National Emission Standards for Hazardous Air Pollutants for Source Categories; National Emission Standards for Hazardous Air Pollutants for Secondary Aluminum Production

SUMMARY: This action proposes national emission standards for hazardous

air pollutants (NESHAP) for new and existing sources at secondary

aluminum production facilities. Hazardous air pollutants (HAPs) emitted

by the facilities that would be regulated by this proposed rule include

HAP organics, inorganic HAPs (hydrogen chloride, hydrogen fluoride, and

chlorine), and particulate HAP metals. Some of these pollutants,

including 2,3,7,8-tetrachlorodibenzo-p-dioxin, are considered to be

known or suspected carcinogens and all can cause toxic effects

following sufficient exposure. Emissions of other pollutants include

particulate matter and volatile organic compounds.

The standards are proposed under the authority of section 112(d) of

the Clean Air Act (the Act) and are based on the Administrator's

determination that secondary aluminum production plants are major

sources of HAP emissions and emit several of the HAPs listed in section

112(b) of the Act from the various process operations found within the

industry. The proposed NESHAP would reduce risks to public health and

environment by requiring secondary aluminum production plants to meet

emission standards reflecting application of the maximum available

control technology (MACT). Secondary aluminum production plants that

are area sources would be subject to limitations on emissions of

dioxins and furans (D/F) only. Implementation of the proposed NESHAP

would reduce emissions of HAPs and other pollutants by about 16,600

megagrams per year (Mg/yr) (18,300 tons per year (tpy)).

DATES: Comments. The EPA will accept comments on the proposed rule

until April 12, 1999.

Public Hearing. If anyone contacts EPA requesting to speak at a

public hearing by March 4, 1999, a public hearing will be held on March

15, 1999 beginning at 10 a.m., at the EPA Office of Administration

Auditorium, Research Triangle Park, NC. For more information, see

section VII.B of the SUPPLEMENTARY INFORMATION section.

ADDRESSES: Comments. Interested parties may submit written comments (in

duplicate, if possible) to Docket No. A-92-61 at the following address:

Air and Radiation Docket and Information Center (6102), U.S.

Environmental Protection Agency, 401 M Street, SW., Washington, DC

20460. The EPA requests that a separate copy of the comments also be

sent to the contact person listed below. The docket is located at the

above address in Room M-1500, Waterside Mall (ground floor).

A copy of today's document, technical background information, and

other materials relating to this rulemaking are available for review in

the docket. Copies of this information may be obtained by request from

the Air Docket by calling (202) 260-7548. A reasonable fee may be

charged for copying docket materials.

Public Hearing. If anyone contacts the EPA requesting a public

hearing by the required date (see DATES), the public hearing will be

held at the EPA Office of Administration Auditorium, Research Triangle

Park, NC. Persons interested in making oral presentations should notify

Ms. Tanya Medley, Minerals and Inorganic Chemicals Group, Emission

Standards Division (MD-13), U. S. Environmental Protection Agency,

Research Triangle Park, NC 27711, telephone number (919) 541-5422.

FOR FURTHER INFORMATION CONTACT: For information concerning the

proposed regulation, contact Juan Santiago, Minerals and Inorganic

Chemicals Group, U.S. Environmental Protection Agency, Research

Triangle Park, NC 27711, telephone number (919) 541-1084, facsimile

number (919) 541-5600, electronic mail address,

``[email protected].''

SUPPLEMENTARY INFORMATION:

Regulated Entities

Entities potentially regulated by this action are ``secondary

aluminum production facilities'' using post-consumer scrap, aluminum

scrap, ingots, foundry returns, and/or dross as the raw material and

operating one or more of the following affected sources: Scrap

shredders, scrap dryer/delacquering/decoating kilns, chip dryers, group

2 process furnaces (i.e., clean charge furnaces using no reactive

flux), sweat furnaces, dross-only furnaces, rotary dross coolers,

secondary aluminum processing units, new and reconstructed group 1

furnaces (i. e., melting, holding, fluxing, refining or alloying), and

new and reconstructed in-line fluxers. The EPA identified more than 400

facilities which include one or more of these affected sources, 86 of

which are estimated to be major sources. Most establishments are

included in SIC 3341 (Secondary Smelting and Refining of Nonferrous

Metals), although others may fall in SIC 3353 (Aluminum Sheet, Plate,

and Foil), SIC 3354 (Aluminum Extruded Products), and SIC 3355

(Aluminum Rolling and Drawing NEC). Affected sources at facilities that

are major sources of HAPs would be regulated under the proposed

standards. In addition, emissions of dioxins and furans (D/F) from

affected sources at facilities that are area sources of HAPs would also

be regulated.

The proposed standards would not apply to facilities in SIC 336

(Nonferrous Foundries/Casting), such as manufacturers of aluminum die

castings (SIC 3363) that use only clean aluminum and aluminum foundries

(SIC 3365) that process only clean aluminum. Secondary aluminum

production facilities that are collocated with primary aluminum

production are regulated under the proposed standard.

Regulated categories and entities include:

------------------------------------------------------------------------

Category Examples of regulated entities

------------------------------------------------------------------------

Industry............................... Owners or operators of

secondary aluminum production

facilities in SIC 3341, 3353,

3354, 3355, or that are

collocated with primary

aluminum production

facilities, that are major

sources of HAPs, or that emit

dioxins and furans and are

area sources of HAPs.

------------------------------------------------------------------------

This table is not intended to be exhaustive, but rather provides a

guide for readers regarding entities likely to be regulated by this

action. This table lists the types of entities that the Agency is now

aware could potentially be regulated by this action. Other types of

entities not listed in the table also could be regulated. To determine

whether your facility is regulated by this action, you should carefully

examine the applicability criteria in Sec. 63.1500 of the proposed

rule. If you have questions regarding the applicability of this action

to a particular entity, consult the person listed in the preceding FOR

FURTHER INFORMATION CONTACT section.

Technology Transfer Network

The proposed regulatory text also is available on the Technology

Transfer Network (TTN), one of EPA's electronic bulletin boards. The

TTN provides information and technology exchange in various areas of

air pollution control.

[[Page 6947]]

The service is free, except for the cost of a phone call. Dial (919)

541-5742 for up to a 14,400 BPS modem. The TTN also is accessible

through the Internet at ``TELNET ttnbbs.rtpnc.epa.gov.'' If more

information on the TTN is needed, call the HELP line at (919) 541-5384.

The help desk is staffed from 11 a.m. to 5 p.m.; a voice menu system is

available at other times.

Electronic Access and Filing Addresses

The official record for this rulemaking, as well as the public

version, has been established under Docket No. A-92-61 (including

comments and data submitted electronically). A public version of this

record, including printed, paper versions of electronic comments, which

does not include any information claimed as confidential business

information (CBI), is available for inspection from 8 a.m. to 5:30

p.m., Monday through Friday, excluding legal holidays. The official

rulemaking record is located at the address in ADDRESSES at the

beginning of this document.

Electronic comments can be sent directly to the EPA's Air and

Radiation Docket and Information Center at: ``A-and-R-

D[email protected].'' Electronic comments must be submitted as an

ASCII file avoiding the use of special characters and any form of

encryption. Comments and data will also be accepted on disks in

Wordperfect 5.1 file format or ASCII file format. All comments and data

in electronic form must be identified by the docket number (A-92-61).

Electronic comments may be filed online at many Federal Depository

Libraries.

Outline

The information in this preamble is organized as shown below.

I. Statutory Authority

II. Introduction

A. Background

B. NESHAP for Source Categories

C. Health Effects of Pollutants

D. Secondary Aluminum Industry

III. Summary of Proposed Standards

A. Applicability

B. Emission Limits and Requirements

C. Operating and Monitoring Requirements

IV. Selection of Proposed Standards

A. Selection of Source Category

B. Selection of Emission Sources and Pollutants

C. Selection of Proposed Standards for Existing and New Sources

1. Background

2. Selection of MACT Floor Technology

3. Consideration of Beyond-the-Floor Technologies

4. Selection of Emission Limits

D. Selection of Operating and Monitoring Requirements

1. Operating and Monitoring Requirements and Options for

Affected Sources and Emission Units

2. Operating and Monitoring Requirements and Options for

Affected Sources and Emission Units Equipped with a Fabric Filter

and Subject to PM Limits

3. Other Operating and Monitoring Requirements and Procedures

E. Selection of Performance Test Methods and Requirements

1. Rationale for Performance Test Methods, Procedures and

Surrogates

2. General Requirements

3. Performance Test Requirements and Options for Affected

Sources and Emission Units

4. Performance Test Requirements and Options for Affected

Sources and Emission Units Equipped with a Fabric Filter or Lime-

Injected Fabric Filter

F. Notification, Recordkeeping and Reporting Requirements

V. Summary of Impacts of Proposed Standards

A. Air Quality Impacts

B. Cost Impacts

C. Economic Impacts

D. Non-air Health and Environmental Impacts

E. Energy Impacts

VI. Request for Comments

VII. Administrative Requirements

A. Docket

B. Public Hearing

C. Executive Order 12866

D. Executive Order 13045

E. Enhancing the Intergovernmental Partnership Under Executive

Order 12875

F. Executive Order 13084: Consultation and Coordination with

Indian Tribal Governments

G. Unfunded Mandates Act

H. Regulatory Flexibility Act

I. Paperwork Reduction Act

J. National Technology Transfer and Advancement Act

K. Pollution Prevention Act

L. Clean Air Act

I. Statutory Authority

The statutory authority for this proposal is provided by sections

101, 112, 114, 116, and 301 of the Clean Air Act, as amended (42 U.S.C.

7401, 7412, 7414, 7416, and 7601).

II. Introduction

A. Background

The EPA estimates that about 28,600 Mg/yr (31,500 tpy) of HAPs and

other air pollutants are released from production processes in 86

major-source secondary aluminum production facilities. The HAPs in

these emissions consist of several organic compounds, including

2,3,7,8-TCDD (a compound in the dioxin/furans (D/F) group); inorganic

``acid gas'' compounds such as hydrogen chloride (HCl), hydrogen

fluroride (HF), and chlorine (Cl2); and 11 nonvolatile HAP

metals. NonHAP particulate matter (PM) and volatile organic compounds

(VOCs) are also emitted.

The proposed standard reduces emissions of HAPs and other

pollutants using a combination of emission limits and pollution

prevention/work practice standards based on MACT floor controls.

Depending on the type of affected source, plants affected by the

standards could achieve the proposed requirements by upgrading or

installing a fabric filter or a lime-injected fabric filter (i.e., a

fabric filter to which lime or other alkaline reagent is continuously

injected). Or, plants may be required to add a thermal incinerator

(also known as an afterburner), a thermal incinerator followed by a

lime-injected fabric filter, and/or apply pollution prevention

techniques to limit the type of scrap charged and the type and amount

of fluxing agents used. Raising the control performance of affected

sources with MACT-level standards would reduce emissions of HAPs by 70

percent and other pollutants by about 42 percent from the current

level, with higher reductions achieved at particular sites. Emissions

of HCl would be decreased by about 74 percent.

The nationwide total capital and annualized costs of control

equipment are estimated at $148 million and $68 million/yr,

respectively. An additional $5.1 million per year is estimated for

monitoring/implementation costs for the first 3 years following

promulgation. The economic impacts of the proposed regulation are

expected to be minimal with price increases and production decreases of

less than one percent. The regulation is not expected to result in a

significant economic impact for a substantial number of small entities.

Only one of the 33 small entities is anticipated to experience

significantly adverse economic impacts as a result of this regulation.

The proposed NESHAP was developed by EPA with input from industry

representatives and associated groups including the Aluminum

Association and STAPPA/ALAPCO (State and Territorial Air Pollution

Program Administrators Association/Association of Local Air Pollution

Control Officials). The rule development process included a cooperative

effort with the industry in identifying data needs; collecting

additional data; planning and conducting emission tests; and meeting

with these representatives to share technical information and resolve

issues.

B. NESHAP for Source Categories

Section 112 of the Act requires that EPA promulgate regulations for

the

[[Page 6948]]

control of HAP emissions from both new and existing major sources. The

regulations must reflect the maximum degree of reduction in emissions

of HAPs that is achievable taking into consideration the cost of

achieving the emission reduction, any nonair quality health and

environmental impacts, and energy requirements. This level of control

is commonly referred to as MACT.

The control of HAPs is achieved through the promulgation of

technology-based emission standards under sections 112(d) and 112(f)

and work practice standards under 112(h) for categories of sources that

emit HAPs. Emission reductions may be accomplished through the

application of measures, processes, methods, systems, or techniques

including, but not limited to: (1) Reducing the volume of, or

eliminating emissions of, such pollutants through process changes,

substitution of materials, or other modifications; (2) enclosing

systems or processes to eliminate emissions; (3) collecting, capturing,

or treating such pollutants when released from a process, stack,

storage or fugitive emissions point; (4) design, equipment, work

practice, or operational standards (including requirements for operator

training or certification) as provided in section (h); or (5) a

combination of the above. (See section 112(d)(2).)

C. Health Effects of Pollutants

The Clean Air Act was created in part to protect and enhance the

quality of the Nation's air resources so as to promote the public

health and welfare and the productive capacity of its population. (See

section 101(b)(1).) Section 112(b) of the Act contains a list of HAPs

believed to cause adverse health or environmental effects. Section

112(d) of the Act requires that emission standards be promulgated for

all categories and subcategories of major sources of these HAPs and for

many smaller ``area'' sources listed for regulation under section

112(c) in accordance with the schedules listed under section 112(c).

Major sources are defined as those that emit or have the potential to

emit at least 10 tons per year (tpy) of any single HAP or 25 tpy of any

combination of HAPs.

In the 1990 Amendments to the Clean Air Act, Congress specified

that each standard for major sources must require the maximum reduction

in emissions of HAPs that EPA determines is achievable considering

cost, health and environmental impacts, and energy impacts. In essence,

these MACT standards would ensure that all major sources of air toxic

emissions achieve the level of control already being achieved by the

better controlled and lower emitting sources in each category. This

approach provides assurance to citizens that each major source of toxic

air pollution will be required to effectively control its emissions. At

the same time, this approach provides a ``level economic playing

field,'' ensuring that facilities that employ cleaner processes and

good emissions control are not disadvantaged relative to competitors

with poorer controls.

