliding hood
"clamshell" hood
fixed hood
2-piece sliding hood
Ratio of
Primary Hood
Draft to
Converter
Blowing Rate
2.5 to 1
3.8 to 1
2.5 to 1
2.6 to 1
2.2 to 1
Secondary Hood
Draft
During Blowing
(dry scfm
110,000(a)
0
120,000
60,000
50,000
(a) Ventilation rate for air curtain hood, does not include ventilation by building evacuation
system
3-7
-------
capture system technology used at any of the five smelters
that operate Fierce-Smith converters.
We believe that the second best performer is the San
Manuel smelter which relies primarily on primary hood
ventilation to affect capture. The San Manuel smelter is
unique in that it has surplus by-product acid plant,capacity
which allows each of the converter primary hoods to operate at
a substantially higher ventilation rate than is usual for
other smelters. The primary hoods at the San Manuel smelter
are operated at a primary hood ventilation rate to converter
blowing rate ratio of 3.8 to 1. In contrast, for the
converter primary hoods at other smelters the ratios are much
lower (ratios in the range of 2.5 to 1. As evidenced by the
building opacity data for the San Manuel smelter, operation of
the primary hoods at a substantially higher ventilation rate
results in enhanced capture efficiency and minimal fugitive
emissions due to leakage about the primary hood.
Our assessment of the remaining three smelters supports
our earlier finding using the performance data approach; the
median or third best performing smelter is the Chino Mines
smelter. All three smelters operate their primary hoods
similarly and each converter is equipped with a secondary
hood. Each of the secondary hoods is, with minor variations,
similar in design. The principal difference is that the
ventilation rate during converter blowing used for the
secondary hoods at the Chino Mines smelter (120,000 scfm) is
approximately twice that used at the Hayden or Hidalgo
smelters (50,000 scfm and 60,COO scfm, respectively). We
believe that by operating at this substantially higher
ventilation rate, the secondary hood system operated at the
Chino Mines smelter is more effective at capturing the process
fugitive emissions that escape from the converter primary hood
during blowing compared with the secondary capture systems
3-1
-------
used at the other two smelters. It: is thus our conclusion
that the emissions capture system applied at the Chino Mines
smelter is the third best among the five smelters that operate
Fierce-Smith converters.
Regardless of whether we base our assessment of
performance on average opacity or on engineering design, the
smelter the uses the third best performing or median control
technology is the Chino Mines smelter. If we had used the
median technology approach at proposal to select the opacity
limit for smelters that operate Fierce-Smith converters we
would have selected 3 percent, the same value we proposed.
3.3 Consideration of Beyond-the-floor Alternatives
Comment. One commenter [docket entry IV-D-8] stated that
we did not consider potential alternatives beyond the floor
despite our acknowledgment that one of the five existing
smelters using Fierce-Smith converters operates air emissions
controls on the converter building to meet a State operating
permit condition of no visible emissions.
Response. Since proposal we evaluated potential
alternatives beyond the MACT floor for control of fugitive HAP
emissions from batch converters. We considered two
alternatives beyond the MACT floor. The first alternative is
to use air curtain hoods for each batch converter. The second
alternative is to use a converter building evacuation system.
For each alternative, the captured emissions are vented to a
baghouse control device.
For each of six smelters using batch converters, we
prepared estimates of the additional HAP emission reduction
and additional cost to implement each of the two alternatives
in place of the control configuration required by the MACT
floor. The results of our analysis are presented in
Appendix A to this document. Taking into consideration the
3-9
-------
costs of implementing either of the alternative beyond the
MACT floor versus the level of addition emission reduction
estimated to be achieved, we concluded that there are no
reasonable alternatives beyond the MACT floor for control of
process fugitive HAP emissions from existing batch converters.
Therefore, we reaffirmed our selection of the MACT floor
(i.e., use of a secondary mechanical hood system vented to a
baghouse) as the basis for the proposed standards to control
process fugitive HAP emissions from existing batch copper
converting operations.
^
3.4 Method 9 Measurement Errors and Interferences '
Comment. One commenter [docket entry IV-D-11] stated
that use of Method 9 for the proposed test protocol to
determine percent opacity of visible emissions from the
converter buildings does not account for measurement errors
and interferences. The commenter interprets our own error
studies for Method 9 to indicate that the method cannot be
used to accurately read visible emissions with less than
5 percent opacity. While the commenter acknowledged that
compliance with the proposed opacity limit would be determined
by averaging a number of opacity readings, the commenter
believed that averaging mitigates but does not eliminate the
possibility of inaccurate results due to measurement errors
and interferences inherent in the proposed test protocol.
Response. The performance test specified in the rule to
determine compliance with the applicable percent opacity limit
for visible emissions from the building housing the batch
converters requires that the opacity readings be made by a
team of observers using Method 9. The commenter is concerned
that Method 9 cannot be used to read visible emissions
accurately at opacity levels less than 5 percent.
3-10
-------
We recognize that Method 9 does not require or allow the
recording'of individual readings less than 5 percent other
than zero percent. However, this does not mean that it is
inappropriate to set an opacity limit based on a 6-minute or
longer averaging period at a value less than 5 percent, or
that such values are unacceptably inaccurate or that Method 9
cannot be used. For the opacity limit for existing Fierce-
Smith converters, the potential error of the measurement
method (Method 9) is accounted for in two ways. First, we
used data collected using Method 9 in establishing the opacity
limit and, therefore, we believe the limit inherently
incorporates potential field measurement error. Second, we
have determined that the three best performing sources can
easily meet the applicable opacity limit. These sources have
an adequate margin of compliance, with any potential error
associated with typical certified readers included in the
data.
3.5 Use of Method 22 for New Source Standard
Comment. Two commenters [docket entries IV-D-9, IV-D-11]
stated that compliance with the new source standard for copper
converter departments (i.e., zero percent opacity) should be
determined using Method 9 instead of Method 22 as we proposed.
The commenters stated that Method 9 is the method we proposed
to be used for existing sources and it is the method specified
in the State operating permit for the primary copper smelter
upon which we based the new source MACT determination.
Response. We agree that a performance test for
determining compliance with a copper converter department
opacity limit for new sources should be determined using
Method 9 instead of Method 22 for two reasons. First, as
noted by the commenters, compliance with the visible emission
standard for the primary copper smelter upon which new source
3-11
-------
MACT floor is based is, according to the facility's operating
permit, to be determined using Method 9. Second, to use the
same test protocol regardless of source status (i.e.,
existing, new, or reconstructed) provides for consistency in
the implementation and enforcement of the rule.
3.6 Effect of Non-HAP Feedstock Impurities on Method 9
Opacity Readings
Comment. Three commenters [docket entries IV-D-6,
IV-D-9, IV-D-11] stated that we did not consider non-HAP
feedstock impurities and their impact on opacity from the
copper converter department when selecting the opacity limits.
The commenters claimed that zinc is not a listed HAP but the
content of zinc oxide impurities in a copper ore concentrate
can significantly impact the Method 9 opacity readings from
the converter building roof monitors because of zinc's
volatility and density characteristics. One commenter [docket
entry IV-D-11] included a set of theoretical calculations for
the formation of zinc oxide solid above copper matte.
Furthermore, the commenters reported that during the periods
when the visible emission observations were conducted, the
zinc content of the copper ore concentrate being processed was
at the lower end of the zinc content range typically processed
or expected to be processed in the future.
Response. We reviewed the theoretical calculations
submitted by the commenter. These calculations do suggest a
theoretical possibility that the presence of zinc in the
copper matter could affect the opacity of plumes emitted from
the converters. However, there is insufficient information to
verify that the opacity readings collected during our field
test program were affected in any appreciable manner by the
amount of zinc in the copper matte or that the zinc levels in
the copper matte being processed were atypically low at the
3-12
-------
time opacity observations were made. We do not believe that
it is appropriate to adjust an opacity limit arbitrarily to
some value higher that the value established by the test data.
3.7 Validity of Field Opacity Data
Comment. Three commenters [docket entries IV-D-5,
IV-D-9, IV-D-11] stated that the opacity limits for existing
Fierce-Smith converters is not based on representative data
and does- not account for variability within the data sets.
Two of the commenters [docket entries IV-D-5, IV-D-9] stated
that we arbitrarily excluded from the data base certain
opacity readings made at the ASARCO Hayden smelter that the
commenters believe are representative of normal operations at
the smelter.
Response. We believe that the opacity limit for existing
Fierce-Smith converters is based on representative data and
does account for variability within the data sets. At
proposal, we included in the docket a summary of the data we
used for the copper converter opacity analysis (docket entry
II-I-20). We did not arbitrarily exclude from this data base
specific opacity readings made at any of the smelters. In the
document we explain that we excluded from further analysis
only those opacity readings made during periods when converter
operations were not representative of normal smelter
operations or when the opacity observation conditions did not
meet Method 9 criteria. We did exclude opacity data that met
this criteria from the data sets for three of the five
smelters used to establish the opacity limit for Fierce-Smith
converters. For each of these smelters we identified the
specific dates, time periods, and our reasons for exclusion of
the opacity data in Table 1 of the docket entry.
We received no comments regarding our exclusion of
certain opacity data from the ASARCO El Paso and BHP Copper
3-13
-------
smelters from data set used to establish the opacity limit for
Fierce-Smith converters. For the ASARCO Hayden smelter, we
excluded the opacity readings taken on April 30, 1997 from
9:16 a.m. to 10:33 a.m.. We did not exclude these data, as
the commenters assert, on the basis of the value of the
opacity readings. Our rationale for excluding the data
collected during this 78 minute period is based on
observations by EPA personnel familiar with copper converter
operations of the con/erter at the time. It was determined
that high opacity readings recorded at the converter building
roof monitors during this period coincided with the occurrence
of a converter malfunction. Abnormally high emissions were
observed from the converter number 1. Inspection of the
hooding system for this converter verified that there was a
leak in the primary hood system during the period of high
opacity observations. Although company representatives argue
that this leaking condition should be considered to be normal
operation, it is our judgement that a leaking hood is clearly
a malfunction and not representative of good operation.
Consequently, the opacity readings during this period should
not be'used for standard setting.
