BACKGROUND INFORMATION
FOR PROPOSED NEW SOURCE
PERFORMANCE STANDARDS:
Asphalt Concrete Plants
Petroleum Refineries
Storage Vessels
Secondary Lead Smelters and Refineries
Brass or Bronze Ingot Production Plants
Iron and Steel Plants
Sewage Treatment Plants
Volume 1, MAIN TEXT
ENVIRONMENTAL PROTECTION AGENCY
-------
APTD-1352 a
BACKGROUND INFORMATION
FOR PROPOSED NEW SOURCE
PERFORMANCE STANDARDS:
Asphalt Concrete Plants
Petroleum Refineries
Storage Vessels
Secondary Lead Smelters and Refineries
Brass or Bronze Ingot Production Plants
Iron and Steel Plants
Sewage Treatment Plants
Volume 1, MAIN TEXT
U. S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Water Programs
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
June 1973
-------
The APTD (Air Pollution Technical Data) series of reports is issued by the Office of Air Quality
Planning and Standards, Office of Air and Water Programs, Environmental Protection Agency, to
report technical data of interest to a limited number of readers. Copies of APTD reports are
available free of charge to Federal employees, current contractors and grantees, and nonprofit
organizations — as supplies permit — from the Air Pollution Technical Information Center,
Environmental Protection Agency, Research Triangle Park, North Carolina 27711 or may be
obtained, for a nominal cost, from the National Technical Information Service, 5285 Port Royal
Road, Springfield, Virginia 22151.
Publication No. APTD-1352 a
-------
NATIONAL AIR POLLUTION CONTROL TECHNIQUES
ADVISORY COMMITTEE
Chairman
Mr. Donald F. Walters
Office of Air and Water Programs
Office" of Air Quality Planning and Standards
Environmental Protection Agency
Research Triangle Park, N. C. 27711
Mr. John W. Blantdn, General
Manager
Advanced Technology Programs
Dept.
General Electric Company
Mail Drop E206 - Jimson Road
Evendale, Ohio 45213
Mr. Charles M. Copley, Jr.
Commissioner, Division of Air
Pollution Control
City of St. Louis
Room 419 City Hall
St. Louis, Missouri 63103
Mr. Rafael Cruz-Perez
Amapola #750, Round Hill
Trujillo Alto, Puerto Rico
00926
Mr. Arthur R. Dammkoehler
Air Pollution Control Officer
Pudget Sound Air Pollution Control
Agency
410 W. Harrison Street
Seattle, Washington 98119
Mr. George P. Ferreri
Acting Director
Bureau of Air Quality Control
Maryland State Dept. of Health-
and Mental Hygiene
610 N. Howard Street,
Baltimore, Maryland 21201
Mr. Charles M. Heinen
Executive Engineer
Materials Engineering
Chrysler Corporation
Box 1118, Dept. 5000
Highland Park, Michigan 48231
Mr. William Munroe
Chief^_ Bureau of Air Pollution
Control
New Jersey State Bureau of Air
Pollution Control
Department of Environmental
Protection
P. O. Box 1390
Trenton, New Jersey 08625
Dr. Robert W. Scott
Coordinator for Conservation
Technology
Esso Research & Engineering Co.
P. O. Box 215
Linden, New Jersey 07036
Dr. R. S. Sholtes
Environmental .Engineer
Environmental Engineering Inc.
2324 Southwest 34th Street
Gainesville, Florida 32601
-------
Mr. W. M. Smith
Director, Environmental Control
National Steel Corporation
Box 431, Room 159, General Office
Weirton, West Virginia 26062
Dr. Aaron J. Teller, President
Teller Environmental Systems, Inc.
295 5th Avenue
New York, N. Y. 10016
Mr. A. J. von Frank
Director, Air and Water Pollution
Control
Allied Chemical Corporation
P. O. Box 1057R
Morristown, New Jersey 07960
Mr. Benjamin *'. Wake
Administrator, Environmental
Sciences Division
Montana State Dept. of Health
and Environmental Sciences
Cogswell Building
Helena, Montana 59601
Dr. Ruth F. Weiner
Chairman, Dept. of Physical
Sciences
Florida International University
Tamiami Trail
Miami, Florida 33144
Mr. Raymond L. Wiggins
Manager, Systems Development
Auto-Trol Corporation
Arvada, Colorado 80002
iv
-------
FEDERAL AGENCY LIAISON COMMITTEE
Mr. Donald F. Walters
Office of Air and Water Programs
Office of Air Quality Planning and Standards
Environmental Protection Agency
Research Triangle Park, N. C. 27711
DEPARTMENT OF AGRICULTURE
Dr. Theodore C. Byerly
Assistant Director of Science and
Education
Office of the Secretary
U.S. Department of Agriculture
Washington, D.C. 20250
DEPARTMENT OF COMMERCE
Dr. James R. McNesby
Room A36l, Materials Building
National Bureau of Standards
Washington, D.C. 20234
DEPARTMENT OF TREASURY
Mr. Gerard Brannon
Director^ Office of Tax Analysis
Room 4217 MT
Department of the Treasury
15th and Pennsylvania Ave. , N.W.
Washington, D.C. 20220
FEDERAL POWER COMMISSION
Mr. T.A. Philips
Chief, Bureau of Power
Federal Power Commission
Room 3011
411 G Street, N. W.
Washington, D.C. 20426
GENERAL SERVICES
ADMINIS TRA TION
Mr. Harold J. Pavel
Director^ Repair & Improvement
Division
Public Building Service
General Services Administration
9th and D Street, S. W.
Washington, D.C.
NATIONAL AERONAUTICS AND
SPACE ADMINISTRATION
Mr. Ralph E. Cushman
Special Assistant
Office of Administration
National Aeronautics and Space
Admini s tr ation
NATIONAL SCIENCE FOUNDATION
Dr. O.W. Adams
Program Director for Structural
Chemistry
Division of Mathematical and
Physical Sciences
National Science Foundation
1800 G Street, N.W.
Washington, D.C. 20550
TENNESSEE VALLEY AUTHORITY
Dr. F. E. Gartrell
Director of Environmental Research
and Development
Tennessee Valley Authority
715 Edney Building
Chattanooga, Tennessee 37401
ATOMIC ENERGY COMMISSION
Dr. Martin B. Biles
Director, Division of Operational
Safety
U. S. Atomic Energy Commission
Washington, D.C. 20545
-------
VETERANS ADMINISTRATION
Mr. Gerald M. Hollander, P. E.
Director of Architecture and
Engineering
Office of Construction
Veterans Administration
Room 619 Lafayette Building
811 Vermont Avenue, N. W.
Washington, D. C. 20420
DEPARTMENT OF JUSTICE
Mr. Walter Kiechel, Jr.
Land and Natural Resources Div.
Department of Justice
Room 2139
10th and Constitution Avenue, N. W.
Washington, D. C. 20530
DEPARTMENT OF LABOR
Mr. Robert D. Gidel
Deputy Director, Bureau of Labor
Standards
Department of Labor
Room 401, Railway Labor Building
400 1st Street, N. W.
Washington, D. C. 20210
DEPARTMENT OF DEFENSE
Harvey A. Falk, Jr.
Commander, USN
Office of the Assistant Secretary
of Defense
Washington, D. C. 20301
AIR FORCE ENVIRONMENTAL
QUALITY OFFICE
Colonel Herbert E. Bell
Hq. USAF (PREV)
Room 5E425, Pentagon
Washington, D. C. 20330
DEPARTMENT OF HOUSING AND
URBAN DEVELOPMENT
Mr. Richard H. Brown
Deputy Director
Office of Community and Environ-
mental Standards
Community Planning and Management
Department of Housing and Urban
Development
451 7th Street, S. W.
Washington, D. C. 20410
POSTAL SERVICE
Mr. Robert Powell
Assistant Program Manager
U. S. Postal Service
Room 4419
1100 L Street
Washington, D..G. 20260
DEPARTMENT OF
TRANSPORTATION
Dr. Richard L. Strombotne
Office of the Assistant Secretary
for Systems Development and
Technology
Department of Transportation
400 7th Street, S. W.
Washington, D. C. 20591
DEPARTMENT OF THE INTERIOR
Mr. Harry Moffet
Deputy Assistant Secretary
Minerals and Energy Policy
Department of Interior
Washington, D. C. 20240
DEPARTMENT OF HEALTH,
EDUCATION, AND WELFARE
Dr. Ian A. Mitchell
Special Assistant to the Assistant
Secretary for Health and
Scientific Affairs
Department of Health, Education,
and Welfare
Room 5620N, North HEW Building
330 Independence Avenue, S. W.
Washington, D. C. 20201
VI
-------
TABLE OF CONTENTS
Section Page
LIST OF FIGURES x
LIST OF TABLES xi
ABSTRACT xii
CONVERSION FACTORS, BRITISH TO
METRIC UNITS - xi i
INTRODUCTION 1
Special Note 1
General Considerations 2
Development Procedures 2
Limits in Terms of Concentration 2
Compliance Testing and Instrumentation 3
Use of Alternative Methods 3
Waiver of Compliance Test 4
Comparisons with State and Local Regulations 4
Environmental Impact 5
Economic Impact 5
Provisions for Startup, Shutdown, and
Malfunction 5
Nomenclature 5
Abbreviations 7
Definitions 7
Code Methods 7
Control Equipment 8
TECHNICAL REPORT NO. 6 - ASPHALT
CONCRETE PLANTS 9
Summary of Proposed Standards 9
Emissions from Asphalt Concrete Plants 9
Rationale for Proposed Standards 10
vii
-------
Section Page
Environmental Impact of Proposed Standards 13
Economic Impact of Proposed Standards 13
References for Technical Report No. 6 15
References Cited 15
Supplemental References 15
TECHNICAL REPORT NO. 7 - PETROLEUM
REFINERIES, FLUID CATALYTIC
CRACKING UNITS 17
Summary of Proposed Standards 17
Standards for Particulates 17
Standard for Carbon Monoxide 17
Emissions from Petroleum Refineries 17
Rationale for Proposed Standards 19
Particulate Matter 19
Carbon Monoxide 21
Environmental Impact of Proposed Standards 22
Economic Impact of Proposed Standards 22
References for Technical Report No. 7 23
TECHNICAL REPORT NO. 8 - PETROLEUM
REFINERIES, BURNING OF GASEOUS FUELS 25
Summary of Proposed Standard 25
Emissions from Petroleum Refineries 25
Rationale for Proposed Standards 26
Environmental Impact of Proposed Standard 27
Economic Impact of Proposed Standards 28
References for Technical Report No. 8 28
TECHNICAL REPORT NO. 9 - STORAGE
VESSELS FOR PETROLEUM LIQUIDS 31
Summary of Proposed Standards 31
Hydrocarbon Emissions from Storage Tanks 31
Rationale for Proposed Standards 34
Environmental Impact of Proposed Standards 35
Economic Impact of Proposed Standards 35
References for Technical Report No. 9 36
TECHNICAL REPORT NO. 10 - SECONDARY
LEAD SMELTERS AND REFINERIES ; 37
Summary of Proposed Standards 37
Standards for Particulate Matter from
Blast and Reverberatory Furnaces 37
Standards for Particulate Matter from
Pot Furnaces 37
Emissions from Lead Furnaces 37
viii
-------
Section Page
Rationale for Proposed Standards 40
Particulate Matter from Blast and
t.
Reverberatory Furnaces 40
Particulate Matter from Pot Furnaces 41
Environmental Impact of Proposed Standards 41
Economic Impact of Proposed Standards 41
References for Technical Report No. 10 42
TECHNICAL REPORT NO. 11 - SECONDARY
BRASS OR BRONZE INGOT PRODUCTION PLANTS 45
Summary of Proposed Standards 45
Standards for Particulates from
Reverberatory Furnaces 45
Standard for Particulates from
Electric and Blast Furnaces 45
Emissions from Secondary Brass and
Bronze Furnaces 45
Rationale for Proposed Standards 46
Particulate Matter from Reverberatory Furnaces 46
Particulate Matter from Blast and
Electric Furnaces 47
Environmental Impact of Proposed Standards 47
Economic Impact of Proposed Standards 48
References for Technical Report No. 11 48
TECHNICAL REPORT NO. 12 - IRON AND
STEEL PLANTS 49
Summary of Proposed Standards 49
Emissions from Basic Oxygen Process Furnaces 49
Rationale for Proposed Standards 51
Environmental Impact of Proposed Standards 54
Economic Impact of Proposed Standards 54
References for Technical Report No. 12 55
TECHNICAL REPORT NO. 13 - SEWAGE
TREATMENT PLANTS 57
Summary of Proposed Standards 57
Emissions from Sludge Incinerators 57
Rationale for Proposed Standards 58
Environmental Impact of Proposed Standards 60
Economic Impact of Proposed Standards 60
References for Technical Report No. 13 61
References Cited 61
Supplemented References 61
IX
-------
LIST OF FIGURES
Figure
Page
1 Uncontrolled Hot-mix Asphalt Concrete Plant 9
2 Controlled Hot-mix Asphalt Concrete Plant 10
3 Particulate Emissions from Asphalt Concrete Plants,
Combined Dryer and Scavenger Exhausts 11
4 Petroleum Refinery Fluid Catalytic Cracking Unit
with Control System 18
5 Fluid Catalytic Cracking Unit Regenerator
with Carbon Monoxide Boiler and
Electrostatic Precipitator 18
6 Particulate Emissions from Petroleum Refineries,
Fluid Catalytic Cracking Units 20
7 Carbon Monoxide Emissions from Petroleum Refineries,
Fluid Catalytic Cracking Units 21
8 Petroleum Refinery Process Gas System 26
9 Petroleum Refinery Process Gas Treating Unit 27
10 Fixed-roof (Cone-roof) Storage Vessel 32
11 Single-deck Floating-roof Storage Vessel 32
12 Double-deck Floating-roof Storage Vessel 32
13 Covered Floating-roof Storage Vessel
with Internal Floating Cover 33
14 Seals for Floating-roof Storage Vessels 33
15 Storage Vessel Vapor Recovery System 34
16 Secondary Lead Smelter Process 38
17 Controlled Lead Blast Furnace,
Afterburner and Baghouse 38
18 Controlled Lead Reverberatory Furnace, Baghouse 39
19 Controlled Lead Pot and Ventilation System, Baghouse 39
20 Particulate Emissions from Secondary Lead
Smelters and Refineries, Blast and
Reverberatory Furnaces 40
21 Controlled Secondary Brass and Bronze Furnaces 46
22 Particulate Emissions from Secondary Brass and Bronze
Ingot Production Industry, Reverberatory and
Blast Furnaces 47
23 Iron and Steel Process System 49
24 Controlled Basic Oxygen Furnace, Open Hood with
Scrubber or Electrostatic Precipitator 51
-------
Figure Page
25 Controlled Basic Oxygen Furnace, Closed Hood
with Scrubber 51
26 Particulate Emissions from Iron and Steel Industry,
Basic Oxygen Process Furnaces 52
27 Controlled Multiple-hearth Furnace, Scrubber 58
28 Controlled Fluidized Bed Reactor, Scrubber 58
29 Particulate Emissions from Sewage Treatment Plant,
Sludge Incinerator 59
LIST OF TABLES
Table Page
1 Representative Data from Process Weight Curve 4
2 Summary of Cost Estimates 6
3 Control Costs for Typical Asphalt Concrete Plants 14
4 Control Costs for Catalytic Cracking Units 23
5 Control Costs for Petroleum Storage Tanks 36
6 Control Costs of Meeting Performance Standard for
Typical Secondary Lead Plants 42
7 Control Costs of Meeting Performance Standard for
Reverberatory Furnaces 48
8 Control Costs of Meeting Performance Standard for
Typical New Two-vessel Basic Oxygen Process Furnaces 55
9 Control Cost for Typical Sewage Sludge Incinerator 61
XI
-------
ABSTRACT
This document provides background information on the derivation of the proposed second
group of new source performance standards and their economic impact on the construction and
operation of asphalt concrete plants, petroleum refineries, storage vessels, secondary lead smelters
and refineries, brass or bronze ingot production plants, iron and steel plants, and sewage
treatment plants. Information is also provided on the environmental impact of imposing the
standards on new installations.
The standards developed require control at a level typical of well controlled existing plants and
attainable with existing technology. To determine these levels, extensive on-site investigations were
conducted, and design factors, maintenance practices, available test data, and the character of
stack emissions were considered. Economic analyses of the effects of the proposed standards
indicate that they will not cause undue reductions of profit margins or reductions in growth rates
in the affected industries.
Multiply
barrels
cubic feet
degrees Fahrenheit*
gallons
grains
inches of watert
pounds
pounds per square inch
square feet
tons (short, 2,000 pounds)
long tons (2,240 pounds)
CONVERSION FACTORS
BRITISH TO METRIC UNITS
By
1.59x 10"1
2.83 x 10'2
5/9
3.79 x 10~3
6.48 x 10'5
2.49 x 102
4.54 x 10~1
6.89 x 103
9.29 x 10"2
9.07 x 102
1.02x 103
*To obtain Celsius (centigrade) temperatures
temperatures (tf), use the formula: t = (tx- 32)1.8.
j... ... . .... ' . C T n
To obtain
cubic meters
cubic meters
degrees Celsius (centigrade)
cubic meters
kilograms
newtons per square meter
kilograms
newtons per square meter
square meters
kilograms
kilograms
from Fahrenheit
tMultiply millimeters of mercury by 1.33 x 102 to obtain newtons per
square meter.
xii
-------
BACKGROUND INFORMATION
FOR PROPOSED
NEW SOURCE PERFORMANCE STANDARDS
INTRODUCTION
This document provides background information on the derivation of the proposed second
group of new source performance standards and their economic impact on the construction and
operation of asphalt concrete plants, petroleum refineries, storage vessels, secondary lead smelters,
and bronze ingot production plants, iron and steel plants, and sewage treatment plants. The
regulation for the proposed standards, published in the Federal Register under Title 40 CFR Part
60, is being distributed concurrently with this document. The information presented herein was
prepared for the purpose of facilitating review and comment by owners and operators of affected
facilities, environmentalists, and other concerned parties prioi to promulgation of the standards.
Information concerning the source categories is provided in Technical Reports 6 through 13. In
the case of petroleum refineries, there are reports covering two affected facilities—catalyst
regenerators and gaseous fuel burning. Technical Reports 1 through 5 were published in 1971 with
the first group of new source performance standards.
The performance standards were developed after consultation with plant owners and operators,
appropriate advisory committees, trade associations, equipment designers, independent experts,
and Federal departments and agencies. Review meetings were held with the Federal Agency
Liaison Committee and the National Air Pollution Control Techniques Advisory Committee. The
proposed standards reflect consideration of comments provided by these committees and by other
individuals having knowledge regarding the control of pollution from the subject source categories.