Emission data, collected during development of this NESHAP, show

that pollutants listed in section 112(b)(1) are emitted by secondary

aluminum production processes and include organic HAPs (e.g., D/F,

benzene, styrene, xylene, acrylonitrile, methylene chloride,

naphthalene, and formaldehyde); inorganic HAPs (HCl, HF, and

Cl2), and HAP metals (antimony, arsenic, lead, manganese,

beryllium, cadmium, chromium, cobalt, mercury, nickel, and selenium).

Emissions of these pollutants would be decreased by implementation of

the proposed emission limits. Some of these pollutants are either known

or probable human carcinogens when inhaled, and can cause reversible

and irreversible toxic effects other than cancer following sufficient

exposure. These effects include respiratory and skin irritation,

effects upon the eye, various systemic effects including effects upon

the liver, kidney, heart and circulatory system, neurotoxic effects,

and in extreme cases, death. Following is a summary of the potential

health and environmental effects associated with exposures, at some

level, to emitted pollutants that would be reduced by the standard.

Almost all metals appearing on the section 112(b) list of HAPs are

emitted from affected sources in secondary aluminum plants. These

metals can cause a range of effects including irritation of the

respiratory tract; gastrointestinal effects; nervous system disorders

(including loss of coordination and mental retardation); skin

irritation; and reproductive and developmental disorders. Additionally,

these metals accumulate in the environment and several of them

accumulate in the human body, and may cause adverse health effects

after exposure has ceased. Cadmium, for example, is a cumulative

pollutant that can cause kidney effects after the cessation of

exposure. Similarly, the onset of effects from beryllium exposure may

be delayed by months to years. Many of the metal compounds also are

known (arsenic, chromium (VI)) or probable (cadmium, nickel carbonyl,

lead, and beryllium) human carcinogens.

Each HAP organic compound has a range of potential health effects

associated with exposures above toxic thresholds. Effects generally

associated with short-term inhalation exposure to these pollutants

include irritation of the eyes, skin, and respiratory tract; central

nervous system effects (e.g., drowsiness, dizziness, headaches,

depression, nausea, abnormal electrocardiograms); and reproductive and

developmental effects. Health effects associated with long-term

inhalation exposure in humans to the organic compounds which will

potentially be decreased by the proposed standard may include mild

symptoms such as nausea, headache, weakness, insomnia, gastrointestinal

effects, and burning eyes; disorders of the blood; toxicity to the

immune system; reproductive disorders in women (e.g., menstrual

irregularity or increased risk of spontaneous abortion); developmental

effects; and injury to the liver and kidneys. In addition to non-cancer

effects, some of the organic HAPs that would be controlled under this

proposed NESHAP are either known or probable human carcinogens.

Hydrogen chloride is highly corrosive to the eyes, skin, and mucous

membranes. Short-term inhalation of HCl by humans may cause coughing,

hoarseness, inflammation and ulceration of the respiratory tract, as

well as chest pain and pulmonary edema. Long-term occupational exposure

of humans to HCl has been reported to cause inflammation of the

stomach, skin, and lungs, and photosensitization.

Acute exposure to hydrogen fluoride will result in irritation,

burns, ulcerous lesions, and necrosis of the eyes, skin, and mucous

membranes. Total destruction of the eyes is possible. Other effects

include nausea, vomiting, diarrhea, pneumonitis (inflammation of the

lungs), and circulatory collapse. Ingestion of an estimated 1.5 grams

produced sudden death without gross pathological damage. Repeated

ingestion of small amounts resulted in moderately advanced hardening of

the bones. Contact of skin with anhydrous liquid produces severe burns.

Inhalation of anhydrous hydrogen fluoride or hydrogen fluoride mist or

vapors can cause severe respiratory tract irritation that may be fatal.

The irritating properties of Cl2 make this HAP a serious

acute respiratory hazard, as well as a skin, eye, and throat irritant.

Prolonged exposure to low concentrations can cause respiratory

problems, tooth corrosion, inflammation

[[Page 6949]]

of the mucous membranes, and susceptibility to tuberculosis. Prolonged

exposure at moderate concentrations can cause decreased lung capacity.

Several of the HAP whose emissions will be reduced by this rule

have been found to cause serious developmental effects in animals or

humans. For example, children are more sensitive than adults to the

neurotoxic effects of lead, suffering neurobehavioral deficits such as

loss of IQ at relatively low exposures. Chlorinated dibenzodioxins and

furans are now understood to be potent developmental toxins, disrupting

a wide variety of developmental events in embryos of numerous

vertebrate species at exposures that are not toxic to adults. Although

this rule is based on emission reduction technology rather than risk

reduction per se, EPA anticipates that reductions in emissions of

developmentally-toxic HAP will especially benefit children.

In addition to the HAPs, the proposed NESHAP also would reduce some

of the pollutants whose emissions are controlled under the National

Ambient Air Quality Standards (NAAQS) program. These pollutants include

particulate matter (PM), volatile organic compounds (VOC--precursors to

tropospheric ozone formation), and lead (also a HAP metal). The health

effects of lead, PM, and VOC are described in EPA's Criteria Documents,

which support the NAAQS. Briefly, PM emissions have been associated

with aggravation of existing respiratory and cardiovascular disease and

increased risk of premature death. At elevated levels, ozone has been

shown in human laboratory and community studies to be responsible for

the reduction of lung function, respiratory symptoms (e.g., cough,

chest pain, throat and nose irritation), increased hospital admissions

for respiratory causes, and increased lung inflammation. Animal studies

have shown increased susceptibility to respiratory infection and lung

structure changes. Exposure to ozone also has been linked to harmful

effects on agricultural crops and forests. Depending on the degree of

exposure, lead can cause subtle effects on behavior and cognition

(particularly in children), increased blood pressure, reproductive

effects, seizures, and even death.

The EPA recognizes that the degree of adverse effects to health can

range from mild to severe. The extent and degree to which the health

effects may be experienced is dependent upon: (1) The ambient

concentrations observed in the area, (e.g., as influenced by emission

rates, meteorological conditions, and terrain), (2) the frequency of

and duration of exposures, (3) characteristics of exposed individuals

(e.g., genetics, age, pre-existing health conditions, and lifestyle)

which vary significantly with the population, and (4) pollutant-

specific characteristics (e.g., toxicity, half-life in the environment,

bioaccumulation, and persistence).

D. Secondary Aluminum Industry

At least 400 facilities which include one or more secondary

aluminum affected sources currently operate in 36 States. Based on

industry responses to EPA's information collection request (ICR) and

responses to a voluntary supplemental industry/EPA survey, the 86

facilities identified as major sources operate at least 69 scrap

shredders, 5 chip dryers, 44 scrap dryers/decoating kilns/delacquering

kilns, 12 sweat furnaces, 15 dross-only furnaces, 86 secondary aluminum

processing units, and 26 rotary dross coolers.

III. Summary of Proposed Standards

A. Applicability

The proposed NESHAP applies to each new, existing or reconstructed

scrap shredder, chip dryer, scrap dryer/ delacquering kiln/decoating

kiln, group 2 furnace, sweat furnace, dross-only furnace, and rotary

dross cooler; each secondary aluminum processing unit (composed of all

existing group 1 furnace emission units and all existing in-line fluxer

emission units); and each new or reconstructed group 1 furnace and in-

line fluxer located at a secondary aluminum production plant that is a

major source of HAP. The proposed NESHAP also applies to each new,

existing or reconstructed chip dryer, scrap dryer/delacquering kiln/

decoating kiln, and sweat furnace; each secondary aluminum processing

unit and each new or reconstructed group 1 furnace and in-line fluxer

located at a secondary aluminum production plant that is an area source

of HAP. The proposed NESHAP also applies to these secondary aluminum

production affected sources if they are collocated at a primary

aluminum production facility that is a major source of HAP.

As discussed further in section IV of this document, the EPA

categorized process furnaces into two classes. A group 1 furnace

includes any furnace that processes aluminum scrap containing paint,

lubricants, coatings, or other foreign materials or within which

reactive fluxing is performed, regardless of the type of scrap charged.

Reactive fluxing means the use of any gas, liquid, or solid flux

(including chlorine gas or magnesium chloride) that results in a HAP

emission.

Group 2 (``clean charge'') furnaces process only molten aluminum,

T-bar, sow, ingot, alloying elements, noncoated runaround scrap,

uncoated aluminum chips dried at 343 deg.C (650 deg.F) or higher, and

aluminum scrap dried, decoated, or delacquered at a temperature at

482 deg.C (900 deg.F) or higher. A group 2 furnace performs no fluxing

or performs fluxing using only nonreactive, nonHAP-containing/nonHAP-

generating gases such as argon and nitrogen.

B. Emission Limits and Requirements

The proposed NESHAP for secondary aluminum production applies to

major sources. In addition, affected sources located at area sources of

HAPs, which emit D/F are regulated for emissions of D/F. The proposed

limits are summarized in Table 1.

BILLING CODE 6560-50-P

[[Page 6950]]

[GRAPHIC] [TIFF OMITTED] TP11FE99.000

[[Page 6951]]

[GRAPHIC] [TIFF OMITTED] TP11FE99.001

[[Page 6952]]

[GRAPHIC] [TIFF OMITTED] TP11FE99.002

[[Page 6953]]

[GRAPHIC] [TIFF OMITTED] TP11FE99.003

BILLING CODE 6560-50-C

PM emission limits would apply to new, reconstructed and existing

scrap shredders, scrap dryer/delacquering/decoating kilns, dross-only

furnaces, rotary dross coolers; secondary aluminum processing units;

and new and reconstructed in-line fluxers, and group 1 furnaces at

secondary aluminum production facilities that are major sources.

Controlling PM emissions would also control emissions of HAP metals. A

surrogate approach to emission limits is used to allow easier and less

expensive measurement and monitoring requirements.

The proposed rule limits total hydrocarbon emissions (THC) from new

and existing chip dryers and from new and existing scrap dryer/

delacquering/decoating kilns at secondary aluminum production

facilities that are major sources. THC represents emissions of HAP

organics. HCl emission limits would apply to new, reconstructed and

existing scrap dryer/delacquering/decoating kilns; new and

reconstructed in-line fluxers and Group 1 furnaces; and secondary

aluminum processing units at secondary aluminum production facilities

that are major sources. HCl serves as a surrogate measure of HAP

inorganics including hydrogen fluoride (HF) and chlorine

(Cl2) emissions. The proposed rule limits emissions of D/F

from new, reconstructed and existing chip dryers, scrap dryer/

delacquering/decoating kilns and sweat furnaces; new and reconstructed

group 1 furnaces; and secondary aluminum processing units at secondary

aluminum production facilities that are major or area sources. No

surrogate is used for D/F emissions. A detailed explanation of the

proposed limits and the rationale for their selection is given in

section IV.C. of this document.

C. Operating and Monitoring Requirements

The proposed NESHAP includes operating and monitoring requirements

for each affected source and emission unit within a secondary aluminum

processing unit to ensure continuous compliance with the emissions

standards. The proposed standard would incorporate all requirements of

the NESHAP general provisions (40 CFR part 63, subpart A). The proposed

operating and monitoring requirements are summarized in Table 2. A

detailed explanation of the monitoring requirements and the rationale

for their selection is given in section IV.D. of this document. \1/

2\Federal Register

Table 2.--Summary of Proposed Operating and Monitoring Requirements for Affected Sources and Emission Units

----------------------------------------------------------------------------------------------------------------

Monitor type/

Affected source/emission unit operation/process Operating requirements Monitoring requirements

----------------------------------------------------------------------------------------------------------------

All affected sources and emission Labeling............. Identification, emission Check monthly to confirm

units. limits and means of that labels are intact

compliance posted on all and legible.

affected sources and

emission units.

All affected sources and emission Emission capture and Design and install in Annual inspection of all

units with add-on control device. collection system. accordance with emission capture,

Industrial Ventilation: A collection, and

Handbook of Recommended transport systems to

Practice; operate in ensure that systems

accordance with O, M & M continue to operate in

plan.b accordance with ACGIH

standards.

All affected sources and emission Charge/feed weight... Operate a device or use an Record the weight of each

units subject to production based equivalent procedure to charge; weight

[lb/ton of feed] emission limits record the weight of each measurement device or

a. charge; operate in other procedure accuracy

accordance with O, M, & M of 1

plan. percent; calibration

every 3 months.

Scrap shredder with fabric filter. Bag leak detector.... Initiate corrective action Install and operate in

within 1 hour of alarm accordance with ``Fabric

and complete in Filter Bag Leak

accordance with O, M, & M Detection Guidance'' and

plan; b operate such that record voltage output

alarm does not sound more from bag leak detector.

than 5% of operating time

in 6-month period.

[[Page 6954]]

or

COM.................. Initiate corrective action Design and install in

within 1-hour of a 6- accordance with PS-1;

minute average opacity collect data in

reading of 5% or more and accordance with subpart

complete in accordance A of 40 CFR 63;

with O, M, & M plan; b. calculate and record 6-

minute block averages.

or

VE................... Initiate corrective action Conduct and record

within 1 hour of any results of 30 minute

observed VE and complete daily test in accordance

in accordance with the O, with Method 9.

M, & M plan.b

Chip Dryer with afterburner....... Afterburner operating Maintain average Continuous measurement

temperature. temperature, averaged device to meet EPA

over each 3-hour period, specifications;

at or above the average calculate and record

operating temperature average temperature for

during the performance each 15-minute block;

test. determine 3-hour block

averages; calibrate

every 3 months.

Afterburner operation Operate in accordance with Conduct annual inspection

O, M, and M plan.b of afterburner internal

parts to maintain good

working order.

Feed material........ Operate using only Record identity of charge

unpainted aluminum chips. daily; certify charge

materials every 6

months.

Scrap dryer/delacquering/decoating Afterburner operating Maintain average Continuous measurement

kiln with afterburner and lime temperature. temperature, averaged device to meet EPA

injected fabric filter. over each 3-hour period, specifications; record

at or above the average temperatures in 15-

operating temperature minute block averages;

during the performance calculate 3-hour block

test. averages; calibration

every 3 months.

Afterburner operation Operate in accordance with Annual inspection of

O, M, & M plan.b afterburner internal

parts; complete repairs

in 10 days.