3.8 Achievability of Opacity Limit for Fierce-Smith
Converters
Comment. Three commenters [docket entries IV-D-5,
IV-D-9, IV-D-11] claimed that the proposed opacity limit for
existing Fierce-Smith converters is set at a value not
achievable by at least two of the five smelters that operate
this type of batch converter. The commenters stated that we
did not identify any new control equipment or modifications
that could be implemented at those existing smelters for which
our opacity observation data indicated do not meet the
proposed opacity limit; nor did we determine the cost,
3-14
-------
economic, environmental, or energy impacts of the controls
these smelters would need to implement to comply with the
limit. The commenters conclude that because the proposed
opacity limits are economically and technically unachievable,
the proposed opacity limit of 3 percent does not meet the
criteria for a MACT standard.
Response. We changed the visible emission standard in
the final rule for existing Fierce-Smith converters to
4 percent opacity. This change was made independent of
comments received on the proposed rule, and our rationale for
the change is presented in the preamble to final rule
promulgation notice. We believe that the 4 'percent opacity
limit for existing Fierce-Smith converters is achievable by
all five existing primary copper smelters using Fierce-Smith
converters and potentially subject to the rule. Using the
field data and following the test protocol of the rule, we
calculated average opacity values for three of the five
smelters that are less than the 4 percent opacity limit -for
existing Fierce-Smith converters specified in the rule. The
calculated average opacity values for the other two smelters
are higher than 4 percent using the field test results.
However, based on our review of the converter capture systems
at these smelters, it is our judgement that the .smelters can
achieve the opacity limit by increasing the ventilation rates
used for the existing secondary hoods.
3.9 Achievability of Opacity Limit for Hoboken Converters
Comment. One commenter [docket entry IV-D-6] stated that
the proposed opacity limit for existing Hoboken converters was
based on a set of opacity readings that was too small to
adequately reflect an achievable emission limit. Furthermore,
the commenter stated that these data are not representative of
normal operating conditions at the one existing smelter using
3-15
-------
Hoboken converters. The commenter submitted additional
opacity data for the existing Hoboken converters. The
commenter stated that these data were more representative of a
two-converter operation which is typical at the smelter and
requested that the data be used to recalculate the opacity
limit.
Response. We examined the new data submitted by the
commenter according to the revised test protocol. It is
important to remember that the test protocol allows
consideration of only those opacity readings that are taken
during converter blowing and when no visible emissions
interferences occur (as defined in the test procedure). Those
opacity readings made when visible emissions interferences
occur are excluded from the calculation. Our analysis of the
new data provided by the commenter yields an average opacity
value of 3.8 percent which supports the 4 percent opacity
limit for Hoboken converters.
3.10 Achievability of Opacity Limit for New Sources
Comment. One commenter [docket entry IV-D-11] stated
that the proposed opacity limit for new copper converter
departments has not been demonstrated to be "achieved in
practice" as required by the Clean Air Act. The commenter
stated that opacity reading da1:a for the smelter upon which
the proposed opacity limit is based does not demonstrate that
no visible emissions from the building housing the copper
converter department can be achieved in practice at all times.
Response. We believe that the opacity limit we selected
for new copper converter departments of zero percent opacity
is in fact demonstrated to be achieved in practice. Field
data gathered for the smelter upon which we based the new
source MACT floor. Furthermore, for this particular smelter,
the requirement to operate with no visible emissions from the
3-16
-------
converter building is specified in the facility's State
operating'permit. Therefore, we believe the requirement under
the final rule for a new or reconstructed copper converter
department to meet a zero percent opacity limit is achievable.
Under the final rule, the owner or operate of a new or
reconstructed copper converter department is not required to
demonstrate that no visible emissions from the building
housing the copper converter department are achieved at all
times. As discussed in the next response, we have revised the
final rule such that the zero percent opacity limit for new or
reconstructed copper converter'departments is determined using
the same test protocol used for existing copper converter
departments. Following this test protocol, the opacity limit
serves as an indicator of converter capture system performance
for those times when a converter is operating in the blowing
mode and when no interferences (as defined in the test
protocol) occur. Opacity readings during periods when
interferences occur are excluded from the calculation. The
final rule requires that at those times when a batch converter
is operating in the blowing mode, the converter capture system
be operated at the ventilation rates and damper settings
established during the most recent test conducted to
demonstrate compliance with the zero percent opacity limit.
The smelter owner or operator is not required by the final
rule to achieve the zero opacity limit at all times regardless
of the copper production operations and other activities
occurring inside the copper converter building.
3.11 "No Visible Emissions" Limit for All Batch Converters
Comment. One commenter [docket entry IV-D-2] stated that
we proposed opacity limits less than 5 percent opacity for
Fierce-Smith and Hoboken converters. The commenter stated
that there is no difference between the proposed limits and
3-17
-------
the no visible emissions limit we proposed for new sources
since opacity levels less than 5 percent are measured by
Method 9 as zero. Therefore, the final rule can be simplified
by setting a single standard of "no visible emissions" for all
batch copper converters.
Response. The rule cannot be simplified by setting a
single "no visible emission" limit for all copper converters.
The standards established for batch converters are based on
application of MACT, as specified by CAA section 112d.
Following the directives of this section, different criteria
are used to establish standards for existing sources and new
sources. This has resulted in opacity limits for existing
Fierce-Smith and Hoboken converters that are below 5 percent
opacity but greater than zero percent opacity. When making
opacity readings using Method 9, the observer does record each
individual opacity reading in 5 percent increments. However,
compliance with the applicable opacity limit for Fierce-Smith
or Hoboken converters is not determined by a single Method 9
opacity reading. Rather, compliance is determined by
averaging many Method 9 opacity readings recorded following
the performance test specified in the rule. Using this
procedure the opacity value compared with the applicable
opacity limit is, at a minimum, the arithmetic average of 960
individual opaciry readings (120 minutes x 4 Method 9 readings
per minute per observer x 2 observers). Averaging such a
large number of data values produce distinct and discernible
average opacity values less than 5 percent but are not zero.
3.12 Expressing Opacity Limit as Whole Percent
Comment. One commenter [docket entry IV-D-8] stated that
in selecting the opacity limits for existing sources, we
improperly rounded up to a whole percent the average opacity
value computed for our data set representing the five best
3-18
-------
performing sources. The commenter believes the opacity limit
should be set at the average value of 2.8 percent.
Response. We established each of the opacity limits
specified in the rule as whole percent values (e.g., 4 percent
and not 4.0 percent). Opacity limit standards in other
Federal, State, and local air regulations are always
established as a whole percent values. We believe that it is
appropriate to do so as well. As a result, we are retaining
the opacity limits as whole percent values.
3.13 Applicability to Continuous "Bath" Converting Technology
Comment. Two commenters [docket entries IV-D-5, IV-D-9]
stated that the new source standards for copper converter
departments should not apply to new or reconstructed sources
that use continuous "bath" converting technology. The
commenters argue that although continuous "bath" converting
technology is not now in widespread use, this technology
eliminates potential air pollutant emission sources associated
with batch converters.
Response. No primary copper smelters in the United States
use continuous "bath" converting technology. It is also our
understanding that none of the smelter owners are planning to
replace the existing Fierce-Smith or Hoboken converters with
this type of converting technology. However, we also
recognize that at some future date it is conceivable that a
continuous "bath" converting technology could be used for a
new or reconstructed copper converter department.
Under the applicability provisions and definitions
specified in the final rule, the rule applies only to those
primary copper smelters at which batch converters are used.
The term "batch converter" is defined in the final rule to
mean either a Fierce-Smith converter or Hoboken converter. If
a smelter does not use a converting process that meet this
3-19
-------
definition, the smelter is not subject to the rule regardless
of whether it uses continuous flash copper converting,
continuous "bath" converting, or some other new, non-batch
copper converting technology.
3-20
-------
Chapter 4
Fugitive Dust Work Practice Standards
4.1 Emission Limit Standards for Fugitive Dust Sources
Comment. One commenter [docket entry IV-D-8] claims that
the CAA requires us to set emission limitations for fugitive
dust sources at primary copper smelters unless we can
demonstrate that a numeric limit is not feasible. The
commenter claims we have provided no documentation to support
such a determination for the fugitive dust sources at primary
copper smelters. The commenter cites Maricopa County (State
of Arizona) "Rule 310 Open Fugitive Dust Sources" as one- •
example where a regulatory agency has established opacity
limits for fugitive dust sources.
Response. In the preamble for the proposed rule we
discuss our rationale for selection of the proposed
requirements that the smelter owner or operator -prepare and
implement a site-specific fugitive dust control plan (see 63
FR 19598). The statutory requirements that we must follow in
establishing standards under a NESHAP are set forth in section
112 of the CAA. Section 112(h) acknowledges that it may not
be feasible to prescribe or enforce a numerical emission
standard for every type of affected source. In these cases,
section 112 (h) (1) allows us to establish work practice
standards (i.e., design, equipment, work practice, operational
standards) in lieu of a numerical emission limit.
4-1
-------
Fugitive dust emissions result from the handling and
storage of dusty material and the entrainment of fine
particles due to wind or mechanically induced forces. Sources
include wind-blown emissions from outdoor stock piles; road
dust from on-site smelter roadways and plant areas due to
vehicular traffic; emissions from material loading and
unloading operations; and conveyors and elevator systems used
to transfer materials within the smelter.
Given the widespread and unconfined/open nature of these
releases, the setting and enforcement of a numerical emission
limit to control fugitive dust emissions from primary copper
smelters is simply not feasible. In addition, we do not
believe that the duty to set numerical emission limitations
extends to opacity limits for fugitive dust sources. We
recognize that opacity limits for open fugitive dust sources
such as Maricopa County Rule 310 have been established by some
State and local regulatory agencies. However, in practice,
determinating compliance with the open source limits is very
difficult and, as a result, the limits are seldom if ever
enforced. Furthermore, these opacity limits are typically
reinforced by companion requirements that the affected source
also develop and implement a fugitive dust control plan.
We strongly believe that the best approach to reducing
potential HAP emissions from fugitive dust sources at primary
copper smelters is through the preparation and strict
adherence to a written, site-specific control plan that
details the control measures to be implemented towards
mitigating emissions from each of the fugitive dust sources at
a given site.