The National Air Pollution Control Techniques Advisory Committee consists of 16 persons who
are knowledgeable concerning air quality, air pollution sources, and technology for the control of
air pollutants. The membership includes State and local control officials, industrial
representatives, and engineering consultants. Members are appointed by the Administrator of the
Environmental Protection Agency (EPA) pursuant to Section 117(d), (e), and (f) of the Clean Air
Act Amendments of 1970, Public Law 91-604. In addition, persons with expertise in the respective
source categories participated in the meeting of the Advisory Committee.
The Federal Agency Liaison Committee includes persons with knowledge of air pollution control
practices as they affect Federal facilities and the nation's commerce. The committee is composed
of representatives of 19 Federal agencies.
The promulgation of standards of performance for new stationary sources under Section 111 of
the Clean Air Act does not prevent State or local jurisdictions from adopting more stringent
emission limitations for these same sources. In heavily polluted areas, more restrictive standards,
including a complete ban on construction, may be necessary in order to achieve National Ambient
Air Quality Standards. Section 116 of the Act provides specific authorization to States and other
political subdivisions to enact such standards and limitations.
SPECIAL NOTE
Subsequent to the development of this document, the Environmental Protection Agency
adopted a-policy of expressing standards in the metric rather than the English system. Consistent
with the proposed standards, emission limits are listed herein in metric units, but English
equivalents are also provided. Due to the complexities involved in recalculating test results, data in
this document have not been converted from the English to the metric system. A table of conversion
1
-------
factors is presented in the preliminary pages, however. To allow comparison with test data, the
standards are frequently referenced in terms of English units.
GENERAL CONSIDERATIONS
The proposed second group of new source standards includes emission limits for particulates
(including visible emissions), as well as sulfur dioxide, carbon monoxide, and hydrocarbons. In
addition, revisions have been incorporated into the General Provisions that were published with
the first group of standards under Title 40 CFR 60. Methods for determining compliance with
particulate and sulfur dioxide limits are published in the Appendix of 40 CFR 60. Methods for
measuring carbon monoxide and hydrogen sulfide are^published with the proposed standards.
Development Procedures
The procedures used to develop the standards were similar for all source categories. In every case,
a screening process was followed to appraise existing technology and to determine the locations of
well controlled sources. Extensive on-site investigations were conducted to identify sources that
appeared to be the best controlled and amenable to stack testing. Design features, maintenance
practices, available test data, and the character of stack emissions were considered. Where
particulate emissions were contemplated, appreciable weight was given to the opacity of stack
gases. In most instances, the facilities chosen for testing were those that exhibited little or not
visible emissions and had a sufficient length of straight ductwork downstream of the collector to
obtain representative source samples.
Observations of stack gases during the screening process and during stack tests furnished the
basis for the proposed visible emission limits. For most of the six particulate standards, several
sites that met the proposed visible emission limits were identified. Mass emissions from many of
them could not be measured because the stack configurations precluded accurate testing. Those
sources that met the proposed mass particulate limits also met the visible emission limits. Thus,
visible emissions in excess of the proposed limits indicate that the mass particulate standards are
also being exceeded.
Condensed water vapor is not considered a visible emission for purposes for this regulation.
Where the presence of uncombined water is the only reason for failure to meet the standards, such
emissions shall not be considered a violation.
This volume contains sufficient data from the tests conducted to support the discussions.
Detailed test data are given in Volume 2 of this document, which was prepared in a limited edition
but is available to those who need the data. The second volume can be obtained from the Emission
Standards and Engineering Division, Environmental Protection Agency, Research Triangle Park,
North Carolina 27711, Attention: Mr. Don R. Goodwin.
Limits in Terms of Concentration
Most of the emission limits included in this group of standards are being proposed in terms of
pollutant concentration. Particulate limits, for example, are being proposed in terms of milligrams
per normal cubic meter of undiluted exhaust gases. This is a deviation from the first group of
performance standards, wherein most of the limits were promulgated in terms of mass per unit of
production, feedstock, or fuel input. The change to concentration units is a result of discussions
with control officials, representatives of affected industries, and others knowledgeable in the field.
Its purpose is to facilitate compliance testing and enforcement of new source performance
standards. Establishing standards in this form obviates the need to determine such things as
production rates and burning rates, which often cannot be ascertained with the same degree of
accuracy as can the pollutant concentration. In some future standards, a pollutant concentration
limit may not be feasible, and other types of standards may be used.
In proposing concentration limits, it is implicit that compliance cannot be achieved by merely
diluting exhaust gases with ambient air. Emission limits are to be achieved through the application
-------
of process changes or remedial equipment that will limit the discharge of pollutants to the
atmosphere. The concentration limits proposed in these regulations will apply to exhaust gas
streams as they are discharged from control equipment. If there is any dilution prior to
measurement, suitable corrections will be made in determining compliance. Provisions have been
incorporated in each standard that preclude dilution as a means of achieving the standard.
The provisions regarding circumvention by dilution, for example, 60.94(c), apply equally to mass
limits and visible emission limits. Where dilution gases are added downstream of air pollution
control devices, owners or operators will be required to demonstrate that the visible emissions
would not constitute a violation of the standard if they were not diluted.
Compliance Testing and Instrumentation
As with the first group of new source performance standards, particulate limits in the proposed
regulation are based on material collected in the probe and filter of the EPA sampling train (see
"Test Methods" section). Impingers, as described in the original proposal for Group 1 source
categories (40 CFR 60), may be utilized; however, the material collected in the impingers is not
considered particulate for purposes of the proposed regulations.
Emissions of hydrocarbons from storage vessels for petroleum liquids will not be measured
directly. This standard is established in terms of emission limitations that can be accomplished
with readily available and standardized control equipment, i.e., floating-roof tanks, vapor recovery
systems, and conservation vents. The standard specifies that these devices or any other device
equally effective for hydrocarbon control shall be utilized. The actual emissions from any specific
storage vessel can be determined by utilizing suitable empirical relationships developed by the
industry.
While the limits for refinery fuel gases are designed to prevent the release of sulfur dioxide, it is
expected that, in essentially all cases, compliance will be determined by analyzing the hydrogen
sulfide content of the fuel gases before they are burned.
The carbon monoxide measurement technique is based on an instrumental method of analyses
of exhaust gases. Instruments specified in the proposed regulations or instruments of essentially
the same type may be utilized to satisfy this requirement. Owners and operators of petroleum
catalyst regenerators may monitor either carbon monoxide or two other significant parameters,
oxygen content and temperature. If they can show by monitoring that there is sufficient oxygen in
the gas stream to provide the necessary degree of carbon monoxide combustion at the firebox
temperature, carbon monoxide monitoring will not be required.
In addition to instruments for the measurement of carbon monoxide and the sulfur content of
fuel, instruments will be required, where feasible, to measure emissions directly or indirectly.
Instruments for recording visible emissions will be required for two source categories for which
particulate limits are proposed.
Use of Alternative Test Methods
A provision has been added whereby the Administrator may accept performance tests conducted
with alternative methods that are not entirely equivalent to the reference method but are suf-
ficiently reliable that they may be used for certain applications. For example, an alternative test
method^ that does not require traversing during sampling for particulate matter may be approved if
such method includes a suitable correction factor designed to account for the error that may result
from failing to traverse, or if it can be demonstrated in a specific case that failure to traverse does
not affect the accuracy of the test. Similarly, use of an in-stack filter for particulate sampling may
be approvable as an alternative method if the method otherwise employs provisions designed to
result in precision similar to the compliance method, and a suitable correction factor is included to
account for variation between results expected due to filter location. In cases where determination
of compliance using an alternative method is disputed, use of the reference method or its
equivalent shall be required by the Administrator.
-------
Waiver of Compliance Test
A provision has been added whereby the Administrator may waive the requirement for
compliance testing if the owner or operator provides other evidence that the facility is being
operated in compliance with the standard. Evidence of compliance may be in the form of: tests of
similar installations and measurement of significant design and operating parameters; observa-
tions of visible emissions; evaluation of fuels, raw materials, and products; and other equally
pertinent information. The Administrator will reserve the authority to require testing of facilities
at such intervals as he deems appropriate under Section 114 of the Act.
Comparisons with State and Local Regulations
In this background document, the proposed new source performance standards frequently are
compared to existing State and local regulations. Process weight regulations are commonly
employed by many State and local jurisdictions to limit particulate emissions from a variety of
industrial sources. In this type of regulation, allowable particulate release is based on the size of
the source. The limit, however, varies from state to state. Consequently, a reference process weight
curve is used for comparison purposes. The reference curve was published as part of an EPA
regulation on the preparatipn of State implementation plans (40 CFT 51); its limitations are given
in Table 1.
Table 1. REPRESENTATIVE DATA FROM PROCESS WEIGHT CURVE
Process weight rate,
Ib/hr
50
100
500
1,000
5,000
10,000
20,000
60,000
80,000
120,000
160,000
200,000
400,000
1,000,000
Allowable
emission rate,
Ib/hr
0.03
0.55
1.53
2.25
6.34
9.73
14.99
29.60
31.19
33.28
34.85
36.11
40.35
46.72
Emissions, E, for process weights up to 60,000 Ib/hr not corresponding to the points given in
Table 1 can be interpolated by the equation: •
E= 3.59 P°-62 (1)
where:
E= emissions, Ib/hr
P =? process weight, Ib/hr
For process weights above 60,000 Ib/hr, interpolation and extrapolation are based on the equation:
E= 17.31 P°-16 (2)
-------
Environmental Impact
All of the proposed standards have the effect of reducing emissions of air pollutants to the
atmosphere. They may also cause an increase in the generation of solid wastes and in some
instances produce liquid wastes.
Six of the standards require control of particulate matter that thereby becomes a potential solid
waste. Nonetheless, it is significant that all six source categories are required by existing State and
local regulations to control particulates to some degree. The effect of the proposed standards is to
require the installation of higher efficiency dust collectors and thus to increase the quantity of
collected solids. In no case is a new type of solid waste created. Some of these collected particu-
lates, e.g., those from secondary lead furnaces and many asphalt concrete plants, can be recycled
back to the system. In others, such as steel furnaces and sludge incinerators, the material must be
disposed of, usually in landfills. None of the materials collected from these facilities are of such
nature that they cannot be successfully handled by landfill.
It is expected that most of the devices installed to meet the proposed standards will collect the
material in the dry state. Dry collection is advantageous because (1) it greatly reduces the pos-
sibility of water pollution and (2) the collected material is more likely to be acceptable for recycle to
the process. Dry dust collectors are feasible with all six source categories, but scrubbers are more
likely to be utilized for basic oxygen process steel furnaces (BOPF) and sewage sludge incinerators.
In addition, some owners and operators of asphalt concrete plants and secondary lead smelters
may choose to utilize wet scrubbing systems rather than dry dust collectors. Since wet scrubbers
have been used extensively in the steel industry and for asphalt concrete plants and sewage sludge
incinerators, techniques are available for recycle of water and for acceptable disposal of solid
wastes. The proposed standards will not require the use of any solid waste or water treatment
practices that are not already utilized to a wide degree. It may increase the complexity and cost of
liquid and solids handling because of the greater quantities of particulate collected.
The proposed standards also require the collection of hydrocarbons and sulfur compounds.
There are no potential adverse effects of the hydrocarbon storage regulation since all hydrocarbons
are retained as product or recycled to petroleum refineries.
Sulfur compounds are recovered as salable by-products, usually elemental sulfur or sulfuric
acid. The most common process generates a liquid waste for which acceptable disposal methods
are available. The process has been in use for many years in the petroleum and natural gas
industry.
Economic Impact
For each of the designated source categories, information is provided on the expected economic
impact of the standard on the industry. Capital and annualized costs (including operating costs)
have been estimated. In addition, the incremental costs of air pollution control on the typical
product have been determined. A summary of pertinent cost items for typical affected source
categories is provided in Table 2.
Provisions for Startup, Shutdown, and Malfunction
Independent of this proposal, the Administrator published on May 2, 1973, a proposed amend-
ment to 40 CFR 60, Subpart A—General Provisions, whereby consideration will be given to condi-
tions that may cause emissions to exceed new source standards during startup, shutdown, and
malfunction. The new provisions are tentative pending a review of the comments and
promulgation of the resulting provision.
NOMENCLATURE
The following lists of abbreviations, definitions, test methods, and control equipment should
help clarify the terms used in the background document text and graphs.
-------
Table 2. SUMMARY OF COST ESTIMATES
Proposed standard
Industry
Asphalt
concrete
plants
Petroleum
refineries
Hydrocarbon
storage
vessels
Secondary
lead
Brass and
bronze
Iron and
steel
Sewage
treatment
Affected
facility
Entire
facility
FCC
catalyst
regenerator
Units
burning
process
gas
Storage
tanks
Furnace
emissions
Furnace
emissions
Basic
oxygen
furnace
Sludge
incinerator
Performance
standard
70 mg/Nm3
Iparticulates)
50 mg/Nm3
(particulates)
0.050 volume %b
(carbon
monoxide)
230 mg/Nm3
of fuel gas
(hydrogen
sulfide)
Require a
floating-
roof tankd
50 mg/Nm3
(particulates )
50 mg/Nm3
(particulates)
50 mg/Nm3
(particulates)
70 mg/Nm3
(particulates)
Basis for cost analysis
Typical
facility
size
150 tons/hr
300 tons/hr
20,000 bbl/day
65,000 bbl/day
80,000 bbl
50- ton/day
reverberatory
furnace
50- ton/day
blast
furnace
50 ton/day
140 tons/melt
250 tons/melt
10 ton,/day
Control
equipment
Fabric filter
or venturi
scrubber
Fabric filter
or venturi
scrubber
Precipitator
Precipitator
'
Floating-roof
tank
Fabric filter
or venturi
scrubber
Fabric filter
or venturi
scrubber plus
afterburner
Fabric filter
Open-hood
scrubbing
Precipitator
Closed-hood
scrubbing
Open-hood
scrubbing
Precipitator
Closed-hood
scrubbing
Venturi
scrubber (low
energy)
Estimated cost
Investment
cost, $
63,000
56,000
92,000
95,000
700,000
1,150,000
27,000
(incremental)
over a fixed
roof)
188,100
125,200
156,600
123,200
110,000
5,720,000
5,880,000
6,760,000
7,400,000
8,000,000
8,400,000
60,000
Annual
cost, S/yr
18,000
21,000
26,000
36,000
150,000
225,000
3,800
50,600
35,600
50,600
79,700
20,070
1,946,000
1,492,000
2,139,000
2,139,000
2,025,000
2,791,000
1 1 ,700
Impact3
$0.16/tonof
product
S0.19/tonof
product
S0.12/ton of
product
$0. 16/tonof
product
$0.022/bbl of
fresh feed
SO.OIO/bbl of
fresh feed
Gasoline-
IS 11,1 00/yr)e
Jet naptha-
$1,000/yr
Crude oil-
($5,200/yr)
Sl.65/tonof
product
$2.85/ton of
product
$4.05/ton of
product
$6.38/ton of
product
$4.01 /ton of
product
$1.1 7 to $1.67/
ton of steel
$0.89 to $1.227
ton of steel
$0. 1 2/person/yr
Estimated product prices: (1) asphalt concrete-$6/ton, (2) brass and bronze-$1100 to $1200/ton, (3) iron and steel-$220/ton (price of
finished steel products for a typical mill product mix), (4) secondary lead-$320/ton.
Carbon monoxide boilers have an attractive economic payout, and, as a result, most new units would be built with such boilers even
without the proposed standards.
It is commonly accepted and necessary practice to treat the various refinery gas and liquid streams for product quality control.
Consequently, there is a 2 to 5 percent increase in investment cost but no discernable difference in operating costs between current
industry practice and the requirements for new source standards.
Floating-roof tanks are required for storage of liquids with vapor pressures between 1.52 and 11.1 psia. Storage of liquids with vapor
pressures above 11.1 psia requires use of recovery or equivalent.
eFigures shown are net costs and include a credit for recovered materials. Figures in parenthesis indicate a savings.
-------
Abbreviations
Definitions
acf
acfm
bbl
dscf
dscfm
°F
ft2
gal
g/Nm3
gr
hr
Ib
min
ou
ppm
psia
scf
scfm
Front half
Back half
Total EPA train
Average
actual cubic feet; volume of gas at stack conditions
actual cubic feet per minute
barrels
dry standard cubic feet
dry standard cubic feet per minute
degrees Fahrenheit
square feet
gallons
grams per normal cubic meter
grains
hours
pounds
minutes
odor units
parts per million by volume
pounds per square inch absolute
standard cubic feet
standard cubic feet per minute
Material captured in probe and filter of EPA train (see test method 2).
Also called "dry filterable particulate."
Material captured in the impinger portion of the EPA train. Also called
"condensables."
Front half plus back half catch (see test method 1).
Arithmetic average of the individual runs.
Code Methods
The following code methods are referred to by number in the technical reports:
1. EPA train with impingers—Isokinetic sampling and traversing of the stack, with analysis of
the probe washings, filter catch, impinger washings, organic extraction, and impinger water.
2. EPA method 5 (as described in the December 23, 1971, Federal Register)—Isokinetic sampling
and traversing of the stack; analysis includes only probe washings and filter catch (also called
"front-half catch," "solids," or "dry filterable particulates").
3. Same as code method 1 except that sampling is conducted at a point of average velocity.
4. Same as code method 2 except that sampling is conducted at a point of average velocity.
5. Isokinetic sampling at point of average velocity with impingers (two containing distilled water,
one dry) followed by Whatman* paper thimble. Gas-meter upstream of pump. Result includes
material collected on the filter and in the impingers (soluble and insoluble) except sulfuric acid
bihydrate.
6. Alundum thimble packed with glass wool followed by a Gelman type A filter. Both thimble and
filter inside stack during test.
7. San Francisco Bay Area Air Pollution Control District Regulation 2 method—Particulate
collected by glass tubes filled with wool located in stack. Gas velocity predetermined by
separate pitot tube and assumed constant throughout test. Samples collected at two to three of
the points of measured velocity during each test.
*Mention of commercial products or company names does not constitute endorsement by the
Environmental Protection Agency.
-------
8. EPA equipment, including impingers, is used, but probe and impinger acetone washings are
combined. Results include washings and filter catch and are therefore higher than those of
code method 2 (filter catch and probe washings only).
9. Adjusted EPA train with impinger results—Data obtained using code method 1 was adjusted
by multiplying it by the average value of the ratio of code method 2 to code method 1 for two
secondary lead blast furnaces.
10. Alundum thimble in stack, packed with glass wool and followed by impingers. Impinger liquid
is filtered and filtrate is included as particulate. Probe is washed and material in washings is
included as particulate.
11. Nondirect infrared test for carbon monoxide—Will appear in the Federal Register as Method
10—Determination of Carbon Monoxide Emissions from Stationary Sources.
