United States Office of Air Quality EPA-450/3-87-024
Environmental Protection Planning and Standards September 1987
Agency Research Triangle Park NC 27711
Fossil and Nonfossil
Fueled-Fired
Industrial Boilers —
Background
Information for
Promulgated S02
Standards
Volume 4
-------
EPA-450/3-87-024
Fossil and Nonfossil Fuel-Fired
Industrial Boilers — Background
Information for Promulgated S02
Standards Volume 4
Emission Standards Division
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
September 1987
-------
This report has been reviewed by the Emission Standards and Engineering Division of the Office of Air Quality
Planning and Standards, EPA, and approved for publication. Mention of trade names or commercial products is
not intended to constitute endorsement or recommendation for use. Copies of this report are available through
the Library Services Office (MD-35), U.S. Environmental Protection Agency, Research Triangle Park N C27711,
or from National Technical Information Services, 5285 Port Royal Road, Springfield VA 22161.
-------
TABLE OF CONTENTS
Section Page
1.0 SUMMARY 1
1.1 SUMMARY OF CHANGES SINCE PROPOSAL 1
1.2 SUMMARY OF IMPACTS OF PROMULGATED ACTION 2
2.0 SUMMARY OF PUBLIC COMMENTS 7
2.1 NEED FOR THE STANDARD 24
2.2 APPLICABILITY 26
2.3 BASIS OF THE STANDARD 41
2.4 STANDARD FOR S02 49
2.5 STANDARD FOR PARTICULATE MATTER 66
2.6 NATIONAL IMPACTS 69
2.7 COST OF THE STANDARD 104
2.8 PERFORMANCE/RELIABILITY OF DEMONSTRATED TECHNOLOGIES 141
2.9 INDUSTRY-SPECIFIC ECONOMIC IMPACTS 155
2.10 SECONDARY ENVIRONMENTAL IMPACTS 161
2.11 REGULATORY IMPACT ANALYSIS 178
2.12 MIXED FUEL-FIRED STEAM GENERATING UNITS 183
2.13 STANDARD FOR COGENERATION UNITS 194
2.14 STANDARD FOR EMERGING TECHNOLOGIES 200
2.15 MONITORING, RECORDKEEPING, AND REPORTING REQUIREMENTS 212
2.16 MISCELLANEOUS COMMENTS 229
-------
1.0 SUMMARY
On June 19, 1986, the Environmental Protection Agency (EPA) proposed
standards of performance limiting emissions of sulfur dioxide (SOg) and
particulate matter from industrial-commercial-institutional steam generating
units with heat input capacities of 29 MW (100 million Btu/hour) or larger
(51 FR 22384; Subpart Db) under authority of Section 111 of the Clean Air
Act. Public comments were requested on the proposal in the Federal
Register. There were 99 commenters, composed mainly of industries, trade
associations, and State regulatory agencies. Also commenting were U. S.
Government agencies, an environmental group, and several nonaffiliated
commenters. The comments that were submitted (see Docket A-83-27), along
with responses to these comments, are summarized in this document. In those
cases where EPA agreed with a comment, appropriate revisions to the final
standards have been made.
1.1 SUMMARY OF CHANGES IN STANDARD
In most cases, the final standards require a 90 percent reduction in
SOg emissions from coal- and oil-fired steam generating units. Maximum
emissions are limited to 520 ng/J (1.2 lb/million Btu) for coal and 340 ng/J
(0.8 lb/million Btu) for oil. These requirements are the same as those
contained in the proposed standards. For steam generating units operating
at an annual capacity utilization factor of less than 30 percent for coal
and oil (or a mixture of coal and oil), no percentage reduction is required,
but emission limits of 520 ng/J (1.2 lb/million Btu) for coal and 130 ng/J
(0.3 lb/million Btu) for oil must be met. This is a new provision not
included in the proposed standards. An exemption from the percentage
reduction requirement has also been added for steam generating units located
in noncontinental areas and firing very low sulfur oil [(130 ng/J (0.3
lb/million Btu)]. In addition, fluidized bed combustion steam generating
units firing coal refuse are allowed to achieve 80 percent reduction in SOg
1
-------
emissions, subject to the same emission rate limit as for other coal-fired
units. The requirements for emerging SO2 control technologies; particulate
matter emissions from oil-fired steam generating units; and monitoring,
recordkeeping, and reporting requirements remain the same as those proposed.
1.2 SUMMARY OF IMPACTS OF PROMULGATED ACTION
1.2.1 Alternatives to Promulgated Action
Regulatory alternatives are discussed in Chapter 8 of "Summary of
Regulatory Analysis for New Source Performance Standards: Industrial-
Commercial -Institutional Steam Generating Units of Greater than 100 Million
Btu/hr Heat Input" (EPA-450/3-86-005), referred to as the Summary of
Regulatory Analysis.
1.2.2 Impacts of Promulgated Action
The proposed impacts associated with the final standards are summarized
in Tables 1 and 2. Table 1 summarizes national impacts and Table 2
summarizes impacts on typical steam generating units.
Projected national impacts associated with the standards can vary
considerably depending on the approach used to estimate these Impacts. The
approach used by the Agency starts with estimates of the growth in national
energy consumption projected by the Department of Energy. These projections
are used to estimate energy consumption in new industrial-commercial-
institutional steam generating units. These energy consumption estimates,
along with projections of future fuel prices, serve as Inputs to a computer
model known as the "Industrial Fuel Choice Analysis Model" (i.e., IFCAM).
With these input assumptions, IFCAM projects a population of new steam
generating units distributed by geographic area, unit size, and operating
level based on historical patterns. For each projected new steam generating
unit, the total cost associated with each type of fuel that could be fired,
including the costs to comply with standards limiting S02 emissions is
calculated and the lowest cost alternative selected for compliance. The
2
-------
TABLE 1. NATIONAL IMPACTS
AIR EMISSION REDUCTIONS
Sulfur Dioxide, thousand tons/yr
Particulate Matter, thousand tons/yr
IFCAM
130 - 360
9 - 24
Sales Data
140
3
LIQUID AND SOLID WASTES GENERATED
Liquid Wastes, million ft /yr
Solid Wastes, thousand tons/yr
neg.
neg.
53
270
ENERGY IMPACTS
Increase in Natural Gas Use, trillion Btu/yr
110 - 310
-
COST IMPACTS
Total Annualized Costs, million $/yr
Average Cost Effectiveness, $/ton
Incremental Cost Effectiveness, $/ton
5 - 50
40 - 130
0
120
890
1,400
TABLE 2. TYPICAL STEAM GENERATING UNIT IMPACTS3
AIR EMISSION REDUCTIONS
Sulfur Dioxide, tons/yr
Particulate Matter, tons/yr
Smalla
1,200
54
Urqeb
780
0
LIQUID AND SOLID WASTE GENERATED
Liquid Wastes, million ft /yr
Solid Wastes, thousand tons/yr
1.3
2.7
COST IMPACTS
Total Annualized Cost, thousand $/yr
Average Cost Effectiveness, $/ton
Incremental Cost Effectiveness, $/ton
860
750
1,900
920
1,200
1,200
Based on a 150 MM Btu/hr oil-fired steam generating unit firing a low
sulfur oil, operating at 55 percent annual capacity factor, and using
a sodium FGD system to reduce S02 emissions, compared to a unit subject
to SOg emissin limits under a typical State Implementation Plan.
bBased on a 400 MM Btu/hr coal-fired steam generating unit firing a
medium sulfur coal, operating at 60 percent capacity factor, and using
a lime spray drying FGD system to reduce SOg emissions, compared to a
unit subject to Subpart D.
3
-------
results are then aggregated to yield estimates of national impacts
associated with standards limiting SOg emissions.
Using this approach, "fuel switching" from coal or oil to natural gas
can occur in response to standards limiting SO2 emissions. For example, in
the absence of SO2 standards, it may be less expensive to fire coal or oil
than natural gas. With SO2 standards, however, it may be less expensive to
fire natural gas than to fire coal or oil. When fuel switching occurs, it
results in lower costs and greater SO2 emission reductions.
The type of fuel projected by IFCAM to be fired in each new steam
generating unit as well as the likelihood of fuel switching occurring in
response to standards, depends primarily on relative fuel prices. Given the
uncertainty in projected fuel prices, a number of different fuel price
scenarios were examined. The range of national impacts associated with
those projections of fuel prices which are currently considered "most
likely" is shown under the "IFCAM" column in Table 1.
Impacts of the final standards on typical steam generating units are
summarized in Table 2. These impacts ignore the possibility of fuel
switching and assume that a new steam generating unit which would fire coal
or oil in the absence of standards will continue to fire coal or oil,
regardless of the costs involved. The actual cost impacts would be lower
and emission reductions higher if fuel switching had been assumed to occur.
A number of commenters stated that the approach used by the Agency to
estimate national impacts (i.e., IFCAM) probably underestimated the costs
and overestimated the emission reductions associated with the standards.
Commenters suggested that both the amount of fuel switching projected to
occur as well as the number of new steam generating units projected to be
built were excessive. To respond to these concerns, an approach based on
historical data was also used to estimate national impacts (shown under the
"SALES DATA" column in Table 1).
This alternate approach is similar to that used to estimate the impacts
on typical steam generating units. It uses annual steam generating unit
sales statistics gathered by the American Boiler Manufacturing Association
(ABMA) for the past five years to project a population of new industrial -
4
-------
commercial-institutional steam generating units expected to be built in the
next 5 years. The costs and emission reductions associated with the
standards are estimated for each new coal- or oil-fired steam generating
unit and no fuel switch is assumed to occur. The results are then
aggregated to yield an alternative estimate of national impacts.
As shown in Tables 1 and 2, the final standards would result in
significant reductions in SOg emissions from new industrial-commercial-
institutional steam generating units on both a national and an individual
basis. Tables 1 and 2 show the standards could, however, result in some
increase in liquid and solid waste. The amount of liquid and solid waste
generated depends on the amount of fuel switching that occurs. Where steam
generating units fuel switch from firing coal or oil to firing natural gas,
the SO2 standards would not result in any waste generation and the standards
could, in fact, result in a net decrease in liquid or solid wastes. Where
fuel switching does not occur, the liquid or solid waste increase would
depend on the type of FGD system installed to control SOg emissions. Some
systems generate only liquid wastes, others generate only solid wastes.
Impacts on energy consumption associated with the final standards also
depend on the extent to which fuel switching occurs. At most, the standards
would result in less than a 5 percent increase in the amount of natural gas
consumed by industrial sources. Much of this increased natural gas
consumption, however, would be "balanced off" by a corresponding decrease in
oil consumption.
The national average cost effectiveness of the standards, based on
IFCAM, is projected to be in the range of $40 to $130/ton SOg removed.
Using the alternative "SALES DATA" approach outlined above, the average cost
effectiveness is projected to be $890/ton. The national incremental cost
effectiveness of the final standards over standards based on the use of low
sulfur fuels is projected to be negligible if fuel switching is assumed to
occur and up to $l,400/ton if no fuel switching is assumed to occur.
Industry specific economic impacts were assessed for six industries
which were considered likely to experience the most severe impacts. For
5
-------
these industries, product prices were projected to increase by less than
0.01 to 1.5 percent in 1990, assuming "full cost pass-through" of all
increased costs associated with standards requiring a percent reduction in
SOg emissions. Assuming "full cost absorption," return on assets was
projected to decrease by 0.03 to 2.8 percentage points.
1.2.3 Other Considerations
1.2.3.1 Irreversible and Irretrievable Commitment of Resources. Other
than the fuels required for power generation and the materials required for
the construction of the control systems, there 1s no apparent Irreversible
or Irretrievable consultment of resources associated with this regulation.
1.2.3.2 Environmental and Energy Impacts of Delayed Standards. The
results of delay 1n the standards are that new 1ndustrial-commercial-
Institutional steam generating units would be built that may not meet the
emission limitations established by these standards. This would delay the
ambient air quality and other environmental benefits associated with this
NSPS.
1.2.3.3 Urban and Community Impacts. Neither plant closures nor
impacts on small businesses are forecast. No significant adverse Impacts on
urban areas or local communities are anticipated as the result of the
promulgation of these standards.
6
-------
2.0 SUMMARY OF PUBLIC COMMENTS
A total of 99 letters commenting on the proposed standards were
received. Comments were provided by industry representatives, governmental
entities, and environmental groups. These comments have been recorded and
placed in the docket for this rulemaking (Docket A-83-27, Category IV).
Table 2-1 presents a listing of all persons submitting written comments,
their affiliation and address, and the recorded Docket Item Number assigned
to each comment letter.
In addition, 19 industry representatives presented oral comments on the
proposed standards at a public hearing held on September 4, 1986. A
verbatim transcript of the comments at the public hearing has been prepared
and placed in Docket A-83-27, Category IV. Table 2-2 presents a listing of
all persons presenting comnents at the public hearing, their affiliation and
address, and the recorded Docket Item Number assigned to the public hearing
transcript.
The comments summarized in this chapter have been organized into the
following categories: '
2.1 Need for the Standard
2.2 Applicability
2.3 Basis of the Standard
2.4 Standard for SOg
2.5 Standard for Particulate Matter
2.6 National Impacts
2.7 Cost of the Standard
2.8 Performance/Reliability of Demonstrated Technologies
2.9 Industry-specific Economic Impacts
2.10 Secondary Environmental Impacts
2.11 Regulatory Impact Analysis
2.12 M1xed Fuel-fired Steam Generating Units
2.13 Standard for Cogeneration Units
2.14 Standard for Emerging Technologies
2.15 Monitoring, Recordkeeping, and Reporting Requirements
2.16 Miscellaneous Comments
7
-------
TABLE 2-1. LIST OF COMMENTERS ON THE PROPOSED STANDARDS FOR
SULFUR DIOXIDE EMISSIONS FROM INDUSTRIAL-COMMERCIAL-INSTITUTIONAL
STEAM GENERATING UNITS
Docket
Commenter Reference
James K. Hambright D-l
Director, Bureau of A1r Quality Control
Pennsylvania State Air Pollution Control Agency
200 N. Third Street
P. 0. Box 2063
Harrlsburg, PA 17120
Bruce Blanchard D-2
Director, Environmental Project Review
U.S. Department of the Interior
Office of the Secretary
Washington, DC 20240
James E. Wilmoth D-3
Manager, Marketing
Combustion Engineering, Inc.
Environmental Systems Division
31 Inverness Center Parkway
P. 0. Box 43030
Birmingham, AL 35243
T. A. Alspaugh D-4
Manager, Water & A1r Resources
Cone Mills Corporation
Greensboro, NC 27405
John E. Pinkerton D-5
A1r Quality Program Manager
National Council of the Paper Industry
for Air and Stream Improvement, Inc.
260 Madison Ave.
New York, NY 10016
Joseph S. Spivey, President D-6
Illinois Coal Association
212 South Second Street
Springfield, IL 62701
8
-------
Coronenter
H. V, Harrel1
Senior Vice President
Freeman United Coal Mining Company
123 South 10th Street
P. 0. Box 1587
Mount Vernon, IL 62864
Richard L. Cook
Executive Director
Commonwealth of Virginia
Air Pollution Control Board
P. 0. Box 10089
Richmond, VA 23240
Jarre!1 S. Mitchell, Colonel, USAF
Chief, Engineering Division
Directorate of Engineering & Services
Department of the Air Force
Headquarters U.S. Air Force
Washington, DC 20332-5000
Lauren W. laabs
Senior Environmental Engineer
A. E. Staley Manufacturing Company
2200 E. Eldorado Street
Decatur, IL 62521
George H. Lawrence, President
American Gas Association
1515 Wilson Boulevard
Arlington, VA 22209
Jack L. Cooper
Director, Environmental Affairs Division
National Food Processors Association
1401 New York Avenue, N.W.
Washington, DC 20005
Winston A, Smith, Director
Air, Pesticides, and Toxics Management Division
U.S. Environmental Protection Agency
Region IV
345 Courtland Street
Atlanta, GA 30365
-------
Commenter
J. Anthony Ercole
Executive Vice President
Pennsylvania Coal Mining Association
212 North Third Street, Suite 201
Harrisburg, PA 17101
A. David Hovarongkura
3510 Klamath Street
Oakland, CA 94602
Francis P. Bonner, Chairman
Anthracite Health and Welfare Fund
Suite 415
2 East Broad Street
Hazleton, PA 18201
J. W. Hughes, President
Turris Coal Company
P. 0. Box 21
Elkhart, IL 62634
George Roskos, Plant Manager
Continental Cogeneration Corp.
P. 0. Box 220
Cohasset, MA 02025
Peter Rozelle
The Pennsylvania State University
Combustion Laboratory
405 Academic Activities Building
University Park, PA 16802
Richard E. Grusnlck, Chief
Air Division
Alabama Department of Environmental Management
1751 Federal Drive
Montgomery, At 36130
John D. Grogan
Alabama Power Company
600 North 18th Street
P. 0. Box 2641
Birmingham, AL 35291
10
-------
Conwienter
W. H. Axtman
Executive Director
American Boiler Manufacturers Association
Suite 160
950 North Glebe Road
Arlington, VA 22203
Albert H. Toma, III
Assistant to the President
Fort Howard Paper Company
P. 0. Box 19130
Green Bay, WI 54307-9130
James 0. Pickard
Secretary of Commerce
Commonwealth of Pennsylvania
Department of Commerce
Harrisburg, PA
William W. Scranton, III
Lieutenant Governor
Commonwealth of Pennsylvania
Lieutenant Governor's Office
Harrisburg, PA 17120-0002
Charles 0. Malloch
Director, Regulatory Management
Environmental Policy Staff
Monsanto Company
800 N. Lindbergh Blvd.
St. Louis, HO 63167
Richard J. Durbin
Member of Congress
U.S. House of Representatives
Washington, DC 20515
William F. Martin
Deputy Secretary
U.S. Department of Energy
Washington, DC 20585
Harold F. Elkin
Director, Environmental Affairs
Sun Company, Inc.
100 Matsonford Road
Radnor, PA 19087-4597
-------
Commenter
Docket
Reference
Carl Avers
President
International District Heating and Cooling Association
1101 Connecticut Ave., Suite 700
Washington, DC 20036
D-30
0.' B. Smith
General Manager
Chevron U.S.A., Inc.
575 Market Street
San Francisco, CA 94105-2856
D-31
M. E. Miller, Jr., P.E.
Manager, Environmental Engineering Unit
R. J. Reynolds Tobacco Company
Winston-Salem, NC 27102
D-32
Richard L. 0'Connell, P.E.
Vice President, Engineering
Hawaiian Electric Company, Inc.
P. 0. Box 2750
Honolulu, HI 96840-0001
D-33
J. R. Smith, Manager
D-34
Air Resources Division
Environmental Protection Department
The Light Company
Houston Lighting & Power
P. O. Box 1700
Houston, TX 77001
W. W. Lyons D-35
Vice President
Nerco, Inc.
Ill SM. Columbia, Suite 800
Portland, OR 97201
John C. Shirvlnsky D-36
President
Keystone Bituminous Coal Association
Suite 301, 208 North Third Street
Harrlsburg, PA 17101
John A. Paul D-37
Supervisor
Regional Air Pollution Control Agency
451 W. Third Street
P. 0. Box 972
Dayton, OH 45422
12
-------
Coronenter
Docket
Reference
Peter W. McCallum
Senior Corporate Environmental Specialist
The Standard Oil Company
200 Public Square
Cleveland, OH 44114-2375
D-38
Charles G. McDowell, P.E.
Manager, Birmingham District Steam System
D-39
Alabama Power Company
15 South 20th Street
Birmingham, AL 35233
Carl E. Bagge
President
National Coal Association
1130 Seventeenth Street, N.W.
Washington, DC 20036-4677
D-40
Richard L. White, Manager
D-41
Environmental Services
Texas Utilities Generating Company
Skyway Tower
400 North Olive Street, L.B. 81
Dallas, TX 75201
John A. Cunningham D-42
Vice President & General Manager
Combustion Engineering, Inc.
1000 Prospect Hill Rd.
P. 0, Box 500
Windsor, CT 06095-0500
Peter A. McGrath D-43
President
American Hydro Power Co,
33 Rock Hill Rd.
Bala Cynwyd, PA 19004-2010
Cogentrix, Inc. D-44
2 Parkway Plaza, Suite 290
Charlotte, NC 28210
Walter Roy Quanstrom D-45
General Manager
Environmental Affairs & Safety Department
Amoco Corporation
200 East Randolph Drive
Chicago, IL 60601
13
-------
Commenter
Kenneth L. Mill lams
Director, Government Affairs
Western Energy Company
16 East Granite
Butte, MT 59701
Keith M. Bentley
Senior Environmental Engineer
Georgia-Pacific Corporation
133 Peachtree Street, N.E.
P. 0. Box 105605
Atlanta, GA 30348-5605
Robert S. Evans, II
Supervisor, Air Programs
Northern States Power Company
414 Nicollet Hall
Minneapolis, MN 55401
John F. McKenzie
Director, Environmental Planning
Environmental Services Department
Pacific Gas and Electric Company
P. 0. Box 7640
San Francisco, CA 94120
William B. Marx
President
Council of Industrial Boiler Owners
5817 Burke Centre Parkway
Burke, VA 22015
C. Richard Cahoon
Vice President for Policy
Petroleum Marketers Association of America
1120 Vermont Ave., N.W.
Suite 1130
Washington, DC 20005
Robert B. Flagg
Manager, Environmental
Mining and Reclamation
1575 Eye Street, N.W.
Suite 525
Washington, DC 20005
and Regulatory Affairs
Council of America
Docket
Reference
D-46
D-47
D-48
D-49
D-50
D-51
D-52
14
-------
Commenter
Docket
Reference
Michael J. Zimmer D-53
President
Cogeneration Coalition of America, Inc.
2 Lafayette Centre
1133 21st Street, N.W.
Suite 500
Washington, DC 20036
Richard E. Eckfield D-54
President
North American District Heating and Cooling Institute
One Thomas Circle, N.W., Suite 725
P. 0. Box 19428
Washington, DC 20036
Lee A. DeHihns, III D-55
Associate General Counsel
Ohio Chamber of Commerce
35 E. Gray Street, 2nd Floor
Columbus, OH 43215-3181
Michael K. Glenn D-56
Porter, Wright, Morris & Arthur
1133 15th Street, N.W.
Suite 1200
Washington, DC 20005
(for The Cincinnati Gas & Electric Company,
Columbus and Southern Ohio Electric Company,
and The Dayton Power and Light Company)
F. William Brownell D-57
Mel S. Schulze
Hunton & Williams
2000 Pennsylvania Avenue, N.W.
Washington, DC 20006
(for the Utility Air Regulatory Group)
James R. Walpole D-58
Mark P. Fitzsimmons
Chadbourne & Parke
1101 Vermont Ave., N.W.
Washington, DC 20005
(for the American Paper Institute/National
Forest Products Association)
15
-------
Commenter
Docket
Reference
James J. Rhoades
State Senator
Senate Post Office
The State Capitol
Harrlsburg, PA 17120-0030
D-59
Campbell Soup Company
Camden, NJ 08101-0391
D-60
Robert Harrison
Vice President and General Manager
Western Oil and Gas Association
727 West Seventh Street
Los Angeles, CA 90017
D-61
James K. Beasom
Staff Governmental Affairs Administrator
Appleton Papers, Inc.
P. 0. Box 359
Appleton, WI 54912
D-62
Roger B. McCann, Director
D-63
Division of Air Pollution Control
Commonwealth of Kentucky
Natural Resources and Environmental Protection Cabinet
Department for Environmental Protection
Fort Boone Plaza
18 Rellly Road
Frankfurt, KY 40601
David 6. Hawkins D-64
Senior Attorney
Natural Resources Defense Council
1350 New York Avenue, N.W.
Washington, DC 20005
John E. Plnkerton D-65
Air Quality Program Manager
National Council of the Paper Industry for Air and
Stream Improvement, Inc.
260 Madison Ave.
New York, NY 10016
T. 0. Andrews D-66
Manager, Environmental Affairs
Hammermi11 Paper Company
1540 East Lake Road
Erie, PA 16533
16
-------
Commenter
Clifford L. Jones
President
Pennsylvania Chamber of Commerce
222 North Third Street
Harrisburg, PA 17101
George H. Lawrence, President
American Gas Association
1515 Wilson Boulevard
Arlington, VA 22209
William C. Ray
United Mine Workers of America
911-914 Northeastern Building
8 West Broad Street
Hazleton, PA 18201
Michael K. Glenn
Porter, Wright, Morris & Arthur
1133 15th Street, N.W.
Washington, DC 20005
Bethlehem Mines Corporation
Stone-Anthracite Business Unit
Annville, PA 17003
Frank P. Partee
Principal Staff Engineer
Stationary Source Environmental Control Offi
Ford Motor Company
15201 Century Drive
Suite 608
Dearborn, MI 48120
Geraldine V. Cox, Ph.D.
Vice President/Technical Director
Chemical Manufacturers Association
2501 M Street, N.W.
Washington, DC 20037
Keith M. Bentley
Senior Environmental Engineer
Georgia-Pacific Corporation
133 Peachtree St., N.E.
P. 0. Box 105605
Atlanta, GA 30348-5605
-------
Commenter
Docket
Reference
William M. Kelce D-75
President
Alabama Coal Association
244 Goodwin Crest Drive
Suite 110
Birmingham, AL 35209
Wendy L. Gramm D-76
Administrator for Information and Regulatory Affairs
Executive Office of the President
Office of Management and Budget
Washington, DC 20503
The Pittston Company . D-77
J. S. Larsen, Vice President D-78
Energy, Environmental, and Regulatory Affairs
Weyerhaeuser Company
Tacoma, WA 98477
U. V. Henderson, Jr. D-79
Associate Director, Environmental Affairs
Texaco, Inc.
P. 0. Box 509
Beacon, NY 12508
Warren W. Tyler D-80
Director
State of Ohio Environmental Protection Agency
P. 0. Box 1049
361 East Broad Street
Columbus, OH 43216-1049
H. E. Cameron D-81
Environmental Activities Staff
General Motors Corporation
General Motors Technical Center
30400 Mound Road
Warren, MI 48090-9015
Paul Bork D-82
Dow Chemical Company
Midland, MI
Island Creek Corporation D-83
Lexington, KY
18
-------
R. Thorne, Director
Corporate Office of Environmental Affairs
Union Camp Corporation
P. 0. Box 1391
Savannah, GA 31402
Commenter
Docket
Reference
D-84
Joel D. Patterson
Manager, Environmental Affairs
Middle South Services, Inc.
Box 61000
New Orleans, LA 70161
D-85
R. Harry Bittle
D-86
Deputy Secretary for Environmental Protection
Commonwealth of Pennsylvania
Department of Environmental Resources
P. 0. Box 2063
Harrisburg, PA 17120
William B. Ericson, P.E. D-87
10 Lakeview Drive
Somerset, PA 15501-8694
C. Richard Cahoon D-88
Vice President for Policy
Petroleum Marketers Association of America
1120 Vermont Ave., N.W.
Suite 1130
Washington, DC 20005
Joseph W. Reitz, Partner D-89
HJ&H Coal Company
P. 0. Box 224
Sunbury, PA 17801
Sidney G. Nelson D-90
President
Sanitech Inc.
1935 East Aurora Road
Twinsburg, OH 44087
Jurgen H. Kleinau D-91
Marketing Manager
Keeler Dorr-Oliver
P. 0. Box 548
Williamsport, PA 17703-0548
19
-------
Commenter
Docket
Reference
Michael Musheno D-92
Senior Program Manager
Combustion Engineering, Inc.
800 Eastowne Drive
Suite 200
Chapel Hill, NC 27514
Peter Rozelle D-93
The Pennsylvania State University
Combustion Laboratory
405 Academic Activities Building
University Park, PA 16802
Jan B. Vlcek D-94
James M. Bushee
Sutherland, Asbill & Brennan
1666 K Street, N.W.
Washington, DC 20006-2803
(for the Council of Industrial Boiler Owners)
Jorge H. Berkowitz, Ph.D. D-95
Director
State of New Jersey
Department of Environmental Protection
Division of Environmental Quality
John Fitch Plaza, CN 027
Trenton, NJ 08625
Douglas A. Riggs D-96
General Counsel
U.S. Department of Commerce
Washington, DC 20230
Peter C. Freudenthal D-97
Director, Air and land Use
Consolidated Edison Company of Hew fork, Inc.
4 Irving Place
New York, NY 10003
Bruce Blanchard D-98
Director, Environmental Project Review
U.S. Department of the Interior
Office of the Secretary
Washington, DC 20240
TRW Energy Products Group D-99
Combustion Business Unit
One Space Park
Redondo Beach, CA 90278-1001
20
-------
TABLE 2-2. LIST OF PUBLIC HEARING SPEAKERS ON THE PROPOSED STANDARDS
FOR SULFUR DIOXIDE EMISSIONS FROM INDUSTRIAL-COMMERCIAL-INSTITUTIONAL
STEAM GENERATING UNITS
Docket
Speaker Reference
H. E. Cameron F-l.l
General Motors Corporation
General Motors Technical Center
30400 Mound Road
Warren, MI 48090-9015
J. H. Kleinau F-1.2
Keeler, Dorr-Oliver Boiler Company
P. 0. Box 548
Williamsport, PA 17703-0548
0. H. Kleinau F-1.3
for North American District Heating and
Cooling Institute
P. 0. Box 19428
Washington, DC 20036
Peter C. Freudenthal F-1.4
Consolidated Edison Company of New York
4 Irving Place
New York, NY 10003
David Pattee F-1.4a
International Paper Company
New York, NY
Francis A. Ferraro F-1.4b
Babcock and Wilcox Company
20 S. Van Buren
Barberton, OH 44203
Joseph W. Mull an F-1.4c
National Coal Association
Washington, DC 20036
Edward Schwartz F-1.5
Peoples Natural Gas of Pittsburgh
Pittsburgh, PA
for American Gas Association
21
-------
Speaker
William Axtman
American Boiler Manufacturers Association
950 N. Glebe Rd., Suite 160
Arlington, VA 22203
John Cuthbertson
James River Corporation
Richmond, VA
for American Paper Institute/National
forest Products Association
Jerry L. Lombardo
Island Creek Coal Company
Lexington, KY
for National Coal Association
Harry L, Storey
Alliance for Clean Energy
555 17th Street
Denver, CO 80202
John Stauffacher
The Dow Chemical Co.
B101 Bldg.
Freeport, TX
Jeffrey Smith
Industrial Gas Cleaning Institute
1707 L Street, N.W.
Suite 570
Washington, DC 20036
William B. Marx
Council of Industrial Boiler Owners
1817 Burke Centre Parkway
Burke, VA 22015
Jan B. Vlcek
Sutherland, Asbill & Brennan
1666 K Street, N.W.
Washington, DC 20006
for Council of Industrial Boiler Owners
David Pattee
International Paper Company
New York, NY
for Council of Industrial Boiler Owners
22
-------
Speaker
John Stier
Anheuser-Busch Companies, Inc.
1 Busch Place
St, Louis, MO 63118
for Council of Industrial Boiler Owners
John C. de Ruyter
E. I. DuPont de Nemours & Co.
Wilmington, DE 19898
for Council of Industrial Boiler Owners
William C. Campbell
Cogentrix, Inc.
4828 Parkway Plaza
Two Parkway Plaza, Suite 290
Charlotte, NC 29210
Dennis Williams
Solid Fuel Technology Energy Resources
P. 0. Box 10340
Wilton, NC 28103
George J. Barkanich
ESI, Inc.
811-C Livingston Court
Marietta, GA 30067
23
-------
2.1 NEED FOR THE STANDARD
1. Comment: Many commenters (IV-D-6, IV-D-7, IV-D-10, IV-D-20, IV-D-26,
IV-D-27, IV-D-28, IV-D-35, IV-D-36, IV-D-38, IV-D-40,
IV-D-50, IV-D-51, IV-D-52, IV-D-54, IV-D-58, IV-D-62,
IV-D-66, IV-D-72, IV-D-74, IV-D-78, IV-D-81, IV-D-84,
IV-D-85, IV-D-88, IV-F-1.12, IV-F-1.19) said the standard
was not necessary because total emissions from new
industrial-commercial-institutional steam generating units
will be insignificant, amounting to less than 1.5 percent of
total U.S. SO2 emissions. In addition, the commenters said,
industrial-commercial-institutional steam generating unit
S02 emissions have declined and will continue to decline due
to energy conservation, reduction in heavy Industrial
capacity, existing State and local emission standards, and
because most new steam generating units are installed to
replace older units.
Another commenter (IV-F-1.14) felt that the effectiveness of
Prevention of Significant Deterioration (PSD) regulations in
limiting new steam generating unit emissions should be taken
into account in developing NSPS. This commenter stated that
the PSD review process alone, Independent of any NSPS, would
be sufficient to limit substantially SOg emissions from new
Installations. Other commenters (IV-D-51, IV-D-88) said
that given the small amounts of oil combusted 1n Industrial-
conrniercial-Institutional steam generating units, they do not
constitute a significant source of S02 emissions. The
commenters felt that 1t would require a large amount of
money to remove just a small amount of air pollution and,
therefore, that the standards were not justified.
24
-------
Response: On August 21, 1979, a priority list for development of
additional NSPS was published in accordance with Sections
111(b)(A) and 111(f)(1) of the Clean Air Act (44 FR 49222).
This list identified 59 major stationary source categories
that were not covered by NSPS, but that were judged to be
"significant contributors," i.e., to contribute
significantly to air pollution that could reasonably be
expected to endanger public health or welfare. Fossil
fuel-fired industrial steam generating units ranked eleventh
on this priority list of sources for which NSPS would be
established in the future.
Of the 10 sources ranked above fossil fuel-fired industrial
steam generating units on the priority list, nine were major
sources of volatile organic compound (VOC) emissions.
Because there are many areas that have not attained the
national ambient air quality standard for ozone, major
sources of VOC emissions were accorded a very high priority.
Of the remaining source categories, fossil fuel-fired
ti
industrial steam generating units were the highest ranked
source of particulate matter and SO^ emissions, and the
second highest ranked source of N0X emissions. The
industrial-commercial-institutional source category is a
significant contributor and, therefore, an appropriate
source category for regulation. In addition, individual
Industrial-commercial-Institutional steam generating units
(both oil- and coal-fired) frequently emit or have the
potential to emit more than 91 Mg/year (100 tons/year) of
sulfur dioxide. Such sources are considered "major sources"
under the Clean Air Act. Further, Section 111 does not
require that NSPS be set for only those sources within a
listed category which are themselves significant
contributors. Instead, it directs that NSPS be set for all
sources within a listed category unless the impacts of such
NSPS would be unreasonable.
25
-------
These standards are designed to achieve reductions in SOg
emissions from all new, modified, and reconstructed
industrial-commercial-institutional steam generating units.
The purpose of an NSPS is to provide a uniform national
standard that requires the best demonstrated level of
control, considering cost, energy, and environmental
factors. Other programs that include local and/or
site-specific requirements, such as PSD and NSR, may be more
stringent than the NSPS.
2. Comment: One commenter (IV-D-4) suggested that instead of regulating
the steam generating unit users, who are many, only the fuel
producers, who are few in comparison, should be regulated.
That way, the commenter said, the sulfur content of the fuel
sold could be regulated, eliminating the need for continuous
stack monitoring or emission control devices and reducing
the paperwork burden on both industry and government.
Response: Section 111 of the Clean Air Act authorizes regulation of
new "sources" of air pollution only. It does not authorize
establishment of national standards for the sulfur content
of fuels. The costs and administrative resources required
to implement the final standards for both government and
industry are considered reasonable, even though the number
of owners and operators affected under the standards is
greater than 1f fuel producers were regulated.
2.2 APPLICABILITY
1. Comment: Several comnenters said that the standards should be limited
to steam generating units with heat input capacities greater
than 73 MW (250 million Btu/hour). One (IV-D-52) stated
that the 1979 decision to regulate utility units larger than
26
-------
73 MW (250 million Btu/hour) fulfilled the Section 111
mandate to regulate this source category, and that emissions
from units in the smaller size category represent only a
very small portion of the total emissions from the source
category. Other commenters (IV-D-14, IV-D-40, IV-D-58,
IV-D-62, IV-D-66, IV-D-74, IV-D-78, IV-D-84, IV-F-1.7,
IV-F-1.8) agreed, saying that these smaller steam generating
units will account for less than 1 percent of total national
SOg emissions. In addition, the commenters said, emissions
from new small steam generating units are already controlled
by State Implementation Plan (SIP), PSD, and other State and
local regulations. Finally, the commenters claimed that the
sophisticated operation and maintenance required to operate
post-combustion S02 control systems will simply be
unavailable or, at best, inadequate in most small industrial
operations.
Response: Section 111(a) of the Act requires that standards of
performance reflecting the degree of emission reduction and
the percentage reduction achievable through application of
best demonstrated technology be established for categories
of fossil fuel-fired sources which are "significant
contributors" to air pollution. The language of
Section 111(a) of the Act does not limit application of NSPS
only to electric utility units or to units above a certain
size. Section 111 requires the establishment of NSPS that
reflect "the best technological system of continuous
emission reduction which... has been adequately
demonstrated" for sources in a listed category. It does not
require that NSPS be set within a listed category only for
classes of sources which are themselves significant
contributors. As discussed above, industrial-commercial -
institutional steam generating units are considered to be
significant contributors of air pollution because of their
particulate matter, N0X, and S02 emissions.
27
-------
Emissions from steam generating units with heat input
capacities less than 73 MW (250 million Btu/hour) are
currently not controlled with any degree of consistency.
While some facilities are well controlled by SIP, PSD, and
other programs, others are subject to much less stringent
standards, or no standards at all. Therefore, this NSPS
will provide a uniform level of control with which all new,
modified, or reconstructed facilities must comply.
With respect to operation and maintenance of control
equipment, the requirements of these standards are considered
reasonable. The additional costs associated with this
operation and maintenance were Included 1n assessing the
Impacts of the standards, as discussed 1n the "SO2 Cost
Report." Proper operation and maintenance may, however,
require an increased management commitment on the part of
some owners/operators. In the event that owners/operators
of smaller steam generating units would prefer not to spend
the additional resources necessary for proper operation and
maintenance of flue gas desulfurlzatlon or fluldlzed bed
combustion systems, alternatives, such as the use of
natural gas, are available.
2. Comment: One commenter (IV-D-37) said that the standards should be
revised to Include those units with heat Input capacities of
22 MW (75 million Btu/hour) or greater. The commenter
claimed that many plants construct multiple smaller units
instead of one larger unit, and said that these sources
should be regulated to dissuade facilities from constructing
smaller units specifically for the purpose of avoiding NSPS
requirements.
28
-------
Response: These standards apply to individual steam generating units
with heat input capacities greater than 29 MW (100 million
Btu/hour) primarily because the steam generating unit
population above that size 1s characterized predominantly by
industrial units. Below 29 MW (100 million Btu/hour) heat
input capacity, the steam generating unit population tends
to be more of a mixture of Industrial, commercial, and
institutional applications. For this reason, it was decided
to develop standards first for steam generating units above
29 MW (100 million Btu/hour) heat input capacity 1n size,
followed by standards for units below that size. Standards
for industrial-commercial- Institutional units with heat
Input capacities of 29 MW (100 million Btu/hour) or less are
currently being developed with final promulgation scheduled
for 1990. Once standards for these smaller boilers are
promulgated, construction of units smaller than 29 MW
(100 million Btu/hour) for the purpose of evading regulation
will not be possible.
3. Comment: Several commenters (IV-D-26, IV-D-30, IV-D-50, IV-D-53,
IV-F-1.12) said that the standards should be limited to
"fossil fuel-fired boilers," I.e., steam generating units
firing more than 50 percent conventional fossil fuel during
a year. According to these commenters, mixed fuel-fired
steam generating units have advantages such as lower SOg
emissions, conservation of fossil fuels, and disposal of
waste materials that should not be discouraged.