4.2 Fugitive Dust Control Plan Requirements
Comment. One commenter [docket entry IV-D-8] stated that
the proposed requirements for a fugitive dust control plan do
4-2
-------
not ensure effective fugitive dust controls will be
implemented at a smelter because there is no requirement for
approval by the EPA of the plan written by the smelter owner
or operator. Also, the proposed requirements provide no
criteria for determining the adequacy of the plans and no
requirements that control measures be used that achieve
emission reductions consistent with the intent of MACT.
Finally, because there is no requirement for the plan to be
incorporated into the Title V operating permit, members of the
public will have no way of knowing whether the smelter is
complying with the provisions of the plan.
Response. We reconsidered our proposal and decided that
it is appropriate to require that the fugitive dust control
plan which the smelter owner or operator is required to
prepare and adhere to at all times be first reviewed and
approved by the appropriate authority responsible for
enforcement of the plan at the smelter. For the final
standards for fugitive dust sources, we added the requirement
that the fugitive dust control plan must be approved by the
State with delegated authority for enforcement. For the
purpose of complying with the final rule, an existing fugitive
dust control plan may be used provided that this plan
addresses the fugitive dust sources and includes the
information specified in the rule. An existing fugitive dust
control plan that meets these conditions and also has been
incorporated into a State implementation plan is considered to
be approved for the purpose of complying with this
requirement.
For many fugitive dust sources, there are several
different control measures available that are effective for
controlling fugitive dust emissions from the source. Our
review of the fugitive dust control measures currently
implemented at smelters shows that the application of the
4-3
-------
different control measures varies depending on the physical
layout of the smelter, the mix of fugitive dust sources at the
smelter, the local meteorological conditions, and the
preferences of the smelter owner or operator for certain types
of control measures. Rather than try to dictate a universal
set of the specific work practice requirements that must be
used at all smelters subject to the NESHAP, we believe that a
better approach is for each affected owner or operator to
implement appropriate control measures tailored to address the
smelter's individual collection of fugitive dust sources and
site conditions. We believe that this site-specific approach
provides the needed flexibility to allow each smelter owner
and operator to use the fugitive dust control options best
suited for their smelter.
The standards include general criteria for determining
the adequacy of the fugitive dust control plan used to comply
with the rule. These criteria list the specific types of
fugitive dust emission sources that must be addressed in the
plan and provide examples of control measure options we
consider to be appropriate for these sources. Adding more
detailed criteria would diminish the site specific flexibility
we want to provide to the smelters subject to the rule.
4.3 Duplication of State Implementation Plan Requirements
Comment. Two conmenters [docket entries IV-D-5, IV-D-9]
stated that the proposed requirement for a written, fugitive
dust control plan can be read as duplicative of State
implementation -plan (SIP) requirements and other independent
legal obligations. Requiring preparation and implementation
of another written fugitive dust plan is unnecessary when a
similar written plan is required for other purposes. The
commenters recommended that we use the same approach that we
used for the primary lead smelter NESHAP and allow the use of
4-4
-------
any existing written, fugitive dust control plan prepared for
other purposes but that nonetheless address appropriately the
fugitive dust sources listed in the NESHAP.
Response. It is not our intention to establish
requirements under the primary copper smelter NESHAP that
duplicate similar requirements already applicable to an
affected smelter under another existing regulatory requirement
or legal obligation. Furthermore, we recognize that many, if
not all, owners and operators of primary copper smelters have
prepared fugitive dust control plans. If an existing plan
addresses the sources and contains the information we have
specified in the rule to be included in such a plan, there is
no reason for the owner or operator to prepare a new, separate
control plan to fulfill the requirements of the primary copper
smelter NESHAP. Therefore, we are adding language to the
final rule that explicitly states that an owner or operator
may use an existing written plan that has been prepared to
comply with an applicable State implementation plan provided
that the plan addresses the fugitive dust sources identified
and includes the information specified in the final rule.
4-5
-------
Chapter 5
Rule-Implementation Requirements
5.1 Compliance Dates
Comment. Three commenters [docket entries IV-D-5,
IV-D-6, IV-D-11] requested that the compliance date for
existing sources be extended to the full 3 years allowed under
the CAA. The commenters, all companies operating primary
copper smelters 'potentially subject to the NESHAP, claimed
that the control measures required to meet the requirements of
the proposed rule cannot be readily implemented within the
proposed 2-year period. The principal reason expressed by the
commenters for extending the compliance period to 3 years is
the rule will require smelters to plan and implement several
significant changes, some of which cannot be completed within
a 2-year period.
.Response. Section 112 (i) (3) of the CAA directs us to
establish a compliance date for existing sources which
provides for compliance with the applicable standards as
expeditiously as practicable but no later than 3 years after
the effective date of the standards. For the final rule, we
reconsidered our proposed compliance date for existing sources
subject to the primary copper smelter NESHAP. We expect that
many of the existing sources that could be subject to the rule
already have the type of controls in place that are needed to
comply with the standards. However, we also recognize that
the control systems for some existing sources subject to the
5-1
-------
rule will likely need to be upgraded to meet the standards.
To allow smelter owners and operators a reasonable period of
time to design, procure, install, and startup these control
upgrades, we decided to establish the compliance date for
existing sources under the final rule at no later than 3 years
after promulgation.
5.2 Performance Test Requirements
Comment. One commenter [docket entry IV-D-8] stated that
the proposed rule would require only an initial performance
test be performed for control devices used to comply with the
rule. The commenter believes that this requirement will not
provide reliable information regarding the levels of HAP
emissions controlled by these devices over time. The
commenter recommended that the rule require performance
testing be conducted by the owner or operator on a "routine"
basis.
Response. We have reconsidered our requirements for
performance testing of affected sources subject to either
particulate matter emission limits or opacity limits, and have
decided to require in the final rule that the owner or
operator conduct performance testing at least once per year.
This change from the proposal is based on our review of the
performance test intervals required at the existing primary
copper smelters under their State operating permits. We found
that source tests to measure particulate matter emissions from
sources at existing smelters are being routinely performed at
least once per year. Furthermore, the results from many of
these annual tests performed to comply with State permit
requirements also are expected to be used by owners and
operators to demonstrate compliance with relevant standards
under the primary copper smelter NESHAP. Thus, we conclude
that requiring annual testing of the control devices should
5-2
-------
not impose any significant additional cost or burdens on
smelter owners and operators subject to the rule since they
already are required to test annually under their existing
State operating permit.
Continent. One commenter [docket entry IV-D-9] stated that
the emission limits in the rule should not be applied as
instantaneous limits, and include an expression of the
averaging time over which compliance with the limit is
determined.
Response. Compliance with the particulate matter
emission limits in the primary copper smelter NESHA'P is not
determined' based on an instantaneous value. Compliance with
the particulate matter emission limits (both total particulate
and nonsulfuric acid particulate) is determined by conducting
an initial and subsequent annual performance tests according
to the test methods and procedures specified in the rule.
These procedures include an expression of the averaging time
over which compliance with a given emission limit is
determined. Depending on the applicable emission limit, the
measurement of particulate matter concentration is performed
using Method 5 or 5B (the rule allows Method 29 to be used for
measurement of total particulate matter emissions). The test
procedure requires that three sampling runs be performed using
the selected test method. The minimum sampling time for each
run is 60 minutes for total particulate matter emission limits
and 240 minutes for the nonsulfuric acid particulate matter
emission limit. The average value of the results for the
three sampling runs is used to determine compliance with the
applicable emission limit.
5.3 Control Device Continuous Monitoring Requirements
Comment. One commenter [docket entry IV-D-8} disagrees
5-3
-------
with our determination that appropriate techniques are not
available'for continuous monitoring of HAP (or an appropriate
surrogate pollutant) from affected sources. In addition, the
commenters stated that there is no information in the record
to support a direct correlation between the control device
operating parameters we propose to be monitored with actual
HAP emissions from the affected sources.
Response. Under the primary copper smelter NESHAP,
initial compliance of a given affected source at a smelter
with the applicable emission or opacity limit is demonstrated
by performance testing using the procedures and methods
specified in the rule. Our selection of control device
operating parameter monitoring to assure continuous compliance
with the applicable emission limit is not intended to provide
a direct correlation with the actual level of HAP that is
emitted from the controlled affected source. Rather, we are
using control device operating parameter monitoring to verify
that the control device continues to operate at the same set
of conditions as the device was operating when the required
emissions testing was performed to demonstrate compliance with
the applicable emission limit.
5.4 Converter Capture System Inspection Requirements
Comment. Three commenters [docket .entries IV-D-5, IV-D-6,
IV-D-11] stated that the requirement to inspect the batch
converter capture systems on a monthly basis should be limited
to those components of the converter capture system that are
readily accessible during normal operations. The proposed
requirement to visually inspect each month all of the capture
system components is not practical, if not impossible to
achieve. For example, the fan blade inspection that would be
required under the proposed rule can only be performed when
the fan housing is opened and operations must be shutdown to
5-4
-------
do this. Another example is the practicality of inspecting
duct components that are covered with insulation.
Response. The intended purpose of the monthly inspection
is to visually check the accessible components of the capture
system for any defects or damage that could diminish or impair
capture system performance from the level that the capture
system is capable of achieving when it is properly operated
and maintained. We also recognize that certain components of
the capture system, such as the examples cited by the
commenters, cannot be inspected by workers without shutdown of
the process or disassembling components. It would be
impractical to inspect these components on a monthly basis.
Therefore, we have revised the wording of the visual
inspection requirement for capture systems in the final rule
to clarify which capture system components are to be inspected
on a monthly basis. The final rule specifies that the owner
or operator inspect those components of the capture system
that can affect the performance of the system to collect the
gases and fumes emitted from the affected source (e.g., hoods,
exposed ductwork, dampers, pressure senors, damper switches).
During each inspection, the inspector must visually check the
physical appearance of the equipment (e.g., presence of holes,
dents, or other damage in hoods or ductwork) and check the
settings for eacn damper and other devices which can be
adjusted to control flow in the capture system.