12. Cadmium salt test for hydrogen sulfide—Will appear in the Federal Register as Method 11—•
Determination of Hydrogen Sulfide Emissions from Stationary Sources.
13. Samples evacuated by air ejector through an in-stack alundum thimble and four impingers
(two containing distilled water). Result consists of material from filter and soluble and
insoluble material collected in impingers.
14. Samples using impingers followed by a Gelman type A glass fiber filter. Result includes filter
and impinger catches.
Control Equipment
Listed below are symbols used in the background document for various types of control devices.
If more than one are used, the order of the letters indicates the arrangement of the control devices,
starting with the one farthest upstream.
s — scrubber
v — venturi scrubber
b — baghouse
e — electrostatic precipitator
a — afterburner
h — open hood
g — closed hood
c — cyclone
m — carbon monoxide boiler
p — plate scrubber
-------
TECHNICAL REPORT NO. 6 -
ASPHALT CONCRETE PLANTS
SUMMARY OF PROPOSED STANDARDS
Standards of performance are being proposed for new hot mix asphalt concrete production
plants. The proposed standards would limit emissions of particulates (including visible emissions)
from the following sources: dryer; hot aggregate elevators; screening (classifying) equipment; hot
aggregate storage equipment; hot aggregate weighing equipment; asphalt concrete mixing
equipment; mineral filler loading, transfer, and storage equipment; and loading, transfer, and
storage equipment that handles the dust collected by the emission control system.
The standards apply at the point(s) where undiluted gases are discharged from the air pollution
control system or from the affected facility if no air pollution control system is utilized. If air or
other dilution gases are added prior to the measurement point(s), the owner or operator must
provide a means of accurately determining the amount of dilution and correcting the pollutant
concentration to the undiluted basis.
The proposed standards would limit particulate emissions to the atmosphere as follows:
1. No more than 70 mg/Nm3' (undiluted) or 0.03 gr/dscf.
2. No more than 10 percent opacity.
EMISSIONS FROM ASPHALT CONCRETE PLANTS
The asphalt concrete industry has been generally recognized as a major source of particulate
emissions. Poorly controlled asphalt concrete plants (Figure 1) can release 10 to 15 pounds of
WEIGH HOPPER
MIXER
MINERAL
FILLER
FINES
STORAGE
PRIMARY DUST COLLECTORx
ASPHALT STORAGE )
TANK
Figure 1. Uncontrolled hot-mix asphalt concrete plant.
9
-------
particulates to the atmosphere per ton of asphalt concrete product. A 200-ton/hr installation,
equipped with only dry centrifugal dust collectors, would emit 2000 to 3000 pounds of particulate
each hour of operation. Because of the large number of plants (approximately 4800), their
collective emissions constitute a significant portion of the total particulate from all industries. EPA
has conservatively estimated that total particulate emissions from this industry were 243,000 tons
in 1967 and would increase to 403,000 tons in 1977 if the '1967 control level of 95 percent were
maintained.2 According to A.E. Vandergrift, et aL, the asphalt industry is the eleventh largest
source of particulate emissions in the nation.
In order to reduce emissions by about 99.7 percent, as required by the proposed standard, fabric
filters or medium energy venturi scrubbers, normally preceded by a cyclone or multiple cyclone,
are used to collect dust from the dryer (Figure 2). Fugitive dust from the hot aggregate conveyor,
EMISSIONS
PRIMARY DUST COLLECTOR
FAN
ASPHALT STORAGE^
TANK
Figure 2. Controlled hot-mix asphalt concrete plant.
screening, mixing, and other process equipment is normally controlled by enclosing these sources
and ducting the dust-laden gases to the dust collection system. The collected dust is normally
recycled to the plant, thereby increasing product yield.
Most S-tate and local regulations limit particulate emissions from asphalt concrete plants, either
on the basis of stack gas concentration or through process weight regulations. The most stringent
State regulation, 0.03 gr/dscf (based on samples collected with both the filter and impingers),
would permit the typical 200-ton/hr plant to emit 5.1 Ib/hr of particulate. The reference process
weight regulation (Table 1) would restrict emissions from this typical asphalt plant to 40 Ib/hr,
which is approximately 0.23 gr/dscf.
RATIONALE FOR PROPOSED STANDARDS
The proposed standard of 70 mg/Nm3 (0.031 dscf) is based on 11 tests of plant emissions
performed by EPA on four asphalt concrete plants. Three of these plants were controlled with
fabric filters, which ranged from 1 to 4 years in age, and one was controlled by a venturi scrubber.
Other data that support the level of the standard were obtained from tests conducted by State and
local agencies and the National Asphalt Pavement Association. The size of plants tested ranged
from 80 to 350 tons/hr. Preliminary investigations by EPA revealed the location of several
reportedly well controlled plants. Si'xty-four were visited, visible emissions were evaluated, and
information was obtained on the process and control equipment. Fifty-two were determined
10
-------
unsatisfactory because of inadequate maintenance or design, often evidenced by excessive visible
emissions, or because the equipment was not suitable for testing (e.g., a pressure-type baghouse
without a stack). Eight of the remaining 12 plants were eliminated because of planned shutdowns
for the winter season. Stack tests were conducted at four locations.
During the initial plant surveys, 12 plants with fabric filter control equipment exhibited no
visible stack emissions other than uncombined water vapor. Nine of these plants were not tested for
reasons listed above.
Results of the four tests (three samples per test) conducted by EPA (Figure 3) reveal that
emissions from the three plants with fabric filter controls (Plants A, B, and D) averaged 0.007,
.10
-t
CO
? -
DEME1
EPAM
FHOD Nl
FTHOn
JMBER
5
T
t
0.05
So
0.04
0.03
CJ
g
a.
0.02
0.01
2
e
3 - EPA METHOD 5 PLUS IMPINGERS;
SAMPLED AT POINT OF
AVERAGE VELOCITY
5 - IMPINGERS PLUS FILTER
—i MAXIMUM
AVERAGE
EPA
OTHER
b-BAGHOUSE
v-VENTURI SCRUBBER
B
Al(b) A2(b) B(b) C(v) D(b) E(b) F(b) G(v) HI(V) H2(v)
PLANT (CONTROL EQUIPMENT)
J(b) K(b) L(b) M(b)
Figure 3. Paniculate emissions from asphalt concrete plants, combined dryer and scavenger
exhausts.
11
-------
0.008, and 0.018 gr/dscf. Individual sample results ranged from 0.005 to 0.024 gr/dscf. The plant
controlled by a venturi scrubber (Plant Hj) emitted 0.031 gr/dscf, with individual tests ranging
from 0.029 to 0.034 gr/dscf. The same plant while operating at a higher scrubber energy con-
sumption, was retested by a State agency (Plant H2), and emissions were 0.011 gr/dscf.
Included in Figure 3 are State and local agency test data, which also support the proposed
standard. Three of the plants tested were controlled by venturi scrubbers (Plants C, G, and I) and
three by fabric filters (Plants E, F, and J). Measured emission rates from the three scrubber
installations ranged from 0.012 to 0.025 gr/dscf and averaged 0.017, 0.015, and 0.025 gr/dscf.
Emissions from the three baghouse installations averaged 0.006, 0.010, and 0.019 gr/dscf with
individual tests ranging from 0.005 to 0.023 gr/dscf. Four of these tests were performed in
accordance with EPA test procedures. The others, although performed with the basic EPA train,
incorporated minor modifications.
A manufacturer of control equipment measured emissions for a prototype baghouse collector
installed at an actual asphalt plant for continuing pilot tests. Results of these 0.012 gr/dscf tests,
are well below the standard.
Additional support for the standard has been provided by the National Asphalt Pavement
Association from tests of four asphalt concrete plants (Plants A2 , K, L, and M). The four plants,
which used fabric filter collectors, had average emissions of 0.007, 0.012, 0.043, and 0.108 gr/dscf.
The two latter values are not considered representative of good maintenance and operation. In
both cases, the dust collectors were inspected by an EPA engineer or by the manufacturer prior to
the test.4 Evidence of poor collector maintenance or operation made the efficiency of the control
equipment suspect.5
Letters on file from three manufacturers of fabric filter collectors and one manufacturer of
venturi scrubbers guarantee emission levels to meet a standard of 0.030 gr/dscf if equipment is
properly installed, operated, and maintained. At least two other manufacturers (one of fabric
filters and one of venturi scrubbers) are guaranteeing an emission level to meet 0.030-gr/dscf State
and local codes.
Two of the three fabric filter installations tested by EPA had recently substituted fuel oil for the
natural gas normally fired in the dryer burner (Plants A and B). The third was burning natural gas.
The replacement of natural gas with fuel oil has been reported to increase particulate mass
emission by 20 to 30 percent.1 Consequently, the emissions measured for the plants using oil for
fuel would probably nave been smaller if the tests had been conducted before the change in fuels.
All EPA tests were performed in the fall, near the end of the asphalt production season, when
the plant is most likely to be in poor repair. The winter months are utilized for maintenance. Thus,
the control devices were tested immediately prior to the annual maintenance cycle. Of the three
fabric filter collectors tested, two had been in service for one season and one for four seasons
without changing the bags. Obviously the control devices were not operating under optimum
conditions; i.e., the filters were not new.
A factor that can affect control equipment performance is the particle size of dust released from
these systems. Since asphalt concrete plants are installed throughout the nation, a wide variety of
aggregate is processed in dryers. In developing the standard, it was necessary to determine the
characteristics of these aggregates and to ascertain that available dust collectors could meet the
proposed emission limits. Particle size has a significantly greater effect on the performance of
high-energy scrubbers than on fabric filters. Particulate emissions from high-energy scrubbers
tend to increase with decreasing particle size.* •5 '9 Where there are large fractions of fines,
scrubbers may require greater energy input. On the other hand, the performance of fabric filter
collectors is relatively unaffected by the size distribution of particulates, such that emission levels
rrom baghouses are nearly the same over a wide variety of aggregate feedstocks.8 -13 This is further
evidenced by the test report of a laboratory study sponsored by the National Asphalt Pavement
Association, in which it is concluded that there is no correlation between particle size and the
capture efficiency of a fabric filter.5
12
-------
There is no evidence that rapid changes in the amount of fine material and transient conditions
during startup and shutdown increase emissions from a fabric filter collector and preclude plants
from achieving the proposed standard. The National Asphalt Pavement Association-supported
laboratory test,5 which did not duplicate actual operating conditions, was partly devoted to
studying the effect on collection efficiency of sudden changes in the airflow through a filter without
appreciable cake. It was found that when asphalt concrete aggregate was used as the test particu-
late, exit concentration varied only from 0.00054 to 0.00012 gr/dscf, a factor of only 4.5:1. If, in
fact, transient conditions during startup and shutdown did affect fabric filter collection efficiency,
such an effect would not preclude plants from achieving the proposed standard. Performance tests
do not begin until the effluent gas temperature stabilizes after plant startup, and tests are
terminated at plant shutdown. Furthermore, Section 60.8 of 40 CFR Part 60, which specifies that
performance tests be conducted during periods of representative performance and consist of three
repetitions of the applicable test method, precludes the possibility that performance tests would be
unduly influenced by routine shutdowns and startups.
The fines content of the process aggregates is reflected in the fraction of -200-mesh material (less
than 74 micrometers). Investigations indicate that 3 to 5 percent of -200-mesh material is typical
for aggregates utilized in asphalt concrete plants.8-14 To assure that EPA tests were
representative, each plant operator was requested to schedule production of a product mix
containing a large fraction of -200-mesh materials. During the four tests conducted by EPA, the
actual fines content of the aggregate ranged from 2 to 7 percent by weight.
The proposed standard of 0.031 gr/dscf is supported by measured emissions from 13 of the 15
source tests presented in Figure 3. Results from 2 of these 15 source tests cannot be considered
representative, since evidence of poor collection, maintenance, and operation made efficiency of
the control equipment suspect. The standard will require installation and proper maintenance of
equipment representative of the best technology demonstrated (considering cost) for the industry.
ENVIRONMENTAL IMPACT OF PROPOSED STANDARDS
Potential adverse environmental effects from implementing the proposed standard include
disposal of collected materials and handling of scrubber liquor to prevent water pollution. At a
typical 200-ton/hr plant from 1 to 1.5 tons/hr of particulate will be collected in either dry or wet
dust collectors. When fabric filters are used, the material is collected dry and can be recycled or
disposed of in that form. In many plants, the collected material is recycled to produce asphalt
concrete. Settling ponds are used in conjunction with scrubbers to separate entrained solids. Water
is recycled, usually in a closed loop, and collected solids are removed from the pond as necessary.
These settled solids are essentially rock and sand and can be safely landfilled. If high-sulfur fuel is
used to fire the aggregate dryer, the scrubbing water will eventually become acidic and require
neutralization to prevent leaching and equipment corrosion. The small quantity of soluble salts
that will be produced as a result of the neutralization should not present a problem. Washing
techniques are available to minimize soluble salt carryover in collected solids.
ECONOMIC IMPACT OF PROPOSED STANDARDS
The production of asphalt concrete has increased at an annual rate of 10 percent over the past
several years. Although growth has been cyclical, it is expected that this average rate will persist
through the near future. To meet increase demand, it is anticipated that 200 new plants will be
constructed each year. In addition, the industry estimates that some 50 obsolete plants will be re-
placed annually. Approximately 250 new plants each year are estimated to be subject to a new
source performance standard.
For a new asphalt concrete plant rated at 150 tons/hr (average on-stream time of 50 percent,
annual production of 112,500 tons) and also for a plant rated at 300 tons/hr (average on-stream
time of 50 percent, annual production of 225,000 tons), three abatement alternatives were
analyzed. Table 3 summarizes the results of these analyses. The objective of the analyses was to
compare the cost effects of two standards: the reference process weight standard and the proposed
13
-------
Table 3. CONTROL COSTS FOR TYPICAL ASPHALT CONCRETE PLANTS3
Plant size.
tons/hr
(acfm)
150
(25,000)
300
(50,000)
Emission
standard
Proposed
performance
standard=
0.031 gr/dscf
Reference
process weight
stand ard=
0.30 gr/dscf
Proposed
performance
standard=
0.031 gr/dscf
Reference
process weight
standard=
0.1 8 gr/dscf
Required
controj
equipment
Fabric filter
Venturi scrubber
Low-energy
scrubber
Fabric filter
Venturi scrubber
Low-energy
scrubber
Control
investment.
$
63,000
56,000
44,000
92,000
95,000
75,000
Annual
cost.
$/yr
18,000
21,000
16,000
26,000
36,000
27,000
Annual
cost per
unit of
production.
$/ton
0.16
0.19
0.14
0.12
0.16
0.12
aModel plant assumptions: (1) 1500 hours on-stream annually, (2) production averages 50
percent of capacity, (3) 10-year straight-line depreciation, (4) 50 percent of retained fines,
valued at $9/ton, recycled, and (5) average product price of $6/ton.
new source performance standard. Estimating the cost to achieve the two standards provides a
measure not only of the total cost but also the incremental cost of control.
Either the fabric filter or the venturi scrubber will enable a new plant to comply with the pro-
posed standard, and the capital costs for these devices do not appear to be significantly different
for either size plant (300 tons/hr or 150 tons/hr). On an annualized cost basis, it appears that the
fabric filter is the lesser-cost device for both plant sizes. The key element is that the fabric filter
collects the particulate material in a useful form, while the material collected by the scrubber must
be disposed of at the operator's expense. An independent study states that in the case of asphalt
concrete plants, properly designed, operated, and maintained fabric filter collectors can be a
profitable investment and not an add-on cost.15 This study concluded that, even for a small plant
(100 tons/hr), use of fabric filter collectors is more economical than wet collector systems.15 Thus,
it may be assumed that most new plants would favor a fabric filter control system on an economic
basis when selecting a control system to comply with the proposed standard.
The installation of a fabric filter on the smaller plant necessitates an increase in capital
investment of 24 percent over the base-plant investment. However, the incremental investment
required to equip the plant with a fabric filter rather than a low-energy scrubber (to comply with
reference process weight curve) is 6 percent. Similarly for the larger plant, the additional capital
investment required by the fabric filter over the base-plant investment is 28 percent, ;while the
incremental investment over equipping the plant with a low-energy scrubber is 4 percent.
The incremental investment required by the proposed standard above that required by State
standards is not anticipated to create any serious additional financing problems for new asphalt
concrete plants.
Since the control cost for a new plant meeting the proposed standard approximates the cost for
an existing plant meeting a less stringent standard, the management of new plant should find that
the market price is sufficient to recover much, if not all, of the cost of complying with the pro-
posed standard. As a result, there should be little or no reduction in earnings for the new plant.
14
-------
REFERENCES FOR TECHNICAL REPORT NO.6
References Cited
1. Air Pollution Engineering Manual. Danielson, J.A. (ed.). National Center for Air Pollution
Control, Public Health Service, U.S. Department of Health, Education, and Welfare.
Cincinnati, Ohio. PHS Publication No. 999-AP-40. 1967.
2. The Economics of Clean Air. Annual Report of the Administrator of the Environmental
Protection Agency to the Congress of the United States in compliance with Public Law 91-604,
the Clean Air Act, As Amended, p. 4-27. March 1972.
3. Vandergrift, A.E., L.J. Shannon, E.E. Sallee, P.G. Gorman, and W.R. Park. Particulate Air
Pollution in the United States. J. Air Pollut. Contr. Assoc. 21(6):321-328, June 1971.
4. Zenach, L.D. Personal communication to F. Kloiber, National Asphalt Pavement Association,
Riverdale, Md. Micropul Division, Slick Corporation, Summit, NJ. August 4 and 24, 1972.
5. Rylander, H.G. and J.O. Ledbetter. Bench Model Filter Tests. University of Texas. Dallas,
Texas. (Sponsored by the National Asphalt Pavement Association) 1972.
6. Jacobson, S. Air Pollution Control for Industry—High Efficiency Venturi Scrubbers for Hot
Mix Asphalt Plants. Poly Con Corporation. (Presented at Air Pollution Control Seminar of the
Asphaltic Concrete Producers Association. Oakland, Calif. May 27, 1971).
7. Environmental Protection Control at Hot Mix Asphalt Plants. The National Asphalt Pavement
Association. Riverdale, Md. Information Series No. 27.
8. Guide for Air Pollution of Hot Mix Asphalt Plants. The National Asphalt Pavement
Association. Riverdale, Md. Information Series No. 17.
9. Friedrick, H.E. Air Pollution Control Practices—Hot Mix Asphalt Paving Batch Plants. J. Air
Pollut. Contr. Assoc. 19:928, December 1969.
10. Control Techniques for Particulate Air Pollutants. National Air Pollution Control
Administration, Public Health Service, U.S. Department of Health, Education, and Welfare.