Response: Mixed fuel-fired steam generating units can be significant
sources of SO2 emissions even when less than 50 percent
fossil fuel 1s fired. Because mixed fuel-fired units tend
to be large, the fossil fuel heat input and, thus, the
emissions of such units can exceed that of smaller units
29
~
-------
<
that fire 100 percent fossil fuel. The control technologies
for reducing S02 emissions are as effective on steam
generating units firing mixtures of fossil fuel with
nonfossil fuel as they are on units firing fossil fuel
alone. Also, the cost and economic impacts of the standards
on mixed fuel-fired steam generating units firing mixtures
of fossil and nonfossil fuels were assessed (see "Impact of
New Fuel Prices on the Costs and Cost Effectiveness of S02
Emission Control of Mixed Fuel-Fired Steam Generating
Units," and the impacts of the final standards on such units
are considered reasonable.
As evidenced in the proposed standards, however, the
analysis identified one situation under which the impacts of
standards requiring a percent reduction in emissions could
be unreasonable for mixed fuel-fired steam generating units.
If the amount of coal or oil fired in a mixed fuel-fired
generating unit is less than 30 percent of its rated heat
input capacity on an annual basis, the costs associated with
achieving a percent reduction in SOg emissions are considered
unreasonable. Consequently, such mixed fuel-fired steam
generating units are exempt from the percent reduction
requirement in the final standards. These units, however,
will be required to comply with an emission limit for S0£.
4. Comment: One commenter (IV-D-95) stated that municipal solid waste
combustion facilities with over 29 MW (100 million Btu/hour)
heat input capacity should be classified as affected
facilities under the standards. The coimnenter noted that
the State of New Jersey requires at least an 80 percent
reduction in SOg from these facilities when potential
emissions exceed 50 ppmv.
30
-------
Response: Although use of municipal solid waste, wood, and natural gas
is subject to regulation under the industrial-commercial-
institutional steam generating unit source category, the
sulfur content of the fuels, and thus their SO2 emission
potential, is generally low. As a result, the costs for
installation of FGD technology for the purpose of S02
control would be high and resulting cost-effectiveness
levels unreasonable. Therefore, these standards are limited
to steam generating units firing coal and oil alone or in
combination with other fuels. If only solid waste is
combusted, the standards would not apply. However, if a
facility combusts solid waste in combination with coal or
oil, it would be subject to the standards. It should also
be noted that the particulate matter standards promulgated
for this source category on November 25, 1986 (51 FR 42768)
would apply to units firing solid waste, either alone or in
combination with other fuels.
As a separate regulatory action, the Agency has also
assessed the unique environmental problems presented by the
combustion of municipal solid waste and, as a result,
concluded that additional regulations specific to the
combustion of municipal solid waste are appropriate
(52 FR 25399). Under this separate action, the Agency has
provided operational guidance to State and local authorities
for use in reviewing prevention of significant deterioration
(PSD) permits under 40 CFR 51.24. Additional regulations
are currently under development.
5. Comment: One commenter (IV-D-28) suggested that a regulatory
clarification be added to state that any change to an
existing steam generating unit, originally designed to
accommodate gaseous or liquid fossil fuels, to accommodate
the use of any other fuel (fossil or nonfossil) does not
31
-------
bring that unit under the applicability of the subpart.
Such a provision, the coirmienter said, was included in the
1979 Utility NSPS at 40 CFR 60.40a(d), and adoption of it in
this standard could significantly mitigate potential adverse
energy impacts associated with the standards. The commenter
also suggested that the preamble to the final standard also
contain a discussion that alteration of these steam
generating units to fire, for example, coal/oil/water
mixtures, coal/oil mixtures, shale oil, or liquified or
gasified coal, would not result in the facility becoming
subject to the standard under the modification/
reconstruction provisions.
Response: Fuel conversions are exempt from the modification provision
under two circumstances: (1) where the facility was
designed to accommodate the alternative fuel prior to the
date of the applicable NSPS subpart [40 CFR 60.14(e)(4)];
or (2) where the facility is required to convert to coal
pursuant to an order issued under Title III of the
Powerplant and Industrial Fuel Use Act of 1978 (42 U.S.C.
8301 et seq.), as amended [40 CFR 60.14(e)(4) and Section
111(a)(8) of the Clean Air Act]. In other instances where
the facility is altered to burn coal or waste fuels, the
facility may be subject to the standards if it qualifies as
a modified or reconstructed unit as defined in 40 CFR 60.14
or 60.15 of the General Provisions.
The modification clause in the General Provisions (60.14)
defines a modification as "any physical or operational
change to an existing facility which results in an increase
in the emission rate...." Therefore, the modification
provision can be avoided by ensuring that the emission rate
(ng/J or lb/million Btu) does not increase as a result of
32
-------
the changes made (generally by the use of fuel with a sulfur
content equivalent to or lower than the fuel used
originally). In addition, the reconstruction provisions
apply only when changes made to the facility exceed
50 percent of the fixed capital costs associated with
constructing a new facility. Few, if any, changes that
could be made to a steam generating unit would be expected
to exceed this limit. Therefore, few reconstructed units
are expected to be affected under the standards.
6. Comment: One commenter (IV-D-55) stated that the proposed standards
should be amended to allow low sulfur content fuels rather
than FGD for modified and reconstructed steam generating
units. The commenter contended that the cost and
feasibility problems associated with retrofitting an
existing unit to accommodate an FGD system would be
substantial.
Response: As discussed above, the General Provisions of 40 CFR Part 60
define modified and reconstructed facilities. A
modification, as defined in 60.14 of the General Provisions,
includes certain physical or operational changes to an
existing facility which result in an increase in the
emission rate. As long as emissions can be maintained at or
below the level measured before the change occurred, the
change would not qualify as a modification. A reconstructed
facility, under 40 CFR 60.15, is one in which replacement of
components constitutes at least 50 percent of the fixed
capital cost of a comparable new facility and which can
technologically and economically meet the standards. In the
case of reconstructed facilities, a determination is made on
a case-by-case basis from an examination of the technical
and economic feasibility of complying with the standards.
33
-------
Thus, the reconstruction provision already provides for
exemptions when costs or technical feasibility create
problems, and the modification provisions can be avoided by
ensuring that there is no increase in the emission rate. In
addition, no general types of modifications or
reconstructions of industrial-commercial-institutional steam
generating units have been identified where compliance with
the standards would be unreasonable. Therefore, no
provisions have been included for an alternative compliance
option for modified or reconstructed units.
7. Comment: A number of commenters felt that steam generating units
operating at low capacity factors should be exempted from
the SOg standards. Several (IV-D-55, IV-D-56, IV-D-57,
IV-D-72, IV-D-74, IV-D-78, IV-D-79, IV-D-80, IV-D-85) said
the percent reduction requirement of the proposed standards
should not be applicable to steam generating units operating
at low capacity utilization rates. One (IV-D-49) suggested
that if a steam generating unit operates at an average fuel
utilization rate (for fuels other than natural gas) of 30
percent or less during a 30-day period, it should be exempt
from compliance with the standard. The commenter said such
a capacity factor exemption could be related to the percent
sulfur in the fuel used. Others (IV-D-26, IV-D-30, IV-D-50,
IV-D-53, IV-D-76) said the rationale for not allowing an
exemption from the percent reduction requirement for low
capacity (less than 30 percent) fossil fuel-fired steam
generating units while allowing it for mixed fuel-fired
units is invalid, since the same cost effectiveness numbers
are used for both. The commenters stated that EPA should
not assume that mixed fuel fired units would behave
differently than low capacity factor units. Several
commenters (IV-D-21, IV-D-34, IV-D-41, IV-D-56, IV-D-57,
34
-------
IV-D-85) felt that electric utility auxiliary steam
generating units which operate at low capacity factors
(e.g., 10 or 20 percent) should not be subject to the
standards.
One commenter (IV-D-12) said special exemptions from
compliance with the standards should be allowed for units
which operate less than 6 months of the year. If a total
exemption is not possible, the commenter suggested an
arrangement should be made so that the average air emission
effects over the entire year can be considered when
determining the applicability of the proposed standards.
Another commenter (IV-D-64), however, stated that there
should be no exemptions from the percent reduction require-
ment given on the basis of size or capacity utilization
rates. The commenter said the difference in emission
reductions achieved by FGD over low sulfur fuel is
significant for all size classes of steam generating units,
and the costs of compliance are only marginally greater for
small or low capacity units.
Response: As noted by the commenters, there are many types of
industrial-comnercial-institutional steam generating units
that operate with low annual capacity factors (capacity
utilization rates). These include auxiliary steam
generating units located at electric utility power plants
that are used to start up main steam generating units, units
operated infrequently as "backup" steam capacity at
industrial plants, mixed fuel-fired steam generating units
burning small amounts of coal or oil, and nonfossil
fuel-fired steam generating units that use oil or coal as a
backup fuel during periods when nonfossil fuel is
unavailable.
35
-------
Review and reconsideration of the initial analysis, recent
steam generating unit sales data, and recent operations data
from new steam generating units leads to the conclusion that
some plants will install coal- or oil-fired steam generating
units even where natural gas is the fuel of economic choice
and despite promulgation of standards requiring a percent
reduction in emissions from coal- and oil-fired steam
generating units. For most applications, the cost of
applying percent reduction requirements to coal- and
oil-fired steam generating units will be reasonable.
However, as in the case of mixed fuel-fired units, low
capacity factor steam generating units which obtain less
than 30 percent of their rated annual heat input capacity
from the combustion of coal or oil could well experience
unreasonable impacts. The final regulation, therefore, does
not require a percent reduction in SOg emissions from steam
generating units operated at low annual capacity utilization
factors, provided a Federally enforceable permit condition
limits the annual capacity utilization factor for coal or
oil to less than 0.3 (i.e., 30 percent of the rated capacity
of the steam generating unit). If a decision is
subsequently made to operate with a coal and oil annual
capacity factor greater than 30 percent, compliance with the
percent reduction requirement would be necessary.
These "low capacity factor" steam generating units would,
however, be required to meet certain emission limits.
Emissions of S02 from steam generating units operating at
annual capacity factors of 30 percent or less would be
limited to 516 ng/J (1.2 lb/million Btu) heat input if coal
is fired and 129 ng/J (0.3 lb/million Btu) if oil is fired.
36
-------
These emission limits were selected based on the
availability of low sulfur coal and oil throughout the U.S.
Continuous emission monitoring, continuous compliance
provisions, and quarterly reporting are required for all
units, including these "low capacity factor" units.
8. Comment: Several commenters (IV-D-11, IV-D-20, IV-D-26, IV-D-30,
IV-D-38, IV-D-42, IV-D-45, IV-D-49, IV-D-50, IV-D-53,
IV-D-58, IV-D-62, IV-D-65, IV-D-66, IV-D-74, IV-D-76,
IV-D-84, IV-F-1.7) said a percentage reduction requirement
should not be issued for steam generating units with low
capacity utilization rates for oil, such as those using oil
for backup or pilot fuel purposes. They felt that, given
the small amounts of oil combusted, these units do not
constitute significant emission sources, and the high cost
of achieving the reduction would be out of proportion to the
amount of pollution removed. One coircnenter (IV-D-92)
requested clarification on this point, asking whether the
SO2 percent reduction requirement applies to steam
generating units that use oil only for startup and then burn
a nonfossil fuel such as municipal solid waste. The
commenter also asked whether the SO2 reductions would apply
if gas is used as the startup fuel instead of oil.
Response: As discussed above, an exemption from the percent reduction
requirement has been granted for steam generating units
operating at low annual capacity utilization factors,
provided a Federally enforceable permit condition limits the
annual capacity utilization factor for coal or oil to less
than 30 percent of the rated capacity of the steam
generating unit. Therefore, the types of steam generating
units mentioned by the commenters would not be required to
achieve a percent reduction in SOg emissions.
37
-------
9. Comment: Several commenters (IV-D-16, IV-D-18, IV-D-19, IV-D-24,
IV-D-25, IV-D-43, IV-D-59, IV-D-66, IV-D-67, IV-D-69,
IV-D-71, IV-D-85, IV-D-89) said sources burning anthracite
coal or anthracite mining waste (culm) should be exempted
from the standards. They said the 1979 Electric Utility
NSPS granted special provisions for anthracite, and felt
that the reasons for granting those provisions are equally
applicable to industrial-commercial-institutional steam
generating units. Several of these commenters (IV-D-18,
IV-D-19, IV-D-24, IV-D-25, IV-D-43, IV-D-67) said that the
secondary environmental benefits associated with burning low
sulfur anthracite coal have been overlooked. In particular,
they noted, the use of anthracite mining waste (culm) in
steam generating units is of great benefit to the local
environment in the anthracite mining areas of Pennsylvania
and should be encouraged, not discouraged. To date, the
commenters said, the only technically and economically
feasible means of disposing of this waste is through
combustion in a fluidized bed steam generating unit.
Response: The exemption from the percent reduction requirement granted
for anthracite in Subpart Da was provided to encourage
reclamation of anthracite mines, resulting in environmental
benefits such as improvement of mine drainage acid-water
conditions, elimination of old mining scars on the
topography, and eradication of dangerous fires in deep mines
and culm banks. At the time of promulgation of Subpart Da
(June 1979), reclamation of areas that had been despoiled by
mining was a high priority, as evidenced by the passage of
the Federal Surface Mining Control and Reclamation Act. The
exemption from the percent reduction requirement provided
under Subpart Da for anthracite created a market for this
fuel in the utility sector, and the environmental benefits
38
-------
associated with this large-scale utility reclamation were
judged to outweigh any ambient air quality impacts of
burning anthracite without a post-combustion SOg control
system.
The small projected overall coal demand in the industrial-
commercial -institutional steam generating unit market,
combined with the predominant use of locally available
coals, would generally result in anthracite being used as a
local fuel only, even if an exemption from the percent
reduction requirement was granted for anthracite. The small
quantities of coal demanded by the industrial sector in
northeastern Pennsylvania and other areas of localized
anthracite deposits would not result in the large-scale
utility-type reclamation of abandoned mines that might have
resulted from the Subpart Da exemption. Therefore, no
special provisions for anthracite have been included in the
final standards.
A different situation exists, however, with the firing of
^ anthracite mining waste and other coal mining and washing
wastes (collectively referred to as coal refuse). These
waste piles are not only unsightly, but they are responsible
for acid drainage problems and can also lead to fires from
spontaneous combustion. Therefore, it is important to
encourage the use of these wastes as fuels in steam
generating units (specifically fluidized bed combustion
steam generating units) to eliminate a potential
environmental hazard. Consequently, a less stringent
percent reduction requirement of 80 percent has been
provided for fluidized bed combustion steam generating units
which fire coal refuse. This action balances the need to
minimize air pollution from combustion of these wastes
39
-------
against the environmental benefits resulting from
eliminating waste fuel piles.
10. Comment: One commenter (IV-D-33) said the standard should include an
exemption for noncontinental areas such as Hawaii, as was
done in Subpart Da. The commenter asserted that these areas
have unique environmental problems associated with disposal
of solid and liquid wastes resulting from SO2 control, as
well as a lack of flexibility in fuel choice (i.e., natural
gas is not available in Hawaii). According to the
commenter, these special situations, plus additional
operating costs such as shipping large quantities of
alkaline chemical reagents, result in exorbitant compliance
costs for sources in island areas.
Response: Facilities in noncontinental areas (Hawaii, the Virgin
Islands, Guam, American Samoa, Puerto Rico, and the Northern
Mariana Islands) constitute a subcategory subject to unique
environmental and economic constraints in complying with
this NSPS. Because of a lack of natural gas supplies, "fuel
switching" to natural gas is not feasible.
P
In addition, the cost of importing FGD reagent and other
materials to noncontinental areas would make the costs
associated with achieving a percent reduction in emissions
much higher in these areas than on the continental mainland.
In light of these unique considerations, an exemption from
the percent reduction requirement has been provided for
steam generating units located in noncontinental areas,
regardless of the capacity factor of the unit. Such
facilities are, however, required to meet the SOg emission
limitations discussed above for units operating at low
40
-------
capacity factors for coal or oil. These emission limits, as
well as the fact that these facilities will be reviewed to
ensure compliance with PSD limitations, will minimize the
impact of these facilities on ambient air quality.
2.3 BASIS OF THE STANDARD
1. Comment: One commenter (IV-D-63) said that the standard should
provide for the reasonable control of emissions from any
fuel a source elects to use and, therefore, the standard
should not directly or indirectly prohibit the use of coal
or oil. The commenter added that, although the nominal
control technique that forms the basis of the standard is
the installation of a flue gas desulfurization (FGD) system,
the actual control system proposed is the use of natural gas
as a steam generating unit fuel, since the cost of
installing and using an FGD system would represent an
unreasonable burden on sources having units in the smaller
size range covered by this standard.
Response: The standard does not prohibit the use of either coal or oil
as a steam generating unit fuel, and their continued use is
anticipated. The increased costs and economic impacts
associated with operating an FGD or FBC system to control
emissions from coal- or oil-fired steam generating units
have been examined and are considered reasonable. It is
also anticipated, however, that a number of new steam
generating units that might have been designed to fire coal
or oil will be designed and constructed to fire natural gas
in response to these standards. The magnitude of this
switch in fuels will vary depending on local fuel prices.
The anticipated reductions in coal and oil use and the
anticipated increases in natural gas use in industrial steam
41
-------
generating units are also considered reasonable in light of
the associated decreases in SO2 emissions.
2. Comment: One commenter (IV-D-76) stated that the need for a percent
reduction requirement should be explicitly examined in terms
of emission reductions achieved and cost, energy and
environmental impacts. The commenter suggested that an
emission limit should be considered as the basis of the
standard.
Response: In the development of this standard, a number of regulatory
alternatives were considered and analyzed. Various
requirements for a percent reduction in SO2 emissions as
well as standards limiting SOg emissions to a specific
emission limit were among the alternatives examined. The
analyses of these alternatives examined the cost, energy,
and environmental impacts, as well as the overall economic
impacts, of each alternative.
Under Section 111(a) of the Clean Air Act, NSPS for fossil
fuel-fired stationary sources are required to include both
an emission limit and a percentage reduction requirement,
unless the imposition of a percentage reduction requirement
would result in unreasonable cost, environmental, or energy
impacts. Percent reduction requirements are considered
reasonable for most subcategories of industrial-commercial -
institutional steam generating units. There are exceptions,
however, for which exemptions have been granted. For
example, steam generating units firing coal or oil, or a
mixture of coal and oil, at less than 30 percent of their
rated capacity on an annual basis, units located in
noncontinental areas, or units firing very low sulfur oil
42
-------
are not required to achieve a percent reduction in S02
emissions, but need meet only certain emission limits.
3. Comment: A number of commenters questioned the legal basis for
establishing a percent reduction requirement as the basis of
the standard. Several (IV-D-26, IV-D-29, IV-D-30, IV-D-35,
IV-D-38, IV-D-40, IV-D-46, IV-D-50, IV-D-53, IV-D-58,
IV-D-62, IV-D-66, IV-D-74, IV-D-75, IV-D-81, IV-D-82,
IV-D-84, IV-D-85, IV-F-1.1, IV-F-1.7, IV-F-1.10, IV-F-1.13)
said the legislative history of the percent reduction
provision indicates that it was intended by Congress to
apply to utility steam generating units only, and not to
industrial steam generating units. Several (IV-D-26,
IV-D-30, IV-D-44, IV-D-50, IV-D-52, IV-D-53, IV-D-58,
IV-D-62, IV-D-66, IV-D-74, IV-D-75, IV-D-78, IV-D-84,
IV-F-1.7, IV-F-1.13) added that Section 111 gives the Agency
the flexibility to forgo a percentage reduction where there
is no demonstrated need for it, or where the costs are too
high compared to the benefits, or where a percent reduction
would create more problems than it solved. The commenters
said that all these things are true of this proposal. Some
(IV-D-26, IV-D-52, IV-D-56, IV-D-58, IV-D-62, IV-D-66,
IV-D-74, IV-D-75, IV-D-81, IV-D-84) said the Agency itself
relied on the "ambiguous legislative history" of the Clean
Air Act Amendments after promulgation of the utility steam
generating unit standards when it was concluded that there
was no obligation to revise the existing large industrial
steam generating unit NSPS to include a percent reduction
requirement. The commenters referred to a brief filed by
the Agency in Sierra Club v. Ruckelshaus in 1984. One
commenter (IV-D-96) said the legislative history of
Section 111 indicates that the percent reduction concept was
introduced into the statute specifically to avoid fuel
43
-------
switching away from coal. Thus, the commenter said, the
90 percent reduction requirement in the proposed standard
defeats the primary purpose of this clause. Another
(IV-D-64), however, stated that Section 111(a)(1)(A)(ii) of
the Clean Air Act mandates the percent reduction approach
and is clearly applicable to this standard.
Response: Section 111(a)(1) defines the term "standard of performance"
as follows:
"(A) with respect to any air pollutant emitted from a
category of fossil fuel fired stationary sources... a
standard -
(i) establishing allowable emission limitations for
such category of sources, and
(ii) requiring the achievement of a percentage
reduction in the emissions from such category of sources
from the emissions which would have resulted from the use of
fuels which are not subject to treatment prior to
combustion."
The percentage reduction requirement was enacted by the 1977
Amendments to the Act. The Conference report characterizes
this requirement as applying to "fossil fuel-fired sources"
generally, not limited to utility steam generating units [S.
Rep. No. 564, 95th Cong., 1st Sess. 130 (1977)]. The House
Report, where the percentage reduction requirement
originated, similarly refers to "fuel-burning new stationary
sources" generally [H. R. Rep. No. 294, 95th Cong., 1st
Sess. 188 (1977); id. at 188-192]. The Conference Report's
discussion of the percentage reduction requirement indicates
that the Agency has the authority to include a percentage
reduction requirement in an NSPS for fossil fuel-fired
sources [H. R. Rep. No. 564, 95th Cong., 1st Sess. 130
44
-------
(1977)]. None of the discussion in the committee reports or
the Congressional Record indicated the percent reduction
requirement was to be limited to only electric utility steam
generating units.
Although the discussion of the percentage reduction
requirement in the legislative history sometimes refers to
particular types of fossil fuel-fired sources, it is not
true that the only type referred to is a utility steam
generating unit. The legislative history also refers to
"boilers" generally and "industrial sources," lid. at 189]
and "mines, processing plants, and factories," lid. at 191]
TAccord. Sierra Club v. Ruckelshaus. Civil Action No.
84-0325 (D.D.C. Sept. 4, 1985), at 5].
The issue in Sierra Club was the scope of the
nondiscretionary duties under Section 111(b)(6). The Agency
did not express any view in that case about the application
to industrial steam generating units of the percentage
reduction requirement under Section 111(a)(1), but simply
argued that Section 111(b)(6) did not impose a
nondiscretionary duty to promulgate revised NSPS for
industrial steam generating units. Similarly, the utility
steam generating unit NSPS rulemaking cited in that brief
did not analyze the applicability of the percentage
reduction requirement to industrial steam generating units
[44 FR 33580 (June 11, 1979); 43 FR 42154 (Sept. 19, 1978)].
The legislative history of the percent reduction standard
does indicate that one of its purposes was to reduce
economic incentives to use low sulfur coal rather than
applying control technology on higher sulfur, but locally
available fuels. However, there is no indication in the
45
-------
legislative history of the percent reduction requirement
that Congress sought to restrict the use of natural gas or
other sulfur-free fuels.
Section 111 does provide the flexibility to forgo the
percentage reduction requirement if the impacts associated
with it would be "unreasonable." However, as discussed
throughout this rulemaking, a thorough analysis of the
economic, environmental, and energy impacts associated with
the standard was conducted, and no unreasonable impacts were
identified for the final standards. The preamble to the
proposed NSPS explained in detail the legal, policy, and
factual bases of the proposal, including an analysis of the
extent to which the standard would discourage the burning of
coal. Therefore, promulgation of the standard under
Section 111 was carefully considered and is considered
appropriate.
4. Comment: Several commenters (IV-D-8, IV-D-22, IV-D-44, IV-D-54,
IV-D-60, IV-D-61, IV-D-83, IV-F-1.2, IV-F-1.4, IV-F-1.6,
IV-F-1.9, IV-F-1.19) said only an emission limit, rather
than a percent reduction requirement, should be established
and sources should be allowed to achieve that limit by
whatever means is appropriate for each source. They said
that industry needs to have clearly understood nationwide
regulations on emission rates while retaining freedom to
choose the engineering method, fuel, and equipment to meet
those emission rates. The commenters felt the percent
reduction requirement preempts this freedom of choice. One
conmenter (IV-D-62) said the language of Section
lll(b)(l)(B)(5), stating that the Administrator is not
allowed to require any new or modified source to install a
particular technological system, also means that the use of
46
-------
low sulfur fuel as a method of compliance cannot be
disallowed. The commenter added that while a demonstrated
percent reduction may be used to establish an emission limit
for new sources based on the dirtiest fuel commercially
available, Section 111 does not authorize the inclusion of
such a percentage reduction as a part of the standard of
performance. According to the commenter, the percent
reduction requirement of the Clean Air Act does not
authorize the setting of a minimum efficiency level that
control equipment must meet. Instead, the commenter said,
it authorizes the Administrator to establish percent
reduction needed via fuel cleaning, pollution control
equipment, etc., for sources using the highest polluting
fuel coiranercially available.
Response: The percent reduction requirement does not force a source to
install any particular technological system of SO2 emission
reduction. It is simply a performance standard that is
based on the performance capabilities of the "best
demonstrated technology" (i.e., flue gas desulfurization or
fluidized bed combustion). Sources are free to use any
technological system that allows compliance with the
standard.
The standard does not require that post-combustion control
technology alone meet the percent reduction requirement. As
such, pretreatment of fuels prior to combustion can be used
to reduce the percent reduction required from the
post-combustion control system.
47
-------
5. Comment: One commenter (IV-D-96) stated that the Clean Air Act
mandates that emissions be controlled by technological
methods. The comnenter said fuel switching is not a
technological method of control and cannot be considered an
"adequately demonstrated" technology.
Response: Section 111 requires NSPS to be set at the level which
reflects the capabilities of the "best demonstrated
technology," i.e., the most effective technology that does
not impose unreasonable costs or other impacts [Essex
Chemical Coro. v. Ruckelshaus. 486 F. 2d 427 (D.C. Cir.
1973)]. The best demonstrated technologies for coal- and
oil-fired industrial-commercial-institutional steam
generating units are flue gas desulfurization and fluidized
bed combustion; therefore the NSPS has been established at
the level reflecting the capabilities of these technologies.
In so doing, the cost and other impacts were considered. As
discussed elsewhere, those impacts are considered
reasonable. Some fuel switching from oil or coal to natural
gas will occur and is an alternative for owners or operators
of affected facilities. However, this does not change the
fact that the standards are based on use of flue gas
desulfurization technology and fluidized bed combustion, and
the impacts of requiring these technologies are reasonable
even without fuel switching considerations. The legislative
history shows that Congress intended that NSPS for coal- and
oil-fired sources should reflect technological systems such
as flue gas desulfurization and fluidized bed combustion [H.
R. Rep. No. 294, 95th Cong., 1st Sess. 183-195 (1977)].
That intent is carried out by this NSPS.
48
-------
2.4 STANDARD FOR S02
1. Comment: One commenter (IV-D-35) felt that coal cleaning should be
considered as a compliance option under the standard. The
commenter stated that the use of cleaned coal is an
. especially important emerging pollution control option, and
cautioned that a standard based solely on the more expensive
scrubbing technology ignores the potential for a more cost-
effective coal cleaning approach.
Response: Coal cleaning is a demonstrated method of removing sulfur
from coal. However, it is currently not capable of
achieving the percentage sulfur removal that is achievable
by the "best demonstrated technology," which is flue gas
desulfurization or fluidized bed combustion. Therefore, the
use of coal cleaning alone would not be sufficient to meet
the promulgated standards. In addition, coal cleaning alone
(which generally achieves only about 20 to 30 percent
reductions in potential SOg emissions) was not intended by
Congress to satisfy the provisions of Section 111 of the
Clean Air Act and, therefore, cannot serve as the basis for
: an NSPS.
Coal cleaning was, however, considered as a means of
producing low sulfur fuels, and the alternative of basing
standards on the combustion of low sulfur fuel was given
; full and complete consideration. In addition, the
1 definition of "potential SO2 emission rate" (60.41b) is
based on the emissions from the combustion of a fuel "in an
uncleaned state," meaning that reductions in SOg emissions
achieved through coal cleaning are creditable toward the
I percent reduction requirement.
49
-------
2. Comment: Several commenters (IV-D-6, IV-D-28, IV-D-40, IV-D-50,
IV-D-52) stated that because there are important differences
between industrial and utility steam generating units,
regulation of these two source categories should be based on
these differences rather than treating industrial and
utility units alike, which, according to the commenters, is
what the proposed regulation does. The commenters
specifically mentioned the following differences:
Industrial steam generating units produce much less
steam per hour than do utility steam generating units.
They said a typical utility unit produces 3.5 million
pounds of steam per hour, compared to only 0:1 million
pounds per hour for a typical industrial unit.
While utility units serve a single purpose - generating
steam at a relatively steady rate to produce
electricity, industrial units serve a variety of
different purposes in various industries. Therefore,
industrial steam generating unit design varies greatly
depending on the fuels burned, the application of the
steam produced, and the dally and seasonal load
variations. According to the commenters, even at a
single industrial operation the steam requirements can
change drastically from day to day, hour to hour, and
sometimes from minute to minute.
Industrial units often burn a wide variety of fuels and
process wastes available on-site, while utility units
burn a relatively homogeneous coal.
The commenters said that all of these characteristics
require a great deal of flexibility and reliability not
required by utilities and need to be considered 1n setting a
regulation.
50
-------
Response: The various design and operational characteristics unique to
industrial-commercial-institutional steam generating units
were considered in developing the standards. It is true
that industrial units are, in most cases, different from
utility units. However, the analyses of performance, cost,
and cost effectiveness of control technologies were
performed on industrial size units with design and operation
features typical of these units. The "model boiler"
analysis was performed using steam generating units of 29,
44, and 117 MW (100, 150, and 400 million Btu/hour) heat
input capacity rather than on larger size units.
Assumptions such as frequent startup and shutdown, and
variable steam demand resulting in load swings, were also
considered when assessing performance of the SO2 control technologies
and factored into the cost algorithms. To account for the smaller lot
size of the fuel purchases made by industrial operations, higher fuel
costs than those attributed to utilities were also assumed.
Differences in F6D malfunction assumptions were also taken into accoun
in the use of natural gas as a backup fuel rather than maintenance of
spare FGD module or steam generating unit shutdown, which is an option
for electric utilities due to the ability to purchase electricity froir
the "grid" system. In addition, the analysis examined the use of
package FGD systems, which are typical of industrial applications,
rather than larger, field-erected FGD systems, which are typical of
utility applications. Finally, separate analyses of mixed fuel-fired
units were performed to account for industrial use of wood, solid
waste, or other alternative fuels in addition to coal, oil, and natura
gas. Thus, the major differences between industrial and utility steair
generating units were considered and the requirements of the standards
reflect these considerations.
51
-------
3. Comment: Several commenters (IV-D-6, IV-D-7, IV-D-12, IV-D-28,
IV-D-40, IV-D-48, IV-D-50, IV-D-52, IV-D-74, IV-D-83,
IV-D-85) said a less stringent standard without a percent
reduction requirement, or no standard, should be set for
steam generating units with heat input capacities between 29
and 73 MW (100 and 250 million Btu/hour) because:
they represent only 18 percent of the aggregate
industrial-commercial-institutional steam generating
unit capacity and less than 0.3 percent of total U.S.
S(>2 emissions;
the S(>2 emissions generated by these units are
insignificant to health and welfare considerations;
capital-related costs for SOg control are
disproportionately higher due to a lack of economies of
scale;
coal transportation costs are higher due to an inability
to obtain volume shipping savings.
A number of commenters (IV-D-6, IV-D-26, IV-D-28, IV-D-30,
IV-D-40, IV-D-46, IV-D-50, IV-D-52, IV-D-53, IV-D-72,
IV-D-75, IV-F-1.16) also felt that even an emission limit of
516 ng/J (1.2 lb/million Btu) is not justified for steam
generating units with heat input capacities less than 73 MW
(250 million Btu/hour) on the basis of the volume of S02
emitted by this group and the disproportionate share of the
costs they would bear. They contended that a higher
standard, such as 688 or 1,033 ng/J (1.6 or 2.4 lb/million
Btu), could be applied to these units with no serious loss
in SOg reduction. According to the commenters, this
increase in the emission limit would produce significant
benefits to smaller users of coal, allowing them to purchase
less expensive, higher sulfur coal. Several of the
52
-------
commenters suggested that this emission limit be set at the
point which minimizes both delivered coal costs and overall
SO2 emissions.
Response: As discussed in the "Summary of Regulatory Analysis" and
"Revised Impacts of Alternative Sulfur Dioxide New Source
Performance Standards for Industrial Fossil Fuel-Fired
Boilers," various regulatory alternatives were considered
and analyzed for steam generating units with heat input
capacities less than 73 MW (250 million Btu/hour). These
alternatives included, as suggested by the commenters,
establishing an emission limit only for these smaller steam
generating units. However, Section 111 of the Clean Air Act
requires standards to reflect application of the "best
demonstrated technology" for which costs, nonair quality
health and environmental impacts, and energy requirements
are considered reasonable. The final standards achieve
greater emission reductions than would be achieved by an
emission limit only, and the costs, nonair quality health
and environmental impacts, and energy requirements of the
final standards are considered reasonable. Therefore, a
less stringent standard for these smaller steam generating
units is not appropriate under Section 111 of the Clean Air
Act.
Comment: Several commenters (IV-D-26, IV-D-28, IV-D-30, IV-D-40,
IV-D-50, IV-D-53, IV-D-83, IV-F-1.8, IV-F-1.16) said the
proposed standards should be withdrawn and a new proposal
submitted that is limited to an emission limit for units
larger than 73 MW (250 million Btu/hour). The commenters
said that if, on reanalysis, an emission limit is found
reasonable for units between 29 and 73 MW (100 and 250
million Btu/hour), that should also be submitted for further
53
-------
public comment as a reproposal. Others (IV-D-10, IV-D-26,
IV-D-45, IV-D-48, IV-D-62, IV-D-73, IV-F-1.12, IV-F-1.16,
IV-F-1.18) agreed, saying that steam generating unit owners
and operators should have the flexibility to use low sulfur
fuel. Another (IV-D-76) felt that the percent reduction
requirement achieves emission reductions in an inefficient
way. Without a percent reduction requirement, the commenter
said, an owner or operator could achieve the same emission
level at a lower cost by using lower sulfur coal or oil.
Response: With several exceptions, as discussed earlier, the impacts
of a percent reduction requirement were thoroughly examined
and found to be reasonable within the meaning of Section 111
of the Clean A1r Act for steam generating units larger than
29 MW (100 million Btu/hour). Therefore, the final
standards include a percent reduction requirement for all
size categories of industrial-commercial-institutional
steam generating units.
5. Comment: One commenter (IV-D-28) said that if the promulgated
standard includes a percent reduction requirement, limits
should be set that provide maximum compliance flexibility.
The commenter suggested that in order to reflect national
energy security and efficiency objectives, the flexibility
could be provided by a requirement 1n the range of
50 percent reduction. Another coranenter (IV-D-99) agreed,
saying that a 90 percent reduction requirement is far too
stringent and not in the nation's best interest. The
coiranenter felt that a more realistic performance standard
permitting use of less expensive coal burning technologies
(such as a 60 percent reduction requirement) would achieve
the dual objectives of S02 control and increased coal
use in the future.
54
-------
Response: The regulatory compliance flexibility of a 90 percent
compared to a lower percent reduction requirement was
carefully evaluated. While it is true that lower percent
reduction requirements (such as 50 percent) would allow the
use of a greater number of SO2 control technologies, this
alternative must be evaluated in relation to the
requirements of Section 111 of the Clean Air Act.
Section 111 requires the NSPS to be established at the level
achievable by the best demonstrated technology for which no
unreasonably adverse cost, environmental, or energy impacts
have been identified. This "best demonstrated technology"
, is flue gas desulfurization and fluidized bed combustion,
which have been demonstrated to achieve a 90 percent
reduction in SOg emissions. Thus, the impacts associated
with achieving a 90 percent reduction as required under the
final standard are considered reasonable.
To encourage the development of alternative SO2 control
technologies, a percent reduction requirement of 50 percent
has been established for emerging technologies. This
provision was discussed in the preamble to the proposed
standards and is retained in the final standards.
6. Comment: Several commenters (IV-D-22, IV-D-26, IV-D-30, IV-D-33,
IV-D-38, IV-D-43, IV-D-44, IV-D-48, IV-D-50, IV-D-53,
IV-D-55, IV-F-1.6) said the requirements in Subpart Da
should be the most stringent scenario for steam generating
units with heat input capacities greater than 73 MW (250
million Btu/hour). They said there is no reason industrial
units should be treated more stringently than utility units.
Two of the commenters (IV-D-43, IV-D-48) specifically stated
that a sliding scale of emission reductions, such as that
: provided in Subpart Da, should be applied to industrial-
55
-------
commercial-institutional steam generating units firing low
sulfur fuels.
Response: Direct comparison of the standard for industrial-commercial -
institutional steam generating units with the Subpart Da
standard promulgated in 1979 is inappropriate for several
reasons.
First, the design and operating characteristics of utility
steam generating units and emission control systems (covered
by Subpart Da) are different from units covered under these
standards. Utility steam generating units are generally
much larger than affected facilities covered under these
standards. A typical 500 MW electric utility unit would
have a heat input capacity of 1,450 MW (5,000 million
Btu/hour). This compares to 44 MW (150 million Btu/hour)
for an industrial type unit, a 30-fold difference in size.
Because of their large size, utility steam generating units
and FGD systems are field erected and are usually custom
designed for a specific site. In the case of coal,
long-term fuel purchase contracts for up to 20 years are
common. As a result, utility steam generating units and FGD
systems can be designed to optimize site- and fuel-specific
factors. Also, to handle the large quantities of flue gas
produced by utility steam generating units, utility FGD
systems typically consist of multiple parallel scrubber
modules (typically 4 or 5 modules), with each capable of
handling part of the total flue gas. To ensure operating
reliability of the total FGD system, an additional FGD
module 1s generally installed as a spare to provide backup
for a module that malfunctions or is idled for preventive
maintenance.
56
-------
In contrast, industrial-commercial-institutional steam
generating units and F6D systems are smaller and are more
likely to be package units that are designed to provide the
operator with maximum flexibility in future fuel purchasing.
Long-term fuel supply contracts are less common with
industrial applications. Industrial-commercial -
institutional steam generating units and FGD systems are
likely to be shop assembled based on standardized designs
and have the ability to handle a wider range of fuels than
typical utility steam generating units. These FGD systems
generally consist of single modules without spares. Because
of their smaller size, switching to natural gas or low
sulfur oil during FGD system malfunctions is a viable
alternative to installing spare FGD modules.
Second, at the time Subpart Da was promulgated,
lime/limestone wet scrubbing was the predominant FGD
technology used by utilities. Newer technologies, such as
lime spray drying FGD, were still in the early stages of
commercial application and concerns existed about the
ability of these newer technologies to achieve 90 percent
SO2 reduction on a reliable basis. Today, a number of
demonstrated FGD technologies including sodium scrubbing,
lime spray drying, and dual alkali, as well as fluidized bed
combustion are available for use by industrial-commercial -
institutional steam generating units. Based on experience
gained with these technologies during the past decade,
including improved preventive maintenance programs, as well
as various technical advantages of these technologies over
wet lime/limestone FGD, industrial FGD systems are expected
to be more efficient and reliable than the utility FGD
systems were a decade ago.
57
-------
Because of the many differences between industrial-
commercial-institutional and utility steam generating units,
it is inappropriate to conclude that the final Subpart Db
standards must mimic Subpart Da.
7. Comment: Two commenters (IV-D-97, IV-F-1.4) noted that the proposed
standards offered four alternative compliance methods, none
of which is feasible in New York City for large-scale
district steam generation. The commenters gave the
following examples:
the use of coal with scrubbers is prohibited by a ban on
coal burning imposed by the City's Air Pollution Control
Code;
the use of high sulfur oil with scrubbers is prohibited
because the Code does not permit the purchase, use, or
storage of high sulfur fuel oil, and this alternative
would be much more costly than firing very low sulfur
oil to meet the same emission level,
the option of burning natural gas 1s unacceptable
because there 1s no assurance of a continuous, long-term
supply;
the use of very low sulfur (0.2 percent) oil is
infeaslble because oil with such a low sulfur content is
not readily available on the East Coast and, even if it
were available, the cost would be prohibitive compared
to the 0.3 percent sulfur oil currently 1n use.