5.5 Baghouse Monitoring and Inspection Requirements
Comment. Three commenters [docket entries IV-D-5, IV-D-6,
IV-D-11] stated that the proposed baghouse monitoring and
inspection requirements are overly burdensome and some
requirements are not achievable in practice. In particular,
the commenters stated that much of the information required to
be collected is duplicative of each other or information
5-5
-------
collection requirements under other rules. The bag tension
inspection requirement is not necessary and serves no useful
purpose. The proposed requirement for use of bag leak
detectors does not provide for alternative or equivalent
methods such as continuous opacity monitors. The marginal
benefits of a facility implementing all of the specified
requirements are unclear especially considering the additional
cost burden. One commenter questions whether the bag leak
detection devices required for baghouses under the proposed
rule are commercially "readily available."
Response. We believe that good inspection and
maintenance practices together with timely corrective action
are critical to maintaining the high level of control that a
well designed and operated baghouse can achieve. A regular
inspection and maintenance program is essential for early
detection of bag leaks and other potential baghouse
malfunctions so that the proper corrective actions can be
taken in a timely manner. The baghouse monitoring and
inspection requirements specified in the proposed rule are
reasonable and do not create significant costs or burdens to
smelter owners and operators.
Since proposal of the primary copper smelter NESHAP, we
made changes to the baghouse inspection and monitoring
requirements under other NESHAP. For the final rule, we
updated the baghouse inspection and monitoring requirements to
be consistent with the requirements in these other NESHAP to
the extent that a particular requirement is also applicable
and appropriate for baghouses operated at primary copper
smelters. Specific changes to the baghouse inspection and
monito'ring provisions in the final rule include replacing the
requirement for a quarterly measurement of bag tension to a
quarterly visual check of bag tension.
We have maintained in the final rule the requirements to
5-6
-------
install and use bag leak detectors. Bag leak detectors are
commercially available and are in use on thousands of
baghouses including baghouses at both primary and secondary
lead smelters. The bag leak detection requirements assist in
early detection of baghouse failures allowing for owners and
operators to implement timely corrective action.
5.6 Operating Limit for Baghouse Leak Detector Alarms
Comment. Six commenters [docket entries IV-D-13, IV-D-14,
IV-D-16, IV-D-17, IV-D-18, and IV-D-19] objected to our
proposed 5 percent limit on baghouse leak detector alarms
during each 6-month reporting period. Reasons cited included:
1) the use of baghouse leak detectors for baghouses operated
at copper smelters is unproven technology; 2) the selection of
the proposed alarm time limit is arbitrary; 3) experience of
commenters has shown that the detectors are subject to false
alarms: 4) any limit on baghouse leak detector time should not
include alarms during periods of startup, shutdown, or
malfunction; and 5) what the EPA means by "initiation of
corrective action" is not clear for the purpose of counting
the elapsed alarm time.
Response. The use of baghouse leak detectors is a proven
technology that can provide an effective means for early
detection of bag failures allowing the baghouse operator to
take timely action to correct the problem and minimize
^xcessive particulate matter emissions that would result if
the problem was not promptly addressed. These detectors
currently are used for baghouse applications at primary lead
smelters and other metallurgical facilities with gas stream
characteristics and operating conditions similar to those
control situations at primary copper smelters for which an
owner or operator also may choose to use a baghouse to comply
with the rule requirements. We believe that there is no
5-7
-------
reason why baghouse leak detectors'cannot similarly be used on
baghouses at primary copper smelters.
The selection of the alarm time limit value is not
arbitrary. We selected this value based on our judgement of
an upper limit to the number of alarms that can reasonably be
expected to occur (excluding false alarms) over a 6-month
period for a baghouse for which the owner or operator
implements good inspection and maintenance practices.
We reviewed the proposed language for use of baghouse
leak detectors with respect to concerns raised by the
commenters about false alarms. For the final rule, we have
revised the requirements for baghouse leak detectors to be
consistent with the requirements we promulgated for the
primary lead smelter NESHAP under 40 CFR 63 subpart TTT.
These requirements include provisions which address the
concerns raised by the commenters about counting false alarms
and alarms during start-up, shutdown, or malfunctions in the
alarm time limit compliance calculation. Under the primary
copper smelter NESHAP, alarms are not included in the sum of
alarm times for purpose of calculating the percentage of time
the alarm on the bag leak detection system sounds if it is
determined that an alarm sounds solely as the result of a
malfunction of the bag leak detection system or if the alarm
sounds as result of a condition that is described in the
smelter's SSMP and the procedures in the plan described to
respond to this condition are implemented.
Finally, when an alarm first sounds from the bag leak
detector, we recognize that there are situations when the
cause of the alarm cannot be corrected or fixed immediately or
within a short period of a few hours. The correction of a
torn bag or other problem which can trip the alarm may require
that the baghouse be shutdown to allow facility personnel to
enter the baghouse when it is safe to do so. We revised the
5-1
-------
language in the final rule to clarify that alarm time is
counted as the time elapsed from when the alarm first sounds
until the owner or operator acknowledges the alarm and
determines the cause of the alarm. Alarm time is not the
total time until the problem which tripped the alarm is
corrected.
5.7 Exemption for Fugitive Dust Control Baghouses
Comment. One commenter [docket entry IV-D-2] objected to
our proposal that baghouses used for fugitive dust control be
exempted from the inspection and monitoring requirements
specified in the rule. The commenter believes that there is
no valid reason for this exemption, and that the baghouse
inspection and monitoring requirements should also apply to
baghouses used for fugitive dust control.
Response. We require that each baghouse used to meet a
particulate matter emission limit under the rule to be
operated according to written standard operating procedures.
These procedures describe in detail the inspections,
maintenance practices, bag leak detection alarm operation, and
corrective actions used for the baghouse. We proposed to
exempt baghouses used exclusively for fugitive dust control
(i.e., not those used to meet a particulate matter emission
limit but rather those used to meet work practice standards
for fugitive dust sources) from having to have these written
standard operating procedures. The type of baghouse that we
expect to qualify for this exemption are the small baghouse
units typically used to collect dust emissions from conveyor
transfer points, feed hoppers, and similar material transfer
operations. These small units are subject to the general
requirement in the rule that at all times, including periods
of startup, shutdown, and malfunction, the baghouses are
operated and maintained in manner consistent with good air
5-9
-------
pollution control practices for minimizing emissions. The
units are'exempted only from the requirements for baghouses
that are intended to ensure continuous compliance with a
particulate matter emission limitation. We believe this is a
reasonable exemption and have included it in the final rule.
5.8 ESP Monitoring Requirements
Comment. Two commenters [docket entries IV-D-5, IV-D-9]
stated that the proposed monitoring requirements can be
problematic for electrostatic precipitators (ESP) because
operating parameters such as voltage, amperage, and volumetric
flow are not reliable indicators of ESP performance. The
commenters recommended that as an alternative to complying
with the proposed monitoring requirements, an owner or
operator be allowed to choose to comply with the rule by using
a continuous opacity monitor to detect malfunctions or unusual
operating conditions.
Response. For particulate matter control devices, other
than a baghouse or a venturi wet scrubber, the proposed rule
does not specify the individual operating parameters required
to be monitored. Instead, the proposed rule requires the
owner or operator to select a set of operating parameters
appropriate for the control device design that the owner or
operator determines to be a representative and reliable
indicator of the control device performance. During the
initial performance test to demonstrate compliance with the
applicable particulate matter numerical emission limit
standard, the owner or operator establishes limiting values
for selected operating parameters based on the actual values
measured during the compliance test.
For other NESHAP, we have required the use of opacity
monitors and have established specifications and test
procedures for opacity continuous emission monitoring systems
5-10
-------
in stationary sources under Performance Specification 1 in
40 CFR part 60, appendix B (PS-1). For control devices other
than baghouses and venturi scrubbers, the primary copper
smelter NESHAP does not preclude the use of a continuous
opacity monitor to comply with the rule's monitoring ,
requirements if an owner or operator chooses to so. An owner
or operator may determine that continuous opacity monitoring
is the representative and reliable indicator best suited for
an ESP or another type of control device used to comply with
the standards. In this case, the opacity readings from the
monitoring device would not be used to determine direct
compliance with a numerical emission limit standard. Instead,
the opacity readings would be used as an indicator of the
control device performance compared to the opacity range
established at the time the performance testing was conducted.
However, we still expect that an owner or operator choosing to
use a continuous opacity monitor to comply with the primary
copper smelter NESHAP will use a monitor that meets the design
and installation specification of PS-1.
5.9 Operating Parameter Excursions
Comment. Four commenters addressed the provisions of the
proposed rule that specified when a control device operating
parameter excursion is a violation of the rule [docket entries
IV-D-5, IV-D-6, IV-D-9, IV-D-11]. Three commenters requested
clarification on what exactly constitutes an "operating
parameter excursion." Several commenters claim that our
selection of six or more excursions in a 6-month period to be
a violation of the rule is arbitrary. It is not clear what
data were used by the EPA to select this number. Commenters
stated that our application of operating parameter excursions
as violations of the rule is inconsistent. For venturi
5-11
-------
scrubber all excursions are violations. For capture systems
there is no limit to the number of allowable excursions, and a
violation can only occur if the owner or operator fails to
take corrective action.
Response. For all of our NESHAP currently under
development, we now use an approach for assuring continuous
compliance with standards that is different than the one we
used at the time we proposed the primary copper smelter
NESHAP. The incorporation of this new continuous compliance
approach as requirements in the final primary copper smelter
NESHAP has clarified the requirements and eliminated the
inconsistencies identified by the commenters.
The rule requires that smelter owners and operators
monitor, record, and report any time a requirement or
obligation established by the NESHAP is not met. This
includes during startup, shutdown, or malfunction, regardless
or whether or not such failure is allowed by a NESHAP. This
requirement applies to all affected sources.
The term "operating parameter excursion" that was used in
the proposed rule has been replaced with the term "deviation"
in the final rule. The term "deviation" is explicitly defined
in the rule to mean any instance in which an affected source
subject to this subpart or an owner or operator of such a
source fails to meet any of the following: 1) any requirement
or obligation established by this subpart, including but not
limited to, any emission limitation (including any operating
limit) or work practice standard; 2) any term or condition
that is adopted to implement an applicable requirement in this
subpart and that is included in the operating permit for any
affected source required to obtain such a permit; or 3) any
emission limitation (including any operating limit) or work
practice standard in this subpart during startup, shutdown, or
malfunction, regardless of whether such a failure is permitted
5-12
-------
by the rule. Furthermore, there is no allowable number of
deviations before a violation can occur.