Washington, D.C. NAPCA Publication No. AP-51. January 1969.
11. Billings, C.E. and J. Wilder. Handbook of Fabric Filter Technology. In: Fabric Filter System
Study (Vol. I). GCA Corporation, GCA Technology Division. Bedford, Mass. Contract No.
CPA 22-69-38. December 1970.
12. Skinner, C.F. New Use for Baghouse Filter — Handling Hot Effluent. Plant Engineering.
23(13):57-59, June 26, 1969.
13. Frey, R. Personal communication to Ken Durkee, U.S. Environmental Protection Agency,
Research Triangle Park, N.C. Micropul Division, Slick Corporation, Summit, N.J. January 28,
1972.
14. Study by American Air Filter Company, Louisville, Kentucky. Conducted for the Plant Mix
Asphalt Industry of Kentucky, Incorporated, Frankfort, Kentucky. 1967.
15. Ashman, R. Filters for Asphalt and Coating Plants. Quarry Managers J. 54(12):433, December
1970.
Supplemental References
16. The Asphalt Handbook. The Asphalt Institute. College Park, Md. Manual Series No. 4. April
1965.
17. Asphalt Plant Manual (3rd Ed.). The Asphalt Institute. College Park, Md. Manual Series No.
3. March 1967.
18. Atmospheric Emissions from the Manufacture of Portland Cement. National Center for Air
Pollution Control, Public Health Service, U.S. Department of Health, Education, and Welfare.
Cincinnati, Ohio. PHS Publication No. 999-AP-17. 1967.
19. Bituminous Construction Handbook. Barber-Greene Company. Aurora, Illinois. 1963.
20. Maintenance and Operation Instructions for Cedarrapids Equipment. Iowa Manufacturing
Company. Cedar Rapids, Iowa. IMCO-038 DRIERS. 1967.
15
-------
21. Maintenance and Operation Instructions for Cedarrapids Equipment — Asphalt Mixing Plant
Model "HC" Series. Iowa Manufacturing Company. Cedar Rapids, Iowa. IMCO-098. 1967.
22. Mix Design Methods for Asphalt Concrete and Other Hot-Mix Types (3rd Ed.). The Asphalt
Institute. College Park, Md. Manual Series No. 2. October 1969.
16
-------
TECHNICAL REPORT NO. 7 -
PETROLEUM REFINERIES, FLUID CATALYTIC CRACKING UNITS
SUMMARY OF PROPOSED STANDARDS
Proposed standards of performance for petroleum refineries will limit emissions of participates
(including visible emissions) and carbon monoxide from new or modified catalyst regenerators on
fluid catalytic cracking units.
Standards for Participates
The proposed standards will limit particulate emissions to the atmosphere as follows:
1. No more than 50 mg/Nm3 (undiluted), or 0.022 gr/dscf.
2. No more than 20 percent opacity except for 3 minutes in any 1 hour.
The proposed visible emission standard is compatible with the mass emission limit; if
particulate emissions are at or below 50 mg/Nm3, visible emissions will be below 20 percent
opacity.
Standard for Carbon Monoxide
The proposed standard will limit carbon monoxide emissions to no more than 0.050 percent by
volume, dry basis.
The proposed carbon monoxide standard can be met by incineration. The most common device
is an incinerator/waste heat boiler, which is normally fired with refinery fuel gas. In the units
tested, only gas was used to supplement the combustion of carbon monoxide. Fuel oil can be used
as the auxiliary fuel, but greater concentrations of particulate would be expected. No emissions
data are available for well controlled units using fuel oil.
The availability of refinery fuel gas and boiler maintenance considerations minimize the use of
fuel oil. For these reasons provisions added to the regulations allow the particulate matter
generated by firing fuel oil to be subtracted from the total particulate matter measured by the
compliance test method. Owners and operators will be required to meet the visible emission
standard regardless of the type of auxiliary fuels burned.
EMISSIONS FROM PETROLEUM REFINERIES
An uncontrolled fluid catalytic cracking unit can release over 300 Ib/hr of catalyst dust.3 Such
installations are equipped only with internal centrifugal dust collectors, which primarily serve to
recycle the catalyst. The proposed standards will require owners and operators of new facilities to
reduce the level of particulate emissions about 93 percent below the level of an uncontrolled unit.
In addition, an uncontrolled unit can release over 15 pounds of carbon monoxide per barrel of
petroleum feedstock processed.4 For a unit processing 40,000 barrels per day (bbl/day), about 20
tons/hr of carbon monoxide would be released. The proposed standard will require owners and
operators of new facilities to reduce carbon monoxide emissions 99.5 percent below those of an
uncontrolled -unit.
At many modern petroleum refineries, an electrostatic precipitator is used to control dust from
the fluid catalytic cracking unit catalyst regenerator. A waste heat boiler fired with auxiliary fuel is
used to control carbon monoxide from the units (see Figures 4 and 5).
The reference process weight regulation (Table 1) is less stringent than the proposed standard
for units of a practical size (less than 150,000 bbl/day). The most stringent State or local
regulations restrict emissions to 30 Ib/hr.
17
-------
WET GAS
RAW GASOLINE
STEAM
LIGHT CYCLE OIL
HEAVY CYCLE OIL
BOTTOMS•
FRACTIONAL
PRECIPITATOR
BOILER
REGENERATOR
REACTOR
TIC STACK
)R
x
-•SS
STEAM
tEJ)
LI-
FEED.
Figure 4. Petroleum refinery fluid catalytic cracking unit with control system.
POWER RECOVERY (OPTIONAL) |f ™Tj $= - ------- mmm
ELECTROSTATIC
PRECIPITATOR
CARBON MONOXIDE
BOILER
Figure 5. Fluid catalytic cracking unit regenerator with carbon monoxide boiler and electro-
static precipitator.
New units will range in size from 10,000 to 100,000 bbl/day of fresh feed, with gas flow rates
varying from 20,000 to 350,000 dscfm, respectively. The proposed standard will allow 3 to 60
pounds of particulates per hour. For a typical unit rated at 50,000 bbl/day of fresh feed at a gas
flow rate of 150,000 dscfm, the proposed standard will allow an emissions of 25.7 Ib/hr of
18
-------
participate matter. The reference process weight regulation will limit emissions to 64 Ib/hr based
on a catalyst recirculation rate of 50 tons/min.
State or local regulations are comparable with the EPA-proposed standard for carbon
monoxide, but are generally framed in different language. Nonfederal standards usually require
the combustion of carbon monoxide for 0.3 second at a temperature above 1300 °F. The same type
of control equipment (carbon monoxide boilers) is required in most cases to meet the proposed
standards. For certain types of catalyst regenerators, the boiler may not be required because the
carbon monoxide is combusted in the regenerator itself. In either case, the proposed standard
requires a 99.5 percent reduction in carbon monoxide emissions over an uncontrolled unit.
uncontrolled unii.
RATIONALE FOR PROPOSED STANDARDS
Preliminary investigations revealed the locations of 17 well controlled cracking units in the
United States. These plants were visited and information was obtained on the type of refinery
process and the control equipment used. Visible emissions at 13 plants were observed to be 20
percent opacity or less. Judgment regarding the feasibility of stack testing was made for each plant.
In this regard, 12 locations were unsatisfactory because the control equipment was judged to be
less than optimum or the physical layout of the equipment made testing unfeasible. One unit could
not be tested because it was undergoing a turnaround. Stack tests were conducted at four
locations.
Particulate Matter
The proposed particulate emission limit is based on tests by EPA, local agencies, and plant
operators and data on control efficiencies and emission levels achieved at similar stationary
sources. The control level required by the standard has been demonstrated on only a few catalyst
regenerators. In proposing new standards, much weight has been given to the fact that higher
efficiency particulate collectors could be installed at refineries and the fact that such collectors
have been installed at both smaller and larger particulate sources, for example, basic oxygen steel
furnaces and secondary lead furnaces.
Of the three catalyst regenerators tested by EPA, all of which were controlled by electrostatic
precipitators, one showed particulate emissions below the proposed standard (Figure 6). Emissions
average 0.014 gr/dscf for three individual runs ranging between 0.011 and 0.016 gr/dscf. This unit
was retested by EPA and showed average particulate emissions of 0.022 gr/dscf with three
individual runs ranging between 0.020 and 0.023 gr/dscf. Emission data gathered by the refinery
over a 7-month period of operation (Figure 6) showed average particulate emission of 0.014 gr/dscf
from 14 individual tests ranging between 0.010 and 0.021 gr/dscf. In addition, emission data
gathered by a second refinery over a 17-month period of operation (Figure 6) showed average
particulate emission of 0.017 gr/dscf from eight individual tests ranging between 0.015 and 0.022
gr/dscf. The refinery test methods were the same in each case. Both refiners employed different
filter media than the EPA method, but neither included impingers.
EPA tests of two units controlled by electrostatic precipitators (Figure 6) average 0.037 gr/dscf
for each test. Results of a fourth unit were invalid because of a process malfunction during testing.
Results of six tests on four fluid catalytic cracking unit regenerators conducted by a local control
agency3 are shown in Figure 6. Emissions from all units were controlled by electrostatic precipi-
tators and carbon monoxide waste-heat boilers. Particulate emissions averaged 0.013, 0.017, 0.018,
0.018, and 0.020 gr/dscf, respectively. The test method used is comparable with, although not
identical to, the EPA method.
Two control equipment designers have stated that they will guarantee particulate emission levels
of about 0.010 gr/dscf. Both of these firms have installed several units on catalyst regenerators.
To determine the level of the proposed standard, further evaluation was made of particulate
collector design. Electrostatic precipitators are the only high—efficiency dust collectors that have
been used with catalyst regenerators. Many of these precipitators are rated at 90 to 95 percent
19
-------
0.04
0.03
00
ISi
0.02
cc
<
a.
0.01
0
e
CODE METHOD NUMBER
2 - EPA METHOD 5
5 - IMPINGERS PLUS PAPER THIMBLE
6 - ALUNDUM THIMBLE; GELMAN "A" FILTER
13 - ALUNDUM THIMBLE PLUS IMPINGERS
r-, MAXIMUM
J-L AVERAGE
U MINIMUM
EPA OTHER
e - ELECTROSTATIC PRECIPITATOR
6 m - CARBON MONOXIDE BOILER
5
13
Aj(em) A2(em) AS (em) B(me)
C(me) E(em) F(me)G(em)H(em)l(em)J(me) K(cm)
PLANT (CONTROL EQUIPMENT)
Figure 6. Particulate emissions from petroleum refineries, fluid catalytic cracking units.
•
efficiency for oil refinery emissions, as compared with the 98 to 99+ percent range encountered in
other industries; however, the exit concentrations at refineries are not as low as with some other
sources. For instance, an electrostatic precipitator cited in Report No. 12 for iron and steel plants
was found to achieve a level of 0.007 gr/scf when applied to a basic oxygen steel furnace. The
efficiency of this precipitator was considerably greater because there was a much greater inlet
loading to the precipitator than is encountered with catalyst regenerators at oil refineries.
Several parameters affect the performance of an electrostatic precipitator, and it is not within
the scope of this document to discuss them all. Other parameters being equal, however, collector
efficiency tends to increase with plate area. It is significant that:
1. The electrostatic precipitator that exhibited the lowest exit concentration during the EPA tests
has considerably greater plate area (250 ft2 /1000 acfm of gases) than the other electrostatic
precipitators (175 and 190 ft2 /1000 acfm) tested by EPA.
2. The previously mentioned precipitator serving a basic oxygen steel furnace has a plate area of
375 ft*/1000 acfm.
3. Precipitators with collection plate areas from 250 to more than 400 ft2/1000 acfm have been
installed at steel furnaces, cement kilns, municipal incinerators, and other sources.
Based on these considerations, it is concluded that exit concentrations of 0.020 gr/dscf can be
'achieved with electrostatic precipitators of the same general design as, but with greater plate area
than those that have already been installed by refiners. In addition, it will probably be necessary
that the precipitators be constructed in modules so that maintenance and repair operations can be
20
-------
conducted while the unit remains in service. Catalyst regenerators frequently are kept on-stream
for 2 years or longer with few shutdowns that would allow time to conduct repairs and
maintenance.
Visible emissions of less than 20 percent opacity were observed at all three of the units tested by
EPA. Ten additional units were observed by EPA engineers to have visible emission levels that meet
the proposed standard. The proposed standard can be exceeded for 3 minutes in any 1 hour to
allow the blowing of soot from the tubes of the carbon monoxide waste-heat boiler.
Carbon Monoxide
In addition to particulate matter, carbon monoxide concentrations were determined during the
EPA tests of well controlled cracking units. The four units, each controlled by a carbon monoxide
incinerator/waste heat boiler, showed carbon monoxide emissions well below the proposed standard
(Figure 7). Carbon monoxide emissions from three tests on two units averaged 5, 10, and 25 ppm
(25 ppm is 0.0025 percent by volume). No measurable carbon monoxide emissions occurred at the
two remaining units tested.
40
11
30
20
x
o
o
CO
Si 10
o
11
11
TEST METHOD NUMBER
11-NDIR
MAXIMUM
AVERAGE
e - ELECTROSTATIC PRECIPITATOT
m - CARBON MONOXIDE BOILER
•t-
H-
B(me) C(me) " D(em)
PLANT (CONTROL EQUIPMENT)
Figure 7. Carbon monoxide emissions from petroleum refineries, fluid catalytic cracking units.
The proposed carbon monoxide standard will require the use of either an incinerator/waste heat
boiler or a regenerator that is capable of the almost complete burning of carbon and carbon
monoxide to carbon dioxide. Burning carbon monoxide in the regenerator (in situ)4 is a relatively
recent innovation that was developed along with improvements in catalytic cracking technology,
which significantly increase the yield of gasoline. In recognition of the more effective use of natural
resources, the standard is being proposed at a level that can be achieved with in situ combustion
even though incinerator/waste heat boilers would provide greater reductions in carbon monoxide
emissions.
21
-------
ENVIRONMENTAL IMPACT OF PROPOSED STANDARDS
The disposal of collected catalyst dust presents a potential adverse environmental effect, as
would the disposal of scrubber liquor if scrubbers were utilized with fluid catalytic cracking units.
Nevertheless, it is expected that electrostatic precipitators will continue to be the principal
collection device used in the near future.
Crystalline zeolite (molecular sieve) catalysts are now in predominant use in the industry. The
bulk of collected particulates is catalyst dust caused by attrition. It has little catalytic cracking
value and is seldom returned to the cracking system. Collected particulates include zeolites,
unburned carbon, trace metals, sulfur compounds, silicates, and alumina, none of which have
appreciable solubility. The usual method of disposal is by landfill.
ECONOMIC IMPACT OF PROPOSED STANDARDS
The growth in catalytic cracking capacity is estimated to be about 685,000 bbl/day of fresh feed
over the next 5-year period. Currently, about 80 percent of existing capacity is operated by
"major" petroleum refiners and 20 percent is operated by "independent" petroleum refiners. The
trend in new refinery construction is to install processing units of increased capacity. For the
purposes of this analysis, it is assumed that about 80 percent of new capacity will be from
construction of large (65,000 bbl/day of fresh feed) units by the major refiners and the remaining
20 percent from construction of small units (20,000 bbl/day of fresh feed) by the independent
refiners. Over the next 5 years, then, it is estimated that nine large units and six small units will be
constructed, or about two large units and one small unit annually.
The costs required to meet the proposed emission standards are proportionately less on larger
sized units. The investment costs for a carbon monoxide boiler and an electrostatic precipitator
installed on a 65,000-bbl/day fresh-feed unit and on a 20,000-bbl/day fresh-feed unit range from
about 25 to 36 percent of the basic process equipment investment cost, respectively. This cost is not
all unproductive investment, however. The cost savings generated from steam production in the
carbon monoxide boiler more than offset the annual cost of the electrostatic precipitator and
carbon monoxide boiler. The value of the stream to the refiner depends on his alternate fuel cost;
and, because the price of natural gas and other fuels is likely to keep rising, the value of the stream
produced will increase in the future. The carbon monoxide boiler investment costs and annual
savings are:
Unit size Investment Annual savings
20,000 bbl/day $1,800,000 $235,000
65,000 bbl/day $3,000,000 $l,25o'oOO
Because the carbon monoxide boiler has an attractive economic payout, most new units would
be built with carbon monoxide boilers even without the requirements of the proposed standards.
The increase in process unit investment that is necessary to install an electrostatic precipitator on a
65,000-bbl/day unit and a 20,000-bbl/day unit, with the carbon monoxide boiler investment cost
included as basic process equipment cost, ranges from about 6 to 8 percent. The increase in annual
operating cost ranges from about 6.2 to 9 percent.
The investment and annualized costs required to meet the new source performance standard
and the reference process weight regulation are shown in Table 4. These costs are based on the use
of electrostatic precipitators as the particulate control device. The basic units were assumed to
have two-stage internal cyclones.
22
-------
Table 4. CONTROL COSTS FOR CATALYTIC CRACKING UNITS
EQUIPPED WITH ELECTROSTATIC PRECIPITATORS
Plant size,
bbl/day
20,000
65,000
Emission
standard
Proposed
performance
standard
0.022 gr/dscf
Reference
process weight
standard
equivalent to
0.09 gr/dscf
Proposed
performance
standard
0.022 gr/dscf
Reference
process weight
standard
equivalent to
0.035 gr/dscf
Control
investment,
$
700,000
470,000
1,150,000
1,050,000
Annual
cost,
$/yr
1 50,000
110,000
225,000
205,000
Annual cost
per unit of
throughput,
tf/bbl
2.2
1.6
1.0
0.9
REFERENCES FOR TECHNICAL REPORT NO. 7
1. National Emission Standards Study; A Report to the Congress of the United States by the
Secretary of Health, Education, and Welfare (Appendix—Vol. I). National Air Pollution
Control Administration, Public Health Service, U.S. Department of Health, Education, and
Welfare. Washington, D.C. March 1970. p. E-54.
2. Control Techniques for Carbon Monoxide Emissions from Stationary Sources. National Air
Pollution Control Administration, Public Health Service, U.S. Department of Health,
Education, and Welfare. Washington, D.C. NAPCA Publication No. AP-65.1970.
3. Murray, R.C., H.E. Chatfield, E.E. Larsson, and M.C. Lawrence. A Report of Source Tests on
Emissions from Catalytic Cracking Unit Regenerators, County of Los Angeles Air Pollution
Control District. U.S. Environmental Protection Agency. Research Triangle Park, N.C. Order
No. 2PO-68-02-332. 1972. 88p.
4. Air Pollution Engineering Manual. Danielson, J.A. (ed.). National Center for Air Pollution
Control, Public Health Service, U.S. Department of Health, Education, and Welfare.
Cincinnati, Ohio. PHS Publication No. 99-AP-40.1967. p. 647-649.
