EMISSION FACTOR DOCUMENTATION
FOR AP-42 SECTION 1.4 NATURAL GAS COMBUSTION
Prepared for:
Office of Air Quality Planning and Standards
U.S. Environmental Protection Agency
Research Triangle Park, NC
Prepared by:
Eastern Research Group
1600 Perimeter Park
Morrisville, NC 27560
March 1998
-------
TABLE OF CONTENTS
Section Page
1.0 Introduction 1.1
1.1 Reasons For Updating 1.1
1.2 References For Section 1 1.2
2.0 Literature Search and Screening 2.1
2.1 Emission Data Quality Rating System 2.1
2.2 Review of Data Sets 2.3
2.3 References For Section 2 2.7
3.0 AP-42 Section Development 3.1
3.1 Revisions to Section Narrative 3.1
3.2 Pollutant Emission Factor Development 3.1
3.2.1 Database Design 3.1
3.2.2 Results of Data Analysis 3.5
3.3 Emission Factor Quality Rating System 3.8
3.4 Emission Factors 3.11
3.5 Peer Review Process 3.11
3.6 References for Section 3 3.11
4.0 AP-42 Section 1.4 4.1
Appendix A - Acid Rain Division Data
Appendix B - Reviewer Comments and EPA Responses
-------
TABLES
Page
2.2-1 SUMMARY OF REFERENCES USED IN THE REVISION OF SECTION 1.4 2.4
3.2-1 SNCR TEST RESULTS FOR WALL-FIRED BOILERS (NOX) 3.9
3.2-2 SNCR TEST RESULTS FOR TANGENTIAL-FIRED BOILERS (NOX) 3.10
3.4-1 SUMMARY OF EMISSION FACTORS FOR AP-42 SECTION 1.4 3.13
in
-------
Emission Factor Documentation for AP-42 Section 1.4
Natural Gas Combustion
1.0 Introduction
The revised AP-42 section described in this report replaces the section published in September
1996 as Supplement B to the Fifth Edition. This background report replaces the Emission Factor (EMF)
Documentation for AP-42 Section 1.4, Natural Gas Combustion, issued April 1993. The purpose of this
background report is to provide technical documentation supporting the Supplement D revisions to
AP-42 Section 1.4.
The EPA publishes emission factors in its Compilation of Air Pollutant Emission Factors, EPA
Publication No. AP-42 (AP-42). The document has been published since 1972 as the primary
compilation of EPA's emission factor information. Federal, State and local agencies, consultants, and
industry use the document to identify major contributors of atmospheric pollutants, develop emission
control strategies, determine applicability of permitting programs, and compile emission inventories for
ambient air impact analyses and State Implementation Plans (SIPs). Volume 1, Stationary Sources is
published by Emission Factor Inventory Group (EFIG) in EPA's Office of Air Quality Planning and
Standards (OAQPS).
1.1 Reasons For Updating
The Clean Air Act Amendments of 1990 added greatly to the number of air pollution sources for
which emission factor development was required, and also called for the improvement of existing factors.
There are several reasons for updating or revising AP-42 sections and emission factors.
• New Standard. After the proposal of a standard, the EPA reviews the available material
to determine if sufficient information has been gathered to support the development of
emission factors for the industry or process being studied. Oftentimes, the proposal or
development of a new standard for a source or source category will trigger a re-
evaluation of emission factors for a particular source. In the proposal of a standard, the
proposal team gathers tremendous amounts of data to support the standard, much more
data than is typically gathered for AP-42. The proposal team may compare their new
data with existing information used to develop AP-42 emission factors. If, in the
comparison, the team discovers a deficiency in the existing information, they may turn
their data over to EFIG, who in turn may use the information to improve emission
factors.
• Outside Requests. The EPA receives requests for better source and emission factor
information. Requests may come from other Office of Air Quality Planning and
Standards (OAQPS) branches, EPA laboratories and regional offices, State agencies,
trade associations, special interest groups, or private individuals. The requests may take
the form of directives, letters, oral inquiries, or comments on published emission factors.