Bag leak detector.... Initiate corrective action Install and operate in

within 1 hour of alarm accordance with ``Fabric

and complete in Filter Bag Leak

accordance with the O, M, Detection Guidance'' and

& M plan; b operate such record voltage output

that alarm does not sound from bag leak detector.

more than 5% of operating

time in 6-month period.

or

COM.................. Initiate corrective action Design and install in

within 1 hour of a 6- accordance with PS-1;

minute average opacity collect data in

reading of 5% or more and accordance with subpart

complete in accordance A of 40 CFR 63;

with the O, M, & M plan.b calculate and record 6-

minute block averages.

Lime injection rate Maintain free-flowing lime Inspect each feed hopper

and schedule. in the feed hopper or or silo every 8 hours to

silo at all times. verify that lime is free-

flowing; record results

of each inspection. If

blockage occurs, inspect

every 4 hours for 3

days; return to 8-hour

inspections if

corrective action

results in no further

blockage during 3-day

period.

Maintain average lime Weight measurement device

injection rate (lb/hr) at accuracy of 1 percent;

during the successful calibration every 3

compliance test and months; record weight of

adhere to the same lime lime injected for each

injection schedule used 15-minute block period

during the performance and determine 3-hour

test for each 3-hour block averages or;

period or

Maintain average lime Weight measurement device

injection rate (lb/ton of accuracy of 1 percent;

rate used during the calibrate every 3

performance test and months; record weight of

adhere to the same lime lime added or injected

injection schedule used for each 15-minute block

during the performance period and determine

test for each operating lime injection rate (lb/

cycle or time period used ton of feed) for each

in performance test or operating cycle or time

period used in

performance test or;

Maintain feeder setting at Record feeder setting

level established at daily.

performance test.

[[Page 6955]]

Fabric filter inlet Maintain average fabric Continuous measurement

temperature. filter inlet temperature device to meet EPA

at or below the average specifications; record

temperature during the temperatures in 15

successful compliance minute block averages;

test +14 deg.C (25 calculate 3 hour block

deg.F) for each three averages; calibration

hour period. every three months.

Sweat furnace with afterburner.... Afterburner operating Maintain average Continuous measurement

temperature. temperature, averaged device to meet EPA

over each 3-hour period, specifications; record

at or above the average temperatures in 15-

operating temperature minute block averages;

during the performance calculate 3-hour block

test. averages; calibration

every 3 months.

Afterburner operation Operate in accordance with Annual inspection of

O, M, & M plan.b afterburner internal

parts; complete repairs

in 10 days.

Dross-only furnace with fabric Bag leak detector.... Initiate corrective action Installation and

filter. within 1 hour of alarm operation requirements

and complete in in accordance with

accordance with the O, M, ``Fabric Filter Bag Leak

& M plan; b operate such Detection Guidance'' and

that alarm does not sound record voltage output

more than 5% of operating from bag leak detector.

time in 6-month period.

or

COM.................. Initiate corrective action Design and install in

within 1 hour of a 6- accordance with PS-1;

minute average opacity collect data in

reading of 5% or more and accordance with subpart

complete in accordance A of 40 CFR 63;

with the O, M, & M plan.b calculate and record 6-

minute block averages.

Feed/charge material. Operate using only dross Record identity of each

as the feed material. charge; certify charge

materials every 6

months.

Rotary dross cooler with fabric Bag leak detector.... Initiate corrective action Install and operate in

filter. within 1 hour of alarm accordance with ``Fabric

and complete in Filter Bag Leak

accordance with the O, M, Detection Guidance'' and

& M plan; b operate such record voltage output

that alarm does not sound from bag leak detector.

more than 5% of operating

time in 6-month period.

or

COM.................. Initiate corrective action Design and install in

within 1 hour of a 6- accordance with PS-1;

minute average opacity collect data in

reading of 5% or more and accordance with subpart

complete in accordance A of 40 CFR 63;

with the O, M, & M plan.b calculate and record 6-

minute block averages.

In-line fluxer with lime injected Bag leak detector.... Initiate corrective action Install and operate in

fabric filter (including those within 1 hour of alarm accordance with ``Fabric

that are part of a secondary and complete in Filter Bag Leak

aluminum processing unit). accordance with the O, M, Detection Guidance'' and

& M plan; b operate such record voltage output

that alarm does not sound from bag leak detector.

more than 5% of operating

time in 6-month period.

or

COM.................. Initiate corrective action Design and install in

within 1 hour of a 6- accordance with PS-1;

minute average opacity collect data in

reading of 5% or more and accordance with subpart

complete in accordance A of 40 CFR 63;

with the O, M, & M plan.b calculate and record 6-

minute block averages.

Reactive flux Maintain the reactive flux Weight measurement device

injection rate and injection rate at or accuracy of 1 percent;

injection rate used calibration every 3

during the performance months; record weight

test and adhere to the and type of reactive

same flux injection flux added or injected

schedule used during the for each 15-minute block

test. period.

Lime injection rate Maintain free-flowing lime Inspect each feed hopper

and schedule. in the feed hopper or or silo every 8 hours to

silo at all times. verify that lime is free-

flowing; record results

of each inspection. If

blockage occurs, inspect

every 4 hours for 3

days; return to 8-hour

inspections if

corrective action

results in no further

blockage during 3-day

period.

[[Page 6956]]

In-line fluxer with lime Maintain average lime Weight measurement device

injected fabric filter (including injection rate (lb/hr) at accuracy of 1 percent;

secondary aluminum processing during the performance calibrate every 3

unit) cont'd test and adhere to the months; record weight of

same lime injection lime injected for each

schedule used during the 15-minute block period

test for each 3-hour and determine 3 hour

period or. block averages or;

Maintain average lime Weight measurement device

injection rate (1b/ton of accuracy of 1 percent;

rate used during the calibrate every 3

performance test and months; record weight of

adhere to the same lime lime injected for each

injection schedule used 15-minute block period

during the test for each and determine lime

operating cycle or time injection rate (lb/ton

period used in of feed) for each

performance test or. operating cycle or time

period used in

performance test or;

Maintain feeder setting at Record feeder setting

level established at daily.

performance test.

Fabric filter inlet Maintain average fabric Continuous measurement

temperature. filter inlet temperature device to meet EPA

at or below the average specifications; record

temperature during the temperatures in 15-

performance test +14 minute block averages;

deg.C (25 deg.F) for each calculate 3-hour block

3-hour period. averages; calibrate

every 3 months.

Clean (group 2) furnace........... Charge materials..... Use only clean charge..... Record identity of all

charge materials;

certify every 6 months.

Flux materials....... Use no reactive flux...... Record identity of all

flux materials; certify

every 6 months that no

reactive flux was used.

Group 1 furnace with lime injected Bag leak detector.... Initiate corrective action Install and operate in

fabric filter (including those within 1 hour of alarm accordance with ``Fabric

that are part of a secondary and complete in Filter Bag Leak

aluminum processing unit). accordance with the O, M, Detection Guidance'' and

& M plan; b operate such record voltage output

that alarm does not sound from bag leak detector.

more than 5% of operating

time in 6-month period.

or

COM.................. Initiate corrective action Design and install in

within 1 hour of a 6- accordance with PS-1;

minute average opacity collect data in

reading of 5% or more and accordance with subpart

complete in accordance A of 40 CFR 63;

with the O, M, & M plan.b calculate and record 6-

minute block averages.

Lime injection rate Maintain free-flowing lime Inspect each feed hopper

and schedule. in the feed hopper or or silo every 8 hours to

silo at all times. verify that lime is free-

flowing; record results

of each inspection. If

blockage occurs, inspect

every 4 hours for 3

days; return to 8-hour

inspections if

corrective action

results in no further

blockage during 3-day

period.

Maintain average lime Weight measurement device

injection rate (lb/hr) at accuracy of 1 percent;

during the performance calibrate every 3

test and adhere to the months; record weight of

same lime injection lime injected for each

schedule used during the 15-minute block period

test for each 3-hour and determine 3-hour

period or; block averages.

Maintain average lime Weight measurement device

injection rate (lb/ton of accuracy of 1 percent;

rate used during the calibrate every 3

performance test and months; record weight of

adhere to the same lime lime injected for each

injection schedule used 15-minute block period

during the test for each and determine lime

operating cycle or time injection rate (lb/ton

period used in of feed) for each

performance test or; operating cycle or time

period used in

performance test or;

Maintain feeder setting at Record feeder setting

level established at daily.

performance test.

Reactive flux Maintain the reactive flux Weight measurement device

injection rate and injection rate at or accuracy of 1 percent;

injection rate used calibrate every 3

during the performance months; record weight

test. and type of reactive

flux added or injected

for each 15-minute block

period.

[[Page 6957]]

Fabric filter inlet Maintain average fabric Continuous measurement

temperature. filter inlet temperature device to meet EPA

at or below the average specifications; record

temperature during the temperature in 15-minute

performance test +14 block averages;

deg.C (25 deg.F) for calculate 3-hour block

each 3 hour period. averages; calibrate

every 3 months.

Maintain molten Operate side-well furnaces Maintain aluminum level

aluminum level. such that the level of operating log; certify

molten metal is above the every 6 months.

top of the passage

between side well and

hearth during reactive

flux injection.

Fluxing in sidewell Add reactive flux only to Maintain flux addition

furnace hearth. the sidewell of the operating log; certify

furnace unless the hearth every 6 months.

is also controlled.

Group 1 furnace without add-on Reactive flux Maintain the reactive flux Weight measurement device

controls (including those that injection rate and injection rate at or accuracy of 1 percent;

processing unit). injection rate used calibrate every 3

during the performance months; record weight

test and adhere to same and type of reactive

flux injection schedule flux added or injected

used in performance test. for each 15-minute block

period.

Feed material (melter/ .......................... Record identity of each

holder). charge; certify charge

materials every 6

months.

Site-specific Operate furnace within the Demonstration of site-

monitoring plan range of charge specific monitoring plan

(approved by materials, contaminant to provide data and show

permitting agency). levels, and parameter correlation of emissions

values established in the across the range of

site-specific monitoring charge and flux

plan.c materials and furnace

operating parameters.

----------------------------------------------------------------------------------------------------------------

a Chip dryers, scrap dryers/delacquering kilns/decoating kilns, dross-only furnaces, in-line fluxers (including

those that are part of a secondary aluminum processing unit) and group 1 furnaces including melter holders

(including those that are part of a secondary aluminum processing unit).

b O, M, & M plan--Operation, maintenance, and monitoring plan.

c Site-specific monitoring plan--Owner/operators of group 1 furnaces without control devices must develop a site-

specific monitoring plan that identifies process or feed parameter-based operating requirements. This plan

would be part of the O, M, & M plan. This plan and the testing to demonstrate adequacy of the monitoring plan

and correlation of parameters over the range of charge materials and fluxing practices must be developed in

coordination with and be approved by the permitting authority.

IV. Selection of Proposed Standards

A. Selection of Source Category

Section 112(c) of the Act directs the EPA to list each category of

major and area sources, as appropriate, emitting one or more of the

HAPs listed in section 112(b) of the Act. The EPA published an initial

list of source categories on July 16, 1992 (57 FR 31576), and may amend

the list at any time. ``Secondary Aluminum Production'' is one of the

174 categories of sources included on the revised list of source

categories (63 FR 7155, February 12, 1998). This list includes major

and area sources of HAPs for which the EPA intends to issue regulations

between November 1992 and November 2000. The category as defined in the

EPA report, ``Documentation for Developing the Initial Source Category

List'' (docket item II-A-6) for the listing includes any facility

engaged in the cleaning, melting, refining, alloying, and pouring of

aluminum recovered from scrap, foundry returns, and dross.

The listing of the secondary aluminum production major source

category was based on the Administrator's determination that some

secondary aluminum production facilities would be major sources of

HAPs. These facilities are known to emit HAPs, including PM metal HAP

(including antimony, arsenic, beryllium, cadmium, chromium, cobalt,

lead, manganese, mercury, and nickel), gaseous organic HAPs (including

dioxins, furans, polycyclic organic matter, benzene and formaldehyde)

and gaseous inorganic HAPs (including hydrogen chloride, hydrogen

fluoride, and chlorine).

A major source must have the potential to emit 9.1 Mg/yr (10 tpy)

or more of a single HAP or 23 Mg/yr (25 tpy) or more of a combination

of HAPs. The EPA has estimated that there are approximately 86 major

source facilities that practice one or more secondary aluminum

production processes.

Section 112(c)(6) of the Act states that by November 15, 2000, EPA

must list and promulgate section 112(d)(2) or (d)(4) standards (i.e.,

standards reflecting MACT) for categories (and subcategories) of

sources emitting seven specific pollutants, including 2,3,7,8

tetrachlorodibenzofurans and 2,3,7,8 tetrachlorodibenzo-p-dioxin which

are emitted by secondary aluminum production affected sources. The EPA

must assure that source categories accounting for not less than 90

percent of the aggregated emissions of the enumerated pollutant are

subject to MACT standards. Congress (docket item II-I-13, p. 155 to 156

(cement) singled out the HAPs enumerated in section 112(c)(6) as being

of ``specific concern'' not just because of their toxicity but because

of their propensity to cause substantial harm to human health and the

environment via indirect exposure pathways (i.e., from the air through

other media, such as water, soil, food uptake, etc.). Furthermore,

these pollutants have exhibited special potential to bioaccumulate,

causing pervasive environmental harm in biota (and, ultimately, human

health risks).

The EPA estimates that secondary aluminum production facilities

emit in aggregate approximately 0.4 lb per year of D/F (from June 20,

1997; 62 FR 33635), or 3.5 percent (from April 10, 1998; 63 FR 17849),

of the total national anthropogenic emissions of D/F per year

[[Page 6958]]

(docket item II-J-2, docket item II-J-4). To assure that this pollutant

is subject to MACT, EPA has added the secondary aluminum production

area source category to the list of source categories and subcategories

listed pursuant to section 112(c)(6). (See 63 FR 17838, 17849; April

10, 1998.) The EPA has done so because area and major source secondary

aluminum D/F emitting processes emit this HAP at about equal rates per

ton of feed, because the D/F emitted by area sources are equally toxic

per amount of emissions as that emitted by major sources (i.e., the

distribution of dioxin and furan isomers is the same for both area and

major sources), and because this is a particularly toxic class of HAP.