5.10 Use of Wet ESP
Comment. One commenter [docket entry IV-D-2] stated that
a dry ESP in combination with a wet scrubber ("wet ESP") is
likely to be the type of control device needed to comply with
the particulate emission limit standards under the rule
because 1) copper ore concentrate is high in sulfate, and
there is no reason to extract heat; and 2)by specifying
compliance with the particulate matter emission limits be
demonstrated by Method 5, the EPA is effectively requiring a
wet scrubber be use since without a wet scrubber there would
be acid in the probe and the test could not be conducted. Two
commenters [docket entries IV-D-5, IV-D-9] stated that the
control device monitoring requirements, even by adding the
alternative of using continuous opacity monitoring, may not
provide reliable indicators of control device performance for
some newer particulate matter control technologies, such as a
wet ESP.
Response. We do not endorse the use of a baghouse over
any other type of particulate control device for purpose of
complying with the standards under the primary copper smelter
NESHAP. In practice, baghouses, electrostatic precipitators,
and wet scrubbing systems are used at existing primary copper
;melters. The primary copper smelter NESHAP establishes
numerical particulate matter emission limits for specific
types of affected sources. An owner or operator can choose to
use any type of particulate control device, or combination of
devices, to comply with the rule provided that owner or
operator demonstrates that the selected control system
achieves the applicable emission limit standard.
The sulfates in the copper ore concentrates are converted
5-13
-------
to gaseous sulfur oxides during the copper smelting and
converting processes, and are subsequently concentrated in the
process off-gas sent to the by-product sulfuric acid plant.
As discussed in Section 3.1.2, Method 5 is the basic reference
test method used for determining particulate matter emissions
from stationary sources. At the sampling temperature of 250°F
required by the method, sulfuric acid mist and waters of
hydration in the sampled gas stream will condense in the
sampling probe and be included in the probe wash and filter
catch. Thus, sulfuric acid mist condensables not related to
the control or emissions of metal HAP may be counted as
particulate matter. For circumstances when appreciable
quantities of condensable sulfuric acid are present in the
stack exhaust to be tested, Method 5B is used which measures
the nonsulfuric acid particulate matter (i.e., does not
include any condensed sulfuric acid mist and waters of
hydration). In the final rule, compliance with the
particulate emission limit for the by-product sulfuric acid
plant tail gas is measured by Method 5B. There are no
appreciable quantities of condensable sulfuric acid in the gas
streams for which compliance with the applicable particulate
matter emission limit is determined by Method 5 (i.e., the
copper concentrate dryer exhaust gas stream and the captured
process fugitive gas streams from the smelting vessels, slag
cleaning vessels, and batch converters). These gas streams
can be effectively controlled using a baghouse or an ESP
without a wet scrubber to meet the required particulate matter
emission limit as measured by Method 5.
If a smelter owner or operator should choose to use a wet
ESP to comply with one of the particulate emission limit
standards under the rule, the rule does not specify the
individual operating parameters required to be monitored for a
wet ESP. Instead, the rule requires the smelter owner or
5-14
-------
operator to select a set of operating parameters appropriate
for the wet ESP design that the owner or operator determines
to be a representative and reliable indicator of the control
device performance.
5-15
-------
Chapter 6
Control Costs and Economic Impacts
6.1 Capital Investment Cost Estimates
Comment. Two commenters [docket entries IV-D-6, IV-D-11]
stated that our estimated costs for the proposed rule are
significantly understated because these estimates do not
include costs for all of the control measures that will need
to be implemented to achieve the requirements of the proposed
rule. One of the commenters [docket entry IV-D-6] estimates
that the capital costs to install controls for the Phelps
Dodge Miami smelter (formerly called the Cyprus Miami smelter)
to comply with the proposed standards could be on the order of
$6 million (if the smelter is required to install a new
control device to meet the particulate matter emission
limitations under the rule). The second commenter [docket
entry IV-D-11] stated that our cost estimates for the Phelps
Dodge Hidalgo smelter do not include any costs for upgrade of
the copper converter hood system, upgrade of the copper
concentrate dryer control system, and preparation and
implementation a fugitive dust control plan.
Response. We reviewed our estimated costs for the
primary copper smelters to comply with standards under the
final rule. Of the five smelters using Fierce-Smith
converters potentially subject to the rule, we believe that
two of the smelters (The ASARCO Hayden and Phelps Dodge
Hidalgo smelters) will need to install additional air
6-1
-------
pollution control equipment to meet the standards. The
additional controls required at both of these smelters consist
of doubling the converter secondary hood ventilation rate and
venting the secondary hoods to a new baghouse (fabric filter).
The Phelps Dodge Miami smelter operates Hoboken converters.
Based on the opacity data and other information available to
us for this smelter, we believe that the Phelps Dodge Miami
smelter can meet the standards under the final rule without
having to install additional air pollution control equipment.
Based on the air pollution control equipment the two
smelters will need to meet the standards, the total capital
costs for the purchase and installation of controls is
estimated to be $8.2 million. Total annual costs (TAG) of
meeting all of the requirements of the rule including
operating and maintenance costs are estimated to be
$1.7 million per year.
6.2 Monitoring, Recordkeeping, and Reporting Cost Estimates
Comment. One commenter [docket entry IV-D-1] stated that
in the SF-83 filing to the Office of Management and Budget
(OMB), we underestimated the burden associated with the
proposed rule for monitoring, recordkeeping, and reporting.
In particular, the commenter stated that we have
underestimated the burden associated with preparation and
implementation of the startup, shutdown, and malfunction plan
and the fugitive dust control plan. One commenter [docket
entry IV-D-11] stated the general belief that the costs to
comply with the monitoring, reporting and recordkeeping
requirements will be significantly higher than the total cost
we estimated but need not provide specific examples of which
cost components are understated.
Response. We submitted an Information Collection Request
(ICR)(EPA ICR No. 1850.01) for the proposed rule to the Office
6-2
-------
of Management and Budget (OMB) for approval under the
Paperwork Reduction Act, 44 U.S.C. 3501 et seq. A revised
version of the ICR (EPA ICR No. 1850.02) was prepared and
submitted to OMB adding the estimated burden for the emission
standard we proposed as a supplement to the proposed rule (see
65 FR 39326). No other changes were made to the burden
estimates presented in this version of the ICR. To respond to
the public comments we received on the ICR, we have prepared
and submitted to OMB a third version of the ICR document for
the primary copper smelter NESHAP (EPA ICR No. 1850.03).
The reporting and recordkeeping estimates presented in
the third version of the ICR (EPA ICR No. 1850.03) have been
revised to address the changes made to the final rule
affecting the smelter owner or operator's recordkeeping and
reporting requirements. For example, the estimates for the
appropriate information collection activities related to
performance testing have been adjusted to reflect the change
to the final rule requiring annual performance testing of
affected sources. In addition, we have reviewed all of the
public comments we received on ICR No. 1850.01 and ICR No.
1850.02 and adjusted the estimates using the labor hour
requirements recommended by the commenters. We revised our
estimates of the time needed to prepare plans required by the
final rule based on information supplied by the affected
companies on the number of hours actually spent by staff in
preparing similar type plans. Also, additional costs were
added to the estimates for equipment and supplies to prepare
and store the required records and reports.
Specific revisions to the estimates in EPA ICR No.
1850.0*3 include the following changes. Labor hours and costs
for annual performance testing of affected sources as required
by the .final rule were included in the burden estimates for
Years 1, 2, and 3. Based on the information provided by the
6-3
-------
copper companies, we increased our estimate for Year 1 of the
amount of time needed to prepare a startup, shutdown, and
malfunction plan from 40 to 80 hours and to prepare a fugitive
dust control plan from 40 to 100 hours.
The revised annual public reporting and recordkeeping
burden for this collection of information (averaged, over the
first 3 years after the effective date of this rule, and
assuming that all six smelters with batch converters are
operating and subject to the rule) is estimated to total
20,500 labor hours per year at a total annual cost of
$923,000. This estimate includes initial notifications,
preparation of a startup, shutdown, and malfunction plan,
preparation of a fugitive dust control plan, annual
performance testing, semiannual compliance reports, and
recordkeeping. 'Total capital costs associated with the
monitoring equipment over the 3-year period of the ICR is
estimated at $276,000. The total annualized cost of the
monitoring equipment is estimated at $98,000. This estimate
includes the capital, operating, and maintenance costs
associated with the installation and operation of the
monitoring equipment.
6.3 Economic Impact Analysis
Comment. Two commenters [docket entries IV-D-6,
IV-D-11] state that it is improper for us to assess the
-conomic impact of the proposed rule by comparing the
estimated annualized compliance costs as a percentage of
copper sales. One commenter [docket entry IV-D-11] stated
that it is more appropriate to assess the economic impact of
the proposed rule in terms of the percentage of current net
profits because the majority of the costs for implementing the
rule must be incurred by the industry within the next 2 or
3 years.
6-4
-------
Response. We agree in principal that use of net profits
is a more appropriate measure of the impact of the cost of the
rule. However, net profit information is often not available
from public information sources or such data will not be made
available to us by companies. Therefore our economic impact
analyses are based on product price information and company
revenue data which are readily available to the Agency.
We revised our economic impact analysis for the final
rule. The economic impact of the rule is determined by
comparing the annualized costs incurred by each smelter to
their estimated annual copper production revenues. The share
of costs to estimated revenues for the affected smelters range
from a low of 0.004 percent to a high of 0.2 percent. Thus,
compared to the estimated production revenues for each
affected smelter, the total annualized costs are minimal.
Based on the smelter-specific TAC/sales ratios, impacts of the
final rule on the companies owning primary copper smelters are
anticipated to be negligible. The economic impact analysis we
prepared to support this finding is available in the docket
for the rulemaking.
6-5
-------
Chapter 7
Legislative Requirements
1.1 Pollution Prevention Act
Comment. One commenter [docket entry IV-D-8] stated that
we violated the statutory requirements of the Pollution
Prevention Act because we failed to consider available process
changes for refining copper ores (such as the solvent
extraction process).