• Improvement of the National Inventory. The EPA may determine that a particular source
category is a significant contributor to the National Inventory and that EPA should
develop or improve emission factors.
B01S04.WPD 1.1
-------
• New Information. New information will be useful that may have been developed
initially for Emission Standards Division (BSD) background documents involving new
source performance standards (NSPS), national emission standards for hazardous air
pollutants (NESHAP), and Control Techniques Guidelines (CTG), and reports by various
EPA laboratories.
• Contractor Expertise. A contractor or consultant may have gained expertise on a source
category during previous work, either for EPA or for other clients, and may warrant
considering a relatively low-expense update and expansion of available information.
Section 1.4 has been updated to incorporate new available data on this source category. New
information has been used to better characterize this source category, develop improved volatile organic
compound (VOC) and particulate matter (PM) emission factors, and update criteria pollutant emission
factors. In response to the upcoming NESHAP for this source category, an expanded hazardous air
pollutant (HAP) emission factor list has also been provided.
This background report consists of four sections. This introduction provides background
information on AP-42 and documents such as this one that are issued to update sections of AP-42.
Section 2 presents the data search and screening steps, discusses the references used to revise AP-42
Section 1.4, and defines the emissions data quality rating system. Section 3 discusses overall revisions
to AP-42 Section 1.4, provides details about the database built for storing the available data, presents the
calculations used to calculate emission factors, and defines the emission factor quality rating system.
Section 4 presents the proposed revision of the existing AP-42 section as it would appear in
Supplement D.
1.2 References For Section 1
1. Procedures For Preparing Emission Factor Documents, Third Revised Draft Version, Office Of
Air Quality Planning And Standards, U.S. EPA, Research Triangle Park, NC 27711,
November 1996.
1.2
-------
2.0 Literature Search and Screening
Data used in this section were obtained from a number of sources within OAQPS and from
outside organizations. The AP-42 background files were reviewed for information on these sources,
demonstrated pollution control technologies, and emissions data. The Factor Information Retrieval
System (FIRE) was searched for emission data on natural gas-fired combustion sources. The Source Test
Information Retrieval System (STIRS) data set, compiled by EFIG, was reviewed and provided emissions
data from several sources. The STIRS data set is a collection of emission test reports that have been
scanned and stored on CD-ROM.
In the review of available references, emissions data were accepted if:
• sufficient information about the combustion source and any pollution control devices
was given.
• the test report identified if the emissions tests were conducted before or after a pollution
control device.
• emission levels were measured by currently accepted test methods.
• emission test results were reported in units which could be converted into the reporting
units selected for this AP-42 section.
• sufficient data existed to characterize operating conditions.
2.1 Emission Data Quality Rating System1
After reviewing the test reports, it should be possible to assign a data quality rating to each
pollutant emission rate for each test series. The individual data quality ratings are not to be confused
with the overall emission factor ratings. The data quality ratings are an appraisal of the reliability of the
basic emission data that will be used to later develop the factor.
Test data quality is rated A through D, based on the following criteria:
A - Tests are performed by a sound methodology and are reported in enough detail for
adequate validation.
B - Tests are performed by a generally sound methodology, but lacking enough detail for
adequate validation.
C - Tests are based on an unproven or new methodology, or are lacking a significant amount
of background information.
D - Tests are based on a generally unacceptable method, but the method may provide an
order-of-magnitude value for the source.
2.1
-------
The quality rating of test data helps identify good data, even when it is not possible to extract a
factor representative of a typical source in the category from those data. For example, the data from a
given test may be good enough for a data quality rating of "A," but the test may be for a unique feed
material, or the production specifications may be either more or less stringent than at the typical facility.
In following the general guidelines discussed above, four specific criteria can be considered to
evaluate the emission data to ensure that the data are based on a sound methodology, and documentation
provides adequate detail. A test series is initially rated "A through D" in each of the following four
areas.
• Source operation. If the manner in which the source was operated is well documented in
the report, and the source was operating within typical parameters during the test, an A
rating should be assigned. If the report stated parameters were typical, but lacked
detailed information, a B rating is assigned. If there is reason to believe operation was
not typical, a C or D rating is assigned.