In addition, EPA's strategy for assuring 90 percent of these pollutants

are addressed includes control of these pollutants from secondary

aluminum production facility area sources through the MACT process.

(See 62 FR 33635, 33636; June 20, 1997.)

The EPA notes, however, as it did in the April 10th document, that

although the section 112(c)(6) listing process makes sources subject to

standards under subsection (d)(2) or (d)(4), the language of section

112(c)(6) does not specify either a particular degree of emissions

control or a reduction in emissions of these specific pollutants to be

achieved by such regulations. Rather, the specific control requirements

will result from determining the appropriate level of control under

MACT (section 112(d)(2), or section 112(d)(4)), and this interpretation

will be made during the section 112(d) rulemakings affecting the

particular source category, not as part of the section 112(c)(6)

listing process. (See 63 FR 17841; April 10, 1998.)

As noted above, EPA is interpreting section 112(c)(6) to require

the EPA to establish standards under section 112(d)(2) or 112(d)(4) for

all sources listed pursuant to section 112(c)(6), whether such sources

are major or area sources. This interpretation reflects the express

language of section 112(c)(6) that sources * * * of each such pollutant

are subject to standards under subsection (d)(2) or (d)(4) and is in

accord with the function of section 112(c)(6):

* * * to assure that sources emitting significant amounts of the

most dangerous HAPs are subject to the rigorous MACT standard-setting

process.

(See S. Rep. No. 228, 101st Cong. 1st Sess., pp. 155, 166.)

In addition, the EPA is interpreting section 112(c)(6) to require

that, for sources listed under section 112(c)(6), MACT (or section

112(d)(4)) controls apply only to the section 112(c)(6) HAPs emitted by

the source. Thus, in this proposed rule, secondary aluminum production

area sources would be subject only to the D/F emission limitations of

the MACT standards. (Since the language of section 112(c)(6) is

ambiguous as to whether the entire source must comply with MACT, or

just for the HAPs enumerated in section 112(c)(6), (see 61 FR 17365, n.

12), either interpretation is legally permissible.) Applying the

provision to the entire source could result in applying MACT to all

HAPs emitted by area sources under circumstances where control would

not otherwise be warranted. The EPA specifically requests comments and

data regarding the decision to include area sources of D/F in this

proposed rule. The Agency seeks information and data regarding the

level of emissions from area sources, the degree to which controls are

in place, and the burden that would be imposed on affected sources.

B. Selection of Emission Sources and Pollutants

The secondary aluminum production source category consists of the

following operations:

(1) Preprocessing of scrap aluminum, including size reduction and

removal of oils, coatings, and other contaminants;

(2) Furnace operations including melting, in-furnace refining,

fluxing, and tapping;

(3) Additional refining, by means of in-line fluxing; and

(4) Cooling of dross.

The following sections include descriptions of the affected sources

in the secondary aluminum production source category, the origin of HAP

emissions from these affected sources, and factors affecting the

emissions. The affected sources for which MACT standards are being

proposed include new, reconstructed and existing scrap shredders, chip

dryers, scrap dryers/delacquering/decoating kilns, group 2 furnaces,

sweat furnaces and dross coolers; secondary aluminum processing units

(composed of all existing group 1 furnace emission units and all

existing in-line fluxer emission units); and new and reconstructed

group 1 furnaces and in-line fluxers. Each of these affected sources

emits one or more of the HAPs listed in section 112 of the Act.

Scrap aluminum is often preprocessed prior to melting.

Preprocessing steps may include shredding to reduce the size of

aluminum scrap; drying of oily scrap such as machine turnings and

borings; and/or heating in a scrap dryer, delacquering kiln or

decoating kiln to remove coatings or other contaminants that may be

present on the scrap. Heating of high iron content scrap in a sweat

furnace to reclaim the aluminum content is also a preprocessing

operation.

Crushing, shredding, and grinding operations are used to reduce the

size of scrap aluminum. Emissions of PM and HAP metals are generated as

dust from coatings and other contaminants contained in the scrap

aluminum. A typical shredder with a capacity of 90,900 Mg/yr (100,000

tpy), is estimated to produce 190 Mg/yr (212 tpy) of PM, before

controls (See docket item II-B-16, impacts memo). PM emitted from

shredders contains HAP metals.

A chip dryer is used to evaporate oil and/or moisture from uncoated

aluminum chips and borings. Chip dryers typically operate at

temperatures ranging between 150 deg.C to 400 deg.C (300 deg.F to

750 deg.F). An uncontrolled chip dryer with a typical capacity of

36,400 Mg/yr (40,000 tons/yr), is estimated to emit 2.4 g TEQ/yr (.0053

lb/yr) of D/F, and 385 Mg/yr (424 tpy) of THC (of which some fraction

is organic HAP) (See docket item II-B-16, impacts memo).

Painted and/or coated materials are processed in a scrap dryer/

delacquering kiln/decoating kiln to remove coatings and other

contaminants that may be present in the scrap prior to melting.

Coatings, oils, grease, and lubricants represent up to 20 percent of

the total weight of these materials. Organic HAPs, D/F, and inorganic

HAPs including particulate metal HAP are emitted during the drying/

delacquering/decoating process.

Used beverage containers (UBC) comprise a major portion of the

recycled aluminum scrap used as feedstock by the industry. In scrap

drying/delacquering/decoating operations, UBC and other post-consumer,

coated products (e.g., aluminum siding) are heated to an exit

temperature of up to 540 deg.C (1,000 deg.F) to volatilize and remove

various organic contaminants such as paints, oils, lacquers, rubber,

and plastic laminates prior to melting. An uncontrolled scrap dryer/

delacquering kiln/decoating kiln with a typical capacity of 45,500 Mg/

yr (50,000 tpy) is estimated to emit 43.3 Mg/yr (47.7 tpy) PM (of which

some fraction is particulate metal HAP), 76.0 Mg/yr (83.6 tpy) HCl, 68

Mg/yr (75 tpy) THC (of which some fraction is organic HAP), and 3.5 g

TEQ/yr (0.0077 lb TEQ/yr) of D/F (See docket item II-B-16, impacts

memo).

A sweat furnace is typically used to reclaim (or ``sweat'') the

aluminum from scrap with high levels of iron. These furnaces operate in

batch mode at a

[[Page 6959]]

temperature that is high enough to melt the aluminum but not high

enough to melt the iron. The aluminum melts and flows out of the

furnace while the iron remains in the furnace in solid form. The molten

aluminum can be cast into sows, ingots, or T-bars that are used as

feedstock for aluminum melting and refining furnaces. Alternately,

molten aluminum can be fed directly to a melting or refining furnace.

An uncontrolled sweat furnace, with a typical capacity of 4,500 Mg/yr

(5,000 tpy) is estimated to emit 0.071 g TEQ/yr (0.00016 lb TEQ/yr) of

D/F (See docket item II-B-16, impacts memo).

Process (i. e. melting, holding or refining) furnaces are

refractory-lined metal vessels heated by an oil or gas burner to

achieve a metal temperature of about 760 deg.C (1,400 deg.F). The

melting process begins with the charging of scrap into the furnace. A

gaseous (typically, chlorine) or salt flux may be added to remove

impurities and reduce aluminum oxidation. Once molten, the chemistry of

the bath is adjusted by adding selected scrap or alloying agents, such

as silicon. Salt and other fluxes contain chloride and fluoride

compounds that may be released when introduced to the bath. HCl may

also be released when chlorine-containing contaminants (such as

polyvinyl chloride coatings) present in some types of scrap are

introduced to the bath. Argon and nitrogen fluxes are not reactive and

do not produce HAPs. In a sidewell melting furnace, fluxing is

performed in the sidewell and fluxing emissions from the sidewell are

controlled. In this type of furnace, fluxing is not typically done in

the hearth and hearth emissions (which include products of combustion

from the oil and gas fired furnaces) are typically uncontrolled.

Process furnaces may process contaminated scrap which can result in

HAP emissions. In addition, fluxing agents may contain HAPs, some

fraction of which is emitted from the furnace. Process furnaces are

large sources of HAP emissions in the secondary aluminum industry. An

uncontrolled melting furnace with a typical capacity of 18,100 Mg/year

(20,000 tpy) which processes contaminated scrap and uses reactive

fluxes is estimated to emit 177 Mg/yr (195 tpy) of PM (of which

approximately 0.80 Mg/yr [0.88 tpy] is particulate metal HAP), 29.7 Mg/

yr (32.6 tpy) of HCl, and 8 g TEQ/yr (0.018 lb TEQ/yr) D/F (See docket

item II-B-16, impacts memo).

As described in section IV.C.1 of this document, process furnaces

have been divided into group 1 (unrestricted scrap content,

unrestricted fluxing) and group 2 (clean charge, no reactive flux).

Existing group 1 furnaces are emission units within the secondary

aluminum processing unit affected source.

Dross-only furnaces are furnaces dedicated to reclamation of

aluminum from drosses formed during the melting/holding/alloying

operations carried out in other furnaces. Exposure to the atmosphere

causes the molten aluminum to oxidize, and the flotation of the

impurities to the surface along with any salt flux creates ``dross''.

Prior to tapping, the dross is periodically skimmed from the surface of

the aluminum bath, and cooled. Dross-only furnaces are typically rotary

barrel furnaces (also known as salt furnaces). A dross only furnace

without controls with a typical capacity of 18,200 Mg/yr (20,000 tpy)

is estimated to emit 113 Mg/yr (125 tpy) of PM (of which some fraction

is particulate metal HAP (See docket item II-B-16, impacts memo).

Rotary dross coolers are devices used to cool dross in a rotating,

water-cooled drum. A rotary dross cooler without controls with a

typical capacity of 9,090 Mg/yr (10,000 tpy) is expected to emit 15.4

Mg/yr (17.0 tpy) of PM (of which some fraction is particulate metal

HAP) (See docket item II-B-16, impacts memo, docket item II-B-15,

Peters Risk Memo 3/27/97).

In-line fluxers are devices used for aluminum refining, including

degassing, outside the furnace. The process involves the injection of

chlorine, argon, nitrogen or other gases to achieve the desired metal

purity. Argon and nitrogen are not reactive and do not produce HAPs.

In-line fluxers are found primarily at facilities that manufacture very

high quality aluminum or in facilities with no other means of

degassing. An in-line fluxer operating without emission controls, of

typical capacity of 45,500 Mg/yr (50,000 tpy) is estimated to emit 60.8

Mg/yr (66.8 tpy) of HCl and 1.9 Mg/yr (2.1 tpy) of PM (see docket item

II-B-16, impacts memo). Existing in-line fluxers are emission units

within the secondary aluminum processing unit affected source.

Given that these processes release significant quantities of HAPs

and the availability of emission control systems, the EPA selected to

develop and propose NESHAP for the following emission sources: New,

reconstructed and existing scrap shredders, chip dryers, scrap dryer/

delacquering/decoating kilns, sweat furnaces, dross-only furnaces,

rotary dross coolers, and group 2 (clean charge, no reactive flux)

furnaces; new and reconstructed group 1 furnaces and in-line fluxers;

and secondary aluminum processing units (composed of existing group 1

furnaces and in-line fluxers).

The proposed standards would limit emissions of metal HAPs, organic

HAPs (including D/F), and HCl from secondary aluminum production

facilities. (Pollutant health effects were discussed in section II.C.

of this document). As described above, these HAPs are emitted in

significant quantities from secondary aluminum production sources.

C. Selection of Proposed Standards for Existing and New Sources

1. Background

After the EPA has identified the specific source categories or

subcategories of major sources to regulate under section 112, MACT

standards must be set for each category or subcategory. Section 112

establishes a minimum baseline or ``floor'' for standards. For new

sources, the standards for a source category or subcategory cannot be

less stringent than the emission control that is achieved in practice

by the best-controlled similar source. (See section 112(d)(3).) The

standards for existing sources can be less stringent than standards for

new sources, but they cannot be less stringent than the average

emission limitation achieved by the best-performing 12 percent of

existing sources for categories and subcategories with 30 or more

sources, or the average or median of the best-performing five sources

for categories or subcategories with fewer than 30 sources.

After the floor has been determined for a new or existing source in

a source category or subcategory, the Administrator must set MACT

standards that are no less stringent than the floor. Such standards

must then be met by all sources within the category or subcategory. In

establishing the standards, the EPA may distinguish among classes,

types, and sizes of sources within a category or subcategory. (See

section 112(d)(1).)

The next step in establishing MACT standards is to investigate

regulatory alternatives. With MACT standards, only alternatives at

least as stringent as the floor may be selected. Information about the

industry is analyzed to develop model plants for projecting national

impacts, including HAP emission reduction levels and cost, energy, and

secondary impacts. Regulatory alternatives (which may be different

levels of emissions control, equal to or more stringent than the floor

levels) are then evaluated to select the regulatory alternative that

best reflects the appropriate MACT level. The

[[Page 6960]]

selected alternative may be more stringent than the MACT floor, but the

control level selected must be technologically achievable. The

regulatory alternatives and emission limits selected for new and

existing sources may be different because of different MACT floors.

The Agency may consider going beyond the floor to require more

stringent controls. Here, the EPA considers the achievable emission

reductions of HAPs (and possibly other pollutants that are co-

controlled) and the cost impacts.

Subcategorization within a source category may be considered when

there is enough evidence to demonstrate clearly that there are

significant differences among the subcategories. The criteria to

consider include process operations (including differences between

batch and continuous operations), emission characteristics, control

device applicability, safety, and opportunities for pollution

prevention.

The EPA examined the processes, the process operations, and other

factors to determine if separate classes of units, operations, or other

criteria have an effect on air emissions from emission sources, or the

controllability of those emissions. Based on differences in emissions,

the type of materials processed and the fluxing practices employed, the

EPA has distinguished two specific classes of melting, holding, and

refining furnaces. Because HAP emission potential is strongly

influenced by the contaminants present in the materials that are melted

and the type and amount of flux added, these furnaces would be subject

to separate standards under the proposed rule.

The classes of process furnaces which are characterized by the

types of scrap charged to the furnace and the operations carried out in

the furnace are: (1) Group 1 (all process furnaces except group 2)

furnaces and (2) group 2 (``clean charge/no reactive flux'') furnaces.