The corranenters requested that "very low sulfur oil" be
defined as 0.3 percent sulfur to provide a reasonable
compliance option for their facilities.
Response: The commenter's claim that there are no feasible compliance
options under the standards for sources in New York City was
investigated. It is true that the use of coal, with or
58
-------
without an FGO system, would likely be precluded due to the
generally tight space restrictions in urban areas and waste
disposal constraints. The use of high sulfur oil with an
FGD system could be permitted under certain circumstances.
The commenter's analysis assumed the use of a dual alkali
FGO system due to suggested liquid waste disposal
constraints, which would noticeably increase annualized
costs over those incurred by other types of FGD systems,
such as a sodium FGD system. While this may be true, the
costs associated with the use of the more expensive dual
alkali FGD systems are considered reasonable, as are the
impacts of the standard under the situation described by the
commenter.
The use of natural gas is also a less costly alternative and
would probably be a more attractive compliance option for
such applications in most cases. The commenter's concerns
about the reliability of natural gas supplies are not borne
out by current data. Natural gas availability would be
expected to compare favorably to that of the low sulfur oil
currently being fired by the commenter. If availability
concerns remain, the source could fire natural gas with very
low sulfur oil as a backup fuel to ensure continuous
operation, or it could simply fire very low sulfur oil.
Additionally, the final standards provide an exemption from
the percent reduction requirement for steam generating units
operating at annual capacity utilization factors of less
than 30 percent for oil or coal. Thus, the final standards
provide two options that would not require the use of an FGD
system. First, natural gas could be used as the primary
fuel with a backup supply of very low sulfur oil maintained
for any periods of gas supply interruption. Second, 1f
natural gas is not available or is more expensive, very low
sulfur oil could be fired.
59
-------
8. Comment: Some commenters (IV-D-15, IV-D-49, IV-D-55, IV-D-57,
IV-D-82) said the proposed S02 emission limit of 86 ng/J
(0.2 lb/million Btu) for an oil-fired steam generating unit
without an FGD system is unrealistic. They said that at
present, a residual oil with a sulfur content of 0.2 percent
is not commercially available. According to the commenters,
the lowest sulfur specification for commercially available
oil is 0.3 percent sulfur. Two other commenters (IV-D-31,
IV-D-34) said the compliance level for very low sulfur oil
should be raised from 86 ng/J (0.2 lb/million Btu) to 215 or
258 ng/J (0.5 or 0.6 lb/million Btu) (equal to an oil of
roughly 0.5 weight percent sulfur or less) because:
oil with a sulfur content equivalent to 86 ng/J
(0.2 lb/million Btu) is not generally available;
most low sulfur residual fuel used in the U.S. is
manufactured by distillation of crude oils yielding
0.2 to 0.5 percent sulfur fuel;
it is not cost effective to desulfurize or to use FGD
systems on fuel oils containing less than
0.5 percent sulfur. The incremental cost effectiveness
is well over the $2,400/Mg ($2,200/ton) cutoff used
elsewhere in this NSPS;
it does not seem reasonable to require construction and
operation of FGD systems with low sulfur fuels when the
SOg emissions without FGD systems are the same or even
lower than when scrubbing emissions from high sulfur
fuels.
60
-------
Response: The basis for the emission limit of 86 ng/J
(0.2 lb/million Btu) in the proposal was to provide an
alternative means of demonstrating compliance with the
percent reduction requirement. The sulfur content of this
oil was so low that it appeared reasonable to assume that
refining the original crude oil had resulted in a
90 percent reduction in potential SO2 emissions. This
provision, therefore, was not based on an assessment of the
availability of such oils, but merely on an assessment of
how low the sulfur content of an oil would need to be in
order to reasonably assume a 90 percent reduction in its
sulfur content had already been achieved.
As a result of these comments, the basis for this provision
was reviewed. A much more detailed assessment than that
done at the time of proposal was undertaken to determine the
means by which most very low sulfur oils are produced.
This assessment concluded that most very low sulfur oils,
even those as low as 86 ng/J (0.2 lb S02/million Btu), are
not produced by desulfurization, but by distillation of very
low sulfur crude oils. As a result, there is no point at
which the sulfur content of an oil is so low that one can
reasonably assume with some confidence that the production
of this oil resulted in a 90 percent reduction in its sulfur
content. Consequently, the final standards contain no
provisions similar to those proposed that provide an
alternative means of demonstrating compliance with the
90 percent reduction requirement.
Although the final standards do not contain these
alternative provisions for demonstrating compliance with the
90 percent reduction requirement based on the sulfur content
of the oil, as discussed earlier, the final standards do
61
-------
exempt certain steam generating units from the percent
reduction requirements. This is the case when the impacts
associated with the percent reduction requirements are
considered unreasonable (i.e., low capacity factor and mixed
fuel-fired steam generating units, steam generating units in
noncontinental areas, and steam generating units firing very
low sulfur oil). These steam generating units, however, are
subject to an SOg emission limit which, in the case of
oil-fired units, is 129 ng/J (0.3 lb SOg/million Btu).
Unlike the 86 ng/J (0.2 lb/million Btu) emission limit
provided at proposal, the 129 ng/J (0.3 1b/m1111 on Btu)
emission limit for steam generating units exempt from the
percent reduction requirement is based on an assessment of
the emissions, costs, and availability of such oils. As
cited by the commenters, this assessment found that the
lowest sulfur content specification placed on commercially
available oils is generally 129 ng/J (0.3 lb/million Btu).
Oils with such low sulfur contents, however, are widely
available. In some areas, these oils may be residual oils
and in other areas they may be distillate oils ("Availabi-
lity of Very Low Sulfur Fuel Oil," 1987). In either case,
such oils are available and the costs associated with
their use are considered reasonable. Consequently, where
oil-fired steam generating units are exempt from the percent
reduction requirements in the final standards, they are
subject to an SOg emission limit of 129 ng/J (0.3 lb
S02/million Btu).
9. Comment: One commenter (IV-F-1,4b) questioned whether firing a very
low sulfur coal with less than 90 percent removal as long as
it met an 86 ng/J (0.2 lb S02/million Btu) emission limit
would also be acceptable. Another (IV-F-1.4c) noted that if
62
-------
an oil with a sulfur content of 1 percent is fired and
emissions are 86 ng/J (0.2 lb/million Btu), only 80 percent
reduction has been achieved but the standard deems this
acceptable. However, the commenter said, if a 1 percent
sulfur coal is fired, 90 percent reduction down to 43 ng/J
(0.1 lb/million Btu) is required. The commenter felt this
was inequitable. Another commenter (IV-D-93) felt that
anthracite should be.considered a "very low sulfur coal" for
compliance purposes and thus be exempt from the percent
reduction requirement. The commenter contended that the
Pennsylvania anthracite deposits represent the largest
single coal reserve in the eastern United States which can
regularly meet a 430 ng/J (1.0 lb/million Btu) emission
limit. According to the commenter, this is due to a
combination of low inherent sulfur content and strict
preparation plant quality control, which ensures a high
heating value.
Response: As discussed above, the final standards do not contain an
emission limit that allows the operator of an FGD system to
achieve less than 90 percent S02 control. For steam
generating units operated at low capacity factors, located
in noncontinental areas, or firing very low sulfur content
oils, an emission limit must be met, but an FGD system is
not required. However, for all other oil- or coal-fired
units, operation of the FGD system at a 90 percent
performance level is required.
63
-------
10. Comment: Several commenters (IV-D-23, IV-D-26, IV-D-30, IV-D-50,
IV-D-53, IV-D-55) said the precombustion coal cleaning
"credit" in the proposed standards does not reflect the
clear meaning of Section 111(a)(1) of the Clean Air Act.
They asserted that full credit against the final percentage
reduction should be allowed for any precombustion cleaning,
including pulverizer rejects and flyash interactions, as was
done in Subpart Da.
Response: Under both the proposed and final standards, full credit
toward complying with the 90 percent reduction requirement
is allowed for all types of precombustion fuel cleaning
technologies, including pulverizer and flyash rejects.
Credit for sulfur removal in the coal bottom ash and flyash
is achieved under the final regulation in the optional
"as-fired" fuel sampling procedures under the SO2 emission
monitoring requirements. By monitoring SOg emissions (ng/J,
lb/million Btu) with an as-fired fuel sampling system
located upstream of coal pulverizers and measuring SOg
emissions with an in-stack continuous SOg monitoring system
downstream of the FGD or FBC system, sulfur removal credits
for the coal pulverizer, bottom ash, and flyash are combined
with removal achieved by the FGD system into one overall
removal efficiency.
11. Comment: Some conmenters (IV-D-15, IV-D-26, IV-D-30, IV-D-33,
IV-D-45, IV-D-50, IV-D-53) said the proposed tracking
procedure for obtaining credits toward the percent reduction
requirement for precleaning oil and coal is unrealistic and
incompatible with mining, refinery, and fuel blending
operations. They felt that low sulfur fuel should be
considered as a pretreated fuel and no pretreatment
documentation should be required.
64
-------
Response: As discussed earlier, review of the means by which very low
sulfur oils are produced concluded that no general
correlation is possible between the sulfur content of a
particular oil and the amount of percent reduction (i.e.,
desulfurization) it has undergone. Therefore, the proposed
requirements for documenting the percent reduction obtained
through the pretreatment of coal and oil have been retained
in the final standards. In order to obtain credit for the
reduction in SO2 emissions achieved by any fuel pretreatment
process, a steam generating unit operator must be able to
certify that the fuel being fired actually had undergone
pretreatment and the extent to which this pretreatment
reduced the sulfur content of the fuel. To this end, both a
certified statement from the fuel pretreatment facility
specifying the amount of sulfur removed and documentation
tracking the shipment of the fuel to the affected facility
are necessary. If no documentation were required, steam
generating unit operators would be able to purchase fuel
with a lower sulfur content and claim a pretreatment credit
when no pretreatment occurred.
65
-------
2.5 STANDARD FOR PARTICULATE MATTER
1. Comment: Two commenters (IV-D-37, IV-D-80) suggested that the
particulate matter standards recognize the differences among
oil grades and specify the oil grades to which the
particulate matter emission limits apply. One (IV-D-80)
said that steam generating units which burn distillate oil
are capable, without any emissions control, of emitting
particulate matter at a rate of less than 8.6 ng/J
(0.02 lb/million Btu) heat input, and are currently required
to do so in some States. Therefore, the commenter said, the
particulate matter emission limit for distillate oil-fired
steam generating units of 43 ng/J (0.10 lb/million Btu) is
too high and should be reduced to 8.6 ng/J (0.02 lb/million
Btu).
Two commenters (IV-D-64, IV-D-80) said that it is
inequitable to establish a particulate matter standard that
allows oil-fired units to emit at twice the rate allowed for
coal-fired units [coal at 27 ng/J (0.05 lb/million Btu)
versus oil at 43 ng/J (0.10 lb/million Btu)]. One commenter
objected that residual oil is an inherently cleaner fuel
than coal and therefore should be able to achieve at least
the level of particulate matter control required for
coal-fired units. The other commenter contended that the
allowance of a more lenient standard for oil-fired units
offers an Inappropriate subsidy for the use of oil 1n
Industrial-commercial- Institutional steam generating units.
This commenter said that the use of electrostatic
precipitators on oil-fired units would achieve substantial
improvement in particulate matter control over scrubbers
alone.
66
-------
Response: As discussed in the Summary of Regulatory Analysis prepared
for the proposed standards, data collected on the
performance of wet FGD systems on residual oil-fired steam
generating units demonstrate that this control system is
capable of reducing particulate matter emissions from
residual oil-fired units to 43 ng/J (0.10 lb/million Btu).
In order to reduce particulate emissions beyond the level
achieved by the scrubber, an electrostatic precipitator
(ESP) would be required. Data collected from residual
oil-fired units controlled by ESP's (both with and without a
scrubber) showed emissions of 22 ng/J (0.05 lb/million Btu).
The incremental cost effectiveness of a 22 ng/J (0.05
lb/million Btu) standard based on the use of an ESP plus an
FGD system compared to the use of the FGD system alone was
estimated to be more than $11,000/Hg ($10,000/ton). This is
considered unreasonable for general application;
consequently, the standard limits emissions to 43 ng/J (0.10
lb/million Btu) for oil-fired steam generating units.
Distillate oil-fired steam generating units may be capable
of achieving particulate matter emission rates as low as 22
ng/J (0.05 lb/million Btu) or possibly less. However,
the same emission rate will be achieved whether the standard
for distillate oil is set at 43 ng/J (0.10 lb/million Btu)
or 22 ng/J (0.05 lb/million Btu), because no emission
control system is necessary to reduce emissions from
distillate oil to these levels. Thus, it makes no
difference whether the standard for distillate oil-fired
steam generating units is 43 or 22 ng/J (0.10 or 0.05
lb/million Btu). Consequently, for convenience and to keep
the testing and reporting requirements as simple and easy to
understand as possible, the final standard limits
particulate matter emissions to 43 ng/J (0.10 lb/million
67
-------
Btu) for all types and grades of oil. In summary, the
emission standard for oil has been set at a higher level
than coal because of cost-effectiveness considerations.
2. Comment: Several commenters (IV-D-26, IV-D-30, IV-D-50, IV-D-53,
IV-F-1.15) said the sodium scrubber test data presented at
proposal do not support a particulate matter emission limit
of 43 ng/J (0.1 lb/million Btu) for oil-fired steam
generating units and are not representative of typical
industrial-commercial-institutional steam generating unit
performance. The commenters added that critical variables
affecting scrubber performance were not recorded.
Specifically, they said that site-specific information about
operating factors, fuel related parameters, and scrubber
parameters must be recorded and considered because they can
have a significant impact on particulate matter formation
and control. According to the conmenters, it is impossible
to extrapolate test data from a small sample population to
all industrial steam generating units without consideration
of these variables.
Response: The data cited in the "Summary of Regulatory Analysis"
were taken from seven oil-fired steam generating units using
wet scrubbers for both particulate matter and S02 control.
The seven wet scrubbing units represented four different
designs and fired oils having potential SOg emissions of 473
to 1,204 ng/J (1.1 to 2.8 lb/million Btu) heat input under a
variety of site-specific operating parameters. Particulate
matter emissions from these seven units ranged from 22 to 43
ng/J (0.05 to 0.10 lb/million Btu) heat input.
Consequently, the test data clearly support a particulate
matter emission limit of 43 ng/J (0.10 lb/million Btu).
68
-------
2.6 NATIONAL IMPACTS
2.6.1. Fuel Market and Energy Impacts
1. Comment: Several commenters questioned the policy of encouraging the
use of natural gas for industrial applications. Some
(IV-D-22, IV-D-54, IV-D-60, IV-D-81, IV-D-84, IV-D-87,
IV-F-1.1, IV-F-1.6, IV-F-1.11) felt that natural gas should
be reserved for uses demanding a more versatile fuel supply,
such as residential, institutional, agricultural, and
transportation uses. Others (IV-D-58, IV-D-62, IV-D-66,
IV-D-74, IV-D-84) said the analyses have not shown that
natural gas is available, practical, and reasonably priced
in all areas of the country. Several commenters (IV-D-26,
IV-D-30, IV-D-50, IV-D-52, IV-D-53, IV-F-1.16) also
expressed concern that if steam generating unit owners
become totally dependent on natural gas to the exclusion of
other fuels, and should curtailments such as those
experienced in the 1970's recur, plant closures could result
due to this loss of fuel flexibility. One commenter
(IV-D-31) said if prices and drilling rates remain at
current very low levels, there may not be sufficient
domestic production capability by 1988 to satisfy industrial
and electric utility natural gas demand. Another
(IV-F-1.18) asserted that no one is buying oil- and
gas-fired steam generating units even with current low
prices, because industry officials have no faith that oil
and gas prices will remain low. Still others (IV-D-31,
IV-D-33, IV-D-51, IV-D-81, IV-D-88, IV-F-1.16) said that
natural gas is not available in all areas of the country,
and in some places, it is available only on an interruptible
basis. In addition, the commenters said that natural gas
currently costs much more than coal or oil in many
locations.
69
-------
Response: The natural gas shortages and curtailments experienced
during the 1970's were not indicative of a shortage of
natural gas reserves, but were rather a response to adverse
supply and demand situations. Because Federal regulation of
natural gas prices kept these prices well below the costs of
alternative fuels, there was a large demand for natural gas.
This led to large decreases in available natural gas
reserves as production exceeded reserve additions, causing
regional shortages in some areas. However, with passage of
the Natural Gas Policy Act of 1978, deregulating the price
of most natural gas by January 1, 1985, an incentive to
locate additional reserves was created and drilling
Increased. This Increased natural gas supply and market
competition with oil created a gas surplus, or "bubble."
This bubble is currently estimated by both the U.S.
Department of Energy (DOE) and the American Gas Association
(AGA) at about 3 trillion cubic feet (Tcf). Although
drilling is currently low, it is anticipated that as this
bubble is used up producers will again have an Incentive to
continue drilling to build up reserves. The American Gas
Association, as well as various reports 1n the literature,
have predicted that gas reserves and demands will come into
balance in the next couple of years.
There has been some concern among corranenters that gas
shortages will occur in the short term -- i.e., 1n the
period between depletion of the gas bubble and resumption of
drilling activity. According to the AGA, there are several'
sources of auxiliary gas supplies that could be made
available on short notice (less than 12 months) to serve as
a "bridge" while industry explores and develops new
long-term supplies. These include Canadian and Mexican
imports, uncoimnitted nonproducing reserves, accelerated
70
-------
infill drilling (drilling into known gas fields), and
liquified natural gas. According to the AGA, these sources
should be more than sufficient to satisfy short-term demand
during resumption of drilling activity.
Projections of natural gas demand through the remainder of
the 1980's have been made by several organizations,
including the AGA and the National Petroleum Council. These
projections are generally within the range of 16 to 18
Tcf/year, with even lower demand expected after 1990. With
a resumption of drilling activity, natural gas supplies
should be more than sufficient to meet this demand. In any
event, fuel switching as a result of these standards
represents less than 1 percent of total U.S. natural gas
consumption. Therefore, the standards will have no
significant impact on overall natural gas supply and demand,
and will not affect the use of natural gas for other
purposes, such as residential, agricultural, or
transportation uses.
While the final standards will encourage the use of natural
gas to some extent, the price of natural gas relative to
other fuels will have a much greater impact on fuel use than
any requirements associated with these standards. Recent
declines 1n the price of natural gas, as well as projections
of continued low prices, indicate that natural gas will
represent the fuel of choice in a major portion of new
industrial-commercial-institutional steam generating units
even in the absence of standards.
For many years, natural gas has been a major energy source
for the industrial sector. A widespread transmission and
distribution system has been developed which serves all
71
-------
areas of concentrated industrial production and even many
isolated plants. Although firm data are lacking, the
largest share of new steam generating units will probably be
installed in existing plants where natural gas hookups
already exist. In some specific instances (such as
noncontinental areas), the availability of natural gas will
be limited. Therefore, an exemption from the percent
reduction requirement has been granted for noncontinental
areas, and facilities in these areas will be allowed to fire
low sulfur fuel to meet the pertinent emission limits.
Another concern raised by commenters, that natural gas
prices could become prohibitive in the future, is also not
supported by the data. The AGA projects that U.S. gas
prices will decline by more than 7 percent annually through
1988 and increase at about the same rate as overall
inflation thereafter. These near-term price declines are
expected as natural gas experiences competition from
residual oil due to current low oil prices and the reaction
of the gas market to recent declines in wellhead prices. As
this occurs, the price advantage historically enjoyed by
coal will shrink, bringing the prices of gas, oil, and coal
closer together. Although delivered coal costs will remain
lower than delivered gas costs, the higher capital,
operating, and maintenance costs associated with firing coal
will tend to make gas more attractive, especially for
smaller industrial applications.
Also, an interruptlble gas contract can be purchased at
lower cost than a firm, noninterruptible contract. As
discussed earlier, the final standard accommodates sources
that choose to purchase natural gas under interruptible
supply contracts and use a backup oil supply by providing an
72
-------
exemption from the percent reduction requirement as long as
very low sulfur oil is fired.
2. Comment: One commenter (IV-D-48) asserted that the proposed
regulation, by encouraging natural gas use, directly
conflicts with the Powerplant and Industrial Fuel Use Act of
1978, which states that natural gas "...shall not be used as
a primary energy source in a new major fuel-burning
installation consisting of a boiler "
Response: Since passage of the Powerplant and Industrial Fuel Use Act
in 1978, the national energy climate has undergone a
significant transformation. Domestic natural gas reserves
have increased and natural gas is no longer in short supply.
Although multi-fuel capability is now an important
consideration in steam generating unit design, natural gas
continues to be the fuel of choice for many industrial-
commercial -institutional applications. In addition,
provisions were included in the Powerplant and Industrial
Fuel Use Act which provide for waivers from this prohibition
against the use of natural gas in cases where this would
result in a substantial increase in costs. In actual
practice, the Fuel Use Act has been administered in a manner
which recognizes this change in energy conditions and the
economic appeal of natural gas. Examination of the record
of requests to the Department of Energy for waivers
indicates that very few requests have been denied. In fact,
repeal of major provisions of the Fuel Use Act occurred in
mid-1987.
73
-------
Therefore, the final standards are believed to further both
national energy and environmental goals, and are consistent
with past energy and environmental practice.
3. Comment: One commenter (IV-F-1.9) said that removing low sulfur coal,
one of the most abundant and economical fuels, from the new
steam generating unit market will guarantee an exponential
increase in the cost of fuels for industrial-
commercial -institutional steam generating unit operators.
Response: While competition from coal can influence the prices of gas
and oil to some degree, the prices of gas and oil are
primarily "driven" by other factors, such as direct
marketplace competition between gas and oil, international
trade regulations and foreign price controls, and domestic
regulation of fuel prices. As a result, the impact of the
standard is unlikely to significantly increase the costs of
fuels for industrial-commercial-institutional steam
generating units.
Both low and high sulfur coal can be used to meet the
percent reduction requirement. Analyses of the impacts of
the standards examined the total costs of generating steam,
including annual fuel costs, annual nonfuel operating and
maintenance costs, and levelized capital costs. By
requiring a percent reduction in SOg emissions from all
regulated fuels, these standards do not discriminate against
the use of either high or low sulfur coal. In addition, the
final standards allow low sulfur coal to be fired without a
percent reduction requirement in "low capacity factor"
coal-fired steam generating units, as discussed previously.
74
-------
4. Comment: Several commenters felt that the standard conflicted with
the national energy policy of encouraging coal use, an
aspect that was not addressed in the proposal. Some
(IV-D-40, IV-D-72, IV-F-1.19) were concerned that the policy
of encouraging domestic coal use was defeated by the
proposed standards. Others (IV-D-12, IV-D-26, IV-D-27,
IV-D-32, IV-D-36, IV-D-39, IV-D-44, IV-D-52, IV-D-58,
IV-D-62, IV-D-66, IV-D-74, IV-D-75, IV-D-84, IV-D-87,
IV-D-96, IV-D-98, IV-F-1.12) said the proposal, by
encouraging the use of natural gas and distillate oil, would
interfere with our national need for increased versatility
in energy sources and force the U.S. further into the
vulnerable position of reliance on foreign and interruptible
energy supplies. One commenter (IV-D-28) expressed the
opinion that the proposal will lead to a significant change
in the mix of fuels used by new facilities. The commenter
felt that significant market distortions without concomitant
gains for society, such as significant environmental
improvements, are contrary to the National Energy Policy
Plan and public policy.
Response: The revised fossil fuel price scenarios indicate that, with
the recent downturn in oil and gas prices, the costs of
generating steam from oil and natural gas are competitive
with those of generating steam from coal. In fact, based on
economic factors alone, it is expected that natural gas and,
to some extent, oil will claim the largest share of the
industrial-conmercial-institutional steam generating unit
market even in the absence of the standards. Thus, the
current level of oil and natural gas prices will have much
more of an impact on coal use in industrial-commercial-
institutional steam generating units than the final
standards. In addition, with oil and gas prices at current
75
-------
levels, national energy policies originally established to
encourage the use of coal have been amended somewhat, with a
focus on encouraging the use of a more versatile mixture of
domestic fuels. This includes domestically produced natural
gas as well as coal. While the standards may cause a larger
proportion of new steam generating units to fire natural gas
than might otherwise do so, much of this "fuel switching"
will be from oil to gas rather than from coal to gas. In
any event, the total amount of fossil fuel demand by
industrial-commercial-institutional steam generating units
is only a very small percentage of total U.S. demand, and
any changes in the fuel mix in this sector will not be
sufficient to cause "significant market distortions."
5. Comment: A number of commenters questioned the effects of the
standard on the industrial coal market. Several (IV-D-6,
IV-D-7, IV-D-17, IV-D-26, IV-D-27, IV-D-28, IV-D-35,
IV-D-36, IV-D-38, IV-D-40, IV-D-50, IV-D-52, IV-D-54,
IV-F-1.11) said the proposed standards would practically
eliminate coal as a viable fuel choice for new industrial
steam generating units, causing displacement of 5.1 to
9 million Mg (5.6 to 10 million tons) per year of potential
coal use by natural gas. Others (IV-D-14, IV-D-17, IV-D-26,
IV-D-29, IV-D-38, IV-D-40, IV-D-46, IV-F-1.8) said that in
addition to a reduction in the future market for coal, the
proposal will result in a reduction of coal's share of the
existing industrial market. According to the commenters,
this will occur because many future steam generating units
will be for replacement of existing capacity. They said the
standards will result in the failure of coal to capture a
share of this replacement market, thus losing a major
portion of the existing market. The commenters claimed that
by the years 2000-2020, the coal industry could face a
76
-------
market share loss of approximately 45 million Mg/year (50
million tons/year), an impact that was not addressed in the
proposed regulation.
Other commenters (IV-D-6, IV-D-10, IV-D-26, IV-D-28,
IV-D-40, IV-D-50, IV-D-52, IV-D-53, IV-D-98, IV-F-1.12,
IV-F-1.8) said the proposed rule would needlessly drive
consumers away from coal, this nation's most secure and
reliable energy source. They questioned this policy, saying
that to remain competitive in national markets, industrial
steam users must be assured a reliable supply of fossil
fuel. One commenter (IV-D-96) said that a comparison of
total costs (on a "per Btu" basis) for coal- and oil-fired
steam generating units complying with the proposed standards
shows that in order for coal to compete effectively with oil
and natural gas as an industrial steam generating unit fuel,
oil prices would have to rise to $35-45/barrel. The
commenter added that given current and anticipated petroleum
prices, this implies that the proposed standards effectively
preclude any burning of coal in new or reconstructed
industrial steam generating units.
Response: The analyses of the national impacts associated with the
proposed and promulgated standards focused on impacts in the
fifth year following proposal of the standards. Two sets of
national impacts projections were prepared prior to the June
1986 proposal. One of these was labeled the "High Oil
Penetration" energy scenario, which reflected a "best
estimate" of future coal, oil, and natural gas prices in
1984. Under this set of assumptions, it was projected that
there would be no displacement of projected industrial coal
demand in new steam generating units due to the proposed
standards. For sensitivity analysis purposes, results for a
77
-------
second scenario, labeled the "High Coal Penetration" energy
scenario, were presented. This scenario was based on
significantly higher oil and natural gas price forecasts.
Under this set of assumptions, a potential 5 million ton
annual displacement of industrial coal demand due to the
proposed standards was projected by the fifth year following
proposal.
Energy price scenarios and the national impacts analysis
were updated between proposal and promulgation, using lower
coal, oil and natural gas price projections. The revised
national impacts analysis concluded that the promulgated
standard will have a negligible effect on coal use in new
industrial steam generating units over the next 5 years.
Using these updated industrial fuel price projections, very
little coal use is forecast in the baseline because the fuel
price differential between coal and natural gas is too small
to pay for the higher nonfuel costs of coal-firing. Because
little coal is forecast in the baseline (based on economic
considerations alone), the updated national impacts analysis
indicates that the promulgated standard will have little
impact on coal use in the fifth year following proposal.
While it is recognized that new coal-fired steam generating
units will continue to be built even with the unfavorable
economics indicated by the new price projections, the
selection of coal as the fuel of choice in these instances
will often be based on site-specific factors or noneconomic
cosiderations.
A few comments suggested that long-term impacts should be
considered and that the annual displacement of industrial
coal demand due to the proposed standards could increase to
approximately 45 million Mg (50 million tons) by the years
78
-------
2000-2020. The concern of these commenters was that
existing coal-fired steam generating units will gradually be
replaced by oil/gas-fired steam generating units, resulting
in an absolute decline in industrial steam generating unit
coal use. Recent data suggest, however, that the life of
many existing steam generating units is being extended and
that those units that are replaced due to poor conditions or
inadequate capacity are generally replaced by new units
firing the same fuel as the existing unit.
However, the most important implication of the national
impacts analysis is that in nearly all of the sensitivity
cases evaluated, and certainly in the most recent analyses
reflecting lower oil and natural gas prices, it is the
relatively low prices for oil and gas, not the final
standards, which may lead to reduced coal use in new
(including replacement) steam generating units. Hence, 1f
the absolute level of industrial coal use declines in the
future, it is more likely to be due to low oil and natural
gas prices than to the final standards.
6. Comment: One commenter (IV-D-28) expressed concern about any
regulation that discourages coal use, unless some degree of
assurance can be offered that the exodus from coal will not
be to oil. The commenter said overly stringent coal
regulations, which discourage coal use, send an undesirable
signal to the marketplace that the Administration does not
i
place a high priority on reducing oil imports. The
commenter added that this discourages technology development
in the private sector, and could encourage accelerated
escalation of oil prices.
79
-------
Response: The promulgated standards include regulations for oil
combustion as well as coal combustion. Because the same 90
percent reduction requirement is generally being applied to
both coal and oil combustion, it is expected that any
"exodus" from coal would not be to oil, but to natural gas.
Under standards requiring the achievement of a percentage
reduction in SOg emissions from both coal- and oil-fired
steam generating units, natural gas will tend to be more
competitive than oil in new steam generating units. If
anything, the standards could reduce the use of imported oil
in new steam generating units in favor of domestic supplies
of natural gas and, as a result, are unlikely to result in
any escalation of oil prices.
Even in the unlikely event that some exodus from coal to oil
occurred, it would represent only a very small fraction of
the current oil market in the U.S. The national impacts
analysis is based on the assumption that sales of new
industrial-commercial-institutional steam generating units
over the next 5 years will result in total annual fossil
fuel consumption of 525 PJ (498 trillion Btu). This
represents less than 2 percent of the total current U.S. oil
market (about 32,500 PJ or 31,000 trillion Btu per year).
7. Comment: Two commenters (IV-D-22, IV-F-1.6) stated that the Clean Air
Act mandates the examination of energy impacts, which were
not fully considered in developing the standards.
Response: The energy impacts associated with the promulgated standards
were thoroughly examined, as discussed above and in "Revised
Impacts of Alternative Sulfur Dioxide New Source Performance
Standards for Industrial Fossil Fuel-Fired Boilers." The
impacts of the proposed standards were also thoroughly
80
-------
examined and discussed in the preamble to the proposed
standards and in the "Summary of Regulatory Analysis."
These impacts are considered reasonable in light of the
significant emission reductions achieved by the standards.
8. Comment: One commenter (IV-D-28) said that if the margin between oil
and coal prices becomes very large, as might occur in a
supply disruption, then retrofitting of oil and gas steam
generating units with coal might be necessary to maintain
industrial production. The commenter asserted that the
energy impacts of the proposed regulation in such an
eventuality were not evaluated.
Response: In assessing the impacts of a standard, it is not possible
to quantitatively address potential impacts that could occur
under every possible eventuality. A number of scenarios
including various ranges of fuel prices were addressed in
the impacts analysis. Given current expectations with
respect to relatively low oil and gas prices, a supply
disruption forcing massive conversions to coal is not
considered likely.
Additionally, from a technical point of view, the vast
majority of gas- and oil-fired industrial-commercial -
institutional steam generating units are package units
specifically designed to accommodate gaseous or liquid
fuels. These steam generating units physically cannot be
converted to fire coal.
9. Comment: One commenter (IV-D-98) noted that a senior interagency
group, under the leadership of the Department of Energy, is
reviewing the question of energy sources and national
security. The commenter said the findings and
81
-------
recommendations of this group should be considered prior to
promulgating standards.
Response: The report issued by this group, "Energy Security - A Report
to the President of the United States," was published by the
Department of Energy in March 1987. The primary national
energy security concern identified and discussed in this
report was a means of limiting the increased use of imported
oil and encouraging the use of domestically produced fuels,
such as coal and natural gas. The final standards treat
coal and oil in equivalent fashions and are expected to
discourage the use of oil in favor of natural gas.
Therefore, these standards are not considered to be in
conflict with national energy policies or the report issued
by this group and referred to by the commenter.
2.6.2 Energy Scenarios
1. Comment: Several coiranenters (IV-D-6, IV-D-28, IV-D-40, IV-D-50,
IV-D-52, IV-D-85) said the fuel price scenarios used to
evaluate impacts of the proposed standards (for both high
coal and high oil penetration) are unrealistic based on
today's environment. In particular, the commenters said,
the high coal penetration scenario, which is used as the
basis for impacts on coal-fired steam generating units,
significantly overestimates the number of coal-fired units
that will be built, given the relatively small current price
differential between coal and natural gas. This has
resulted in an exaggerated and unrealistic estimate of the
impacts of the standard on coal firing. One (IV-D-28) said
the baseline use of coal, which was small in the analysis,
could drop to near zero with current price projections. The
commenter said this would reduce the projected benefit and
cost of the proposed regulation.
82
-------
Response: The high coal penetration scenario was analyzed in the
proposed regulation in response to concerns that the "most
likely" (high oil penetration) scenario could underestimate
the impacts of the standard on coal-fired steam generating
units. Therefore, it was structured as a "conservative"
estimate of the impacts that could occur if more coal-fired
units than expected were built.
As a result of changes in fossil fuel prices since early
1986, the energy prices were revised to evaluate potential
changes in impacts. As expected, the projected amount of
coal use in new industrial-commercial-institutional steam
generating units decreased to almost zero. The revised
energy price scenarios and their potential impacts are
discussed in "Revised Impacts of Alternative Sulfur Dioxide
New Source Performance Standards for Industrial Fossil
Fuel-Fired Boilers."
Despite these projections of little or no coal use in
industrial-commercial-institutional steam generating units,
some coal use will occur. Therefore, impacts were
reassessed for coal-fired steam generating units. This
reassessment indicated that the impacts associated with the
final standards were essentially the same as those cited at
the time of proposal in terms of costs and emission
reductions from individual coal-fired steam generating
units. These revised model steam generating unit impacts
are discussed in "Impact of New Fuel Prices on the Costs and
Cost Effectiveness of S02 Emission Control of Industrial
Coal- and Oil-Fired Model Steam Generating Units."
83
-------
While the national impacts of the final standards on
coal-fired steam generating units could be lower than those
associated with the "high coal penetration" scenario
discussed at proposal, SOg emissions from coal-fired steam
generating units are still significant on an individual
steam generating unit basis. Because the costs of
controlling emissions from these steam generating units are
considered reasonable in light of the resulting S02 emission
reductions, the final standards will apply to coal-fired
steam generating units.
2. Comment: One commenter (IV-F-1.2) said the Industrial Fuel Choice
Analysis Model (IFCAM) cost projections are outdated and far
too high. The commenter asserted that fuel costs are only
50 percent of what they were expected to be and are not
likely to rise to that higher level. Another (IV-D-28)
suggested that, due to the high oil price projections, the
energy scenario may have led to the proposal of
inappropriate standards. In fact, the commenter said,
alternative energy assumptions may lead to the conclusion
that no revision of the current standards is warranted.
Therefore, the commenter said, a greater range of energy
prices should be considered in order to bracket likely
energy impacts. One commeter (IV-D-44) suggested that the
new energy scenarios should include an analysis of the risk
of energy price and supply volatility. Two other commenters
(IV-D-31, IV-D-76) also agreed that lower crude oil prices
than those assumed in the analysis should be used. They
said that changes in fuel price assumptions would reduce the
price differential between low and high sulfur oil, with a
concomitant deterioration in the attractiveness of FGD.
84
-------
Response: As discussed above, new industrial fuel price forecasts were
prepared based on lower crude oil prices than those used in
the proposal. These new forecasts include lower projections
of industrial coal, natural gas, residual oil, and
distillate oil prices. The national impacts analysis has
been revised using these new industrial fuel price
forecasts.
The updated national impacts analysis includes a variety of
scenarios based on a wide range of future crude oil prices
in order to address the risks of energy price and supply
volatility. These revised national impacts analyses are
detailed in "Revised Impacts of Alternative Sulfur Dioxide
New Source Performance Standards for Industrial Fossil
Fuel-Fired Boilers."
The use of alternative energy price assumptions to evaluate
the national impacts of the final standards under a variety
of energy scenarios did not result in identification of any
significant adverse impacts, nor did it alter the previous
conclusions that the standards are appropriate and
reasonable. While it is true that the differential in high
and low sulfur oil prices is smaller than that identified at
proposal, resulting in somewhat greater costs associated
with the use of FGD compared to very low sulfur oil, the
costs and benefits of the final standards are still
considered reasonable.
85
-------
3. Comment: One commenter (IV-D-28) said that a spectrum of possible
energy scenarios should be evaluated and new assumptions and
analyses should be subject to public review and comment
prior to further action on this rulemaking. The commenter
suggested that the following three scenarios should be
considered:
The cost of oil could become enough lower than the
marginal price of gas that the response to regulation
would not be switching to gas, but to controlling
emissions from firing oil. According to the commenter,
this would reduce benefits, but probably increase costs
and result in less favorable cost-effectiveness ratios.
If the gas prices are projected to be lower than oil
prices until after the year 2000, almost all steam
generating units would use natural gas, with or without
a regulation, and there would be no environmental
justification for the current rulemaking. The commenter
said this scenario, as well as a low oil price scenario,
should be evaluated using IFCAM.
Legislation now under consideration by Congress would
repeal provisions of the Fuel Use Act which authorize
special exemptions for industrial steam generating units
undergoing coal retrofits. The commenter said EPA has
not published any analysis of such a scenario where such
massive conversions to coal occur and are subject to
NSPS. Should these modifications be subject to NSPS,
the commenter said, a third energy scenario is created
for evaluation.
The commenter suggested that the first two scenarios (gas
more expensive than oil, oil more expensive than gas) be
considered the basis for judging the need and benefit of the
standard. The commenter also said that the third scenario,
86
-------
while less likely, needs to be considered because it is
possible, and because if it occurs, standards based only on
the first two scenarios could have unacceptable energy and
economic impacts.
Response: Although considered unlikely, as part of the revised
(post-proposal) national impacts analysis using new
industrial fuel price projections, two scenarios were
examined where gas is more expensive than oil. With these
scenarios, there is some switching to gas under the final
standards, but many units choose to continue to fire oil.
Emission reductions associated with the final standards are
substantial and the national average cost effectiveness of
the final standards compared to the baseline for these two
scenarios remains in the same range as those presented at
proposal in the "Summary of Regulatory Analysis."
In addition, a scenario was also examined where gas prices
are projected to be lower than oil prices until after the
year 2000. Under this scenario, almost all steam
generating units would use natural gas, with or without a
regulation. The total benefits and total costs are
significantly lower than the estimates published with the
June 1986 proposal. However, these results do not mean that
there is no environmental justification for the current
rulemaking. Even under this scenario, the benefits
associated with the final standards are expected to outweigh
; their costs.
It has also been suggested that the analysis consider a
scenario where massive conversions to coal occur and are
subject to NSPS. As mentioned earlier, it is not possible
to examine every conceivable energy scenario. This scenario
87
-------
of massive conversions to coal is not considered reasonable
given current expectations with respect to future oil and
gas prices. Furthermore, as also discussed earlier, energy
scenarios characterized by high coal use in new industrial-
commercial -institutional steam generating units were
examined prior to proposal and the impacts associated with
standards requiring a percent reduction in emissions were
considered reasonable under these scenarios.