• Test method and sampling procedures. In developing ratings, the accuracy of the test
method as well as the adequacy of the documentation are considered. In general, if a
current EPA reference test method appropriate for the source was followed, the rating
should be higher (A or B). If other methods are used, an assessment is made of their
validity. If it is judged that the method was likely to be inaccurate or biased, a lower
rating (C or D) is given. A complete report should indicate whether any procedures
deviated from standard methods and explain any deviations. If deviations were reported,
an evaluation is made of whether these were likely to influence the test results.
• Sampling and process data. During testing, many variations can occur without warning
and sometimes without being noticed. Such variations can induce wide deviations in
sampling results. If a large spread between test run results cannot be explained by
information contained in the site test report or from test reports of other sources, the data
are suspect and are given a lower rating. However, it should be recognized that a process
may have highly variable emissions and a lower rating may not be appropriate solely on
the basis of wide deviations in sampling results.
• Analysis and calculations. Ideally, test reports should contain original raw data sheets
and other QA documentation. If there are data sheets, the nomenclature and equations
used are compared with those specified by EPA to establish equivalency. The depth of
review of the calculations is dictated by the reviewers' confidence in the ability and
conscientiousness of the tester, based on such factors as consistency of results and
completeness of other areas of the test report. Reports may indicate that raw data sheets
were available but were not included. If the test report is of high quality based on the
other criteria, the quality rating should not be lowered due to a lack of data sheets.
An overall emission data quality rating is developed considering the scores on the four criteria.
There is no precise equation for the relative weighting of the factors, because each report presents
different issues, and the rating system needs to provide flexibility to consider the strengths and
weaknesses of each test series and reach a judgment on the overall rating. However, the two criteria
concerning (1) the test method and sampling procedures and (2) the sampling and process data should be
2.2
-------
weighted most heavily. If either of these two criteria are assigned a low rating, this low rating should be
assigned as the overall data quality rating, no matter how complete the documentation is.
2.2 Review of Data Sets
A total of 42 documents were reviewed in the process of developing emission factors for this
revision to AP-42 Section 1.4, Natural Gas Combustion. A summary review of the references used to
develop emission factors and their associated database identification numbers is presented in Table 2-1,
following this section. The majority of the references which were used to revise the emission factors for
natural gas combustion sources were either compliance test reports or summaries of compliance test
results. Seven of the references used in the development of this data were the results of research or
specific information gathering efforts. Furthermore, NOX emission factors for several natural gas
combustion sources were developed from an electronic database received from the Acid Rain Division
(ARD) of EPA. The data received from the ARD, and the corresponding emission factor averages, are
presented in Appendix A.
References 2 Through 6. 8. and 41
References 2 through 6, reference 8, and reference 41 are the results of several research or
specific information gathering efforts on natural gas-fired boilers. The data extracted from these reports
make up the vast majority of all the HAP information contained in the revision of AP-42 Section 1.4.
Pollutants tested in references 2, 3, and 41 also included speciated poly cyclic aromatic hydrocarbons
(PAH) and speciated metals. The test results reported in these references were all from emission
measurements conducted on tangential- and wall-fired utility boilers. Most of the sources detailed in
these references were uncontrolled, however, some incorporated flue gas recirculation (FGR) forNOx
control. All of the emission test data contained in these references were assigned a rating of A due to the
detailed information provided.
References 7. 9 Through 40. and 43 Through 44
These references were all compliance test results from both utility and industrial boilers firing
natural gas. The majority of these compliance tests focused on NOX and CO emissions, however, several
tests included results of total hydrocarbon (THC), non-methane hydrocarbon (NMHC), methane, and
particulate matter (PM) measurements. Some of the boilers reported in these references were operated
with low-NOx burners, FGR, or selective non-catalytic reduction (SNCR) for NOX control. All of the
emission test data contained in these references were assigned a rating of A due to the detailed
information provided.