23
-------
TECHNICAL REPORT NO. 8 -
PETROLEUM REFINERIES, BURNING OF GASEOUS FUELS
SUMMARY OF PROPOSED STANDARD
The proposed standard of performance for petroelum refineries will limit emissions of sulfur
dioxide from process heaters, boilers, and waste gas disposal systems that burn process gas
generated in the refinery. The proposed standard does not apply to extraordinary situations, such
as emergency gas releases, or to the burning of liquid or solid fuels in the same heaters and boilers.
The proposed standard will limit sulfur dioxide emissions to the atmosphere from heaters,
boilers, and flares by specifying that the fuel gas burned shall contain no more than 230 mg/Nm3 of
hydrogen sulfide, or 0.10 gr/dscf, unless resultant combustion gases are treated in a manner
equally effective in preventing the release of sulfur dioxide to the atmosphere.
Compliance with the standard is based upon measurement of the hydrogen sulfide concen-
tration in the fuel gas or the sulfur dioxide concentration in the exit gases. The proposed standard
is equivalent to a sulfur dioxide content of approximately 20 gr/100 scf of fuel gas burned. Burning
such fuel will result in a concentration of 15 to 20 ppm of sulfur dioxide in the combustion
products.
The regulation would have the effect of requiring hydrogen sulfide removal from all refinery-
generated fuel gases used to fire new boilers and heaters. The extracted sulfur compounds cannot
be burned in flares, heaters, or any other sources unless the control devices used (for example,
flue gas scrubbers) are equally effective as fuel desulfurization.
EMISSIONS FROM PETROLEUM REFINERIES
Refinery processes, such as distillation and fluid catalytic cracking, produce substantial
quantities of "process gas" (Figure 8) that may contain more than 5 percent hydrogen sulfide by
volume. If this untreated gas is burned in heaters, boilers, or flares, substantial quantities of sulfur
dioxide will be emitted to the atmosphere. Monoethanolamine (MEA) and diethanolamine (DBA)
scrubbing units (Figure 9) are widely used to remove the hydrogen sulfide from both refinery
process gases and natural gas.1 In addition, new processes that employ other scrubbing media are
being applied to refinery process gases. The proposed standard will require owners and operators
of new facilities to reduce the hydrogen sulfide content of refinery-derived fuel gases to levels that
are consistent with these technologies. For most such gases, the proposed limit represents a
reduction of more than 99 percent in sulfide levels. For a fuel gas equivalent to methane, the
resultant emission of 16 ppm sulfur dioxide is roughly equivalent to the burning of fuel oil
containing 0.04 percent sulfur by weight.
Approximately 1 million tons of sulfur charged to U.S. refineries could not be accounted for in
1970. The majority of this sulfur was probably burned and emitted to the atmosphere as sulfur
dioxide. If all sources were controlled to the level of the standard, these emissions would be
reduced by 95 percent. (Most of the difference between the 99+ percent mentioned in the
preceding paragraph and 95 percent is the result of losses in conversion of the recovered gases into
sulfur or sulfuric acid.)
25
-------
At the present time, only one local regulation restricts sulfur dioxide emissions from the burning
of refinery process gas. Some State and local agencies have proposed regulations with limits
ranging from 10 to 50 grains of hydrogen sulfide per 100 scf of fuel gas burned (19 to 94 grains of
sulfur dioxide per 100 scf).
PROCESS GAS
SOUR PROCESS GAS
TO AMINE TREATING
ATMOSPHERIC
TOWER
SOUR PROCESS GAS
TO AMINE TREATING
CRUDE DISTILLATION
Figure 8. Petroleum refinery process gas system.
RATIONALE FOR PROPOSED STANDARD
The proposed sulfur dioxide standard is consistent with the capability of a well designed and
properly operated amine treating unit that is used to scrub typical refinery process gases at the
moderate pressures available in the refinery.2'4 Amine treating technology is well demonstrated
and has been widely used to reduce hydrogen sulfide concentrations in gas streams to levels less
than that required to meet the proposed standard.
Three refineries were visited by EPA representatives, and information was obtained on the
operation of amine systems. All systems were stated to be operating with exit gas concentrations of
less than 13 grains of hydrogen sulfide per 100 scf. Diethanolamine (DEA) and monoethanolamine
(MEA) scrubbers are found in almost every U.S. refinery, and hundreds are operated in natural gas
fields throughout the country. Amine treating is used to reduce the hydrogen sulfide content of
natural gas to the pipeline specification level of 0.25 gr/100 scf. It would be difficult, however, to
consistently achieve this level in a refinery where treating pressures are lower than in natural gas
26
-------
^TREATED PROCESS GAS
,REGENERATED
AMINE; FOUL AMINE
SOUR PROCESS GAS
WATER
AMINE
Figure 9. Petroleum refinery process gas treating unit.
fields, since higher treating pressures favor hydrogen sulfide removal. The refinery gases also
contain unsaturated hydrocarbons not usually found in natural gas. These unsaturates tend to
accelerate fouling of the amine solutions and to reduce scrubbing efficiency. There is no
discernible difference in plant hardware or design operating parameters for an amine treating unit
designed to treat refinery process gas to 10, 50, or 100 grains of hydrogen sulfide per 100 scf.2"4
Exit gas levels are apparently associated with operating practices; a standard equivalent to 13
grains of hydrogen sulfide per 100 scf requires good operating practice and has, therefore, been
chosen as the basis for the standard.
The proposed standard is expressed in terms of hydrogen sulfide, but can be measured as either
hydrogen sulfide in the fuel gas or total sulfur compounds in stack gases. Process gas streams also
contain small amounts of other sulfur compounds, which are not removed by the amine scrubbing
system. These materials would be included in the total sulfur compounds measured.
ENVIRONMENTAL IMPACT OF PROPOSED STANDARD
Due to thermal and chemical degradation, amine solutions require periodic replacement or
treatment. In MEA treating systems, it is common practice to send a continuous slipstream of
amine solution to a heated reclaimer. There, caustic soda is added and MEA is disassociated from
complex salts, distilled, and returned to the regenerator tower. As the salt content in the reclaimer
increases, it is necessary to purge the salts and recharge the reclaimer. For a typical 200,000-
bbl/day refinery producing 30 x 106 dscf of process gas, about 2000 Ib/month of waste salts may be
formed. Water is added to reduce the viscosity of salt slurry before disposal by incineration or
landfill.
Diethanolamine has a higher boiling point than MEA and cannot be similarly treated. Usual
practice is to continue operating until the solution is spent and the hydrogen sulfide content of
treated gas reaches undesirable levels. The entire solution is then replaced with fresh solution. In
the typical refinery cited above, approximately 50,000 gallons of solution containing 20 percent
DBA and 10 percent complex salts would have to be removed annually.
27
-------
ECONOMIC IMPACT OF PROPOSED STANDARD
Treatment of the various refinery gas and liquid streams to control product quality is a
commonly accepted and necessary practice. Consultation with several engineering companies that
design amine absorption systems, which are the most commonly used control devices, indicates
that there could be a 2 to 5 percent increase in investment cost, but no discernable difference in
operating costs, between a new plant designed to meet the equivalent of 13 grains of hydrogen
sulfide per 100 scf and a new one designed to meet 100 grains of hydrogen sulfide per 100 scf
(typical current practice). Therefore, there is a small increase in amine treating cost to refiners as a
result of the proposed standard. In addition, increased operator effort and attention may be
required to maintain the design efficiency of the process during actual operation. Because this cost
factor is quite variable depending on the individual Company's present operating practice and
should be of minor consequence, it has not been quantified. If the refiner chooses to run an
increased volume of gas through an existing amine absorption system, he may incur costs in
upgrading the existing system to meet the proposed standard. Because each system must be
examined individually to determine the cost of upgrading, no attempt has been made to give costs
for this type of modification.
It is the intent of the proposed standard that hydrogen-sulfide-rich gases exiting the amine
regenerator be directed to an appropriate recovery facility, such as a Claus sulfur plant. A medium-
size refinery that processes crude oil containing 0.92 weight percent sulfur, the national average in
1968, would have an emission potential of over 100 tons/day of sulfur dioxide (50 tons/day of
sulfur) from the amine regenerator. The annualized cost was calculated for a range of Claus plant
sizes. A discontinuity occurs in the cost-capacity curve at about 10 long tons/day. The reason for
the discontinuity is that for plants up to about 10 long tons/day, less costly prefabricated package
units can be used. Units producing more than 10 long tons/day are generally field-erected and
considerably more expensive.
For each size unit, the required sulfur sales price to break even was calculated. At a sales price of
$20/long ton, the break-even size for package Claus units is about 5 long tons/day. The plant
investment for a 5-long-ton/day package Claus unit is about $90,000 exclusive of possible future
investment for control of the sulfur dioxide in the tail gas. The break-even size (at a sulfur price of
$20/long ton) for field-erected Claus units is about 15 long tons/day, which represents an
investment of about $350,000.
No data are presented to show the cost that refineries would incur if their hydrogen sulfide
removal systems were required to meet the 0.25 gr/100 scf achieved by plants processing natural
gas. There are several reasons why one should not compare natural gas processing plants with
refinery fuel gas systems. The natural gas plant processes gas at a high pressure, with a stable gas
composition, and with low levels of impurities. These conditions allow better hydrogen sulfide
absorption. Refinery gas is at lower pressure, has a variable composition, and has a variety of
impurities that reduce the ability of an absorption system to reach the low levels of hydrogen
sulfide achieved in a natural gas plant. The refinery gas pressure could be increased at a high cost,
but other limitations would still prevent the absorption system from achieving the low levels found
in natural gas plants.
Cost data have not been developed for higher pressure absorption systems, but the small
incremental reduction in hydrogen sulfide would make such a system highly questionable from a
cost-effective point of view.
REFERENCES FOR TECHNICAL REPORT NO. 8
1. Kohl, A.L. and F.C. Riesenfeld. Ethanolamines for Hydrogen Sulfide and Carbon Dioxide
Removal. In: Gas Purification. New York, McGraw Hill Company, Inc., 1960.
2. Thompson, H.L. Private communication to R.K. Burr, U.S. Environmental Protection Agency,
Research Triangle Park, N.C. UOP, Process Division, Des Plaines, 111. November 15,1971.
28
-------
3. Parnell, D.C. Private communication to R.K. Burr, U.S. Environmental Protection Agency,
Research Triangle Park, N.C. Ford, Bacon and Davis Texas, Garland, Texas. January 4, 1972.
4. Mayes, J.R. Private communication to R.K. Burr, U.S. Environmental Protection Agency,
Research Triangle Park, N.C. Graff Engineering Corporation, Dallas, Texas. January 6,1972.
29
-------
TECHNICAL REPORT NO. 9 -
STORAGE VESSELS FOR PETROLEUM LIQUIDS
SUMMARY OF PROPOSED STANDARDS
Standards of performance are being proposed for new storage vessels that have capacities
greater than 245,000 liters, or 65,000 gallons, and that are used for the storage of gasoline, crude
oil, or petroleum distillates.
The proposed standards will limit hydrocarbon emissions from any storage vessel that contains
any petroleum product having a true vapor pressure, at actual storage conditions, as follows:
l.No more than 78 mm Hg, or 1.52 psia: the storage vessel must be equipped with a
conservation vent or equivalent.
2. More than 78 mm Hg but not more than 570 mm Hg, or 11.1 psia: the vessel must be equipped
with a floating roof or equivalent.
3. More than 570 mm Hg: the vessel must be equipped with a vapor recovery system or
equivalent.
In contrast to other new source performance standards, the standards for storage vessels are not
proposed in terms of allowable hydrocarbon emissions. Nevertheless, the standards do limit
emissions to specific levels, and hydrocarbon release rates can be calculated from empirical
relationships developed for such equipment. Any device capable of providing comparable control
of hydrocarbon emissions may be used in lieu of the specified device.
HYDROCARBON EMISSIONS FROM STORAGE TANKS1
Hydrocarbon emissions from storage vessels depend on three basic mechanisms: breathing loss,
working loss, and standing storage loss. Breathing and working losses are associated with cone-
roof tanks2 (Figure 10), and standing storage losses are associated with floating-roof tanks (Figures
11, 12, and 13).
Breathing losses are hydrocarbon vapors expelled from the vessel by expansion of existing
vapors due to increases in temperature or decreases in barometric pressure. Working losses are
hydrocarbon vapors expelled from the vessel during emptying or filling operations. Emptying
losses result from vapor expansion caused by vaporization after product withdrawal. Filling losses
are the amount of vapor (approximately equal to the volume of input liquid) vented to the
atmosphere by displacement.
Breathing and emptying losses are usually restricted to fixed-roof tanks vented at what is
essentially atmospheric pressure. Filling losses are experienced in fixed-roof tanks and low-
pressure storage tanks vented to the atmosphere. Both working losses and breathing losses can be
significant and are therefore taken into consideration when proposing the standards.
Standing storage losses are hydrocarbon emissions from floating-roof tanks. They are caused by
the escape of vapors through the seal system between the floating roof and the tank wall (Figure
14), the hatches, glands, valves, fittings, and other openings.
31
-------
CONSERVATION —
(PRESSURE-VACUUM)
VENT
'MANHOLE
GAUGEx
HATCH \
NOZZLE
MANHOLE
Mil
Figure 10. Fixed-roof (cone-roof) storage vessel.
Figure 11. Single-deck floating-roof storage vessel.
D:
SEAL HATCH SUPPORT DRAIN
VENTx GUIDE, SEAL
NOZZLE
-COMPARTMENT
PARTITIONS
MANHOLE
Figure 12. Double-deck floating-roof storage vessel.
32
-------
GUIDE AND
STRUCTURAL COLUMNS
Figure 13. Covered floating-roof storage vessel with internal floating cover.
SECONDARY SEAL OPTIONAL
PRIMARY FABRIC SEAL
WEATHER
SHELL,
SEAL CONTAINING
LIQUID, GAS, OR
RESILIENT MATERIAL
METALLIC SEAL
TANK SHELL-
TANK SHELL
METALLIC TYPE SEAL
NONMETALLIC TYPE SEAL
Figure 14. Seals for floating-roof storage vessels.
The magnitude of hydrocarbon emissions from storage vessels depends on many factors,
including the physical properties of the material being stored, climatic and meterological
conditions, and the size, type, color, and condition of the tank. To quantify emissions from existing
storage tanks, emission factors were calculated based on correlations developed by the American
Petroleum Institute.2'3 These emission factors are based on national average wind velocities,
average ambient temperature changes, physical characteristics of typical products, average tank
size and mechanical conditions, volume throughputs, and other pertinent parameters.
33
-------
Using these factors and estimated current control levels, annual hydrocarbon emissions from
crude oil, gasoline, and distillate tanks are estimated at 1.3 million tons. At present, 75 percent of
these tanks are equipped with floating roofs. The annual estimated emissions constitute about 3
percent of the total national hydrocarbon emissions and about 7 percent of the 18.6 million tons/yr
emitted from all stationary sources. Using current controls, hydrocarbon emissions from new
gasoline and crude oil storage vessels would be about 41,000 tons/yr. The proposed standard will
reduce this loss to about 7700 tons/yr. Based on the current emissions rate, this represents a
reduction of about 80 percent.
State and local regulations that limit hydrocarbon emissions from petroleum storage vessels are
similar to the proposed standard. Typically, cone-roof tanks are not allowed for the storage of
materials having true vapor pressures in excess of 1.5 psia, and floating-roof tanks are not allowed
for storage of materials having a true vapor pressure in excess of 11.0 psia.
RATIONALE FOR PROPOSED STANDARDS
A contract study made for EPA and an analysis of capacities of tanks installed in petroleum
bulk terminals5'6 and plants indicate that tanks with capacities greater than 65,000 gallons
account for over 95 percent of hydrocarbon emissions from storage tanks. These data also indicate
that few tanks with capacities less than 65,000 gallons will be constructed to store the specified
products and that the relative cost of control devices increases sharply at the lower capacities.
Hydrocarbon emissions from the storage of jet fuels, volatile crude oils, and gasolines can be
considerable. A true vapor pressure greater than 1.52 psia requires a high degree of control, and
this level provides a distinct break point in those products that can be stored in cone-roof tanks.
The proposed standard will allow less volatile distillate fuel oils, kerosenes, heavy catalytic cracked
naphthas, and heavy crude and residual oils to be stored in cone-roof tanks. Conservation vents
would not be required on cone-roof tanks used to store fuel or residual oils.
The proposed standard requires the use of a vapor recovery system (Figure 15) or its equivalent
when storing a material having a true vapor pressure greater than 11.1 psia. The materials most
likely to exceed this vapor pressure at actual storage conditions are certain volatile winter-grade
northern gasolines and volatile gasoline-blending components that are stored in areas such as the
FUEL GAS
STORAGE TANKS
VAPOR TANK
COMPRESSOR
Figure 15. Storage vessel vapor recovery system.
34
-------
Gulf Coast of the United States. Vapor pressures greater than 11.1 psia would probably occur in
such an area over a very short period in the fall when winter-grade gasolines are accumulated in
storage for shipment north. During this period, Gulf Coast ambient temperatures remain high.
However, data obtained from Gulf Coast refineries, where the worst climatic conditions are likely
to be encountered, show that essentially all gasolines and most blending components can be stored
in floating-roof tanks in any part of the country if the proposed standard is used. This is contingent
upon using water cooling systems designed and properly operated to ensure adequate cooling of
the product prior to storage.
Beyond a true vapor pressure of 11.1 psia, losses from a floating-roof tank increase very rapidly,
and surface boiling, with concurrent high losses, is likely to occur. Accordingly, storage of
materials having a true vapor pressure greater than 11.1 psia at actual storage conditions should
be controlled by vapor recovery systems, pressure storage, refrigeration, or combinations thereof.
Vapor recovery systems have been used to a small extent to control hydrocarbon emissions from
large tank farms and bulk terminals, and were considered as a possible means of controlling
emissions from the storage of all liquids with high true vapor pressures. However, they have not
been demonstrated to be reliable in all areas of the country. These systems have generally been
reliable in regions of moderate climate where excessive, long-term vapor loads on the system
caused by high summer temperatures are minimized. In addition, when a vapor recovery system is
shut down by compressor failure or for maintenance, no controls exist for the entire tank farm.
ENVIRONMENTAL IMPACT OF PROPOSED STANDARDS
A substantial portion of the hydrocarbon emissions from storage tanks are compounds that
react in the atmosphere to form photochemical oxidants. Typical emissions from gasoline storage
tanks are C4 through C6 paraffins, C4 and C5 olefins, and small quantities of propane. Available
information indicates that all of these compounds, with the exception of propane, are
photochemically reactive.
