United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
EPA-453/R-98-005
September 1998
Air
EPA New Source Performance Standards,
Subpart Da and Db - Summary of
Public Comments and Responses
-------
EPA-453/R-98-005
New Source Performance Standards,
Subparts Da and Db
Summary of Public Comments and Responses
Emission Standards Division
U.S. Environmental Protection Agency
Office of Air and Radiation
Office of Air Quality Planning and Standards
Research Triangle Part, North Carolina 27711
-------
EPA-453/R-98-005
September 1998
in
-------
DISCLAIMER
This Report has been reviewed by the Emissions Standards Division
of the Office of Air Quality Planning and Standards, EPA, and
approved for publication. Mention of trade names or commercial
products is not intended to constitute endorsement or
recommendation for use.
IV
-------
TABLE OF CONTENTS
1.0 SUMMARY 1-1
2.0 BEST DEMONSTRATED NOX CONTROL TECHNOLOGY 2-1
2.1 SELECTIVE CATALYTIC REDUCTION (SCR) 2-1
2.2 SELECTIVE NONCATALYTIC REDUCTION (SNCR) 2-9
2.3 NATURAL GAS REBURN 2-11
3.0 CONTROL TECHNOLOGY COSTS 3-1
3.1 ESTIMATED COSTS ARE TOO HIGH 3-1
3.2 ESTIMATED COSTS ARE REASONABLE 3-2
3.3 ESTIMATED COSTS ARE TOO LOW 3-2
3.4 OTHER COST ISSUES 3-9
3.4.1 Fuel Switching Costs 3-9
3.4.2 Energy Pricing 3-10
3.4.3 Proposed Standards Not Cost Effective . . 3-11
4.0 REGULATORY APPROACH 4-1
4.1 APPLICABILITY 4-1
4.1.1 "No New Exemptions" Policy 4-1
4.1.2 NOX Emissions Limits for Existing Boilers . 4-3
4.1.3 Existing Sources Should Be Exempt from NSPS
4-4
4.1.4 Modification Criteria 4-8
4.1.5 Applicability in NO.. Attainment Areas . . . 4-9
4.2 FUEL NEUTRAL APPROACH VERSUS SUBCATEGORIZATION . . 4-9
4.2.1 Support Fuel Neutral Approach 4-9
4.2.2 Oppose Fuel Neutral Approach 4-10
4.2.3 Distinguish between Classes, Types and Sizes
4-14
4.3 PROMULGATION SCHEDULE AND COORDINATION WITH ICCR 4-15
4.4 OVERALL MONITORING, REPORTING, AND RECORDKEEPING
REQUIREMENTS 4-18
5.0 ESTABLISHING OUTPUT-BASED FORMAT FOR UTILITY BOILERS . . 5-1
5.1 OVERALL APPROACH 5-1
5.1.1 Support Output-Based Format 5-1
5.1.2 Oppose Output-Based Format 5-1
5.2 INPUT TO OUTPUT CONVERSION ASSUMPTIONS 5-4
5.2.1 Support the 38-Percent Baseline Efficiency . 5-4
5.2.2 Oppose the 38-Percent Baseline Efficiency . 5-5
5.2.3 Support Net Heat Rate of 9.000 Btu/kWh . . . 5-7
-------
5.2.4 Oppose Net Heat Rate of 9.000 Btu/kWh . . . 5-7
5.2.5 Efficiency Calculation for Cogeneration Units
5-10
5.3 GROSS VERSUS NET OUTPUT VARIABLE IN EQUATION . . 5-13
6.0 REVISED STANDARD FOR ELECTRIC UTILITY STEAM GENERATING UNITS
(SUBPART Da) 6-1
6.1 SUPPORT THE LEVEL OF THE STANDARD 6-1
6.2 STANDARD IS TOO LENIENT 6-2
6.3 STANDARD IS TOO STRINGENT 6-3
7.0 REVISED STANDARD FOR INDUSTRIAL-COMMERCIAL-INSTITUTIONAL
STEAM GENERATING UNITS (SUBPART Db) 7-1
7.1 EXCLUSIONS 7-1
7.2 LEVEL OF THE STANDARD 7-3
7.3 INPUT-BASED FORMAT 7-5
7.3.1 Support Input-Based Format 7-5
7.3.2 Oppose Input-Based Format 7-6
CONTINUOUS EMISSION MONITORING (CEM) REQUIREMENTS . . . 8-1
8.1 GENERAL 8-1
8.2 APPLICABILITY TO SMALL/SEASONAL UNITS 8-1
8.3 CONSISTENCY BETWEEN PROGRAMS 8-2
8.4 AVERAGING PERIODS 8-5
8.4.1 Support 30-Day Averaging Period 8-5
8.4.2 Oppose 30-Day Averaging Period 8-5
8.5 SUPPORT ELECTRONIC FILING 8-6
8.6 NEW MONITORING AND PERFORMANCE TESTING REQUIREMENTS
9.0 OTHER 9-1
9.1 COST, ENVIRONMENTAL, ENERGY, AND ECONOMIC IMPACTS . 9-1
9.2 EDITORIAL 9-2
9.3 GLOBAL WARMING 9-3
9.4 BEST AVAILABLE CONTROL TECHNOLOGY 9-3
9.5 APPLICABILITY OF THE CREDIBLE EVIDENCE RULE .... 9-3
9.6 ADDITION OF TECHNICAL DOCUMENTS TO THE RECORD . . . 9-4
9.7 FEDERAL INTERVENTION 9-4
VI
-------
1 . 0 SUMMARY
The EPA proposed amendments to subparts Da and Db of 40 CFR
part 60 on July 9, 1997. The purpose of this document is to
present a summary of the public comments received on the proposed
amendments to subparts Da and Db of 40 CFR part 60 and the
responses developed by the EPA. This summary of comments and
responses serves as the basis for revisions made to the standards
between proposal and promulgation.
The EPA received 70 public comment letters on the proposed
rule changes. The commenters represent the following
affiliations: government (5), utility industry (26), industrial
boiler users (13), public interest and environmental groups (7),
private citizens (4), fuel producers (11) and other (4). This
document incorporates all the comments in the docket and some
additional comments that will be added to the docket. Table 1-1
presents a listing of all persons submitting written comments,
their affiliation, and their docket number (if available). No
comments were received at the public hearing.
1-1
-------
TABLE 1-1. LIST OF COMMENTERS ON THE PROPOSED REVISIONS TO
SUBPARTS Da AND Db
Number3
1
2
3
4
IV-D-01
IV-D-02
IV-D-03
IV-D-04
IV-D-05
IV-D-06
IV-D-07
IV-D-08
IV-D-09
Commenter, Addressee, Title or
Description, etc.
J. Brax, Air Quality Intern,
Environmental Defense Center, Santa
Barbara, CA
R. Machaver, RJ Associates, Lincoln, MA
A. Bodnarik, ICCR Boiler Workgroup
Member, State of New Hampshire
Department of Environmental Services Air
Resources Division, Concord, NH
G. Kamaras, Director, Energy Advocacy
Program, Legal Environmental Assistance
Foundation, Tallahassee, FL
J.D. Baird, Manager, Environmental
Services, Hunt Wesson, Inc., Fullerton,
CA
K. Bailey, Chadbourne & Parks, LLP,
Washington, DC on behalf of the American
Forest & Paper Association, Inc.
R.I. Zvaners, Senior Manager,
Environmental Policy, Chemical
Manufacturers Association, Arlington, VA
F.W. Hottenroth, Private Citizen, Seal
Beach, CA
J.W. Clarke, Private Citizen, Rockville,
MD
C.W. Whitmore, Principal, Whitmore
Associates, Shawnee Mission, KS
M.A. Curtis, Executive Director, New
Jersey Environmental Lobby, Trenton, NJ
T.A. Elter, Sr., Environmental Analyst,
Niagra Mohawk Power Corporation,
Syracuse, NY
R.M. Salmon, Coordinator, Environmental
Services/Public Works Projects, City of
Tampa, FL
Date of
Document
09/2/97
09/3/97
09/5/97
08/20/97
08/5/97
08/7/97
08/20/97
06/19/97
08/20/97
08/19/97
08/21/97
08/25/97
09/5/97
1-2
-------
TABLE 1-1. LIST OF COMMENTERS ON THE PROPOSED REVISIONS TO
SUBPARTS Da AND Db
Number3
IV-D-10
IV-D-11
IV-D-12
IV-D-13
IV-D-14
IV-D-15
IV-D-16
IV-D-17
IV-D-18
IV-D-19
IV-D-20
IV-D-21
IV-D-22
Commenter, Addressee, Title or
Description, etc.
S. Shell, Manager, Environmental,
Safety, and Health, Lockheed Martin
Utility Services, Inc., Paducah, KY
T.J. Porter, Director, Air Quality
Management, Wheelabrator Environmental
Systems, Inc., Hampton, NH
G.K. Crane, Executive Vice President,
Environmental, Ogden Projects, Inc.,
Fairfield, NJ
M. Zannes, President, Integrated Waste
Services Association, Washington, DC
E.D. Yates, Sr. Vice President,
California League of Food Processors,
Sacramento, CA
D. Hearth, Bracewell & Patterson,
L.L.P., Washington, DC
J.W. Dwyer, President, Lignite Energy
Council, Bismarck, ND
J.A. Miakisz, Director Environmental
Regulatory Affairs, Niagara Mohawk Power
Corporation, Syracuse, NY
B. Mathur, Chief, Bureau of Air,
Illinois Environmental Protection Agency
A. Lee, Senior Staff Environmental
Engineer, Texaco, Inc., Beacon, NY
B.A. Craig, Director, Utility and
Environmental Regulatory Affairs,
Natural Gas Supply Association,
Washington, DC
L.J. Becker, Environmental Analyst, San
Diego Gas & Electric, San Diego, CA
N.L. Morrow, Exxon Chemical Americas,
Houston, TX
Date of
Document
09/5/97
09/5/97
09/5/97
09/5/97
09/5/97
09/5/97
10/3/97
10/2/97
10/7/97
10/3/97
10/6/97
10/3/97
10/6/97
1-3
-------
TABLE 1-1. LIST OF COMMENTERS ON THE PROPOSED REVISIONS TO
SUBPARTS Da AND Db
Number3
IV-D-23
IV-D-24
IV-D-25
IV-D-26
IV-D-27
IV-D-28
IV-D-29
IV-D-30
IV-D-31
IV-D-32
IV-D-33
IV-D-34
Commenter, Addressee, Title or
Description, etc.
S.B. Peirce-Sandner, KP Environmental
Services, Eastman Kodak Company,
Rochester, NY
N. Stafki, Senior Environmental Analyst,
Northern States Power Company,
Minneapolis, MN
N. Ford, Sierra Club, Ohio Chapter
Energy Committee, Cincinnati, OH
J.J. Mayhew, Assistant Vice President
Environmental & Policy Analysis,
Chemical Manufacturers Association,
Arlington, VA
T. Romero, U.S. Generating Company,
Bethesda, MD
B.E. Ramsey, Executive Director,
Anthracite Region Independent Power
Producers Association, Lemoye, PA
D.W. Marshall, Staff Attorney,
Conservation Law Foundation, Concord, NH
K.A. Colburn, Director, Air Resources
Division, New Hampshire Department of
Environmental Services, Concord, NH
G. Schaefer, Director, Government Issue
& Analysis, ARCO Coal Company, Denver,
CO
R.L. White, Vice President,
Environmental Services, Texas utilities
Services, Inc., Dallas, TX
E.S. Roy, Vice President,
Intercontinental Energy Corporation/R. D.
Ain, Senior Vice President, Cogen
Technologies
M. Spurr, Legislative Director,
International District Energy
Association, Washington, DC
Date of
Document
10/6/97
10/3/97
10/7/97
10/8/97
10/7/97
10/7/97
10/7/97
10/7/97
10/7/97
10/7/97
10/7/97
10/7/97
1-4
-------
TABLE 1-1. LIST OF COMMENTERS ON THE PROPOSED REVISIONS TO
SUBPARTS Da AND Db
Number3
Commenter, Addressee, Title or
Description, etc.
Date of
Document
IV-D-35
M.R. Robida, Manager - Air Quality,
American Electric Power, Columbus, OH
10/7/97
IV-D-36
M.J. Ruffatto, President, North American
Electric Power Group, Ltd., Greenwood
Village, CO
10/7/97
IV-D-37
C. Seidlits, President & CEO,
Association of Electric Companies of
Texas, Inc., Austin, TX
10/7/97
IV-D-38
S.M. Ruffin, Environmental Services
Department, South Carolina Electric &
Gas Company, Columbia, SC
10/7/97
IV-D-39
M.C. Hall, Manager, Legislative and
Regulatory Affairs, Trigen Energy
Corporation, White Plains, NY
10/7/97
IV-D-40
P. Glaser, Attorney at Law, Doherty,
Rumble & Butler, Washington, DC on
behalf of F.D. Palmer, GM & CEO, Western
Fuels Association, Inc., Denver, CO
10/8/97
IV-D-41
R.L. Brubaker/C.F. Barry, Attorneys at
Law, Porter, Wright, Morris & Arthur,
Columbus, OH on behalf of Ohio Edison
Company
10/8/97
IV-D-42
D.J. Jezouit, Counsel to the Class of
'85 Regulatory Response Group, Baker &
Botts, LLP, Washington, DC
10/8/97
IV-D-43
T.L. Fisher, Chairman, American Gas
Association/Chairman, Natural Gas
Council; L.D. Hall, Chairman, Interstate
Natural Gas Association of America; L.O.
Ward, Chairman, Independent Petroleum
Association of America; M.E. Wiley,
Chairman, Natural Gas Supply Association
10/8/97
IV-D-44
R. Cooper, Senior Vice President,
Government Relations, American Gas
Association, Arlington, VA
10/8/97
1-5
-------
TABLE 1-1. LIST OF COMMENTERS ON THE PROPOSED REVISIONS TO
SUBPARTS Da AND Db
Number3
IV-D-45
IV-D-46
IV-D-47
IV-D-48
IV-D-49
IV-D-50
IV-D-51
IV-D-52
IV-D-53
IV-D-54
IV-D-55
Commenter, Addressee, Title or
Description, etc.
R.C. Kaufmann, Director, Air Quality
Program, American Forest & Paper
Association, Washington, DC
Coalition for Gas-Based Environmental
Solutions, Arlington, VA
F.W. Brownell/C . S . Harrison, Hunton &
Williams, Washington, DC on behalf of
Utility Air Regulatory Group and the
National Mining Association
S.H. Segal, Counsel to the Council of
Industrial Boiler Owners, Bracewell &
Patterson, LLP, Washington, DC
S. Hedman, Environmental Law & Policy
Center, Chicago, IL
J. Grumet, Executive Director, Northeast
States for Coordinated Air Use
Management (NESCAUM) , Boston, MA
M.S. Brownstein, Esq., Environmental
Policy Manager, Air Quality, Public
Service Electric and Gas Company,
Newark, NH
B. Green, Environmental Manager,
Kennecott Energy Company, Gillette, WY
M.W. Stroben, Manager, EHS Technical
Analysis Corporate Environment, Safety &
Health, Duke Energy Corporation,
Charlotte, NC
C. Johnson, Deputy Commissioner, New
York State Department of Environmental
Conservation, Albany, NY
A.W. Hadder, Manager, Environmental
Policy and Compliance, Virginia Power
Date of
Document
10/8/97
undated
10/8/97
10/8/97
10/8/97
10/8/97
10/8/97
10/8/97
10/8/97
10/8/97
10/8/97
1-6
-------
TABLE 1-1. LIST OF COMMENTERS ON THE PROPOSED REVISIONS TO
SUBPARTS Da AND Db
Number3
IV-D-56
IV-D-57
IV-D-58
IV-D-59
IV-D-60
IV-D-61
IV-D-62
IV-D-63
IV-D-64
IV-D-65
IV-D-66
Commenter, Addressee, Title or
Description, etc.
M.G. Dowd, McGuire, Woods, Battle &
Booth, L.L.P., Richmond, VA on behalf of
P.J. Margaritis, Senior Vice President
Tractebel Power, Inc., Houston, TX
M.J. Wax, Deputy Director, Institute of
Clean Air Companies, Washington, DC
D. Heminway, Assistant Director,
Citizens' Environmental Coalition,
Medina, NY
L.E. Watkins, Jr., General Counsel,
Sunflower Electric Power Corporation,
Hays, KS
L.S. Beal, Director, Environmental
Affairs, Interstate Natural Gas
Association of America, Washington, DC
J.L. Woolbert, Engineering Associate,
Eastman Chemical Company, Longview, TX
D. Marrack, M.D., Fort Bend Medical
Clinic, Houston, TX
W.R. Watson, Sr. Environmental
Professional, Illinois Power Company,
Decatur, IL
A. Deshmukh, Environmental Specialist-
Air Quality, Occidental Chemical
Corporation, Dallas, TX
P. Bailey, Director, Health and
Environmental Affairs, American
Petroleum Institute, Washington, DC
A. Titus, A. Bisantz, Private Citizens,
Batavia, NY
Date of
Document
10/9/97
10/7/97
9/29/97
undated
10/8/97
10/7/97
10/4/97
10/6/97
undated
10/9/97
7/1/97
The docket number for this rulemaking is A-92-71.
1-7
-------
2.0 BEST DEMONSTRATED NOX CONTROL TECHNOLOGY
2.1 SELECTIVE CATALYTIC REDUCTION (SCR)
Several commenters raised concerns that the determination
that SCR represents the best demonstrated technology is not
adequate. Following is a summary of their comments, and the
EPA's response.
Comment: Coal-fired industrial boilers. Commenters IV-D-23
and IV-D-26 stated that the EPA should not consider SCR as the
best demonstrated technology for coal-fired industrial boilers.