Reference 42
Reference 42 is a NOX emission summary for all national gas-fired utility boilers required to
submit CEM data to the ARD as required by Title IV of the Clear Air Act Amendments. These data
represent average NOX emissions from these boilers for the 3rd quarter of 1996. This data set included
NOX emissions from 121 wall-fired boilers, 62 tangential-fired boilers, and five wall-fired boilers with
low-NOx burners. The data received from ARD, and the corresponding emission factor averages, are
presented in Appendix A.
2.3
-------
Table 2.2-1. SUMMARY OF REFERENCES USED IN THE REVISION OF SECTION 1.4
Reference
Number3
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
General Information Concerning Document
Source Test on a Tangential-Fired Utility
Boiler
Source Test on a Wall-Fired Utility Boiler
Source Test on a Wall-Fired Utility Boiler
Source Test on a Wall-Fired Utility Boiler
Source Test on a Package Boiler
Compliance Test on a Package Boiler
Source Tests on Seven Wall-Fired Utility
Boilers
Source Test on a Wall-Fired Utility Boiler
With SNCR Control
Compliance Test on a Wall-Fired Utility
Boiler With SNCR Control
Compliance Test on Two Tangential-Fired
Utility Boilers With SNCR Control
Compliance Test on Two Wall-Fired Utility
Boilers With SNCR Control
Compliance Test on a Tangential-Fired
Utility Boiler With SNCR Control
Compliance Test on a Tangential-Fired
Utility Boiler With SNCR Control
Compliance Test on a Tangential-Fired
Utility Boiler With SNCR Control
Compliance Test on a Wall-Fired Utility
Boiler
Compliance Test on a Tangential-Fired
Utility Boiler With SNCR
Pollutants Tested
NOX, CO, speciated
HAP's, metals
NOX, CO, speciated
HAP's, metals
Benzene,
Formaldehyde
Benzene,
Formaldehyde
NOX, CO, Methane,
Ethane, PM
NOX, CO
Benzene,
Formaldehyde
NOX, CO,
Hydrocarbons, PM
NOX, CO,
Hydrocarbons, PM
NOX, CO,
Hydrocarbons, PM
NOX, CO,
Hydrocarbons
NOX, CO,
Hydrocarbons, PM
NOX, CO,
Hydrocarbons
NOX, CO,
Hydrocarbons
NOX, CO,
Hydrocarbons
NOX, CO,
Hydrocarbons
Data
Qualit
y
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
Database
I.D.
1
2
3
4
6
7
8,9, 10, 11,
12, 13, 14
15
23
17
18
19
20
22
16
21
2.4
-------
Table 2.2-1. SUMMARY OF REFERENCES USED IN THE REVISION OF SECTION 1.4
(Continued)
Reference
Number3
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
General Information Concerning Document
Compliance Test on a Boiler
Compliance Test on a Boiler
Compliance Test on a Boiler
Compliance Test on Two Boilers
Compliance Test on a Boiler
Compliance Test on Two Boilers
Compliance Test on a Boiler
Compliance Test on Two Boilers
Compliance Test on a Boiler
Compliance Test on a Boiler
Compliance Test on a Boiler
Compliance Test on a Boiler
Compliance Test on a Boiler
Source Test on a Boiler
Source Test on a Boiler
Compliance Test on a Boiler
Compliance Test on a Boiler
Compliance Test on a Boiler
Compliance Test on Two Boilers
Compliance Test on a Boiler
Compliance Test on a Boiler
Pollutants Tested
NOX, CO
NOX, CO
NOX, CO
NOX, CO
NOX, CO,
Hydrocarbons
NOX
NOX, CO,
Hydrocarbons
NOX
NOX, CO, PM
NOX, CO,
Hydrocarbons
NOX, CO
NOX, CO,
Hydrocarbons
NOX
NOX, CO
NOX, CO,
Hydrocarbons, PM
NOX
NOX
NOX, CO
NOX, CO
PM
NOX, PM
Data
Qualit
y
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
Database
I.D.