4. Comment: One commenter (IV-D-28) said projections of national impacts
should be made to later years, perhaps at 5-year increments
to the year 2000. According to the commenter, this would
better reflect the impacts of alternative regulations and
permit more reasoned selection of standards. The commenter
said that such longer term projections, for example, were
made to support the proposal of the electric utility NSPS.
The commenter added that the effective life of the standards
in terms of the number of sources that will be subject to
the standards, the length of time and progress of technology
required before a standard will be revised, and the
operating life of the affected facilities, requires a longer
time period to be analyzed than that provided in the
proposal.
Response: The national impacts associated with the standards were
projected for the fifth year following proposal of the
standards (i.e., 1990). This 5-year period allows
projection of impacts with a greater degree of confidence
than would be possible under longer time periods.
Projections of impacts under longer time periods would lead
to increases in projected costs as well as projected
emission reductions. The balance, however, between these
costs and the projected emission reductions would be
88
-------
unlikely to change significantly. As a result, judgments
regarding the reasonableness of the final standards would
not change. In addition, the Clean Air Act provides for
review of standards every 4 years. Therefore, impacts of
these standards will be updated and reassessed in the future
during the regular review process.
5. Comment: Two commenters (IV-D-23, IV-D-98) said that even though the
Agency indicated that it intends to update its energy price
scenarios between the time of proposal and promulgation of
the standards to determine whether the costs, emission
reductions and cost effectiveness of the proposed rule will
be altered significantly, such an update is required prior
to proposal of any standard to provide an adequate
opportunity for notice and comment.
Response: The revised energy price scenarios and national impacts
analysis were entered into the docket for this rulemaking
and distributed to industry trade associations and other
interested groups. The results of these new analyses did
not significantly alter any of the conclusions drawn from
the analyses supporting the proposed standards; therefore, a
formal period of public comment was not judged to be
necessary.
2.6.3 New Steam Generating Unit Population Impacts
1. Comment: Several commenters felt that the increase in total
industrial-commercial-institutional steam generating unit
population was overestimated in the analyses. Some
(IV-D-22, IV-D-26, IV-D-28, IV-D-30, IV-D-50, IV-D-53,
IV-D-96, IV-F-1.2, IV-F-1.6, IV-F-1.14) said this
overestimation could be as much as double the number of new
89
-------
units that may actually materialize by 1990. According to
the commenters, the market for steam generating units has
declined by 50 percent in the last 9 years, and the new
capacity ordered each year is only about 0.8 percent of
existing installed capacity. Also, the commenters claimed
that 70-80 percent of "new" units installed are actually
replacing existing capacity, rather than 27 percent as
estimated in the proposal. Therefore, the commenters said,
the overall effect of the standard on air quality
improvement was vastly overstated.
One commenter (IV-D-91) indicated that industry-wide
replacement of existing capacity could be substantially
higher than 75 percent of steam generating unit sales. This
was based on the commenter's observation that between 1977
and 1983, only 21 new "grass roots" plants were constructed
in the energy intensive industries identified by SIC Codes
22 (textile), 26 (pulp and paper), 29 (petroleum), and 33
(primary metals), whereas over this same period, over
2,500 industrial watertube steam generating unit orders were
placed.
Response: The projections of new steam generating unit installations
used in the national impacts analysis were derived from
estimates of future energy consumption in the industrial
sector by the Department of Energy and are higher than those
made by other organizations, such as the American Boiler
Manufacturers Association (ABMA). It is also true that
steam generating unit sales have decreased and stabilized
since the 1970 levels, and some new units are replacements
for existing units. Thus, emission reductions as well as
costs attributed to the proposed standards may be
overestimated. However, this overestimation does not
90
-------
significantly influence the overall balance between the
costs and benefits or the cost effectiveness of the
standard. Even if fewer steam generating units are built
than projected, the S02 reductions achieved by the standard
are still considered significant. Also, while steam
generating unit sales are lower than the 1970 levels, new
units are being built and will continue to be built in the
future. As more and more units are constructed, further SO2
reductions will be achieved by the promulgated standards.
The cost, environmental, and energy impacts of the final
standards, therefore, are considered reasonable, even
assuming smaller population growth projections.
In response to comments concerning the proportion of new
steam generating units that are replacements for existing
steam generating units, a survey of steam generating units
constructed between 1981 and 1984 was conducted. The
results of this survey are discussed in "Survey of New
Industrial Boiler Projects - 1981-1984" (EPA-450/3-87-006).
This survey indicated that about 50 percent, rather than 70
to 80 percent, of new steam generating units were for
replacement of existing steam generating units. Assuming
recent declines in the price of oil and natural gas result
in a curtailment of new steam generating units installed for
the purpose of switching from firing oil or gas to firing
coal, the percentage of new steam generating units sold as
replacements for existing units will be even lower.
2. Comment: Several commenters expressed concern that the standard would
discourage the replacement of existing steam generating
units with new, less polluting units. One (IV-D-23) said
that a standard providing an incentive to keep such old,
91
-------
"grandfathered" units in service is poor policy. Several
(IV-D-52, IV-D-62, IV-D-66, IV-D-75, IV-D-78, IV-D-81,
IV-D-87) said this disincentive to install new units would
be caused because the standard increases the costs of new
units, makes their operation less reliable because of the
effects of scrubber operation, and requires reliance on
uncertain natural gas for backup fuel. According to the
commenters, industrial operators cannot take such risks and
likely will continue to operate older, existing units,
thus making the standards environmentally counterproductive.
Another commenter (IV-F-1.12) said that the proposed
standard will further depress the already sluggish
industrial-commercial-institutional steam generating unit
market, causing research and development in the area of new
steam generating unit design and efficiency to slow
drastically.
Response: The steam generating unit sales market is smaller than it
was in the mid-1970's, but this is due more to sluggish
industrial growth and the depressed nature of the
manufacturing sector than any existing or planned
environmental standards. According to steam generating unit
sales data compiled by ABMA, sales of industrial-commercial -
institutional steam generating units peaked in 1974, dropped
thereafter, but appear to have leveled off in the last
couple of years.
New steam generating unit purchases can be divided into two
categories: discretionary, meaning the purchase could be
deferred if economic or other factors changed, and
nondiscretionary, meaning a new unit must be installed.
Discretionary steam generating unit purchases could be
discouraged by environmental requirements. As discussed
92
-------
above, a survey of new steam generating units ordered
between 1981 and 1984 indicated that nondiscretionary
installations accounted for about half of the new steam
generating unit projects surveyed. These installations are
expected to be neither discouraged nor delayed by an NSPS.
The survey results also indicated that many of the
"discretionary" installations were relatively insensitive to
cost changes. Nearly three-fourths of all respondents
Indicated that the discretionary projects would have
proceeded as designed if costs increased by up to 20
percent. In most cases, the percent reduction requirement
would increase costs by less than this amount; therefore,
the standards would not influence construction of most steam
generating unit projects.
The use of an F6D system, if properly designed, constructed,
operated, and maintained, should not reduce the overall
reliability of the production unit.
Although natural gas availability is not expected to be a
problem, reliance on natural gas as a backup fuel is not
mandatory. The final standards also accommodate the use of
low sulfur oil as a backup fuel in steam generating units as
long as oil use does not exceed 30 percent of the annual
rated heat input capacity of the steam generating unit.
Consequently, the standards should not significantly promote
the continued operation of older existing steam generating
units, should not depress the sales of steam generating
units, and should not reduce new steam generating unit
research and development efforts.
93
-------
3. Comment: Five commenters (IV-D-6, IV-D-28, IV-D-40, IV-D-50, IV-D-52)
were concerned that, because the Clean Air Act mandates
compliance as of the date of proposal of standards, the
uncertainty inherent in the post-proposal period combined
with the expenses associated with the percent reduction
requirement could cause costly project delays and, in some
cases, abandonment of new steam generating unit projects.
Response: The applicability date of new source performance standards
is established by the Clean Air Act [Section 111(a)(2)].
The purpose of proposed standards is to notify new source
owners and operators of the requirements that may apply to
sources commencing construction after the date of proposal.
Although the commenters did not submit any specific examples
or information to support their opinion, it is possible that
some new steam generating unit construction projects could
be delayed until the final standards are published. Any
delays that might result, however, should be only temporary.
Also, the promulgation of these standards is mandated by a
court-ordered promulgation date, which is a matter of public
record. Therefore, there is no uncertainty related to
completion of a final standard in this case.
4. Comment: One commenter (IV-F-1.2) said the forecast of new coal-fired
steam generating units is overly optimistic. At most, the
commenter said, 10 to 15 new coal-fired units in the size
range between 29 and 73 MW (100 and 250 million Btu/hour)
have been ordered annually over the last few years. Another
(IV-D-99) said that very few coal-fired steam generating
units are being sold today, due to their higher capital and
nonfuel operating costs and the current small price
differentials between coal, oil, and natural gas. The
commenter asserted that this situation is not likely to
94
-------
change in the near future, and expressed the concern that
the NSPS would apply additional negative pressure on the new
coal-fired steam generating unit market.
Response: The national impacts analysis was recently revised using
lower projections of oil and gas prices. This revised
national impacts analysis shows essentially no demand for
new coal-fired industrial-commercial-institutional steam
generating units, even in the absence of standards, due to
the competitive prices of oil and gas. Thus, future oil and
gas prices are of much more significance in determining
future coal use in new industrial-commercial-institutional
steam generating units than the final standards.
In practical terms, however, new coal-fired steam generating
units will be ordered and installed in the future. While it
is possible that some steam generating unit operators
contemplating installing a new coal-fired unit would elect
to use natural gas instead as a direct result of the
standards, the amount of "fuel switching" from coal to gas
is expected to be minimal due to factors favoring the use of
coal in those situations where coal is selected.
2.6.4 Emissions Impacts
1. Comment: Many commenters (IV-D-6, IV-D-10, IV-D-14, IV-D-26, IV-D-28,
IV-D-29, IV-D-30, IV-D-35, IV-D-36, IV-D-40, IV-D-44,
IV-D-45, IV-D-48, IV-D-50, IV-D-52, IV-D-53, IV-D-58,
IV-D-62, IV-D-66, IV-D-73, IV-D-74, IV-D-84, IV-D-96,
IV-D-98, IV-D-99, IV-F-1.12) felt that without an option to
use compliance coal, steam generating unit operators will
have less economic incentive to install new units, thus
inhibiting the replacement of old, existing high sulfur
95
-------
fuel-fired units. The commenters said this would inhibit
the gradual reduction in emissions that is currently taking
place and must be taken into account when calculating the
net emission reductions due to the proposed standard.
One commenter (IV-D-62) further stated that plants with both
new and existing steam generating units will choose to
operate older units with higher emissions during maintenance
or malfunction periods of the FGD system on the new unit,
resulting in an increase in emissions. If they were allowed
to continue operation of the new steam generating unit
firing low sulfur coal, the commenter said, overall
emissions would be lower.
Response: To assess the impacts of the standard on steam generating
unit replacement, a survey was conducted of all new
industrial steam generating unit projects undertaken between
January 1981 and August 1984. Responses were received on
168 new projects, encompassing a total of 229 new steam
generating units. Of these, 151 steam generating units were
in the regulated size category and formed the basis for the
analysis discussed below. Comparison with steam generating
unit sales data collected by ABMA indicated that these units
represent almost all of the industrial steam generating
units in this size category sold between 1981 and 1984.
The results of the survey indicate that about 50 percent of
the new industrial steam generating unit projects undertaken
during the timeframe surveyed were nondiscretionary and
would be expected to be completed regardless of the
requirements of the NSPS. The remainder were
discretionary installations and could be affected by
regulatory requirements. However, the survey found that
96
-------
nearly three-fourths of all projects would have proceeded as
designed even if project costs increased by 20 percent.
Therefore, while some projects could be affected by the
NSPS, most of the new steam generating unit installations
would remain viable, since the NSPS will generally increase
costs for new steam generating units by less than
20 percent.
The effect of replacing existing steam generating units on
overall S0£ emissions was also analyzed in this survey.
Analysis of S02 emissions before and after installation of
new steam generating units found that although the new coal-
and oil-fired steam generating units generally emit less S02
per million Btu of fuel fired than the existing (replaced)
steam generating units, the total annual emissions from
plants installing new steam generating units increased by
roughly 70 percent. This increase results from 1)
installation of new steam generating capacity that is not
replacing existing units, 2) replacement of existing units
with new units that are significantly larger, 3) fuel
switches from natural gas and oil to coal, and 4) continued
operation of the existing steam generating units, which were
reportedly replaced by new units, although at reduced load.
A large portion of the emissions increase was due to steam
generating units installed either to switch base fuel
(usually from gas or oil to coal) or to cogenerate steam and
electricity. Because these types of projects are more
vulnerable to cancellation based on cost considerations than
projects undertaken for other reasons, calculations were
also made excluding these types of projects. Even without
these significant sources of SO2 emissions, total emissions
97
-------
at plants installing new steam generating units increased by
roughly 20 percent.
These results indicate that the replacement of existing
steam generating units does not appear to result in any
"natural" decrease in overall SOg emissions from this source
category, as suggested by the commenters. As a result, for
the reasons discussed above, the standards are not expected
to significantly discourage the replacement of existing
steam generating units, nor will they result in an increase
in SO2 emissions.
While it is possible that plants with both new and existing
steam generating units would operate the existing units
during periods of FGD malfunction, this should not
significantly affect overall emissions. Periods of
malfunction should be infrequent and temporary occurrences
in a well operated and maintained system.
2. Comment: Five commenters (IV-D-26, IV-D-30, IV-D-50, IV-D-53,
IV-D-81) said that because as much as 80 percent of the
projected "new" steam generating unit population will
actually be replacement units, they will not constitute new
sources of SOg emissions. They stated that by including
these emissions in the analysis, the actual "new" emission
reductions resulting from the proposed standards were
overestimated.
Response: As discussed above, a survey of new industrial steam
generating unit projects undertaken between 1981 and 1984
indicates that only about 50 percent of new steam generating
units were installed to replace existing capacity, rather
than 80 percent. Emissions from all new steam generating
98
-------
units (including those that replaced existing steam
generating units) were correctly included as "new" capacity
emissions in the impacts analyses. The SO2 emission
reduction attributed to the NSPS is a reduction in emissions
from new steam generating units compared to baseline
emissions that would occur from those new units in the
absence of the NSPS. Because new industrial-commercial-
institutional steam generating units are a major source of
SOg emissions, regardless of whether they are replacement or
"new capacity" units, they are covered by the final
standards.
3. Comment:
Three commenters (IV-D-12, IV-D-39, IV-D-99) expressed
concern that the high compliance costs would force the
construction of large numbers of smaller steam generating
units which will not be subject to the standards, thereby
increasing emissions and increasing problems for future
regulation of these sources.
Response: Although this may be the case in a few situations, in most
cases it would be more expensive to build and operate two or
more smaller steam generating units than one larger one,
even with the added SO2 control costs associated with the
larger steam generating unit. In any event, development of
standards for steam generating units less than 29 MW
, (100 million Btu/hour) heat input capacity is currently
underway, and their issuance will, in effect, eliminate any
temporary incentive of this nature.
4. Comment:
One commenter (IV-F-1.2) stated that as much as 50 percent
of what is referred to as "industrial" steam generating
units are actually purchased by institutional or commercial
installations for heating and are used at capacity
99
-------
utilization rates far lower than Industrial operations.
According to the commenter, institutional steam generating
units are operated only during the heating season, and even
then they rarely have more than 80 percent of full load, so
the total capacity utilization, and the resulting SOg
emissions, are much lower than those for industrial steam
generating units. Therefore, the commenter said, the
emissions impacts of the proposed standards have been
overestimated.
Response: Data on new steam generating unit sales between 1979 and
1985, taken from Power magazine, listed 20 units with heat
input capacities greater than 29 MW (100 million Btu/hour)
that were ordered for institutional or district heating
purposes over this period. According to Power, the total
number of steam generating unit orders was 205. Therefore,
during this period, institutional and district heating
systems represented only 10 percent of new steam generating
unit orders in this size range. Even this figure, however,
may be too high, because a disproportionately large number
of units (12) were sold to Federal military installations.
The reason for this large number of Federal orders is not
known, but the sales may have resulted from a Department of
Defense directive to reduce steam generating unit
consumption of oil for national security reasons. If these
installations are excluded from the figures because they are
non-recurring, the institutional and district heating
systems represent only about four percent of new steam
generating unit sales over this time period. This is
roughly comparable to information from the Department of
Energy for new steam generating units ordered between 1977
and 1981, which indicated that 22 out of approximately 400
orders were for institutional or district heating purposes.
100
-------
There is no reason to expect this percentage to increase
significantly in future years. Therefore, commercial and
institutional steam generating units are expected to
represent only a very small percentage of all steam
generating units subject to this NSPS, as discussed in the
proposal.
In any event, an exemption from the percentage reduction
requirement has been granted for steam generating units with
capacity utilization factors for coal or oil of less than
30 percent. Therefore, any commercial and institutional
steam generating units that would be operated below this
level would not required to install an FGD or FBC system,
but must meet certain emission limits for SOg.
5. Comment: Several commenters questioned the baseline emission levels
used in the analyses. Two (IV-D-28, IV-D-32) said that the
baselines in both the model boiler analysis and IFCAM do not
reflect existing new source review regulations or PSD
permits. This resulted in overestimating the probable SO2
reduction of the proposal by over 90 percent. According to
the commenters, the actual new source requirement for
Eastern States is 688 ng/J (1.6 lb/million Btu), and for
Western States is 516 ng/J (1.2 lb/million Btu). Therefore,
the commenters said, these are the maximum reasonable levels
for projecting future baseline emissions. One commenter
(IV-D-28) further stated that the mean State Implementation
Plan (SIP) values for baseline emissions from steam
generating units with heat input capacities less than 73 MW
(250 million Btu/hour) should be 1,334 ng/J (3.1 lb/million
Btu) for coal and 817 ng/J (1.9 lb/million Btu) for oil.
Commenters IV-D-28 and IV-D-96 added that under the proposed
regulations, the net effect of the proposal was calculated
101
-------
in the IFCAM analysis to be an SOg emission reduction of
211,000 Mg/year (232,000 tons/year) under the high oil
penetration scenario or 269,000 Mg/year (296,000 tons/year)
under the high coal penetration scenario. Using the same
analytic approach, but a baseline based on recent new source
review practices (as discussed above), would yield a net
emissions change of 109,000 Mg/year (120,000 tons/year)
under the high oil penetration scenario or 149,000 Mg/year
(164,000 tons/year) under the high coal penetration
scenario. Therefore, according to the commenters, use of an
incorrect baseline in the IFCAM analysis led to
overestimating emission reductions by 81-93 percent. The
commenters said that simply assuming the correct emissions
baselines of 688 ng/J (1.6 lb/million Btu) in the East and
516 ng/J (1.2 lb/million Btu) in the West reduces the
probable environmental benefit of the regulations to a
reduction of less than 0.5 percent of national SOg emissions
in 1990.
Response: Emission levels established under PSD and NSR programs
are determined on a case-by-case basis and reflect local
and/or site-specific conditions. Because the limits are
site-specific, they cannot be predicted with certainty.
The "baseline" emission levels used in the IFCAM analysis
are actual SIP emission levels applicable in each Air
Quality Control Region (AQCR) throughout the country. The
AQCR-SIP levels represent the minimum level of control
imposed on new units. The use of this minimum level of
control as the baseline for purposes of analysis produces
"upper limit" estimates of the cost impacts associated with
the standards and, therefore, ensures that these costs are
not underestimated. It is true that using more stringent
102
-------
baselines (such as those reflected in PSD and NSR programs)
would result in smaller emission reductions attributable to
the standards. The costs attributable to the standards,
however, would also be lower since the baseline annualized
costs would increase to account for the additional PSD/NSR
compliance costs. For example, as shown in the Summary of
Regulatory Analysis, if an emission limit of 520 ng/J
(1.2 lb/million Btu) was established as the "baseline"
level, the annual baseline emissions would be 340 Mg/year
(370 tons/year) for a 44 MW (150 million Btu/hour) unit in
Region V. Total annualized costs under the baseline would
increase from $6,160,000 to $6,340,000. More importantly,
however, changing the baseline would not change any of the
incremental impacts. Incremental impacts were used as the
primary indicator of the reasonableness and effectiveness of
the various regulatory alternatives. The baseline emission
rates suggested by the commentor are reasonable, but even if
used in the analysis, would not have changed the conclusion
that the standard is reasonable.
6. Comment: One commenter (IV-D-48) asserted that 90 percent S02
reduction by FGD was not achievable when firing low sulfur
; coal and the 90 percent reduction requirement will force
units to burn higher sulfur fuels in order to meet the
percent reduction requirement. Therefore, the commenter
said, the actual emission rate under 90 percent reduction
standard could be greater than that which would result from
the use of a low sulfur fuel with a less stringent, 70-90
percent "sliding scale" reduction requirement.
Response: The standards will not "force" steam generating unit
operators to purchase high sulfur fuels. The percent
reduction requirement is achievable on both high and low
sulfur fuels, and the choice of fuel will likely depend more
103
-------
on local availability and other site-specific factors than
on the requirements associated with the NSPS. Thus, a
sliding scale will not result in lower SOg emissions.
2.7 COST OF THE STANDARD
2.7.1 General
1. Comment: Several commenters felt that the cost analyses associated
with the proposed standards were deficient or inaccurate:
A. Five commenters (IV-D-58, IV-D-62, IV-D-66, IV-D-74,
IV-D-84) said that the use of capital cost estimates
rather than actual data was a major deficiency.
B. These same conmenters said that the use of significantly
lower capital cost estimates per unit of heat input for
industrial steam generating units as compared to utility
steam generating units (i.e., failure to consider
economies of scale) resulted in the capital costs being
greatly underestimated.
C. Several commenters (IV-D-6, IV-D-28, IV-D-35, IV-D-40,
IV-D-44, IV-D-50, IV-D-52) said that the cost estimates
did not adequately reflect problems associated with
sludge disposal.
D. These commenters also said that problems associated with
the reliability of SO2 control technology were not
considered in developing cost estimates.
E. Six commenters (IV-D-6, IV-D-27, IV-D-28, IV-D-40,
IV-D-50, IV-D-52) stated that the cost estimates did not
reflect the coal transportation cost penalties that are
likely to be incurred by smaller coal users. The
commenters said these penalties are incurred due to the
inability of smaller users to obtain volume discounts,
and should be included 1n compliance cost estimates for
these steam generating units.
104
-------
F. Two commenters (IV-D-22, IV-F-1.6) said that the cost
analyses did not allow for manufacturers' profit
margins.
G. One commenter (IV-D-28) said that the high emissions
baseline used in the cost analysis resulted in
underestimating the cost effectiveness of the standard
for many steam generating unit categories. The
commenter stated that using more reasonable baseline
levels of 516 and 688 ng/J (1.2 and 1.6 lb/million Btu),
cost effectiveness ratios for small coal-fired units
would be $2,200-$2,750/Mg ($2,000-$2,500/ton), while the
model steam generating unit ratios in the analysis were
roughly $880-$l,300/Mg ($800-$l,200/ton). According to
the commenter, this discrepancy would be even greater in
the IFCAM analysis, which results in an average cost
effectiveness figure for all steam generating units of
$150/Mg ($140/ton).
Response: A. The cost estimates used to assess the impacts of various
alternatives were generated from cost algorithms
developed from data obtained from vendors. The validity
of the cost algorithms was examined by comparison with
I costs associated with actual installations. The
agreement between these algorithms and actual
installations was generally found to be very good --
well within the general criterion of + 30 percent which
is normally associated with "budget cost" estimates.
Validity of the costs used in the cost analyses was also
confirmed by statements from several commenters,
including some industry trade associations. This does
not mean that the cost estimates generated by the cost
algorithms will agree in every case with actual
installed costs. Unique design requirements related to
105
-------
site-specific factors may well cause actual costs to be
higher or lower than those generated by the cost
algorithms used in the cost analyses. The costs
generated by the cost algorithms, however, are
considered representative of the costs associated with
installation and operation of steam generating units and
emission control systems.
B. The cost analyses assessed the costs associated with
various emission control requirements for steam
generating units as small as 29 MW (100 million
Btu/hour) heat input capacity and as large as 117 MW
(400 million Btu/hour) heat input capacity. Costs were
also assessed for units of 44, 58, and 73 MW (150, 200
and 250 million Btu/hour). Consequently, the effects of
"economies of scale" were considered in the cost
analyses. For example, the capital cost of SOg control
for a 29 MW (100 million Btu/hour) steam generating unit
firing bituminous coal is approximately $10.8 million.
For a 117 MW (400 million Btu/hour) steam generating
unit, the capital cost of SOg control is approximately
$34.6 million. Therefore, while the capacity of the
second unit is four times the capacity of the first, the
cost increases only by a factor of 3.5.
In addition, the capital cost estimates were based on
vendor quotes for industrial, not utility, steam
generating units. As discussed above, these costs have
been confirmed by comparison with costs associated with
actual installations and are considered representative
of industrial S02 control systems in general.
106
-------
C. Sludge disposal costs were Included in the cost
analyses. The costs reflect the typical costs
associated with off-site land disposal of sludges from
emission control systems. These costs were based on
information provided by steam generating unit vendors
and operators, and are considered representative of
sludge disposal costs in general. They may, however, be
lower or higher than actual costs experienced at
specific locations where unique or site-specific
requirements may apply.
0. The cost analyses assumed, based on operating data from
a number of flue gas desulfurization (FGD) systems on
industrial steam generating units, that FGD systems are
capable of 95 percent reliability with proper design,
operation, and maintenance. The cost analysis,
therefore, considered control system reliability
problems and examined various approaches to reducing
emissions during periods of control system malfunctions.
These approaches ranged from the use of spare absorber
modules to the firing of very low sulfur backup fuels
such as natural gas. The assessment concluded that the
costs associated with the use of spare absorber modules
or the firing of very low sulfur fuels were of the same
order of magnitude. Thus, either the use of spare
absorber modules or the firing of very low sulfur fuels
could be used in the cost analyses to represent the
additional costs of reducing emissions during periods of
emission control system malfunctions. For convenience
and ease of calculation, the cost analyses reflected the
firing of natural gas to reduce SO2 emissions during
periods of emission control system malfunction. The
capital and annualized costs of spare equipment to fire
107
-------
very low sulfur fuels, such as valves, atomizers,
controls, etc., were also added to the control costs.
Consequently, the cost analyses considered the increased
costs necessary to address reliability problems.
E. The coal prices used in the cost analyses related to
this rulemaking were developed specifically for
industrial-commercial-institutional steam generating
units. The coal transportation costs assumed single
coal car rates to reflect the lack of volume discounts
such as those obtainable by utilities, which use much
greater quantities of fuel.
F. As discussed above, the cost data used in developing the
cost algorithms were developed from data submitted by
vendors, and were confirmed by comparison with costs
associated with actual installations. The cost
algorithms, therefore, reflect the charges made for
vendor services and products and thus include vendors'
profit margins.
G. The baseline used in the national impacts analysis
reflects existing SIP emission limits within each AQCR
as discussed earlier. The baseline used in the "model
boiler" cost analysis was chosen to represent a typical,
although somewhat lenient, SIP emission limit. Other
baselines could have been selected. However, if more
stringent baselines were used, the annualized costs for
SO2 control at the baseline would also have to be
increased to account for the additional compliance costs
associated with more stringent regulations. Selection
of different baselines would also change the "average"
cost effectiveness of various regulatory alternatives
108
-------
since "average" cost effectiveness compares the costs
and emissions under the alternative in question to the
costs and emissions under the baseline. Of much greater
significance, however, are comparisons of the
incremental impacts among alternatives. Varying
assumptions regarding the baseline have no effect on
these comparisons and, as a result, would not change any
of the comparisons of incremental impacts among
alternatives, including comparisons of incremental cost
effectiveness.
2. Comment: Several commenters (IV-D-26, IV-D-30, IV-D-50, IV-D-53,
IV-D-58, IV-D-62, IV-D-66, IV-D-74, IV-D-84, IV-F-1.15) felt
that the +30 percent accuracy of the cost estimates
significantly affected the range of cost effectiveness
values cited in the proposal preamble.
Response: Depending on how one views the cited cost effectiveness
figures, the commenter is correct. The accuracy of the cost
estimates can significantly affect the range of cost
effectiveness values. In terms of individual steam
generating units, the cost effectiveness figures cited
could be considered as a range of expected values
within + 30 percent of the stated value. In terms of
industrial steam generating units as a group, however, it is
appropriate to focus on the cost effectiveness figures
cited. Within a group of similar steam generating units,
the subtle differences between units that give rise to the
+ 30 percent accuracy range ascribed to the cost algorithms
will tend to average out. As a result, the average costs
are more likely to be much closer to the median or midpoint
estimates provided by the cost algorithms than to either
extreme of the + 30 percent range. Thus, in terms of groups
109
-------
of industrial steam generating units, the + 30 percent
accuracy ascribed to the cost algorithms has little impact
on the cited cost effectiveness figures and is considered
appropriate for assessing the impacts associated with
regulatory actions.
3. Comment: Three commenters (IV-D-29, IV-F-1.9, IV-F-1.17) said that a
percent reduction requirement would increase the
maintenance, operating, and energy costs while reducing the
reliability of the total production unit.
Response: Any emission standard will increase the costs for a
production unit to some extent. The impacts associated with
these increased costs were considered and included in the
analyses of various alternatives. Similarly, the impact on
reliability was also considered. Experience with industrial
steam generating unit FGD systems shows them to be highly
reliable technologies 1f they are properly operated and
maintained. More important, however, the reliability of the
FGD system does not affect the reliability of the steam
generating unit or of the production units. Alternative
fuels such as natural gas or very low sulfur oil can be
employed to maintain steam generating unit operation and
production at a reasonable cost during those times when an
FGD system is shut down for maintenance or repair.
4. Comment: Three commenters (IV-D-55, IV-D-56, IV-D-70) said that all
memoranda associated with the determination of the cost
effectiveness of the standard and concerning the
determination of a "reasonable" level of cost effectiveness
should be identified and placed in the docket for this
rulemaking for public review and comment.
110
-------
Response: The notice of proposed rulemaking and materials already in
the docket gave the public full notice of, and opportunity
to comtient on, the facts and arguments relevant to the
consideration of cost effectiveness in this rulemaking. The
specific memoranda referred to by the commenters are related
to other regulations and proceedings and are not part of
this rulemaking. Therefore, they need not be placed in the
docket [Sge Clean Air Act Section 307(d)(3)(C), Section
307(d)(4)(B)(i), and Section 307(d)(6)(C)]. In any case,
the memoranda are exempt from disclosure under the
deliberative process privilege of the Freedom of Information
Act, 5 U.S.C. 552(b)(5).
5. Comment:
Three commenters (IV-D-31, IV-D-51, IV-D-88) asserted that
in most situations, the NSPS would require investment in
FGD systems for backup oil firing on gas-fired steam
generating units due to the interruptible nature of natural
gas pipeline supplies (especially during the heating
season). Therefore, the commenters said, this should be
reflected in the cost of using natural gas for compliance.
Response: Natural gas would only be unavailable in situations where
the steam generating unit operator purchases an
interruptible gas supply contract. Non-interruptible gas
supply contracts may be purchased in many locations. The
costs associated with firing natural gas assumed that
noninterruptible gas supplies were purchased. If a steam
generating unit operator elected to use an interruptible gas
supply, the most likely and least expensive backup system
would be to maintain a supply of very low sulfur oil capable
of meeting an emission limit of 129 ng/J (0.3 lb/million
Btu) heat input. Under the final standards, an FGD system
111
-------
to reduce SO2 emissions when firing oil would not be
required as long as a very low sulfur oil was fired.
6. Comment: One commenter (IV-D-12) felt that the costs associated with
the standards are too high for a modified steam generating
unit. For example, the commenter said that a simple
capacity expansion involving a minimal Increase in air
emissions would require the retrofit of an FGD system, which
often would far exceed the original modification cost.
Response: As discussed previously (see Section 2.2), no general types
of modifications to Industrial steam generating units have
been identified for which compliance with the standards is
considered unreasonable. To be considered a modification,
there must be an Increase in the emission rate beyond the
original design limits and there are ways to prevent
emission rate Increases, such as firing a lower sulfur fuel,
to avoid invoking the modification provisions.
7. Comment:
One commenter (IV-F-1.9) said industrial steam generating
unit operators are faced with the probability of two
separate regulatory programs that will impose stringent and
costly emission controls (acid rain legislation and the
proposed NSPS). According to the commenter, the costs will
be approximately 5 times greater than the cost of reducing
the same emissions in large steam generating units to
control only a small percentage (1.5 percent) of total U.S.
SOg emissions.
Response: To address the concern that the relative costs of emission
control are much greater for industrial-commercial-
Institutional steam generating units than for utility steam
112
-------
generating units, the relative cost effectiveness of S02
control for industrial and utility steam generating units
was assessed. The average cost effectiveness of comparable
S02 control, using the same assumptions for both industrial
and utility units, are similar for both types of steam
generating units. Therefore, there is no reason to assume
that the cost of reducing a given amount of SOg emissions
would be substantially greater for industrial-commercial -
institutional steam generating units than for utility steam
generating units.
Although industrial-commercial-institutional steam
generating units emit less SOg than their utility
counterparts, these emissions are still considered
significant and, therefore, are subject to regulation under
Section 111 of the Clean Air Act.
8. Comment: One commenter (IV-D-4) expressed the opinion that there can
be no total calculation of the ultimate cost of this
regulation because it has so many ramifications in so many
areas. The commenter said any cost estimates will be at
least 2-10 fold off the mark because all cross-media costs
are never included.
Response: While it may be impossible to assess all possible costs
associated with any regulation, the analysis supporting the
standards attempted to address all costs. Cross-media
costs, such as waste disposal and other associated
environmental costs, for example, were considered. Based on
a thorough examination of all of these factors, the costs
associated with the final standards are considered
, reasonable.
113
-------
9. Comment: Two commenters (IV-D-81, IV-F-1.1) said the proposal did not
state what cost effectiveness figure was used to conclude
that all of the substantive requirements of the proposed
standard are reasonable.
Response: Although cost effectiveness is an important factor in
determining the "reasonableness" of an NSPS, it is not the
only factor that is considered. The Clean Air Act states
that the selection of a standard must be based on the "best
demonstrated technology," taking into account cost, nonair
quality health and environmental impacts and energy
requirements. All these things were considered in
determining that the requirements of the final standards are
reasonable, and were addressed in the "Summary of Regulatory
Analysis" and elsewhere in the docket for this rulemaking.
10. Comment: One coiranenter (IV-D-44) said the costs associated with
purchasing low sulfur coal were overestimated in the
proposal. For example, the commenter said, one long-term
coal contract allows the purchase of 0.85 percent sulfur
coal for S2.08/million Btu versus the proposal estimate of
$3.32/million Btu.
Response: As discussed elsewhere 1n this document, new fuel price
estimates have been made since the proposal of the
standards. The coal prices 1n these new forecasts are
generally lower than those used 1n the proposal, and the
difference in price between high and low sulfur coal (i.e.,
the "sulfur premium") is less than previously estimated.
Therefore, the new coal prices associated with low sulfur
coal have been amended to reflect these new estimates. It
should be noted, however, that the fuel prices used in the
cost and national impacts analyses reflect average costs
114
-------
that might be incurred by a typical steam generating unit
firing that fuel in the region being analyzed. The actual
prices for a particular steam generating unit may be higher
or lower than these average prices depending on
site-specific factors.
11. Comment: One commenter (IV-D-45) said the high costs associated with
tracking the desulfurization of oil, combined with the high
desulfurization costs themselves, will make "compliance
fuel" very expensive. The commenter felt this should be
recognized in the costs of the regulation.
Response: The provisions 1n the proposed regulation pertaining to
tracking the desulfurization of oil apply only to those
facilities wishing to use credit for fuel pretreatment
toward the percent reduction requirement. The use of
compliance fuel as mentioned by the commenter referred to
the provision in the proposal stating that facilities with
potential S02 emissions of 86 ng/J (0.2 lb/million
Btu) or less would be deemed to comply with the percent
reduction requirement. This was not a compliance fuel
provision, but merely an alternative means of demonstrating
compliance with the 90 percent reduction requirement. As
discussed previously, a reassessment of this issue indicated
that there 1s no point at which the sulfur content of an oil
is so low that one can determine with a reasonable degree of
confidence that it has undergone a 90 percent reduction 1n
its sulfur content. Therefore, this provision has been
dropped in the final standards. In the final standards, a
compliance fuel alternative exists for steam generating
units operating at low capacity utilization rates for oil or
coal and for units firing very low sulfur oil. No fuel
tracking is required for steam generating units subject to
these alternative requirements.
115
-------
2.7.2 Control Technology Costs
1. Comment: Many commenters felt that the cost estimates associated with
flue gas desulfurization (FGD) were too low. Several
(IV-D-6, IV-D-28, IV-D-40, IV-D-50, IV-D-52) said that while
they appear to agree well with results obtained from an
independent survey, the FGD systems surveyed were dry
scrubbing systems whereas the cost estimates were based on
wet scrubbing. According to the commenters, studies of
utility FGD systems have shown that annualized costs for dry
FGD systems are usually about 10 to 30 percent less than for
wet FGD systems. Therefore, they said, the estimates for
wet FGD systems should be proportionately higher.
Other commenters (IV-D-10, IV-D-26, IV-D-27, IV-D-30,
IV-D-35, IV-D-36, IV-D-38, IV-D-48, IV-D-50, IV-D-52,
IV-D-73, IV-D-96, IV-F-1.12) said that the costs to install
FGD systems on industrial-commercial-institutional steam
generating units are considerably greater than for large
utility units due to lack of economies of scale and will
greatly increase total steam generating unit costs with
little or no air quality benefit. For instance, the
commenters stated that a dry FGD system would add about
17 percent to the cost of a new steam generating unit, while
wet FGD systems would add 25 to 33 percent. The commenters
asserted that no explanation was given as to why these
higher costs are justified.
116
-------
Other commenters (IV-D-5, IV-D-58, IV-D-62, IV-D-65,
IV-D-66, IV-D-74, IV-D-84) stated that because sodium
once-through FGD was selected as the basis for the national
cost Impact analysis, the costs were underestimated. The
commenters said that many steam generating unit operators
will select other methods, such as dual alkali or Hme spray
drying, for operational or waste disposal reasons.
According to the commenters, capital costs for sodium
throwaway systems are roughly 40 percent less than for lime
spray dryers and 60 percent less than for dual alkali
systems, and total annualized costs are likewise lower. The
commenters felt that this "real world" variability should be
taken into account in the NSPS analysis.
Response: As mentioned by a number of commenters, the economics of FGD
systems are different for utility and Industrial steam
generating units. Because of their much larger size and the
continuous nature of utility operation, operating costs play
a more dominant role in the economics of FGD systems for
utility units than they do for industrial units. Thus, the
least expensive FGD system for a utility steam generating
unit 1s frequently one which minimizes operating costs,
often at the expense of higher capital costs. In contrast,
the least expensive FGD system for an industrial steam
generating unit 1s frequently one which minimizes capital
costs, often at the expense of higher operating costs.
Sodium FGD systems are characterized by relatively low
capital costs, but relatively high operating costs due to
the use of soda ash as a reagent. L1me spray drying
systems, on the other hand, are characterized by relatively
high capital costs, but relatively low operating costs due
to the use of Hme as a reagent. As one might expect,
therefore, when the economics of these two FGD systems are
117
-------
compared, lime spray drying is less expensive for utility
steam generating units, but sodium scrubbing 1s less
expensive for many Industrial steam generating units.
Sodium scrubbing 1s currently the most widely used FGD
technology for controlling SOg emissions from Industrial
steam generating units. In addition, sodium scrubbing is
generally less expensive than other FGD technologies for
industrial units. Consequently, the sodium scrubbing cost
algorithm was used in the analyses prior to proposal of
standards to generate costs which were viewed as
representative of the type of FGD system that would be most
widely used to control SOg emissions from new industrial-
commercial -Institutional steam generating units.