Commenter IV-D-23 recommended that adequate pilot-plant testing
be conducted for these boilers. Commenter IV-D-31 added that "it
is doubtful whether any of the SCR units that EPA points to could
operate under an emission limit this low." Commenter IV-D-23
noted that SCR is installed on only 7 coal-fired units in the
U.S., all of which are electric utility units. In addition, none
of the 200 European and Japanese units with SCR cited by the EPA
are industrial units. Because the EPA has cited no industrial
units that use SCR successfully, Commenter IV-D-23 asserted that
this technology is not adequately demonstrated.
Commenter IV-D-48 stated that the EPA presented no evidence
of any coal-fired industrial boilers that employ SCR. This lack
of demonstrated technology "does not support imposition of SCR as
the minimum NSPS control level." The commenter recommended that
the EPA consider the potential problems associated with SCR,
including costs, catalyst poisoning, and oil ash coating the
catalyst, when finalizing the NSPS. The commenter suggested that
the standards for coal- and oil-fired boilers be based on the use
of low NOX boilers, staged combustion, and/or selective
noncatalytic reduction (SNCR) which have had some demonstration
in industrial units.
Two major problems cited by Commenter IV-D-60 were
2-1
-------
deactivation of the catalyst from alkali sulfates, and excess
sulfur trioxide (S03) in the flue gas. The commenter contended
that the EPA casually dismissed alkali poisoning without
justification. According to the commenter, excess S03 can lead
to increased downstream corrosion and negative impacts on the
heat rate of the unit.
Commenters IV-D-38 and IV-D-41 stated that "the relevant
technological art is immature... standards rooted in it will not
be attainable on a sustained basis unless they are flexible,"
and that the results of the EPA's examination of SCR and SNCR
were inconclusive. Commenter IV-D-38 added that the flexibility
would need to account for variabilities in the empirical data,
and need "to accommodate phenomena about which the EPA has no
data." Commenter IV-D-63 remarked that "the standards set in
this rulemaking are beyond the envelope of today's technology."
Commenters IV-D-32 and IV-D-37 stated that the coal-fired
and natural gas power plants could not meet a 0.15 Ib/MMBtu
standard without implementing costly SCR technology. The
commenter remarked that the reported cases of successful SCR
applications are extremely limited, with success being measured
on the basis of short-term performance and without cost
considerations.
Coal-fired utility boilers. Commenter IV-D-52 stated that
SCR has only been applied to coal-fired [utility] boilers over
that past two years. The commenter added that this is
"indicative of a developmental phase of technology," noting that
these sample sizes are not valid for any verifiable statistical
comparisons." The commenter also noted that there appears to be
a discrepancy between when the EPA stated that SCR and SNCR
technologies "have been applied widely to commercial-scale gas-
and oil-fired steam generating units" and when the EPA explained
that statistical analysis of combustion control was not performed
2-2
-------
since either (1) no applicable operating subpart Da units are
known to exist, or (2) during development of the proposal, long-
term CEM data were unavailable.
Appropriateness of SCR at pulp and paper mills. Commenter
IV-D-45 indicated that SCR is not appropriate for industrial
boilers, particularly combination boilers at pulp and paper mills
that burn wood and fossil fuels. The commenter explained that
boilers at paper mills are subject to wide, sudden changes in
load that complicate the use of SCR. Other potential problems
include high particulate loadings, high potential for sulfur
poisoning of the catalyst, and difficulty in maintaining the
temperatures necessary to minimize NOX and HAP generation.
Residual oil-fired boilers. Commenters IV-D-19 and IV-D-65
stated that the EPA's data have not demonstrated that SCR
technology reduces emissions from residual oil-firing steam
generating units for the Db standard. Therefore, the commenters
recommended that the EPA retain the current standard of 0.30 Ib
NOx/MMBtu. Commenter IV-D-19 added, that if the EPA insists on a
single performance standard of 0.20 Ib per MMBtu, that the EPA
allow for an annual averaging period for this performance
standard.
Response: The first issue raised by several of the
commenters is that EPA's determination that SCR represents BDT
for a range of boiler type and operating conditions is not
adequate. The EPA disagrees and believes the data base that
supports the BDT decision is adequate for two reasons. First,
the proposal data base resulted from an extensive review on the
available domestic and international SCR units in use in the
industry at the present time. However, in response to the
comments, the EPA has obtained data from three more utility
boilers that utilize SCR and represent a range of operating
conditions and coal types. The first utility boiler (U.S.
2-3
-------
Generating Company's Logan plant) is a 225-megawatt pulverized-
coal cogeneration facility, and is operated under cycling
conditions. This facility submitted 3 months of NOX emission
data to EPA. The analysis of these data indicate that the
facility is capable of achieving the input-based NOX standard of
0.15 Ib/MMBtu and the revised output-based standard of 1.6 Ib/MWh
on a 30-day rolling average. (See section 5.2 for a discussion
of the development of the revised output-based standard.) The
second plant is the Birchwood Power Facility (jointly owned by
Southern Energy Incorporated and Cogentrix), which was described
in Power Engineering (December 1997, pp. 28-30). Birchwood is a
240-megawatt cogeneration facility with cycling load that began
operation in 1996. Birchwood is required to meet a NOX emission
standard of 0.10 Ib/MMBtu on a 30-day rolling average. Actual
test results show that the facility achieves NOX emissions of
0.77 Ib/MWh at low load conditions, easily attaining the output-
based standard. The third facility is a 464-megawatt utility
boiler firing bituminous coal (Stanton Energy, Florida). This
facility is currently meeting its permitted emission limit of
0.17 Ib/MMBtu. If this facility were to improve the performance
of its SCR to 0.15 Ib/MMBtu, this facility would be capable of
meeting the 200 ng/J (1.6 Ib/MWh) output-based limit.
Second, this data base is adequate to evaluate the factors
that can potentially affect SCR performance in a wide range of
operating conditions. According to the subpart Da Background
Information Document (BID), the performance of an SCR system is
influenced by six factors: flue gas temperature, NH3/NOX ratio,
NOX concentration at the SCR inlet, gas flow rate, and catalyst
condition. Low temperatures result in a failure or slowdown in
NOX reduction, which is a particular issue when the boiler is
operating at a low-load condition. Fundamentally, like all post-
combustion control devices, SCR is designed to respond to the
2-4
-------
characteristics of the stack gas. The primary difference between
utility and nonutility boiler types may be that, on average, non-
utility boilers may be more likely to operate with fluctuating
loads. This difference in operating pattern may appear to have
an impact on the characteristics of the stack gas. However, the
NSPS is based on a 30-day averaging period to accommodate normal
fluctuations in performance. Further, as discussed above, new
analyses of two facilities that operate under cycling conditions
have shown that SCR can meet the proposed standard over a 30-day
averaging period. The Birchwood facility reports daily cycle
variations from 32 percent to 100 percent of load. The Logan
facility's daily cycles ranged from 28 percent to 84 percent in
the 3-month period for which data were supplied.
Another load-related technical issue raised in the case of
pulp and paper is the difficulty in maintaining the temperatures
necessary to minimize NOX and HAP generation. In general, while
designing an SCR system for a boiler, the boiler duty is taken
into consideration. Specifically, the expected temperature range
at the exit of the economizer is factored into the selection of
an SCR catalyst formulation.
There are other steps that operators can take to ensure the
desired SCR performance under variable or low load conditions.
For example, if low load contributes to insufficient gas velocity
to keep the flyash in suspension, the operator can add an ash
hopper to divert the ash from the reactor and catalyst face.
Alternatively, good ductwork system design can avoid these
problems. Also, low boiler exit temperatures can be avoided by
adding a economizer by-pass to keep the gas temperature higher at
low loads. Finally, good flue gas mixing can overcome
differences in gas flows and boiler firing conditions.
(Robinson, T. And Croteau, P., "Adapting the German Coal-Fired
SCR Experience to the U.S." Presented at the Council of
2-5
-------
Industrial Boiler Owners NOX Control XI Conference, February
1998.) Taking into consideration all of the above, in general,
the EPA does not believe that SCR use is constrained by boiler
duty.
Several commenters raised catalyst poisoning as an
illustration that SCR is not suitable for all units. As a result
of developments in catalyst technology, formulations are
currently available that minimize the impact of poisoning.
Nevertheless, the EPA believes this issue is really related to
the cost of operating the SCR; catalyst; appropriate catalyst
management plans now make it possible to maximize catalyst life
under plant operating conditions.
Another issue raised by commenters is that the SCR
technology is immature and insufficiently demonstrated. The EPA
disagrees with this comment. One recent study (Khan, S., et al,
"SCR Applications: Addressing Coal Characteristic Concerns."
Presented at the EPRI-DOE-EPA Combined Utility Air Pollutant
Control Symposium, August 1997) identified at least 212 worldwide
SCR installations on coal-fired units, which cover different
types of boilers subjected to varying operating conditions and
firing a variety of coals. Some of these installations were
designed for and have achieved high NOX reduction levels,
exceeding 90 percent. Plants in Europe have been continuously
using SCR for over 10 years. (Robinson, T., and Croteau, P.,
"Adapting the German-Coal-fired SCR Experience in the U.S."
Presented at Power-Gen International 97, December 1997.)
Finally, SCR-equipped units located in the U.S., such as the
Logan, Birchwood, and Stanton facilities are meeting some of the
most stringent NOX limits in the country.
Comment: Coal-related issues. Commenter IV-D-47 provided a
rigorous description of what would legally be considered
"adequate demonstration", and concluded that the proposed NSPS
2-6
-------
are not adequately demonstrated for all U.S. coals, particularly
medium- and high-sulfur coals. Previously, the commenter had
urged the EPA not to base the NSPS on undocumented experience in
Germany or Japan. The commenter also rejected the Department of
Energy Plant Crist high-sulfur coal demonstration project because
of its limited scope. Commenter IV-D-60 reported the same
comment.
Additionally, Commenter IV-D-47 claimed that EPA's analysis
of U.S. coal usage is misleading. The commenter noted that "EPA
claims that high-sulfur coal technical issues are irrelevant,
because 85% of the coal fired in this nation has 2% or less
sulfur content." The EPA reported coal consumption on a mass
basis, which is biased toward high moisture, high ash content
coal. The commenter indicated that coal use should be depicted
on a Btu basis. Additionally, the commenter stated that coal use
should be described on a regional basis. The commenter claimed
that an analysis based on heat value and regional consumption
would show that 48 percent of the coal burned east of the
Mississippi is high-sulfur coal.
Commenter IV-D-63 added that the Japanese demonstration is
on low-dust environment using a hot-side electrostatic
precipitator (ESP), compared to most of the U.S. boilers which
use cold-side ESP's.
Response: The EPA disagrees that the use of SCR for high-
sulfur coal applications is unsupported. As noted in the Acid
Rain Phase II NOX Rule Response to Comments Document (p. 171), in
addition to one coal-fired plant in Japan and another in Austria
firing coals with sulfur contents of 2.5 percent of higher, there
are two coal-fired SCR installations in the U.S., Chamber Works
and Keystone Plants in New Jersey, that are firing coals with
sulfur contents close to 2 percent. Northampton, which is
equipped with SNCR, successfully burns waste coal, and meets some
2-7
-------
of the most stringent NOX limits in the U.S. (0.10 Ib/MMBtu). In
the Plant Crist demonstration project, catalysts from various
suppliers performed successfully. Criteria for successful
performance at this demonstration included ammonia slip less than
5 ppm and S02 oxidation less than 0.75 percent.
In view of the experience both in the U.S. and abroad, the
commenters' concerns over the use of SCR for high-sulfur coal
applications is unsupported. In general for these installations,
design features such as low ammonia slip, a catalyst that
minimizes S03 conversion, and an economizer bypass to maintain
proper flue gas temperatures at low loads are provided.
The commenters said that the NSPS was not adequately
demonstrated for the range of U.S. coals, particularly medium-
and high-sulfur coals and that EPA's analysis of U.S. coal usage
is misleading. First, EPA's analysis did specifically address
medium- and high-sulfur coals. For example, page 6-12 of Subpart
Da's BID estimates an indirect cost factor of 1.45, where it
stated that "For the application of SCR to boilers burning
medium- to high-sulfur coals, indirect costs may be greater than
45 percent of the process capital due to factors discussed in
Chapter 3." In any case the key issue is the impact that burning
lower grade, higher sulfur coals has on SCR performance. Once
again, this is more of a cost issue than a performance issue,
because the major effect of burning some coal types is that the
SCR catalyst may wear out more quickly or that problems such as
plugging of the catalyst or additional cleaning requirements may
add to the costs of using SCR in some applications. This issue
is discussed further in section 3.3.
One commenter also says that the Japanese demonstration is
on low-dust environment using a hot-side ESP, compared to most
U.S. boilers, which use cold-side ESPs. Once again, this is a
catalyst life issue.
2-8
-------
2.2 SELECTIVE NONCATALYTIC REDUCTION (SNCR)
Other commenters argued that SNCR was not adequately
demonstrated.
Comment: Fluidized bed combustion boilers (FBCs).
Commenter IV-D-56 reported that SNCR has not been adequately
demonstrated for use on circulating FBCs. Commenter IV-D-56
added that "due to the inherently low combustion temperature of
circulating FBCs, SNCR cannot work properly on that type of
boiler unless the boiler is operating at its maximum capacity
rate." The commenter explained that the flue gas must be,
generally, between 1700°F and 1800°F, in order for the chemical
reaction that removes NOX to occur. Further, Commenter IV-D-56
reviewed the EPA questionnaire and found that three of the five
circulating FCBs that use SNCR stated that SNCR did not work
properly when the units were operated at anything less than
maximum capacity. Commenter IV-D-56 concluded this discussion by
stating that the "EPA has no basis whatsoever for extrapolating
data obtained from the application of SNCR to other types of
boilers to conclude that SNCR is appropriate emission control
technology for circulating FBC boilers that cycle their load."
Large boilers. Commenter IV-D-56 commented that SNCR "has
not been adequately demonstrated to work on large boilers [with a
rated capacity greater than 390 MMBtu/hr], whether circulating
bed or not." Commenter IV-D-56 reviewed the data and reported
that the rated capacity of the FBC boilers using SNCR that were
analyzed by the EPA in developing the proposal ranged from 389
MMBtu/hr to 290 MMBtu/hr. The commenter concluded by stating
that SNCR cannot be considered an adequately demonstrated
emission control technology for FBC boilers greater than 390
MMBtu/hr rated capacity.
Response: According to the subpart Db BID (p. 3-43), flue
gas temperatures exiting the furnace can range from 1,200 °C ±
2-9
-------
110 °C (2,200 °F ± 200 °F) at full load down to 1,040 °C ± 70 °C
(1,900 °F ± 125 °F) at half load. At similar loads, temperatures
can increase by as much as 30 to 60 °C (50 to 110 °F) depending
on the extent of ash deposition on heat transfer surfaces. Due
to these variations in the temperatures, it is often necessary to
inject the reagent at different locations or levels in the upper
furnace or convective pass for effective NOX reduction. A recent
publication summarized the successful retrofit of retractable
lances on a 100 MWe coal-fired utility boiler equipped with SNCR,
which greatly improved low load performance. (Hunt, T. , et. al,
"Using Retractable Lances to Maximize SNCR Performance."
Presented at the EPRI-DOE-EPA Combined Utility Air Pollutant
Control Symposium, August 1997) Finally, as noted in the subpart
Db BID, the addition of hydrogen or other hydrocarbon reducing
agent can be injected with the NH3 to lower the effective
temperature range. Similarly, additives can increase the
temperature range of urea application. By taking these sorts of
steps, the EPA believes that operators can successfully operate
SNCR, even under low load conditions.
Recent analysis of NOX emissions data from a 110-megawatt,
base-loaded, circulating fluidized-bed boiler equipped with SNCR
(U.S. Generating Company's Northampton plant) indicates that the
facility is quite capable of meeting the proposed standard. This
facility achieves average input-based emissions of 0.089 Ib/MMBtu
and output-based emissions of less than 0.8 Ib/MWh, well below
the output-based standard of 1.6 Ib/MWh.
Regarding SNCR on large boilers, the Acid Rain Phase II NOX
Response to Comments Document (p. 212) notes that SNCR has been
demonstrated on coal-fired units as large as 1,230 MMBtu/hr
(Germany) and on oil-fired units as large as 2,900 MMBtu/hr
(Niagara Mohawk's Oswego Station). The SNCR application on
Oswego shows that injectors can effectively penetrate the
2-10
-------
combustion gas flow in large boilers. Since the effectiveness of
injecting SNCR reagent into large boiler casings has been proven,
and SNCR has been applied to a variety of boilers, the EPA does
not see boiler size as a restriction for applying SNCR to NSPS
sources.
2.3 NATURAL GAS REBURN
Comment: Commenters IV-D-19, IV-D-20, IV-D-61 and IV-D-65
recommended that the EPA recognize natural gas reburn, as well as
SCR, as the demonstrated technology basis for Subpart Da sources.
The commenters pointed out that this approach would be consistent
with the Acid Rain NOX programs.
Commenter IV-D-61 listed several advantages of reburn
technology over the add-on controls proposed by the EPA:
(1) lower capital costs (one-third to one-half of SCR);
(2) minimal boiler modifications required;
(3) lower maintenance requirements;
(4) no costly catalysts;
(5) no downtime for catalyst replacement, and;
(6) demonstrated effectiveness on coal-fired industrial
boilers in the U.S.
The commenter said that EPA's assertion that the maximum
potential emission reductions from this technology is only 50
percent, and therefore less than the other technologies
considered, is in error. The commenter said that there are at
least five commercial installations of the reburn technology in
the U.S., and they are achieving NOX emission reductions of 58-77
percent. In addition, coal, natural gas and other fuels can be
utilized in the fuel rich zone.
Response: Commenter IV-D-61 refers to a paper from the
August 1997 EPRI megasymposium, "Field Experience--Reburn NOx
Control," which presents the results of five full-scale retrofit
applications of reburn technology. The paper describes design
2-11
-------
considerations and advantages of the technology. One unit is
being equipped with a nitrogen agent injection system, which is
expected to reach of goal of total NOx of 0.15 Ib/million Btu (to
date has achieved 0.2 Ib/million Btu). Other commenters were
worried that EPA's apparent exclusion of reburn is based on
faulty rationale and contradicts acid rain rulemakings, which
place SCR and reburn on the same level of effectiveness.