Response. In developing the primary copper smelter
NESHAP, we have complied fully with the statutory requirements
of the Pollution Prevention Act of 1990 (42 U.S.C. 13101 et
seq., Pub. L. 101-508, November 5, 1990). This act
establishes the national policy for pollution prevention by
declaring that: 1) pollution should be prevented or reduced
whenever feasible; 2) pollution that cannot be prevented or
reduced should be recycled or reused in an
environmentally-safe manner wherever feasible; 3) pollution
that cannot be recycled or reused should be treated; and 4)
disposal or release into the atmosphere should be chosen only
if none of the other options is available.
We assessed the feasibility of implementing different
pollution prevention options at primary copper smelters using
batch copper converting consistent with the directives of the
Pollution Prevention Act. Opportunities for implementing the
policy of the Pollution Prevention Act at these smelters are
basically limited to applications of control measures that
7-1
-------
reduce metal HAP emissions. Material substitutions, process
modifications, or recycling measures are not feasible
pollution prevention options for the types of copper ores
processed at primary copper smelters using batch copper
converting.
The HAP emissions from primary copper smelters originate
primarily from metal impurities (e.g., arsenic, lead, cadmium,
antimony, and other heavy metal species listed as HAP) that
naturally occur in copper sulfide ore concentrates. Each
company obtains most, if not all, of the copper sulfide ore
concentrate that it processes at a given smelter from nearby
open pit copper mines that the company also owns and operates.
The natural concentrations of the trace metals in the ore can
vary significantly within the ore deposit at these mines.
This natural variability makes it difficult for a company to
assure that its mines can provide sufficient quantities of as-
mined ore with metal HAP concentrations consistently below a
specified maximum level.
The copper smelters potentially subject to the primary
copper smelter NESHAP are not designed to process other types
of copper ore (e.g., copper oxide ore). Furthermore, these
smelters were built to be located near the company's mines
that supply the copper sulfide ore concentrates to the
smelters. Purchasing other copper ore concentrates from third
party suppliers is not an economically practical alternative
for these smelters. Thus, switching to copper sulfide ore
concentrates with lower HAP metal contents is not a realistic
option for controlling HAP emissions from the primary copper
smelters potential subject to the NESHAP.
The commenter specifically suggested using a solvent
extraction process for pollution prevention. The solvent
extraction process is suitable only for processing copper
oxide ores. The process cannot be used for the copper sulfide
7-2
-------
ore concentrates processed at the smelters using batch
converters.
7.2 Endangered Species Act
Comment. One commenter [docket entry IV-D-8] stated that
the record does not show any attempt by us to comply with
section 7 of the Endangered Species Act (ESA), 16 U.S.C.
§1536. The commenter stated that numerous species listed
under the ESA inhabit areas potentially impacted by the HAP
emissions from primary copper smelters.
Response. Under section 112(d) of the Clean Air Act
(CAA), 42 U.S.C. 7412(d), as amended in 1990, the EPA is to
set a first generation of emission standards for HAP that
"require the maximum degree of reduction in emissions of
hazardous air pollutants ... that the Administrator, taking
into consideration the cost of achieving such emissions
reduction and any non-air quality health and environmental
impacts ... determines is achievable ...." CAA 112(d). For
new sources, the maximum degree of reduction in emissions
"shall not be less stringent than the emissions control that
is achieved in practice by the best controlled similar source"
and for existing sources, the maximum degrees of reduction in
emissions shall not be less stringent than "the average
emission limitation achieved by the best performing 12 percent
of the existing sources" or the "average emission limitation
achieved by the best performing 5 sources," for categories
with less than thirty sources. Id.
Thus, the .EPA sets the first generation standards under
CAA section 112(d) on the basis of technological
achievability, after considering costs, non-air quality health
and environmental impacts, and energy requirements associated
with implementation of the standards. Cf. Sierra Club v. EPA.
167 F.3d 658, 660 (D.C. Cir. 1990)(describing the parallel
7-3
-------
MACT standard for solid waste incinerators in CAA section 129,
U.S.C. 7429). Congress established this structure for
establishing technology-based HAP emission standards in
response to what had been the slow pace of risk-based HAP
regulation and the many uncertainties inherent in trying to
evaluate and quantify risks posed by HAP emissions. See
generally, H.R. Rep. No. 101-490, pt. 1, at 150-54, 316-18,
322-24 (1990), reprinted in 2 A Legislative History of the
Clean Air Act Amendments of 1990 at 3174-78, 3340-42, 3346-48
(Comm. Print 1993).
Section 112(f) of the CAA also requires a second
generation of risk-based standards (not yet issued) to protect
public health with an ample margin of safety, and to prevent
"adverse environmental effects," which may be needed in order
to address residual risks remaining after application of the
technology-based MACT standards. 42 U.S.C. 7412(f)(2). Under
section 112(a)(7) of the CAA, the "adverse environmental
effects" that are to be addresses in the later residual risk
standards, may include "adverse impacts on populations of
endangered or threatened species [.]" 42 U.S.C. 7412(a)(7).
Although EPA takes into account non-air quality health
and environmental impacts when setting the first generation
technology-based standards, as required under the statute, EPA
does not consider emissions-based impacts on endangered
species to constitute such effects. Rather, Congress directed
EPA to establish emission reduction requirements taking into
account the possible adverse environmental consequences of
increasing other kinds of pollution, such as wastewater that
results when wet scrubbers are used to control air emissions.
In light of the language of sections 112(f) and 112(a)(7),
there is no reason to believe that emissions-based impacts on
listed -species are actually "non-air quality" effects within
the meaning of section 112(d); nor has the commenter suggested
7-4
-------
that this is so.
Consequently, EPA interprets the explicit language and
structure of section 112 as directing the Agency to reserve
consideration of air quality related impacts, including
potential emissions-based effects on listed endangered and
threatened species, in setting HAP regulations for the
subsequent residual risk stage under 112 (f), rather than
evaluating such risks in establishing MACT standards. This is
further evident from che fact that Congress specified the
precise factors the EPA was to consider in setting MACT
standards and deliberately chose not to include air quality
related environmental impacts among those factors.
In fact, it is only at the later residual risk stage,
which occurs 8 years after the promulgation of the technology-
based section 112(d) standards, that EPA can properly and
effectively consider air quality related risk-based
environmental effects, including effects on endangered
species. Prior to that point, under section 112(d), EPA is
without latitude or ability to consider such air quality
related environmental impacts, since doing so would
necessarily slow down and frustrate the first step in the two-
step process Congress established for regulating HAP emissions
under the 1990 Amendments to the CAA.
7-5
-------
Chapter 8
Other Comments on Proposal Preamble
8.1 Major Source Status of Kennecott Utah Copper Smelter
Comment. We received two comments challenging our
statement in the proposal preamble that the 'Kennecott Utah
Copper Corporation smelter located near Garfield, Utah, does
not emit HAP at major source levels and is therefore an area
source. The Utah Department of Environmental Quality (DEQ)
commented that the information that we used to characterize
the emissions potential of the smelter is incorrect or
outdated. Data in the smelter's emission inventory report for
the year 1997 indicate that the smelter did emit and has the
potential to emit HAP at major source levels. The Kennecott
Utah Copper Corporation (hereafter referred to as
"Kennecott"), owner and operator of the smelter, commented and
acknowledged that the HAP emissions from its smelter in 1997
exceeded the major source threshold levels, but that the
company planned to install new air pollution control equipment
in the anode furnace and casting departments that will reduce
HAP emissions, especially emissions of lead compounds, to well
below major source levels.
Response. The proposed rule was developed before any HAP
emissions data were available based on the full-time operation
of the Kennecott smelter. At the time, all the available
evidence indicated that the smelter would not be a "major
source" of HAP emissions because of the smelter's unique
8-1
-------
design and anticipated level of emission control. In their
comments on the proposed rule, the Utah DEQ presented HAP
emissions data obtained in 1997, the first full year of
operation of the new smelter. Contrary to the company's, the
State's, and our expectations, total annual HAP emissions from
the smelter in 1997 exceeded the major source threshold level.
Specifically, lead emissions, the most prominent HAP emitted,
were reported to exceed 23 tons/year. This level is well
above the 10 tons/year single HAP threshold level for major
sources and exceeds substantially the smelter's Title V
permitted lead emission rate of 1.3 pounds per hour, which is
equivalent to about 6 tons/year.
Extensive in-plant testing by Kennecott determined that
the primary source of the excess lead emissions were the two
anode furnaces used to fire refine the blister copper flowing
from the flash converting furnace prior to anode casting. At
the time the combined off-gas from both furnaces was treated
in two high-energy wet scrubbers installed in series and
designed to achieve both sulfur dioxide and particulate matter
control. Testing of the anode furnace off-gas and the
scrubber system outlet gas stream showed much higher levels of
fine particulate and lead emissions than originally
anticipated. Results of particle size measurements performed
on the anode furnace off-gas indicated that more than half of
the particulate matter was less than 1 micron in diameter with
significant portions less than 0.3 microns.
During 1999 and 2000, Kennecott installed additional air
pollution control equipment to better control the fine
particulate and lead compounds in the anode furnace process
off-gas. A quench tower, a lime injection system, and a
baghouse were installed upstream of the two wet scrubbers.
With the installation and startup of the new controls, the
levels of fine particulate matter and HAP metal compounds
1-2
-------
emitted in the anode furnace off-gas have been significantly
reduced. ' Based on results from a month-long test program
conducted in January 2001, total annual lead emissions from
the smelter were determined to be approximately 1.75 tons/year
and the emissions of all metals to be approximately 2.6
tons/year. These annual HAP emissions levels are well below
the 10 tons/year major source threshold level for a single HAP
and 25 tons/year major source threshold level for total HAP.
Consequently, the smelter is no longer a major source of HAP
emissions.
On February 15, 2001, Kenhecott submitted to the Utah DEQ
a notification of compliance with all Title V operating permit
limits and conditions including its lead limit of 1.3 pounds
per hour. The requirements of the smelter's Title V operating
permit are federally-enforceable, and both the State of Utah
and the EPA have authority to take enforcement action should
Kennecott fail to continue to operate the smelter in
compliance with its permitted emission limits.
8.2 Health Impact Characterization
Comment. One commenter [docket entry IV-D-6] disagrees
with our preamble discussion on health effects because it
implies there is an adverse health effect associated with
primary copper smelter air emissions.