In response to comments concerning the use of the sodium
scrubbing cost algorithm 1n this manner, however, the costs
of various FGD systems were reviewed and compared again.
The costs of sodium scrubbing were generally found to be
somewhat lower than the costs of other FGD systems, such as
dual alkali and lime spray drying, as well as fluldlzed bed
combustion. From the standpoint of overall project
economics, the total annualized cost of the steam generating
unit and SOg control system varied by less than 10 percent
for all technologies examined. However, when only the cost
associated with SOg control was considered, this variation
was much larger, ranging from 30 to 100 percent. Therefore,
because of variations 1n project-specific factors that could
Influence the choice of SOg control system, the use of any
single technology as a surrogate for all SOg control systems
in subsequent cost analyses was judged to be Inappropriate.
Consequently, to avoid underestimating costs, generic cost
estimates representing the mid-point in the range of cost
118
-------
estimates for all types of industrial FGO systems (i.e.,
sodium scrubbing, lime spray drying, and dual alkali) were
used to represent the costs of industrial-
commercial -institutional steam generating unit FGD systems
in subsequent analyses. While this resulted.in somewhat
higher cost estimates than at proposal, it did not lead to
any significant differences in conclusions.
2. Comment: Several commenters said the operation and maintenance (O&M)
costs associated with flue gas desulfurization were
underestimated:
A. Some (IV-D-12, IV-D-58, IV-D-66, IV-D-74, IV-D-84) felt
that the O&M costs used in the cost analysis were not
adequately documented, and that there is considerable
uncertainty associated with the estimated O&M costs.
The commenters requested that additional data on this
point be produced and made public.
B. Four commenters (IV-D-10, IV-D-23, IV-F-1.15, IV-F-1.16)
said the analysis did not consider the additional
chemical or reagent costs required to achieve high
percentage reductions and minimize the effect of load
swings on FGD performance. Such costs, the commenters
said, will add significantly to total operating costs.
C. Two of these commenters (IV-D-10, IV-D-23) also said
that the increased costs of liquid and solid waste
disposal at higher percent reduction requirements were
not considered.
D. Several commenters (IV-D-26, IV-D-30, IV-D-50, IV-D-51,
IV-D-53, IV-D-88, IV-F-1.15) said that the effects of
other regulations on FGD system operation and cost
should be considered. The commenters specifically
mentioned that FGD system cost calculations should
account for the larger equipment that will be needed to
119
-------
meet the N0X standards through the use of staged
combustion air, as well as the fact that stack gas
reheat could be required to avoid violations of ambient
standards or PSD increments.
Response: A. The development of O&M costs is fully discussed and
documented in the background information documents
included with the proposed standard. A discussion of
this development can be found in the "Industrial Boiler
SO2 Cost Report" and in a memorandum entitled "Sodium
Scrubbing Cost Algorithm Development." In addition to
the "SOg Technology Update Report" and "Fossil
Fuel-Fired Boilers: Background Information," the basis
for many of the assumptions used in developing these
costs was the "Technology Assessment Report for
Industrial Boiler Applications: Flue Gas
Desulfurization." There are also other reports and
memoranda available for review in the docket which
discuss various scrubbing systems. Finally, various
changes and revisions of the cost algorithms are
suiranarlzed 1n "Summary of Revisions to the FGD Control
Cost Algorithms Since 1982."
B. The chemical and reagent costs used to assess the total
annualized cost Impacts of the standard were based on an
average SOg removal efficiency of 92 percent on both
high (3.54 wt. percent) and low (0.6 wt. percent) sulfur
coal. In addition, chemical and reagent costs were also
calculated for lower percent reduction requirements,
such as 50 and 70 percent. In terms of total annualized
costs, the Incremental cost effectiveness of a 90
percent reduction 1n SOg emissions over lower percent
reduction requirements was found to be very reasonable.
This was discussed in "Summary of Regulatory Analysis."
120
-------
The additional reagent costs associated with maintaining
a high percentage reduction in S02 emissions during
periods of load swings were included in the cost
algorithms. These algorithms were developed based on
actual operating information provided by steam
generating unit users and vendors. Therefore, the cost
data encompassed a wide variety of actual industrial
operating conditions, including periodic load swings.
While an individual plant could experience more severe
or frequent load swing episodes than those included in
the analysis, the cost estimates are based on conditions
found at a "typical" Industrial plant and are considered
to be representative of FGD costs in general.
The waste disposal costs Included 1n the cost algorithms
were based on an average SOg removal efficiency of 92
percent. Therefore, the costs associated with waste
disposal at higher percent reduction requirements were
considered in the analyses, and these costs are
considered representative of typical disposal costs.
The "baseline" for analysis of various standards
limiting SOg emissions from new Industrial-commercial -
institutional steam generating units represents
compliance with existing SIP requirements as well as
compliance with the final NSPS limiting particulate
matter and nitrogen oxides (N0X) emissions from new
industrial-commercial-Institutional steam generating
units. As a result, the costs to comply with ambient
SO2 standards were Included 1n baseline costs.
Similarly, costs for N0X control (the use of low excess
air and staged combustion air) were Included 1n the
baseline costs.
121
-------
In addition, the analysis of secondary environmental
impacts also examined the potential impact on ambient
air SOg concentrations of various standards limiting S02
emissions. The analysis indicated that none of the
ambient air quality standards or the limits on
deterioration of air quality, such as those under the
PSD program, were exceeded. Consequently, no additional
requirements to comply with ambient S02 standards or PSD
requirements were identified.
3. Comment: Some commenters mentioned site-specific differences in costs
that could be encountered. Several (IV-D-58, IV-D-62,
IV-D-66, IV-D-74, IV-D-84) said that the use of multiple
package FGD systems to represent a single large system is
inaccurate, because many large steam generating unit owners
will be forced to use a single field-erected unit due to
space limitations.
Another commenter (IV-D-21) expressed the opinion that the
cost analysis did not give adequate consideration to the
space required for a dry FGD system. The commenter said the
space needed for the actual control equipment, the water
treatment equipment, and for waste disposal would be
especially critical for steam generating units in an urban
area.
Response: Standards of performance are based on an analysis of impacts
associated with various regulatory alternatives. This
analysis focuses primarily on cases or scenarios which are
considered representative of the types of situations and
constraints individuals might generally face. In this
respect, as discussed earlier, the cost analysis did
consider the costs associated with additional space
122
-------
requirements for emission control equipment, the costs of
waste disposal, and the type of steam generating unit
installed (i.e., package or field erected). In addition,
the analysis also examined cases or scenarios representative
of specific situations where special provisions may be
appropriate. Thus, the analysis also examined, for example,
impacts on small as well as large steam generating units,
units operating at low capacity factors as well as high
capacity factors, units firing low sulfur as well as high
sulfur fuels, cogeneration systems as well as conventional
steam generating units, units located 1n the western United
States as well as those located 1n the eastern United
States, and steam generating units 1n specific Industries
likely to be most affected by standards.
In some cases, especially where space limitations restrict
the SO2 control equipment, a steam generating unit operator
will elect to Install a single, large, field-erected FGD
system rather than the smaller multiple package units
assumed for purposes of analysis. However, this should not
significantly affect the costs associated with most steam
generating units in the size range covered by the NSPS.
Although equipment costs for the shop-fabricated (package)
approach are higher than those for larger field-erected
systems, the total Installation costs (material plus labor)
are lower. Therefore, the underestimation of installation
costs for field erection will be largely offset by the
overestimation of equipment costs. Review of the analysis
also indicates that where the costs of FGD become excessive,
these costs can be avoided by firing an alternative fuel,
such as natural gas. The Impacts of such fuel switching
were evaluated and are considered reasonable. Thus, no
123
-------
special provisions for site-specific factors were judged to
be appropriate.
4. Comment: Two commenters addressed the impact of the capital costs of
FGD systems on oil-fired steam generating units. One
(IV-D-79) said the regulatory impacts analysis is
significantly understated and does not adequately address
the cost impact of the NSPS on the petroleum industry.
Specifically, the commenter said, forecasts that the capital
cost of a typical industrial steam generating unit would
increase by 4 to 25 percent under the proposed standard were
incorrect. The commenter's experience indicates that the
capital cost increase for a 73 MW (250 million Btu/hour)
unit would be 89 percent. The other commenter (IV-D-12)
also said that the capital costs of a new oil-fired steam
generating unit would be increased by 30-100 percent as a
result of the proposed rule. According to the commenter,
this will be amplified by the added costs to operate the
FGD system and dispose of the sludge.
Response: The capital costs associated with Installing FGD systems on
oil-fired steam generating units were assessed and discussed
in the "S02 Cost Report." This report discusses the
specific cost assumptions and methodologies used 1n
calculating the capital costs for both the uncontrolled
steam generating unit and the SO2 control device. For a 73
MM (250 million Btu/hour) heat Input capacity steam
generating unit 1n Region V, the capital cost for an
uncontrolled o1l-f1red steam generating unit was $4,579,000,
compared to $5,500,000 for a steam generating unit equipped
with an FGD system (including waste handling equipment).
This represents only a 20 percent Increase 1n capital cost,
rather than the 30 to 100 percent figures cited by the
124
-------
comnenters. It Is possible that steam generating unit
operators could experience costs that are higher or lower
than this depending on s1te-spec1f1c factors, but the
capital cost estimates 1n the "S02 Cost Report" are
considered representative of typical steam generating unit
costs.
5. Comment: Two comnenters felt that the capital costs of F6D systems
for coal-fired steam generating units were too high. One
(IV-D-87) felt that, based on capital costs for FGD systems
alone, smaller entitles cannot afford the S02 controls for
coal and residual oil-fired steam generating units. The
commenter said the cost of a new 44 MM (150 million
Btu/hour) unit with a baghouse and spray dryer for S02
control (or conversion to fludlzed bed combustion) 1s
approximately $6 million. Another (IV-F-1.18) contended
that a requirement for FGD systems will Increase project
capital costs by at least $1 million and will not result in
sufficient operating cost savings to justify the Investment.
As a result, the commenter said, few Industrial coal-fired
units will be built in the future.
Response: The capital costs associated with Installing FGD systems on
coal-fired steam generating units were discussed 1n the "S02
Cost Report." As discussed above for oil-fired units, the
assumptions used 1n calculating these costs for both the
uncontrolled steam generating unit and the S02 control
device were detailed in this report. For a 44 MW
(150 million Btu/hour) heat Input capacity steam generating
unit 1n Region V, the capital cost for an uncontrolled
bituminous coal-fired steam generating unit was $14,050,000,
compared to $14,899,000 for this same steam generating unit
equipped with a sodium FGD system. This represents an
125
-------
increase of about $850,000, or 6 percent of the capital
cost. The capital cost of a lime spray drying system, while
higher than that associated with sodium scrubbing, would add
only about 10 percent to the uncontrolled steam generating
unit costs. Therefore, the costs associated with the steam
generating unit itself will likely play a far more important
role in determining affordability than the cost of the S02
control system.
Data from the previously discussed boiler replacement survey
indicate that of the 89 projects surveyed, only 7 projects
(8 percent) would have been changed if project costs
increased by 10 percent. These data indicate that while
some steam generating unit projects may be altered based on
the results of the survey, most projects would have been
built as planned.
6. Comment: One commenter (IV-D-42) said the cost impacts of adding an
FGD system to a municipal resource recovery incinerator
which fires oil as a startup fuel only were much higher than
those cited in the proposal. Specifically, the commenter
said the annualized costs of a dry FGD system when examined
over a 20-year period could add 25 to 45 percent to the
operating cost of a municipal resource recovery incinerator
when considering all cost factors, including the loss in
revenue because of a decrease in net electric generation.
The commenter noted that this is considerably higher than
the 6 to 22 percent increase projected at proposal.
Response: In the final standard, a municipal resource recovery
incinerator firing a very low sulfur oil would not be
subject to the percent reduction requirement.
126
-------
7. Comment: One commenter (IV-F-1.19) discussed the viability of steam
generating unit installations under various alternatives.
The commenter said that at one recent installation, the
payback period was 3.5 years for low sulfur coal, 4.2 years
for a dry scrubber, and 4.3 years for a wet scrubber.
According to the commenter, this small range was enough to
cause the steam generating unit not to be built if scrubbers
were required. At another installation, the payback period
was 2.5 years for low sulfur coal, versus 3.1 years for a
dry scrubber and 3.3 years for a wet scrubber. Again, the
commenter said, with a scrubber requirement, the project
would not have been built.
Response: The economics of installing a steam generating unit, with or
without an FGD system, will vary from site to site. As
evidenced by the commenter's example, a payback period of
3.5 years may be acceptable for one facility, whereas a
payback period of 3.3 years may be unacceptable for another.
In developing national standards of performance, it is not
possible to predict impacts on a site-specific basis. The
impacts of the standards have been determined to be
reasonable on a national basis, as well as for individual
! industrial-commercial-institutional steam generating units
in general. If the costs associated with installing and
operating an SOg control system would prove to be too high
for a particular steam generating unit, the steam generating
unit operator could elect to fire natural gas or very low
sulfur oil in order to avoid the costs of SOg control.
8. Comment:
One commenter (IV-D-48) said the cost per ton of SO2 removed
under a percent reduction requirement is higher for low
sulfur coals than for high sulfur coals due to the smaller
127
-------
amounts of sulfur present. The commenter contended that it
makes no sense to penalize sources using lower sulfur fuels
by not providing a "sliding scale" percent reduction
requirement.
Response: The commenter is correct in saying that the cost per ton of
SO2 removed could be higher for low sulfur coals. However,
this does not "penalize" sources using low sulfur fuels. Of
more concern to the steam generating unit operator 1s the
total annualized cost associated with SOg control. Total
annualized costs actually can be lower for lower sulfur
coals due to the reduced operational and waste disposal
costs associated with lower sulfur coal. Analyses of a
sliding scale standard analogous to Subpart Da for
industrial-commercial-institutional steam generating units
found that the additional cost of 90 percent SO2 removal was
reasonable relative to the sliding scale.
9. Comment: Two conmenters (IV-D-45, IV-D-76) said the proposal states
that the percent reduction standard 1s not significantly
more costly than a low sulfur fuel standard. However, by
the proposal's estimates, the national annualized cost
associated with a standard based on percent reduction is
$133 million, compared to $57 million for a standard based
on the use of low sulfur fuel. The commenters noted that
this 1s an Increase of more than 130 percent.
One (IV-D-76) added that the national cost effectiveness
calculation, which projects "average" national incremental
removal costs of $940/Mg ($850/ton) over a low sulfur fuel
standard, underestimates the cost of a scrubber-based
percent reduction requirement for steam generating units
using coal or oil due to the fuel switching assumption.
128
-------
According to the commenter, this number is well below the
costs that could be incurred by individual owners and
operators not switching to natural gas, which could range as
high as $4,400/Mg ($4,000/ton). Also, the commenter said,
some steam generating unit classes operating at lower
capacity factors incur control costs that are comparable to
the $1,760 Mg (SI,600/ton) that was considered
"unreasonable" for mixed fuel units. Another commenter
(IV-D-96) agreed, saying that the implications of the fuel
switching assumption on whether the standard would have
unreasonable impacts were not properly considered.
Response: The national impacts associated with the standard were
reassessed using the Industrial Fuel Choice Analysis Model
[i.e., IFCAM) and revised fuel price projections (see
"Revised Impacts of Alternative Sulfur Dioxide New Source
Performance Standards for Industrial Fossil Fuel-Fired
Boilers"). The new analysis projects that even in the
absence of an NSPS, no new coal-fired steam generating units
would be built due to the capital-intensive nature of these
units and the relatively small price differential among
coal, oil, and natural gas. In addition, the prices of
! natural gas and oil appear to be so competitive that even an
NSPS establishing an emission limit only (i.e., without a
percent reduction requirement) causes almost all
industrial-commercial-institutional steam generating units
to fire natural gas. Thus, the analysis of national impacts
projects little or no difference at the national level
between the costs associated with firing low sulfur fuels
and those associated with achieving a percent reduction in
S02 emissions. Either alternative would be expected to
result in total nationwide annualized costs of $5 million
(assuming high oil prices) to $50 million/year (assuming low
129
\
-------
oil prices). Because of the similarity in cost estimates
under both regulatory options, the incremental cost
effectiveness of a percent reduction standard compared to a
low sulfur coal standard is projected to be negligible. An
alternate analytical approach developed to estimate national
impacts under the assumptions that many
industrial-commercial- institutional steam generating units
will be coal fired and that fuel switching does not occur
was also used. The national cost estimates under these
assumptions are $23 million for a low sulfur fuel standard
and $124 million for a percent reduction standard. The
estimated Incremental cost effectiveness of a percent
reduction standard under these assumptions is $1,600/Mg
($l,400/ton). These costs are considered reasonable
considering the significant emission reductions achieved by
a percent reduction requirement.
It is possible that individual steam generating unit
operators could incur costs that are significantly higher
(or lower) than the national average cost effectiveness
values. In particular, the costs associated with achieving
a percent reduction in SO2 emissions could be unreasonably
high for steam generating units that receive only a small
amount of their total heat Input from coal or oil, or those
that operate at only a small portion of their total annual
capacity. For this reason, 1n the final standard, steam
generating units operating at an annual capacity utilization
factor for coal or oil, or a mixture of coal and oil, of
less than 30 percent are not subject to a percent reduction
requirement, but need meet only an emission limit.
130
-------
2.7.3 Waste Disposal Costs
1. Comment: Several commenters (1V-D-58, 1V-D-62, 1V-D-66, 1V-D-74,
IV-D-84) said that waste disposal costs will greatly
increase if FGO system sludge is designated a hazardous
waste subject to disposal requirements under the Resource
Conservation and Recovery Act (RCRA). They felt this had
not been considered in developing the standards.
Response: Sludge generated by FGD systems 1s not currently considered
a hazardous waste under RCRA and, thus, 1t would be
inappropriate to consider it as such in the analysis. If
this should happen, however, it could alter the relative
economics of various SOg control systems, making the costs
of those systems producing larger volumes of wastes or
wastes with certain undesirable characteristics higher than
the costs for other types of systems. Thus, such a decision
could alter the choice of SOg control system. In addition,
if the costs increased to the point where they were regarded
as excessive in certain cases or at specific sites, then
such a decision could alter the selection of fuel for the
steam generating unit. In such situations, use of very low
sulfur fuels, such as natural gas, would avoid these waste
disposal costs. As mentioned earlier, the impacts
associated with such fuel switching are considered
reasonable. Thus, regardless of whether FGD or FBC system
waste is designated a hazardous waste under RCRA, the
impacts associated with the standards are considered
reasonable.
2. Comment: One commenter (IV-D-44) asserted that the ash disposal costs
associated with the standards were greatly underestimated.
The commenter said that without even considering the extreme
131
-------
costs of disposal in lined landfills, ash disposal costs
alone are 512/ton in Michigan, 545/ton in New Jersey, and
$25/ton in Virginia.
Response: Ash handling and disposal systems and their associated costs
were included in the cost algorithms as part of the
uncontrolled steam generating unit (i.e., "baseline") costs.
The costs of ash disposal, however, do not arise as a result
of compliance with SO2 control requirements under standards
of performance. The need for ash disposal results from
burning the fuel itself, not from controlling SO2 emissions.
Thus, these costs have little bearing on considerations
regarding selection of standards limiting SOg emissions.
3. Comment: One commenter (IV-D-48) said that the costs for industries
which dispose of their own wastes were not considered.
According to the commenter, these costs should include
capital costs relating to landfill construction, monitoring
costs, site closure costs, oxidation equipment and operating
costs, and permit application and monitoring costs.
Response: The cost algorithms assume that most steam generating unit
operators would elect to dispose of their wastes off-site
rather than invest in waste treatment and disposal systems.
The waste disposal costs used in the algorithms reflect
the typical costs of off-site disposal. The costs
associated with on-site disposal, for industries which
dispose of their own wastes, could be higher or lower than
the off-site costs assumed by the cost algorithms, depending
on site-specific situations or local requirements. If
on-site disposal costs are lower, the cost algorithms will
overestimate costs. If on-site disposal costs are higher,
operators will likely select off-site disposal and the cost
132
-------
algorithms will more accurately reflect costs. Thus, the
assumption of off-site disposal is a "conservative"
assumption which, if anything, is more likely to
overestimate costs in cases of on-site disposal.
4. Comment: Several commenters felt that sodium FGD system wastewater
treatment costs were underestimated. Some (IV-D-26,
IV-D-30, IV-D-50, IV-D-53, IV-D-73, IV-F-1.15) said that
sodium scrubber wastewater treatment costs such as 1on
exchange or reverse osmosis technology could be incurred at
some facilities and should be considered. According to the
commenters, these costs can be significant
($200,000-$500,000).
One commenter (IV-F-1.15) said that the estimate of total
costs for oxidation of the sodium FGD system waste stream
($60,000) was low by a factor of 2 to 3, assuming discharge
to a publicly owned treatment works (POTW).
Response: The cost associated with sodium FGD wastewater treatment and
disposal was represented 1n the analyses by oxidation of the
wastewater stream and discharge to a receiving water body or
POTW. These costs were discussed in "SO2 Re-em1ssions from
the Sodium Scrubbing Wastewater Stream 1n Aerobic
Environments." While total direct costs for a wastewater
stream with a small sulfite loading were estimated at
$60,000, the total turnkey costs associated with Installing
an oxidation tower ranged from approximately $95,000 for a
small wastewater stream to $430,000 for a large wastewater
stream. These costs were based on actual vendor data and
are considered to be accurate within the range of accuracy
of the cost estimates in general. Costs for other
wastewater treatment techniques, such as ion exchange or
133
-------
reverse osmosis technology, were not considered in the
analysis. Although these costs could be incurred at some
facilities depending on site-specific disposal requirements,
they are not considered representative of "typical" costs
that would be incurred at most facilities. If the costs
increased to the point where they were regarded as excessive
by the steam generating unit operator, an alternative FGD
system, such as lime spray drying, could be used to change
the nature of the waste or, alternatively, a very low sulfur
fuel such as natural gas could be fired in the steam
generating unit to avoid these costs.
2.7.4 Startup. Shutdown, and Malfunction Costs
1. Comment: Many commenters (IV-D-6, IV-D-22, IV-D-26, IV-D-28, IV-D-30,
IV-D-32, IV-D-40, IV-D-44, IV-D-50, IV-D-51, IV-D-52,
IV-D-53, IV-D-58, IV-D-61, IV-D-62, IV-D-65, IV-D-66,
IV-D-73, IV-D-74, IV-D-84, IV-D-88, IV-F-1.6, IV-F-1.16)
expressed concern that the capital costs of auxiliary fuel
systems necessary for startups, shutdowns, and upsets were
not considered. Conanenters specifically mentioned costs for
burners, piping, storage, and other equipment, and said that
these costs are significant. In addition, commenters noted
that if very low sulfur oil is used as the backup fuel, the
costs for low-NOx burners should be included to achieve the
NOx limits. According to the commenters, typical costs for
providing low sulfur oil backup capability could be
$300,000-$500,000.
Response: Many steam generating units are already designed and
constructed with alternative fuel firing capability to allow
greater flexibility in fuel selection and to provide for
steam generating unit startup capability. In these cases,
134
-------
the only additional costs for firing natural gas or very low
sulfur oil during periods of startup, shutdown, and
malfunction would be the difference in the price between
natural gas or very low sulfur oil and coal. However,
because not all boilers have this capability, additional
capital costs were included in the cost analysis to provide
the flexibility of switching from firing coal or oil to
firing natural gas or very low sulfur oil during periods of
malfunction. For example, in the case of switching from
coal to a very low sulfur oil, additional costs were
included for oil guns, piping, flow control valves, burner
control system, fuel oil heaters, pumps, oil storage tanks,
additional draft control, coal shutoff gates, cinder
reinjection cutoff gates, and, in the case of stoker coal
steam generating units, thermocouples on the stoker gate.
For all model steam generating unit sizes analyzed (100,
150, 250 and 400 million Btu/hour), the capital cost of
providing alternative fuel firing capacity to a coal-fired
steam generating unit was about two to three percent of the
total steam generating unit cost. The annualized cost of
providing alternative fuel firing capability was three to
four percent of the annualized cost of the steam generating
unit system. Therefore, while the additional capital costs
of providing very low sulfur fuel backup capability could be
in the range suggested by the commenter, they would still
represent only a very small percentage of the total steam
generating unit costs while providing significant benefits
in terms of additional SOg control.
Relative to N0X control costs, they were fully addressed in
the promulgation of the N0X standard on November 25, 1986
(51 FR 42768). Additionally, even if they were not included
in the analysis, the cost of low N0X burners compared to
135
-------
conventional burners 1s Insignificant, and 1n relation to
S02 control cost where percent reduction requirements are
applicable, would not affect any conclusions related to the
reasonableness of percent reduction requirements.
2. Comment: One commenter (IV-D-32) said the rationalization for a 90
percent reduction requirement seems to rely heavily on more
costly backup fuels which may or may not be available. The
commenter said that fuel costs appear to have been
underestimated, especially considering the likelihood of
frequent startup and shutdown events experienced by some
Individual steam generating units such as those operated
only on weekdays.
Response: The Impact of startup and shutdown procedures was
considered, including the potential need to fire alternative
fuels during these procedures. Information provided by
scrubber vendors, however, Indicates that the
startup/shutdown time for FGD systems 1s minimal and that
this time can be integrated 1n such a way with normal
startup and shutdown procedures associated with the steam
generating unit that periods during which the control system
might need to be bypassed would be extremely brief. As a
result, 1n most cases there should be no necessity, beyond
conventional practice, for firing alternative fuels during
such periods.
In most cases steam generating units subject to frequent
startups and shutdowns are likely to be gas-f1red, not only
for ease of operation and simplicity, but also for economic
reasons. Steam generating units subject to frequent
startups and shutdowns are also likely to operate at low
annual capacity utilization rates. Thus, as discussed
136
-------
previously, such low capacity units would not be subject to
a percent reduction requirement. The economics of low
capacity factor steam generating units also favor the
selection of alternative fuels, such as natural gas, over
selection of coal. In addition, many coal-fired steam
generating units normally use premium fuels such as natural
gas for startup. This is done not only to bring the steam
generating unit up to operating temperature, but to serve as
a source of ignition for the coal. Thus, the standards are
likely to have little impact on steam generating units
subject to frequent startups and shutdowns.
However, in order to consider those few situations where the
need to cope with frequent startups and shutdowns might
impose additional costs, such as the need to fire natural
gas or very low sulfur oil, the increased costs associated
with those situations were examined. Assuming that a
typical 44 MW (150 million Btu/hour) coal-fired steam
generating unit (operating at an overall annual capacity
utilization factor of 0.6) undergoes startup or shutdown
about five percent of the time it is in operation, the
increased costs of firing natural gas during these periods
i would increase the annualized cost of the steam generating
unit and FGD by less than one percent. More important,
however, the incremental cost effectiveness of the
additional SOg emission control achieved would be less than
$1,100/Mg ($l,000/ton). This is not considered
unreasonable. Consequently, no special provisions are
considered necessary to accommodate startup/shutdowns.
3. Comment: Two commenters (IV-D-31, IV-F-1.16) felt that higher costs
should be assigned to FGD system malfunction because
downtime for manufacturing firms can be very costly.
137
-------
Response: The reliability of the emission control system and any FGD
malfunction does not affect the reliability of the
production unit. The model systems analyzed in the
development of this standard were designed to include
alternative fuel firing capabilities which would preclude
the necessity for any downtime in the production unit as a
result of FGD malfunction. By switching to an alternative
fuel, such as natural gas, the production unit would be able
to continue operations with essentially no interruptions.
Outside of FGD system malfunction, routine inspection and
maintenance of the FGD system could be scheduled for periods
when the steam generating unit and/or the production units
are also shut down for routine inspection and maintenance,
thus avoiding any impact on the operation of production
units.
2.7.5 Monitoring. Recordkeeping, and Reporting Costs
1. Comment: One commenter (IV-D-28) said the assumption that steam
generating units using low sulfur fuels still be required to
perform continuous emission monitoring results in monitoring
costs that exceed pollution control costs in some cases.
The commenter suggested that provisions for alternative
monitoring or sampling procedures such as those in the
current NSPS would markedly improve the cost effectiveness
of compliance fuels and coal cleaning technologies, while
not affecting the cost effectiveness of FGD.
Response: Under the final standards, alternative monitoring procedures
are available in those cases an owner or operator judges
continuous emission monitoring system (CEMS) costs to be
unreasonable. Fuel sampling and analysis at the inlet to
the SOg control device can be used in lieu of a CEMS, and
138
-------
Method 6B stack sampling (see 40 CFR Part 60, Appendix A,
Method 19) can be used in lieu of an outlet CEMS for those
facilities where percent reduction requirements are
applicable. For steam generating units operating with
annual capacity utilization factors for coal or oil of less
than 30 percent or firing very low sulfur oils, any one of
three monitoring methods may be used: inlet fuel sampling,
outlet CEMS, or stack sampling using Method 6B. Only a
single monitoring alternative is necessary for these units.
In addition, the NSPS General Provisions in 40 CFR 60.13(i)
provide for alternative monitoring procedures, subject to
approval by the Administrator, under various conditions.
Therefore, if monitoring costs would be excessive in certain
circumstances, alternatives are available to reduce these
costs.
2. Comment: Several commenters (IV-D-26, IV-D-30, IV-D-32, IV-D-50,
IV-D-53) said that costs for backup monitoring equipment,
backup data processors, and the manhours needed to comply
with the monitoring, reporting, and recordkeeping
requirements were not considered.
Response: The minimum data availability requirement of 22 out of
30 steam generating unit operating days was established to
minimize the need for backup monitoring equipment. This
level of data availability is achievable unless the
equipment is not being properly operated or maintained. To
account for infrequent and unusual circumstances, however,
in which data may not be available, data estimation
procedures have been provided in Method 19, Section 7.
139
-------
Manhours associated with data collection and preparing
periodic reports were calculated and considered, and a
detailed breakdown of the manhours associated with the
reporting and recordkeeping requirements 1s Included 1n
docket for this rulemaking.
140
-------
2.8 PERFORMANCE/RELIABILITY OF DEMONSTRATED TECHNOLOGIES
2.8.1 Flue Gas Desulfurization
1. Comment: Many commenters (IV-D-26, IV-D-30, IV-D-32, IV-D-40,
IV-D-50, IV-D-53, IV-D-62, IV-D-65, IV-D-66, IV-D-72,
IV-D-73, IV-D-74, IV-D-84, IV-D-96, IV-F-1.1, IF-F-1.8,
IF-F-1.12, IV-F-1.16, IV-F-1.18) said there is inadequate
proof that there are demonstrated technologies which can
meet the 90 percent removal requirement on a continuous
basis. They contended that the performance of
"demonstrated" technologies, such as flue gas
desulfurization (FGD), has largely been gathered on utility
steam generating units while ignoring the totally different
design and operating requirements of industrial units.
Response: The performance data for FGD systems discussed in the
"Summary of Regulatory Analysis, Fossil Fuel-Fired
Industrial Boilers-Background Information," and "SOg
Technology Update Report" are based on experience with
industrial steam generating units. Moreover, as also
discussed, for several types of FGD systems this experience
is supported by experience with utility FGD systems.
Lime and limestone FGD systems, because of higher capital
and maintenance costs, have had limited application in the
industrial sector. In a long-term test at an industrial
steam generating unit that operated a lime/limestone FGD
system, the S02 removal efficiency was over 91 percent for
lime and 94 percent for limestone. Removal efficiencies
were insensitive to changes in steam generating unit load.
In addition, lime and limestone wet FGD systems are proven
processes in the utility industry. Lime and limestone
141
-------
FGD systems account for approximately 68 percent of the
total number of utility FGD systems. Although few, if any,
lime or limestone FGD systems are expected to be installed
on industrial-commercial-institutional steam generating
units, there are no technical limitations which would make
this technology less effective on industrial steam
generating units than on utility steam generating units.
The performance and reliability of lime spray drying systems
and fluidized bed combustion (FBC> steam generating units
have also been demonstrated in industrial applications and
are discussed elsewhere in this section.
Dual alkali is the second most prevalent FGD system in use
on industrial steam generating units. Tests of industrial
dual alkali FGD systems have shown average SOg removal
efficiencies of around 90 percent, with long-term
efficiencies of around 92 percent. Removal efficiencies of
95 percent or greater have been achieved on a continuous
basis under certain operating conditions (such as operating
in a dilute mode with low TDS concentrations).
Sodium FGD systems have been the FGD system most often used
on Industrial steam generating units. Therefore, the
industrial data base for this technology is more extensive
than those for other FGD technologies. Both short- and
long-term tests of sodium FGD systems on industrial steam
generating units showed consistent SO2 removal efficiencies
of greater than 90 percent. Emissions data from
45 industrial FGD systems, representing 18 sites, showed an
average S02 removal efficiency of greater than 96 percent.
142
-------
Consequently, the ability of FGD systems to achieve a
90 percent reduction in S02 emissions is well demonstrated
by the experience of a number of different FGD systems
currently operating on industrial steam generating units.
The experience from similar systems operating on utility
steam generating units only serves to confirm and add
additional support to this experience.
2. Comment: One coironenter (IV-F-1.15) stated that the data presented
with the proposal do not support the 90 percent S02 removal
standard for oil-fired steam generating units and are not
representative of typical industrial steam generating unit
performance. Specifically, the commenter said there are no
long-term test data available during oil firing, and the
short-term data are heavily dependent upon Kern County oil
field steam generating units. The commenter asserted that
these base load units operate at constant loads, in contrast
to typically widely swinging industrial steam generating
unit loads, and therefore are not representative of
industrial steam generating unit performance.
Response: Industrial FGD experience on oil-fired steam generating
units has been generally limited to sodium FGD systems. As
discussed in the "S02 Technology Update Report," a large
portion of the sodium FGD system performance data base was
gathered from oil field steam generating units which are
used to enhance oil recovery. The S02 removal efficiency of
sodium FGD systems operating on these units averaged over
95 percent. These steam generating units do operate at
fairly constant load; however, data were also gathered from
sodium FGD systems operating on oil-fired steam generating
units 1n other industrial applications, Including those with
typically widely swinging loads. A series of short-term
143
-------
compliance tests conducted on a number of sodium FGO systems
operating on oil-fired steam generating units in various
industrial manufacturing plants showed SO2 removal
efficiencies ranging from 89.3 to 99.4 percent and averaging
96.5 percent. Also, long-term data have shown average S02
removals of 95 percent at high (98 percent) reliability
levels for sodium FGO systems operating on oil-fired steam
generating units.
Although dual alkali FGD systems have generally not been
applied on oil-fired steam generating units, they should
perform just as well as sodium FGD systems. The SOg control
loop of dual alkali FGD systems operates 1n basically the
same manner as a sodium FGD system; thus, dual alkali
FGD systems can attain SO2 control performance levels
equivalent to sodium FGD systems.
In addition, emissions from coal-fired steam generating
units contain more trace metals and other elements that
could adversely affect the SOg removal efficiency of an FGD
system than emissions from oil-fired steam generating units.
Therefore, FGD performance on oil-fired units should be at
least as good as FGD performance on coal-fired units.
3. Comment: One commenter (IV-D-62) said that an average 90 percent
reduction 1s feasible only when a steam generating unit 1s
using a high sulfur fuel; 1t 1s less feasible when a low
sulfur fuel is used because of the lower dilution of the
pollutant to be removed.
Response: The ability of an FGD system to achieve high levels of S02
removal during combustion of low sulfur fuels is determined
by the concentration of SOg in the flue gas exiting the FGD
144
-------
system. Test data from sodium and Time spray drying FGD
systems and from FBC systems indicate S02 concentrations of
less than 15 ppm are achievable (1n some tests, S02
concentrations of less than 5 ppm were measured); a
concentration of 15 ppm 1s roughly equal to 0.025 lb
S02/m1ll1on Btu. This equates to 90 percent removal on an
inlet flue gas with 0.25 lb SOg/mllHon Btu, or 98 percent
on an Inlet flue gas of 1.2 lb SOg/mllHon Btu. These
performance levels measured 1n operating FGD systems are
supported by kinetic data from laboratory studies and by FGD
vendor claims. Consequently, the ability to achieve a
90 percent reduction when firing low sulfur fuels 1s
considered well demonstrated.
4. Comment: Several commenters questioned the adequacy of the data
presented on FGD performance. Some (IV-D-6, IV-D-26,
IV-D-28, IV-D-30, IV-D-40, IV-D-50, IV-D-52, IV-D-53,
IV-D-73, IV-D-96, IV-F-1.15, IV-F-1.16) stated that the
analysis of FGD performance significantly underestimates the
Impacts on industrial steam generating unit owners of a
mandatory percent reduction requirement. They claimed that
variable operating modes and high capacity load swings
typical of Industrial operations can cause severe upsets in
scrubber efficiency and reliability. Also, the commenters
said, the test data 1n most cases did not reflect full load
operations, and, therefore, did not reflect typical
operating conditions.
Others (IV-D-26, IV-D-30, IV-D-50, IV-D-51, IV-D-53,
IV-D-88, IV-F-1.15, IV-F-1.16) noted that the assessments of
FGD performance 1n removing S02 stressed the Importance of
several critical variables to FGD system performance. They
said, however, that the performance test data presented did
145
-------
not report these variables in many cases, making it
impossible to correlate emissions data with these key
variables.
Response: The FGD system performance data base is primarily composed
of data collected from industrial steam generating unit
installations. These steam generating units were located at
plants representative of the industrial-commercial -
institutional steam generating unit population, including
steam generating units operated under many different
conditions. Steam generating units with average loads
(capacity utilization factor) ranging from 5 to 100 percent
were included in the data base. In addition, steam
generating unit loads were varied during tests of individual
units to simulate load swings that might be experienced in
some industrial applications. Based on this data, SOg
removal efficiency was found to be insensitive to changes in
steam generating unit load over the ranges observed.
The primary concern for FGD systems operating on steam
generating units which experience load swings is a sudden
increase in the SOg loading. This can result from an
increase in either the flue gas flow rate or the flue gas
SO2 concentration. As discussed in the "SOg Technology
Update Report," changes in flue gas flow rate are matched by
corresponding changes in the scrubbing solution flow rate
according to a set liquid-to-gas (L/G) ratio. In a well
designed and operated system, a safety margin is maintained
in the L/G ratio to account for delays in control loop
response; thus, an increase in flue gas flow rate would be
adequately handled. Also, FGD systems which experience
highly variable SOg loadings typically operate at high
alkaline reagent concentrations. This provides a buffering
146
-------
capacity against large swings 1n solution pH caused by
dramatic changes in S02 concentration. As a result,
sufficient excess alkaline reagent 1s present to ensure
adequate SO2 removal performance during load swings.
5. Comment: A number of commenters questioned the reliability of FGD
systems. Several (IV-D-6, IV-D-26, IV-D-28, IV-D-30,
IV-D-40, IV-D-50, IV-D-52, IV-D-53, IV-D-73, IV-F-1.16,
IV-F-1.17) said that although the proposed regulation
assumes that reliable FGD systems can be economically
constructed and operated, the operational history of FGD
systems does not bear this out. Two other commenters
(IV-D-44, IV-F-1.19) asserted that no environmentally
acceptable FGD system has been demonstrated to meet the high
availabilities (around 85 percent) required by some
facilities, such as cogeneration facilities. In addition,
two commenters (IV-D-26, IV-D-35) expressed concern that
decreased steam generating unit reliabilities due, in part,
to the uncertainty of the FGD system can result in extremely
high costs incurred due to process shutdowns in many plants
at which continuous steam generating unit operation 1s
required. The commenters felt that this incompatibility
between FGD efficiency levels and industrial steam
generation requirements has not been addressed.
Four commenters (IV-D-26, IV-D-30, IV-D-50, IV-D-53) also
maintained that utility FGD operability and reliability
experience cannot be extrapolated to industrial operation
because most utilities install spare scrubbers to improve
reliability, do not experience the radical load fluctuations
typical of industrial operations, and have enough operation
and maintenance personnel to keep equipment running, unlike
small industrial facilities.