No adverse environmental effects will occur as a result of meeting the requirements of the
proposed standard. Floating-roof tanks and conservation vents increase and preserve the yield of
salable products. Hydrocarbons collected in vapor recovery systems are normally recycled to the
refinery. In no case will the standard cause the generation of solid or liquid wastes.
ECONOMIC IMPACT OF PROPOSED STANDARDS
Over the next 5 years, approximately 175 new gasoline storage tanks and 420 new crude oil
storage tanks will be constructed annually in the United States. The number of new storage tanks
depends on the growth rate of the crude oil refining industry, on gasoline production, and on the
actual size of the tanks constructed. The estimated annual growth rate for both crude oil refining
and gasoline production is 4 percent. This growth rate will require storage for approximately
20,900,000 barrels of crude oil and 12,500,000 barrels of gasoline annually. The growth in
production of military jet naphtha, the least volatile material covered by the proposed standards, is
uncertain but will probably be small. For this reason national investment-cost projections have not
been made.
Tanks storing crude oil, gasoline, and petroleum distillates having a true vapor pressure greater
than 1.52 psia require a floating roof to meet the proposed standard. The increased investment
cost over a cone roof is 12 to 25 percent, depending on size. However, the savings from the product
recovered exceeds the annualized cost of the floating-roof installation when storing gasoline or
when storing crude oil in tanks greater than 20,000-barrel capacity. For an 80,000-barrel jet-
naphtha-fuel tank under average storage conditions, the annual cost is estimated at $1,000, or 0.1
cent per barrel of throughput. Costs for two sizes of tanks are shown in Table 5.
35
-------
Table 5. CONTROL COSTS FOR
PETROLEUM STORAGE TANKS
Tank capacity.
bbl
20,000
80,000
Incremental investment
cost above cone-
roof tank.
$
20,000
27,000
Material
stored
Gasoline
Crude oil
Jet naphtha
Gasoline
Crude oil
Jet naphtha
Annual cost.
$
savings 1,1 40a
480
2,100
savings 11,100
savings 5,200
1,000
aSavings from the product recovered exceed the annualized cost.
Vapor recovery systems are required for some of the materials covered by the standard. These
systems are considerably more costly than floating roofs. For some products (for example, winter-
grade northern gasolines and gasoline-blending stocks) having a true vapor pressure greater than
11.1 psia, the incremental recovery over a floating-roof tank with a capacity of 50,000 barrels is 7
percent. If the increase in control costs for the vapor recovery system is divided by the increased
product recovered, the cost per barrel is about 20 times the average control cost per barrel for the
floating-roof system. However, proper cooling at the production unit will keep the true vapor
pressure of these materials below 11.1 psia at actual storage conditions.
Other materials with a true vapor pressure greater than 11.1 psia at actual storage conditions
must be stored in vessels controlled by vapor recovery systems, or their equivalents, regardless of
cost in order to prevent excessive losses caused by surface boiling.
REFERENCES FOR TECHNICAL REPORT NO. 9
1. Evaporation Loss in the Petroleum Industry, Causes and Control. American Petroleum
Institute, Division of Refining. Washington, D.C. API Bulletin 2513.1959.
2. Evaporation Loss from Fixed Roof Tanks. American Petroleum Institute, Division of Refining.
Washington, D.C. API Bulletin 2518. 1962.
3. Evaporation Loss from Floating Roof Tanks. American Petroleum Institute, Division of
Refining. Washington, D.C. API Bulletin 2517. 1962.
4. Hydrocarbon Pollutant Systems Study; Volume I—Stationary Sources, Effects and Control.
MSA Research Corporation. Evans City, Pa. 1972.
5. U.S. Petroleum Inventories and Storage Capacity, National Petroleum Council. Washington,
6. 1967 Census of Business, Wholesale Trade, Petroleum Bulk Stations and Terminal. Bureau of
Census. Washington, D.C. 1968.
36
-------
TECHNICAL REPORT NO. 10 -
SECONDARY LEAD SMELTERS AND REFINERIES
SUMMARY OF PROPOSED STANDARDS
Standards of performance being proposed for new secondary lead smelters and refineries would
limit emissions of particulate matter (including visible emissions) from blast (cupola) and
reverberatory furnaces. Pot furnaces with a charging capacity of more than 250 kilograms (550
pounds) would be subject to visible emission limitations only.
Standards for Particulate Matter from Blast and Reverberatory Furnaces
The proposed standards would limit particulate emissions from blast and reverberatory furnaces
to the atmosphere as follows:
1. No more than 50 mg/Nm3 (undiluted), or 0.022 gr/dscf.
2. No more than 20 percent opacity.
The proposed visible emission standard is compatible with the mass emission limit for blast and
reverberatory furnaces; if particulate emissions are at or below 50 mg/£jm3, visible emissions will
be below 20 percent opacity. Observations of pot furnaces have shown that visible emissions will be
less than 10 percent opacity if commonly used dust equipment is installed and properly
maintained.
Standard for Particulate Matter from Pot Furnaces
The proposed standard for pot furnaces will limit visible emissions to less than 10 percent
opacity. ,
i
EMISSIONS FROM LEAD FURNACES
A poorly controlled (80 to 85 percent collection efficiency) secondary lead furnace can release 30
to 40 pounds of dust and fume per ton of lead produced.1 Such installations are likely to be
equipped with centrifugal dust collectors,, settling chambers, or low-energy scrubbers. The
approach results in a loss of valuable product,2 since average smelter dust is estimated to be 63
percent lead,3 and the dust can amount to 2 percent of the lead product. On the basis of this
poorly controlled emission rate, a collection efficiency of about 97 percent is required to meet
the particulate standards.
At well controlled secondary lead smelters (Figure 16), either baghouses or high-energy
scrubbers are used to collect dust and fumes from the furnace. When fabric filters are used to
control blast furnace emissions, they are normally preceded by an afterburner (Figure 17) to
incinerate oily and sticky materials to avoid blinding the fabric. This afterburner has the added
advantage of converting carbon monoxide to carbon dioxide. An afterburner is not needed in the
reverberatory furnace (Figure 18) since the excess air and temperature are sufficient to incinerate
carbon monoxide and hydrocarbons.
Emissions from blast and reverberatory furnaces are normally released into the atmosphere
through stacks with an average height of 150 feet; however, stack heights may range from a few
feet above the top of the control device (about 30 feet above ground level) to 300 feet.
37
-------
TO BLAST FURNACE
CONTROL SYSTEM
LEAD HOLDING,
WELTING
AND REFINING POTS
TO VENTILATION
CONTROL SYSTEM
GAS OR
FUEL OIL
TO REVERBERATORY
FURNACE
CONTROL SYSTEM
BLAST FURNACE
Figure 16. Secondary lead smelter process.
EMISSIONS
TO VENTILATION SYSTEM
JODOP
BLAST FURNACE AFTERBURNER
COOLING TOWERS
COOLING BLEED AIR
BAGHOUSE
DUST RECYCLED TO REVERBERATORY FURNACE
Figure 17. Controlled lead blast furnace, afterburner and baghouse.
Baghouses and scrubbers are also used to control pot furnaces (Figure 19). During melting and
holding operations associated with pot furnaces, uncontrolled emissions are quite low because the
vapor pressure of lead is low at the melting temperature. During dross skimming and refining
operations, emissions are substantially increased, and adequate ventilation must be provided to
protect the health of the workers. The latter requirements govern the volume of exhaust gases.
Emissions from pot furnaces are typically released into the atmosphere through short stacks, 15 to
35 feet in height.
State and local particulate regulations are less stringent than the proposed standards for blast
and reverberatory furnaces. The most stringent standards restrict particulate emissions from 20- to
80-ton furnaces to from 4 to 8 Ib/hr, which corresponds to from 0.02 to 0.08 gr/dscf. Some of these
standards are based on particulate sampling methods that differ from the EPA technique in that
they include material collected in wet impingers.
38
-------
EMISSIONS
TO VENTILATION SYSTEM
.j
Jnaan
REVERBERATORY FURNACE
COOLING TOWERS
COOLING BLEED AIR>
BAGHOUSE
DUST RECYCLE
Figure 18. Controlled lead reverberatory furnace, baghouse.
FROM FURNACES 1
U\J
HOLDING, LEAD MELTING,
AND REFINING POTS
DUST RECYCLED TO REVERBERATORY FURNACE
Figure 19. Controlled lead pot and ventilation system, baghouse.
For a typical blast (cupola) furnace rated at 50 tons/day at a flow rate of 15,000 dscf, the
proposed standard will allow the furnace to emit 2.6 Ib/hr of particulate matter. The reference
process weight regulation (Table 1) will limit emissions to 7.7 Ib/hr for a charging rate of 6900
Ib/hr. New furnaces will range in size from 20 to 80 tons/day in ingot production, with respective
gas flow rates of 10,000 to 40,000 dscfm.
39
-------
RATIONALE FOR PROPOSED STANDARDS
Particulate Matter from Blast and Reverberatory Furnaces
Preliminary investigations revealed the location of 11 well controlled plants. These plants were
visited, and information was obtained on the process and control equipment. Visible emissions at
the plants were observed to be less than 10 percent opacity. The feasibility of stack testing was
determined in each case. Six locations were unsatisfactory for testing because control equipment
was inadequate or because the physical layout of the equipment made testing unfeasible. Stack
tests were conducted at five locations, including three blast and two reverberatory furnaces.
All furnaces tested showed average particulate emissions below the proposed standard (Figure
20). The blast furnaces were controlled by (1) an afterburner and baghouse; (2) an afterburner,
baghouse, and venturi scrubber; and (3) a venturi scrubber. Particulate emissions averaged,
respectively, 0.003, 0.009, and 0.015 gr/dscf. The reverberatory furnaces were controlled by
baghouses, with particulate emissions averaging 0.004 gr/dscf in both cases.
0.04
0.03
0.02
o
0.01
-------
Designers and manufacturers of control equipment will guarantee efficiencies that achieve
outlet concentrations between 0.015 and 0.020 gr/dscf.
No visible emissions were observed at three of the furnaces tested; the other two had visible^
emissions of 15 percent opacity or less. Six additional furnaces were observed by EPA engineers to
have visible emissions within the proposed standard, although moisture-condensation plumes were
present in cold weather from those furnaces controlled by scrubbers.
Standard for Particulate Matter from Pot Furnaces
Other than visible emission limits, no emission standard has been proposed for pot furnaces.'
Nine smelters with pot furnaces controlled by baghouses or high-energy scrubbers have been
observed to have visible emissions less than 10 percent opacity. It is estimated that particulate
emissions are less from pot furnaces than from blast and reverberatory furnaces, but no tests have
yet been conducted.
ENVIRONMENTAL IMPACT OF PROPOSED STANDARDS
No significant quantity of solids will require disposal as a result of implementing the proposed
standard since, in most instances, the collected solids are lead compounds that are recycled back to
the process.
The predominant control devices for the secondary lead industry are expected to be fabric
filters, along with a small number of high-energy scrubbers. Dust collected in baghouses can be
recycled directly back to the furnace. When wet scrubbers are used, settling tanks and ponds have
been installed to precipitate the collected solids. The precipitate is removed, dried, and fed back to
the furnace. Scrubbing water will pick up sulfur dioxide from the gas stream, causing the water to
become acidic. Alkali can be added to the scrubber to control pH. Salts that precipitate with
collected dust are also returned to the furnace and usually become part of the slag.
ECONOMIC IMPACT OF PROPOSED STANDARDS
At the end of 1971, there were 23 firms operating approximately 45 secondary lead smelting
plants in the United States. The four largest companies account for approximately 72 percent of
the output. Total production has been cyclical but tending upward at a yearly rate of 3.2 percent.
Consumption of lead-acid storage batteries, the major market for secondary lead, has been
growing at a rate of 5.1 percent annually. In general, the industry expects these trends to continue,
with no major problems in the forseeable future.
It is anticipated that two new secondary lead plants will be installed and one to two plants will be
modified in the United State each year. Table 6 gives estimates of control costs for two model units
representative of the size and type expected to be installed. For a new plant consisting of a blast
furnace rated at 50 tons/day with auxiliaries, two abatement alternatives were analyzed. If an
afterburner, U-tube cooler, and fabric filter were installed, the annualized control costs (including
charges for labor, materials, utilities, depreciation, interest, property taxes, and an allowance for
recovered materials) would amount to about to about $4.OS/ton of output. In the worst-case
situation (that is, if the costs could not be passed forward or backward) this level of expense would
cause a reduction in typical net earnings of approximately 15 percent. If the alternative venturi
scrubber system were installed, annualized costs would amount to approximately $6.40/ton of
output. This approach would reduce typical net earnings about 25 percent in the worst-case
situation.
For a new secondary lead plant consisting of a reverberatory furnace rated at the same capacity
and having equivalent auxiliary equipment, two similar abatement alternatives were considered. In
this case, however, afterburners need not be added to prevent blinding of the baghouse filter
material. If a U-tube cooler and fabric filter were installed, annualized costs for the control
equipment would be about $1.65/ton of product. With no ability to shift costs, this approach
would reduce typical net income approximately 7 percent.
41
-------
Table 6. CONTROL COSTS OF MEETING
PERFORMANCE STANDARD (0.022 gr/dscf)
FOR TYPICAL SECONDARY LEAD PLANTS3
Plant type
Blast furnace.
50 tons/day
Reverberatory
furnace, 50 tons/day
Required
control
equipment
Afterburner,
U-tube cooler.
fabric filter
Afterburner,
water quench,
venturi scrubber
U-tube cooler.
fabric filter
Water quench.
venturi scrubber
Control
investment,
$
157,000
123,000
188,000
125,000
Annual
cost.
$/yr
51,000
80,000
21,000
36,000
Annual cost per
unit of
production.
$/ton
4.05
6.40
1.65
2.86
Major assumptions: (1) production rate, 4,000 Ib/hr; (2) annual production, 12,500 tons;
(3) recoverable dust is recycled at a value of 2.25 cents/lb, except for reverberatory dust
recovered from fabric filters at value of 4.5 cents/lb; (4) fabric filter systems depreciated
straight-line, 15-year life; (5) venturi scrubber systems depreciated straight-line, 10-year life;
and (6) estimated average product price $320/ton.
If the alternative control system, consisting of a water quench and venturi scrubber, were
installed, the annualized control costs would be approximately $2.86/ton of output and would
reduce typical net income about 12 percent.
The costs shown in Table 6 are total in the sense that they account for complete control systems
added to new, uncontrolled plants. The incremental control costs to meet the proposed standard
beyond those required to meet the reference process weight standard are minimal. Many State and
local agencies presently have regulations for secondary lead smelters that require the same types of
dust-control equipment necessary under the proposed standards. The industry has also practiced
relatively effective control in the past in order to minimize occupational health hazards.
It is estimated that the 1967 level of control for the industry was 90 percent. New secondary lead
facilities will be introduced into a market situation in which the price of the product or the prices
paid for raw material scrap already reflect, to some degree, the increased expenses from air
pollution control Since the incremental control costs for a new plant versus an existing unit are
rmmmal, profitability at least equaling that of existing industry should be achievable for a new
Secondary producers compete with their primary counterparts and are subject to the cyclical
nature of the lead industry as a whole. Total control costs for the secondary facilities are small in
absolute terms and in relation to the expected control costs for the primary producers. Control
llTr ^mafy ^f arC °n* the °rder °f $0-022/1b, or $44/ton. Costs for new secondary
plants are n the range of $2/ton to $6/ton. With full implementation by both types of producers,
secondary lead producers should not be placed at a competitive disadvantage.
REFERENCES FOR TECHNICAL REPORT NO. 10
1. Control Techniques for Lead Emissions. U.S. Department of Health, Education and Welfare,
Public Health Service, Environmental Health Services. Washington, D.C. September 1970.
42
-------
2. Brainbridge, C.A. Fume Control and Recovery in Lead Smelting Furnaces. Chemical and
Process Engineering. August 1960.
3. Bray, J.L. Non-Ferrous Production Metallurgy (2nd Ed.). New York, John Wiley and Sons, Inc.
-I 7O*J.
4. Williamson, I.E., J.F. Nenzell, and W.E. Zwiacher. A Study of Five Source Tests on Emissions
from Secondary Lead Smelters. County of Los Angeles Air Pollution Control District. Los
Angeles, Calif. Order No. 2PO-68-02-3326. 1972. 45p.
43
-------
TECHNICAL REPORT NO. 11 -
SECONDARY BRASS OR BRONZE INGOT PRODUCTION PLANTS
SUMMARY OF PROPOSED STANDARDS
The proposed performance standards for new secondary brass or bronze ingot production plants
will limit particulate emissions (including visible emissions) from reverberatory furnaces and will
limit visible emissions from electric and blast (cupola) furnaces. The standards will apply to batch
furnaces with a capacity of 1000 kilograms (2205 pounds) or greater per heat, and to continuous
(blast) furnaces capable of producing 250 kilograms (550 pounds) or more of metal per hour. The
standards do not apply to the manufacture of brass or bronze from virgin metals or to brass or
bronze foundry operations. Furthermore, the standards apply to particulate emissions from
furnaces only. Other sources of particulate emissions may exist in plants affected by the proposed
standards, but further study will be required to delineate such sources and to recommend
appropriate levels of control.
No mass standard is proposed for electric and blast furnaces because (1) 95 percent of the
production is carried out in reverberatory furnaces, (2) the emissions from blast furnaces are about
the same as those from reverberatory furnaces and far less than those from electric furnaces, (3)
well controlled blast and electric furnaces can meet the visible emission standard, (4) the
expenditure of EPA resources for the testing needed to support a specific mass standard is not
warranted, and (5) the visible emission standard is an adequate enforcement criterion and can be
met only by well controlled units.
Standards for Particulates from Reverberatory Furnaces
The proposed standards will limit emissions to the atmosphere as follows:
1. No more than 50 mg/Nm3 (undiluted), or 0.022 gr/dscf.
2. No more than 10 percent opacity.
Standard for Particulates from Electric and Blast Furnaces
The opacity of visible emissions shall be no more than 10 percent.
EMISSIONS FROM SECONDARY BRASS AND BRONZE FURNACES
Particulate emissions from brass and bronze furnaces (Figure 21) vary with the content of the
alloy being produced, and with the presence of impurities in the scrap feed. Most of the particulate
emissions are metal oxides, predominantly zinc oxides (45 to 77 percent) and lead oxides (1 to 13
percent). Uncontrolled reverberatory furnaces can emit as much as 80 pounds of particulate
matter per ton of ingot produced. The level of emissions from blast furnaces (cupolas) is
approximately equal to that from reverberatory furnaces; the level of emissions from electric
furnaces is typically far lower.1'2'3 The composition of emissions from blast furnaces is similar to
that from reverberatory furnaces.1'2 Emissions from electric furnaces are also expected to be
similar because the process and raw materials are identical.