106
107
108
109
110
111
112
113
114
115
116
117
119
120
121
122
123
125
126
131
132
2.5
-------
Table 2.2-1. SUMMARY OF REFERENCES USED IN THE REVISION OF SECTION 1.4
(Continued)
Reference
Number3
39
40
41
42
43
44
General Information Concerning Document
Compliance Test on a Boiler
Compliance Test on Four Boilers
Source Tests on Two Wall-Fired and Two
Tangential-Fired Boilers
CEM Data Submitted to ARD
Compliance Test on One Boiler
Compliance Test on One Boiler
Pollutants Tested
NOX, CO,
Hydrocarbons
NOX, CO,
Hydrocarbons
NOX, CO, speciated
HAP's, metals
NOX,
PM
PM
Data
Qualit
y
A
A
A
A
A
A
Database
I.D.
133
134
200
Not in
Database
201
202
Reference number corresponds to the reference listing at the end of this section.
2.6
-------
2.3 References For Section 2
1. Procedures for Preparing Emission Factor Documents, Third Revised Draft Version, Office of
Air Quality Planning and Standards, U.S. EPA, Research Triangle Park, NC 27711,
November 1996.
2. PICES Field Chemical Emissions Monitoring Project Site 120 Emissions Report. Carnot, Tustin,
CA, December 1995.
3. PICES Field Chemical Emissions Monitoring Project Site 121 Emissions Report. Carnot, Tustin,
CA, December 1995.
4. Emission Inventory Testing at El Segundo Generating Station No. 1 for Southern California
Edison Company, Carnot, April 1990.
5. Air Toxics Emissions Inventory Testing at Alamitos Unit 5, Carnot, May 1990.
6. Gas Research Institute/WP Natural Gas @ Boise Cascade Timber and Wood Products Division
#2 Package Boiler, Amtest Air Quality, Inc., May 1995.
7. Source Test For Measurement Of Nitrogen Oxides And Carbon Monoxide Emissions From
Boiler Exhaust At GAF Building Materials, Pacific Environmental Services, Inc., Baldwin Park,
CA, May 1991.
8. Field Chemical Emissions Monitoring Project: Emissions Report For Sites 103-109.
Preliminary Draft Report. Radian Corporation, Austin, TX, March 1993. (EPRI Report)
9. Urea Permit Compliance Testing at Alamitos Generation Station Unit 2, Carnot, November
1992.
10. Emissions Source Test Report For Urea Injection Compliance Testing Huntington Beach Unit 1
Permit Application No. R-249463, Geraghty & Miller, March 1994.
11. SCE Etiwanda Units 1 and 2 Urea Compliance Source Test Report, Final Report, Volume 1 of II,
Radian Corporation, March 1994.
12. Source Test Report For Urea Permit Compliance Testing Redondo Beach Generating Station
Units 5 and 6, Sierra Environmental Engineering, Inc., October 1992.
13. Urea Permit Compliance Testing at Alamitos Generation Station Unit 4, Carnot, April 1993.
14. Urea Permit Compliance Testing at El Segundo Generating Station Unit 3, Carnot,
September 1993.
15. Emissions Source Test Report For Recirculation Gas By-Pass and Urea Compliance Testing
Etiwanda Unit 3 Permit Application No. 261513, Acurex Environmental, March 1994.
2.7
-------
16. Emissions Source test Report: Permit Application No. R-249462, Huntington Beach Generating
Station, Acurex Environmental, March 1996.
17. Urea Permit Compliance Testing at El Segundo Generating Station Unit 4, Carnot,
September 1993.
18. California Fruit Produce, Fresno, Ca. Boiler Emissions Test 12-4-92. Best Environmental, Inc.,
San Leandro, CA, December 17, 1992.
19. California Fruit Produce, Madera, Ca. Boiler Emissions Test 12-2-92. Best Environmental Inc.,
San Leandro, CA, December 17, 1992.
20. Emission Testing at Zacky Farms Kewanee Boiler, Dinuba, California. Steiner Environmental,
Inc., Bakersfield, CA, July 1993.
21. Compliance Test Report Determination ofNOx emission rates From Boilers 3, 4, and 5.
Harrison Radiator, Dayton, Ohio. Hayden Environmental Group, Inc., Miamisburg, OH, March
20, 1990.