Dross-only furnaces and sweat furnaces are distinctly different

from the other types because they each specialize in recovering

aluminum from a particular type of raw material. As the name implies,

``dross-only'' furnaces charge only dross collected from other furnace

operations. Sweat furnaces recover aluminum from materials with a high

iron (or other ferrous material) content. Both of these furnaces are

unique in their method of operation and are treated as separate sources

in development of the proposed NESHAP.

2. Selection of MACT Floor Technology

In establishing these proposed emission standards, the technology

representative of the MACT floor level of control was determined for

each affected source. Add-on control technologies were considered as

well as work practices and pollution prevention techniques. Data

related to operating procedures and emissions for secondary aluminum

plants were obtained through a combination of site visits, an ICR, an

EPA/industry voluntary follow-up questionnaire, and emissions tests.

Emission tests were conducted at 12 facilities to measure

uncontrolled and controlled emissions from selected production

processes and to evaluate the effectiveness of the technology

representative of the MACT floor level of control. Sites for these

tests were selected jointly by the EPA and industry as operating

technology representative of the MACT floor level of control. Funding

for tests was provided by the EPA, The Aluminum Association, and

individual facilities. The EPA also met frequently with industry

representatives to discuss the test program and available data, and to

identify and resolve issues. In addition to the data from the emission

testing program, the Agency also used emissions data from the ICR

database (docket item II-D-105, ICR database). Data from all these

sources were considered in the selection of emission limits for

individual emission points at secondary aluminum plants. Additional

details on the emission test data can be found in the docket. (See

Docket Item II-B-17. Memorandum. M. Wright, Research Triangle

Institute, to J. Santiago, EPA:MICG. Summary of Emissions Data. 1998.)

One important aspect of the more effective control technologies is

the system that captures and collects the HAPs generated by each of the

processes. Well-designed hoods and their proper placement, adequate air

flows or ventilation rates, and adequately sized ductwork and fans, in

well-maintained systems are representative of the MACT floor technology

control systems. These well-designed capture and collection systems can

be achieved by following the design standards in the American

Conference of Governmental Industrial Hygienists (ACGIH) ``Industrial

Ventilation: A Manual of Recommended Practice.'' The standards

described in Chapters 3 and 5 of this manual are incorporated by

reference in the rule as a requirement applicable to affected sources

equipped with add-on control devices.

Scrap shredders. Based on information provided in the ICR

responses, the EPA identified 69 shredding and crushing operations at

51 facilities. Emissions test measurements show that shredders and

crushers are sources of PM (containing particulate metal HAP). Fabric

filters are used to control emissions at 49 of the 69 shredders and

crushers in the industry. The best performing 12 percent of the

existing 69 scrap shredders and crushers are equipped with a fabric

filter for controlling PM and HAP metals. Therefore, the floor level of

control for existing sources is determined by the average/median of the

best performing 8 sources within the category. This median level of

control is represented by a well designed and operated pulse-jet fabric

filter using fiberglass bags with an air to cloth ratio of about 6.0.

This same level of control is also the MACT floor for new sources

since it is also the level of control achieved by the best controlled

source.

Chip dryers. The EPA identified five chip dryers based on

information provided in the ICR responses. Emissions test measurements

show that these sources emit THC (containing organic HAP) and D/F. Four

of these five dryers are equipped with an afterburner. The MACT floor,

for categories of less than 30 sources is determined by the median of

the five best controlled sources in the category. The best performing 4

of the existing 5 chip dryers are equipped with an afterburner for

organics (i.e., THC and D/F) control. Therefore, the floor level of

control for existing sources is determined by the median of the best

performing 5 sources within the category. This median level of control

is represented by a well designed and operated afterburner with a

minimum of 1-second residence time and operated at a temperature of

1,200 deg.F.

The same level of control which represents the existing source MACT

is also the MACT floor for new sources since it is also the level of

control achieved by the best controlled source.

Scrap dryers/delacquering kilns/decoating kilns. Based on

information provided in the ICR responses, the EPA identified 46 scrap

dryers, delacquering kilns, and decoating kilns. Emissions test

measurements show that these sources emit PM (containing particulate

metal HAP), HCl, THC (containing organic HAP) and D/F.

Afterburners followed by a lime injected fabric filter system are

used to control emissions at 13 of the 46 scrap dryers/delacquering

kilns/decoating kilns in the industry. The best performing 12 percent

of the existing 46 scrap dryers/delacquering kilns/decoating kilns are

equipped with an

[[Page 6961]]

afterburner for organics (i.e., THC and D/F) control and a lime

injected fabric filter for controlling HCl, D/F, PM and HAP metals.

Therefore, the floor level of control for existing sources is

determined by the average/median of the best performing 6 sources

within the category. This median level of control is represented by a

well designed and operated afterburner with a minimum of 1-second

residence time and operated at a temperature of 1400 deg.F followed by

a pulse-jet fabric filter using fiberglass bags with an air to cloth

ratio of about 4.0 and continuous lime injection.

The existing source MACT is also the MACT floor for new sources

since it is also the level of control achieved by the best controlled

source.

Sweat furnaces. Based on data provided in the ICR responses, the

EPA identified 12 sweat furnaces in the industry. These sources reclaim

aluminum from scrap containing high levels of iron by heating the scrap

to a temperature above the melting point of aluminum but below that of

iron. Emissions test measurements show that these sources emit THC and

D/F. Six of the 12 sweat furnaces are equipped with afterburners to

control THC and D/F. The MACT floor, for categories of less than 30

sources is determined by the median of the five best controlled sources

in the category. Therefore, afterburners represent the MACT floor level

of control for existing sweat furnaces. An afterburner representative

of this median level of control is designed for a minimum of 1-second

residence time and operated at a temperature of 1600 deg.F.

The existing source MACT is also the MACT floor for new sources

since it is also the level of control achieved by the best controlled

source.

Group 1 furnaces. Existing group 1 furnaces are emission units

within a secondary aluminum processing unit affected source. Each new

and reconstructed group 1 furnaces is a separate affected source. The

EPA identified 528 Group 1 furnaces based on information provided in

the ICR responses. Approximately one-half of these furnaces operate

with no add-on air pollution control devices. Emissions test

measurements show that these sources emit PM (containing particulate

metal HAP), HCl, and D/F. The add-on controls used on group 1 furnaces

include fabric filters, lime coated fabric filters, lime injected

fabric filters, cyclones, incinerators and wet scrubbers.

Other furnaces in group 1 limit emissions through the use of work

practices, design practices, and pollution prevention approaches. These

techniques include, but are not limited to, charging only clean scrap

to the furnaces and design and work practice approaches for fluxing,

limiting oil and coatings content of furnace charges through the use of

scrap purchasing specifications and scrap inspection, fluxing only in

holding furnaces, fluxing in in-line fluxers, and limiting the use of

reactive fluxes. Work practices and pollution prevention approaches may

also be combined with add-on controls to achieve HAP reductions.

Lime injected fabric filter systems are used to control emissions

at 68 of the 528 group 1 furnaces in the industry. The best performing

12 percent of the existing 528 group 1 furnaces are equipped with a

lime injected fabric filter for controlling HCl, PM and HAP metals, and

for controlling D/F from those furnaces which process scrap containing

oil and coatings. Therefore, the floor level of control achievable by

existing emission units is determined by the average/median of the best

performing 63 sources within the category. This median level of control

is represented by a well designed and operated pulse jet fabric filter

with an air to cloth ratio of about 6.5 and continuous lime injection.

The level of control achievable by existing emission units

represents the MACT floor for new sources since it is also the level of

control achieved by the best controlled source.

Group 2 furnaces. Based on the ICR data, the EPA estimates that

about 75 group 2 furnaces are currently in operation. None of the

furnaces in group 2 are equipped with add-on air pollution control

devices. Emissions from these furnaces are typically controlled by work

practices that require charging only clean charge materials, coupled

with fluxing operations using only non-reactive agents (i.e. fluxes

which do not contain or produce HAPs). Since emissions from these units

are at very low levels and considering the cost of emissions testing,

the application of emission measurement methodology and setting

specific emissions limits for this particular class of source is not

practicable due to economic limitations. Thus, work practice procedures

under section 112(h) of the Act (limitations on type of charge and type

of flux used) constitute the MACT floor level of control for existing

Group 2 furnaces as well as MACT for new group 2 furnaces.

Dross-only furnaces. Based on the information reported in the ICR,

the EPA identified 15 dross-only furnaces. Emissions test measurements

show that these sources emit PM (containing particulate metal HAP). All

dross-only furnaces are equipped with control systems that include a

fabric filter, some of which have lime injection systems. The MACT

floor, for categories of less than 30 sources is determined by the

median of the five best controlled sources in the category. The ICR

data show that the control technology in place at the five best-

controlled sources is a lime injected fabric filter. Therefore, lime

injected fabric filters represent the MACT floor level of control for

existing dross-only furnaces. The technology at the median level of

control is represented by a well designed and operated fabric filter

with polyester bags at an air to cloth ratio of 6.5 to 1 with

continuous lime injection.

The existing source MACT floor is also the MACT floor for new

sources since it is also the level of control achieved by the best

controlled source.

Rotary dross coolers. The EPA identified 26 rotary dross coolers

based on the information provided in the ICR responses. Emissions test

measurements show that these sources emit PM (containing particulate

metal HAP). All 26 rotary coolers are equipped with fabric filters. The

MACT floor, for categories of less than 30 sources is determined by the

median of the five best controlled sources in the category. Therefore,

fabric filters represent the MACT floor level of control for existing

rotary dross coolers. A fabric filter representative of the median of

the best 5 controlled sources is a well designed and operated pulse-jet

fabric filter system using polyester bags with an air to cloth ratio of

3.0.

The existing source MACT floor is also the MACT floor for new

sources since it is also the level of control achieved by the best

controlled source.

In-line fluxers. Existing in-line fluxers are emission units within

a secondary aluminum processing unit affected source. Each new and

reconstructed in-line fluxer is a separate affected source. The EPA

identified a total of 120 in-line fluxers (also referred to as

degassing boxes) from the information reported in the ICR responses.

Emissions test measurements show that in-line fluxers are sources of

low concentrations of PM (containing particulate metal HAP) and HCl.

Eleven in-line fluxers are controlled by fabric filters and 7 of these

have lime (or other alkaline reagent) injection systems. The average of

the best performing 12 percent of the existing 120 in-line fluxers is

represented by a lime injected fabric filter for controlling HCl, PM

and HAP metals. The level of control achievable by existing emission

units is represented by a well designed and operated pulse-jet fabric

filter using

[[Page 6962]]

fiberglass bags with an air to cloth ratio of about 7.0 and continuous

lime injection.

The level of control achievable by existing emission units

represents the MACT floor for new sources since it is also the level of

control achieved by the best controlled emission unit.

Secondary aluminum processing units. A secondary aluminum

processing unit consists of all of the existing group 1 furnace

emission units and all of the existing in-line fluxer emission units at

a secondary aluminum production facility. The MACT floor level of

control is determined by applying the level of control achievable to

each emission unit within the affected source. As described in the

paragraphs in this section of the document which address the

determination of the MACT floor for group 1 furnaces and in-line

fluxers, this is represented by the level of control achieved by a lime

injected fabric filter of appropriate design, coupled with continuous

lime injection. Each new or reconstructed group 1 furnace or in-line

fluxer is a separate affected source subject to the MACT floor emission

limitations as described in the paragraphs in this section of the

document which address the determination of the MACT floor for group 1

furnaces and in-line fluxers.

3. Consideration of Beyond-the-Floor Technologies

The EPA investigated beyond-the-floor controls for each pollutant

and affected source regulated by the proposed rule. For each of the

cases evaluated, the Agency did not identify cost-effective emission

control technologies that would accomplish additional emission

reductions to a level below that achieved by the MACT floor technology.

Therefore, the Agency is proposing emission limits at the MACT floor

level of control.

4. Selection of Emission Limits

The EPA and industry conducted comprehensive emission tests at 12

facilities to characterize uncontrolled and controlled emissions from

the various processes and to evaluate the effectiveness of existing

control devices and work practice and pollution prevention approaches.

Sites with add-on control technologies selected for emission testing

represented the use of technology identified by the EPA as the MACT

floor technology. Other sites were tested where work practice and

pollution prevention approaches were used to achieve HAP emission

reductions. Data from these sites showed that work practices and

pollution prevention approaches could achieve HAP emission levels

similar to those achieved with add-on MACT floor technologies.

Therefore, the EPA is proposing a combination of work practice/

pollution prevention based standards and MACT floor control technology

based numerical emission limits for control of HAP from affected

sources subject to the proposed rule.

The EPA is, in most cases, proposing emission limits in a mass per

unit (e.g., kg/Mg or lb/ton) of feed format. This format provides

several advantages. For example, for process units that release

emissions from more than one stack and where multiple similar affected

sources are controlled by a common control device, total emission rates

can be determined by measuring emissions for a particular pollutant

from each stack or discharge point, e.g. lbs/hr, adding those, and

dividing by the sum of all affected source feed rates, e.g. tons/hr. In

addition, this format is tied to production and the emission limits are

unaffected by dilution. In specific cases, concentration based

numerical emission limits, or minimum percentage reduction standards

are appropriate; the format of these standards is explained in the

discussion of these emission standards.

All limits on particulate metal HAP emissions are expressed in

terms of a surrogate pollutant, PM. The use of the surrogate PM

emissions limit will require the installation and operation of the

appropriate MACT floor technology for metal HAPs control from new and

existing sources. Use of PM as a surrogate for metal HAPs also has the

advantage of simplifying and reducing the cost of performance testing

and monitoring.

Except for D/F which merits special consideration due to high

toxicity, all emission standards for gaseous organic HAPs are expressed

in terms of a surrogate pollutant, THC. The use of a surrogate THC

emissions limit for gaseous organic HAPs will require facilities to

install and operate the appropriate MACT floor technology for gaseous

organic HAPs from new and existing sources.