The EPA agrees that reburn technology may be a viable
alternative to SCR in some situations. As structured, the NSPS
would not preclude application of this or other innovative
technologies, so long as they meet the emission standard.
However, the EPA believes the existing analysis supporting SCR is
adequate for purposes of supporting the selection of BDT.
2-12
-------
3.0 CONTROL TECHNOLOGY COSTS
Several comments addressed the cost analysis performed by
the EPA in support of the proposed standards.
3.1 ESTIMATED COSTS ARE TOO HIGH
Comment: Commenter IV-D-39 remarked that the EPA's NOX
control costs are too high and gave two reasons why: (1) the
control costs should not be based on add-on control technologies;
(2) the utilization of the industrial boilers is underestimated.
Commenter IV-D-57 asserted that the EPA has overestimated
the cost of post-combustion NOX controls, and that aggregate
costs of the proposed standards would be less than the EPA
estimates. The EPA cites costs of 2.1-3.3 mills/kWh and cost-
effectiveness estimates of $1,460-2,270/ton for SCR on coal-fired
electric utility boilers. The commenter cites one unit where
actual SCR costs are 0.98 mills/kWh and approximately $l,200/ton.
The commenter also states that the EPA has not considered recent
strides in reducing reagent use, and operating cost, for SNCR
installations. The commenter refers to one coal-fired utility
boiler that reduced reagent use by 50 percent through a control
upgrade, including continuous ammonia and temperature monitors,
improved control hardware and software, and additional injector
pressure controls.
Response: The EPA considered both the use of add-on
controls and process modifications, including fuel switching, at
proposal. That analysis showed that add-on control technology
represented BDT in this case. The EPA's analysis did consider
the utilization rate of industrial boilers, which contributed to
the selection of a higher emission limit (0.2 Ib/MMBtu vs. 0.15
Ib/MMBtu). As for commenter IV-D-57's example, EPA expects that
costs of operating SCR will decrease as facilities gain
experience in maintaining and operating these units. However,
3-1
-------
EPA believes the overall cost analysis presented at proposal
fairly represents average costs to the industry.
3.2 ESTIMATED COSTS ARE REASONABLE
Comment: Commenters IV-D-19, IV-D-20, IV-D-26 and IV-D-65
voiced support of the Agency's conclusion that additional
controls for new gas-fired and distillate oil-fired units are not
cost effective. Commenters IV-D-19, IV-D-26 and IV-D-65 added
"that is not clear whether EPA has taken into account the cost of
"scope adders" in the construction of a new boiler or
reconstruction of an existing one." Both commenters explained
that the "scope" of the project reflects reconstruction of the
boiler and "scope adders" may include significant site work,
rerouting of lines, relocation of other equipment, and/or the
costs of shutting down production. The commenters added that
these costs may add 100 percent to the costs of simply
constructing or reconstructing the [same] boiler (at a different
site).
Response: The Agency appreciates the feedback from the
commenters. Retrofit costs were included in the cost estimation,
as noted in Docket Item II-A-21, App. A, page 4-3.
3.3 ESTIMATED COSTS ARE TOO LOW
Comment: Commenter IV-D-47 contended that the EPA cost
estimates for SCR in the proposed rule were much too low, and
that the cost analysis was inadequate. The commenter stated that
the EPA extrapolated their costs "from an earlier study that had
very different technical premises for SCR." This earlier study
focused on retrofit costs for existing plants, and did not
consider site layout, with boiler conditions not typical of new
units. The commenter reported that the EPA estimates of SCR
capital costs are only 65 percent of recently estimated values
that were summarized at an EPA-Department of Energy (DOE)-
Electric Power Research Institute (EPRI) technical conference.
3-2
-------
The commenter asserted that the "EPA has not met the requirements
of §111 and should withdraw its NSPS proposal." The commenter
recommended that the EPA analyze options other than SCR for coal-
fired boilers, and then "determine whether an SCR standard for
coal-fired units is appropriate."
Commenter IV-D-37 reported that SCR systems require more
energy to operate due to a pressure drop associated with the
catalyst bed. The commenter stated that "by effectively
requiring the use of SCR for solid fuel-fired units, EPA is
encouraging the use of an energy-intensive emission control
method to achieve marginally lower NOX emissions..." In addition
to the additional operational costs, Commenter IV-D-37 reported
that there is fouling of air heater surfaces by ammonium salts,
and waste disposal costs for the spent catalyst.
Commenters IV-D-23 and IV-D-45 pointed out several costs
associated with SCR and fuel switching for coal-fired industrial
boilers that were not considered by the EPA cost estimates.
Commenter IV-D-23 provided items (1) through (5) and Commenter
IV-D-45 provided item (6).
(1) There are several components in U.S. coals (e.g.,
alkaline metals, heavy metals, chlorine, and fluorine) that
could significantly shorten the catalyst life. The EPA
estimate assumed a catalyst life of 5 years. If the
lifespan is reduced to 2-3 years, the effect is a doubling
in the cost of the catalyst, which is already estimated to
be 30 percent of the SCR cost with a 5-year lifespan.
(2) Ash from SCR installations will have different
characteristics, such as higher nitrogen content, and may
have additional regulatory requirements and costs.
(3) Sulfur in coal and ammonia from the SCR can react to
form ammonium bisulfate, which can plug air heaters. Other
calcium and ammonium salts can foul the catalyst. The
3-3
-------
commenter stated that the costs of these maintenance
problems were apparently not considered in the proposal.
(4) The costs associated with storing large quantities of
ammonia for the SCR were not adequately considered. Health
and safety hazards and the potential for leaks will
necessitate alarm systems and evacuation plans.
Additionally, ammonia can cause "fogging" of photographic
film, so SCR would be highly undesirable at film
manufacturing sites.
(5) SCR requires significant open space for the catalyst
bed. The commenter believed that newly constructed units
could accommodate SCR through advance planning. However,
for existing units located in dense industrial facilities,
the lack of space presents a technical feasibility issue.
(6) The EPA did not consider the significant costs
associated with handling and disposal of spent ammonia
catalyst from SCR installations.
Commenter IV-D-45 also wrote that the EPA's estimate that
SCR would cost $2,000/ton was "significantly understated." The
commenter explained that most paper mills have smaller sized and
lower capacity boilers than electric utility units. These
smaller boilers are expected to have a much higher cost per ton
of NOX reduction associated with SCR. The commenter stated that
"the estimate of $2,000/ton is still too high to be considered
cost-effective for control of a criteria pollutant like NOX."
Commenters IV-D-26 and IV-D-61 did not agree with the cost-
effectiveness values that the EPA calculated for NOX control
technology for coal-fired industrial boilers. The commenters
cited a best control technology (BACT) analysis conducted by a
State regulatory agency for a Prevention of Significant
Deterioration Application written in 1992. Commenter IV-D-26
quoted the report to state: "in transferring SCR technology from
3-4
-------
the commercial applications in Japan to European sources,
technical problems arose..." Commenters IV-D-26 and IV-D-61
quoted "it can be expected that almost every application will
need to be verified in a pilot plant... Therefore, SCR is
presently not considered technically feasible, and thus not BACT,
for the proposed project." Commenter IV-D-26 summarized the
following technical concerns that resulted from the 1992 BACT
analysis:
(1) Catalyst costs are "over half of the operating and
maintenance costs..."
(2) The reaction of S03 with ammonia to form ammonium
bisulfate, which in turn can foul the catalyst and
downstream equipment, and;
(3) The difference in coal characteristics between foreign
and domestic coals.
In addition to the citations from the 1992 BACT analysis,
Commenter IV-D-26 reported the following concerns with the EPA's
own evaluation of the use of SCR for NOX control for coal-fired
industrial boilers.
(1) The Agency's analysis was based on seven coal-fired
utility boilers with SCR in the Eastern United States only.
The Agency did not demonstrate a solution on industrial
boilers for all of the coal characteristics that will be
encountered in the United States.
(2) Technologies utilized for utility boilers, especially
multiple control devices that increase the risk of
breakdowns, are not always directly transferable to
industrial boilers. This is because of the different
operating and maintenance practices between the two sources.
Commenter IV-D-26 explained that the unplanned shutdown for
a utility boiler can be managed by shifting the electric
power generation to other available units of power, which
3-5
-------
can be bought from a regional power grid. However, in the
case of industrial boilers, the steam demand of the chemical
manufacturing operation cannot be met, which results in
shutting down the chemical operations and in considerable
economic penalty.
Commenter IV-D-16 added that the EPA has not demonstrated
anywhere that SCR or SNCR can be used cost-effectively with North
Dakota lignite. The commenter recommended that the EPA not
proceed with the rulemaking process until it can adequately
demonstrate that SCR and SNCR technologies are cost effective
with a variety of coals including lignite.
Commenter IV-D-53 stated that the cost-effectiveness value
for coal units using SCR is calculated using a baseline NOX
emission rate of 0.45 Ib/MMBtu, which in turn, "artificially
inflates" the cost effectiveness of SCR for new coal fired units.
The commenter stated that low NOX burners can "easily meet" a
0.30 Ib/MMBtu NOX emission rate. The commenter recommended a
higher emission standard in the range of 0.20 to 0.25 Ib/MMBTU
for all fuel sources.
Response: Several commenter's said that the EPA's cost
estimates understated SCR costs and failed to represent the range
of boiler conditions, particularly industrial boilers, in the
U.S. The Agency is satisfied that the proposal cost analysis
adequately represents the average nationwide costs to comply with
BDT for new sources, and has not revised the analysis at
promulgation. However, the Agency will take this opportunity to
respond to the less-generic comments summarized above.
Commenter IV-D-47 speculated that the EPA extrapolated cost
from an earlier study and did not account for all the capital
costs of the SCR system. Please note that BID cost estimates
were revised in a memorandum dated June 10, 1997 explaining that
the costs were based on more recent information obtained by the
3-6
-------
Acid Rain Division's cost estimates from their draft report
entitled "Cost Estimates for Selected Applications of NOX Control
Technologies on Stationary Combustion Boilers." The EPA cost
estimates for SCR used in the impacts analysis, and summarized in
the preamble, were different than those used in the BID. The
costs in the preamble and impacts memoranda (Docket Items, II-B-
8, II-B-9, and II-B-10) were made using actual baseline emissions
from the planned, new units in the country.
In response to the commenter who said that SCR is too energy
intensive, the EPA notes that a detailed regulatory impact
analysis was performed. The EPA believes that the energy impacts
of SCR, which are only 0.4 percent of the boiler output, are
justified.
The Agency offers the following response to address
Commenters' IV-D-23 and IV-D-45 six items:
(1) The assertion that EPA based costs on a 5-year catalyst
life is incorrect. The EPA used 3 years for coal-fired units.
(2) The Agency realizes that ash from SCR installations
will have different characteristics (higher nitrogen content) and
additional regulatory requirements and costs. The Agency did
account for different types of coal, with varying ash contents,
in the costing analysis.
(3) The Agency realizes that there are downstream effects
from SCR. The cost estimates included indirect costs that
accounted for these effects. Further, since proposal, the Agency
has received cost estimates from two facilities with SCR that
validate the Agency estimates. The indirect costs of SCR
maintenance, ammonia and catalyst management were estimated to be
approximately 1 percent of EPA's total SCR capital costs. (Memo
to Project File, "Indirect SCR Costs.")
(4) The additional storage costs for ammonia were
considered in the indirect costs of SNCR and SCR. Because
3-7
-------
anhydrous ammonia has been used safely for many years in the U.S.
in a variety of industrial and agricultural applications, the EPA
believes that any safety, environmental, or operational concerns
can be fully addressed by proper planning and design of the
ammonia handling system. These concerns are not a factor against
adopting the emission limits that are based on SCR.
(5) This regarding space constraints is similar to
Commenters' IV-D-19, IV-D-20, IV-D-26 and IV-D-65 comment about
"scope adders" addressed above.
(6) Spent catalyst costs were also addressed in the
indirect costs of SCR (BID for Subpart Db , page 6-11)
Commenter IV-D-26 raised concerns about the ability of
industrial units to operate reliably when equipped with SCR and
the resulting cost impacts of downtime due to control device
malfunctions. While the Agency realizes that control devices do
malfunction, and in fact, accounted for extra maintenance costs
of SCR, both process and control device malfunctions are a fact
of life in any complex operation. This is why many facilities
are equipped with back-up or standby boilers. In the case of a
malfunction, the NSPS provisions would not apply during the
period of the malfunction, assuming the source acted to repair
the malfunction as soon as practicable.
With respect to Commenter IV-D-16's comment regarding the
use of SCR or SNCR with North Dakota lignite, the EPA's cost
estimates did project costs for lignite use and did not find its
impacts different from the impacts of using different coal types.
(Docket Item No. II-A-33.)
Regarding baseline emission rate, model plants used a higher
emission rate (0.45.b/MMBtu), but the impacts analysis presented
in the preamble used emission rates based on projected permit
limits, which are lower. There is also a tradeoff in assuming a
higher emission rate compared to a lower rate when looking at
3-8
-------
cost effectiveness. Higher baseline emissions would require a
larger SCR and more catalyst, which would increase the cost side
of the equation. Lower baseline emissions would require a
smaller SCR, but would result in lower potential emission
reductions from the controls.
Comment: Commenter IV-D-52 argued that EPA provided an
inadequate basis for its conclusions, particularly the EPA's
assumption that because gas- and oil-fired units are not expected
to need SCR, it was not costed for these units. The commenter
took exception to this assumption because "insufficient data is
presented to warrant such assumptions and exclusions." The
commenter also stated that EPA must provide complete cost and
performance models.
Response: As stated previously, the EPA is satisfied that
the proposal cost analysis adequately represents the average
nationwide costs to comply with BDT. Because gas and oil-fired
units should be able to perform close to or at the NSPS emission
limits (particularly in the case of industrial boilers) with low-
NOX burners or other combustion controls, the basis for the EPA's
statement in the preamble was the EPA's assumption that these
units would choose to either use SNCR to meet the limit or to
simply improve the efficiency of their existing systems. The EPA
did cost SNCR and SCR for oil- and gas-fired utility units in the
proposal Background Information Documents, but since the EPA was
implementing the philosophy of a "fuel neutral" approach the cost
effectiveness calculations were conducted based on projected
coal-fired steam generating units using coal.
3.4 OTHER COST ISSUES
3.4.1 Fuel Switching Costs
Comment: Commenter IV-D-23 noted that the costs of fuel
switching were not analyzed in the proposal. The commenter
estimated that natural gas costs more than twice as much as coal
3-9
-------
(on a Btu basis) when purchased on a "curtailment basis" (which
favors residential customers over industrial customers during
shortages). Natural gas supplied on a "non-curtailment basis"
would be prohibitively expensive. The commenter advised the EPA
that capital retrofit costs to accommodate a switch from coal to
natural gas may be significant for some industrial units.
Response: The fuel neutral format of the proposed standards
would allow for the use of natural gas, but would not require it
in cases where the costs of using natural gas exceeded the costs
of meeting the standard using alternative means, i.e., the
application of SCR or other similar technology.
3.4.2 Energy Pricing
Comment: Commenters IV-D-24 and IV-D-42 noted that the
EPA's economic analysis used an electricity rate structure with
average costs because many corporations own multiple facilities.
Both commenters agreed that with utility deregulation and
restructuring in the near future, the averaging of costs over
several facilities is outdated, unreasonable, and unacceptable.
Additionally, Commenter IV-D-42 stated that "even if EPA's
analysis were correct, that total annualized costs are 2.1-3.3
mills/kWh for SCR on a coal-fired unit, this does not justify
requiring such technology based on cost."
Response: The Agency's economic analysis used projected
energy rates from the DOE's Energy Information Administration for
1996 through 2000 to serve as the baseline and projected the
incremental increase in electricity rates for each year to be
equal to the weighted average of compliance costs across affected
utility boilers. This national-level approach does not account
for the more local nature of markets under regulated monopoly or
deregulation. However, the uncertainty regarding which customers
would be subject to higher rates and the future competition in
electricity provision necessitated this national-level analysis.
3-10
-------
In Section 5 of the RIA, the Agency states that its approach
"will understate the expected increase in market prices under
regulated monopoly and understate or overstate expected price
increases for specific customers due to the use of average
national price of electricity and measure of compliance costs."
The total annualized costs from EPA's engineering cost analysis
vary from 0.13 to 2.9 mills/kWh as compared to the projected
national electricity price of 69.0 mills/kWh. Although beyond
the scope of the economic analysis, the Agency does expect that
price changes in local markets under regulated monopoly or
deregulation will vary according to the actual costs incurred by
the utility boiler servings these markets.
3.4.3 Proposed Standards Not Cost Effective
Comment: Commenter IV-D-48 noted that the incremental cost-
effectiveness numbers for industrial, spreader-stoker coal-fired
boilers with SCR and combustion controls versus combustion
controls alone are exorbitant and totally unjustified. The
commenter recommended that the standard be revised to impose
limits which reflect demonstrated technology at a reasonable
cost.
Commenter IV-D-26 quoted cost-effectiveness values for NOX
removal with SCR from a BACT analysis conducted in 1992 to be
$ll,541/ton of NOX removed, as compared to the range quoted in
Table 3 of the NSPS, which was $1,460-2,270. The commenter
concluded that the costs of the SCR for NOX control for
industrial coal-fired boilers is currently not cost effective.