22. R. F. MacDonald Source Emissions Testing at Tomatek, Inc. Ecoserve Environmental Services,
Inc. Pittsburg, CA, October 1989.
23. Nitrogen Oxide Emission Tests Boilers Number 4 and 5. Whiteman Air Force Base. Shell
Engineering and Associates, Inc., August 20 and 21, 1990.
24. Source Emissions Survey of Firestone Synthetic Rubber & Latex company Boiler EB-114
Exhaust Stack, Orange, Texas. METCO Environmental, Addison, TX, November 1990.
25. A Compliance Emission Test Report Determination of Nitrogen Oxides. Dual-Fuel Generating
Units Nos. 1 and 2. Greiner, Incorporated, Grand Rapids, MI, September 2, 1993.
26. Texaco Refining & Marketing, Inc. P. O. Box 1476, Bakersfield, California. Boilers A and B.
Annual Compliance Test. Steiner Environmental, Inc., Bakersfield, CA, June 19, 1992.
27. Source Emission Test for NOX, CO, and ROC From Conventional Steam Boiler at Thomas Plant,
Building 373, Naval Construction Battalion Center, Port Hueneme, California. Naval Energy
and Environmental Support Activity, October 1990.
28. Chevron U.S.A., Inc. Section 26C Steam Plant Steam Generator # 50-6 and 50-7. Initial
Compliance Test. Genesis Environmental Services Company, Bakersfield, CA, June 11, 1991.
29. Source Test for Measurement of Oxides of Nitrogen, Carbon Monoxide and VOCfrom Boiler
Exhaust at Candlewick Yarns, Lemoore, California. Pacific Environmental Services, Inc.,
Baldwin Park, CA, April 21, 1993.
30. Compliance Test for NOX Siemens Energy and Automation Natural Gas Fired Boiler #2. K&B
Design, Inc., August 26, 1994.
2.8
-------
31. Source Test Report Gibson 7028-01, Gibson Oil and Refining Company, Bakersfield, California.
Brown and Caldwell, Pleasant Hill, CA, September 11, 1992.
32. Source Test Report Gibson Oil and Refining Company, Inc. Bakersfield, California. Brown and
Caldwell, Emeryville, CA, May 14-17, 1990.
33. Compliance Test Report: Determination of Nitrogen Oxide Emissions, Annapolis Hospital
Westland Center Boilers #1, 2, and 3, OakwoodHospital, WestlandMichigan.
WW Engineering & Science, Grand Rapids, MI, November 1993.
34. Report on Compliance Testing for General Motors Corporation, Fort Wayne Assembly Plant,
Roanoke, Indiana, Clean Air Engineering,
35. Stella Cheese. P. O. Box 1379, Tulare, California. Superior Mohawk Boiler. Initial
Compliance Test. Steiner Environmental, Inc., Bakersfield, CA, July 30, 1993.
36. Crystal Geyser Water Company. 1233 East California Avenue. Bakersfield, California. Boiler
#1 & 2, Initial Compliance Test. Steiner Environmental, Inc., Bakersfield, CA,
February 26, 1993.
37. Results of the Emissions Testing Services at Minnesota Corn Processors. Mar shall Minnesota.
December 20-21, 1994. Nova Environmental Services, Inc., Chaska, MN, January 31, 1995.
38. Results of the July 27, 1994 Air Emission Compliance Testing of the No. 10 Boiler at the Virginia
Public Utilities Plant in Virginia, Minnesota. Interpoll Laboratories, Inc., Circle Pines, MN,
August 17, 1994.
39. Los Gatos Tomato Products Compliance Emissions Testing. Best Environmental, Inc.,
Hayward, CA, April 1991.
40. Gallo Winery Fresno Plant Boilers # 1, 2, 3, & 4 Emissions Compliance Testing. Best
Environmental, Inc., San Leandro, CA, May 1992.
41. Gas-Fired Boiler and Turbine Air Toxics Summary Report. Prepared by Carnot Technical
Services, Tustin, CA, For the Gas Research Institute and The Electric Power Research
Institute, August 1996.