All limits on D/F emissions are expressed in units of toxic

equivalent (TEQ). Toxic equivalent refers to the international method

of expressing toxicity equivalents for dioxins and furans as defined in

the EPA report, ``Interim Procedures for Estimating Risks Associated

with Exposures to Mixtures of Chlorinated Dibenzo-p-dioxins and -

dibenzofurans (CDDs and CDFs) and 1989 Update'' (docket item II-A-1).

In addition to the emission limits discussed below, the EPA is also

proposing a 10 percent opacity limit applicable to affected sources

with fabric filter control devices that choose to monitor with a COM

and affected scrap shredders that choose to monitor with a COM or by

visible emissions monitoring. During the course of many emission tests

conducted at secondary aluminum facilities, the EPA has determined that

the exhaust gases from properly designed, operated, and maintained

fabric filters have essentially zero opacity. An opacity of 10 percent

or greater following a successful performance test on a fabric filter

controlled affected source is a clear indication that the control

device is not functioning properly.

Scrap shredders. The proposed PM limit for scrap shredders and

crushers of 23 mg/dscm, (0.010 gr/dscf) is based on test results from

four facilities equipped with well designed and operated fabric filters

representative of the MACT floor technology for new and existing

sources where PM measured emissions ranged from 0.0002 gr/dscf to

0.0069 gr/dscf. The EPA took into consideration the wide variation in

controlled emissions for the four MACT floor fabric filter systems in

selection of the emission limits of 23 mg/dscm (0.010 gr/dscf). Such a

range in performance represents the typical variations associated with

the process and with application of the floor technology. The proposed

PM emission limit represents a level that can be achieved by all scrap

shredders and crushers using the MACT floor technology. The supporting

emissions data are presented in Figure 1 and Table 3 below.

BILLING CODE 6560-50-P

[[Page 6963]]

[GRAPHIC] [TIFF OMITTED] TP11FE99.004

BILLING CODE 6560-50-C

Table 3.--Summary of Scrap Shredders and Crushers Particulate Emissions Test Data

----------------------------------------------------------------------------------------------------------------

Average PM emissions

Plant Control device -------------------------------

PM (gr/dscf) PM (mg/dscm)

----------------------------------------------------------------------------------------------------------------

24......................................... Fabric Filter...................... 0.0022 5.0

25......................................... Fabric Filter...................... 0.0069 15.8

26......................................... Fabric Filter...................... 0.0002 0.46

27......................................... Fabric Filter...................... 0.0008 1.8

----------------------------------------------------------------------------------------------------------------

For this affected source, a concentration format is appropriate

because PM concentration is easily and reliably measured from these

sources and PM concentration reflects fabric filter performance, the

technology representative of MACT for new and existing sources.

The EPA is also proposing a 10 percent opacity limit applicable to

fabric filters applied to scrap shredder waste gas streams if the owner

or operator chooses to monitor either with a COM or by visible

emissions monitoring. As noted above, the EPA has determined that the

presence of a 10 percent or greater opacity discharge from a fabric

filter following a successful performance test is a clear indication

that the device is not functioning properly.

Chip dryers. One chip dryer with a well designed and operated

afterburner representative of the MACT floor was tested. The controlled

THC emissions from tests at this facility averaged 0.21 kg/Mg (0.42 lb/

ton) of feed and the D/F emissions averaged 1.3 /Mg D/F TEQ

(1.7 x 10 -5 gr/ton) of feed. The data are shown in Figure 2

below.

BILLING CODE 6560-50-P

[[Page 6964]]

[GRAPHIC] [TIFF OMITTED] TP11FE99.005

BILLING CODE 6560-50-C

[[Page 6965]]

Based on these data the EPA is proposing a THC limit of 0.40 kg/Mg

(0.80 lb/ton) of feed and a D/F (TEQ) limit of 2.5 ``g/Mg (3.5

x 10-5 gr/ton) of feed.

Scrap dryers/delacquering kilns/decoating kilns.

The same process equipment can function as a scrap dryer, a

delacquering kiln, or a decoating kiln. Equipment of an identical

design is capable of performing different functions by changing the

operating temperature and charge make-up. In addition, the control

technology representative of MACT for new and existing sources is the

same for kilns operating as scrap dryers and kilns operating as

delacquering/decoating kilns. The EPA/industry task group spent

considerable effort trying to define scrap dryers and delacquering/

decoating kilns such that separate emission standards could be set for

each. Despite this substantive effort, the task group was unable to

develop consistent, unambiguous definitions which would permit the

establishment of different classes of scrap dryers, delacquering kilns,

or decoating kilns. In recognition of the different operating modes

applicable to these affected sources such as operating temperatures,

charge make-up, difference in uncontrolled emission levels; to provide

operational flexibility; and to ensure that the technology

representative of the MACT floor for new and existing sources is

installed and properly operated at these sources, the EPA is proposing

two alternate sets of emission standards.

One set of emission standards is based on emissions data obtained

from a kiln operating as a delacquering/decoating kiln with an

operating temperature about 1,000 deg.F and processing only coated

materials, such as painted siding and used beverage containers, and

operating a well designed afterburner/lime injected fabric filter

system representative of MACT for new and existing sources. This set of

standards for PM, HCl, THC, and D/F is summarized in Table 4.

Table 4. Summary of Emission Limits for Scrap Dryers, Delacquering Kilns, and Decoating Kilns Operating as

Delacquering Kilns

----------------------------------------------------------------------------------------------------------------

PM (lb/ton of HCl (lb/ton of THC (lb/ton of D/F (g/

Process feed) feed) feed) Mg of feed)

----------------------------------------------------------------------------------------------------------------

Scrap Dryer, Delacquering Kiln, Decoating

Kiln....................................... 0.080 0.80 0.060 0.25

----------------------------------------------------------------------------------------------------------------

The other set of emission standards is based on the emissions data

obtained from a kiln that had an operating temperature of about

700 deg.F and was processing scrap with oils, coatings, paints,

insulation, etc. The control technology in use was an afterburner/lime

injected fabric filter system representative of MACT for new and

existing sources. That set of standards and control device design and

operating requirements is summarized in Table 5.

Table 5.--Summary of Alternate Emission Limits and Control Equipment Requirements for Scrap Dryers, Delacquering Kilns, and Decoating Kilns Operating as

Scrap Dryers

--------------------------------------------------------------------------------------------------------------------------------------------------------

Afterburner design and operating

requirements

Process PM (lb/ton of HCl (lb/ton of THC (lb/ton of D/F (g/---------------------------------

feed) feed) feed) Mg of feed) Temperature ( Residence timea

deg.F) (seconds)

--------------------------------------------------------------------------------------------------------------------------------------------------------

Scrap Dryer, Delacquering Kiln, Decoating Kiln.... 0.30 1.50 0.20 5.0 1,400 1.0

--------------------------------------------------------------------------------------------------------------------------------------------------------

a Afterburner design residence time.

The first set of proposed emission limits for scrap dryers,

delacquering kilns, decoating kilns in Table 4 is supported by the

delacquering emissions data summarized in Table 6 and Figure 3. Under

this set of standards an operator is required to meet a more stringent

set of emission limits, but the afterburner design parameters are not

requirements.

BILLING CODE 6560-50-P

[[Page 6966]]

[GRAPHIC] [TIFF OMITTED] TP11FE99.006

[[Page 6967]]

[GRAPHIC] [TIFF OMITTED] TP11FE99.007

BILLING CODE 6560-50-C

[[Page 6968]]

Table 6.--Summary of Scrap Dryer, Delacquering Kiln, Decoating Kiln Emissions Data With MACT Controls

----------------------------------------------------------------------------------------------------------------

D/F (g/Mg of

feed) feed) feed) feed)

----------------------------------------------------------------------------------------------------------------

2--Scrap Dryer.................................. 0.167 0.827 .............. ..............

3--Scrap Dryer.................................. 0.214 1.26 a 0.072 a 2.66

4--Delacquering................................. b 0.00057 b 0.544 b 0.006 b 0.118

5--Delacquering................................. c 0.024 .............. c 0.037 ..............

d 0.051 .............. d 0.035 ..............

----------------------------------------------------------------------------------------------------------------

a Calculated by applying the afterburner efficiency to the uncontrolled fugitive emissions escaping from the

kiln product discharge point. These emissions are supposed to be captured and controlled by the afterburner

but problems during testing allowed emissions to escape from the kiln end where material leaves the process.

b Emissions test of kiln processing used beverage containers for D/F test and painted siding for all other

tests.

c Emissions test of kiln processing used beverage containers.

d Emissions test of kiln processing painted siding.

Because of the lower level of uncontrolled emissions generated when

a kiln is operated as a delacquering kiln (i.e., operating temperature

of about 1,000 deg.F and processing used beverage containers and

painted siding only), an operator could conceivably operate a kiln

primarily as a delacquering/decoating kiln but add a small amount of

materials, such as oils or insulation, and classify it as a scrap

dryer. In this case the operator could thereby operate with less than

the MACT floor control equipment 1400 deg.F and 1 second residence time

afterburner design, while only reducing emissions to the level of the

less stringent alternate emission

limits. To preclude this, the EPA is specifying minimum afterburner

design and operating requirements of 1 second residence time and

1400 deg.F, MACT floor technology, for those operators electing to

process material with oils, coatings, and insulation, in addition to

used beverage containers and painted siding, thus operating the

equipment as a scrap dryer rather than a delacquering/decoating kiln.

The EPA is proposing the second, or alternate, set of emission

standards based on data obtained from a kiln being operated as a scrap

dryer. These alternate limits are combined with control device design

and operating requirements to ensure that

control technology representative of MACT is used when an operator

chooses to comply with the higher, or less stringent, emission limits

associated with a scrap dryer processing scrap with oils, coatings,

paints, etc.

As noted above, the emissions data supporting the second or

alternate emission limits were obtained from a kiln operating as a

scrap dryer at a temperature of about 700 deg.F. These data are

summarized in Table 6 and shown in Figure 4. The control technology in

use was an afterburner/lime injected fabric filter system

representative of MACT for new and existing sources.

BILLING CODE 6560-50-P

[[Page 6969]]

[GRAPHIC] [TIFF OMITTED] TP11FE99.008

[[Page 6970]]

[GRAPHIC] [TIFF OMITTED] TP11FE99.009

BILLINC CODE 6560-50-C

[[Page 6971]]

The EPA is also proposing a 10 percent opacity limit applicable to

fabric filters applied to scrap dryer, and delacquering and decoating

kiln waste gas streams if a COM is chosen as the monitoring option. As

noted above, the EPA has determined that the presence of a 10 percent

or greater opacity discharge from a fabric filter following a

successful performance test is a clear indication that the device is

not functioning properly.

Sweat furnaces. EPA tested one sweat furnace equipped with a well

designed and operated afterburner representative of MACT for new and

existing sources.

Controlled D/F emissions averaged 0.35 ng/dscm (1.5 x

10-10 gr/dscf) and are shown in Figure 5. Based on these

data, the EPA is proposing a D/F limit for sweat furnaces of 0.80 ng/

dscm D/F TEQ (3.5 x 10-10 gr/dscf) corrected to an 11

percent oxygen basis.

BILLING CODE 6560-50-P

[GRAPHIC] [TIFF OMITTED] TP11FE99.010

BILLING CODE 6560-50-C

A concentration limit, as opposed to a production based limit, is

proposed for this source because materials charged to these furnaces

are typically introduced in a random fashion without being weighed.

Consequently, determining an emission rate per unit of feed is not a

practical option as a format for the emission limit.

Dross-only furnaces. The EPA/industry tested one dross only furnace

equipped with a well designed and operated fabric filter representative

of the MACT floor for new and existing sources. The PM emissions from

tests at this facility averaged 0.104 kg/Mg of feed (0.207 lb/ton).

Based on these data as shown in Figure 6, the EPA is proposing a PM

limit of 0.15 kg/Mg of feed (0.30 lb/ton).

BILLING CODE 6560-50-P

[[Page 6972]]

[GRAPHIC] [TIFF OMITTED] TP11FE99.011

BILLING CODE 6560-50-C

The EPA is also proposing a 10 percent opacity limit applicable to

fabric filters applied to dross-only furnace waste gas streams if a COM

is chosen as the monitoring option. As noted above, the EPA has

determined that the presence of a 10 percent or greater opacity

discharge from a fabric filter following a successful performance test

is a clear indication that the device is not functioning properly.

Rotary dross coolers. The EPA/industry tested two rotary dross

coolers equipped with a well designed and operated fabric filter

representative of the MACT floor technology for new and existing

sources. The PM emissions from tests at these facilities averaged 2.29

and 75.5 mg/dscm (0.001 and 0.033 gr/dscf), respectively. These data

are summarized in Table 7 and Figure 7.

Table 7.--Summary of Rotary Dross Cooler Emission Data

------------------------------------------------------------------------

PM (gr/

Plant PM (mg/dscm) dscf)

------------------------------------------------------------------------

21........................................... 2.29 0.001

22........................................... a 75.5 a 0.033

------------------------------------------------------------------------

a Plant 22 is equipped with a lime-injected fabric filter.

BILLING CODE 6560-50-P

[[Page 6973]]

[GRAPHIC] [TIFF OMITTED] TP11FE99.012

BILLING CODE 6560-50-C

Based on these data the EPA is proposing a PM limit of 92 mg/dscm

(0.040 gr/dscf). The proposed PM emission limit represents a level that

can be achieved by all rotary dross coolers using the floor technology

for new and existing sources.

The EPA is also proposing a 10 percent opacity limit applicable to

fabric filters applied to rotary dross cooler waste gas streams if a

COM is chosen as the monitoring option. As noted above, the EPA has

determined that the presence of a 10 percent or greater opacity

discharge from a fabric filter following a successful performance test

is a clear indication that the device is not functioning properly.

In-line fluxers. The EPA/industry tested one in-line fluxer

equipped with a well designed and operated fabric filter with

continuous lime injection representative of the control which is

achievable for these emission units. Additional performance test data

from the same in-line fluxer was also available (see docket item II-B-

19, historical data memo). The PM emissions from tests performed at

this facility averaged 0.00170 kg/Mg (0.00340 lb/ton) of feed and are

shown in Figure 8. Based on these data the EPA is proposing a PM limit

of 0.005 kg/Mg (0.01 lb/ton) of feed for new and reconstructed in-line

fluxers. The HCl emissions from tests at this facility averaged 0.0072

kg/Mg (0.014 lb/ton) of feed and are also shown in Figure 8. Based on

these data the EPA is proposing an HCl limit of 0.02 kg/Mg (0.040 lb/

ton) of feed for new and reconstructed in-line fluxers.