Response. The proposal preamble includes a general
discussion of the health impacts to humans from the airborne
exposure to the type of metal HAP that potentially can be
emitted from primary copper smelters. Primary copper smelters
process copper sulfide ore concentrates that contain naturally
occurring metal impurities (e.g., arsenic, lead, cadmium,
antimony, and other heavy metal species) listed as HAP.
During the smelting process of these copper ore concentrates
and the subsequent converting process to produce blister
8-3
-------
copper, these HAP metal species either are eliminated in the
molten slag tapped from the process vessels or are vaporized
and discharged in the process vessel off-gas. Upon cooling of
the process off-gas, the volatilized HAP metal species
condense, form aerosols, and behave as particulate master.
Unless control measures are taken at a smelter to capture and
control the process off-gas containing metal HAP, as well as
control other fugitive HAP emission sources, people living in
the vicinity may be exposed to ambient metal HAP
concentrations which can produce adverse health effects.
We did not state or imply any judgement regarding the
degree or extent of adverse heath effects associated with a
specific primary copper smelter located in the United States.
The extent and degree to which the health effects may be
experienced in the vicinity of a given primary copper smelter
are dependent upon a combination of site-specific factors
including: 1) the level of metal HAP impurities in the copper
ore concentrate processed and effectiveness of control
measures implemented at the smelter; 2) the ambient HAP
concentrations that occur in the area (e.g., as influenced by
the HAP emission rates from the smelter, local meteorological
conditions, and the local terrain); 3) the frequency and
duration that people living in the area are exposed to these
ambient HAP concentrations; 4) the health characteristics of
the exposed people (e.g., genetics, age, pre-existing health
conditions, and lifestyle); and 5) pollution specific
characteristics associated with each of the metal HAP species
(e.g., toxicity, half-life in the environment,
bioaccumulation, and persistence). Irrespective of impact of
these site-specific factors at an individual smelter location,
we believe that there is the potential for adverse health
effects associated with nationwide primary copper smelter air
emissions and, therefore, a NESHAP to ensure adequate control
-------
of metal HAP emissions from primary copper smelters using
batch converting processes is necessary for the protection of
human health and the environment.
8.3 Primary Copper Smelter Descriptions
Comment. Two commenters [docket entries IV-D-6, IV-D-11]
provided minor corrections or clarifications to the technical
descriptions of primary copper smelting as presented in the
proposal- preamble. Most of these comments pertain to how the
description presented in the proposal preamble relates to the
actual operations at the Phelps Dodge primary copper smelters
(including the formerly called Cyprus Miami smelter). In
particular, one commenter [docket entries IV-D-6] noted that
the Phelps Dodge Miami smelter uses the IsaSmelt® smelting
technology which is not flash smelting.
Response. We appreciate receiving information from the
commenters to correct our understanding of the copper
production operations at specific primary copper smelters. In
particular, the following corrections and clarifications are
made to our background description of primary copper smelting
that we presented in Section III.A of the proposal preamble
(see 63 FR 19583) .
Primary copper smelters process copper ore concentrates.
This copper ore concentrate is often obtained from relatively
nearby copper ore mines. However, primary copper smelters can
also process ore concentrate received from copper mines
located hundreds of miles away or, even from mines located
outside the United States. At some smelters operated in the
United States, the copper concentrate receives little or no
additional processing before it is charged to the smelting
furnace. All of the primary copper smelters, with one
exception, dry the concentrate in a rotary or fluidized-bed
dryer to reduce moisture content to a prescribed level. Some
8-5
-------
smelters further process the copper ore concentrate by sizing
and additional crushing to prepare the material to the proper
size needed for feeding into the particular type of smelting
furnace design operated at the facility.
The IsaSmelt® technology used at the Phelps Dodge Miami
smelter (formerly called the Cyprus Miami smelter) in Arizona
is not considered to be a flash smelting technoloy. Because
of the small size of the Isa vessel, the molten mixture is
tapped into an electric furnace with sufficient holding
capacity that there is time for the matte and slag to
separate. Matte and slag are then tapped from this furnace,
whereas in the flash smelting process, both are tapped from
the flash furnace.
Molten copper matte from the smelting furnace is
processed in batch converters to remove most of the iron,
sulfur, and other impurities to yield blister copper. In
addition to molten copper matte, solid materials are charged
to the converters to promote slag formation and for process
operating temperature control. Solid silicate flux materials
are charged with the molten matte to facilitate the formation
of irofi oxide slag. Solid materials other than flux (e.g.,
revert materials, copper scrap) can also be added at various
times during the batch cycle to cool the molten bath in the
converter vessel and maintain the process temperature in its
optimum operating range.
The blister copper produced in the batch converter must
be processed further to reduce the copper that was oxidized in
the converting process before transfer of the copper to a
refining facility. The molten blister copper is transferred
from the batch converters to the anode vessels where air is
blown through the molten blister copper to oxidize any
remaining sulfur. Flux materials may or may not be added
during this process.
J-6
-------
The applicability language of the final rule has been
revised to address other comments. These revisions have
eliminated the need to use in the final rule the term "flash
smelting." We have reviewed the regulatory language of the
final rule to ensure that all references to smelting furnaces
correctly characterize the different smelting furnace types
used at those smelters potentially subject to the primary
copper smelter NESHAP.
i-7
-------
APPENDIX A
Analysis of Beyond-the-floor Alternatives for
Control of Process Fugitive HAP Emissions from
Pierce-Smith and Hoboken Converters
Table A-l. Primary Copper Smelter Batch Converter
Configurations A-3
Table A-2. Estimated Baseline Process Fugitive HAP Emissions
From Pierce-Smith and Hoboken Converters . . A-4
Table A-3. Batch Converter Control Beyond-the-floor
Alternatives A-5
Table A-4. Air Pollution Control Equipment Cost Estimates
for Beyond the-floor Alternatives A-5
Table A-5. Estimated Costs and HAP Emission Reductions for
Alternative 1 ' A-7
Table A-6. Estimated Costs and HAP Emission Reductions for
Alternative 2 A-8
A-l
-------
The following series of Table A-l through A-6
presents the estimates of additional HAP emission reduction
and costs of implementing two beyond-the-floor alternatives
for control of process fugitive HAP emissions from existing
smelters operating Fierce-Smith or Hoboken converters.
A-2
-------
tn
c
SS
3
O)
C
C
o
o
1
0)
c
O
O
x:
o
•4-«
(0
00
CO
l_
0)
a
a
o
O
(Q
.0
(0
_o
2
3
0)
C
c
o
o
t Control Device
1
0
Q.
O)
C
t5
'x
LU
a>
*:
£, O
ffl ^
•~ *A
^ j2
fl) 5
V)
E
UJ
ndary Emissions
Controls
o
o
0)
w
—
1
*8
re w
Prim
lissions
fa
HI
o S
& fl)
E ^
z °
fc
O 0) ^
'fg > W
OQ £ |S*
o
£>« »
n a±i
l&i
0. O V)
I
73 r- CU
:= t (/)
D JU 3
, J^ j|*~
(J) ^* Q)
~c ro
lio
sr
OJ
o
r curtain hood
ed to baghouse
.= +-•
Q)
>
t) „
73 "5 TO
o "°-
1^1
°-|1
cu ">
CO
E
CO
cu
o
CU
Q.
ASARCO
El Paso
0)
c
o
c
'tractable hood
ted to baghouse
cc. £
"5 ^
^is
o ED.
CT3 O o
•ii'i
°"f 1
in
1
CO
CD
CD
si
ASARCO
Hayden
CD
C
O
c
tractable hood
ented to ESP
cu >
o:
75 „
73 •§ TO
§ Q. °-
<5 o o
ll|
(D W
„.
E
CO
CD
CU
^
8.
Q.
O
O
D.
X
CO
CD
C
O
c
tractable hood
ted to baghouse
£|
"C ^
73 ^ J5
1 o o
£||
d) w
^
1
CO
o
CD
b.
0) u)
f *
c§f
(A O
Q. C
cu !c
£«
0)
c
o
c
0)
to
Is
.><73
1 1 CU
0)
t5 ^
73 "5 TO
0 E ^.
(T3 O o
•ii =
°"ll
n
E
CO
o
cu
CL
CU
0)
Q.
CD
C
O
c
0)
o
13
til
w -2 .9
-------
2
u.
w
o
'35
(0
0.
(0
X 0>
§> c
.- o
•a"
c
£ «
(/> O
id-o
0) o
0 -f
o J-
o re
.E -c
^ 4-1
0 -e
(0 E
«
ffl d)
"§ ^
,J2 Q)
Is
+•<
(A
UJ
(VI
re
ES
O (A
11
•2 2
*&
C (~>
!§?
O.|
xl
0) C
£o
+•* W
'5) i-
zi
w o
IS
«s
ry Copper Smelter
re
E
Q.
1 Batch
Converter
Design
u.
^.
c
2
CD
ASARCO
El Paso
|-
0
m
O5
ASARCO
Hayden
L_
^,
O
•
f"
Si
(A O
Q. C
CD x:
n=°
|
0
o
00
CD
D)
21
P
a.
.c
CO
£
0)
Q.
g
^
2
in
CD
O)
•o
O 'c
Qcl
ell
CD
Q.
Hoboken
0)
CO
(U
(/I
ro
E
(/>
(A
CO
-o
CD
CD
(A
CD
re
0
o
_ro
Q.
C
8
C35
C
'x
0)
ro
to
•
o
'
CD
0.
o
a.
0)
w in
CO CD
£> CO
^r
w ro
CO "O
CO •^>
UJ CM
A-'
-------
C/3
O-
(0
5
*=
(0
o
o
^
o
r
•o
c
o
>»
0)
QQ
"o
is
o
o
I
0)
c
o
o
(Q
CO
CO
fl)
XI
_
a^
"lT_o
fl) 'Z
53 fll
ll
3 O
•1°
n
Q.
I
£
f
U.
n
U
(0
c
o
'
c
o
o
£
*^
CO
•o
1
c
'co
•c
3
T-
(U
"ro
c
CD
CO
o
O)
to
CO
O5
oi
E
OJ
CO
^
g
3
CO
5
C7)
'5
.Q
t-
1
o
o
CN
V
"co
C
5
rr
0
CO =
._ to
S
0-0
o.