42. NOX Emission Reporting for Utility Boilers for 3rd Quarter 1996. Acid Rain Division, U.S.
EPA.
43. Compliance Paniculate Matter Source Emissions Measurement Program: Nebraska Package
Boiler, Kimberly-Clark Corporation, Neenah, WI. Geraghty & Miller, Inc., July 1994.
44. Results of the September 14 and 15, 1994 Air Emission Compliance Tests on the No. 11 Boiler at
the Appleton Paper Plant in Combined Locks, WI. Interpoll Laboratories Inc., October 1994.
2.9
-------
3.0 AP-42 Section Development
3.1 Revisions to Section Narrative
The technical discussion in AP-42 Section 1.4 did not need major revisions because no
significant technological changes in this source category were identified since the last publication. Some
of the discussion on NOX and PM formation was revised to better describe emissions from this source
category.
3.2 Pollutant Emission Factor Development
3.2.1 Database Design
The emission data assembled for the development of natural gas combustion emission factors
were stored in a database except for the data received from ARD. A database approach was chosen to
easily access and manipulate the large amount of data collected for this section and to facilitate data
transfer within other concurrent projects at EPA. The design of this database was accomplished in
conjunction with the Industrial Combustion Coordinated Rulemaking (ICCR) effort ongoing within the
Emission Standards Division (ESD). Data entered under either of these projects were easily transferred
between databases. Furthermore, the common design of the database will allow for future additions to
the database and simple recalculation of emission factors.
Within the database, data were stored in two tables to reduce repetitive entry of data. These
tables, and the data fields associated with each table are as follows:
Facilities Table
Facility name
Location
Testing Company
Date of Test
Boiler Manufacturer
Boiler Type (wall-fired, tangential-fired, etc.)
Air Supply (forced draft, induced draft, balanced draft etc.)
Capacity (MW)
Load (percent of capacity)
Fuel Type
Fuel Higher Heating Value
Heat Input (MMBtu/hr)
Post-combustion Emission Controls
Application (electrical generation, process steam, etc.)
Test Data Table
Pollutant
Test Method
Pollutant Concentration (as reported)
Detection Limit
3.1
-------
• Exhaust Oxygen Percentage
• Data Rating
• Fuel Exhaust Factor (F-Factor)
• Exhaust Flow Rate
• Fuel Flow Rate
• Exhaust Moisture Fraction
• Molecular Weight of Pollutant
The database was programmed to merge the data in the two tables and calculate emission factors
for the available pollutants in units of pounds of pollutant per million standard cubic feet of fuel burned.
To ensure consistent calculation of emission factors, the database was programmed to use the emission
concentration data and process data taken during the testing period to calculate the emission factors.
Emission factors provided in test reports were not used. The EPA concluded that this method of
calculation would provide the highest quality emission factors. This method of calculating emission
factors was chosen because different methods of calculation emission factors were used in some of the
references and in some cases, the method of calculating emission factors was not given. Equations used
to calculate emission factors for this section were dependent on the pollutant concentration units.
The following equations were used to convert concentration data to the selected emission factors
used in this section.