Commenter IV-D-61 stated that the docket does not support
the EPA's conclusion that SCR is cost effective for coal-fired
units. Table 4 (62 FR 36953) shows the range of cost
effectiveness for SCR on industrial units as $0 - $4,800 per ton
NOX removed. The average is shown as $2,030. However, Table 3
(62 FR 36951) shows the range of cost effectiveness for coal-
3-11
-------
fired industrial units as $1,590 - $8,700 per ton. The commenter
was unable to locate the basis for the ranges in Table 3. In the
background document, New Source Performance Standard, Subpart Db
- Technical Support for Proposed Revisions to NOX Standard, Table
C-l shows a range of $2,780 - $29,950 per ton for incremental
cost effectiveness for combustion controls and SCR versus
combustion controls alone at coal-fired model boilers. The
commenter noted that the values in Table 3 (62 FR 36951) appear
to have been taken in error from the "Cost Effectiveness" column
rather that the "Incremental Cost Effectiveness" column in Table
C-l. The commenter also pointed out that the values in Table 4
(62 FR 36953) are based on an estimate of 381 new industrial
units in the next 5 years, and that only 22 of these were
projected to be coal-fired. Because the gas and distillate oil
units would not have any control costs, the low average of $2,030
obscures the high costs incurred by the few coal-fired units.
Response: Commenter IV-D-48 listed as exorbitant all of the
incremental cost-effectiveness (CE) numbers for industrial coal-
fired boilers with SCR and combustion controls versus all the
other possible control options. However, only the incremental CE
of SCR with combustion controls versus combustion controls alone
is relevant, because the baseline level of control is combustion
controls. Therefore this scenario is the appropriate basis for
comparison.
Commenter IV-D-26 compared the CE value from a 1992 BACT
analysis conducted by a State agency to that calculated by the
EPA for this NSPS revision. The EPA stands behind its original
CE calculation and does not deem the CE to be exorbitant.
Commenter IV-D-61's report of being unable to locate the
basis for the ranges in Table 3 is understandable. The EPA
realizes that the derivation of the values in Table 3 of the
preamble may not have been documented adequately for the
3-12
-------
proposal. The CE values summarized in Table 3 of the preamble
originated from the impacts memo for subpart Db, Docket Item II-
B-9, Table 8. The cost data in the BID Tables C-l and C-2 are
for the range of boiler sizes and capacity factors. The EPA
determined that the capacity factor of 0.1 was too small for
coal-fired boilers, therefore the cost-effectiveness values from
all coal-fired boilers with this capacity factor were not used in
the cost analysis or summarized in the Table 3 of the preamble.
Commenter IV-D-61's speculation that the numbers in Table 3
of the Preamble are the "Cost Effectiveness" values from the BID
instead of the "Incremental Cost Effectiveness" values was
incorrect, even though it may have appeared that way. The EPA
used the overall CE numbers with the current NSPS level of
control as the baseline.
Commenter IV-D-61 noted that the CE values in Table 4
obscure the larger coal CE values because they represent a
smaller portion of the new boiler projections. The high numbers
in Table 4 are CE values for oil-fired boilers where the CE
values for coal-fired boilers are within the range listed in
Table 4.
Comment: Commenter IV-D-31 stated that EPA addressed coal-
fired power plants that use SCR technology for the cost analysis,
but failed to acknowledge other high performance power plants
that do not. The commenter cited the example of the Neil Simpson
II unit, which is an 80-MW conventional boiler that has only low-
NOx burners and achieved an output-based emission rate of 0.18
Ib/million Btu in the second quarter of 1997. Commenter IV-D-31
calculated the cost effectiveness of the Neil Simpson II unit
using the EPA's cost estimate and the operational data from the
second quarter of 1997, which was extrapolated to one year. The
commenter reported that "the results of this analysis showed that
the incremental cost of reducing NOX would be in the range of
3-13
-------
$10,600 to $16,625 per ton. The commenter quoted the President's
address on the Implementation of Revised Air Quality Standards
for Ozone and Particulate Matter when he said: "It was agreed
that $10,000 per ton of emission reduction is the high end of the
range of reasonable cost to impose on a source."
Response: The EPA appreciates the data provided by the
Commenter IV-D-31, however, the EPA did cost a comparably sized
unit. In the National Impacts Memorandum, Docket item II-B-8,
the EPA estimated the cost impacts of controlling NOX emissions
from an 80-MW boiler. The national impacts were calculated from
actual facility data, and the NOX emissions from this unit were
controlled with SNCR. The EPA would not assume that SCR would be
cost effective on such a small unit that is operating at low
boiler outlet NOX levels.
Comment: Commenter IV-D-31 stated that the cost
effectiveness of NOX controls on utility steam generating units
is incorrect. The commenter explained that the NSPS represents a
significant relaxation of standards from the NSR limits for
natural gas units; therefore, there should be no incremental or
annualized costs for these units. The commenter argues that the
appropriate baseline for the cost analysis should be the NSR
program, not the NOx levels being achieved with technologies to
meet the current NSPS.
Response: The EPA used two different baselines in its
analyses. In the model plant analysis the existing NSPS level of
control was the baseline used. However, for the impacts
analysis, the results of which are presented in the proposal
preamble, the baseline limits used were current/expected permit
limits, which were more stringent than the baseline limits used
for the model plant analysis presented in the BID. Further, new
units tend to have limits based on NSR decisions, and the EPA
believes that the limits used for the impacts analysis reflect
3-14
-------
NSR decisions in most cases.
3-15
-------
4.0 REGULATORY APPROACH
4.1 APPLICABILITY
4.1.1 "No New Exemptions" Policy
Comment: Commenter IV-D-49 recommended that the rule
"expressly state that [the] NSPS must be met in cases where a
utility plant is transferred to a new owner." Commenter IV-D-49
referred to this concept as "no new exemptions." This
recommendation was made in response to the deregulation of the
electricity production industry. The commenter speculated that
utilities will start to market electricity from "grand fathered"
(preexisting, and therefore exempt from the NSPS) power plants to
customers outside of their service territories, and, in some
cases, sell entire power plants to other utilities or independent
power producers.
Commenter IV-D-49 recommended that the EPA adopt a "no new
exemptions" policy in two areas:
(1) The rule should require utilities to count all emission
increases attributable to off-system sales when calculating
increased emissions associated with a major modification.
EPA currently exempts emission increases attributable to
increased demand in a utility's service territory, because
of the utility's obligation to serve. This rationale does
not apply in the case of off-system sales, which are wholly
discretionary. If a utility makes a major modification to
upgrade a plant to sell power outside of its service
territory, the costs of that decision should be borne by the
utility's stockholders -- not the environment. Requiring
such a utility to count all emission increases attributable
to off-system sales as increased emissions would trigger the
new NSPS for NOX, thereby preventing the utility from
externalizing at least some environmental costs associated
4-1
-------
with power plant NOX emissions.
(2) The rule should also expressly state that [the] NSPS
must be met in cases where a utility power plant is
transferred to a new owner. When a utility purchases a
power plant from another utility -- as is happening with
increasing frequency -- the power plant is "new" from the
perspective of the new owner, and [the] NSPS should apply.
In this type of situation, the new owner has a choice
between purchasing a power plant or building a power plant.
In the latter case, the plant would have to meet [the] NSPS.
To exempt the former "new source" from [the] NSPS would be
contrary to the express purpose of Title IV of the Clean Air
Act, to reduce the adverse effects of NOX emissions from
fossil fuel combustion by implementing standards of
performance that reflect improvements in methods for
reduction of NOX emissions.
Response: The Clean Air Act itself limits the scope of
changes considered modification to "any physical change in, or
change in the method of operation of, a stationary source which
increases the amount of any air pollutant emitted by such source
or which results in the emission of any air pollutant not
previously emitted." [Section 111(c)(4)] Section 60.14 of the
Subpart A General Provisions provides additional guidance on
EPA's interpretation of this definition, and specifically
excludes changes in ownership of an existing facility from being
considered a modification. In addition, a key aspect to the
definition of modification is that the change to the facility
must result in an emissions increase. If the owner or operator
can offset the increase, an NSPS modification is not established.
The commenter also noted that the EPA currently exempts
emission increases attributable to increased demand in a
utility's service territory, because of the utility's obligation
4-2
-------
to serve . The commenter said that this "exemption" should not
apply in the case of off-system sales, which are wholly
discretionary. The EPA believes that the guiding exemption is
the ability of the source to increase its load in cases where:
(1) the increase can be accomplished without a capital
expenditure, (2) the increase results from an increase in the
hours of operation of the facility, or (3) the use of an
alternative fuel or other raw material takes place at a facility
that was previously designed to accommodate the alternative use.
None of these changes would be considered modifications.
4.1.2 NOX Emissions Limits for Existing Boilers
Comment: Commenters 1, 4, IV-D-05, and IV-D-07 stated that
the new source standards "do not address the overwhelming problem
of NOX emissions from utility power plants that were built prior
to 1977." The commenter noted that these sources constitute 70
percent of the utility fossil fuel plants. Commenter IV-D-05
added that an emission reduction from these source as proposed in
the NSPS would result in roughly a 75-percent reduction in NOX
emissions.
Commenter IV-D-50 recommended that the EPA explore how the
current proposal can be interfaced with existing units, which
operate with a wide range of efficiencies, have an extremely low
retirement rate, and will continue to generate most of the
electricity in the future.
Commenter IV-D-33 recommended that the output-based standard
should be applied to all existing facilities. The commenter
noted that output-based standards would promote economic
development by removing market barriers for new generators. The
commenter felt that an output-based standard combined with
emission allowance trading mechanisms would ensure cost-effective
emission reductions.
Response: The commenters' suggestions are beyond the scope
4-3
-------
of this rulemaking. In any case, NOx controls developed under
the acid rain program, the OTAG program, ozone SIP calls, etc.
are all designed to specifically address existing sources.
4.1.3 Existing Sources Should Be Exempt from NSPS
Comment: Commenters IV-D-22, IV-D-32, IV-D-35, IV-D-47, IV-
D-55, and IV-D-63 expressed opposition to the applicability of
the NSPS to modified units. Commenters IV-D-32 and IV-D-63 both
explained that in Section 111 of the CAA, "Congress was careful
to limit the applicability of NSPS to sources that could be
designed to include state-of-the-art pollution control
technology." Commenters IV-D-32 and IV-D-63 continued by
explaining that "NSPS were not made applicable to existing
sources because Congress recognized the difficulty and expense of
retrofitting control technology on such sources" adding that "the
capital costs of retrofitting SCR at existing natural gas or
coal-fired boilers are far more expensive than the costs of
deploying SCR at new natural gas or coal-fired boilers."
Commenters IV-D-47 and IV-D-55 agreed.
Commenter IV-D-41 said that the EPA was "acting unlawfully
by failing to consider the costs that will be incurred by
existing sources that become the subject of the proposed NOX
standard." The commenter proposed that existing coal-fired
sources are likely to become subject to this rule eventually,
unless they are specifically excluded. If this occurs, the
existing sources will be faced with excessive retrofit costs in
order to attain the standard. The commenter added that because
the proposed standards were not based on sound science, they
conflicted "with principles adopted by the President and Vice
President for Reinventing Environmental Regulation and endorsed
through the Administrator's Common Sense Initiative."
Commenter IV-D-55 stated that "the installation of SCR on
existing units... would be economically infeasible." A possible
4-4
-------
solution proposed by Commenter IV-D-55 was that the EPA propose a
standard that modified units could meet without SCR, or justify
the use of the same standards as for new units. Commenter IV-D-
47 reasoned that "since EPA states that few modified sources will
be affected, adding specific language clarifying that such units
are not subject to the NSPS would raise few, if any, policy
implications." Another possible solution presented was that the
EPA specifically exclude modified boilers from the final NSPS.
Commenters IV-D-16 and IV-D-17 stated that modified coal-
fired boilers should be explicitly excluded from the Subpart Da
standard. The reason reported was because the capital costs of
retrofitting SCR to an existing source is significantly more than
applying the SCR technology to a new source. Commenter IV-D-17
quoted the EPA's estimate that it is 27 percent less expensive to
outfit a new source compared to retrofitting an old plant.
Commenter IV-D-22 stated that the proposed NOX emission
limit was not demonstrated for non-gas-fired modified sources and
that the new limit should not apply to sources that come under
the NSPS through modification. In situations where liquid or
solid fuel is fired, it is not always possible or reasonable to
comply with the proposed limit. For instance, the commenter has
a residual oil-fired boiler that could not be retrofitted to meet
the proposed standard, and add-on controls would not be feasible
because of limited space and unreasonable cost.
Commenter IV-D-35 wrote that the EPA claimed this was not a
concern in the proposal. However, the commenter pointed out that
EPA is aggressively pursuing businesses that have made efficiency
improvements to force the units to meet NSPS under the
modification provisions in 40 CFR 60. The commenter stated that
the EPA "clearly has the discretion and duty to distinguish
between new and existing sources which become subject to this
rule."
4-5
-------
Response: As described in the response to the comment under
section 4.1.1, the General Provisions already provide several
limitations to changes that might be considered modifications.
For example, sources that offset their increased emissions are
not subject to the NSPS because of modification. These
provisions serve to effectively limit the application of the
modification provisions to existing sources.
Section 111(b)(1)(B) of the Clean Air Act (the "Act")
requires the Administrator to promulgate standards of performance
for "new sources" in each category of sources which in the
Administrator's judgment causes, or contributes significantly to,
air pollution which may reasonably be anticipated to endanger
public health or welfare. Section 111(a)(2) of the Act defines
"new source" to include stationary sources which are modified
after an applicable standard of performance is proposed. The EPA
finds nothing in the comments that would justify ignoring this
clear statutory mandate. In developing standards of performance,
section 111(a)(1) of the Act does, however, allow the
Administrator to take into consideration the cost of achieving
the required reduction and any nonair quality health and
environmental impact and energy requirements. As noted at
proposal, the efficiency of most existing electric utility steam
generating plants ranges from 24- to 38-percent efficient. The
EPA selected 38-percent efficiency as the baseline reflective of
NSPS units. The EPA believes that selecting the 38-percent
efficiency level for new electric utility steam generating units
was an appropriate exercise of its discretion based on the
available information. The EPA realizes, however, that existing
units are likely to operate in the lower end of this range, with
higher associated heat rates, which would make it more difficult
to meet an output-based standard. These sources would have to
compensate with higher control device performance (up to a 40-
4-6
-------
percent increase in performance), which would be more costly. As
a result, as discussed below in section 5.2.2, the EPA will allow
existing units that become subject to the NSPS because of
modification or reconstruction to meet an equivalent input-based
standard of 0.15 Ib/MMBtu. This change will eliminate the
concern that lower boiler efficiencies at existing units could
adversely affect a source's ability to meet an output-based
standard. This level of control represents the same overall
level of SCR performance that would be required of new units, but
lacks the benefits attributed to promoting energy efficiency that
the output-based format provides.
Comment: Commenter IV-D-42 expressed concern that the
current NSPS, new source review (NSR), and prevention of
significant deterioration (PSD) programs punish unit owners for
improving the efficiency and performance of existing units. The
commenter pointed out that if a coal-fired unit changed burner
systems to improve heat rate and annual availability, the owner
could be subject to NSR, technology analysis, preconstruction
delays, administrative costs, potential emission control
upgrades, emission offsets, and compliance with the proposed NSPS
limit. The commenter proposed that the EPA "couple the current
NSPS proposal for an efficiency-based standard with an
enforceable policy that physical changes to existing fossil-fuel-
fired steam generating units which result in a reduction in the
Ibs/MWh of pollutant emissions would not trigger NSR or PSD."
The commenter would support the efficiency-based standard if
efficiency upgrades for existing units were not penalized.
Response: A reduction in maximum hourly emissions would not
trigger the NSPS modification provisions. As for applicability
under NSR, the applicability criteria for utility boilers as well
as for other sources is the subject of an ongoing NSR rulemaking,
which was proposed on July 23, 1996 (see 61 FR 38250) . The
4-7
-------
comment period for that rule has closed, and the final rulemaking
will address the issue of NSR applicability for utility boilers
and other sources.
4.1.4 Modification Criteria
Comment: Commenter IV-D-06 noted that, in the applicability
exemptions to the rule, "no mention is made of routine
maintenance, repair and replacement." The commenter explained
that the routine replacement of boiler steam tubes may result in
increased efficiency. This increased efficiency may increase the
heat input capacity, and the hourly emissions, given the same
emissions rates. The commenter asked if this change would make
the boiler subject to the revised NSPS as a modification.
Response: The EPA, upon request, will determine the rule's
applicability on a case-by-case basis in accordance with the
requirements of sections 60.14 and 60.15 of the part 60 General
Provisions.
4.1.5 Applicability in NO.. Attainment Areas
Comment: Commenter IV-D-59 said that the new emission limit
is not needed in portions of the United States that already
comply with current air standards for NOX. Commenter IV-D-59
concluded that, in certain regions of the United States, the
proposed limits "do not result in any improvement in air
quality." Further, the commenter stated that the proposed rule
would penalize units which already "pay a production penalty due
to the installation of the same control equipment."
Response: The NSPS program is intended to be a national
program that serves, in part, to "level the playing field"
between similar sources and to contribute to nationwide
attainment (and maintenance of attainment) of the criteria
pollutants, of which NOX is one. In addition, in revising the
National Ambient Air Quality Standards (NAAQS) in July 1997, the
Agency recognized the regional role that NOX emissions play in
4-8
-------
ozone formation.
4.2 FUEL NEUTRAL APPROACH VERSUS SUBCATEGORIZATION
4.2.1 Support Fuel Neutral Approach
Comment: Commenters 1, IV-D-05 and IV-D-07 supported a cap
on NOX emissions at the same level for nearly all fuel types.
Commenters IV-D-05 and IV-D-07 reasoned that this allows fuel
switching as a control technology. Commenter 1 added that it was
an "important and positive step toward cleaner air ... across the
nation."
Commenters 4, IV-D-20, IV-D-25, IV-D-29, IV-D-44 and IV-D-46
also expressed support for the proposed fuel neutral standard.
The commenters stated that currently, natural gas-fired units are
subject to the most stringent standard while coal and residual
oil are allowed to emit much larger quantities of NOX. The
proposed rule will remove the disincentive toward natural gas
that has been created. One commenter wrote that a fuel neutral
standard would not penalize any particular industry, but would
encourage competition for new efficient boilers and cogeneration
units, and would be consistent with the EPA's emphasis on
pollution prevention.