BILLING CODE 6560-50-P

[[Page 6974]]

=======================================================================

[GRAPHIC] [TIFF OMITTED] TP11FE99.013

BILLING CODE 6560-50-C

[[Page 6975]]

The EPA is also proposing a 10 percent opacity limit applicable to

fabric filters applied to in-line fluxer waste gas streams if a COM is

chosen as the monitoring option. As noted above, the EPA has determined

that the presence of a 10 percent or greater opacity discharge from a

fabric filter following a successful performance test is a clear

indication that the device is not functioning properly.

Furnace Operations The EPA spent considerable effort analyzing ICR

data and emissions data to evaluate the need for different classes for

the remaining furnace types and configurations. Operating practices,

control practices, work practices, pollution prevention efforts,

furnace charge materials, flux rates and methods, and emissions vary

widely within the industry. All of these factors entered into the

consideration of different classes (Ref. ICR database, emission data

summaries). In addition, there were many meetings and discussions with

the industry to discuss and evaluate a multitude of options and issues

associated with each factor. At one time, as many as five potential

classes were under consideration and discussion. As analyses of the

potential classes progressed, many issues were raised regarding

definitions of the classes, process operating practices, and control

approaches. Further, as potential emissions limits for these classes

were discussed, it became evident to the EPA that these furnaces could

be compressed into two classes. Therefore, based on evaluation of these

options, the EPA is proposing two classes for process furnace

operations:

Group 2 furnaces--clean charge materials with no reactive

fluxing.

Group 1 furnaces--furnaces charging different gradations

of clean materials with reactive fluxing to dirty materials with

various fluxing amounts/techniques.

Group 2 furnaces. For group 2 furnaces the EPA is proposing work

practice/pollution prevention practices under section 112(h) of the

Act. Section 112(h) of the Act provides for the establishment of work

practice standards where it is not feasible to prescribe or enforce an

emission standard.

The MACT floor for new and existing sources for this group of

furnaces consists of work practices/pollution prevention practices

including charging and melting only ``clean'' charge materials, as

defined in the proposed regulation (molten aluminum, T-bar, sow, ingot,

alloying elements, uncoated aluminum chips, aluminum scrap dried/

delacquered/decoated, and noncoated runaround scrap), and no reactive

fluxing. Compliance with the standard would be demonstrated by labeling

of the furnace as group 2, and record keeping of charge and flux

materials along with certification every six months that only clean

charges were used and that no reactive flux was used in the furnace.

The Administrator has determined it is not feasible to prescribe an

emission standard for this class of furnaces because the application of

measurement methodology is not practicable due to economic limitations.

Group 1 furnaces. Group 1 furnaces consist of all process (melting,

holding, refining) furnaces that do not meet the requirements for a

group 2 furnace. These include combinations of:

(1) Dirty furnace charge materials and fluxing with or without

reactive fluxes, and

(2) Clean furnace charge materials (work practices) with use of

reactive fluxing.

The achievable emissions limitation for group 1 furnace emission

units and the standard for new and reconstructed group 1 furnaces is

based on furnaces in which dirty charge materials and unlimited fluxing

are used, and that are equipped with the MACT floor control technology,

a fabric filter with a continuous lime injection system. The proposed

limits for new and reconstructed group 1 furnaces are shown in Table 8.

The basis and rationale for these limits are provided in the emission

test data graphs and discussion below.

Table 8.--Summary of Group 1 Furnace Emission Limits for New and Reconstructed Sources (Except Melter/Holders

Processing Clean Charge)

----------------------------------------------------------------------------------------------------------------

HCla

Process PM (lb/ton) D/F (g ---------------------------------

TEQ/Mg) (lb/ton) Removal (%)

----------------------------------------------------------------------------------------------------------------

Group 1 Furnaces............................ 0.40 15 0.40 90

----------------------------------------------------------------------------------------------------------------

a Facilities with add-on control devices will choose which requirement to comply with.

To meet the emission limits based on MACT floor technology, not all

new and reconstructed group 1 furnaces will have to be equipped with

lime injected fabric filter systems. Work practices, pollution

prevention practices, process design changes, charging clean or almost

clean materials, and reduced use of reactive fluxes while controlling

the reactive flux injection rate are some control approaches that may

be applied to some group 1 furnace installations with varying add-on

control approaches such that the resulting HCl and other HAP emissions

are below the emission limits being proposed.

To determine the emissions limitations achievable by group 1

furnace emission units and to establish the emission limits for new and

reconstructed group 1 furnaces, the EPA and industry tested furnaces in

6 facilities (Plants 6 through 11) with the MACT floor technology

applied. The emissions data are presented in Figures 9, 10, and 11

below. The furnace emissions data with control status labeled as ``lime

baghouse'' were equipped with the MACT floor technology.

BILLING CODE 6560-50-P

[[Page 6976]]

[GRAPHIC] [TIFF OMITTED] TP11FE99.014

[[Page 6977]]

[GRAPHIC] [TIFF OMITTED] TP11FE99.015

[[Page 6978]]

[GRAPHIC] [TIFF OMITTED] TP11FE99.016

BILLING CODE 6560-50-C

[[Page 6979]]

In addition, the EPA and industry tested group 1 furnaces that had

no add-on control technologies, but used work practices/pollution

prevention practices such as process design changes that allowed

reduced levels of reactive fluxing, as well as selective scrap charging

(but not ``clean charge''), to achieve lower levels of HAP emissions.

Both melting and holding furnaces were included in these tests. These

results are also shown in Figures 9, 10, and 11. These furnace data are

labeled with control status as ``work practice.''

All of the data in Figures 9, 10, 11 were considered in determining

the achievable emissions limitations for group 1 furnace emission units

and in establishing the proposed emission limits for new and

reconstructed individual Group 1 furnaces that are listed in Table 8

above. Some of the variations in the work practice/ pollution

prevention emissions are due to different design of process, work

practice, and pollution prevention alternatives, and the fact that

these emissions will vary with the differing grades of aluminum

produced.

Average PM emission levels from group 1 furnaces equipped with MACT

floor add-on air pollution control devices varied from a low of 0.029

to a high of 0.28 lb/ton of feed. Average HCl emission levels from

furnaces equipped with MACT floor add-on air pollution control devices

varied from a low of 0.07 to a high of 0.36 lb/ton of feed. The

equivalent ranges of emissions for the work practice/pollution

prevention practice furnaces were 0.019 to 0.37 lb/ton and 0.001 to

0.36 lb/ton of PM and HCl, respectively.

The three test results for average D/F emissions from group 1

furnaces equipped with MACT floor add-on air pollution control devices

ranged from a low value of 0.46 to a high value of 4.5 g D/F

TEQ/Mg of feed. For the four work practice/pollution prevention

practice furnaces, the range was 0.21 to 0.41 g D/F TEQ/Mg.

To provide another perspective on the achievable D/F emission

limitation, the 15 g/Mg of feed emission limit (proposed for

new and reconstructed group 1 furnaces) expressed on a concentration

basis for the furnaces tested would be about 0.9 to 15.5 ng D/F TEQ/

dscm depending on the quantity of waste gas flow from the furnace.

The proposed standards for new and reconstructed group 1 furnaces

shown in Table 8 provide the option of achieving a 90 percent emission

reduction in HCl discharged from the furnace in lieu of meeting an

emission limit of 0.40 lb/ton. The EPA considered that group 1 furnaces

can be used to process a wide variety of scrap types (i.e., clean, with

insulation, oils, coated, painted, etc.) and perform various fluxing

operations with multiple agents including HAP producing and non-HAP

producing fluxes (i.e., salts, chlorine gas, nitrogen/chlorine bi-gas,

etc.) to produce a wide range of aluminum alloys. Because of the

potential differences in charge make-up, fluxing, work practices, and

final aluminum properties, there is potential for variability in HCl,

organic HAPs, particulate metal HAPs, and D/F emitted by the group 1

furnaces. In recognition of the different operating modes applicable to

these emission units and affected sources and to promote the most cost-

effective and economical approach to MACT controls while achieving the

MACT add-on air pollution control device equivalent reductions, the EPA

is proposing a dual HCl emission standard for new and reconstructed

group 1 furnaces. Both a numerical emission limit and an alternate

percent reduction requirement are being proposed. Some furnaces process

scrap that contains relatively large amounts of chloride compounds.

This factor in combination with high fluxing rates necessary to refine

some aluminum can yield control device inlet HCl quantities in excess

of 4 lbs/ton of feed. In these circumstances the floor technology may

not be able to meet the limit of 0.40 lb/ton, but can comply with the

90 percent removal requirement which is representative of what the MACT

floor technology is capable of achieving. Test results from Plants 7,

9, and 10, shown in Figure 10, indicated that HCl efficiencies in

excess of 90 percent removal were achieved. The range of variation in

measured efficiencies was significant at two facilities with some test

results below 90 percent. In these tests the lime usage rates were not

adequately controlled to achieve consistent HCl removal, hence a wide

variation in HCl removals resulted.

The level of removal achievable became an issue with the industry

and to resolve this issue the EPA tested another group 1 furnace in

Plant 11 with a lime injected fabric filter. During these tests the

lime injection rate was controlled to consistently achieve greater than

90 percent removal of HCl. Individual test results for this furnace are

shown in Table 9. These and other data demonstrate that fabric filters

operated with continuous lime injection into the gas stream upstream of

the fabric filter inlet are capable of consistently achieving at least

90 percent removal.

Table 9.--Plant 11 HC1 Individual Test Results

------------------------------------------------------------------------

Inlet lb/ Outlet Percent

Test No. ton lb/ton removal

------------------------------------------------------------------------

1...................................... 2.64 .018 99.3

2...................................... 2.66 0.020 99.2

3...................................... 1.31 0.050 96.2

4...................................... 2.10 0.028 98.7

------------------------------------------------------------------------

New and reconstructed group 1 furnaces processing clean charge

materials only, that perform both melting and holding functions

including reactive fluxing within the same unit (i.e., melter/holder),

and that do not transfer molten aluminum to or from another furnace

would be subject to alternate standards. These units perform the

operations normally carried out in two or more separate furnaces within

the confines of one furnace. Emission data obtained from tests on a

melter/holder furnace are shown in Figure 12.

BILLING CODE 6560-50-P

[[Page 6980]]

[GRAPHIC] [TIFF OMITTED] TP11FE99.017

BILLING CODE 6560-50-C

Emission limits are proposed for PM and HCl emissions from new and

reconstructed group 1 melter/holders. Those limits are shown in Table

10. The PM standard for new and reconstructed group 1 melter/holder

furnaces processing only clean charge materials is 0.40 kg/Mg (0.80 lb/

ton) of charge and the alternate HCl standard is 0.20 kg/Mg (0.40 lb/

ton) of charge.

Table 10.--Summary of New and Reconstructed Group 1 Melter/Holder Emission Limits

----------------------------------------------------------------------------------------------------------------

D/Fb (g TEQ/Mg) HCl (lb/ton)

----------------------------------------------------------------------------------------------------------------

Group 1 Melter/Holder Furnaces a. 0.80 .............. 0.40 or 90 percent removal.

----------------------------------------------------------------------------------------------------------------

a Performing both melting and holding functions in the same furnace and processing only clean charge materials.

b No dioxin limit because this furnace uses clean charge.

[[Page 6981]]

Operators of group 1 side-well furnaces would be permitted to

conduct reactive fluxing operations in the furnace side-well only. If

reactive fluxing operations are conducted in the furnace hearth, those

emissions must be captured and ducted to a control device. In this

event total furnace emissions (hearth plus side-well) would be subject

to the new and reconstructed group 1 furnace emission limits.

In addition to the above standards, the EPA is also proposing a 10

percent opacity limit applicable to the waste gas discharge from any

fabric filter applied to a group 1 furnace if a COM is chosen as the

monitoring option. As noted above, the EPA has determined that the

presence of a 10 percent or greater opacity discharge from a fabric

filter following a successful performance test is a clear indication

that the device is not functioning properly.

Secondary aluminum processing units. Available data from existing

group 1 furnace emission units and existing in-line fluxers were

analyzed to determine the emissions limitations which could be realized

through the application of add-on control devices and pollution

prevention/work practices. These data have been presented in the

paragraphs in this section of this document relating to group 1

furnaces and in-line fluxers. A secondary aluminum processing unit is

composed of all of the existing group 1 furnace emission units and all

of the existing in-line fluxer emission units at a secondary aluminum

production facility. Emission standards for this affected source have

been proposed, based on throughput weighted processing of material in

emission units controlled to achievable emission limitations. Limits

for PM, HCl and D/F have been proposed on a production basis.

(Operators of group 1 furnaces with very high potential HCl emissions

may choose to calculate the HCl limit for any or all individual group 1

furnace emission units on the basis of achieving a 90 percent reduction

in potential HCl emissions.) Based on the emissions achievable by

individual emission units, the following standards are proposed:

[GRAPHIC] [TIFF OMITTED] TP11FE99.018

[GRAPHIC] [TIFF OMITTED] TP11FE99.019

[GRAPHIC] [TIFF OMITTED] TP11FE99.020

Where:

LiPM=the PM emission limit for individual emission unit i in

the secondary aluminum processing unit kg/Mg (lb/ton) of feed]

Ti=the feed rate for individual emission unit i in the

secondary aluminum processing unit

LtPM=the overall PM emission limit for the secondary

aluminum processing unit [kg/Mg (lb/ton) of feed]

LiHCl=the HCl emission limit for individual emission unit i

in the secondary aluminum processing unit [kg/Mg (lb/ton) of feed].

Operators may choose to calculate this limit on the basis of 90 percent

reduction in potential HCl emissions.

LtHCl=the overall HCl emission limit for the secondary

aluminum processing unit [kg/Mg (lb/ton) of feed]

LiD/F=the D/F emission limit for individual emission unit i

[g/Mg (gr/ton) of feed]

LtD/F=the overall D/F emission limit for the secondary

aluminum processing unit [g/Mg (gr/ton) of feed], and

n=the number of units in the secondary aluminum processing unit.