147
-------
Response: The reliability of various types of FGD systems for
industrial applications was discussed in the "Background
Information Document," the "Summary of Regulatory Analysis,"
and the "SOg Technology Update Report." For lime spray
drying systems, reliability levels ranging from 70 to 97
percent were reported for various test sites. Some decrease
in reliability was reported with increasing SO2 removal
efficiency. However, an examination of the reasons for
decreased reliability indicated that the FGD system failures
were not generally the result of increased system stress,
but were due to other factors unrelated to SO2 removal
efficiency and could often have been prevented with improved
operating and maintenance procedures or maintaining a spare
parts inventory. In fact, one vendor of lime spray dryers
is prepared to offer a guarantee of 95 percent reliability
providing the customer follows a preventive maintenance
program. The ability of Hme spray dryers to achieve high
reliability levels 1s also supported by the history of one
Hme spray dryer operating on a 132 MW (450 million
Btu/hour) steam generating unit in Michigan, which over
2 years (1985 and 1986) has achieved 1n excess of 98 percent
reliability.
During a long-term (85-day) performance test of a
lime/limestone FGD system on an Industrial steam generating
unit, FGD system reliability was greater than 90 percent
(91 percent when lime was the reagent and 94 percent when
limestone was the reagent). Limestone FGD systems have also
demonstrated high reliabilities when applied to utility
steam generating units. Data from one utility
lime/limestone FGD system firing coal indicate reliabilities
close to 100 percent over a period of several years.
148
-------
Data for sodium FGD systems also indicate high reliabilities
for industrial applications. Average reliability levels
of 98-100 percent have been reported for over 250 coal- and
oil-fired steam generating units for long periods of time
(several months to several years in length) at high SO2
removal levels.
The reliability of fluidized bed combustion systems is
discussed in a subsequent section of this document.
The above discussion indicates, therefore, that with proper
operation and maintenance, high FGD reliabilities can be
achieved and maintained on industrial steam generating units
operating at high SOg removal levels. For compliance during
periods of FGD malfunction, the cost of firing alternative
low sulfur fuels, such as natural gas, in the steam
generating unit was included in the cost calculations. As
an alternative approach, installation of backup FGD modules,
common in the electric utility sector to ensure system
reliability, may be used on industrial-commercial-
institutional steam generating units.
6. Comment: Several coiranenters expressed the opinion that the
performance and reliability of dry FGD systems have not been
demonstrated in industrial applications. Three commenters
(IV-D-44, IV-F-1.17, IV-F-1.18) said that lime spray drying
systems have a dismal performance and reliability record in
the few installations at which they are operating, and
should not be included in the standards. Another (IV-D-48)
said the 90 percent removal requirement will preclude dry
FGD technologies, forcing facilities firing low sulfur coals
to live with wet SO£ removal options. Others (IV-D-26,
IV-D-30, IV-D-50, IV-D-53) stated that the "Summary of
149
-------
Regulatory Analysis" Identifies no exceptions to the
application of spray dryers to coal-fired steam generating
units, while "Projected Environmental, Cost and Energy
Impacts of Alternative SOg NSPS for Industrial Fossil
Fuel-Fired Boilers" notes that lime spray drying is not a
control option for medium and high sulfur coals regardless
of the S02 removal rate. They felt that this was
inconsistent with the statement that lime spray drying is a
demonstrated technology.
Response: The performance and reliability of lime spray drying has
been demonstrated, as discussed in the "Summary of
Regulatory Analysis" and the "SOg Technology Update Report."
The citation quoted by the commenters to the contrary 1s the
result of an error that had been overlooked prior to
publication. Although little long-term data are available
to demonstrate high SO2 removal levels, short-term tests
have indicated that Hme spray drying systems are capable of
achieving performance levels 1n excess of 93 percent. The
long-term average removal levels of 60 to 80 percent that
have been observed to date are not Indicative of an Inherent
Inability to achieve high levels of performance over long
periods of time, but rather reflect the fact that these
systems have not been required to achieve such high removal
levels. Therefore, owners and operators have operated them
at the minimum efficiency required to save on operating
costs. As discussed above, high FGD reliabilities have been
demonstrated for Hme spray dryers. Although most Hme
spray dryers today are operated at moderate performance
levels, there 1s no reason to believe 11me spray dryers
could not maintain these high reliabilities with Increasing
S02 removal. Nothing 1n the results demonstrating the
capability to achieve high performance levels of 90 percent
150
-------
reduction suggests any reasons why high reliabilities cannot
be sustained along with 90 percent removal levels. In fact,
those periods of time during which high removal levels have
been achieved 1n performance tests Indicate that lime spray
drying systems are capable of achieving high performance
levels with high reliability levels. In addition,
performance guarantees for commercial spray drying systems
are available for 90 percent S02 removal and 95 percent
system reliability.
2.8.2 Fluldized Bed Combustion
1. Comment: A number of coimienters expressed the opinion that fluldlzed
bed combustion (FBC) has not been demonstrated to achieve
90 percent SO^ removal 1n Industrial applications. Several
{IV-D-23, IV-D-26, IV-D-30, IV-D-40, IV-D-43, IV-D-50,
IV-D-53, IV-D-58, IV-D-62, IV-D-66, IV-D-73, IV-0-74,
IV-D-81, IV-D-84, IV-F-1.1, IV-F-1.8, IV-F-1.16) said the
data base cited in the proposal does not support the
contention that an FBC unit can achieve 90 percent SOg
control on a 30-day rolling average basis. They contended
that the data on which this conclusion was based are
one-time, short-term programs which are not representative
of long-term emissions. Further, the commenters said, the
monitoring used in collecting the data did not include
continuous monitors on the Inlet and outlet, or coal
sampling and analysis on units of a size similar to those
being regulated. The commenters felt that until sufficient
data are collected, FBC should be classified as an emerging
technology.
151
-------
Others (IV-D-6, IV-D-26, IV-D-28, IV-D-30, IV-D-38, IV-D-40,
IV-D-50, IV-D-52, IV-D-53, IV-F-1.18) agreed, saying that
new technologies, such as FBC, currently lack sufficient
operating data upon which to base NSPS regulations. The
commenters expressed concern that the Inclusion of such
units In the standards could seriously Impede the full
demonstration and acceptance of this technology. Another
commenter (IV-F-1.19) added that Industry as a whole has
little more faith 1n FBC units than they do in other FGD
systems, since their availability and reliability have not
yet met industry expectations. One commenter (IV-F-1.16)
also felt that the high reliabilities required by industrial
steam users have not been demonstrated with FBC systems.
The commenter said that the only long-term reliability data
presented 1n support of the proposed standards show 92 and
93 percent, which is not up to normal Industrial
requirements.
Response: Sufficient short-term performance test data exist to
demonstrate that FBC technology is capable of achieving
90 percent SOg removal. Short-term data presented in the
"Summary of Regulatory Analysis" show that several FBC units
achieved 90 percent or greater SOg control during the test
periods.
Long-term data also demonstrate that FBC units are capable
of achieving greater than 90 percent SO2 removal at high
reliability levels. A recent 30-day test on a bubbling bed
FBC unit burning high sulfur coal showed SO2 removal
efficiencies averaging 93.5 percent. A 30-day test
conducted at a different site showed an average SO2 removal
of 90 percent with greater than 99 percent reliability.
During a 67-day period at this site, the FBC unit had a
152
-------
reliability of 97 percent. In addition, vendors have stated
that FBC units can be designed to achieve well over 90
percent SOg removal at high reliability levels.
Approximately 21 coal-fired FBC units are currently
operating in the industrial, commercial, and institutional
sectors in the U.S. with heat inputs of 22 MW (75 million
Btu/hour) or greater. Nineteen more in this size range are
now being planned or are under construction. Given the
numbers of FBC units in operation and in the
planning/construction stage and the performance of operating
units, FBC is considered a demonstrated technology.
2. Comment: Several comnenters (IV-D-26, IV-D-30, IV-D-50, IV-D-52,
IV-D-53, IV-F-1.16) stated that the analysis of FBC
performance in support of the proposed standards did not
recognize the technological differences between atmospheric
FBC and pressurized FBC, nor the variety of atmospheric FBC
technologies such as bubbling bed, circulating bed, and dual
bed.
Response: Pressurized FBC technology 1s still 1n the development
stages and 1s not considered a demonstrated technology for
the purpose of setting standards. As a result, 1n-depth
analyses were not conducted on pressurized FBC systems. As
discussed 1n the "Summary of Regulatory Analysis," however,
atmospheric FBC is considered a demonstrated SOg control
technology; thus, discussion of FBC is limited to
atmospheric FBC systems.
¦Differences 1n the design and performance of bubbling,
i circulating, and dual bed atmospheric FBC units are
! discussed in the "S02 Technology Update Report" and in
153
-------
"Fluidized Bed Combustion: Effectiveness as an SO2 Control
Technology for Industrial Boilers." Relative advantages and
disadvantages among these designs are also discussed.
Emission test data for five FBC units presented in the
"Summary of Regulatory Analysis" demonstrate that greater
than 90 percent SO2 removal can be achieved using FBC
technology. Although these data are based on bubbling bed
designs, equal or better performance is expected from
circulating and dual bed systems because of more rapid
carbon burnout, higher limestone particle densities in the
freeboard area, and more uniform gas-solid contact between
SO2 and limestone.
2.8.3 Other Technologies
1. Comment: Three commenters (IV-D-40, IV-D-52, IV-F-1.8) said that
although the use of low sulfur coal or physical coal
cleaning were determined to be "demonstrated technologies,"
1
they have been virtually eliminated by the percent reduction
requirement. The commenters felt that all demonstrated
technologies should be available for use in complying with
the standards.
Response: The use of low sulfur coal and physical coal cleaning to
control emissions of SOg was discussed in "Summary of
Regulatory Analysis." While these two techniques were
determined to be demonstrated methods of reducing S02
emissions, neither one was found to be capable of achieving
reductions equivalent to those achieved by SO2 control
systems considered the "best demonstrated technology" under
Section 111 of the Clean Air Act.
154
-------
It should also be noted that the Congressional Record for
the 1977 Clean Air Act amendments indicated that a percent
reduction should be "based" on a "best demonstrated
technology" capable of achieving emission reductions in the
range of 85 to 90 percent [H. R. Rep. No. 1175, 94th Cong.,
2nd Sess. 162 (1976)]. This report continued, "...use of
coal washing alone (which results in up to 40 percent sulfur
removal) would not constitute a suitable substitute, even
though the economic and energy impacts of mechanical coal
washing may be significantly lower than flue gas
desulfurization." Therefore, conventional coal cleaning
technologies clearly cannot serve as the "basis" of an NSPS.
Both low sulfur coal and physical coal cleaning, however,
are available under certain circumstances. The use of low
sulfur coal to meet an S02 emission limit is allowed for
steam generating units operating at low capacity utilization
factors. In addition, any reduction in potential SOg
emissions achieved through the use of physical coal cleaning
(or other fuel pretreatment methods) can be credited toward
the percent reduction requirement, thus reducing the SO2
removal efficiency required by the F6D system.
2.9 INDUSTRY-SPECIFIC ECONOMIC IMPACTS
1. Comment: Several coiranenters (IV-D-16, IV-D-19, IV-D-24, IV-D-25,
IV-F-1.2) said the anthracite mining industry in
Pennsylvania would be dealt a severe blow by a percentage
reduction requirement because a very low sulfur coal would
no longer be marketable. One (IV-D-16) added that the
elimination of anthracite as a viable fuel choice would
place a severe strain on the pension funds of over
7,000 former and 1,000 active anthracite miners. The
155
-------
commenter supported this by saying that the pension fund is
financed on the basis of the amount of anthracite produced
and sold.
Response: The final standards do not eliminate anthracite as a viable
fuel choice for industrial-commercial-institutional steam
generating units.
As discussed previously, an exemption from the percent
reduction requirement has been provided for steam generating
units operating at low capacity utilization rates for coal
or oil and meeting certain emission limits. Fuels with
relatively low sulfur contents, such as anthracite, may
therefore become more attractive for steam generating units
operating at low capacity factors, especially for those
located near anthracite deposits 1n the northeastern United
States.
2. Comment: Three commenters (IV-D-28, IV-D-96, IV-F-1.2) said no
thought has been given to the negative Impact of pending new
tax laws on capital investment. They said capital
investments are going to become even more dire than they
have been 1n recent years, further suppressing the already
sluggish steam generating unit market.
Response: The Tax Reform Act of 1986 repealed the investment tax
credit and modified the depredation periods and methods for
capital Investments. These Federal Income tax changes are
offset, to some extent, by the lower business income tax
rates that were instituted. Although repeal of the
investment tax credit does Increase the after-tax costs of
capital-Intensive projects such as the construction of a
steam generating unit, there 1s no reason to assume that any
156
-------
additional capital costs associated with compliance with the
NSPS, over the costs of the steam generating unit itself,
would cause a decision regarding whether or not to purchase
a steam generating unit to change. It could, however, have
an effect on the fuel selection for the new steam generating
unit. In those instances where the additional capital
investment required to install an FGD system is judged to be
significant, source operators would be more likely to fire
alternative fuels, such as natural gas, to avoid these
additional capital costs.
3. Comment: Several commenters (IV-D-26, IV-D-46, IV-D-51, IV-D-52,
IV-0-60, IV-0-73, IV-0-77, IV-0-88, IV-F-1.9) said the costs
associated with the regulation will be reflected in the cost
of goods, affecting the competitiveness of U.S. industry in
the world market. They asserted that marginal changes in
production costs can often push product prices beyond the
limits of market acceptance. Another commenter (IV-D-44)
was also concerned about the international competitiveness
of U.S. industry. The commenter said that there is a
substantial risk that oil and gas prices will skyrocket, and
if U.S. industry is dependent on those fuels as a result of
: the NSPS, it will be in a very poor position to compete in
international markets. According to the commenter, this
would negatively affect the trade deficit.
Response: As discussed in the "Summary of Regulatory Analysis," an
economic analysis was performed to assess the impacts of the
standards on individual plants, with an emphasis on
steam-intensive industries. The analysis was done assuming
that all facilities would be subject to a percent reduction
requirement, and thus represents a "worst case" assessment.
; This analysis estimated that product prices would increase
157
-------
by less than one percent, even assuming full cost
pass-through of the costs imposed by the standards to
product prices. In addition, the analysis concluded that
th6re would be negligible impacts on return on assets and
debt/equity ratios as a result of the standards. Steam
generating costs generally represent a small fraction of
total manufacturing expenses.
The standard does not force industrial steam users to become
dependent on any particular fuel. However, 1t is true that
some operators will select natural gas to avoid the costs
associated with the standard; therefore, there could be a
greater proportion of natural gas-f1red steam generating
units than would occur 1n the absence of the NSPS.
Projections of natural gas supply and price, however (see
Section 2.6.1), Indicate that natural gas supplies will
remain more than sufficient to meet the demand, and prices
are expected to Increase very gradually throughout the
remainder of this century. Therefore, the impacts of the
standard as 1t relates to national energy markets, and thus
to foreign competitiveness, are small.
4. Comment: Several commenters were concerned about the impact of the
standards on the coal mining industry. Some (IV-D-14,
IV-D-26, IV-D-36, IV-D-52) said the negative employment
consequences in the coal mining industry, as well as the
manufacturing and commercial sectors, far surpass the air
quality benefits of the proposed standards. They estimated
that the total coal mining ancillary jobs lost as a result
of this NSPS could amount to 36,000 persons. Others
(IV-D-40, IV-D-52, IV-D-83, IV-D-87) agreed, saying that
because the standard will affect the existing market share
of coal in the industrial fuel market as well as the new
158
-------
~ market share, not only new jobs will be affected. They
asserted that this impact has not been analyzed. The
commenters estimated that rather than a loss of 5 million
tons and 1,000 jobs, the actual displacement will be 40 to
50 million tons and 10,000 to 29,000 existing jobs. Another
commenter (IV-D-27) expressed concern that the economy of
those States in which coal production plays a major role
could experience significant adverse impacts as a result of
the standards with no significant benefit in terms of
improved air quality.
Response: No negative employment consequences in the coal mining
industry are expected to result from these standards. It is
anticipated that the recent decline in oil prices will
result in few coal-fired steam generating unit orders in the
near future, even in the absence of the NSPS. Because of
this, the revised national impacts analysis shows
essentially no projected impacts on coal use due to the
promulgated standard in the fifth year following proposal.
The levels of oil and natural gas prices in the future will
have a far greater Impact on coal use in new steam
generating units than any emission control standards.
The current industrial coal-fired steam generating unit
market (about 50 million tons/year) is only a small fraction
of total current coal production (over 800 million
tons/year). The amount of this coal market that could be
displaced as a result of the standard ranges from zero to
five million tons/year in 1990, a very small fraction of the
total coal market. In addition, total coal production is
expected to increase to meet the demand from new electric
utility power plants, more than offsetting any decline in
coal demand that might occur in the industrial sector.
159
-------
5. Consent: Two commenters (IV-D-51, IV-D-88) addressed the impacts of
the standards on small businesses. They said the standards
will affect small businesses, especially those which sell
oil for use in industrial steam generating units. They
further stated that standards will cause many steam
generating unit owners who fire oil to switch to natural
gas, causing the oil marketers to lose both present and
future customers. The commenters felt that an initial
Regulatory Flexibility Act analysis should be conducted to
address the impact of the proposed standards on small
businesses, particularly petroleum marketers who may be
greatly affected by fuel switching from oil to natural gas.
Response: The Regulatory Flexibility Act requires assessment of
impacts on "affected facilities" (in this case, the steam
generating units) operated by small businesses, rather than
on suppliers to the affected facilities. As a result,
formal review of the impact of the Regulatory Flexibility
Act on oil suppliers is not required.
While it is true that some steam generating unit operators
may elect to fire natural gas instead of oil in response to
the standards, this is expected to have little overall
impact on the total U.S. oil market. Oil consumption by
U.S. industry in 1986 was 8,340 PJ (7,900 TBtu). By
comparison, the quantity of fuel switching from oil to
natural gas projected by IFCAM is 120-330 PJ (110-310 TBtu)
per year, or roughly 1-4 percent of 1986 industrial oil use.
Further, because of factors not evaluated by IFCAM this
estimate of fuel switching may be overestimated. Therefore,
the NSPS will have little or no effect on most oil
distributors.
160
-------
2.10 SECONDARY ENVIRONMENTAL IMPACTS
2.10.1 Air
1. Comment:
Response:
Two commenters (IV-D-21, IV-F-1.4) felt that if district
steam systems are forced to shut down due to excessively
stringent standards, steam customers would have to produce
their own steam. According to the commenters, this would be
done largely with less efficient and less well-controlled
steam generating units that would not be subject to the
NSPS. Therefore, the commenters said, local emissions would
increase, a factor that was not considered in the impacts
analyses.
The standards would not require any existing facility to
shut down, since the standards apply only to new steam
generating units. If as a result of the standard there are
more new small steam generating units built and fewer large
district heating system steam generating units built, this
would not necessarily result in an increase in emissions.
The smaller the steam generating unit, the more attractive
natural gas and oil become over coal. Coal is an attractive
fuel only in large units. Consequently, smaller steam
generating units would be expected to fire natural gas or
premium oils (residual oils frequently require heating and
pumping systems for constant circulation and heating to
reduce viscosity). Premium oils also tend to have low
sulfur contents. As a result, the emissions from the small
steam generating units could well be less than the emissions
from the large coal-fired district heating system steam
generating units, even with controls applied. In any event,
standards of performance for small steam generating units
are under development and those standards will eliminate any
161
-------
Incentive created by these standards to replace large
district heating system steam generating units with numerous
small steam generating units.
2. Comment: Several comnenters addressed the analysis of ambient Impacts
presented 1n the proposal. Some (IV-D-26, 1V-D-30, IV-D-50,
IV-D-53, IV-F-1.14) said the analysis showed that the
proposed regulations may cause a deterioration 1n ambient
air quality. However, they said, not enough Information was
provided to duplicate the modeling analysis or to determine
a 3-hour or a 1-hour Impact. The comnenters also said there
were no modeling results for hilly terrains, which are
common 1n the eastern United States. One coramenter
(IV-D-81) noted that the Impact on ambient air quality may
be greater for a steam generating unit with an FGO system
than for one using low sulfur fuel. The comenter said that
a modeling demonstration 1n the "Sutmary of Regulatory
Analysis" Indicates that the maximum 24-hour downwind
concentration for a 44 MW (150 million Btu/hour) steam
generating unit would be higher with a 90 percent reduction
and an emission rate achieving 50 percent of the standard
than with a low sulfur fuel that just meets the standard.
Response: Modeling of ambient air quality Impacts 1n the "Summary of
Regulatory Analysis" was limited to two 150 million Btu/hour
stean generating units, one firing oil and one firing coal.
In both cases, percent reduction was based on "wet"
scrubbing of the highest sulfur fuel reasonably available.
Due to flue gas cooling with wet FGD and the resulting
decrease 1n plume buoyancy, use of wet FGD will have poorer
air dispersion characteristics than dry FGD systems.
Despite this "worst case" approach, of the four cases
examined (I.e., annual and 24-hour ambient concentrations
162
-------
from oil- and coal-fired units), ambient impacts were 20 to
30 percent lower with percent reduction in both of the oil
cases. In the two coal-fired cases, ambient impacts are
roughly equal (FGD ambient impacts were roughly 5 percent
lower on an annual basis, but 7 percent higher on a 24-hour
basis).
Given this uncertainty, there are probably some situations,
such as hilly terrain as mentioned by the commenter and
short averaging times, in which the use of wet FGD systems
could result in higher ambient air quality impacts than use
of low sulfur fuel. It is expected that where site-specific
modeling shows this to be the case, other air quality
programs, such as prevention of significant deterioration
and new source review (I.e., PSO and NSR), will ensure that
appropriate techniques are employed to maintain and protect
ambient air quality. Such techniques include dry FGD
processes and stack gas reheat, both of which result in
higher flue gas temperatures and greater flue gas
dispersion.
Finally, protection of air quality cannot be limited to
assessment of short-term, short-range dispersion of
pollutants. Of particular concern with SO2 emissions is
their impact on sulfate levels, which are of significance to
acid rain. While the magnitude of the acid rain problem and
the benefits of various solutions are still the subject of
research and debate, it is clear that sulfates are a
contributor. From the perspective of total sulfate
loadings, FGD technologies are more effective than use of
low sulfur coals and thus have benefits not captured in the
ambient air modeling cases discussed above.
163
-------
Each of the above factors, plus many others, were taken into
consideration prior to adoption of the final standards.
2.10.2 Water
1. Comment: Several commenters said the impacts of a standard "based" on
the use of FGD systems had not been evaluated in relation to
other regulations. Two commenters (IV-D-26, IV-D-78) said
the analyses did not address the Impact of sodium FGD system
waste on water quality regulations such as effluent
limitations guidelines for various industrial categories or
water quality standards which could Impose stringent
effluent limitations on wastewater treatment plants.
Another (IV-D-62) stated that no consideration has been
given to the fact that some States' water quality criteria
Include limits on osmotic pressure, dissolved solids, and
sodium sulfate because of the potential for causing harm to
irrigated crops. Several commenters (IV-D-26, IV-D-30,
IV-D-50, 1V-D-53, 1V-F-1.15) noted that California has
classified the effluent from sodium FGD systems as a
"designated" waste requiring special handling, such as
hauling to a secured disposal site. The commenters felt
that other States may follow suit on this lead, creating a
major waste disposal problem. Also, the commenters said,
the standards encourage complete oxidation of sulfites to
sulfates prior to discharge of sodium FGD system wastewater
streams. However, they noted that some municipalities have
adopted a sulfate limitation of 250 mg/1 for discharges into
treatment works and said this has not been considered in
developing the NSPS. Another commenter (IV-D-78) said that
the cost implications of FGD system wastewater treatment for
zero-discharge facilities were not adequately discussed.
164
-------
Response: The limits imposed by existing regulations on the disposal
of wastewater streams from sodium FGD systems were examined
by reviewing current disposal practices for these types of
wastes. Wastewater streams from sodium FGD systems are not
considered hazardous wastes, even under the most stringent
State or local regulations. In the West, disposal of these
types of wastes 1s generally by deep well Injection or
above-ground evaporation or percolation ponds. In the East,
disposal of these types of wastes is generally by direct
discharge to a receiving water body or indirect discharge
through a POTW. These streams are often treated prior to
discharge by oxidation, dilution, and/or removal of
suspended solids to comply with State or local effluent
limitations, water quality standards, or POTW pretreatment
standards. Thus, while pretreatment may be necessary 1n
some cases, these types of wastewater streams are currently
being disposed of by several methods in compliance with
State and local regulations.
The cost algorithm for sodium FGD systems included costs for
oxidation 1n order to reflect some form of pretreatment
prior to disposal. In some cases, however, even with
pretreatment of the wastewater streams, some forms of
disposal, such as direct or Indirect discharge, may not be
permitted. This could happen 1n areas of high Industrial
usage where the receiving water body and/or POTW has reached
Its maximum pollutant load. In addition, 1t is also
possible 1n some cases that disposal by deep well Injection,
evaporation ponds, or landfill containers may not be
permitted. This could happen 1n areas where concerns about
possible contamination of underground aquifers are
paramount, effectively prohibiting disposal of any liquid
wastes by such means.
165
-------
For this reason, as well as other reasons, the standard is
not "based" on the use of sodium FGD systems alone, nor for
that matter does the standard require the use of any
particular FGD system or S02 control technology. Rather,
the standard reflects the level of control that is
achievable through the use of any one of several
technologies, including sodium FGD systems, dual alkali FGD
systems, lime/limestone FGD systems, lime spray drying, and
fluidized bed combustion. In addition, where an individual
confronted with the standard may view the "burden" of using
these control technologies as excessive, fuel switching to
natural gas can be employed. In this manner, the necessity
of using these SOg control technologies can be avoided.
As discussed throughout the "Background Information
Document," the "SOg Control Technology Update Report," and
the "Summary of Regulatory Analysis," the environmental
(i.e., air, water, and solid waste), energy, cost, and
economic impacts associated with use of all of the
above-mentioned SOg control technologies, as well as those
associated with fuel switching, were reviewed and are
considered reasonable. Consequently, in areas where
disposal of wastewater streams from sodium FGD systems, or
for that matter wastewater streams from any "wet" FGD
system, is found to be very costly or essentially prohibited
by local regulation, steam generating units would be
expected to select an alternative approach to complying with
the standard. Such alternatives could range from the use of
"dry" scrubbing systems (i.e., lime spray drying or
fluidized bed combustion) to the use of natural gas.
2. Comment: One commenter (IV-D-48) expressed concern that for plants
producing only solid wastes, addition of a liquid waste
stream would represent a major shift 1n disposal
requirements.
166
-------
Response: As discussed above, the standard 1s not "based" solely on
"wet" SO2 control technologies (I.e., technologies which
produce wastewater streams requiring disposal). In
addition, the standard does not require the use of any
particular control technology. Consequently, a plant that
produces only solid wastes from Its production process would
most likely select an SOg control option that produces a
solid waste, such as Hme spray drying or fluidlzed bed
combustion, or, alternatively, would fire a fuel not
requiring the use of S02 control, such as natural gas.
Comment: Five commenters (IV-D-26, IV-D-30, IV-D-50, IV-D-53,
IV-F-1.16) stated that no data were presented on the
characteristics of wastewater from sodium F6D systems
serving coal-fired steam generating units. According to the
commenters, treatment of this wastewater could Include
neutralization, oxidation, metals precipitation, and sol Ids
sedimentation, and could be very costly. Moreover, the
commenters said, there may be discharge problems with even
this treated waste stream due to high total dissolved sol Ids
(TDS) content.
Response: The typical concentrations of trace metals in the wastewater
stream from a sodium FGD system on a coal-fired steam
generating unit were estimated based on a review of the
trace metal concentrations typically present 1n the
wastewater streams generated from other "wet" scrubbing
systems, such as lime/limestone scrubbing systems, operating
on coal-fired steam generating units, as well as
consideration of the known entities present 1n the coal and
the reagents used in the FGD system. This was discussed in
the "Summary of Regulatory Analysis" and a memorandum
entitled "Overview of the Sodium Wet Scrubbing Technology."
167
-------
Other characteristics, such as the dissolved sol Ids
concentration, were also estimated from review of the range
of values typically observed In the wastewater streams
discharged from sodium FGD systems operating on oil-fired
steam generating units and adjusted for various differences
between oil and coal, such as the typically higher S02 and
trace metal loadings 1n the flue gas from coal-fired units.
A wide range of concentrations was presented for each
element, 1n order to Include values for low and high sulfur
coal, and pulverized and spreader stoker type steam
generating units.
No data were presented by commenters, nor did any commenters
challenge the validity of these estimates by pointing out
errors that may have been made. Consequently, these
estimates of the characteristics of the wastewater streams
discharged from sodium FGD systems operating on coal-fv
steam generating units are considered valid.
4. Comment: Several commenters (IV-D-26, IV-D-30, IV-D-50, IV-D-53,
IV-D-73, IV-F-1.15) said the assumption that FGD system
water usage will be only a small fraction of the total plant
demands may be true for water-intensive Industries such as
Iron and steel or refineries, but could be significant for
other Industries. The commenters felt that these other
industries should be examined.
Response: Again, as mentioned above, the standards do not require the
use of "wet" FGD systems. In addition, the amount of water
required by wet FGD systems should not pose a problem 1n
most cases. However, there may be some situations (such as
location 1n an arid area or at a plant that has a low
overall water need) where this additional water usage would
168
-------
be significant. In such situations, alternative control
options could be selected (such as dry scrubbing, fluidized
bed combustion, or firing an alternative fuel such as
, natural gas) to minimize or eliminate this additional water
requirement. As discussed above, the costs associated with
I these alternative approaches to S02 control are considered
reasonable.
5. Comment: Two commenters (IV-D-26, IV-D-73) said the sodium salts
which are discharged to a wastewater treatment plant as part
of the effluent from a sodium FGD system Increase the total
Influent total dissolved sol Ids (TDS) concentration. This,
in turn, causes high effluent total suspended sol Ids (TSS)
concentrations and also Inhibits the ability to reduce
effluent TSS by the use of polyelectrolytes. They said that
higher TDS and temperature of the wastewater effluent can
also act to Inhibit the mass transfer of oxygen into water
and encourage oxidation of sulfite to sulfate, requiring
additional deaeration capacity 1n the wastewater treatment
plant.
Response: This may be true 1n some cases. In fact, as discussed
earlier, the wastewater streams from sodium FGD systems are
generally pretreated by dilution, oxidation, and/or removal
of suspended sol Ids prior to discharge. Thus, the cost
algorithm for sodium FGD systems Included costs for
oxidation to reflect additional costs for some type of
pretreatment prior to disposal.
As also discussed earlier, however, the standard does not
require the use of sodium FGD systems or, for that matter,
any specific SOg control technology. Thus, 1n cases where
the use of sodium FGD systems would result in deterioration
169
-------
of the overall quality of the wastewater effluent discharged
from a manufacturing plant to the point where additional and
costly increases to the wastewater treatment system would be
needed, other alternatives such as Hme spray drying,
fluidized bed combustion, or combustion of natural gas could
be selected.
6. Comment: Several commenters expressed concern about the effects of
wastewater effluent from a sodium FGD system on small
wastewater treatment plants. Two (IV-D-26, IV-D-73) said
for a small wastewater treatment plant near its hydraulic
capacity, the additional flow from a sodium FGD system may
be enough to overload the system. The consequence of this
overload, the commenters said, would be a reduction 1n
effluent quality. Others (IV-D-26, IV-D-30, IV-D-50,
IV-D-53, IV-F-1.15) noted that the conversion of sulfates
and sulfites 1n the wastewater stream from a sodium FGD
system to hydrogen sulfide can create an odor problem. They
said the analysis of this problem focused on large POTW's
where no problems were encountered. However, the commenters
felt that analysis of a small municipal treatment plant with
less flow should be conducted.
Response: Small POTW's or municipal wastewater treatment plants could
be unable to accept the wastewater stream from a sodium FGD
system 1f they are already operating near their maximum
pollutant load. Also, poorly designed or poorly operated
wastewater treatment plants, which fail to maintain aerobic
conditions throughout the treatment system, could give rise
to odor problems as a result of conversion of
sulfates/sulfites to hydrogen sulfide. Odor problems can be
mitigated by oxidizing the wastewater stream from a sodium
FGD system prior to discharge and by proper operation of the
170
-------
wastewater treatment plant to maintain aerobic conditions.
This 1s generally done at POTW's by injecting air Into the
sewer lines and ensuring that the sewer flow does not become
stagnant, and by aeration of ponds and lagoons. Odor
problems are generally not related to the size of the POTW,
but rather to the proper maintenance of aerobic conditions
in the system. Again, as mentioned above, the standards do
, not require the use of any particular control system and
where problems of this nature may exist, alternatives to the
use of sodium FGD systems could be selected.
2.10.3 Solid Waste
1. Comment: Several conmenters expressed concern about the availability
of adequate landfill capacity to dispose of FGD system
waste. Some (IV-D-26, IV-D-29, IV-D-30, IV-D-50, IV-D-52,
IV-D-58, IV-D-62, IV-D-66, IV-D-74, IV-D-84, IV-F-1.16)
stated that most FGD processes produce a waste sludge that
requires costly and space consuming disposal in increasingly
scarce landfill capacity. For example, they asserted, to
remove 600 Mg/year (660 tons/year) of SOg from high sulfur
coal, 7,250 Mg (8,000 tons) of solid waste would be
generated. The commenters felt that the focus on relative
rather than absolute Increases in waste volume 1s
inappropriate. Another (IV-D-44) said that although the
proposal claims that there is adequate capacity for
disposal of wastes generated by FGD systems, no study was
: cited to support this claim. The commenter said that
according to figures contained 1n the proposal document, the
use of FGD systems could generate approximately 15 times as
much waste as the use of low sulfur coal. The commenter
asserted that there is no evidence 1n the record of landfill
capacity sufficient to handle a fifteen-fold Increase 1n
waste.
171
-------
Another commenter (IV-D-78) added that State and local solid
waste disposal regulations are forcing the closure of many
existing landfills, making the siting of proximal landfills
difficult or Impossible, and driving up the cost of disposal
In available municipal landfills. The commenter said these
Impacts were not addressed 1n the proposal. Another
commenter (IV-D-62) said that until a means is found to site
landfills where they are needed without endless lawsuits and
other legal maneuvers Impeding the process, the Agency
should allow the use of "compliance fuels" having a sulfur
content low enough to meet the numerical emission limit
without the use of FGO systems.
Response: Steam generating units do not generally operate as
Independent entitles, but are most often part of an
Industrial plant which Itself produces wastes requiring
disposal. In addition, coal-fired steam generating units
generate fly ash which also requires disposal. Thus, use of
an FGD system to control S02 emissions generally does not
create a new problem (I.e., a need to dispose of wastes
where no such need existed before).
As discussed 1n the "Summary of Regulatory Analysis," the
use of dual alkali or Hme spray drying FGD generates less
than twice as much solid waste as the use of low sulf.
coal, rather than fifteen times as suggested by the
commenters. For example, the use of low sulfur coal would
generate about 3,200 Mg/yr (3,500 tons/yr) of solid waste
(steam generating unit blowdown and ash) from a typical 44
MW (150 million Btu/hour) steam generating unit. The use of
a dual alkali FGD system to achieve a 90 percent reduction
in SO2 emissions from a steam generating unit firing this
same coal would generate an additional 1,200 Mg/yr (1,300
tons/yr) of solid waste (FGD sludge).
172
-------
In most cases, disposal of wastes generated by FGD systems
presents no more of a problem than disposal of plant wastes
or steam generating unit fly ash. As a result, FGD system
wastes may generally be disposed of by the same means as
these wastes. In fact, since the wastes from some
Industrial plants are considered toxic or hazardous and
FGD system wastes are not, disposal of wastes from FGD
systems may present less of a problem than disposal of plant
wastes.
Consequently, 1n those specific locations where landfill
capacity may be limited, disposal of plant wastes is likely
to present as many problems -- and 1n some cases more
problems, given the nature of certain plant wastes -- as
disposal of wastes from FGD systems. For the plant, such
constraints may necessitate substantial changes to the
manufacturing process 1n order to minimize the wastes
generated or to alter their characteristics. For the steam
generating unit and SOg control system, this may necessitate
selection of one type of control system over another (I.e.,
a "dry" system over a "wet" system, for example) or
selection of an alternative fuel, such as natural gas, with
little or no waste disposal requirements.
2. Comment: Several conmenters (IV-D-26, IV-D-30, IV-D-50, IV-D-53) said
the potential effects of fabric filter bag failures on the
operation of a wet FGD system which is not designed for
particulate removal were not evaluated. They felt that the
consequent effect of the increase 1n particulate loading on
the FGD system operation and the FGD system effluent should
be evaluated.
173
-------
Response: A fabric filter bag failure should only be a temporary and
infrequent occurrence in a well operated and maintained
system. In addition, the typical baghouse is constructed
with multiple compartments, allowing one compartment to be
shut down (bypassed) for routine cleaning, maintenance, or
repairs without affecting the overall particulate matter
collection efficiency of the baghouse. Within each
compartment are multiple fabric filter bags, so that even if
one of these bags were to rupture or fail, the amount of
particulate matter passing through the compartment that
remained uncollected would not be significant and should not
appreciably affect either the F6D efficiency or the F6D
system effluent.
3. Comment: Several commenters (IV-D-10, IV-D-44, IV-D-82, IV-D-83)
asserted that higher percent reduction requirements increase
solid waste impacts due to additional absorbent use with no
discernible air quality benefits over lower percent
reduction requirements.
Response: Higher percent reduction requirements do result in
additional absorbent use and, therefore, larger quantities
of solid waste than do the lower percent reduction
requirements. However, the impacts associated with waste
disposal are unlikely to be significantly different whether
the percent reduction requirement is 90 percent, 70 percent,
or even 50 percent. The estimates of waste quantities
developed for this analysis were based on 92 percent S02
removal and, therefore, represent conservative estimates.
In addition, a 90 percent reduction requirement results in
significant emission reductions over a 50 or 70 percent
reduction requirement, and the impacts associated with 90
percent reduction (including waste disposal) are considered
reasonable.
17*
-------
4. Comment r"
One commenter (IV-D-48) claimed that FGO waste disposal 1s
not "easy" as was stated 1n the proposal. The commenter
cited an example 1n which a recently granted permit to
dispose of dry FGD waste in Minnesota required a leachate
collection system, an impermeable plastic Uner, a detailed
closure plan, pre- and post-operational monitoring plans,
and a performance bond. Others (IV-D-44, IV-D-58, IV-D-62,
IV-D-66, IV-D-74, IV-D-84) agreed, saying some States, such
as North Carolina and Minnesota, do not allow disposal of
FGD system waste 1n an unllned landfill, due to the risk of
groundwater contamination. They warned that this could
create a waste disposal problem.
Response: As discussed previously, steam generating units do not
generally operate independently, but as part of an
industrial plant which, even in the absence of the steam
generating unit, would produce wastes requiring disposal.
In addition, the steam generating unit Itself produces
fly ash that must be disposed of even in the absence of FGD
system wastes. The wastes produced by dry SOg control
techniques, such as Hme spray drying and fluldlzed bed
combustion, are not classified as hazardous and generally
can be disposed of 1n landfills without special handling.
In fact, the wastes produced as a result of the
manufacturing process may actually be more difficult to
dispose of than FGD system wastes, due to hazardous or toxic
materials present during the manufacturing process.
There may be specific locations, however, where strict State
or local waste disposal requirements exist. In these cases,
constraints on disposal of plant wastes would likely be as
stringent as, or even more stringent than, constraints on
FGD system waste disposal (depending on the nature of the
175
-------
plant waste). Consequently, 1n those specific locations
where landfill capacity may be limited, disposal of plant
wastes 1s likely to present as many problems -- and 1n some
cases more problems, given the nature of certain plant
wastes --as disposal of wastes from FGD systems. For the
plant, such constraints may necessitate substantial changes
to the manufacturing process 1n order to minimize the wastes
generated or to alter their characteristics. For the steam
generating unit and SOg control system, this may necessitate
selection of one type of control system over another (I.e.,
a "dry" system over a "wet" system, for example) or
selection of an alternative fuel, such as natural gas with
little or no waste disposal requirements. As mentioned
previously, the costs associated with the use of alternative
FGD systems or the use of alternative fuels, such as natural
gas, were examined and are considered reasonable.