Fabric filters are extensively used to control emissions from all three types of furnace; only
recently have electrostatic precipitators been adopted as control devices. Although no scrubber has
yet been used to control emissions to the level of the proposed standard, such levels are within the
capability of scrubbing technology.
45
-------
EMISSIONS
REVERBATORY
FURNACE
DOMESTIC &
INDUSTRIAL SCRAP
FLUX
METAL PRODUCT CONTROL DEVICE
/TO INGOT MOLD
mm ING SURFACE FABRIC
CUULINla MJKt-AUt COLLECTOR 0R
/A\ A A A A ELECTROSTATIC
//\\ /\ /A\ AN /AN PREC,p|TATOR
FUEL
(GAS OR OIL)
AIR METAL PRODUCT TO INGOT WIOLD
Figure 21. Controlled secondary brass and bronze furnaces.
No State or local agency now has an emission standard specifically for the brass and bronze
industry. General restrictions applied to this industry are based on the process weight regulations
or on emission concentrations.4 These concentration restrictions range from 0.05 to 0.3 gr/scf.
Some restrictions are based on particulate sampling methods that differ from EPA method 5
because they include material collected in a wet impinger.
The reference process weight regulation (Table 1) would restrict emissions from a typical 25-ton
brass furnace (24-hour cycle) to 3.6 Ib/hr. The proposed mass standard is more restrictive than any
existing process weight curve for furnace sizes appropriate to the brass and bronze industry. The
standard will limit these emissions to between 1.0 and 1.5 Ib/hr.
Lead emissions during production of a 5 percent lead alloy comprised 4 to 7 percent of the total
particulate matter emitted. Production of alloys with a higher lead content would probably
increase the lead content of the total emissions. Because there is no known control technique
specific for lead, the maximum possible control of lead emissions can be obtained by using the
most effective particulate collectors.
RATIONALE FOR PROPOSED STANDARDS
Based on the results of preliminary screening, eight plants were inspected as candidates for
source testing. Visits to these plants revealed that five plants operated with no visible emissions.
Four of these five were selected for source testing, and three plants were successfully tested. The
fourth test was aborted because plant malfunctions during testing rendered the test results invalid.
Particulate Matter from Reverberatory Furnaces
All furnaces tested by EPA showed average particulate emission rates below the proposed
standard. Emission rates from the reverberatory furnaces, all controlled by fabric filters, averaged
0.001, 0.006, and 0.008 gr/dscf.
At least three heats were tested at each plant. The tests began when the first scrap'was charged
into the furnace and ended when the pouring of ingots began. The pouring phase of the heats was
not tested because none of the facilities adequately collected the emissions from this phase of the
heat. During some of the tests, individual samples were collected during different phases of the
46
-------
heats in order to determine fluctuations of emissions during the heat. The EPA data points in
Figure 22 represent plant emission levels determined by averaging the data acquired from the
individual samples. Four of the 31 tests were aborted because of sampling irregularities or process
upsets. These samples were not used to determine furnace emissions.
Results of other tests performed by Federal, State, and local agencies showed emission rates of
0.002, 0.005, 0.010, 0.0125, 0.014, and 0.017 gr/dscf from reverberatory furnaces. The furnaces
tested, which ranged in production capacity from 7.5 to 100 tons, were all controlled by fabric
filters.
During EPA tests at plants A and B (Figure 22), there were no visible emissions. At plant D,
visible emissions of 10 percent opacity were observed during the fabric filter cleaning cycle. No
visible emissions were reported for plants E and F, which were tested previously by EPA.
Participate Matter from Blast and Electric Furnaces
Results of one blast furnace test revealed emissions of 0.013 gr/dscf; recent EPA inspections
showed that the furnace operates with no visible emissions. Although no electric furnaces have
been source-tested, it should not be difficult for electric furnaces to meet the proposed standards
for reverberatory furnaces because their process cycle is similar.
ENVIRONMENTAL IMPACT OF PROPOSED STANDARDS
Fabric filters are the primary devices used to limit emissions from brass and bronze furnaces.
Although scrubbers can be used to meet the proposed standards, it appears that new facilities will
0.04
0.03
00
cxs"
1 0.02
LU
£ o.oi
CODE METHOD NUMBER
1 - TOTAL EPA TRAIN
2 - EPA METHOD 5
4 - EPA METHOD 5;
SAMPLED AT POINT
OF AVERAGE VELOCITY
10 - CALIFORNIA TEST
MAXIMUM
AVERAGE
EPA OTHER
b-BAGHOUSE
Aj(b) A2(b) F(b) B(b) D(b) l(b) H(b) C(b) E(b) G(b)
PLANT (CONTROL EQUIPMENT)
Figure 22. Particulate emissions from secondary brass and bronze ingot production industry,
reverberatory and blast furnaces.
47
-------
continue to use fabric filters. There is now only one brass and bronze facility that uses a wet
scrubber (low efficiency). This facility utilizes settling ponds to separate collected solids and
recycles the water in a closed-loop system.
Until recently, the particulate matter collected from brass and bronze furnaces was marketable;
however, this market has been declining, and an excess of collected zinc and lead oxides now
exists. Several methods of disposal are being used, some of which are: bagging, 55-gallon drums,
open piles, and landfills. Techniques are available to prevent leaching and water contamination
due to the storage of water-soluble solid wastes. For example, it is common practice in the
chemical processing industry to use plastic-lined, watertight disposal pits in order to prevent
leaching or runoff. Since the total annual tonnages involved in the brass and bronze industry are
small (about 10,000 tons/yr for the entire industry), this disposal technique can be used if
necessary.
ECONOMIC IMPACT OF PROPOSED STANDARDS
Brass and bronze ingot production has grown at an average annual rate of 1.2 percent over the
last 10 years. Production reached a peak in 1965 and 1966 and has declined somewhat since that
time. For this reason, it is believed that excess capacity exists in the industry and few, if any, new
plants will be constructed in the next few years. It is probable, however, that some obsolete
furnaces will need to be replaced. Such replacements are expected at a rate of one or two furnaces
per year; these new furnaces will be required to comply with the new source performance
standards.
Although there are a few wet scrubbers and electrostatic precipitators in use in the industry, the
fabric filter has been the most common control device used in the past. The fabric filter will most
likely be the control device used to meet the proposed new source performance standards. Control
cost for different sizes of reverberatory furnaces are shown in Table 7.
It is possible to channel the exhaust from several furnaces into a common control system, and
thus achieve the economy of a large-scale system. The extent that this economy can be realized will
depend on the characteristics of the individual plant in which the furnace replacement is made.
The proposed standard is not likely to require expenditures above those already required by
existing State or local standards.
Table 7. CONTROL COSTS OF MEETING PERFORMANCE STANDARD
(0.022 gr/dscf) FOR REVERBERATORY FURNACES
Furnace
capacity, tons/day
20
50
75
Investment,
$
74,000
110,000
130,000
Annual cost.
$
13,000
20,070
34,300
Annual cost per
ton of product.
$
6.52
4.01
3.24
REFERENCES FOR TECHNICAL REPORT NO. 11
1. Air Pollution Aspects of Brass and Bronze Smelting and Refining Industry. National Air
Pollution Control Administration, Public Health Service, U.S. Department of Health,
Education, and Welfare. Raleigh, N.C. NAPCA Publication No. AP-58. November 1969. 63 p.
2. Air Pollution Engineering Manual. Danielson, J.A. (ed.). National Center for Air Pollution
Control, Public Health Service, U.S. Department of Health, Education, and Welfare.
Cincinnati, Ohio. PHS Publication No. 999-AP-40. 1967.
3. Compilation of Air Pollutant Emission Factors (Revised). U.S. Environmental Protection
Agency. Research Triangle Park, N.C. Office of Air Programs Publication No. AP-42.
February 1972.
4. A Compilation of Selected Air Pollution Emission Control Regulations and Ordinances.
National Center for Air Pollution Control, Public Health Service, U.S. Department of Health,
Education, and Welfare. Washington, D.C. PHS Publication No. 99-AP-43.1968.146 p.
48
-------
TECHNICAL REPORT NO. 12 -
IRON AND STEEL PLANTS
SUMMARY OF PROPOSED STANDARDS
Standards of performance being proposed for the iron and steel industry will limit emissions of
particulates, including visible emissions, from new basic oxygen process furnaces (BOPF's).
1. No more than 50 mg/Nm3 (undiluted), or 0.022 gr/dscf.
2. No more than 10 percent opacity.
The proposed standard for visible emissions is compatible with the 50-mg/Nm3 mass emission
limit. The proposed particulate limits can be achieved with high-energy venturi scrubbers or
electrostatic precipitators.
EMISSIONS FROM BASIC OXYGEN PROCESS FURNACES
In the steel industry, there are several processes that are major sources of particulate emissions if
not properly controlled. These processes include the basic oxygen process; operation of open
hearth, blast, and electric furnaces; and operation of coke ovens and sintering plants (Figure 23).
IRON ORE
COKE OVEN
LIMESTONE
BLAST
FURNACE
SLAG
v CONTINUOUS CASTING
OPEN
HEARTH
FURNACE
ELECTRIC
FURNACE
Figure 23. Iron and steel process system.
49
BILLETS
INGOTS
SOAKING
r"
-------
The proposed standards would apply only to basic oxygen process furnaces. Other pollutant
sources in this industry will be covered by standards to be developed at a later date.
The BOPF is a vertical cylindrical container that is open at one end. During the steel-making
process, oxygen at high velocity is directed at the surface of the molten mix, violently agitating the
mix and causing a large quantity of particulate matter and carbon monoxide to be emitted through
the open end of the furnace. As much as 40 pounds of particulate matter is emitted per ton of
steel. The emissions are drawn into a hood, which is similar to that used with kitchen stoves to
draw off steam and cooking odors. From the hood, the hot, dirty air is ducted to cleaning devices,
usually electrostatic precipitators or high-pressure venturi scrubbers, which remove much of the
particulate matter before the air is vented to the outside.
There are two different hood systems used to capture the BOPF emissions. One system uses an
open (or combustion) hood, which has 1.5 to 2 feet of clearance above the furnace rim. The other
system uses a retractable closed hood, which fits rather closely around the top of the furnace and
prevents additional air from being drawn into the exhaust system.
The closed hood was designed to minimize the exhaust volume and to reclaim carbon monoxide.
In many countries, this carbon monoxide is collected for use as a fuel or as a feed gas for
petrochemical processing operations; however, in the two plants in the United States using closed
hoods, the exhaust gases are currently flared with no heat recovery.
From an air pollution standpoint, there are two factors pertinent to closed hoods: (1) the high
concentrations of combustible carbon monoxide make the hot gases potentially too hazardous to
clean in the arcing electric field of an electrostatic precipitator (in the open hood, oxygen in the air
reacts with carbon monoxide to form nonexplosive carbon dioxide) and (2) the rate of the
volumetric flow (cubic feet of gas per minute) through the cleaning system and out the stack is less
than 20 percent of the rate of flow in an open-hood system. The first factor limits the choice of
cleaning equipment to a single type, the high-energy venturi scrubber. The second factor leads to
lower stack emissions per unit time (pounds per hour) than with an open hood. This is true because
the venturi scrubber achieves about the same degree of cleanliness (0.02 grain of particulate matter
per cubic foot of air) whether it is fed extremely dirty air or moderately dirty air. (The extremely
dirty air from the closed hood comes out just as clean as the moderately dirty air from the open
hood.) The amount of particulate matter coming out of the stack per unit time (pounds per hour) is
dependent, therefore, upon how many cubic feet of air come out of the stack per unit time. As
previously mentioned, the air flow in a closed system is less than one-fifth that of the open system,
and the emission of particulate matter would be correspondingly lower.
Both open and closed hoods allow some air contaminants to escape through the roof ventilators
to the atmosphere during charging, turn down, tilting, tapping, and ladle additions. Because the
closed hood may be withdrawn to the up position during these operations, it is less efficient in
collecting resultant emissions. During the oxygen blow, a small portion of the particulate matter
also escapes to the building ventilation system, regardless of the collection device used.
Collectively, these uncaptured emissions are estimated to be only a small percentage of the total
quantity from BOPF's.
In the United States, BOPF's range from 100 to 325 tons in capacity. Emission volumes vary
from 200,000 to 600,000 dscfm for open-hood systems. A typical 250-ton furnace has a gas volume
of 200,000 to 500,000 dscfm. With a 90 percent yield and a particulate concentration of 0.022
gr/dscf, the furnace would produce 470 tons of steel and emit between 36 and 90 pounds of
particulate per hour, depending on the quantity of excess air permitted to enter the combustion
hood.
The requirements of existing State and local regulations that are specifically for BOPF facilities
range from 0.1 to 0.2 lb/1000 Ib of stack gas. This limitation is equivalent to 0.045 to 0.090 gr/scf.
Such regulations would permit the furnace in the example given above to emit 77 to 386 pounds of
particulate per hour. State limitations submitted pursuant to Section 110 of the Clean Air Act will
require control only slightly less stringent than the new source standard.
50
-------
RATIONALE FOR PROPOSED STANDARDS
f °f iV^1, comPanies' which ^rate 26 of the 36 BOPF facilities in the
actual emd d ^ U T* ^ controlled P^nts. from these 12, five were chosen for
S fnn £ T I 8 m * t0 °btain data °" a wide ran8e of furnace sizes <140 to 325 tons)
and on the three basic types of emission control systems. The control systems are: the open hood
* "ei sfrub,bf ' the °Pen hood with an electrostatic precipitator (both illustrated in
P ,, , ' cosac prec
Hgure 24), and the closed hood with a high-energy scrubber (Figure 25)
WATER
OXYGEN LANCE
COOLING
CHAMBER
ELECTROSTATIC
PRECIPITATOR
AA.
DUST VENTURISCRUBBER
EMISSIONS
STACK
SLUDGE
COMBUSTION
AIR
WATER
HOT METAL
SCRAP
RECYCLED WATER
\ .
SLAG
STEEL
SLUDGE
FAN
MIST ELIMINATOR
WATER TREATMENT PLANT
Figure 24. Controlled basic oxygen furnace, open hood with scrubber or electrostatic
precipitator.
.OXYGEN LANCE
EMISSIONS
TO GAS STORAGE
IF APPROPRIATE
WATER TREATMENT PLANT
1 RECYCLED WATER
r*
Figure 25. Controlled basic oxygen furnace, closed ho6d with scrubber.
51
-------
Three of the five plants tested had average particulate emissions below the proposed standard.
Figure 26 shows the results of the tests. Plants A and B were equipped with closed hoods and high-
energy venturi scrubbers. Test ^ is a second test of the same facility 2 months after test A2. In
plants C and E, open hoods and electrostatic precipitators were used. Plant D was equipped with
an open hood and a venturi scrubber. (Emissions from plant D included some particulate matter
that was formed from supplementary fuel oil that was burned in the hood to provide a more
uniform heat source for generation of steam in the hood cooling coils.)
0.05
0.04
c/>
•a
M
0.03
<
o
0.02
0.01
0
CODE METHOD NUMBER
2 - EPA METHOD 5
MAXIMUM
AVERAGE
EPA
v VENTURISCRUBBER
h- COMBUSTION HOOD
e - ELECTROSTATIC PRECIPITATOR
g - CLOSED HOOD
E(he)
Aj(gv) A2(gv) B(gv) C(he)
PLANT (CONTROL EQUIPMENT)
Figure 26. Particulate emissions from iron and steel industry, basic oxygen process furnaces.
A series of three runs comprised a test of a BOPF, each run lasting approximately 2 hours, long
enough to include from four to six heats. Two of the three runs performed on plant D were invalid
because rupture of the filters prevented accurate calculation of particulate concentration. Of the
remaining 16 runs at the five facilities, 14 showed emissions of less than 0.022 gr/dscf.
The length of the sampling period was designed to permit measurement of all emissions
controllable with existing technology. By beginning the test immediately after the furnace was
charged and ending it immediately prior to tapping, it was possible to use data on emission from
the preheat, oxygen blow, and all reblows in preparing the standard.
Test results show that the proposed concentration standard of 0.022 gr/dscf is representative of
the lowest particulate concentration that can be achieved by control devices for BOPF emissions.
52
-------
They also reveal that the closed hood, which prevents induction of ambient air into the hood,
minimizes the mass emission rate of particulate matter from the process. Designers and manu-
facturers of control equipment will guarantee efficiencies that will achieve an average outlet
concentration of 0.020 gr/dscf from either open- or closed-hood collection devices.
Existing open-hood systems are characterized by widely varying gas flow rates. For instance, an
open-hood system applied to a 250-ton furnace might handle as little as 150,000 dscfm or as much
as 500,000 dscfm, depending on the control equipment and operating practices of the particular
firm. The lowest gas flows are used with scrubbers and the highest with precipitators, where there
is a serious explosion hazard. If the technology were sufficiently developed, the regulation would
include limits on exhaust gas flow rates in open-hood systems, thereby further restricting mass
emissions into the atmosphere. Because of explosion hazards, such a secondary limitation is not
practical at this time if dry collectors (electrostatic precipitators and fabric filters) are to be a viable
control option. Nonetheless, economic considerations will dictate that operators hold exhaust gas
rates to the minimum compatible with their systems. Both capital and operating costs of control
equipment are significant and are proportional to the gas volume handled.
The proposed standard will permit industry to utilize either the open or closed hood for future
installations, even though the closed hood provides better air pollution control. The decision to
propose this less stringent standard was made only after intensive investigation into the
consequences of a standard that would require the closed-hood system on all new steel facilities.
Several of these consequences are considered in the following paragraphs.
Manganese is used in all steel manufactured by the BOPF process to improve the fluidity of slag,
which reduces splashing and permits increased production rates. Those BOPF's controlled by
closed hoods, however, require higher manganese levels. The closed hood has minimum clearnace
between the hood, furnace, and removable oxygen lance; and slag that splashes on the lance or the
hood-furnace juncture and solidifies will halt production. A requirement that would necessitate
increased use of manganese, a strategic raw material essential to the national defense but available
within the continental United States in only limited quantities, would be undesirable.
If the closed-hood system were used on all new steel production facilities in the United States,
the nation's capability to recycle scrap steel would diminish. Contamination in poor (dirty) grades
of scrap causes excessive splashing, which the closed hood cannot tolerate. Furthermore, low-grade
scrap contains appreciable grease, paint, and other contaminants. Some of these materials are
burned with the carbon monoxide in flares or boilers; however, a portion of the hydrocarbon
emissions escapes from the closed hood to become either an air or water pollution problem.
The closed-hood system tested by EPA is of Japanese design and patent. A single U.S. company
has been licensed to market the system. Although other systems are available, none was tested and
all are of foreign design. The implication of having a foreign supplier and the associated adverse
economic effects for the United States were considered.