8
s
2-s
m 0)
•c O)
CO >
— ro
-S
I
I
r
•
Si
.S u' CO
Q. 10 C
CO Q) O
O C -J
- 5 £
to
-------
f
+3
CO
O
O
5=
c
o
Q)
CO
£
(0
to
to
UJ
to
o
o
a
'5
or
UJ
C
o
o
c
o
o
a.
JQ
CO
sis for Cost Estimate
£
n
"o
"D
o
0
8
0
T3
4J
TO
E
43
1/1
UJ
&
8
o
15
3
C
"*
J2
10
o
o
J3
'5.
n
0
•s
0)
II
3 =
— or
O UJ
ll
o
o
._ o
CO ^j x"
(f^ll-
1 2 i» ^ £
t ? * g .£
EPA for its review of the
50/3-83-01 8a, pp. 8-32 ai
melter operating 4 conve
Chemical Engineering PI
next whole million dollar
f<™|I
= rr" .y> 's §-
|||l|
•s ^ — ^ .);
to p (/j o -o
O 00 O CM CO
S
C
g
i=
'E
CO
1o
x_ ^
c
g
E
o
28
T3 >
CO ?
>- CO
^t
- £
n. m T3 X"
fHf f ^
£ 06 2 ~ *
'= T-I <5 1: O
Q- ? -c co .E
EPA for its review of the
50/3-83-01 8a, pp. 8-32 ai
melter operating 4 conve
Chemical Engineering PI
next whole million dollar
>* ^ w CD CD
|~ w-io.
illll
*J "55 *- g ^
O CO 8 CM CO
g
I
00
«/»
c
g
^
E
CO
CM
E
"55 co
C 0) 0
° 3 S1
•^ o o
CO £ CO
3 £ J3
If!
C ft)
^ w
'5
CO
«s
?€
•- ,2
"§ CO
$z
-7
In 92
81-
*!
§i
•«= CO
ic
3= >
o <„
B£
££
(D CO
it
°t.
18fi
§0:0
c < *-
.5 00 °
s< fe
*1
««
o >-
O -Q
O
CO
s
CO
•o
CO
r T3
0 CD
Q.
05 i- '^=
Q) 0) "^
0)
3
O
CO
§ t co
•§ ""s
c o> co
cu £ —
'I ™ 'co
•c y> 2
% 2
!„
i§i
-c o o
1 co 3>
c ^2 CO
O C "O
•^ 3 if
^C CO
ii?
3i°
1H
a> -D Q.
2^c5
— o *=
w
I'-e
« Q
_ w
2|
0.-D
.52 co
r ^«
£S2
A-6
-------
^
2
•js
to
s
«
_>
^
< E
b &
£ <
CO o
C ,2
0 J
S 1
3 3
Tl C
« 1
a ,|
c m
.2
'55
to
E
UJ
Q.
<
TJ
C
AS
£
tl\
0
o
c
0
*3
o
•o
Q)
os-c-
C i1
0 C
8&
Wl ^"^
E
UI
Q.
1
8 =
5^
§1
c «>
J2~
CO C
31
a E
&«a
o
•o
Estimate
in
<
Si
JO
n
h—
£>fe fe
1|1
£51
L.
^^ c
ol 0)
S c 8
m DO
o
TO
O
S
O
s
o
0 o
o «>
%.£
«iD
*^
m
CD
^~
00
CO
in
rvi
0 c
O 4>
o: "5,
< TO
^^
*H.
i^
^_j
CO
o
I
Q.
O
O
Q.
I
m
m
CO
CO
o
-o
•S-I
(U
CL
^
1
CO
§
.a
Q_
CO
TT
CO
O
T-
0}
0)
° p
"1
Q.5
,
T5
m
S
TO
U>
§
£
C
'TO
•e
3
i—
<
s
A-7
-------
CM
Q)
V
re
c
(0
c
o
'•5
o
n:
c
o
'w
<2
ill
Q.
•o
C.
(0
(0
o
o
•o
s
re
E
'•ff
tn
LU
CM
i
IS
£
£
^
|
i
•6
c
o
c
o
"o
3
TJ
0)
OS T
§t
'•2 -2
(0 —
E
UJ
Q.
I
«-£
O ~
11
c —
^ "c"
o o
o =
IE
TO • —
0
m Q.—
i&j
t-
fi{£
an
o
O
^
o
"S"
o
0 o
O <2
%£
WuJ
^
oi
00
CO
CM
ASARCO
Hayden
CO
CM
00
CO
o
Q.
Q.
O
O
Q_
I
m
q
00
CO
CM
05 a?
in o
a. c:
0) !c
O)
00
CO
CM
0>
D)
"O o
10 TJ
s-if
a.
op
o3
E
o
in
00
CO
CM
-------
APPENDIX B
Summary of Pierce-Smith Converter Opacity Data for
1-minute Intervals When
"Blowing Without Interferences"
Table B-l. Summary of 1997 Field Test Data for Fierce-Smith
Converters for 1-minute Intervals When
"Blowing Without Interferences" B-2
Table B-2. Fierce-Smith Converter Opacity Averages For
Simulated Performance Tests Using 1997 Field
Test Data B-3
B-l
-------
Table B-1. Summary of 1997 Field Test Data for Pierce-Smith Converters for
1-minute Intervals When "Blowing Without Interferences"
Primary
Copper
Smelter
ASARCO
El-Paso
BHP Copper
Phelps Dodge
Chino Mines
Phelps Dodge
Hidalgo
ASARCO
Hayden
Total
1 -minute Intervals
in Data Base When
"Blowing Without
Interferences" (a)
236 (c)
311
262
224
167
Smelter Average
Average Opacity for
Total 1 -minute Intervals in Data Base
When "Blowing Without Interferences"
Calculated From
Field Test Data (b)
0 %
0.7 %
2.4 %
3.2 %
9.5 %
3.2 %
Rounded to Next
Whole Percent
0%
1 %
3%
4%
10%
3.6 %
(a) "Blowing without interferences" is when at least one converter is blowing and none of the
interferences listed in the test procedure in the final rule were observed by the inside process
monitor to have occurred during the 2 minutes prior to the clock time for the 1-minute interval. See
test procedure in final rule for details
(b) Opacity field data recorded using the test procedure in the final rule by observer teams during
smelter site visits in April and May of 1997. Average opacity for each 1-minute interval represents
the average of the readings by at least two observers using Method 9 (i.e., average opacity value
for the 1-minute interval is the average of 8 opacity readings [4 readings per observer at 15 second
intervals]).
(c) No inside process monitor during field opacity readings. Data comprised of a total of 236 minutes
of opacity readings at the smelter by a pair of certified opacity readers from the local office of the
Texas Air Control Board.
5-2
-------
Table B-2. Fierce-Smith Converter Opacity Averages
For Simulated Performance Tests Using 1997 Field Test Data
Primary
Copper
Smelter
ASARCO
El Paso
BHP Copper
Phelps Dodge
Chino Mines
Phelps Dodge
Hidalgo
ASARCO
Hayden
Performance Test
Performance Test 1
{1st 120 1 -minute intervals)
Performance Test 2 (b)
(Last 1 16 1 -minute intervals)
Performance Test 1
(1st 120 1 -minute intervals)
Performance Test 2
(2nd 120 1 -minute intervals)
Performance Test 3 (b)
. (Last 71 1 -minute intervals)
Performance Test 1
(1st 120 1 -minute intervals)
Performance Test 2
(2nd 120 1 -minute intervals)
Performance Test 3 (b)
(Last 22 1 -minute intervals)
Performance Test 1
(1st 120 1 -minute intervals)
Performance Test 2 (b)
(Last 104 1 -minute intervals)
Performance Test 1 (c)
(all 167 1 -minute intervals)
Average Opacity for
Performance Test
Calculated From
Field Test Data (a)
0 %
0%
0.3 %
0.4 %
1.8%
2.7 %
2.3 %
1 .5 %
4.9 %
1.2 %
9.5 %
Rounded to Next
Whole Percent
0%
0%
1 %
1%
(d)
3%
3%
(d)
5%
2%
10%
(a) Field test data are average opacity values for 1-minute intervals when "blowing without
interferences" recorded using the test procedure in the final rule by observer teams during smelter
site visits in April and May of 1997 (see Table B-1).
(b) Does not meet minimum number of 120 1-minute intervals "blowing without interferences"
required for a performance test using the procedures in the final rule. However, listed as a
"performance test" in this table for the sole purpose of presenting the additional minutes of data
for this smelter.
(c) Insufficient total number of qualifying 1-minute intervals in data base to simulate two performance
tests Used all 167 minutes for average opacity calculation.
(d) Too few 1 -minute intervals to consider further as a "performance test."
B-3
-------
TECHNICAL REPORT DATA
(Please read Instructions on reverse before completing)
1. REPORT NO
EPA- 453/R-02-003
3 RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
National Emission Standards for Hazardous Air Pollutants
(NESHAP) for Primary Copper Smelters-Background
Information for Promulgated Standards
5. REPORT DATE
December, 2001
6 PERFORMING ORGANIZATION CODE
v AUTHOR(S) Research Triangle Institute
8. PERFORMING ORGANIZATION REPORT NO.
EPA-453/R-02-003
9. PERFORMING ORGANIZATION NAME AND ADDRESS
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Research Triangle Park, NC 27711
10 PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-D6-0014
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
13. TYPE OF REPORT AND PERIOD COVERED
Background Information for
Promulgated Standards
14. SPONSORING AGENCY CODE
EPA/200/04
15. SUPPLEMENTARY NOTES
16. ABSTRACT
The response to comments made on the proposed NESHAP Rule and Supplemental proposal for
Primary Copper Smelters are documented in this Background Document.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b. IDENTIFIERS/OPEN ENDED TERMS
c. COSAT1 Field/Groueshap
NESHAP, Copper Smelting, Final Rule,
Background Document.
Air Pollution control
18. DISTRIBUTION STATEMENT
Release Unlimited
19. SECURITY CLASS (Report)
Unclassified
21. NO. OF PAGES
103
20. SECURITY CLASS (Page)
Unclassified
22. PRICE
EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDITION IS OBSOLETE
-------
U.S. Environmental Protection Agency
Region 5, Library (PL-12J)
77 West Jackson Boulevard, 12th Floor
Cniwgo, IL 60604-3590
------- |