5. Comment: Several commenters mentioned the uncertainty concerning the
definition of FGD system sludge under the Resource
Conservation and Recovery Act (RCRA). Some (IV-D-58,
IV-D-62, IV-D-66, IV-D-74, IV-0-84) said that although
FGD system sludge at present 1s specifically exempted from
the definition of hazardous waste under RCRA, that exemption
1s not necessarily permanent. They stated that 1f a study
which 1s currently being conducted Indicates that FGD system
sludge should be reclassified as hazardous, this could
significantly affect the NSPS analysis. Even 1f FGD system
sludge 1s not designated a hazardous waste, the commenters
said, the Agency is required under the new RCRA amendments
to develop Subtitle D Non-Hazardous Waste Regulations. The
commenters felt these regulations could significantly
Increase the requirements for disposal of FGD system sludge.
176
-------
One commenter (IV-D-66) said the analysis of the secondary
Impacts of the proposed standards leaves questions
unresolved: Would State agencies allow the disposal of FGO
system solid and sludge wastes with other nonhazardous
wastes? W111 developments In waste testing procedures
result 1n determining FGD system wastes to be hazardous?
Did the Agency test actual sludges for hazardous
constituents? If FGD system wastes are found to be
hazardous, would companies be required to establish
hazardous waste landfills?
Response: In developing emission standards, 1t 1s necessary to
consider the Impacts of these standards on existing
regulations, as well as the Impacts of existing regulations
on the standards. Sludge produced by FGD systems 1s
currently considered to be a nonhazardous waste under RCRA.
If further study Indicates that 1t should be reclassified as
hazardous, this will be taken Into account during the
regular NSPS review process 1n assessing the Impacts of
revising the standards.
6. Comment: Three commenters (IV-D-44, IV-D-82, IV-F-1.10) said the
Increase 1n FGD system sludge resulting from this regulation
1s counter to the principle of not blindly requiring the
shifting of pollutants between media to solve a pollution
problem. Another commenter (IV-D-83) said too little or no
attention has been given to the cross-media aspects of these
regulations. The commenter said that although the proposal
recognized that the standards will Increase the output of
solid and liquid wastes, 1t stated that this Increase would
be Insignificant compared to the large amount of waste
already generated by these facilities. According to the
commenter, this runs counter to the current trend that the
177
-------
generation of all solid wastes should be reduced, not
increased. The commenter contended that any regulation
which results in the generation of more wastes should be
subjected to stricter scrutiny than these appear to have
been.
Response: The secondary environmental impacts of the final standards
were examined, including impacts on ambient air quality,
water quality, and solid waste generation and disposal. All
regulations which control pollution in one area will result
in some "cross-media" pollution impacts in another; a
determination must be made as to whether these impacts are
"unreasonable" given the benefits of pollution reduction
associated with the standard. The impacts of these
standards on other environmental media were reviewed by the
Agency and are considered reasonable in light of the
reduction in emissions of SO2 achieved by the standards.
2.11 REGULATORY IMPACT ANALYSIS
1. Comment: One coiranenter (IV-D-64) stated that the Regulatory Impact
Analysis (RIA) fails to account for the full range of
benefits that would accrue from these emission reductions.
Specifically, the commenter said, the RIA acknowledges that
benefits due to reductions 1n acid deposition are Ignored,
as are the health benefits of expected reductions 1n sulfate
concentrations. The commenter suggested that fuller
elaboration of these two important items would provide an
even stronger justification for the proposed standards.
Response: The commenter 1s correct 1n stating that the RIA does not
present the full range of possible benefits from the
proposed standards. Although complete coverage is
178
-------
desirable, 1n this Instance complete quantification 1s not
needed to demonstrate the appropriateness of the standards.
2. Comment: One commenter (IV-D-76) asserted that there are some
important shortcomings in the RIA that have resulted in an
overstatement of benefits of SOg control. Specifically, the
commenter said:
the contingent valuation study used for visibility
estimates has serious shortcomings that result 1n an
upward bias in the benefit estimates for visibility;
the morbidity benefits estimate omits a key variable -
ozone - that results in Inclusion of ozone morbidity in
the PM/sulfates morbidity estimates, overstating the
morbidity by one-half to one-third.
Response: The decision to use the contingent valuation study by
Tolley, et al., to calculate the visibility benefits was
reexamined in light of the commenter's concerns. In
response to concerns that the Tolley study overestimated the
benefits of visibility Improvements, four other contingent
valuation studies were examined for comparison.
Comparison of Tolley with these four studies leads to the
following conclusions:
The benefits estimated using the Tolley study were not
significantly different from the benefits estimated
using the other four studies.
The use of Tolley, therefore, does not bias the
visibility benefits upward.
As a result of this comparison and to provide a rigorous
analysis, the RIA has been updated to include all five
contingent valuation studies 1n estimating visibility
benefits rather than relying solely on Tolley.
179
-------
The commenter also noted that the study used to calculate PM
morbidity (Ostro 1983) omitted ozone as a key variable. A
more recent study by Ostro (1986) has been reviewed and
replaces Ostro 1983 in the updated RIA.
The morbidity measure used in Ostro 1986 is limited to acute
respiratory conditions only and is therefore a better
measure of PM-induced acute morbidity than that used in his
earlier study. The 1986 study also uses fine particles
rather than total suspended particulate (TSP) as the measure
of exposure. Since the smaller (or fine) particles are the
most damaging to health, they are probably the most
appropriate indicator of PM exposure.
Due to these analytical improvements made in the Ostro 1986
study, the reported relationship between PM and acute
morbidity in adults is considered consistent, vigorous, and
reliable.
3. Comment: Several commenters (IV-D-26, IV-D-30, IV-D-50, IV-D-53,
IV-F-1.14) noted that in the RIA analysis of sulfate levels,
there are only four States where estimated sulfate levels
decrease from the base case to those achieved by a low
sulfur fuel standard, and none of those is decreased further
by imposing a percent reduction requirement. They further
pointed out that, of the five other States showing any
improvement by imposition of a percent reduction standard
over a low sulfur fuel standard, none shows more than a
3
0.1 ug/m improvement. The commenters felt that this cannot
be deemed significant. The commenters also contended that
similar results are seen in the analysis of visibility
improvements contained in the RIA.
180
-------
Response: In the draft RIA, Table III-3 (page III-7) shows that all 31
States analyzed under the low sulfur fuel alternative have
estimated 1995 sulfate concentrations that are from 0.1 to
3.9 percent lower than the 1995 baseline emissions.
State-by-State 1995 sulfate concentrations under the percent
reduction alternative (Table II1-4, page III-8) are 0.3 to
4.0 percent lower than the 1995 baseline. The apparent lack
of change in sulfate concentrations noted by the commenters
1s due to round off of actual numbers to two digits when
reported 1n the RIA. The draft RIA shows that the benefits
resulting from the proposed standard are Indeed significant
and of the same order of magnitude as the costs.
4. Comment: Four commenters (IV-D-26, IV-D-30, IV-D-50, IV-D-53)
asserted that the RIA shows that the actual costs of the
standard can be higher than the value of the purported
benefits.
Response: The draft RIA shows that the benefits and costs are of the
same order of magnitude. The draft RIA also states that
there are potentially significant benefit categories that
are not Included and other categories that are only
partially covered. For example, the SO2 benefit estimates
include only calculations for reductions 1n residential
materials damage. No calculations are Included for
commercial or Industrial facilities. Potentially
significant benefits are not Included from the nonhuman
biological effects category (e.g., acid deposition) and from
the S02 health categories. In addition, as discussed
earlier, the impacts associated with various standards were
reexamined under revised energy scenarios. This analysis
Indicates that the costs associated with the final standards
are considerably lower than those presented 1n the draft
RIA.
181
-------
5. Comment? One commenter (IV-D-96) said that the RIA should consider
the effect of the recent Federal tax reforms on
discretionary capital expenditures.
Response: Although the new Federal tax reforms have not been
Incorporated Into the RIA, the revised tax system should not
have a significant Impact on capital expenditures for
Industrial-commercial-Institutional steam generating units.
A survey of new steam generating unit projects conducted 1n
1986 Indicated that capital costs could Increase by as much
as 30 percent without significantly affecting the decision
to purchase a new steam generating unit. Although the new
tax revisions abolish the Investment tax credit, this credit
amounts to only 10 percent of the capital cost of the steam
generating unit and, therefore, the decision to buy a new
steam generating unit would not be affected.
6. Comment: One commenter (IV-D-96) felt that the baseline level of
emissions used 1n the benefit analysis was too high,
resulting 1n a significant overestlmatlon of the benefits of
the standards.
Response: The cost and benefit estimations are calculated from the
same baseline and, as a result, 1f benefits are erroneously
high then so are costs. The draft RIA presents benefits and
costs that are of the same order of magnitude.
Consequently, 1f the baseline emissions are reduced, then
benefits and costs will also be reduced but will remain In
the same order of magnitude. Also, as noted above, under
the revised energy price scenarios, the cost estimates are
greatly reduced, making the benefit-cost ratio associated
with the final standards even more positive.
182
-------
2.12 MIXED FUEL-FIRED STEAM GENERATING UNITS
1. Comment: Several commenters (IV-D-58, IV-D-62, IV-D-65, IV-D-66,
IV-D-74, IV-D-84) said there is no need to revise the
current NSPS for mixed fuel-fired steam generating units due
to the small amounts of SOg emitted.
Response: Based on an analysis of mixed fuel-fired steam generating
units, there is no justification for retaining the current
NSPS for these units. Significant reductions in SOg
emissions can be achieved by the revised standards, and the
cost, energy, and environmental impacts associated with the
revised standards are considered reasonable.
Mixed fuel-fired steam generating units are generally larger
than many other types of industrial steam generating units
and emit significant amounts of SOg as individual sources as
well as from the group as a whole (a projected 69,000 tons
year by 1990). For example, a typical 117 MW (400 million
Btu/hour) heat input capacity mixed fuel-fired steam
generating unit firing a mixture of 50 percent coal and 50
percent nonfossil fuel under the regulatory baseline would
emit nearly 2,000 Mg (2,100 tons) of SOg annually. Even
firing only 20 percent coal, this steam generating unit
would emit almost 800 Mg (900 tons) of SOg annually.
Therefore, mixed fuel-fired steam generating units are
considered significant sources of SOg emissions.
The costs of controlling SOg emissions from these steam
generating units were examined thoroughly and discussed in
"An Analysis of the Costs and Cost Effectiveness of S02
Control for Mixed Fuel-Fired Steam Generating Units." This
report analyzed the impacts associated with a number of
183
-------
regulatory alternatives ranging from standards based on an
emission limit only to standards requiring a percent
reduction in SO2 emissions from steam generating units
firing various mixtures of sulfur-bearing and
nonsulfur-bearing fuels. The impacts associated with
standards requiring a percent reduction in SOg emissions
were considered reasonable, with one exception. The cost
effectiveness of standards requiring a percent reduction in
emissions was considered unreasonable for mixed fuel-fired
steam generating units with annual capacity utilization
factors for coal or oil of 30 percent or less. Therefore,
this subcategory of mixed fuel-fired steam generating units
was granted an exemption from the percent reduction
requirement but must comply with an emission limit to
minimize SO2 emissions.
2. Comment: A number of commenters felt that emission credits should be
granted for mixed fuel-fired steam generating units. Two
commenters (IV-D-26, IV-D-73) said many industrial steam
generating units are burning more and more waste fuels which
would otherwise be destined for off-site disposal. They
felt this environmentally beneficial and cost effective
effort should be encouraged, not discouraged. Several
(IV-D-5, IV-D-26, IV-D-30, IV-D-50, IV-D-53, IV-D-58,
IV-D-62, IV-D-65, IV-D-66, IV-D-74, IV-D-78, IV-D-84,
IV-F-1.7) said the conclusion that there is no environmental
benefit associated with allowing emission credits for the
amount of nonfossil fuel fired in the steam generating unit
is incorrect. They further stated that the assumption that
steam generating unit operators increase the sulfur content
of the coal or oil burned as the amount of nonfossil fuel
burned in the unit increases is not borne out by operational
experience, nor would this usually be allowable under PSD or
184
-------
State regulations. Therefore, the commenters suggested that
the conclusion not to allow emission credits be reexamined.
If there is still concern on this matter, the commenters
said, some minimum amount of nonfossil fuel firing (e.g.,
5 percent) could be specified for a steam generating unit to
qualify as a combination unit.
Other commenters focused on units firing natural gas in
combination with coal or oil. Some (IV-D-11, IV-D-49,
IV-D-57, IV-F-1.5) felt that the standard does not give
natural gas equal treatment when used in combination with
coal or oil because it does not consider gas as an energy
input when calculating average emissions. They said this
could deter the use of new and existing gas-related
technologies such as co-firing, re-burn, or re-burn combined
with sorbent injection. The commenters suggested that the
standard be revised to allow any source that burns gas and
another fuel to receive emission reduction credit for the
amount of gas used. One commenter (IV-F-1.5) added that
natural gas should be allowed a greater opportunity to aid
reduction of S0£ as well as N0X emissions through co-firing
with coal and oil as well as being a direct substitute for
either. Others (IV-D-11, IV-D-15, IV-D-26, IV-D-28,
IV-D-73) felt the regulation should allow steam generating
unit operators to use any mixture of fuels to meet an
applicable emission limit and to calculate emissions based
on total energy input from all fuels. They said this would
greatly improve the cost effectiveness of compliance for
those sources using inherently clean fuels.
Response: Emission credits effectively negate any environmental
benefits, in terms of reduced S0£ emissions, associated with
the combustion of nonsulfur bearing fuels in mixed
185
-------
fuel-fired steam generating units. Emission credits would
permit S02 emissions from a mixed fuel-fired steam
generating unit to increase to the same level they would be
if the steam generating unit fired only oil or coal. A
mixed fuel-fired steam generating unit firing a 50/50
mixture of coal and wood waste, for example, would be
permitted to emit twice the S02 emissions with an emission
credit as it would without an emission credit.
The merits of emission credits for mixed fuel-fired steam
generating units were thoroughly examined and discussed in
"Summary of Regulatory Analysis," "An Analysis of the Costs
and Cost Effectiveness of S02 Control for Mixed Fuel-Fired
Steam Generating Units," and "Impacts of New Fuel Prices on
S02 Emission Credits for Cogeneration Systems and Mixed
Fuel-Fired Steam Generating Units." To assess the merits of
emission credits, the costs, S02 emissions, and cost
effectiveness of S02 control were analyzed with and without
an emission credit. This analysis shows that granting an
emission credit for mixed fuel-fired steam generating units
results in very small reductions in costs while allowing
significant increases in S02 emissions. Therefore, the
incremental cost effectiveness of the additional reduction
in S02 emissions achieved by not providing emission credits
for mixed fuel-fired steam generating units is low,
generally in the range of $220-330/Mg ($200-300/ton). These
costs are considered reasonable in view of the significant
additional emission reductions achieved by not providing
emission credits. Consequently, the standards do not
include provisions for emission credits for mixed fuel-fired
steam generating units.
186
-------
The absence of provisions for emission credits should not,
however, discourage the use of on-site wastes as steam
generating unit fuels. In most cases, the disposal of such
wastes in this manner would represent the most cost
effective method of disposal regardless of the NSPS. In
addition, the final standards include an exemption from the
percent reduction requirement for steam generating units
firing oil or coal at 30 percent or less of their rated
annual heat input capacity. This provides substantial
incentive to fire on-site or off-site wastes in order to
reduce the amount of oil or coal burned in the steam
generating unit.
3. Comment: Two commenters (IV-D-5, IV-D-58) felt that the emission
baselines used for mixed fuel-fired steam generating units
were unrealistically high and did not reflect current
emission levels. Specifically, they said that the use of a
baseline emission level of 1,076 ng/J (2.5 lb/million Btu)
for mixed fuel-fired units firing coal is unrealistic.
According to the commenters, units larger than 73 MW
(250 million Btu/hour) are already subject to a 516 ng/J
(1.2 lb/million Btu) emission limit, and it is likely that
smaller units would be subject to similar limits due to
PSD/BACT limitations or State regulations. Therefore, the
commenters said, a baseline of 516 ng/J (1.2 lb/million Btu)
should be used and all cost effectiveness and emission
calculations should be revised accordingly. The commenters
also felt that a baseline of 1,290 ng/J (3.0 lb/million Btu)
for units firing mixtures of oil and wood is unrealistic,
and a more logical baseline would be 344 ng/J (0.8
lb/million Btu).
187
-------
Response: The S0£ emission limits of 1,076 ng/J (2.5 lb/million Btu)
for coal-fired steam generating units and 1,290 ng/J (3.0
lb/million Btu) for oil-fired steam generating units
represent the "average" emission limits currently required
under SIP's for units with heat input capacities below 73 MW
(250 million Btu/hour). Steam generating units larger than
this size are, as noted by the commenters, subject to the
Subpart D emission limits of 516 ng/J (1.2 lb/million Btu)
for coal and 344 ng/J (0.8 lb/million Btu) for oil. The
baselines for these larger units have been revised to
reflect these levels in subsequent analyses.
4. Comment: One commenter (IV-D-79) said the final standards should
clarify that only S0£ emissions resulting from combustion of
regulated fuels are subject to the standard. The commenter
added that emissions of S0£ that result from the combustion
of other fuels, as in carbon monoxide boilers, should not be
covered by the standard.
Response: The promulgated standard applies only to S0£ emissions
resulting from the combustion of coal and/or oil.
Consequently, if a sulfur-bearing fuel other than coal or
oil is combusted in combination with coal or oil, and if the
contribution to emissions from that fuel can be separated
from the contribution from coal or oil, those emissions
would not be subject to the standards. The final standards
have been clarified with regard to this point.
5. Comment: Several commenters (IV-D-5, IV-D-58, IV-D-62, IV-D-65,
IV-D-66, IV-D-74, IV-D-84) requested that the source of
"historical data" used to geographically distribute the
projected 35 new mixed fuel-fired steam generating units be
identified. They questioned why all of these units were
188
-------
placed in Regions I, IV and X when there are numerous mixed
fuel-fired steam generating units in other regions. For
example, the commenters said, of the 35 mixed fuel-fired
units installed from 1980-1984, 21 were in the South. They
felt that the distribution used in the analysis gives undue
emphasis to Regions I and X, and should be revised to give
more weight to Regions IV and VI where the compliance costs
are much higher.
Response: Regions I, IV, and X were selected for the purpose of
analyzing the potential impacts of standards on new mixed
fuel-fired steam generating units. Historically, these
regions have had the highest concentration of the pulp and
paper and forest products industries, and most of the
existing mixed fuel-fired steam generating unit population
is located in these regions. The information used to
geographically situate the mixed fuel-fired steam generating
unit population was discussed in "Nonfossil Fuel Fired
Industrial Boilers - Background Information." This
information was obtained from various directories of the
pulp and paper and forest products industries.
In determining which regions to use for purpose of analysis,
a number of factors were considered. While it is true that
mixed fuel-fired steam generating units are located
throughout the U. S., most new mixed fuel-fired steam
generating units were expected to be sited in the northeast
and southeast and on the west coast. Based on
this determination, Region I (with 6 boilers) was selected
to represent the northeast, Region IV (with 16 boilers) was
selected to represent the southeast, and Region X (with
13 boilers) was selected to represent the west coast.
Subsequent data collected as part of the boiler replacement
189
-------
survey identified 30 mixed fuel-fired boilers built between
1981 and 1984. Of these, 21 were designed to burn a
combination of coal and wood, while 8 were capable of also
firing oil or natural gas. By geographic area, the 30 units
were distributed as follows: 16 were located in the
southeast, 8 were on the west coast, 3 were in the upper
midwest, 2 were in the northeast, and 1 was in the
southwest. Of the 16 in the southeast, 12 were in
Region IV, 3 were in Region III, and 1 was in Region VI.
Based on this review of regional data and comparison of
regional fuel price estimates, it was concluded that the
original placement of boilers and the analysis of cost
impacts for compliance with a percent reduction standard
were valid.
6. Comment: Several commenters (IV-D-5, IV-D-26, IV-D-30, IV-D-50,
IV-D-53, IV-D-58, IV-D-62, IV-D-66, IV-D-74, IV-D-84) said
the projection that 35 new mixed fuel-fired steam generating
g
units with a total heat input capacity of 19x10 Btu/hour
and firing wood with coal, oil, or natural gas will be
installed in the next 5 years is not in line with the
current pace of installation. They suggested that a better
estimate would appear to be 15 new units with a heat input
g
capacity of 10x10 Btu/hour, or about one-half of this
projection.
Response: The purpose of the analysis was to assess the potential
impacts associated with alternative standards on mixed
fuel-fired steam generating units firing mixtures of
nonsulfur-bearing fuels with coal or oil. Thus, the
analysis includes mixed fuel-fired steam generating units
firing other fuels, such as municipal solid waste, for
example, as well as wood. Therefore, while the projection
190
-------
of 35 new mixed fuel-fired steam generating units may not be
exact if only units firing wood with coal or oil are
considered, the addition of units firing municipal solid
waste, combined with expected growth in this source
category, resulted in this figure being selected for
purposes of analysis.
7. Comment: Two commenters (IV-D-5, IV-D-58) said the selection of coal
type for meeting a 516 ng/J (1.2 lb/million Btu) standard is
not correct in Table 16 of the mixed fuel-fired steam
generating unit report for the 117 MW (400 million Btu/hour)
unit cases. They asserted that this error affects the
annualized costs and emissions, and thus the cost
effectiveness numbers, and should be revised. In addition,
the commenters said the selection of oil type in Table 18
for units firing an 80 percent oil/20 percent wood mixture
and meeting a 344 ng/J (0.8 lb/million Btu) emission limit
appears to be incorrect.
Response: The costs cited in Table 16 of the subject report are based
on the coals specified in Table 15. The commenters are
correct in pointing out that some of the coals selected for
meeting a 516 ng/J (1.2 lb/million Btu) standard are in
error. This applies to the 50 percent coal/50 percent
nonsulfur-bearing fuel mixtures in Regions I and IV and the
80 percent coal/20 percent nonsulfur-bearing fuel mixtures
in Regions I, IV, and X. In the case of the 344 ng/J (0.8
lb/million Btu) standard for mixed fuel-fired units firing
oil, the oil type specified for the 80 percent oil/20
percent nonsulfur-bearing fuel was also in error.
191
-------
Each of these oversights has been addressed and corrected in
a subsequent analysis performed to evaluate the impacts of
new fuel prices on the previous analysis. The revised
analyses are discussed in the memoranda entitled "Impact of
New Fuel Prices on the Cost and Cost Effectiveness of SO2
Emission Control of Mixed Fuel-Fired Steam Generating Units"
and "Impact of New Fuel Prices on the National Impacts
Analysis for Mixed Fuel-fired Steam Generating Units."
8. Comment: Several commenters (IV-D-26, IV-D-30, IV-D-50, IV-D-53,
IV-D-58, IV-D-62, IV-D-65, IV-D-66, IV-D-74, IV-D-84,
IV-F-1.7, IV-F-1.14) stated that the impacts of a percent
reduction requirement on mixed fuel-fired steam generating
units should be further analyzed. They noted that
combination units have advantages such as lower SO2
emissions, conservation of fossil fuels, and disposal of
waste materials that should not be discouraged. Also, they
said, mixed fuel-fired steam generating units will account
for, at most, only 0.06 million Mg (0.07 million tons)
SC^/year of a nationwide total of 23 million Mg (25 million
tons) S02/year. They expressed concern that imposing a
percentage reduction on these units could provide a
disincentive to some companies considering installing them.
Response: The impacts of alternative SO2 control options on mixed
fuel-fired steam generating units were reanalyzed in light
of recent declines in oil and gas prices. New fuel prices
were developed for coal, oil, and natural gas and used to
assess the impacts of the standards on mixed fuel-fired
steam generating units. The impacts associated with this
updated analysis are generally lower than those in the
original analysis, and are considered reasonable. The
requirements associated with the final standards are not
192
-------
expected to discourage future installations of mixed
fuel-fired steam generating units. The costs associated
with SOg control are only a small proportion of total steam
generating unit costs, and the advantage of firing wastes
generated on-site as fuel will generally help offset any
increased costs associated with SO2 control. In addition,
as discussed above, the exemption from the percent reduction
requirement included in the final standards for steam
generating units firing coal or oil at 30 percent or less of
their total annual capacity will serve as an incentive for
these units to fire more waste fuels and less coal or oil.
9. Comment: One commenter (IV-D-20) said that the standard listed in 40
CFR 60.45b(f) appears to apply to the combustion of oil
"only" and does not clarify whether it applies to oil fired
in combination with other fuels. Units fired with oil in
combination with natural gas, waste wood, etc., should be
provided a similar exemption as that available for coal
firing under 60.42b(c). Another commenter (IV-D-92)
requested clarification on 60.42b(c)(1). The commenter
asked whether subparagraph (ii) applies only to units that
burn coal in combination with other fuels or whether it may
also apply to situations in which oil or natural gas is
burned in combination with other fuels such as municipal
solid waste where such waste constitutes 10 percent or more
of the annual capacity utilized. The commenter said if
60.42b(c)(1) does not apply to other situations where oil or
gas is co-fired with other nonfossil fuels, why is an
exemption reserved for coal since coal emits more SOg than
oil or natural gas?
Response: As discussed earlier, the final standards provide an
exemption from the percent reduction requirement for steam
generating units which obtain 30 percent or less of their
193
-------
annual maximum heat input capacity from the combustion of
oil or coal (or a combination of oil and coal). This
provision applies regardless of whether the coal or oil is
fired alone or in combination with natural gas or nonfossil
fuels.
2.13 STANDARD FOR COGENERATION UNITS
1. Comment: Several commenters expressed the opinion that emission
credits should be allowed for cogeneration systems. Some
(IV-D-26, IV-D-38, IV-D-60, IV-D-73, IV-D-79, IV-D-84)
argued that a cogeneration system which produces both steam
and electricity from one heat source will result in lower
emissions than separate units for each purpose. They also
felt that if full credit for this difference is not
considered cost effective, the Agency should at least look
at the cost effectiveness of some fraction of the difference
in emissions being applied as a credit.
Another commenter (IV-D-28) said the analysis of
cogeneration contained in the proposal ignored the fact that
processes which are more efficient use less of natural
resources and emit less waste to the environment. The
commenter felt that these benefits should be encouraged
through the allowance of credits. Also, the commenter said,
the marginal cost effectiveness for the additional S02
reduction that results from denying these credits to
cogeneration units ranges from $330-$550/Mg ($300-$500/ton)
for FGD-based systems and from $l,650-$3,300/Mg
($1,500-$3,000/ton) for compliance fuel-based systems. The
commenter said credits also encourage energy efficiency and
offer nonSC^ environmental benefits by requiring less
194
-------
resource extraction, transportation, etc., and stated that
these benefits and improved cost effectiveness should be
encouraged.
Response: As discussed above for mixed fuel-fired steam generating
units, emission credits would permit S02 emissions
from a cogeneration steam generating unit to increase to the
level that would have existed if a conventional, rather than
a cogeneration, steam generating unit had been installed.
As a result, emission credits effectively reduce the
environmental benefits, in terms of reduced S02 emissions,
associated with cogeneration.
The merits of emission credits for cogeneration systems were
thoroughly examined and discussed in the "Summary of
Regulatory Analysis," "An Analysis of the Costs and Cost
Effectiveness of Allowing SO2 Emission Credits for
Cogeneration Systems," and "Impact of New Fuel Prices on S02
Emission Credits for Cogeneration Systems and Mixed
Fuel-Fired Steam Generating Units." To assess the merits of
emission credits, the costs, S02 emissions, and cost
effectiveness of S02 control were analyzed with and without
an emission credit. This analysis concluded that granting
an emission credit for cogeneration steam generating units
results 1n very small reductions 1n costs while allowing
significant Increases 1n S02 emissions. Therefore, the
Incremental cost effectiveness of the additional emission
reductions achieved by not providing emission credits for
cogeneration steam generating units 1s low, generally in the
range of $220-$660/Mg ($200-$600/ton). Similar cost
effectiveness values exist relative to not providing partial
emission credits. These costs are considered reasonable in
view of the significant additional emission reductions
195
-------
achieved by not providing emission credits. Consequently,
the final standards do not include provisions for emission
credits for cogeneration systems.
The absence of emission credits for cogeneration systems is
not expected to discourage their use. The non-SOg
environmental benefits listed by the commenters, such as
lower fuel costs resulting from greater energy efficiency,
should still provide incentive for the use of this
technology.
2. Comment: One commenter (IV-D-44) said that because they often do not
produce other plant wastes, FGD system waste disposal costs
could be a significant burden for cogeneration systems.
Response: Even those cogeneration systems which are not part of a
manufacturing plant that produces plant wastes (such as
those used solely for "third party" generation of steam and
electricity for sale) will produce some waste from operation
of the steam generating unit itself. For example, firing
coal produces an ash that must be disposed of regardless of
other regulatory requirements. In addition, the steam
generating unit will produce waste from steam generating
unit blowdown and other routine steam generating unit
operation and maintenance procedures.
As discussed previously (see Section 2.10.1), the standard
is not "based" on the use of any particular SOg control
technology. Therefore, where disposal of waste associated
with a "wet" FGD system, for example, would be burdensome,
other technologies (such as lime spray drying or fluidized
bed combustion) could be selected to change the nature of
the waste produced. In addition, where an individual
196
-------
confronted with the standards may view the burden associated
with disposal of any wastes as excessive, the use of
alternative fuels, such as natural gas, would substantially
reduce or eliminate the need for waste disposal.
3. Comment: One commenter (IV-D-44) asserted that if the proposed
standards are promulgated, the economics of cogeneration
projects will be severely affected. The commenter stated
that because the prices for steam and electricity are
determined by factors that will not be significantly
affected by the regulation, the additional operating costs
incurred cannot be "passed through" to the steam purchaser.
Therefore, the coiranenter said, the added costs of complying
with a percent reduction requirement would significantly
reduce the number of viable cogeneration projects. The
commenter requested that cogeneration facilities be exempted
from the standards or less stringent standards adopted for
these facilities.
Response:' The final standards will increase the project costs of coal-
and oil-fired cogeneration steam generating units. This
could preclude some smaller, less profitable cogeneration
projects. However, most coal- and oil-fired cogeneration
steam generating units covered by the standards will not be
severely affected. For example, a cost analysis of the
impact of standards on a typical coal-fired cogeneration
steam generating unit showed that a percent reduction
requirement increased the total annualized steam generating
unit costs by about 11 percent. This can be compared to the
results of the "Survey of New Industrial Boiler Projects,"
which indicate that more than 80 percent of the cogeneration
projects completed in 1981 through 1984 would have gone
forward as designed even if costs increased by 10 percent.
197
-------
In addition, more than 60 percent of the cogeneration
projects would have gone forward as designed if costs
increased by as much as 20 percent.
To put the impact of the standards on cogeneration projects
into perspective, however, it must be realized that not all
cogeneration projects will be affected by the final
standards. In fact, given current fuel prices, most new
cogeneration projects are expected to be based on firing
natural gas (both natural gas-fired turbines and natural
gas-fired boilers with steam turbines) or municipal solid
waste. Cogeneration projects firing these fuels are not
affected in any way by the S02 standards. Thus, the
standards will have no impact on most new cogeneration
projects.
Finally, another factor to consider is that new coal-fired
cogeneration steam generating units compete with new
coal-fired electric utility stean generating units, which
are subject to percent reduction requirements. Application
of percent reduction to coal-fired cogeneration steam
generating units, therefore, is a "neutral" position which
treats both utility steam generating units and
industrial-commercial-institutional cogeneration steam
generating units in a similar manner. Consequently, the
final standards include no special provisions for
cogeneration steam generating units. They are treated just
like any other industrial-commercial-institutional steam
generating unit.
4. Comment: Two commenters (IV-D-44, IV-F-1.17) said the standards would
make many cogeneration projects uneconomical by increasing
the capital costs of a standard 55 MW cogeneration facility
198
-------
by about $5.3 million. In addition, the commenter said that
operating costs, disposal costs, and costs due to decreased
steam generating unit availability would be significantly
higher than those estimated in the proposal. Increased
annual costs of a 90 percent reduction requirement were
estimated by the commenter at $3.7 million. The cost
effectiveness of SO2 removal would be over $1,760/Mg
($1,600/ton) compared to the proposal estimates of $430 to
$475/Mg ($390 to $430/ton). This, the commenters said,
would put builders of cogeneration systems out of business.
Response: The Agency reviewed its cost estimates for compliance with a
percent reduction standard on a 55 MW cogeneration facility
and found them to be consistent with those cited by the
commenter. Similarly, the Aency's estimates of cost
effectiveness of percent reduction are similar to those
cited by the commenter. Further, these costs are similar to
those imposed on a conventional coal- or oil-fired steam
generating unit of the same size and are not believed to
impose an unreasonable burden on cogeneration facilities.
The commenter's citation of $430 to $475/Mg ($390 to
$430/ton) is a national average, and is not intended to
imply that higher cost effectiveness values for individual
units are reasonable or unreasonable.
In summary, the impact of the final standards on
cogeneration facilities was examined in some detail and is
considered consistent and reasonable for cogeneration, we
well as conventional, steam generating units. Also, as
noted in the previous response, the final standards for
industrial-commercial-institutional cogeneration systems are
consistent with those for electric utility steam generating
: units.
199
-------
5. Comment: One commenter (IV-D-44) said that in conformance with
national energy policy goals, the Powerplant and Industrial
Fuel Use Act of 1978 stated that cogeneration facilities
must be constructed with "the capability to use coal or any
other alternate fuel as a primary energy source." Thus, the
commenter said, it is not reasonable to assume that these
facilities will be able to switch to natural gas. The
commenter claimed that the effect of EPA's fuel switching
assumption is to underestimate the costs of compliance for
cogeneration systems.
Response: Regulations under the Fuel Use Act, which prohibited the use
of natural gas in most new industrial-commercial -
institutional steam generating units, were repealed by
P.L. 100-42, signed on May 22, 1987. Under the new law,
conventional steam generating units and cogeneration systems
which sell less than 50 percent of the total electrical
production to a utility are exempted from the Fuel Use Act's
natural gas use restrictions. Cogeneration units operating
in excess of 3500 hours per year and selling in excess of
50 percent of their electrical output to a utility must be
designed with the capability to burn coal or coal-derived
fuels, but can obtain an exemption to burn natural gas if
using coal would impose an economic hardship or penalty.
Because a facility switching from coal to natural gas in
response to the standards would be doing so to avoid the
higher costs associated with firing coal, obtaining such an
exemption should be straightforward.
2.14 STANDARD FOR EMERGING TECHNOLOGIES
1. Comment: A number of commenters (IV-D-23, IV-D-26, IV-D-30, IV-D-40,
IV-D-43, IV-D-50, IV-D-52, IV-D-53, IV-D-58, IV-D-62,
IV-D-66, IV-D-73, IV-D-74, IV-D-75, IV-D-81, IV-D-84,
200
-------
IV-D-85, IV-F-1.1, IV-F-1.8, IV-F-1.16) said fluidized bed
boilers are only beginning to show promise for low cost
pollution control in some energy intensive industries. The
commenters felt that this technology should be considered an
emerging, rather than demonstrated, technology.
Response: As discussed previously (see Section 2.8.2), fluidized bed
combustion is considered a demonstrated technology capable
of achieving 90 percent reduction on a continuous basis at
high reliability levels.
2. Comment: One commenter (IV-D-43) said that fluidized bed steam
generating units firing culm (anthracite mining waste) and
gob (bituminous coal mining waste) should be considered an
emerging technology. According to the commenter, this
technology is different from that used to burn conventional
fuels.
Response: As discussed above, fluidized bed combustion is considered a
demonstrated, rather than emerging, technology. As also
discussed earlier, however, concerns remain about the
economic viability of using this technology to achieve a 90
percent reduction in SOg emissions when firing culm or other
types of coal refuse. As a result, a less stringent percent
reduction requirement has been established for FBC units
firing coal refuse.
3. Comment: One commenter (IV-D-40) said that, by definition, an
emerging technology has not been "adequately demonstrated."
Therefore, the commenter said, a standard for emerging
technology cannot be set under Section 111. Others (IV-D-6,
IV-D-7, IV-D-28, IV-D-38, IV-D-40, IV-D-50, IV-D-52,
IV-D-75) agreed, saying new technologies lack sufficient
201
-------
operating data upon which to base NSPS. Several commenters
(IV-D-6, IV-D-26, IV-D-28, IV-D-40, IV-D-50, IV-D-52,
IV-D-72) added that Federal standards regulating emissions
from steam generating units using emerging technologies
should not be proposed until valid test data from a
representative group of such installations across U.S.
industry have been tabulated and reviewed to determine the
appropriate emission levels for such technologies. They
said premature proposal of standards would inhibit and
retard development and commercial demonstration of these
technologies. Another commenter (IV-D-99) stated that the
emerging technology provision is beneficial only to
technologies having substantial near-term promise of
exceeding 90 percent SOg control at a relatively low cost.
This, the commenter said, would terminate development and
commercialization of many technologies capable of
substantial, cost effective SOg control, but less than 90
percent reduction. One commenter (IV-D-76) said no
quantitative or cost effectiveness analysis was provided for
the proposal to allow emerging technologies to achieve a
less stringent 50 percent reduction requirement.
Response: In the absence of provisions for emerging technologies, all
sources would have to achieve performance levels that
conform with best technical system of continuous emission
reduction. The commenters do not dispute that EPA has
authority to make provisions designed to encourage the
emergence of new technology. That being the case, there can
be no question that EPA has the authority to define
parameters conducive to that goal.
202
-------
As discussed at proposal and also mentioned by the
commenters, however, standards requiring a high level of
performance, such as 90 percent reduction, could act to
discourage continued development of some new technologies.
Owners and operators of new steam generating units could
simply view the risks of using a new and untried emission
control technology to achieve a 90 percent reduction in
emissions as too great. Thus, to encourage the continued
development of emission control technologies that show
promise of achieving levels of performance comparable to
those of existing technologies, special provisions are
included in the standards to accommodate and, it is hoped,
foster the continued development of new technologies.
These provisions do not reflect the perceived performance
capability of any specific new or emerging technologies. As
discussed at proposal, these provisions reflect a reasoned
judgment of an appropriate balance between a level of
performance which is low enough to significantly reduce the
risks associated with use of a new technology, but also high
enough to ensure that with continued development the
technology appears to have the potential to achieve
performance levels comparable to those achieved by existing
technologies, as well as ensuring that the increased
emissions resulting from use of the new technology are
minimized.
A standard requiring a 50 percent reduction in SO2
emissions, but also limiting emissions to 172 and 258 ng/J
(0.4 and 0.6 lb/million Btu) heat input for oil and coal,
respectively, appears to strike this balance. A minimal
percent reduction requirement of 50 percent should
effectively eliminate the risk of failure for any technology
203
-------
which has the potential to reduce emissions by 90 percent.
It is difficult to conceive of a new control technology that
would be incapable of achieving at least a 50 percent
reduction in SOg emissions during development, and still
have potential for achieving 90 percent reduction when
development is completed.
When combined with the emission limits stated above, this
standard minimizes SC^ emissions by limiting these emissions
to 50 percent or less of those resulting from the use of low
sulfur coal or oil. The emission limits will encourage the
further development of these emerging technologies by
requiring higher percent reductions when firing higher
sulfur fuels. Thus, the final standards retain the special
provisions, as they were proposed, for emerging
technologies.