Routine maintenance of the closed-hood system is far more expensive than that for the open-
hood system. Since the closed system is designed to prevent intrusion of dilution air, even simple
repairs can become complex and time consuming, often requiring arc-cutting and rewelding of
connections that in the open-hood system are merely bolted together.
In a facility where an open-hood system is used, the capital cost for installation of a third
furnace controlled by a closed-hood system is several million dollars more than would otherwise be
required. Thirty percent of the existing BOPF shops were designed to accommodate a third
furnace at some future date when steel demand would justify the investment. A new open hood can
normally be manifolded to the existing control device. A closed-hood installation, however, would
require a hood, ducting fans, and new control equipment at a premium of $7 million to $8 million.
Some steel facilities that cannot physically accommodate the high vertical profile required by the
retractable closed-hood system would also need building modifications that could cost up to $30
million. The proposed standard, which will allow the existing control device to be improved to
service the third furnace, will result in a reduction of emissions from the older vessels even though
they are not subject to the standard.
53
-------
ENVIRONMENTAL IMPACT OF PROPOSED STANDARDS
Most new units are expected to utilize scrubbers to meet the standard. In typical installations,
including all three scrubbing systems tested during the EPA development program, water is
recirculated. Two to 10 percent purge may be necessary (40 to 200 gal/min). This purge can be
treated in the plant water treatment facilities with existing techniques. Guidelines for water
effluents from steel plants are currently under preparation.
At the present time, the high zinc content of the collected dust prevents it from being used as
blast furnace charge. It is possible that technology may be developed to enable the dust to be
recycled, but at the present time it is landfilled. The material may be in the form of sludge or dust,
or it may be pelletized. Landfill sites are usually segregated and mapped for possible future
reclamation.
Contamination of ground water by BOPF dust disposal sites has not been identified as a
problem. The dust is primarily iron and zinc oxide and calcium fluoride precipitated from the
scrubber solution through lime addition. None of these materials has significant solubility in
water.
ECONOMIC IMPACT OF PROPOSED STANDARDS
At the end of 1971, there were 36 basic oxygen steel furnace facilities in the United States, owned
by 19 different companies. Of a total of 120 million tons of new steel produced that year, these
facilities accounted for 64 million tons. Only three of the major integrated iron and steel firms do
not utilize the basic oxygen furnace steel facilities.
It has been estimated that approximately 8 million tons of additional capacity will come on-
stream between 1974 and 1977. This projection is based on two factors: (1) an expected growth rate
of 4.5 percent in raw steel production by the BOPF process and (2) a recovery from the 1971
production-to-capacity ratio of 86 percent to the historic ratio of 92 percent. At present, it is not
known how many new facilities will be constructed or how many existing two-vessel facilities will
add a third vessel with an open-hood control device.
Three types of control systems can meet the proposed regulations:(l) open hood with scrubber,
(2) open hood with precipitator, and (3) closed hood with scrubber. Costs of controlling particulate
emissions from new two-vessel facilities are shown in Table 8. These costs cover gas-cleaning
devices, hood, duct work, cooling towers (for open-hood scrubbers only), fans, pumps, motors,
slurry settlers and filters (for scrubbers), and dust-removal and storage equipment (for
precipitators).
Many states formulating plans for air quality implementation are developing particulate
standards that limit emissions from steel furnaces to 0.03 to 0.05 gr/dscf. These values are fairly
close to current industry performance for all BOPF shops. Meeting the new source performance
standards would not increase costs over the requirements of the current industry practice of
installing electrostatic precipitators at a new plant. Employing an open-hood scrubber to meet the
performance standards at a new plant would increase costs about $0.10/ton more in comparison
with current industry practice in BOPF shops using such control devices. The difference is due to
increased power consumption. This cost penalty is negligible compared with a price of $220/ton of
finished steel products for a typical mill product mix.
Plants expanding from two- to three-vessel facilities may be required to incorporate increased
cleaning capability into their operations, either with (1) larger fans and bigger motors (including
cooling towers for those facilities that do not have them) for scrubbers or (2) additional cleaning
sections in precipitation systems. It is expected that an individual shop with two 200-ton vessels
may spend up to $1 million to upgrade the existing control equipment to meet the proposed
performance standards covering the third vessel. It seems that this same investment may be
required to comply with State regulations as proposed in the implementation plans, especially
where expansion of facilities is concerned.
54
-------
Table 8. CONTROL COSTS OF MEETING PERFORMANCE STANDARD
(0.022 gr/dscf) FOR TYPICAL NEW TWO-VESSEL BASIC
OXYGEN PROCESS FURNACES3
Plant size,
tons/melt
140
250
Required
control
equipment
Open hood.
scrubber
Open hood.
ESPb
Closed hood,
scrubber
Open hood.
scrubber
Open hood.
ESP
Closed hood.
scrubber
Control
investment,
$
5,700,000
5,900,000
6,800,000
7,400,000
8,000,000
8,400,000
Annual cost.
$/yr
1,950,000
1,500,000
2,140,000
2,750,000
2,000,000
2,800,000
Annual cost per
unit of
production.
$/ton
1.52
1.17
1.67
1.20
0.89
1.22
aMajor assumptions: (1) production of
straight-line depreciation.
bESP-electrostatic precipitator.
140 tons/melt = 2,300,000 tons/yr; (2) 18-year
The standard should not impede conversion of existing open hearth furnaces to basic oxygen
steel production. The $1 million cited above for upgrading controls amounts to 5 percent of the
total investment required to add a third vessel to an existing facility, and open hearth furnaces will
likely require comparable control investment to comply with a State implementation plan.
This standard should not prove a deterrent to growth in raw steel production nor to conversion
of open hearth facilities. With such minimal cost penalties, profit margins should not be affected
by the standard.
REFERENCES FOR TECHNICAL REPORT NO. 12
1. A Systems Analysis Study of the Integrated Iron and Steel Industry. Battelle Memorial
Institute. Columbus, Ohio. Contract No. PH 22-68-65. May 1969.
2. Iron and Steel Industry. Environmental Engineering Incorporated and Herrick Associates.
Gainesville, Fla. CPA 70-142, Task O, No. 2. March 1971.
55
-------
TECHNICAL REPORT NO. 13 -
SEWAGE TREATMENT PLANTS
SUMMARY OF PROPOSED STANDARDS
Standards of performance being proposed for municipal sewage treatment plants would limit
emission of particulate matter (including visible emissions) from new incinerators used to burn
sludge generated in the treatment plant. These standards would apply to all sewage treatment
plants that incinerate sludge from primary or secondary treatment. For plants processing
industrial wastewaters, further restriction might be required to prevent the release of specific
metals, toxic, organics, or radioactive substances.
The proposed standards would limit particulate emissions to the atmosphere as follows:
1. No more than 70 mg/Nm3 (undiluted), or 0.031 gr/dscf.
2. No more than 10 percent opacity.
The proposed visible emission standard is compatible with the mass emission limit; if
particulate emissions are below 70 mg/Nm3, visible emissions will be less than 10 percent opacity.
EMISSIONS FROM SLUDGE INCINERATORS
Sludge incinerators (Figures 27 and 28) differ from most other types of incinerators, primarily in
that the refuse does not supply enough heat to sustain combustion. Further, there is less emphasis
on retaining ash in the incinerator and much of it is discharged in stack gases. In one type of
incinerator, the fluidized bed reactor, all of the ash is carried out with the gases. Particulate
emissions into the atmosphere are almost entirely a function of the scrubber efficiency and are only
minimally affected by incinerator conditions. All sludge incinerators in the United States are
equipped with scrubbers of varying efficiency; the scrubbers range from simple bubble-through
units to venturi scrubbers with pressure drops of up to 18 inches of water.
Available data indicate that, on the average, uncontrolled multiple-hearth incinerator gases
contain about 0.9 gr/dscf of particulate matter. Uncontrolled fluid bed reactor gases contain about
8.0 gr/dscf. For average municipal sewage sludge, these values correspond to about 23 Ib/hr in a
multiple-hearth unit and about 205 Ib/hr in a fluid bed unit. Particulate collection efficiencies of
96.6 to 99.6 percent will be required to meet the standard, based on the above uncontrolled
emission rate. Emissions will be on the order of 1.0 Ib/hr.
Existing State or local regulations tend to regulate sludge incinerator emissions through
incinerator codes or process weight regulations. The most stringent State or local limit, 0.03
gr/dscf, is based on a test method that is different from the reference method in that it includes
impingers.1 Many State and local standards are corrected to a reference base of 12 percent carbon
dioxide or 6 percent oxygen. Corrections to carbon dioxide or oxygen baselines are not directly
related to the sludge incinerator rate because of the high percentage of auxiliary fuel required. In
some regulations, the carbon dioxide from fuel burning is subtracted from the total in
determinations of compliance.
For a typical incinerator with a rated dry solids charging rate of 0.5 ton/hr at a gas flow rate of
3000 dscfm, the proposed standard would allow the incinerator to emit 0.8 Ib/hr of particulate
matter. The reference process weight regulation (Table 1) would limit emissions to 6.3 Ib/hr, based
on a charging rate of wet sludge (80 percent water) of 5000 Ib/hr. Dry solids charging rates for new
incinerators will range from 0.5 to 4.0 tons/hr, with gas flow rates of 1,000 to 20,000 dscfm.
57
-------
SLUDGE
CAKE
SHAFT COOLING
AIR FAN
Figure 27. Controlled multiple-hearth furnace, scrubber.
EMISSIONS,
ASH PIT
f^\ l—l
AIR BLOWER AIR BLOWER
Figure 28. Controlled fluidized bed reactor, scrubber.
RATIONALE FOR PROPOSED STANDARDS
Preliminary investigations revealed the location of 30 reportedly well controlled sewage sludge
incinerators. These plants were visited, and information was obtained on the process and control
equipment. At 15 of the plants, visible emissions were observed to be less than 10 percent opacity.
58
-------
Determination was made as to the feasibility of stack testing in each case. Stack tests were
conducted at five locations, including three multiple-hearth incinerators and two fluid bed
reactors. Four incinerators tested were controlled by impingment-type scrubbers, and one by a
venturi scrubber. Pressure drops across the scrubbers range from 2.5 to 18 inches of water.
Of the incinerators tested, one fluid bed reactor and one multiple-hearth incinerator showed
particulate emissions at or below the proposed standard (Figure 29). Particulate emissions
averaged 0.010 and 0.030 gr/dscf, respectively. A previsous test by a local control agency2 using
the reference method on the fluid bed reactor (Figure 29) indicated average emissions of 0.009
gr/dscf. The other multiple-hearth incinerators tested had erroneously low exit particulate
concentrations as a result of dilution by shaft cooling air prior to sampling. Estimated undiluted
exit concentrations (Figure 29) are 0.050 and 0.055 gr/dscf. Emission from the second fluid bed
reactor (Figure 29) averaged 0.060 gr/dscf.
The fluid bed reactor on which the standard is based is controlled by a venturi scrubber with a
pressure drop of 18 inches of water. Because of the limited application of this type of control device
0.10
0.08
, 0.06
3
o
a:
a.
0.04
0.02
CODE METHOD NUMBER
2 EPA METHOD 5
8 - STATE OF NEW JERSEY TEST
MAXIMUM
AVERAGE
" EPA OTHER
s-SCRUBBER
p- PLATE SCRUBBER
* - CORRECTED TO ACCOUNT
FOR DILUTED AIR
Al(s)
~A2(s) B(p)* C(p) D(p)
PLANT (CONTROL EQUIPMENT)
E(s)*
Figure 29. Particulate emissions from sewage treatment plant, sludge incinerator.
59
-------
on sludge incinerators, the standard has been set at a level somewhat higher than that obtained
during the tests of the unit. The remaining installations tested had impingement-type scrubbers,
which operated at considerably less pressure drop (2.5 to 6 inches of water). The lower-efficiency
impingement scrubbers are adequate to meet opacity and process weight regulations, but do not
represent the best control technology. Designers and manufacturers of control equipment will
guarantee an outlet concentration of less than 0.030 gr/dscf.
No visible particulate emissions were observed at the five incinerators tested, although moisture-
condensation plumes were sometimes present. Ten additional incinerators were observed by EPA
engineers to have visible emissions that were within the proposed standard.
ENVIRONMENTAL IMPACT OF PROPOSED STANDARDS
Incineration consumes hydrocarbons and reduces the volume of solid wastes by up to 95 percent.
Incinerated sludge is usually acceptable for landfill. If raw sludge is not incinerated, it must be
digested or otherwise stabilized to render it acceptable for landfill. Water from the plant is used as
the scrubbing medium and recycled to treatment facilities. In no case is new solid or liquid waste
created.
For municipal treatment plants, the combination of high-temperature incineration and high-
efficiency scrubbing will provide sufficient safeguards against the release of highly toxic air
pollutants. Nevertheless, this treatment may not be adequate for industrial installations where
there are significant concentrations of mercury or other toxic materials in the sewage. In such
instances, other means of sludge handling and disposal should be evaluated.
ECONOMIC IMPACT OF PROPOSED STANDARDS
Over the next few years, it is estimated that 70 new municipal sewage sludge incinerators will be
constructed annually in the United States. Factors such as the availability of alternative methods
of sludge disposal will have a significant effect on the actual rate of construction.
To estimate the economic impact of the proposed new source performance standards, a model
sewage sludge incinerator (multiple-hearth furnace) serving a population of 100,000 persons was
utilized. Investment and annual cost to achieve the proposed standard were estimated. To provide
a basis for cost comparison, investment and annual costs to comply with a process weight standard
for the incinerator were also estimated. Table 9 gives the results of these analyses. Cost information
is based upon private communication with manufacturers of sludge incinerators and
manufacturers of air pollution control equipment.
Investment costs in air pollution control equipment (low-energy impingement scrubbers) to meet
the process weight standard^were found to be approximately 4.0 percent of the total installed cost
of the sludge incineration facility. The control cost (for a low-energy venturi scrubber) to achieve
the proposed new source performance standard represents approximately 4.3 percent of the total
installed cost. The increase in installed cost from 4.0 to 4.3 percent is due primarily to the
additional fans and motors required for the venturi scrubber.
Annual costs to meet the process weight standard were estimated to be 4 percent of the total
annual cost of the sludge incinerator facility. The annual cost of control to comply with the
proposed new source standard is estimated to be 6 percent of the total annual cost of the
incinerator facility. Increases in the power requirements of the venturi scrubber were found to be a
major cause for the increases in annual cost of control. On a per capita basis (population of
100,000 persons), meeting the proposed new source performance standard is estimated to cost
$0.04/year more than a process weight standard of 0.10 gr/scf.
In financing the required investment, municipalities have several alternatives, such as issuing
bonds or securing money through pledges of ad valorem tax revenues. The proposed new source
performance standard is not anticipated to cause additional difficulties.
60
-------
Table 9. CONTROL COSTS FOR TYPICAL SEWAGE
SLUDGE INCINERATOR3
Plant size,
tons/day
(cfm)
10
(10,000)
100
(17,800)
Emission
standard
Performance
standard =
0.031 gr/dscf
Typical local
standard =
0.10 gr/dscf
Performance
standard =
0.031 gr/dscf
Typical local
standard =
0.10 gr/dscf
Required control
equipment
Low-energy
venturi scrubber
Low-energy
impingement
scrubber
Low-energy
venturi scrubber
Low-energy
impingement
scrubber
Control
investment,
$
60,000
55,000
132,000
120,000
Annual cost,
$/year
11,700
8,400
34,200
21,100
Annual cost per
person, $
0.12
0.08
0.03
0.02
aModel plant assumptions: (1) 10 tons/day—3640 hours of operation per year, 100 tons/day—
8736 hours of operation per year; (2) sinking fund depreciation over 12.5 years; and (3) interest
at 6 percent.
REFERENCES FOR TECHNICAL REPORT NO. 13
References Cited
1. Implementation Plan, Metropolitan Baltimore Intrastate Air Quality Control Region. State of
Maryland, Department of Health and Mental Hygiene, Environmental Health Administration.
Baltimore, Md. January 28,1972. (addendum April 4,1972).
2 Test Report on the N.W. Bergin Sewage Authority Sludge Incinerator for the State of New
' Jersey. Engineering-Science, Inc. Washington, D.C. Contract No. 68-02-0225. May 1972.
Supplemental References
3. Burd, R.S. A Study of Sludge Handling and Disposal. Ohio Basin Region, Federal Water
Pollution Control Administration. Cincinnati, Ohio. FWPCA Publication No. WP-20-4. May
1968.
4. State of the Art Review on Sludge Incineration Practice. Resource Engineering Associates.
Wilton, Conn. Contract No. 14-12-499. April 1970.
61
-------
Tule and Subtitle BACKGROUND INFORMATION FOR PROPOSED NEW SOURCE
PERFORMANCE STANDARDS: Asphalt Concrete Plants, Petroleum Refin-
eries, Storage Vessels, Secondary Lead Smelters and Refineries,
Brass or Bronze Ingot Production Plants. Iron and Steel Plants.
BIBLIOGRAPHIC DATA
SHEET
1. Report No.
APTD-1352a
3. Recipient's Accession No.
5- Report Date
June' 1973
Author(s)
Sewage Treatment Plants; Volume 1, Main
Text
8. Performing Organization Kept.
No.
Performing Organization Name and Address
U.S. Environmental Protection Agency
Office of Air and Water Programs
Office of Air Quality Planning and Standards
Research Triangle Park. North Carolina 27711
10. Project/Task/Work Unit No.
11. Contract/Grant No.
2. Sponsoring Organization Name and Address
13. Type of Report & Period
Covered
14.
15. Supplementary Notes
sttactsThis document provides background information on the derivation of the proposed
second group of new source performance standards and their economic impact on the con-
struction and operation of asphalt concrete plants, petroleum refineries, storage vessels
secondary lead smelters and refineries, brass or bronze ingot production plants, iron
and steel plants, and sewage treatment plants. Information is also provided on the en-
vironmental impact of imposing the standards. The standards require control at a level
typical of well controlled existing plants and attainable with existing technology. To
determine these levels, extensive on-site investigations were conducted, and design fac-
tors, maintenance practices, available test data, and the character of emissions were
considered. Economic analyses of the effects of the standards indicate they will not
cause undue reductions of profit margins or reductions in growth rates.
17. Key Words and Document Analysis. 17o. Descriptors
Air Pollution
Pollution control
* Performance standards
* Asphalt concrete plants
* Petroleum refineries
* Lead smelters and refineries
* Brass ingot production
* Bronze ingot production
* Iron production
17b. Identifiers/Open-Ended Terms
* Air pollution control
* Steel production
* Sewage treatment
17c. COSATI Field/Group
18. Availability Statement
Unlimited
19. Security
Report)
UNCl
ASS1FIEP
20. Security Class (This
Page
UNCLASSIFIED
61
22. Price
FORM NT1S-35 (REV. 3-72)
------- |