Commenter IV-D-50 generally supported the fuel neutral
proposal because it "provides a level playing field for different
fuels and promotes the use of natural gas and clean oil-based
fuels, while at the same time it avoids unnecessary burdens on
coal-fired units." Commenters 4 and IV-D-29 added that a uniform
emission limit is needed to encourage fuel switching as a control
option.
Response: The Agency appreciates the commenters' support.
4.2.2 Oppose Fuel Neutral Approach
Comment: Commenters IV-D-17 and IV-D-31 opposed the same
NOX emission limit for all fuel types because "EPA's proposal
sets a lower than lowest achievable emission rate (LAER) level
4-9
-------
for coal-fired boilers, while significantly relaxing standards
for natural gas units by a factor of two to four times."
Commenter IV-D-31 noted that BACT for coal-fired boilers is
currently about 0.23-0.25 Ib NOx/MMBtu, and LAER is about 0.15
Ib/MMBtu; whereas the proposed standard appears to be 0.13-0.14
Ib/MMBtu. Further, for natural gas units, BACT is currently
around 0.07-0.08 Ib/MMBtu and LAER is on the order of 0.03 to
0.04 Ib/MMBtu.
Commenter IV-D-50 noted that the EPA is requiring much less
stringent control for gas- and oil-fired units. The commenter
pointed out that a number of gas- and oil-fired units in the U.S.
currently achieve approximately one-tenth of the proposed limit
with the application of SCR.
Commenters IV-D-16, IV-D-24, IV-D-28, IV-D-56 and IV-D-61
stated that the "proposal violates the Act by providing an
overwhelming incentive for new and modified electric generating
units to burn natural gas to the exclusion of coal." Commenter
IV-D-56 continued by stating that "the plain purpose of the
percent reduction requirement was to protect Appalachian and
Midwestern high sulfur coal...by requiring all new modified coal-
fired units to be scrubbed." Further, Commenter IV-D-56 reported
that the S02 allowance trading program created by the 1990
Amendments was intended in part to create flexibility for sources
to continue to use high sulfur coal..." Commenter IV-D-56
recommended that the EPA withdraw the proposal.
Commenter IV-D-24 pointed out that "a varied mix of energy
sources should be supported for the stability of the U.S.
generation system." The commenter stated that coal use should
not be discouraged, and that natural gas could meet a NOX limit
lower that those for other fuels.
Commenter IV-D-61 expressed opposition to the fuel neutral
approach because of fuel availability and cost factors. The
4-10
-------
commenter noted that the natural gas industry has not adequately
supplied areas of the U.S. that have an abundant supply of coal.
The commenter stated that natural gas is not uniformly
distributed and evenly available to all industrial users. The
commenter asserted that the proposed emission limit "favors
industrial development in regions that have an ample supply of
natural gas and penalizes regions that have no practical option
for steam production at industrial facilities other than coal."
Commenter IV-D-59 said that the fuel neutral emission rate
may inadvertently be a dis-benefit to the introduction of low NOX
technology. The commenter postulated that "the result then might
be continued operation of older more polluting sources than might
otherwise occur."
Response: The EPA disagrees with the commenters who contend
that the proposed fuel neutral format creates an overwhelming or
disproportionate incentive to use fuels other than coal. The
EPA's approach is designed to allow the continued use of coal as
a fuel in those cases where it is desirable. At the same time,
the standard would not preclude conversion to natural gas where
it makes sense in the individual application.
The EPA believes the fuel neutral approach would expand the
control options available to owners and operators by allowing the
use of clean fuels as a method for reducing NOX emissions. Since
projected new utility steam generating units are predominantly
coal-fired, the use of clean fuels (i.e., natural gas) as a
method of reducing NOX emissions from these coal-fired steam
generating units may give the regulated community a more cost-
effective option than the application of SCR for meeting the NOX
limit. Similarly, for industrial units, the use of clean fuels
as a method of reducing emissions may be a cost-effective
approach for coal-fired and residual oil-fired industrial steam
generating units.
4-11
-------
The fuel neutral approach also fits well with section
101(a)(3) of the Clean Air Act's emphasis on pollution
prevention, which is one of the EPA's highest priorities.
Because natural gas is essentially free of sulfur and nitrogen
and without inorganic matter typically present in coal and oil,
S02, NOX, inorganic particulate, and air toxic compound emissions
can be dramatically reduced, depending on the degree of natural
gas use. With these environmental advantages, gas-based control
techniques should be viewed as a sound alternative to flue gas
treatment technologies for coal or oil burning.
Finally, the proposed amendments do not relax the existing
NSPS for natural gas units. In fact, the 0.15 Ib/million Btu
heat input reflects a 50- and 25-percent reduction in NOX
emissions over the current Subpart Da limits for oil-fired and
gas-fired units, respectively. Revised Subpart Db would not
require any additional controls for new gas-fired and distillate
oil-fired units over the current NSPS because of the costs
associated with additional controls. However, subpart Db does
not relax the existing standards for these units either.
Historically, projections for new utility boilers have tended to
be for coal-fired units. Stricter NOX controls for gas might
make co-firing less attractive, while a fuel neutral approach
facilitates adoption of some natural gas firing, which has
environmental and other benefits over straight coal-burning
units.
4.2.3 Distinguish between Classes, Types and Sizes
Comment: Commenter IV-D-17 recommended that the EPA
establish separate standards for coal-, oil-, and gas-fired
units. The commenter noted that EPA has subcategorized utility
boilers by fuel type in all previous NSPS for NOX. The commenter
stated that EPA has chosen to ignore differences between
categories of sources. The commenter voiced concern that the
4-12
-------
proposed standard will result in a bias toward natural gas in
electric generation, which could be a risk considering supply and
availability factors. Additionally, the commenter did not agree
with establishing a standard that "stretches one source category
(coal) to the limits of economic efficiency, while requiring
little from the other source categories (oil and gas)."
Commenters IV-D-47, IV-D-52, and IV-D-63 asserted that the
EPA has not justified its decision to abandon fuel specific
standards in favor of the fuel neutral approach. The commenters
noted that there was "little discussion or analysis concerning
cost, feasibility and other issues regarding the
subcategorization of types of coals." Commenter IV-D-47 noted
that the EPA's Regulatory Impact Analysis did not consider any
option that would have subcategorized types of coal, or any
control technology other than SCR. The commenter stated that
"Because Congress gave EPA discretion to subcategorize on the
basis of fuel type, and because EPA has previously determined
that subcategorization is necessary to satisfy the §111 decision
making criteria in the case of NOX standards for utility boilers,
the Agency has an obligation to explain why it has decided to
reject its prior rulemaking conclusions that subcategorization is
necessary to satisfy the §111 statutory criteria."
Commenters IV-D-38 and IV-D-52 noted that the proposal did
not distinguish between classes, types and sizes within
categories. Commenter IV-D-52 stated that the approach taken is
counter to all previous NSPS rulemakings for NOX, and requested
that the EPA explain why.
Commenter IV-D-35 stated that the EPA has not justified the
rationale for not creating subcategories among coal-fired units
based on fuel sulfur content.
Response: Past regulatory approaches were based on boiler
modification techniques, which made fuel selection more closely
4-13
-------
related to performance compared to the post-combustion control
techniques that comprise the currently proposed subparts Da and
Db. The performance characteristics of the SCR technology do not
justify the creation of subcategories based on sulfur content of
coal (although the EPA has revisited some of the cost analyses
related to the catalyst life issue, see section 3.3.). Section
lll(b)(2) of the Clean Air Act allows the Administrator to
distinguish among classes, types and sizes of sources, but does
not require the Administrator to do so. As discussed elsewhere,
the EPA does not believe that the format of the proposed rules
creates a bias in the use of natural gas in electrical
generation, but rather, provides owners and operators with
additional flexibility in meeting the NOX limit.
4.3 PROMULGATION SCHEDULE AND COORDINATION WITH ICCR
Comment: Commenter IV-D-30 expressed opposition to any
further delays in the promulgation and implementation of the
proposed NOX NSPS. The commenter pointed out that "EPA should
have promulgated this NOX NSPS over 3 years ago and further
delays at this time are unjustified." The commenter urged EPA to
adopt the proposed revisions as soon as possible in order to
achieve needed reductions in NOX emissions from all sectors.
Commenters IV-D-30 and IV-D-62 stated that "EPA should not
delay the implementation of the proposed NOX NSPS in order to
coordinate it with other ongoing actions such as the Industrial
Combustion Coordinated Rulemaking (ICCR) process." Commenter IV-
D-30 speculated that the NSPS could be used as a NOX benchmark in
the ICCR process when establishing the MACT floor.
In contrast, Commenter IV-D-47 noted that Executive Order
12866 directs regulatory agencies to "avoid regulations that are
inconsistent, incompatible, or duplicative with its other
regulations." Combined cycle units emit NOX from a combustion
turbine and a duct burner. At this time, the combustion turbine
4-14
-------
emissions are regulated by NSPS Subpart GG, while the duct burner
emissions are regulated by Subpart Da or Db. The commenter
stated that it would be difficult for combined cycle units to
show compliance with the proposed output-based Subpart Da
standard. The commenter noted that the ICCR is considering
revising Subpart GG. The commenter urged the EPA to "undertake a
separate rulemaking that results in a single rule that regulates
all NOX emissions from such units."
Commenter IV-D-26 explained that the ICCR committee was
established to coordinate the rulemaking for industrial-
commercial-institutional combustion sources under Section 111,
112 and 129 of the Clean Air Act. The commenter stated that the
proposed NOX was drafted outside of the ICCR process, and
recommended that the EPA determine how the modified NSPS will
impact the ICCR process and whether the ICCR should alter the
scope of its rulemaking.
Commenters IV-D-45 and IV-D-48 recommended that the proposed
Subpart Db NOX NSPS should be combined with the EPA's ICCR
procedure. Commenter IV-D-61 recommended that the NOX NSPS
proposal for industrial-commercial-institutional boilers should
be withdrawn and considered as part of the ICCR. If the EPA
issues the proposed NSPS and the ICCR analysis results in a
proposal that would be different, the EPA would have to revise
the NSPS again. "This would cause the regulated community undue
hardship in trying to comply with multiple and possibly differing
control requirements."
However, Commenter 3 stated that the industrial portion of
the NOX NSPS should not be included within the ICCR and gave six
reasons:
(1) It is not appropriate for either the U.S. EPA or the
ICCR to attempt to circumvent the court-ordered deadlines by
using the ICCR as an excuse.
4-15
-------
(2) NOX reductions are needed from all sectors, including
nonutility units, as soon as possible in order to reduce the
current ozone problem...
(3) The EPA has already performed an adequate analysis of
the impact of the proposed NOX NSPS on a source's ability to
control other air pollutants.
(4) The EPA has already performed an adequate cost analysis
on the cost impacts of the rule for both utility and
nonutility units and has already proposed to substantially
reduce the cost impact for new industrial steam generating
units, by about 70%, by proposing NOX emission limit at 0.20
Ib/million Btu rather than 0.15 Ib/million Btu.
(5) The proposed NSPS can be used by the ICCR boiler
workgroup as a NOX benchmark when establishing the MACT
floor and therefore need not conflict with any of the work
already performed by the ICCR.
(6) The simplest way to avoid any conflict between the
proposed NOX NSPS and the ICCR Boiler Workgroup's work is to
accept the proposed NOX NSPS as the NOX emission limit for
fossil fuel-fired boilers when establishing the MACT floor
for these units.
Response: The EPA is under a court-ordered deadline to
promulgate revisions to the NSPS by September 1998. The July
1997 promulgation of revisions to the ozone NAAQS lends
increasing urgency to the development of national standards and
other tools that will assist the States in developing
implementation plans to meet the new standards. The NOX NSPS
revisions are one such tool that would be used by the States in
their attainment planning. However, the EPA agrees with
commenters that the outcome of the NSPS should be considered in
the ICCR process.
The EPA agrees that ICCR-driven revisions to subpart GG,
4-16
-------
standards of performance for stationary gas turbines, pose a
potential conflict with the subparts Da and Db standards, if they
extend the applicability of subpart GG to the duct burner, which
is currently covered by subparts Da and Db. Therefore, the EPA
will revise subparts Da and Db to exempt sources that may also
become subject to subpart GG, should such revisions to subpart GG
occur.
4.4 OVERALL MONITORING, REPORTING, AND RECORDKEEPING
REQUIREMENTS
Comment: Commenter IV-D-01 noted that small units (maximum
heat input of 100 MMBtu/hr or less, but greater than or equal to
10 MMBtu/hr) subject to Subpart DC have no NOX limit, monitoring
or recordkeeping requirements. The commenter recommended NOX
requirements that are intermediate to those of Subpart DC (none)
and the current requirements of Subpart Db (extensive) for low
emitting units subject to Subpart Db.
Response: The EPA believes the current Subpart Db
requirements are the minimum needed to ensure compliance with the
standard. However, owners or operators of low-emitting boilers
subject to the requirements of this rule may petition the U.S.
EPA Regional offices for alternative monitoring methods,
according to section 60.13(1) of the part 60 General Provisions.
The EPA will consider these petitions on a case-by-case basis.
Comment: Commenter IV-D-21 recommended that language be
added to 40 CFR Part 60 to "exempt from measuring and reporting
gas- and/or oil-fired boilers that currently meet any state or
local NOX emission standard that is equivalent to or more
stringent than the federal regulation."
Response: Since State and local regulations are usually not
Federally enforceable, EPA regulations must be enforced. If the
State/local regulations are more stringent than the applicable
EPA regulations, the affected facilities may individually
4-17
-------
petition the EPA for relief under the alternative monitoring
provisions. Alternatively, Title V streamlining can coordinate
the State and Federal requirements.
4-18
-------
5.0 ESTABLISHING OUTPUT-BASED FORMAT FOR UTILITY BOILERS
5.1 OVERALL APPROACH
5.1.1 Support Output-Based Format
Comment: Several commenters (1, 4, IV-D-05, IV-D-07, IV-D-
18, IV-D-19, IV-D-20, IV-D-25, IV-D-27, IV-D-29, IV-D-33, IV-D-
34, IV-D-39, IV-D-42, IV-D-43, IV-D-44, IV-D-46, IV-D-49, IV-D-
50, IV-D-51, IV-D-54, IV-D-60, and IV-D-65) expressed support for
the output-based format of the proposed standard. These
commenters indicated that the output-based format would reward
energy-efficient generators.
Commenter IV-D-39 cited the use of the following design
options to improve efficiency: air or water preheaters,
economizers, fans, and/or heat exchangers. The commenter added
that it is "common today for boiler efficiency to deteriorate
over the life of the unit, and the efficiency calculation would
ensure that the operator properly maintained the unit."
Response: The EPA appreciates the commenters' support.
5.1.2 Oppose Output-Based Format
Comment: Commenters IV-D-11, IV-D-17, IV-D-32, IV-D-36, IV-
D-37, IV-D-47, IV-D-53, and IV-D-63 opposed the output-based
format noting the following reasons:
(1) The incentives to be efficient have recently increased
due to the newly competitive nature of the industry, and
will continue to increase without output-based standards.
(2) The format would add significant burdens to an already
complicated monitoring system for utilities.
(3) There are inconsistencies between the proposed NSPS
output-based format and the following input-based
regulations also applicable to these sources: existing
boilers NOX units, sulfur dioxide and particulate matter
limits, electric generating units under NSPS Subpart D,
5-1
-------
existing regulations implementing reasonable available
control technology (RACT) for NOX in ozone non-attainment
areas, proposed NOX limitations for states included in the
Ozone Transport Assessment Group (OTAG), and the NOX
Emission Reduction Program under 40 CFR 76, and Section 407
of the Acid Rain Program requires output-based reporting.
(4) NOX averaging of NSPS units with existing units would
be very complicated.
(5) The output-based format is inappropriate and inaccurate
for cogeneration facilities that produce steam in addition
to or in place of electric generation. The commenter
explained that customers dictate the temperature and
pressure conditions of the steam that is produced. The
generator has no choice and must produce the desired
product. The commenter indicated that the EPA method of
equating steam production to electric production was over-
simplified and punitive in that it does not consider all of
the potential steam production conditions. The commenter
reported that this would increase the cost of efficient
cogeneration. The commenter concluded that the input-based
standard is more appropriate, fair, and environmentally
protective.
(6) An output-based NSPS does not promote energy efficiency
because it "makes no allowance for the use of low Btu fuels
(such as waste coal) that would otherwise go unused...By
encouraging consumption of less expensive low Btu fuels, the
EPA would promote generation of electric power at costs
below those presently realized." Commenter IV-D-36 added
that not "penalizing" utilities for burning low Btu fuels
would promote discovery and utilization of these fuels, and
thereby contribute to national energy self-sufficiency.
Further, commenter IV-D-36 argued that the proposed NSPS "is
5-2
-------
not keeping with recent utility deregulation," because "an
important goal of recent utility de-regulation was to allow
market forces to minimize the cost of electric power to
consumers, without eroding environmental protection."
(7) The EPA's proposal should encourage consumption of low-
cost fuels.
Response: The EPA continues to believe in the benefits
associated with an output-based standard for new sources that
encourages energy efficiency. The changes in the output-based
format, discussed below in section 5.2, will simplify the
compliance demonstration for sources by eliminating the need to
convert input values to output values. While, the EPA is
concerned about apparent inconsistencies in monitoring
requirements associated with various programs to which individual
sources might be subject, the EPA also feels that the
requirements of the NSPS stand on their own merits. The NSPS
provisions do not require any additional monitoring at sources
beyond what is already required by the Acid Rain program. In
some instances, the Title V permit process and activities such as
permit streamlining may provide relief to sources on a case-by-
case basis. In addition, the EPA will continue to explore
additional ways to provide monitoring relief that do not
compromise the ability of EPA to adequately enforce Federal
standards.
As discussed below in section 5.2.5, the EPA did examine
possibilities to revisions to the steam credit allowance for co-
generation facilities. These issues are further addressed in
that section.