4. Comment: A number of commenters (IV-D-23, IV-D-26, IV-D-28, IV-D-30,
IV-D-38, IV-D-45, IV-D-50, IV-D-51, IV-D-52, IV-D-53,
IV-D-58, IV-D-62, IV-D-66, IV-D-73, IV-D-74, IV-D-75,
IV-D-80, IV-D-84, IV-D-85, IV-D-88, IV-F-1.16) expressed the
opinion that requiring emerging technologies to meet a
maximum emission cap of 258 ng/J (0.6 lb/million Btu) is
overly restrictive, arbitrary and will operate as a
disincentive for industries wishing to pursue the
installation of promising but unproven technology. They
said the use of emerging technology should not be
discouraged by regulations or otherwise. The commenters
asserted that this emission cap should be raised to 516 ng/J
(1.2 lb/million Btu) for emerging technologies. This, they
said, is consistent with the cap for adequately demonstrated
technology and would allow affected sources to burn a wider
range of coals, thereby generating a better data base for
204
-------
future decision making. One commenter (IV-D-52) added that
the requirement that facilities using emerging technologies
meet an emission limit of 258 ng/J (0.6 lb/million Btu) heat
input is unnecessary as a safeguard against the failure of
the new technology to meet the required degree of reduction.
The commenter said any potential problems with the emerging
technologies can be addressed and resolved through the use
of the current permitting procedures implemented by the
States.
Response: As discussed above and in the preamble to the proposed
standards, limiting the emerging technology provision to the
use of low sulfur fuels will minimize any increase in SO2
emissions as a result of the use of an emerging technology.
For coal-fired steam generating units, an emission limit of
258 ng/J (0.6 lb/million Btu) essentially requires the use
of coal with potential SO2 emissions of 516 ng/J
(1.2 lb/million Btu) or less if the minimum 50 percent
reduction is being achieved. However, to the extent that an
emerging technology achieves emission reductions greater
than 50 percent, coal with higher sulfur contents can be
used to comply with the emission limit. Therefore, this
"emission cap" provides a further incentive for the
continued development of emerging technologies. The purpose
of this emission limit is not to provide a "safeguard"
against the failure of emerging technologies to achieve a
50 percent reduction in emissions, but rather to minimize
total emissions from technologies achieving 50 percent
reduction. Any technology subject to the emerging
technology provisions must achieve at least a 50 percent
reduction in emissions, or it will be subject to the same
enforcement requirements and/or penalties as conventional
demonstrated technologies.
205
-------
5. Comment: One commenter (IV-D-64) said the proposed standards for
emerging technologies are too lenient, because steam
generating units using "emerging technologies" would be
allowed to emit more than facilities using conventional
technologies due to the lower percent reduction requirement.
The commenter said the maximum emission limit should be no
greater than the equivalent of a 90 percent reduction from
the average sulfur content of coal used in industrial steam
generating units.
Response: It is possible that sources subject to the emerging
technology provisions will have higher SOg emission rates
than sources using conventional demonstrated SOg control
technologies. The emission limits established for emerging
technologies serve to minimize this difference in emissions
by essentially restricting the use of these technologies to
sources firing low sulfur fuels. The imposition of more
stringent emission limits could provide too great a
disincentive to the use of emerging technologies, contrary
to the intent of establishing an emerging technology
provision.
6. Comment: Several commenters (IV-D-26, IV-D-52, IV-D-75, IV-D-80,
IV-F-1.12) said the proposal would discourage the
development of inexpensive innovative technologies because
it would require high-risk investment, since any technology
would have to meet both highly restrictive percentage
reduction figures (50 percent) and an emission rate of 258
ng/J (0.6 lb/million Btu). According to the commenters, it
would also reduce the capital available for continued
research and commercial development since a major portion of
a company's finite dollars would have to be spent on
FGD systems and/or backup systems to ensure immediate and
continued compliance with the standards.
206
-------
Any SO2 control technology with the potential of achieving
90 percent emission reductions in the future should be able
to meet a 50 percent reduction requirement at initial
application. This is not considered overly restrictive
since the aim of the emerging technology provision is to
foster development of SO2 control technologies with promise
of achieving "demonstrated technology" status (i.e., 90
percent reduction). As discussed previously, the emission
limits for emerging technologies are also considered
reasonable in light of the purpose of the emerging
technology provision.
7. Coiranent: One commenter (IV-D-3) suggested that the Agency should
consider how it plans to limit the emerging technology
provision to: (1) new technologies or processes which are
unique relative to demonstrated technologies rather than
simply modified versions of currently demonstrated
technologies, and (2) those which are likely to be able to
meet 90 percent removal after a reasonable demonstration and
development time.
Response: As in the proposed standards, the final standards include
definitions of demonstrated technologies. In some cases,
these definitions have been revised in response to the
concerns expressed by the coiranenter. Technologies
considered as nothing more than slightly modified versions
of existing demonstrated technologies will not be viewed as
emerging technologies by the Agency.
As with all new source performance standards, steam
generating units subject to the final standards must notify
the Agency within 30 days of the date of commencement of
construction of the affected facility. If the owner or
operator of the steam generating unit plans to use an
Response:
207
-------
generating unit plans to use an emerging technology, and
thereby operate under a 50 percent reduction requirement, a
full and complete description of this technology must be
submitted to the Agency along with a discussion of how or
why this technology does not meet any of the definitions of
demonstrated technologies. Technologies not considered
demonstrated by the Agency will be designated as emerging
technologies, and the steam generating unit owner or
operator will be so notified. If the steam generating unit
commences operation with the technology and fails to comply
with the 50 percent reduction requirement and the
established emission limits for emerging technologies, it
will be subject to enforcement action by the Agency. In
addition, to ensure and maintain consistent enforcement of
the special provisions for emerging technologies, these
provisions will not be delegated to States.
The emerging technology provision will be reviewed regularly
during the course of the review process associated with all
NSPS. As appropriate, the percent reduction requirements
will be revised upward in light of additional performance
data available at that time for various emerging
technologies. As a result of these reviews, emerging
control technologies that do not demonstrate improvements in
performance capabilities, or show no promise of achieving
emission reductions greater than 50 percent, will no longer
be considered emerging technologies and all subsequent
installations would be subjected to the same requirements as
those included in the standards for conventional
demonstrated control technologies.
9. Comment: Two commenters (IV-D-26, IV-D-73) felt that the procedures
in 60.49(b)(4) for establishing controls as emerging
technologies are too complex, time consuming, and
208
-------
discretionary. They said the procedure should be more
streamlined and straightforward in the steps for gaining
approval as an emerging technology.
Response: The procedures for establishing controls as emerging
technologies consist of submitting a description of the
technology to the Administrator for review and approval.
This is not considered to be an unreasonable requirement,
given the significant benefits associated with qualifying as
an emerging technology.
10. Comment: Several commenters (IV-D-26, IV-D-28, IV-D-30, IV-D-42,
IV-D-50, IV-D-53, IV-D-73, IV-F-1.16) said disallowance of
credit for precombustion cleaning of fuel toward the percent
reduction requirement for emerging technologies 1s
Inappropriate and counter to the Agency's past philosophy of
preferring the cleanup of fuels prior to combustion, rather
than post-combustion cleanup. Another (IV-D-52) felt that
fuel precleaning should be Included among the emerging
technologies, consistent with the Agency's previous position
of encouraging fuel precleaning.
Response: While the allowance of full credit for fuel pretreatment
toward the 90 percent reduction requirement 1s appropriate
and 1s allowed in the final standards, giving credit for
fuel pretreatment toward the 50 percent reduction
1 requirement for emerging technologies is inappropriate for
several reasons. First, the primary objective of the
50 percent reduction requirement is to stimulate and
encourage the development and use of emerging SOg control
technologies which show promise of achieving significant
emission reductions 1n the future. Any emerging technology
that is unable to reduce emissions by 50 percent without the
209
-------
use of fuel pretreatment credits is unlikely to have any
potential for achieving SOg removal levels of 90 percent in
the future.
Second, allowance of fuel pretreatment credits toward the
50 percent reduction requirement would give a
disproportionately larger benefit to emerging technologies
than to technologies subject to the 90 percent reduction
requirement. For example, a steam generating unit firing
coal with potential S02 emissions of 4,300 ng/J
(10 lb/million Btu) and subject to a 90 percent reduction
requirement would be required to reduce emissions to
430 ng/J (1 lb/million Btu). Allowance of a 30 percent fuel
pretreatment credit would reduce the effective sulfur
content to 3,010 ng/J (7 lb/million Btu). In order to
achieve an emission level of 430 ng/J (1 lb/million Btu), an
85.7 percent reduction in SOg emissions is still required.
However, a steam generating unit firing this same coal and
subject to the 50 percent reduction requirement would be
required to reduce emissions to only 2,150 ng/J
(5 lb/million Btu). Allowance of a 30 percent fuel
pretreatment credit would reduce the effective sulfur
content to 3,010 ng/J (7 lb/million Btu), requiring only a
28 percent reduction in SC^ emissions to be achieved by the
control device. This is clearly contrary to the intended
purpose of encouraging the development of SO2 control
technologies capable of achieving a 90 percent reduction in
emissions.
The final standard, therefore, does not allow fuel
pretreatment to be credited against the percent reduction
requirement for emerging technologies. The final standard
has, however, been amended to allow this fuel pretreatment
210
-------
credit if all of the required 50 percent reduction is
achieved by the fuel pretreatment.technology. This will
serve to encourage the development of fuel pretreatment
technologies that show promise of achieving significant
(i.e., 90 percent) SO2 reductions.
11. Comment:
One commenter (IV-D-85) said that lime spray drying should
be considered an emerging technology for oil-fired steam
generating units.
Response: As discussed above in Section 2.8, lime spray drying is
considered to be capable of achieving a 90 percent reduction
in SO2 emissions from oil-fired steam generating units, and
is therefore considered a "demonstrated" technology for this
application.
12., Comment:
Two commenters (IV-D-11, IV-F-1.5) requested that natural
gas re-burn/sorbent injection be considered a new SOg
control technology subject to the 50 percent reduction
requirement.
Response: If natural gas re-burn/sorbent injection is capable of
achieving at least a 50 percent reduction in S02 emissions
from coal- or oil-fired steam generating units, it would
qualify as an emerging technology under the final standards.
13. Comment:
One commenter (IV-D-42) said the municipal waste handling
techniques of source separation and waste processing and
recycling should be considered emerging technologies subject
to the 50 percent reduction requirement.
211
-------
Response: Municipal waste is not a regulated fuel under the standards
and is therefore not subject to a percent reduction
requirement.
2.15 MONITORING, RECORDKEEPING, AND REPORTING REQUIREMENTS
2.15.1 Continuous Emission Monitoring Systems
1. Comment: One commenter (IV-D-38) said the standard should allow for
alternatives to continuous emission monitoring systems
(CEMS), such as monitoring fuel sulfur content. The
commenter asserted that CEMS are extremely difficult to
operate and are unreliable. Another commenter (IV-D-74)
said sources which are exempt from the percent reduction
requirement should be allowed to monitor fuel sulfur content
to demonstrate compliance with the standards.
Response: The standard does allow alternatives to CEMS. As mentioned
previously (see Section 2.7.5), steam generating units
subject to a percent reduction requirement can use inlet
fuel sampling and analysis in place of inlet CEMS and Method
6B stack testing can be used in place of the outlet CEMS.
For steam generating units exempt from the percent reduction
requirement, where only one sampling device is required,
either fuel sampling or Method 6B may be used 1n place of a
CEMS.
2. Comment: One commenter (IV-D-62) claimed that it is not reasonable to
apply the requirements for continuous emission monitoring
prior to control equipment, since emissions prior to the
control equipment do not have an impact on the environment
or on public health or welfare. The commenter said it is
212
-------
only actual emissions entering the atmosphere which may have
an effect on public health and welfare, and which may have
other environmental effects.
Response: In order to determine whether a percent reduction is being
achieved by the SO2 control device, it is necessary to know
what the potential emissions of the flue gas were before it
passed through the control device. Therefore, some type of
inlet monitoring is necessary. This is commonly achieved
with a CEMS, but can also be achieved by fuel sampling and
analysis.
3. Comment: Four commenters (IV-D-28, IV-D-56, IV-D-81, IV-F-1.1) said
the CEMS requirements, which are basically the same as those -
for utility steam generating units, are unreasonable for
industrial-commercial-institutional steam generating units.
They said the provision that units using low sulfur fuels
still be required to monitor emissions continuously results
in monitoring costs that exceed pollution control costs in
some cases, and creates an artificial bias against the use
of low sulfur coals. The commenters suggested that
provisions similar to those in the current NSPS providing
for alternatives to CEMS would markedly improve the cost
effectiveness of coal cleaning and low sulfur fuel use
without affecting the cost effectiveness of FGD. They
asserted that for a monitoring system to exceed the cost of
a control system by several fold 1s contrary to policy
implicit in existing regulations.
Response: As discussed above, the final standard does allow
alternatives to CEMS, such as Inlet fuel sampling in lieu of
an inlet CEMS. Steam generating units obtaining 30 percent
or less of their maximum heat input capacity on an annual
213
-------
basis from coal or oil may monitor emissions using as-fired
fuel sampling and analysis or Method 6B, as provided in
Method 19.
The annualized costs for emissions monitoring (assuming both
an inlet and outlet SOg monitor and diluent monitor) were
estimated in the "SO2 Cost Report" to be approximately
$100,000/year, compared to total annualized emission control
costs of $900,000/year for a typical 44 MW (150 million
Btu/hour) steam generating unit firing low sulfur coal.
This represents only about 10 percent of total S0£ control
costs.
4. Comment: Several commenters questioned the performance and
reliability of CEMS. Two (IV-D-81, IV-F-1.1) asserted that
the reliability of CEMS has not been proven, especially in
the acidic atmosphere found in FGD systems. Others
(IV-D-22, IV-D-26, IV-D-30, IV-D-50, IV-D-53, IV-D-72,
IV-D-78, IV-F-1.6) felt that the need for CEMS should be
eliminated due to high maintenance requirements, shortages
of instrument technicians, high capital costs, and the
consequent likelihood of unavailability. They said that
small companies with limited staffing will be hard pressed
to deal with the monitoring and reporting requirements.
Response: The reliability of S02 and diluent CEMS were extensively
studied during the development of Subparts 0, Da, and J and
Appendix F of 40 CFR Part 60. The gas streams found at
these sources are similar to those found at industrial-
commercial -institutional steam generating units. These
studies found CEMS to be reliable and capable of meeting the
minimum data availability requirements described in this
standard.
214
-------
In addition, the costs associated with the monitoring and
reporting requirements were examined. This included the
costs associated with the initial performance test and
performance demonstration of the CEMS, as well as operation
and quality assurance procedures, data collection, and
report preparation. For most facilities, the costs would be
less than $20,000/year after the first year. These costs
are considered reasonable. If, in an individual situation,
a steam generating unit operator perceives the CEMS
requirements as too burdensome, alternative monitoring
procedures may be used, as described above.
5. Comment: One commenter (IV-D-62) said that it is neither reasonable
nor necessary to require continuous emission reduction
monitoring and reporting, if an effort is made to set and
enforce reasonable emission limits.
Response: The final standards are based on the "best" technological
system of continuous SOg emission reduction, as required by
Section 111 of the Clean Air Act. This "best demonstrated
technology" is the use of FGD to control SOg emissions. In
order to determine that a percentage reduction is being
achieved by the control device, continuous emission
monitoring at both the inlet and outlet to the control
device is necessary. As discussed above, the costs
associated with all monitoring and reporting requirements in
the final regulation were assessed and are considered
reasonable.
2.15.2 Averaging Time
1. Comment: Several commenters (IV-D-23, IV-D-58, IV-D-62, IV-D-66,
IV-D-74, IV-D-84) felt that the proposed 30-day rolling
average compliance period is unreasonably short. They
215
-------
suggested that compliance be determined on at least a
quarterly or 90-day basis to provide operational
flexibility. The commenters added that the 30-day average
does not ensure that a well designed and operated source
will comply with the rule despite its best efforts. They
contended it is a violation of Section 111 to promulgate a
standard that allows a well operated facility to violate the
standards.
Response: As discussed in the "Summary of Regulatory Analysis,"
various averaging times were considered. The primary
purpose of establishing an averaging time for compliance
purposes is to minimize the effect of variability in fuel
sulfur content and short-term perfomance of control devices
on compliance with the standards. The longer the period
selected for averaging S02 emissions data, the lower the
variability exhibited by the data and the more realistically
it reflects the long-term or average performance of the
system. However, in terms of enforcing compliance with the
standards, this averaging period must also be short enough
to permit timely enforcement of a standard once a source
begins operation. An averaging period of 30 days is
considered long enough to yield data representative of
long-term performance, while also being short enough to
allow timely enforcement of the standards. In addition, use
of a 30-day rolling average (as opposed to a discrete
average) allows enforcement of the standard on a daily
basis.
Based on the assessment of demonstrated performance levels
for the various SOg control systems discussed in the
"Summary of Regulatory Analysis," a properly designed, well
216
-------
operated, and properly maintained control system will be
able to comply consistently with the standards on a 30-day
rolling average basis.
2. Comment: Two commenters (IV-D-1, IV-D-20) said the requirement to
calculate rolling 30-day average SOg emissions using only
steam generating unit operating days rather than consecutive
calendar days makes comparison with other data (e.g.,
ambient) difficult if not impossible. Three commenters
(IV-D-1, IV-D-20, IV-D-86) suggested that a daily average
emission limit should be added to the 30-day rolling average
emission limit to prevent steam generating unit owners or
operators from burning very high sulfur fuel at a very high
rate one day, then burning very low sulfur fuel another day
to compensate.
Response: The calculation procedures as proposed are considered
appropriate. Some industrial steam generating units may
experience sporadic usage patterns that result in the steam
generating units only being operated for a few days in a
calendar month. Using 30 consecutive calendar days to
determine average emissions would permit steam generating
unit operators to "trade off" emissions during those days
when the steam generating unit unit was operating against
days when no emissions occurred because the steam generating
unit did not operate. This approach would effectively
defeat the objective of the standards -- i.e., to minimize
emissions from combustion of fuels in steam generating
units. Allowing a "credit" for periods of time when the
steam generating unit is not operated would reduce the
reduction in emissions necessary from combustion of fuels
when the steam generating unit is operated.
217
-------
The CEMS requirements under Subpart Db are associated with
Subpart Db technical S02 control requirements. If other
averaging times are judged necessary under SIP requirements
or other environmental programs, the operating permit issued
under PSD determinations can require CEMS data to be
provided for that site-specific averaging time.
3. Comment: One commenter (IV-D-38) said the rules should be reviewed to
allow for block rather than rolling averages as is allowed
for utility sources. The commenter asserted that there is
no reason for this standard to be more stringent than the
utility standard.
Response: The utility standard (Subpart Da) requires compliance
determination on a 30-day rolling average basis and, thus,
is the same as this standard. As mentioned above, the
purpose of a rolling average is to provide compliance data
on a continuous, daily basis. A block average would not
allow this daily record, which is important in enforcing the
standards to ensure that compliance is being achieved and
maintained on a continuous basis.
4. Comment: One commenter (IV-D-32) felt that the emission monitoring
requirements should be simplified, and suggested that a
30-day composite sampling program be instituted in place of
rolling averages.
Response: Compositing fuel samples collected over a 30-day period for
analysis would not provide daily compliance data, which is
necessary for enforcement. A 30-day composite would
effectively convert the standard from a rolling average to a
block average. As discussed above, a block average would
preclude the ability of enforcement personnel to enforce the
218
-------
standard on a continuous, daily basis. Thus, a 30-day
composite sampling program would unnecessarily reduce the
enforceability of the standards.
2.15.3 Data Availability Requirements
1. Comment:
Two commenters (IV-D-1, IV-D-86) stated that if enforcement
actions are taken for failure to meet minimum data
availability requirements, the estimation procedures of
Method 19 would be unnecessary as sources would maintain the
CEMS so as to obtain the minimum amount of data in order to
avoid penalties.
Response: The minimum data availability requirements are directly
enforceable, and failure to meet them is considered a
violation. However, this does not preclude the need to
calculate average percent reduction and emission values for
the period. The data estimation procedures in Method 19 are
provided to enable a judgment to be made with a measure of
confidence using available data when the minimum amount of
data is not obtained.
2. Comment:
One commenter (IV-D-13) noted that Section 60.47b(d)
requires at least two data points to calculate each 1-hour
average, while 60.13(h) of the General Provisions requires
four or more equally spaced data points over each 1-hour
period. For the sake of consistency, the commenter said,
the Subpart should agree with the General Provisions unless
substantial reasons can be given for doing otherwise. Two
other commenters (IV-D-1, IV-D-86) suggested that the
minimum amount of data necessary for a valid hourly average
should be specified as a percent rather than the proposed
"two points." They said the proposal would allow the use of
219
-------
readings recorded during the first 2 minutes of the hour to
be used to calculate the average for the hour. The
commenters pointed out that sources in Pennsylvania must
collect 75 percent valid data for a valid hourly average.
Response: The requirements in 60.13(h) are minimum CEMS design
criteria; those in 60.47b are minium data capture or CEMS
performance criteria. These minimum performance criteria
are sufficient to indicate that the CEMS is being properly
operated and maintained. Facilities are still required to
collect all data; the "2 points" requirement is a minimum
standard that must be met. There is no need for this to be
the same as the design criteria in the General Provisions.
It is true that, as proposed, the regulation allowed 2
minutes of data to be used to represent an hour average.
This was judged to be inappropriate, and the monitoring
requirements have been clarified in the final standards.
The final standards require at least 30 minutes of
continuous operation in order to obtain a valid 1-hour
average. In addition, one of the two required data points
must be obtained during each of the two 15-minute periods of
operation.
3. Comment: Two commenters (IV-D-1, IV-D-86) said the use of the
proposed "75 percent of operating hours" criterion would
allow a single hourly average to represent the emissions for
the entire day 1f the unit were operated only during that
hour for the day. They felt that some criteria based on
total hours rather than operating hours should be used so
that short-term data would not be used to determine
compliance with long-term averages. Another commenter
(IV-D-64) said that because the proposed regulation requires
220
-------
emissions data for only 75 percent of the operating hours in
22 of 30 days, only 65 percent of the total operating time
for the facility must be included. The commenter said the
monitoring requirements should be amended to ensure that
complete and accurate data on emissions are generated, and
to not reward operators who do not operate their monitors
properly.
Response: The requirement for "... a minimum of 75 percent of the
operating hours in at least 22 of 30 successive steam
generating unit operating days" is solely related to CEMS
performance criteria and is not related to procedures for
calculating 30-day average emission rates. The 30-day
average emission rate which is used to determine compliance
with the standard is calculated as the average of all valid
hourly emissions data in the past 30 steam generating unit
operating days. All hourly emissions data are given equal
weight; therefore, it does not matter if the steam
generating unit is operated for one hour or 24 hours in any
particular steam generating unit operating day.
Additionally, all valid hourly emissions data are used in
the calculation, whether the "75 percent" criterion is or is
not met. Because of these procedures, the operation of the
steam generating unit at a low emission rate for 1 hour in a
steam generating unit operating day does not get equal
weight in offsetting 24 hours of poor performance in another
steam generating unit operating day. In summary, the
commenter's concerns are valid but the calculation procedure
adequately addresses them.
Minimum data capture requirements provide for downtime, but
limit the amount of lost data before supplemental sampling
is required. The requirements provide the owner or operator
221
-------
with time to maintain and calibrate the CEMS, correct minor
malfunctions, and arrange for supplemental sampling if
necessary, while at the same time providing sufficient data
for compliance determinations. Minimum data capture
requirements also prevent the possibility of an affected
facility operating for unreasonably long periods without
collecting data. This does not reward operators; the extra
cost of supplemental sampling provides incentive to operate
the CEMS properly. Well operated and maintained CEMS will
routinely operate better than the minimum data requirements
and thus supplemental sampling should rarely be required.
4. Comment: Two commenters (IV-D-1, IV-D-86) said sources should not be
allowed to arbitrarily exclude any data. They suggested
that specific data validation criteria be developed so as to
retain control of which data are included or excluded.
Response: Procedure 1 (Appendix F) provides specific criteria for
defining when CEMS data are not valid for purposes of
meeting the minimum data capture requirements. Sections
60.12 and 60.13 of the General Provisions require continuous
operation except for periods of system breakdown, repair,
calibration checks, and zero and span adjustment. Section
60.47b provides minimum data capture requirements, and
60.49b requires identification of times when emission data
have been excluded and justification for excluding data. As
long as the CEMS is not "out of control" as defined by
Procedure 1 (Appendix F) and the source is combusting
sulfur-bearing fuels, there appear to be few instances when
CEMS data can be excluded. The burden of proof is on the
operator to justify exclusion of data from average emission
rate calculations. If the CEMS is operating, the data are
222
-------
required to be made available upon request. This appears to
provide sufficient authority while allowing enforcement
discretion for unforeseen, yet justifiable, reasons.
5. Comment: Two commenters (IV-D-1, IV-D-86) said that if indication of
daily averages that are valid due to excess drift were
required, reporting of each individual drift calculation
would be unnecessary.
Response: This has been examined and appropriate changes have been
incorporated in the regulation. There is no need to
routinely report daily CEMS drift test results. However,
periods when the CEMS is out of control are required to be
reported by 60.49b(j)(5).
2.15.4 Performance Testing
1. Comment:
One commenter (IV-D-13) noted that nowhere 1n the proposed
regulation does it specifically state that the CEMS data are
to be used for the initial or subsequent performance tests.
According to the commenter, neither Method 19 nor 19A
specifies a method for measuring the SOg concentration in
the stack. The commenter said the subpart should explicitly
stipulate how the source should obtain these values for
determining compliance.
Response: This has been clarified 1n the final standard, which
requires an initial 30-day performance test to be conducted
using the CEMS or alternative monitoring procedures under
Method 19.
2. Comment:
One commenter (IV-D-13) said that if the CEMS values are to
be used in the initial performance test, then the
Performance Specification Test should be conducted and
223
-------
accepted prior to the initial performance test. The
commenter stated that this must be stipulated in the
subpart, because the General Provisions require the CEMS
performance evaluations to be conducted during or within 30
days after the initial performance test.
Response: The final standard stipulates that the CEMS Performance
Specification test must be conducted prior to the initial
performance test of the SO2 control system.
2.15.5 Startup. Shutdown, and Malfunction
1. Comment: Several commenters (IV-D-21, IV-D-23, IV-D-26, IV-D-44,
IV-D-55, IV-D-56, IV-D-57, IV-D-58, IV-D-62, IV-D-66,
IV-D-73, IV-D-74, IV-D-79, IV-D-84, IV-D-85, IV-F-1.7) felt
that emissions during startup, shutdown, and malfunction
should not be included when determining compliance with
percent reduction requirements. They said a 30-day rolling
average does not provide sufficient allowance for these
periods with current FGD system reliabilities and the
uncertainty of availability of very low sulfur fuels, and
the costs of maintaining a backup compliance system are
prohibitive and unrealistic.
•Response: Emissions from industrial-commercial-institutional steam
generating units during periods of startup, shutdown, and
malfunction can be significant. In addition, a review of
the factors affecting FGD performance indicates that SO2
removal efficiency can be maintained at high levels even
during periods of startup and shutdown. Therefore, there
generally should be no need, beyond conventional practice
(such as firing natural gas during startup of a coal-fired
steam generating unit for ignition purposes), for firing
224
-------
alternative fuels during periods of startup and shutdown.
Overall, the 30-day averaging time is long enough to "dampen
out" any slight variations in short-term SO2 control
technology performance during these periods. During periods
of FGD system malfunction, however, some type of alternative
compliance procedure may be necessary to maintain compliance
with the standards on a 30-day rolling average basis. The
costs of various alternative compliance procedures
(maintaining a spare scrubber module or firing natural gas
or very low sulfur oil) were examined and found to be
reasonable. Therefore, emissions during periods of startup,
shutdown, and malfunction are not exempt from compliance
with the standards.
2. Comment: Five commenters (IV-D-58, IV-D-62, IV-D-66, IV-D-74,
IV-D-84) said that by requiring the burning of an
alternative fuel such as natural gas during periods of
startup, shutdown, and malfunction, the standard appears to
be in contravention of Section 111. They claimed the
requirement for "continuous" emission reduction has been
interpreted by the courts as likely banning intermittent
controls, including the temporary use of low sulfur fuels.
Response: Flue gas desulfurization, supplemented by low sulfur fuels,
constitutes a system of continuous emission reduction.
Indeed, it is more continuous than FGD alone, since the low
sulfur fuels minimize emission peaks during startup,
shutdown, or malfunction of the FGD system. The requirement
that NSPS reflect "continuous" systems was meant to avoid
"intermittent" control systems, which reduce emissions only
when atmospheric dispersion conditions are poor [H. R. Rep.
No. 294, 95th Cong., 1st Sess. 81-92 (1977)]. The
225
-------
requirement was not intended to prevent the inclusion in an
NSPS of systems that minimize emissions during startup,
shutdown or malfunction.
2.15.6 Miscellaneous Monitoring Comments
1. Comment: One commenter (IV-D-20) said that under the alternative to
CEMS, fuel sampling and analysis, specified in 40 CFR
60.47b(b)(l), the requirement for daily fuel oil samples is
unnecessary for affected facilities firing fuel oil from
homogeneous storage tank mixtures. The commenter stated
that upon receipt of additional fuel oil shipments, it would
then be appropriate to require additional fuel oil samples
and analyses.
Response: The requirement that samples of fuel oil be analyzed dally
for sulfur content has been retained for two reasons.
First, in many large manufacturing facilities, multiple oil
storage tanks feeding into the same fuel lines could result
in a mixture of oils with different sulfur contents being
fired in the steam generating unit. Second, fuel oil placed
in storage prior to being fired can mix with fuel oil from
prior or subsequent shipments (or stratify, depending on
tank design) and result in a change in the sulfur content of
the fuel oil being fired. For these reasons, continuous
compliance with the S02 emission limits could not be
determined accurately without daily analyses of the sulfur
content of fuel oil fired in a steam generating unit. Daily
sampling and analysis of oil sulfur content is uncomplicated
and not time consuming, and should not represent a
significant burden to steam generating unit operators.
226
-------
As a result of refining, however, low sulfur fuel oils show
very little variability in heating value. Data obtained
from one electric utility plant, for example, show that the
heating value (J/kg; Btu/lb) of the fuel oil received at the
plant over a 77-day period, comprising four separate
shipments of fuel oil, varied by less than one-half of 1
percent. Because the variability in the heating value of
low sulfur fuel oil is negligible, the final regulation
allows the use of a single heating value in calculating the
30-day rolling average emission rates for each calendar
quarter for such oil. The final regulation requires owners
or operators of oil-fired steam generating units to
determine the heating value of the fuel oil sampled during
the first steam generating unit operating day in each
calendar month. The lowest of the three heating values
obtained during each calendar quarter is used to calculate
all of the SO2 emission rates for the calendar quarter. If
a steam generating unit operator believes that, for some
reason, the first steam generating unit operating day was
not representative of the fuel fired for the remainder of
the calendar month, more frequent measurement of the heating
value may be conducted.
2. Comment: One commenter (IV-D-72) said that bi-weekly testing of EPA
reference samples and splitting additional fuel samples with
independent laboratories as a quality assurance procedure
when performing fuel sampling and analysis is excessive.
The commenter asserted that split samples for quality
assurance should be conducted monthly, and reference samples
should be analyzed quarterly.
Response: The requirements mentioned by the commenter are not included
in either the proposed or promulgated standards.
227
-------
2.15.7 Recordkeeping and Reporting
1. Comment:
Two commenters (IV-D-1, IV-D-86) said all data collected by
the CEMS should be required to be recorded, to provide a
convenient record of CEMS activity, even if some of the data
are not required to be reported.
Response: A requirement that all data collected by the CEMS be
recorded and retained for 2 years has been included in the
final standard.
Comment: Four commenters (IV-D-26, IV-D-30, IV-D-50, IV-D-53) said
the sheer complexity of some of the reporting requirements
will make it difficult to obtain accurate calculations. For
example, the commenters said, when burning mixed fuels, the
proper "F" factors have to be determined based on a complex
set of operating variables. They claimed that having to
calculate all these variables will unnecessarily contribute
to less accurate and less meaningful data.
Response: The reporting and calculation requirements reflect only the
information necessary to determine that the CEMS is
operating properly and the unit is in compliance. The
requirements are clearly described in the regulation and in
Method 19 and efforts have been made to keep them as
uncomplicated as possible.
3. Comment:
Four commenters (IV-D-26, IV-D-30, IV-D-50, IV-D-53) said
the time that will be required for other agencies to review
the reported data was not considered. They said past
experience indicates that State agencies are inadequately
staffed to cope with the volume and sophistication of the
information which would be provided.
228
-------
Response: In developing the reporting and recordkeeping requirements
associated with these standards, the burden associated with
review of the reports by enforcement personnel was assessed.
This burden, which may be assumed by Federal, State, or
local enforcement personnel, was determined to be reasonable
and, therefore, submittal of reports from steam generating
unit operators is required. If State or local enforcement
officials find these burdens to be too high, given specific
State or local circumstances, they may elect not to request
delegation of authority from the Federal government to
administer and enforce this NSPS.
16 MISCELLANEOUS COMMENTS
. Comment: One commenter (IV-D-9) felt that the proposed effective date
of the standard (June 19, 1986) is too restrictive.
According to the commenter, most projects of this size take
from 9 to 18 months to design and it would force a massive
redesign if the percent reduction requirement were made
retroactive. The commenter suggested that the effective
date be revised to 6 months after promulgation of the final
standard.
Response: Section 111(a)(2) defines the "new sources" subject to an
NSPS to be those built (i.e., on which construction
commences) after the NSPS is proposed. This does not impose
retroactive requirements or require redesign of sources.
The notice of proposed rulemaking put potential owners and
operators of sources to be built thereafter on notice that
those sources would be subject to the NSPS. Sge Denial of
Petitions for Reconsideration of Final [Utility Boiler NSPS]
Regulations, 45 FR 8210, 8232-8233 (Feb. 6, 1980).
229
-------
2. Comment: Two commenters (IV-D-55, IV-D-56) said the failure to
promulgate the revision to the source category list to
include nonfossil fuel-fired steam generating units and
commercial and institutional steam generating units before
proposing the standards is in violation of the procedure
required by the Clean Air Act. The commenter asserted the
Agency should first complete the rulemaking regarding the
list of source categories and then repropose a standard.
Response: Section 111(b) requires the Agency to list, and set NSPS
for, "categories of stationary sources...[that] cause[], or
contribute!] significantly to, air pollution which may
reasonably be anticipated to endanger public health or
welfare." These source categories are referred to as
"significant contributors" [National Asphalt Pavement Ass'n
v. Train. 539 F. 2d 775 (D.C. Cir. 1976)]. Section 111(f)
required the Agency to add all "major" source categories
that are significant contributors to the Section 111(b)
11st, and to set NSPS for them by August 7, 1982. A "major"
source is one that emits more than 100 tons per year of any
air pollutant [Section 302(j) .
In 1971, the Agency listed and set NSPS for fossil fuel
steam generating units of more than 250 million Btu/hour
heat input [36 FR 5931 (March 31, 1971)]. In 1979, the
category of fossil fuel industrial steam generating units
(without regard to heat input rate) was added to the
priority 11st [44 FR 49222 (Aug. 21, 1979)]. In 1984, it
was concluded that this latter source category should be
broadened to include commercial, institutional and nonfossil
fuel steam generating units. This conclusion was based on
the determination that the design, emission rates, and
230
-------
applicable control techniques for fossil, nonfossil and
mixed fuel steam generating units were substantially
similar:
In fact, any practical difference between fossil and
nonfossil fuel-fired boilers has virtually disappeared
as many new boilers have interchangeable fossil fuel,
nonfossil fuel and mixed fuel capability [49 FR 25156
(June 19, 1984)].
In addition, commercial and institutional steam generating
units were found to be substantially similar to industrial
steam generating units:
These boilers emit similar pollutants, fire the same
fuels, and may enjoy the same emission control
techniques. Their impacts on human health and welfare
are similar... [Id].
The Agency therefore proposed to broaden the source category
to include commercial, institutional and nonfossil fuel
steam generating units, and simultaneously proposed NSPS
(controlling emissions of PM and N0X) for the broadened
source category [49 FR 25102 (June 19, 1984)]. Final action
was taken on both proposals simultaneously, broadening the
source category as proposed, and promulgating an NSPS for
the broadened source category [51 FR 42794, 42768 (Nov. 25,
1986)]. The final NSPS, controlling S02 emissions, also
applies to the broadened source category.
There is no requirement to conduct multiple consecutive
rulemakings, the first to determine the scope of the source
j category, and later ones to determine the terms of the NSPS
fNational Asphalt Pavement Ass'n v. Train. 539 F. 2d 775
(D.C. Cir. 1976)]. In National Asphalt, a single rulemaking
231
-------
was conducted to consider both listing a source category and
setting an NSPS for that source category. The Court upheld
that procedure:
Thus, the EPA can continue to have one informal
rulemaking proceeding as long as that proceeding
considers both the "significant contributor" designation
and the proposed standards. Indeed, in rulemaking
proceedings such as this one, where the data underlying
the "significant contributor" designation is likely to
overlap substantially with that underlying the proposed
standards, the most sensible course for an agency is to
have one proceeding directed at both issues [539 F. 2d
at 779 n.2. Accord. Thomas v. State of New York. 802 F.
2d 1443, 1443, 1446, 1447 (D.C. Cir. 1986)].
The addition of subsection (f) to Section 111 by the 1977
Amendments to the Act does not change the result.
Subsection (f) simply requires the Agency to list by
regulation major source categories that are significant
contributors, and to set NSPS for them. The legislative
history shows that the purpose of subsection (f) was simply
to expedite the setting of NSPS for these source categoric
[H.R. Rep. No. 294, 95th Cong., 1st Sess. 193-195 (1977)].
Neither the text nor the legislative history of subsection
(f) suggests that National Asphalt was being overruled, or
that EPA was barred from acting on listing a category and
setting NSPS for it simultaneously.
In addition, when it was able to propose broadening the
source category and propose an NSPS for it in 1984, the
Agency was already in default of the statutory requirement
that the NSPS be promulgated by 1982. See Sierra Club,
supra. If EPA had further delayed the NSPS rulemaking while
it first completed a separate listing rulemaking, that would
have worsened the Agency's violation of the statutory
dead!ine.
232
-------
The rulemaking notices proposing the broadening of the
source category and proposing NSPS have been clear and have
given commenters full and fair opportunity to comment.
233
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO. 2.
EPA 450/3-89-024
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE _ „
Fossil and Nonfossil Fuel-Fired
Industrial Boilers - Background Information for
Promulgated SO2 Standards, Volume 4.
6. REPORT DATE
September 1987
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Radian Corporation
Research Triangle Park, North Carolina 27711
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Office of Air Quality Planning and Standards
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-02-3816
12. SPONSORING AGENCY NAME AND ADDRESS
Office of Air and Radiation
U.S. Environmental Protection Agency
401 M Street, S.W.
Washington, D.C. 20460
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
EPA/200/04
15. SUPPLEMENTARY NOTES
16. ABSTRACT
This document summarizes EPA's response to public comments received on proposed
new source performance standards for sulfur dioxide emissions from new coal- and
oil-fired industrial-commercial-institutional steam generating units and particulate
matter emissions from oil-fired units (51 FR 22384, June 19, 1986). Alternative
SO, control technologies and regulatory options are discussed in terms of S0?
emission reduction capability, costs of control, secondary environmental impacts,
national impacts, industry-specific economic impacts, emerging technologies, and
monitoring, recordkeeping, and reporting requirements. In addition, the impacts
of allowing emission credits for cogeneration and mixed fuel-fired steam generating
units are reviewed. This document is intended to serve as an overview of the
analyses and regulatory alternatives considered during the standards development
process.
17. 1 KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
b. 1 DENTlF1ERS/OPEN ENDED TERMS
c. COSATI Field/Croup
Air pollution
Pollution control
Standards of performance
Steam generating units
Fossil fuel-fired
industrial boilers
Mixed fuel-fired
industrial boilers
Cogeneration systems
Air pollution control
13B
18. DISTRIBUTION STATEMENT
Release unlimited
19. SECURITY CLASS (TtUs Report)
Unclassified
21. NO. OF PAGES
233
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (R«». 4-77) previous edition is obsolete
------- |