The emissions limits LiPM, LiHCl, and

LiD/F to be used in calculating the proposed standards for

secondary aluminum processing units are those proposed for individual

new and reconstructed in-line fluxers and group 1 furnaces. Production

in clean charge group 1 furnaces can not be included in calculating the

overall D/F emission limit, because it is assumed that these furnaces

are capable of operation with no D/F emissions, and because these

emission units are not subject to D/F limits. In-line fluxers that

operate using no reactive flux materials cannot be included in the

calculations of the overall PM and HCl emission limits since they are

not subject to emission limits for PM and HCl.

In addition to the above standards, the EPA is also proposing a 10

percent opacity limit applicable to the waste gas discharged from any

fabric filter applied to a furnace process train if a COM is chosen as

the monitoring option. As noted above, the EPA has determined that the

presence of a 10 percent or greater opacity discharge from a fabric

filter following a successful performance test is a clear indication

that the device is not functioning properly.

D. Selection of Operating and Monitoring Requirements

The EPA identified and analyzed the hierarchy of monitoring options

available for this source category. The array of monitoring options

includes the direct measurement of HAP or HAP surrogates by a CEM or

COM, periodic performance tests, continuous monitoring of process or

control device operating parameters that are related to emissions of

HAP, and recordkeeping and certification requirements. Each option that

was relevant to a process or add-on control device was evaluated

relative to its technical feasibility and cost.

A CEM provides a direct measurement of emissions of HAP or HAP

surrogates. CEMs are commercially available for HCl and THC. PM CEMs

are also available, however, the technical feasibility of these devices

for monitoring affected sources and emission units in this source

category has not yet been demonstrated, and the estimated capital cost

of PM monitoring systems is $213,000 with annual costs of $66,000 (see

docket item II-B-24, enhanced monitoring options memo). These costs are

significantly higher than those of other available options.

Continuous opacity monitoring systems (COMs) do not provide a

direct measurement of PM emissions but do provide continuous indication

of fabric filter performance. These devices are presently in use on

affected sources and emission units within this source category. Bag

leak detection systems also provide a continuous indication of fabric

filter performance and are less expensive to install and operate than

COMs.

Periodic performance tests by established EPA test methods are

required by the proposed rule. These tests provide important

information about HAP emissions. The expense of conducting performance

tests (see docket item II-B-24, enhanced monitoring options memo)

limits their usefulness as a means of ensuring continuous compliance

with an emission standard.

Another option for compliance assurance is monitoring control

device operating parameters coupled with repeat emission tests prior to

permit renewal (i.e., every 5 years). Control

[[Page 6982]]

device operating parameters can be monitored to ensure continued good

operation and maintenance. Test data and operating experience have

shown that maintaining operating parameters within a specified range of

values (those established based on existing data or performance tests)

can be used to ensure that the control device is operating properly and

is well maintained. Operating parameters and defined work practices

consistent with pollution prevention can also be used to maintain

emissions within limits.

In selecting monitoring requirements to ensure continuous

compliance with the proposed emission standards, the EPA has considered

technical feasibility and cost for all applicable options for each

combination of pollutant, affected source and control technique. In

some cases, where several monitoring options are technically feasible

and equally reliable, and where the operator has already installed a

particular type of monitor, the proposed rule allows the owner or

operator to select a monitoring technique such that a presently

installed, appropriate monitor may continue to be used.

Finally, the proposed rule recognizes that the owner or operator

may, through performance testing under varying conditions, be able to

devise and demonstrate the feasibility of certain monitoring parameters

and procedures. The proposed rule provides a procedure by which site-

specific monitoring plans for certain affected sources and emission

units can be submitted with appropriate documentation for consideration

by the permitting authority. A site-specific monitoring plan, when

approved, would provide alternate monitoring procedures and parameter

levels for secondary aluminum processing units, emission units and

combinations of emission units. Performance testing requirements,

discussed in section IV. E. of this preamble, are proposed to ensure

that each affected source is capable of meeting the applicable emission

standards for HAP or HAP surrogates. Operating requirements are

proposed to ensure that affected sources continuously meet these

emission standards. Monitoring requirements are proposed to ensure that

each owner or operator can demonstrate that the operating requirements

have been met.

1. Operating and Monitoring Requirements and Options for Affected

Sources and Emission Units

Owners or operators of affected sources would be required to submit

an O, M, & M plan as part of their applications for a part 70 or part

71 permit. The plan would include procedures for the proper operation

and maintenance of affected sources and control devices used to comply

with the emission limits as well as the corrective actions to be taken

when control devices or process parameters deviate from allowable

levels established during performance testing. The plan would also

identify the procedures for proper operation and maintenance of

monitoring devices including periodic calibration and verification of

accuracy.

Operating requirements. The proposed rule provides specific

operating requirements for each affected source, and for emission units

within a secondary aluminum processing unit, which are necessary to

ensure that the conditions during initial and periodic performance

tests are not changed between performance tests in such a way as to

increase emissions beyond the proposed standards. Owners or operators

of affected sources are required to operate the affected source and

controls within established parameter ranges. In addition, the proposed

operating requirements incorporate the applicable provisions of the

site-specific O, M, & M plan. These plans include specific corrective

actions to be taken to maintain emissions within acceptable levels.

Operating requirements are also proposed which specify work

practices for group 2 ``clean charge'' furnaces; require labeling of

all affected sources and emission units to facilitate compliance

assurance; specify capture system design and operating parameters for

all affected sources and emission units with add-on control devices;

restrict operation and fluxing practices conducted in group 1 sidewell

furnaces; and establish a means by which site-specific operating plans

for group 1 furnaces without add-on control devices can be developed

and approved.

Monitoring requirements. The EPA is proposing monitoring procedures

for each emission limitation proposed under the rule. The EPA is not

requiring the use of CEMs. PM CEMs have not been demonstrated for use

with affected sources and emission units in this source category. PM

CEMs, as well as HCl CEMs and THC CEMs, are substantially more

expensive than other effective monitoring methods (see docket item II-

B-24, enhanced monitoring options memo).

(a) Scrap Shredder. The proposed monitoring alternatives for scrap

shredders are COMs, bag leak detectors or daily visual emissions

testing by EPA Method 9 of appendix A to 40 CFR part 60. Continuous

opacity monitoring systems (COMs) provide a continuous indication of

fabric filter performance. These devices are presently in use on

affected sources within this source category. Bag leak detection

systems also provide a continuous indication of fabric filter

performance and are less expensive to install and operate than COMs.

Requirements for COMs and bag leak detectors are discussed in section

IV.D.2 of this document, Operating and Monitoring Requirements and

Options for Affected Sources and Emission Units Equipped with Fabric

Filters or Lime Injected Fabric Filters.

Under the visible emission monitoring option, a certified observer

would perform daily visible emissions observations (five 6-minute

readings in a 30-minute period) for each fabric filter according to the

requirements of Method 9 of appendix A to 40 CFR part 60 and the

general provisions in subpart A of 40 CFR part 63. If any visible

emissions were observed, the owner or operator would be required to

initiate corrective actions in accordance with the O, M, & M plan

within 1-hour to correct the cause of the emissions. Visual emissions

monitoring by Method 9 is an appropriate monitoring option for scrap

shredders because these affected sources are intermittently operated

and Method 9 can be used to determine opacity during periods of

operation.

(b) Chip Dryer. Monitoring requirements for chip dryers under the

proposed NESHAP include feed/charge weight monitoring as discussed in

section IV.D.3 of this document, Other Operating Requirements,

Monitoring Systems and Procedures: Feed/Charge Weight, afterburner

temperature monitoring as discussed in section V.D.3 of this document,

Other Operating Requirements, Monitoring Systems and Procedures:

Afterburner Operating Temperature. The identity (i.e. uncoated,

unpainted aluminum chips) of each batch of material charged must be

recorded to ensure compliance with the requirement to process only

uncoated, unpainted aluminum chips.

(c) Scrap Dryer/delacquering kiln/decoating kiln.

Monitoring requirements for scrap dryers/delacquering kilns/

decoating kilns under the proposed NESHAP include feed/charge weight

monitoring as discussed in section IV.D.3 of this document, Other

Operating Requirements, Monitoring Systems and Procedures: Feed/Charge

Weight, afterburner temperature monitoring as discussed in section

IV.D.3 of this document, Other Operating Requirements, Monitoring

Systems and Procedures: Afterburner Operating Temperature, and fabric

filter

[[Page 6983]]

monitoring as discussed in section IV.D.2 of this document, Operating

and Monitoring Requirements and Options for Process Units Equipped with

Fabric Filters or Lime-injected Fabric Filters.

(d) Clean Charge (Group 2) Furnace. Monitoring requirements for

clean charge (group 2) furnaces under the proposed NESHAP are charge

makeup and flux identity recordkeeping, and periodic certification that

only clean charge has been processed and that no reactive flux has been

used. No numerical emission limits are proposed for clean charge

furnaces as discussed in section D.2. of this document, Selection of

MACT Floor Technologies: Group 2 furnaces. Recordkeeping and

certification requirements are necessary to ensure that the affected

sources are operating as clean charge (group 2) furnaces.

(e) Sweat Furnace. The monitoring requirement for sweat furnaces

under the proposed NESHAP is afterburner temperature monitoring as

discussed in section IV.D.3 of this document, Other Operating

Requirements, Monitoring Systems and Procedures: Afterburner Operating

Temperature.

(f) Dross-only Furnace. Monitoring requirements for dross-only

furnaces under the proposed NESHAP include feed/charge recordkeeping as

described in section IV.D.3 of this document, Other Operating

Requirements, Monitoring Systems and Procedures: Feed/Charge Weight,

and fabric filter monitoring, (bag leak detection systems or COMs) as

discussed in section IV.D.2 of this document, Operating and Monitoring

Requirements and Options for Process Units Equipped with Fabric Filters

and Lime-injected Fabric Filters.

(g) In-line Fluxer. Monitoring requirements for in-line fluxers

under the proposed NESHAP include feed/charge weight monitoring as

discussed in section IV.D.3 of this document, Other Operating

Requirements, Monitoring Systems and Procedures: Feed/Charge Weight,

monitoring of chlorine injection rate as described in section IV.D.3 of

this document, Other Operating Requirements, Monitoring Systems and

Procedures: Total reactive chlorine flux injection rate and schedule,

and, for in-line fluxers equipped with add-on control devices, fabric

filter monitoring as discussed in section IV.D.2 of this document,

Operating and Monitoring Requirements and Options for Process Units

Equipped with Fabric Filters and Lime-injected Fabric Filters.

(h) Rotary Dross Cooler. Monitoring requirements for rotary dross

coolers are to comply with one of two monitoring options to demonstrate

continuous compliance with the PM standard. These options (bag leak

detection systems or COMs), and the applicable monitoring requirements,

are discussed in section IV.D.2 of this document, Operating and

Monitoring Requirements and Options for Process Units Equipped with

Fabric Filters and Lime-injected Fabric Filters.

(i) Group 1 Furnace With Add-on Controls. Monitoring requirements

for group 1 furnaces with add-on controls under the proposed NESHAP

include feed/charge weight monitoring as discussed in section IV.D.3 of

this document, Other Operating Requirements, Monitoring Systems and

Procedures: Feed/Charge Weight, monitoring of chlorine injection rate

as described in section IV.D.3 of this document, Other Monitoring

Systems and Procedures: Total reactive chlorine flux injection rate and

schedule, and fabric filter monitoring as discussed in section IV.D.2

of this document, Operating and Monitoring Requirements and Options for

Process Units Equipped with Fabric Filters and Lime-injected Fabric

Filters.

(j) Group 1 Furnace Without Add-on Controls and Using Pollution

Prevention/Work Practices (Processing Only Clean Charge). Monitoring

requirements for group 1 furnaces without add-on controls (processing

only clean charge) and employing pollution prevention/work practices to

limit emissions under the proposed NESHAP include feed/charge weight

monitoring as discussed in section IV.D.3 of this document, Other

Operating Requirements, Monitoring Systems and Procedures: Feed/Charge

Weight, monitoring of chlorine injection rate as described in section

IV.D.3 of this document, Other Operating Requirements, Monitoring

Systems and Procedures: Total reactive chlorine flux injection rate and

schedule and a semi-annual certification that only clean charge had

been processed.

(k) Group 1 Furnace Without Add-on Controls Using Pollution

Prevention/Work Practices Processing Scrap Other Than Clean Charge.

Proposed monitoring requirements for group 1 furnaces not equipped add-

on controls using pollution prevention/work practices and processing

scrap other than clean charge include feed/charge weight monitoring as

discussed in section IV.D.3 of this document, Other Operating

Requirements, Monitoring Systems and Procedures: Feed/Charge Weight and

monitoring of chlorine injection rate as described in section IV.D.3 of

this document, Other Operating Requirements, Monitoring Systems and

Procedures: Total reactive chlorine flux injection rate and schedule.

Operators of these furnaces would be required to develop a site-

specific monitoring plan acceptable to the permitting authority. The

plan would include additional parameters to be monitored, based on

supporting information provided by the operator and developed in

coordination with the permitting authority, which demonstrates the

correlation between these parameters and the actual emissions from

these furnaces.

If the site-specific monitoring plan includes scrap sampling as a

means of monitoring, the scrap sampling program must, at a minimum,

include the elements described in section IV.D.3 of this document,

Other Operating Requirements, Monitoring Systems and Procedures: Scrap

inspection program. If the site-specific monitoring plan includes the

use of CEMs, the operator must install, operate and maintain the CEMs

as described in section IV.D.3 of this document, Other Operating

Requirements, Monitoring Systems and Procedures: Continuous emission

monitoring systems. If the site-specific monitoring plan includes

limitations on the chlorine injection rate, the operator must monitor

reactive flux injection as described in section IV.D.3 of this

document, Other Operating Requirements, Monitoring Systems and

Procedures: Total reactive chlorine flux injection rate and schedule.

The specific parameters monitored under a site-specific monitoring plan

must be proposed by the owner or operator along with supporting

documentation and approved by the permitting authority.

(l) Secondary Aluminum Processing Units. All of the existing group

1 furnaces and all of the existing in-line fluxers within a facility

make up the secondary aluminum processing unit. Each g

This text is long and has been trimmed here. Open the source document for the complete record.