Finally, the EPA believes that low-cost fuels can be used
effectively at facilities subject to the final standards. As
discussed, the U.S. Generating Company's Northampton facility is
currently performing better than would be required under the
5-3
-------
amended NSPS and uses waste coal as its sole energy source.
5.2 INPUT TO OUTPUT CONVERSION ASSUMPTIONS
The EPA has decided to revise the compliance demonstration
for affected sources, based on analysis of comments submitted on
the input to output conversion assumptions. As discussed in
detail in this section, the EPA will finalize the standard for
new sources at a level of 200 ng/J (1.6 Ib/MWh) gross energy
output. This change incorporates concerns related to overall
heat rates, steam credits for cogeneration facilities, and gross
versus net output. In addition, the key underlying assumption
inherent in the selection of the level of the final standards at
200 ng/J (1.6 Ib/MWh) gross output, i.e., the input-based
standard of 0.15 Ib N0x/million Btu, is maintained. The effect
of this change is that sources would no longer be required to
calculate output emissions based on a measurement of input
converted to the output format. The EPA believes this change
will be simpler for sources to comply with and for enforcement
agencies to monitor compliance.
5.2.1 Support the 38-Percent Baseline Efficiency
Comment: Commenters IV-D-34, IV-D-50, and IV-D-54 noted
that the application of a baseline efficiency factor was an
appropriate means of establishing the output-based limit.
Commenters IV-D-18, IV-D-34, IV-D-50 and IV-D-54 stated that the
38-percent efficiency factor was reasonable.
Commenter IV-D-20 did not challenge the EPA's selection of
38-percent efficiency for new boilers, corresponding to a heat
rate of 9,000 Btu/kWh. However, the commenter believed that EPA
should be consistent and "choose a representative, sustainable
heat rate for new boilers after 5 years of operation."
Response: As discussed below, the selection of a baseline
efficiency value is intimately tied to the selection of a
corresponding heat rate. Based on information received by
5-4
-------
commenters, the EPA has decided to revisit the heat rate issue.
5.2.2 Oppose the 38-Percent Baseline Efficiency
Comment: Commenters IV-D-19 and IV-D-65 stated that the EPA
adoption of a single heat rate was indefensible. The commenters
remarked that the EPA "had ample time" to convert each boiler's
input-based emission data to an output-based emission rate.
Further, the commenters noted that the EPA "must be careful in
choosing the single heat rate factor for conversion;" adding that
"merely picking a 38 percent efficiency based on anecdotal
evidence is not sufficiently rigorous."
Response: As discussed below, the EPA has used information
provided in the public comment period to reevaluate its
assumptions regarding the underlying assumptions in the output
conversion equation. We believe our analysis is adequate and
sufficient to demonstrate the feasibility of the final approach.
Comment: Commenter IV-D-36 stated that "the proposed NSPS
defines a NOX emission limit that is a function of NOX emission
rates and plant-wide thermal efficiency and in so doing, favors
water-cooled condensers over air-cooled condensers." The reason
is that "the air-cooled heat rejection systems are inherently
less thermally efficient than water-cooled systems." The
commenter explained that in the western United States water is at
a premium, adding that the "best allocation of water resources in
these areas does not always include water-cooled power plants."
The commenter recommended that "some allowances must be made,"
and offered the possibility of second equation for air-cooled
units, "replacing the assumed 38-percent thermal efficiency in
the current equation with a representative air-cooled efficiency
(probably in the order of 31 percent)."
Response: The proposed output-based standard has been
revised to 1.6 Ib/MWh gross output. This standard corresponds to
a gross heat rate of 10,667 Btu/kWh and a gross thermal
5-5
-------
efficiency of 32 percent, which should satisfy the commenter's
concerns about air-cooled units.
Comment: As discussed in section 4.1.3., several commenters
were concerned about the ability of existing boilers to comply
with the NSPS should they become affected sources through
modification or reconstruction.
Response: The EPA agrees with the concerns raised by
commenters that the inherent efficiencies of existing boilers may
be less than the efficiency that new boilers are capable of
achieving. Lower boiler efficiency translates to higher average
heat rates, which would make it more difficult for existing
sources to meet an output-based standard without increasing SCR
performance significantly (up to a 40 percent improvement could
be required.) Therefore, the EPA has revised the final rule to
allow existing boilers that might become subject to the NSPS
through modification or reconstruction to meet an equivalent
input-based standard of 65 ng/J (0.15 Ib/MMBtu). This level of
control represents the same overall NOX reduction efficiency that
would be required of new units, but lacks the benefits attributed
to promoting energy efficiency that the output-based format
provides. The actual environmental impact of the change should,
therefore, be negligible.
5.2.3 Support Net Heat Rate of 9.000 Btu/kWh
Comment: Commenter IV-D-50 supported the EPA's assumed
"baseline" efficiency of 38 percent, which corresponds to a heat
rate of 9,000 Btu/kWh. The commenter noted that most U.S. boiler
heat rates range from 9,000 to 13,000 Btu/kWh. Because the
intent of the EPA is to encourage efficiency, the 9,000 Btu/kWh
heat rate is appropriate.
Response: As discussed below, the EPA has reconsidered the
heat rate assumption, based on data obtained by the EPA since
proposal and received from commenters.
5-6
-------
5.2.4 Oppose Net Heat Rate of 9.000 Btu/kWh
Comment: Commenters IV-D-17, IV-D-19, IV-D-27, IV-D-28, IV-
D-37, IV-D-47, IV-D-52, IV-D-53, and IV-D-65 questioned the
proposed heat rate standard of 9,000 Btu/kWh. The commenters
stated that the "proposal fails to provide necessary discussions
justifying the selection of the highly restrictive baseline
threshold of 9000 Btu/kW-hr." Further, Commenter IV-D-52
recommended that the EPA review all available heat rate data for
U.S. utilities, and reconsider applicable modifications to the
proposed baseline..." One commenter noted that the rate may be
appropriate for gas-fired combined cycle units, but would
discourage the use of coal and waste coal. Commenter IV-D-37
specified that heat rates in the 9,000 Btu/kW-hr are typically
limited to those operating at supercritical steam pressures and
temperatures along with combined cycle gas turbine (CCGTs) units.
Commenter IV-D-37 continued by stating that "because many Da
units are subcritical and fire solid fuel, imposition of a one-
size-fits-all net efficiency constitutes a bias against these
types of units."
Commenter IV-D-28 stated "the only type of solid fuel
facility that could meet a low heat rate standard of 9,000 Btu
would be a huge (1,000 MW) super critical coal unit with
extremely high operating temperatures. Small waste coal
facilities with circulating fluidized bed boilers could not meet
this standard." Commenter IV-D-55 elaborated by stating that one
of their pulverized coal-fired utility power plants has an
average net heat rate of 9,808 Btu/kWh. Commenter IV-D-27 stated
that their coal burning facilities are state-of-the-art from an
emissions standpoint, and operate at heat rates of up to 11,000
Btu/kWh. This figure agrees with conceptual designs for a future
coal-fired plant, which assumed heat rates of 9,900 to 13,757
Btu/kWh. The commenter stated that these data suggest that an
5-7
-------
output-based standard of 1.35 Ib/MWh is equivalent to an input-
based standard of 0.1 to 0.137 Ib/MMBtu, not 0.15 Ib/MMBtu as
assumed by the EPA.
Commenter IV-D-53 reported that the net heat rate for new
coal fired generation will fall in the range of 9,400 to 9,600
Btu/KWh at full load, 9,600 to 11,000 Btu/KWh at mid load and
over 13,000 Btu/KWh at minimum load. The commenter recommended a
baseline efficiency of 10,500 Btu/KWh. Commenters IV-D-17, IV-D-
47 and IV-D-37 stated that, assuming one typical heat rate is
appropriate, an analysis of data from Subpart Da boilers
indicates that a heat rate of 10,500 Btu/kWh would be more
appropriate, whereas Commenter IV-D-27 recommended a heat rate of
10,000 Btu/kWh. This would result in an output-based limit of
1.58 Ib NOx/MWh for the 10,500 heat rate and 1.5 Ib/MWh for the
10,000 Btu/kWh heat rate.
Commenters IV-D-16 and IV-D-63 stated that there is no data
by EPA to show a 9,000 BTU/kWh heat rate can be obtained.
Response: The proposed heat rate was a major concern of
both commenters and the EPA. In light of additional data
supplied by commenters and collected by EPA, the EPA has decided
to revise the assumed heat rate. First, the output-based
standard is now based on gross output instead of net output, so
the following discussion will be in terms of gross heat rates.
The decision to switch from net to gross output is discussed in
section 5.3.
The commenters indicated that net heat rates of 10,000 to
10,500 Btu/kWh are typical of state-of-the-art units. The EPA
collected data from four additional utility boiler that are
considered to be new and state-of-the-art from an emissions
standpoint. The first boiler was a base-loaded, fluidized bed
combustion cogeneration unit that fired waste coal and was
equipped with SNCR (Northampton). This unit's average gross heat
5-8
-------
rate (with 50 percent credit for export steam) was less than
9,000 Btu/kWh. The second unit was a pulverized coal-fired,
cogeneration unit that operated under cycling load and was
equipped with SCR (Logan). This unit's average gross heat rate
(with 50 percent credit for export steam) was approximately
10,250 Btu/kWh. The third utility boiler (Stanton) had an
average heat rate of 10,250 Btu/kWh. The Birchwood cogeneration
unit, the fourth facility, reported that they cycle between heat
rates of approximately 10,700 Btu/kWh at 32 percent load and
9,000 Btu/kWh at 100 percent load. The heat rates reported by
the Birchwood cogeneration unit are based on a 100 percent credit
for export steam.
The EPA conducted statistical analyses in which the
objective was to assess long-term NOX emission levels, on an
output basis, that can be achieved continuously. Statistically,
Logan, Northampton, and Birchwood can meet the revised output-
based standard of 1.6 Ib/MWh (gross) on a 30-day rolling average.
5.2.5 Efficiency Calculation for Cogeneration Units
Comment: Commenters IV-D-18, IV-D-19, IV-D-34, IV-D-39, IV-
D-44, IV-D-53 and IV-D-65 asserted that using only 50 percent of
the thermal energy from the steam generated at cogeneration
facilities in calculations of output-based emission rates is
inappropriate. The commenters reported that the 50-percent
allocation is from a section of the Public Utility Restructuring
Policy Act (PURPA) in which the 50-percent thermal output is used
as part of a definition of a PURPA-qualifying facility. Further,
the commenter stated that the calculation should use either the
electric output converted to MMBtu plus the enthalpy of the full
steam or hot water output in MMBtu, or the electric output in
MWhel plus the enthalpy of the full steam or hot water output in
MWhth. Further, Commenter IV-D-39 reported that the efficiency
of new industrial boilers typically ranges from 78 to 83 percent
5-9
-------
depending on fuel and other design features. The commenters
reasoned that each application would differ in efficiency, and
credit should be given for the heat actually used. Commenters
IV-D-19 and IV-D-65 added that the restriction of the steam
credit to 50 percent is "arbitrary and capricious."
Commenters IV-D-20, IV-D-44, and IV-D-46 supported the
output-based standard and stated that the "appropriate output
measure for industrial boilers would be pounds of NOX per million
Btu of steam produced at the boiler steam header." The
commenters saw no reason to penalize cogeneration units by
calculating output as electric output plus 50 percent of the
thermal output, as suggested by EPA. The output calculation
should give full value to the steam produced. Output from a
cogeneration unit should be measured as the electric output plus
the full thermal output in consistent units.
Commenter IV-D-46 said that "the output for a cogeneration
facility should be the electric output and the full thermal
output expressed in consistent units (MWh or MMBtu) where 1 kWh =
3413 Btu."
Commenters IV-D-39 and IV-D-46 insisted that efficiency
should not be used as a compliance measure. The commenter
explained that the efficiency calculation is an extra, unneeded
step. The commenters reported that all that is needed is a
continuous emission monitoring system (CEMS) to directly measure
NOX and an electric or thermal measurement for output in units of
MMBtu or MWh.
Response: The EPA considered three approaches to resolve
the issue of steam credit for cogeneration facilities: 1) Allow
credit for steam as if it were being converted into electricity;
2) Allow credit in the form of 50 percent of the thermal value
(enthalpy) of the steam; and 3) Allow credit for greater than 50
percent of the value of the steam, up to 100 percent.
5-10
-------
The EPA decided not to allow credit for steam as if it were
being converted into electricity because the EPA wants to
encourage cogeneration. Allowing credit as if electricity would
only provide credit for up to 38 percent of the value of the
steam, which is the reported maximum of the efficiency of steam
to electricity conversion.
The EPA also decided not to allow for greater than 50-
percent credit for the steam. Based on analysis of heat rates
from cogeneration facilities, the EPA has determined that once
you exceed 50 percent and approach 100 percent credit for the
steam there is a disproportionate lowering effect on heat rate,
particularly at high steam export rates. This would result in
artificially low NOX emission rates. As another option, the EPA
considered allowing 100 percent credit for steam, but capping the
amount of steam for which credit could be received to a certain
percentage of total output. This approach was deemed to be too
complex from a monitoring standpoint.
Therefore, the EPA has decided to retain the proposed 50-
percent credit for export steam from cogeneration facilities on
the basis that it encourages cogeneration, will not artificially
lower NOX emissions, and will not require complex monitoring.
Comment: Commenter IV-D-39 reported that steam metering has
been well established, especially for companies that sell thermal
as well as electric energy. Commenter IV-D-39 estimated the cost
of thermal measurement equipment to range from $7,000 to $15,000
per boiler depending on the specific requirements of the system.
The commenter also provided two pages of cost data.
Response: Owner's or operator's would be allowed to request
the approval of alternative monitoring procedures. However,
with the change in the format of the standard, the use of the
proposed input to output conversion equation would no longer be
5-11
-------
necessary. The EPA anticipates that most sources will comply
with the standard through the continuous monitoring of NOX outlet
emissions.
5.3 GROSS VERSUS NET OUTPUT VARIABLE IN EQUATION
Comment: Commenter IV-D-54 supported the use of the net
output format. The reason given was that this format will
encourage owners and operators to lower the auxiliary power
requirements at the facility. Commenter IV-D-50 added that the
emission limit should be based on the net energy leaving the
facility.
Commenter IV-D-46 stated that the output-based standards
should be defined as Ibs NOx/MWh net for utility boilers or Ib
NOx/MMBtu at the steam header for other boilers.
In contrast, Commenters IV-D-37 and IV-D-42 opposed basing
the output standard on the net output term. Commenter IV-D-37
specified that the certified monitoring of electric power output
would add another layer of monitoring requirements while
providing "no real benefit." Further, Commenter IV-D-37 reported
that the output-based format would "require significant and
costly changes to the software of monitoring and reporting
systems." Further, the commenter explained that the issue of
measurement location is unresolved by noting the discrepancy in
the definition as "the net electrical output (i.e. net busbar
power leaving the plant) from the turbine generator set." The
term "net" means the sum of the power leaving the generating
units minus the power required to drive auxiliary equipment.
Commenter IV-D-42 recommended basing the standards on gross
rather than net output to account for the power drain associated
with many types of control technologies.
Commenters IV-D-36 and IV-D-47 reported that electrical
output cannot be measured directly because it is dependent on the
"electrical usage by hundreds of motors and other auxiliary
5-12
-------
equipment located throughout the plants." The commenter claimed
that net generation cannot be measured "by simply installing a
wattmeter."
Response: The EPA has reconsidered its position, and has
decided to finalize the rule based on the use of gross output
because of the monitoring difficulties inherent in the net output
methodology. In particular, measuring net output at facilities
with both affected and nonaffected units could be problematic,
because a single meter on the electricity leaving the facility
could not sufficiently allocate the electricity leaving the
affected boiler. The EPA reserves the opportunity to revisit
this issue should EPA develop a methodology to determine the net
heat output in all circumstances.
Comment: Commenters IV-D-32, IV-D-37, IV-D-47, IV-D-52, IV-
D-54, and IV-D-55 protested that the proposal did not include a
specific methodology for determining the unit net output, but
that a methodology will be in the final rule. One commenter
pointed out that this does not provide for a subsequent comment
period on a "significant component" of the proposal. Commenter
IV-D-52 urged the EPA to "withdraw this proposal until a complete
and thorough package can be provided for full public review and
comment, as required."
Response: The changes made in the final standard make the
commenters' concerns about a specific methodology to determine
unit net output immaterial.
5-13
-------
6.0 REVISED STANDARD FOR ELECTRIC UTILITY STEAM GENERATING UNITS
(SUBPART Da)
6.1 SUPPORT THE LEVEL OF THE STANDARD
Comment: Commenters IV-D-07, IV-D-33, and IV-D-57
supported the level of the standard. Commenter IV-D-57 stated
that the proposed 1.35 Ib/MWh standard was achievable for
electric utility steam generating units with post-combustion NOX
controls, either alone or combined with combustion controls. The
commenter pointed out that oil- and gas-fired units in California
use SCR and have emissions below 1.0 Ib/MWh. The commenter also
indicated that there are coal-fired units that emit less than the
proposed standard. One example is a new coal-fired boiler in
Virginia that achieves SCR reductions of 60-65 percent and has
outlet emissions below 0.08 Ib/MMBtu (heat input). Commenter IV-
D-07 complimented the EPA on the fact that these standards are
stricter than all of the previous non-gas standards.
In support of the EPA's analysis, Commenter IV-D-51 referred
to a report entitled "Fuel Choice for New Electric Generating
Capacity in the Next Century: Coal or Natural Gas", and provided
a copy of this report for EPA review. Commenter IV-D-51 stated
that the report confirmed the application of SCR or FBC with SNCR
can achieve 80- to 90-percent NOX removal, yielding a
representative emission rate of 0.05 Ib/MMBtu to 0.20 Ib/MMBtu.
The commenter also stated that emission reductions from existing
utility plants would be required to achieve the new federal
standard for ozone, and that retrofit of existing plants to meet
the proposed standard would be feasible and cost-effective.
Commenter 1 supported the proposed NOX limit of 1.35 Ib/MWh.
However, the commenter noted that only 17 new utility boilers
were planned for the next five years and that "the real NOX
problem will likely come from industrial boilers."
6-1
-------
Response: The EPA appreciates the commenters' support.
6.2 STANDARD IS TOO LENIENT
Comment: Commenter IV-D-58 stated that the proposed
standard is much too lenient and does not reflect the "best
demonstrated technology." Commenter IV-D-58 reported reviewing
NOX data for Phase 1 and Phase 2 coal-fired units and found that
the average Btu-weighted emissions rate in 1995 was 0.58
Ib/MMBtu. Commenter IV-D-58 then assumed an 80-percent NOX
removal rate for SCR, and calculated a controlled emission rate
of 0.12 Ib/MMBtu, which is lower than the presumptive limit upon
which the proposed standard is based. However, a new boiler,
which should be equipped with low-NOx burners, should have a much
lower uncontrolled baseline, resulting in much lower emissions
from the SCR. The commenter believes an emission limit of 0.08
Ib/MMBtu is achievable, especially because 25 percent of all
coal-fired boilers in the EPA's Acid Rain Program inventory have
a NOX emission rate of 0.40 Ib/MMBtu, and when reduced by 80
percent, yields 0.08 Ib/MMBtu. The commenter added that 1995
CEMS data show that 41 percent of those units in the EPA'S Acid
Rain database emitted at 0.05 Ib/MMBtu or lower. For oil-fired
units, commenter IV-D-58 noted that in the Acid Rain database the
average NOX emission rate was 0.184 Ib/MMBtu, with 35 percent of
the facilities in the database below the NSPS limit. Commenter
IV-D-58 concluded by stating that is "inappropriate for EPA to
establish limits that can already be met by a substantial portion
of the existing fossil fuel-fired population."
Response: The commenter's assumption of an 80 percent NOX
removal rate is based on a unit that emits more NOX than would be
emitted at baseline by units considered under the NSPS analysis.
For example, the Logan SCR has an average NOX removal rate of 65
percent. Therefore, to approach the NSPS limit with a
requirement for this high level of removal does not reflect
6-2
-------
actual conditions at new units. Also, there are additional costs
associated with meeting more stringent limits, especially with
respect to the amount of ammonia and catalyst required and the
increase in ammonia slip.
6.3 STANDARD IS TOO STRINGENT
Comment: Commenters IV-D-16, IV-D-25, and IV-D-59 stated
that the standard is too restrictive. Commenter IV-D-25
speculated that a standard that is too strong might discourage
construction of new, clean, and efficient plants, or eliminate
the use of lignite. The commenter recommended reviewing the
standards if no new plants are built to conform to them within
two years of finalization.
Commenter IV-D-59 also noted that the standard imposed a 35-
percent emission reduction for natural gas-fired units, and this
percent emission reduction should also be applied to coal-fuel
emission rates. The commenter recommended an emission limit for
coal-fired units of 0.35 Ib/MMBtu.
Commenter IV-D-31 reported that the NSPS level for coal-
fired boilers is lower than the lowest emitter in the database.
The commenter reviewed the EPA Acid Rain database on the Internet
for the facilities in the NSPS database. The commenter found
that the Merrimack 2, from the NSPS database, is not a new
boiler, but an SCR retrofit on an older cyclone boiler. Further,
the Stanton 2 plant's emission rate of 0.163 Ib NOx/MMBtu (second
quarter 1997) which equated to an emission rate of 1.67 Ib
NOx/MW-hr. This emission rate is higher than the EPA's proposed
limit of 1.35 Ib NOx/MW-hr. The commenter asked, "How does EPA
justify a rate lower than a state of the art plant?"
Response: As discussed, the EPA has revisited the format of
the final standard and has revised it accordingly. Regarding the
performance of the Stanton plant, the EPA's analysis shows that
it would meet the revised standard of 1.6 Ib/MWh, based on a 30-
6-3
-------
day rolling average.
6-4
-------
STEAM GENERATING UNITS (SUBPART Db)
7.1 EXCLUSIONS
: Commenters IV-D-22 and IV-D-26 stated that the EPA
should not apply the proposed standard to modified and
waste heat systems are typically installed in the ductwork of a
gas turbine exhaust and are not amenable to significant
x control because of their configuration.
reconfiguration is extremely limited, and possible back pressure
impacts on the upstream device are a major concern. Applying the
because the NO from the upstream device (i.e., combustion
turbine) cannot be separated from the steam generator NO for
purposes of add-on control. The commenters said that add-on
controls are not demonstrated for such systems.
: As described in the response to the comment under
section 4.1.1, the EPA is not aware of any instances where
to the NSPS, nor does the EPA anticipate such instances in the
future. The General Provisions already provide several
For example, sources that offset their increased emissions are
not subject to the NSPS because of modification. These
modification provisions to existing sources.
The systems described by the commenters would be subject to
As discussed earlier, the EPA agrees that ICCR-driven revisions
to subpart GG could pose a potential conflict with the subparts
7-1
-------
Da and Db standards, if they extend the applicability of subpart
GG to the duct burner, which is currently covered by subparts Da
and Db. Therefore, the EPA will revise subparts Da and Db to
exempt sources that may also become subject to subpart GG, should
such revisions to subpart GG occur.
Comment: Commenters IV-D-09, IV-D-11, IV-D-12, IV-D-13, IV-
D-15, and IV-D-45 noted that the proposed revision appears to
apply to all steam generating units, including units that are
excluded from the current standard because they fire 10 percent
or less fossil fuel. The commenters did not believe that the EPA
intended that the revised NOX limit should apply to facilities
that combust a limited amount of fossil fuel. Several commenters
suggested clarifying the following language to at the end of 40
CFR 60.44b(1)(1): "...86 ng/J(0.20 Ib/million Btu) heat input
unless the affected facility has an annual capacity factor for
coal, oil, and natural gas of 10 percent (0.10) or less and is
subject to a Federally enforceable requirement that limits
operation of the facility to an annual capacity factor of 10
percent (0.10) or less for coal, oil, and natural gas; or ...."
Response: The EPA did not intend to remove the 10-percent
exemption from the revised NSPS. The EPA will add the suggested
regulatory language to clarify that this exemption still holds.
Comment: In addition to recommending the revised language
cited above, commenters IV-D-09, IV-D-12, and IV-D-13 pointed out
that, as written, the proposed NOX revisions would include
municipal solid waste combustors that only use a limited amount
of fossil fuels for startup purposes and supplemental fuel during
those periods when the heat content of the waste is low, in order
to maintain good combustion conditions. The commenters stated
that the proposed Subpart Db NOX emission limit revisions would
be approximately 120 parts per million, by volume, dry (ppmdv)
(corrected to 7% 02) , as compared to the revised NSPS for large
7-2
-------
municipal waste combustors (MWC) units (December 19, 1995,
Federal Register 40 CFR 60 Subpart Eb) that limits NOX
to 150 ppmd, corrected to 7% 02
three years of operation. Commenter IV-D-09 added that existing
(EG) in 40 CFR 60 Subpart Cb to meet a NO emissions limit of 205
ppmdv corrected to 7% 0 (daily arithmetic average). The
commenters suggested that the addition of the 10-percent
exemption, discussed above, would alleviate this concern. In
sense to exempt facilities entirely that are subject to the
subpart Eb and Cb requirements.
: As discussed above, the EPA has included the
language regarding the 10-percent exemption to the final rule,
will revise the final rule to exempt units that are subject to
subpart Eb to avoid any possible conflicts.
Comment: Commenter IV-D-57 stated that the proposed
quite reasonable. According to the commenter, several units
burning a wide variety of fuels currently use SCR and other
x emissions below this level.
numerical standard to encourage innovation in pollution
prevention. The commenter stated that the 0.20 lb\ MMBtu NO
limit would require the application of at least one control
technology or a combination of the technologies cited in the
will add complexity to the operation of the affected source. The
commenter was unable to determine an appropriate recommendation
-------
the EPA in order to develop a reasonable value that allows the
necessary flexibility.
Commenter 1 stated that the NOX emission limit of
0.2 Ib/MMBtu was not adequate. The commenter pointed out that
the EPA has estimated that 381 new industrial boilers will be
built in the next five years (293 natural gas/distillate oil, 66
residual oil, and 22 coal). The commenter indicated that the new
standards, as currently proposed, would ignore the 293 new
natural gas and distillate oil units that are predicted over the
next five years. The commenter requested that the standard be
set at 0.15 Ib/MMBtu (except for low-heat gas and distillate oil
units), and enforced using the best available control technology.
Commenter IV-D-05 agreed with this comment, reasoning that a 0.15
Ib/MMBtu emission level was achievable for industrial sources
stating that the best available technology should be used on all
new sources.
Response: The EPA believes that the proposed 0.20 Ib/MMBtu
is the appropriate level for the subpart Db standard, and will
finalize this limit in the promulgated standard. The EPA
evaluated the costs associated with controlling natural gas and
distillate oil units to the 0.15 Ib/MMBtu level, and found that
their smaller size and lower capacity factors resulted in much
higher cost-effectiveness values associated with the application
of flue gas treatment than do coal-fired units. As stated in the
proposal Federal Register notice, the 0.20 Ib/MMBtu limit would
result in approximately a 70-percent reduction in the annual
nationwide costs for new industrial steam generating units
compared to establishing a limit at 0.15 Ib/MMBtu for all new
units. However the 0.20 Ib/MMBtu limit reflects about a 50- to
70-percent reduction in NOX emissions over the current subpart Db
limits for coal-fired and residual oil-fired units. Based on
these cost considerations, the EPA has determined that
7-4
-------
establishing a lower limit for natural gas and distillate units
is not justified.
7.3.1 Support Input-Based Format
: Commenters IV-D-10, IV-D-19, IV-D-22, IV-D-24, IV-
D-26, IV-D-36, IV-D-37, IV-D-45, IV-D-64, and IV-D-65 supported
to be input-based. Commenter IV-D-22 opposed the output-based
standard for industrial boilers, because they operate complex
The commenter wrote that "it would be extremely difficult, if not
impossible, to reliably monitor an output-based standard."
would add complication for the following reasons:
(1) Permit limits, Title IV limits, and other regulations
(2) The output-based standard cannot be applied equitably
to all sources because the standard used a rate of 9,000
(3) The NSR program would be a disincentive to improving
efficiency. Further, if an output-based standard is set, it
net, to account for power drain from pollution control
equipment.
monitoring equipment.
(5) The industrial sector contains economic drivers to
(6) Implementation of an output standard would be almost
impossible for the industrial sector due to the variety of
and IV-D-26 added that downstream conditions limit the
-------
achievable efficiency of the boiler, adding that they may be
required to reduce the pressure of the steam generated in
order to supply steam at a lower pressure for a particular
process. These constraints by the production processes
lowers the calculated efficiency of the boiler. Also, steam
is generated from waste heat systems where possible. Waste
heat availability is highly variable, which means the
efficiency of a waste heat system is highly variable.
Response: The EPA appreciates the commenters' support. The
EPA continues to believe that the input-based format is
appropriate for industrial boilers.
7.3.2 Oppose Input-Based Format
Comment: Commenters IV-D-18, IV-D-39, and IV-D-54
recommended that the EPA consider an output limit for industrial
steam generating units. Commenter IV-D-18 recommended that the
limit be determined on a case-by-case basis, where it can be
applied as an alternative to the input limit. In contrast,
commenter IV-D-54 recommended a unilateral limit because it would
give preference to higher efficiency systems and because the
owner/operator would be more attentive to plant operations to
"ensure efficient operation."
Commenter IV-D-39 said that a longer averaging period, i.e.,
12 months, would address the load variability issues associated
with industrial boilers meeting an output-based standard. The
commenter also argued that output-based standards will promote
improved boiler efficiency at both initial installation and over
the life of the boiler. The commenter said that factors such as
basic combustion design, use of air or water preheaters,
economizers, fans, and heat exchangers all affect efficiency and
will be the basis for efficiency improvement if the regulations
provide the incentive. In addition, an output-based standard
will encourage the owner or operator to maintain the efficiency
7-6
-------
of his or her new boiler over the life of the unit or to find
ways to offset decreases in efficiency. Finally, the commenter
argued that the measurement issue is addressed if the output is
measured as the full enthalpy of thermal energy (steam or hot
water) leaving the boiler. The enthalpy can be calculated
automatically from the temperature and pressure sensors that are
part of the steam flow metering system. According to the
commenter, the cost of thermal measurement equipment can range
from $7,000 to $15,000 per boiler.
Commenter IV-D-50 noted that the current input-based
standards have not provided incentives for efficiency and
pollution prevention. The continuation of input-based standards
would not encourage efficiency in planned units.
Response: The EPA continues to believe in the value in
promoting pollution prevention and energy efficiency in the
regulatory process. Unfortunately, in this case, the nature of
industrial boilers and their use patterns seems to preclude the
practical application of an output-based format. As stated at
proposal, the EPA did consider an output-based format option of
Ib N0x/million Btu steam output, which could be applicable to all
new industrial boilers. However, this output-based format option
provides the owners with only minimal opportunities for promoting
energy efficiency at their respective facilities, because it
accounts only for boiler efficiency and ignores both the turbine
cycle efficiency and the effects of energy consumption internal
to the plant. The boiler efficiency is mainly dependent on fuel
characteristics. Beyond the selection of fuels, plant owners
have little control over boiler efficiency. In addition, an
output-based format would require additional hardware and
software monitoring requirements for measuring the stack gas flow
rate (for determining the mass rate of NOX emissions), steam
production rate, steam quality, and condensate return conditions.
7-7
-------
Instrumentation to conduct these measurements may not generally
exist at industrial facilities as they do at utility plants.
Commenter IV-D-18's suggestion to allow the use of an
output-based format as an alternative to the input-based format
on a case-by-case basis would overcome the difficulties
associated with the variability in baseline efficiencies of
industrial boilers. In some cases, energy efficiency might be
effectively encouraged, but the difficulties associated with
monitoring such systems on a routine basis would still be
present. Therefore, the EPA has not changed the rule to reflect
this option. However, the NSPS would not preclude individual
States or sources (through Title V permit streamlining) from
pursuing this option when it can be demonstrated that equivalent
emission reductions could be obtained as under the NSPS.
Regarding the proposal to measure steam output from the
boiler as the means of demonstrating compliance with the output
standard, owners or operators could request use of such
alternative means on a case-by-case basis under the part 60
General Provisions.
7-8
-------
8.0 CONTINUOUS EMISSION MONITORING (CEM) REQUIREMENTS
8.1 GENERAL
: Commenter IV-D-08 requests that the EPA make the
reference to Part 75 CEMS much broader, so that it is readily
ranges, quality assurance, etc.) satisfy Subpart Da and Db
provisions.
: In the past, the EPA determined that Acid Rain
CEMS can be used as NSPS Subpart Da CEMS. That determination is
Assurances's web site. However, all of these CEMS must generate
reports according to the requirements of the applicable subpart
(State) in the regulatory format and by means acceptable to that
authority. The EPA is adding language to both subparts Da and Db
the part 60 requirements.
8.2 APPLICABILITY TO SMALL/SEASONAL UNITS
: Commenter IV-D-14 requested that the CEM
requirements for new low-NOx
reviewed, especially for boilers only used on a seasonal basis.
nitrogen oxide emission limits on old and new boilers that "far
exceed the limits contemplated in the proposal." Commenter IV-D-
processors in California provide no added assurance of
compliance, but merely add significant costs which are not
Response: As discussed in section 7.1, the EPA is
should also address the concerns of small, seasonal boilers.
-------
Comment: Commenter IV-D-01 recommended that the EPA
eliminate the CEMS requirements and associated reporting and
recordkeeping requirements in Subpart Db for steam generating
units that have a heat input capacity of 250 MMBtu/hy or less,
fire natural gas, and whose NOX emissions are 30 ppm (0.037
Ib/MMBtu) or less; and, instead require only initial and annual
emission testing. The commenter noted that their proposed
emission limit for CEMS requirement is about one third of the 0.1
Ib/MMBtu limit and one fifth of the 0.2 Ib/MMBtu limit in the
proposal. The commenter explained that this emission rate is the
limit in California. Commenter IV-D-01 stated that the testing,
monitoring, reporting, and recordkeeping requirements for steam
generating units regulated by Subpart Db are extensive as well as
costly. The commenter also noted that this would give the
facility the choice between installing a low NOX emitter (NOX
emissions of 30 ppm (0.037 Ib/MMBtu) or less) or a higher NOX
emitter with a CEMS. Further, the commenter provided recommended
language to revise the proposed NSPS.
Response: As discussed at proposal, the EPA believes the
monitoring costs associated with Subpart Db are reasonable and
necessary. In any case, low emitters have an option to petition
the EPA for alternative monitoring methods according to section
60.13(1) of the part 60 General Provisions.
8.3 CONSISTENCY BETWEEN PROGRAMS
Comment: Commenter 2 stated that "several Subpart Db NOX
monitoring procedures could benefit from revision or elaboration
to clarify ambiguities in the existing rule and eliminate
inconsistencies with overlapping specification imposed by other
programs, such as Acid Rain, NOX Budget, and NOX RACT. " The
issues commenter 2 identified as warranting review include data
validation procedures, continuous emissions monitoring system
(CEMS) configuration specifications, and methods of compliance
8-2
-------
determination. Additionally, commenter IV-D-54 recommended that
the EPA adopt the Title IV requirements whenever there are common
requirements between the NSPS and Title IV.
Data Validation. For data validation of the monitoring
process, commenter 2 indicated that the requirements for Subpart
Db differ compared to the criteria under the Acid Rain and NOX
Budget monitoring programs. Commenter 2 stated that two parallel
data processing systems and databases would have to be maintained
to meet the different requirements.
Definition of Operating Status. Commenter 2 reported that
the definition of the operating status of partial operating hours
also differs between Subpart Db and the Acid Rain and NOX Budget
Programs. Specifically, commenter 2 reported that for Subpart
Db, any hour in which a combustion unit is on-line for less than
30 minutes should be ignored for purposes of monitoring
compliance. However, both the Acid Rain and NOX Budget Programs
require that all operating time be accounted for, so that no CEMS
data can be ignored during on-line periods. As in the data
validation case stated above, the commenter indicated that these
differences lead to apparent reporting conflicts.
Span Value of the CEMS. Commenter 2 noted differences in
the span value of the CEMS. Commenter 2 explained that the span
value of the CEMS is the primary determinant of the allowable
daily calibration drift. Commenter 2 continued by stating that
Subpart Db establishes 500 ppm as the span for NOX if oil or gas
are being combusted, and 1000 ppm if coal is being fired. The
commenter reported that this value is "markedly lower than 500
ppm" for new boilers under the Acid Rain procedures. Commenter 2
concluded that it is possible for the measured daily calibration
error to be small enough to be acceptable under NSPS, but large
enough to trigger an out-of-control condition under Acid Rain,
NOX Budget or NOX RACT.
8-3
-------
Compliance Determination: Commenter 2 addressed the
difference in compliance determination depending on the fuel
used. Commenter 2 reported that under Subpart Db compliance is
evaluated on a 30-day rolling average basis; however, there are
different compliance determination periods for low-heat release
units that fire gas and distillate, and those that fire other
fuels. He pointed out this discrepancy, presumably to allow the
EPA to better coordinate the compliance determination between the
different fuel types.
Treatment of Emission Limits During Hours of Invalid Data
Collection. Commenter 2 reported that there is no explicit
guidance on the treatment of emission limits during hours of
invalid data collection. Commenter 2 recommended that if no
valid measurement data are available for an hour, then the
accompanying emission limit should also be omitted in the
calculation of the 30-day average. Done this way, commenter 2
indicated that the emission rate and emission limit would be
calculated using the same hourly data set (i.e., considering
valid CEMS hours only).
Response: A subpart Db boiler equipped with an acid rain
CEMS can use this CEMS as a subpart Db CEMS. The reports
generated by this CEMS must be generated according to the
provisions of subpart Db and submitted to the authority in charge
of the NSPS program, because the NSPS and acid rain programs have
different requirements and are managed by different authorities.
Regarding data validation procedures, the EPA headquarters
already maintains the acid rain data base and the AIRS data base,
which is suitable for reports from non-acid rain programs. In
addition, several States maintain their own data bases. The EPA
believes that the data validation issue should not lead to any
conflicts considering that the acid rain and the subpart Db
report formats must follow their own requirements.
8-4
-------
The EPA headquarters has addressed a few span-related issues
upon request and will continue this practice under the part 60
Finally, emission limits during hours of invalid data must
be met using other means than CEMS data according to the
applicable. As discussed above, the EPA has added language to
sections 60.47a(c) and 60.48b(b) to clarify the relationship
to demonstrate compliance with the part 60 standards.
8.4 AVERAGING PERIODS
Support 30-Day Averaging Period
Comment
stated that the 30-day rolling average period should be
sufficient to account for operating efficiency variability.
compliance requirements consistent under Subparts D and Da,
adding that Subpart D units should be allowed to demonstrate
IV-D-08 added that "it doesn't make sense that the averaging
period is shorter for an older standard."
: The EPA appreciates the commenters' support.
However, expanding the averaging period in subpart D is beyond
time.
8.4.2
Comment: Commenter IV-D-50 stated "EPA should recognize the
x emissions on a daily basis, evaluate the
x emissions from power plants on
emission limit on a 24-hr basis."
Commenters IV-D-19, IV-D-20, IV-D-24, IV-D-34, IV-D-39, IV-
-------
D-42, IV-D-44, IV-D-46, IV-D-52, and IV-D-65 expressed concern
that a 30-day rolling average may be insufficient to account for
variability in operating efficiency. Commenters IV-D-19, IV-D-
24, IV-D-34, IV-D-39, IV-D-42, IV-D-44, IV-D-46 and IV-D-65
recommended a 12-month averaging period, while commenter IV-D-52
recommended a 6-month period. The commenters explained that this
period is consistent with the NOX standards under the Acid Rain
Program. Commenters IV-D-19 and IV-D-65 explained that the 12-
month period will be "environmentally neutral" explaining that
the mass of emissions is equivalent regardless of the averaging
period. Another reason stated by commenters IV-D-19, IV-D-20,
and IV-D-65 was that a longer averaging period will allow further
opportunities for pollution prevention. Another reason given by
Commenter IV-D-39 was that the effect of partial load operation
on the efficiency can be accommodated with a 1-year averaging
period.
Response: The EPA has not proposed any change to the
averaging period in the NSPS, and will not do so now. As
demonstrated by the four facilities analyzed after proposal who
all meet the revised output standards, 30-days is sufficient to
account for operational variability.
8.5 SUPPORT ELECTRONIC FILING
Comment: Commenter IV-D-47 supported the EPA's attempt to
streamline reporting by allowing quarterly electronic reports and
consolidation of NSPS Subpart Da and Part 75 reporting
requirements. However, the commenter "does not believe that the
proposed language provides a meaningful standard for determining
when reports are acceptable or for resolving any of the issues
that are likely to arise in implementing consolidated reporting."
The commenter stated that if the EPA intends to limit the new
option to reporting under a specific format, the EPA should
propose language indicating that, and "commit to working with
8-6
-------
utilities to ensure that the format is consistent with existing
formats, like Part 75." Additionally, the commenter emphasized
optional."
Commenters IV-D-24, IV-D-37, IV-D-42, IV-D-44, IV-D-52, and
reporting for Subpart Da and Db units. The commenters stated
that this action will reduce the burden on affected units and
the Acid Rain Program requirements under 40 CFR 75. The
following recommendations were made:
reporting under Subpart Da (inlet SO and C02
concentrations, S02
electronic format along with outlet emission parameters.
data algorithm required by Part 75 should treat periods of
missing or invalid data for the purposes of electronic
(3) Existing Subpart D units be allowed to file
electronically, as they should not be required to file
switch to electronic reporting.
Commenter IV-D-54 recommended that the EPA adopt the Title
the NSPS and Title IV.
Commenter IV-D-59 noted that the voluntary provision for
additional reporting requirements than might already be required
for a specific source. The commenter noted that the opacity
submitting other required data in electronic format.
-------
Response: In general, the EPA supports electronic submittal
of the reports, provided that those reports are generated in the
format required under all applicable regulations and submitted to
the appropriate authorities. A facility choosing to submit
reports electronically must obtain an agreement from the EPA
Regional offices and the State authority.
The missing data procedures required by part 75 are not
acceptable under subpart Da.
As discussed above, the EPA has added language to section
60.47a(c) to clarify that "If the owner or operator has installed
a nitrogen oxides emission rate continuous emission monitoring
system (CEMS) to meet the requirements of part 75 of this chapter
and is continuing to meet the ongoing requirements of part 75 of
this chapter, that CEMS may be used to meet the requirements of
this section, except that the owner or operator shall also meet
the requirements of §60.49a. Data reported to meet the
requirements of §60.49a shall not include data substituted using
the missing data procedures in subpart D of part 75 of this
chapter, nor shall the data have been bias adjusted according to
the procedures of part 75 of this chapter." Similar language has
also been added to section 60.48b(b) to clarify the use of part
75 CEMS with subpart Db affected facilities.
8.6 NEW MONITORING AND PERFORMANCE TESTING REQUIREMENTS
At a February 18, 1998 meeting with representatives from the
Utility Air Regulatory Group and the National Mining Association,
representatives identified the following issue related to the
potential variability in data from monitoring systems. A
complete summary of this meeting is in the project docket.
Comment: Mr. Kanary noted that when the EPA Acid Rain
Division studied the problem of heat input they back-calculated
the heat input from the flow monitors and the carbon monoxide
monitors. They found that there was a difference of as much as
-------
20 percent between the CEM data and the fuel analysis. Mr.
Wilson reported that the differences vary from site-to-site
because the error depends on the location of the monitor and the
design of the ductwork. This difference is the result of flow
meter errors. Flow meters are designed for laminar flow, and
actual duct conditions are not laminar flow.
Mr. Wilson explained that to have an accurate flow meter
reading, the ductwork would need to be extremely long and
straight and use larger fans. This type of ductwork would be
extremely costly to implement, assuming that space was not an
issue, which it often is. As an alternative flow measurement
approach, Mr. Wilson recommended using venturi nozzle to measure
the pressure drop. Mr. Harrison concluded the discussion of this
topic by stating that there was no proposal on the table on how
to resolve the issue and added that he is requesting 3 weeks to
comment on whatever method the Agency proposes.
Response: Under Part 75, flow monitors are used with S02
CEMs to determine S02 mass emissions and to verify that utility
units meet their allowance obligations each year. Recently,
utilities have expressed concern that Test Method 2, EPA's
reference method for certifying flow monitors, may cause flow
monitors to read high under certain flow conditions. This could
cause S02 emissions to be over reported. Because flow monitors
and C02 CEMs are used by utilities to calculate heat rate, there
is also concern that some heat rates may be overestimated.
Because of these concerns, the Acid Rain Program conducted
three field studies this past summer at two gas-fired and one
coal-fired power plant to test the performance of seven probes
and several new procedures being considered for revisions to Test
Method 2. In addition, several wind tunnel studies have or will
be performed, including pre- and post-test probe calibrations at
North Carolina State University, post-test probe calibrations at
8-9
-------
the National Institute of Standards and Technology, possible
Reynolds Number (temperature) effects on probe calibration at a
variable density wind tunnel at the Massachusetts Institute of
Technology, and testing in an Electric Power Research Institute
^swirl' wind tunnel to determine probe performance under
controlled yaw and pitch conditions.
Work is under way on a draft findings report along with
draft Test Method revisions. These documents will be peer
reviewed in the next several months and will be the basis for
Test Method 2 revisions, which are expected to be published by
the end of 1998.
In the meantime, if a utility has swirling flow, and
suspects their flow monitor measurements, flow straighteners may
be installed, without pressure drop penalty in many instances.
Several utilities have installed flow straighteners and have
found reductions in their volumetric flow measurements and heat
rate disparity. Other utilities have reported improvements
through the use of automated implementations of Method 2 and
through taking measurements at more traverse points than the
minimum required under Method 1. Heat rate disparities can also
be reduced through tighter quality assurance of (1) fuel sampling
and analysis procedures to ensure that their calculations are not
biased low, and (2) of C02 CEMs to ensure that their measurements
are not biased high.
Finally, the EPA believes that new units, which are the
primary types of sources to be affected by the NSPS, can be
designed to overcome measurement problems.
8-10
-------
9.0 OTHER
9.1 COST, ENVIRONMENTAL, ENERGY, AND ECONOMIC IMPACTS
Comment: Commenters IV-D-31, IV-D-38, IV-D-42, IV-D-47, and
IV-D-52 stated that the EPA did not adequately perform the nonair
quality, health and environmental impact and energy requirements
analysis. Commenter IV-D-63 quoted Section 111 of the Clean Air
Act to say that rulemaking must balance the environmental and
cost factors. The commenter stated that this rulemaking "does
not produce any tangible health and welfare benefits," and would
cause an increase in energy costs.
Commenter IV-D-50 suggested that the EPA expand on the
environmental impacts discussed in Section VI of the proposal.
The commenter encouraged the EPA to note the role of NOX in local
and regional ozone formation.
Response: As demonstrated by the regulatory impact
analysis, the EPA believes that the impact analysis, with the
modifications conducted as the result of the evaluation of public
comments, is sufficient to support our regulatory decisions.
Comment: Commenter IV-D-40 urged the EPA not to adopt the
proposed NSPS because the rules could have a negative effect on
the cost of producing electricity, particularly in the case of
coal-fired electric utilities. The commenter asserted that an
increase in the cost of electricity would reduce the use of
electrotechnologies throughout the economy. The commenter
stated, "By reducing economic growth, the proposed rules would
hinder the ability of Americans to purchase a safer environment
and improved health care. And by raising the relative price of
electricity, the proposed rules would impede market penetration
of a wide array of technologies that could produce major benefits
for American health and safety."
Response: The EPA believes that the impacts of the
9-1
-------
standards have been adequately addressed, that the final rule is
justified based on these impacts, and that no revisions to the
impacts analysis are needed.
9.2 EDITORIAL
Comment: Commenter IV-D-27 pointed out that the equation in
section III.D. of the proposal preamble used to calculate the
output based standard is incorrect. The commenter stated that
the equation should be E0 = (EJ (n)/1000.
Response: The commenter correctly noted the error in the
equation. The EPA has used the correct equation in the actual
analyses.
Comment: Commenter IV-D-50 requested clarification of the
units for emission limits. The proposed units are pounds NOX per
MWh or MMBtu. The commenter noted that most NOX is emitted as NO
and a small fraction is N02. The commenter stated that it would
be "useful to clearly state that the NSPS limits are in pounds of
NOX as N02, as is traditionally done."
Response: As noted by the commenter, the current section
60.44a does not distinguish NOX as N02, nor did the proposed
amendments. However, section 60.44b, both in its current version
and as proposed does. In order to correct this discrepancy, the
EPA will revise the final section 60.44a to express NOX as N02.
Comment: Commenter IV-D-47 requested "that the regulatory
language in § 60.44a(d) explicitly state that the basis for
compliance is a 30-day rolling average." The commenter noted
that "The proposed rule is somewhat ambiguous, and the mention of
the 30-day rolling average in § 60.44a(a) on its face only
applies to the standards in § 60.44a(a)."
Response: The EPA agrees that the addition of the language
specifying that the emission limits are to be based on a 30-day
rolling average to section 60.44a(d) is a useful clarification.
This change will be made in the final rule.
9-2
-------
9.3 GLOBAL WARMING
Comment: Commenter IV-D-04 warned that "all air temperature
readings on which decisions on matters pertaining to global
warming are made are suspect." The commenter recommended that
the readings include humidity as a factor and decisions regarding
global warming and NOX control regulations be revised
accordingly. Further, the commenter warned against establishing
policy and regulations based on "suspect" data.
Response: The EPA appreciates the commenter's input. The
EPA does its best to ensure that its regulations are based on
sound science.
9.4 BEST AVAILABLE CONTROL TECHNOLOGY
Comment: Commenter IV-D-37 recommended that the EPA issue
guidance stating that gas-fired boilers and gas-fired combustion
turbines with duct-firing, which are able to meet the new NOX
standard, constitute best available control technology (BACT) for
the purposes of Prevention of Significant Deterioration (PSD)
permitting.
Response: This request is beyond the scope of the
rulemaking.
9.5 APPLICABILITY OF THE CREDIBLE EVIDENCE RULE
Comment: Commenter IV-D-47 noted that the EPA recently
amended Part 60 to include the credible evidence (CE) rule. The
commenter expressed opposition to the CE rule. Because the
proposed NSPS clearly indicates how compliance is determined, the
commenter believes "that the CE rule has no application to this
NSPS," and requested that the EPA clearly state so.
The commenter stated that if the EPA intends for the CE rule
to apply to this NSPS, the EPA must supplement the proposal and
allow additional comment regarding application of the CE rule.
The commenter requested clarification as to what kind of
information could be "considered to be evidence of a violation of
9-3
-------
this NSPS." The commenter also requested a discussion of the
effect of data other than compliance method data on the ability
of source owners or operators to determine compliance, and to
certify compliance under Title V, for this NSPS.
Response: The CE rule is applicable to all NSPS
regulations, as well as to other programs, as determined in the
rule. The public comment period for the CE rule was closed a few
years ago. The NOX limit compliance determination method under
this NSPS NOX rule is a NOX CEMS. In addition, other credible
evidence such as evidence of tampering with the CEMS or
destruction of valid data, could also be used to allege non-
compliance .
9.6 ADDITION OF TECHNICAL DOCUMENTS TO THE RECORD
Comment: Commenter IV-D-47 attached "a large number of
pertinent technical reports and studies that are not in the
rulemaking record" along with the comments. The commenter urged
EPA to "review and evaluate all pertinent technical literature,
not simply those papers that might support a preconceived
position."
Response: The EPA has considered all of the material
provided by commenters in developing the final rules.
9.7 FEDERAL INTERVENTION
Comment: Commenter IV-D-66 supported federal intervention
in regulations that improve the health and environment of the
American citizens.
Response: The EPA appreciates the commenter's support.
9-4
-------
TECHNICAL REPORT DATA
(Please read Instructions on reverse before completing)
1. REPORT NO. 2.
EPA-453/R-98-005
4. TITLE AND SUBTITLE
New Source Performance Standards, Subpart Da and Db - Summary
of Public Comments and Responses
7. AUTHOR(S)
9. PERFORMING ORGANIZATION NAME AND ADDRESS
EC/R Incorporated
2327 Englert Drive
Durham, NC 27513
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency
Emission Standards Division
Office of Air Quality Planning and Standards
Research Triangle Park, NC 2771 1
3. RECIPIENT'S ACCESSION NO.
5. REPORT DATE
September 1998
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-D6-0008
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
EPA/200/04
15. SUPPLEMENTARY NOTES
Published in conjunction with promugated revised air emission standards for NOx emissions from utility and
nonutility boilers. EPA Contact: James Eddinger, 919-541-5426
16. ABSTRACT
This document presents a summary of public comments regarding development of revised air emission
standards for NOx emissions from electric utility boilers and industrial boilers and EPA's responses to
those comments.
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
Nitrogen Oxide (NOx) Emissions
NOx Control Technology
Utility Boilers
Industrial Boilers
18. DISTRIBUTION STATEMENT
Release Unlimited
b. IDENTIFIERS/OPEN ENDED TERMS c. COSATT Field/Group
Air Pollution control
19. SECURITY CLASS (Report) 21. NO. OF PAGES
Unclassified 98
20. SECURITY CLASS (Page) 22. PRICE
Unclassified
EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDITION IS OBSOLETE
------- |