For concentration in parts per million by volume - dry (ppmvd), the following equation was used:
(C ,*F* 1,020 *MW)
EF f= — *temperature correct!on*oxygen correction
(106*385.5)
For concentration in parts per million by volume - wet (ppmvw), the following equation was used:
_ (Cnnmvw*F*l,020*MW)
scf
EF f = ppmvw -_ - *temperature correct!on*oxygen correction
(10b*385.5)*(l-Wc)
For concentration in micrograms per dry standard cubic feet, the following equation was used:
(Cf*F* 1,020)
EF f = — * oxygen correction
(106*453.6)
3.2
-------
For concentration in parts per billion by volume - dry, the following equation was used:
(C ,d*F* 1,020*MW)
EF f = — *temperature correction*oxygen correction
(109*385.5)
For concentration in volume percent, the following equation was used:
(C./o*F*l,020*MW)
EF f = *temperature correct!on*oxygen correction
scf (100*385.5)
For concentration in nanograms per dry standard cubic feet, the following equation was used:
(C f*F* 1,020)
EF f = *oxygen correction
(109*453.6)
For concentration in grains/dscf, the following equation was used:
EF§cf = (C f*F* 1,020*1.43*10~4) * oxygen correction
For concentration in micrograms per dry standard cubic meter, the following equation was used:
__(CUBm*F*l,020)
scf
EF f = — Mgm — * oxygen correction
(10b*453.6*35.31)
Where:
Efscf = Emission factor (pounds per million standard cubic feet of fuel input)
Cppmvd = Concentration (parts per million by volume, dry)
Cppmvw = Concentration (parts per million by volume, wet)
C^gf = Concentration (micrograms per dry standard cubic foot)
Cppbvd = Concentration (parts per billion by volume, dry)
Co/0 = Concentration (percent by volume)
Cngf = Concentration (nanograms per dry standard cubic foot)
Cgrf = Concentration (grains per dry standard cubic foot)
C^g,,, = Concentration (micrograms per dry standard cubic meter)
F = F-Factor (dry standard cubic feet per million Btu)
MW = Molecular weight (pounds per pound-mole)
Tstd = Reference temperature of F-Factor
3.3
-------
%O2 = Percent of oxygen in exhaust, by volume
1,020 = Natural gas heating value (MMBtu per 106 scf)
385.5 = Volume occupied by 1 Ib-mole of gas at 68°F (standard cubic feet per Ib-mole)
60 = Conversion factor (minutes per hour)
Wc = Water volume fraction in exhaust
453.6 = Conversion factor (grams per pound)
1.43 * 10"4 = Conversion factor (pounds per grain)
35.31 = Conversion factor (dry standard cubic feet per dry standard cubic meter)
Temperature correction / CIQOD
£• r r j. J2o K.
for F-Factor
(to68°F)
Oxygen correction _ [ 20 9
(to 0% 02) = 20.9 - o/oCL
Detection Limits
Test results from several tests of trace organic and metallic compounds reported concentrations
below the method detection limits. If a detection limit was provided in the test report, EPA used that
information in the development of AP-42 emission factors. To effectively use this data, two methods
were employed. For cases where a portion of the test data for a specific pollutant were below the method
detection limit but other test data report detection of that compound, then one-half of the detection limit
was averaged with the detected concentrations to calculate of the emission factor for that pollutant. In
cases where all of the test data for a specific pollutant reported concentrations below the method
detection limit, the lowest detection limit was reported for the emission factor for that pollutant, and that
factor noted as a detection limit. If an emission factor for an individual boiler was developed from a
detection limit and the resulting emission factor was higher than the emission factors generated from
detected concentrations, the emission factor based on a detection limit was removed from the average.
The goal of this decision was to prevent an unusually high detection limit from artificially increasing an
average emission factor. These methods for addressing detection level issues were provided in the
Procedures For Preparing Emission Factor Documents.1
Calculation of Average Emission Factors
To provide average emission factors for these sources, the arithmatic average of the emission
factors from all tests on a specific source type was calculated in the database. For tests that consisted of
multiple runs, the arithmetic average of the runs was used to develop the emission factor of that test.
Individual tests were given equal weight in the calculation of average emission factors for each boiler
group. In the case of NOX data received from ARD, the quarterly average from each boiler was treated
like an individual test.
3.4
-------
Presentation of Data
Due to the size of the database, a printout of all the test data used to generate the boiler emission
factors in Section 1.4 is not presented. The NOX data provided by the Acid Rain Division is provided in
Appendix A. For the remaining data that was stored in the database, EPA is providing an electronic copy
of the database on the Technology Transfer Network (TTN). This decision has resulted in a substantial
decrease in paper needed for this background information document and will provide users with a more
detailed background data set for this section. Providing the database to the public will allow anyone to
use or augment the database for their individual needs, providing a substantial building block for anyone
interested in compiling an extensive database on natural gas-fired combustion sources. An electronic
copy of the database in Microsoft Access® format, can be downloaded from the TTN at
http://www.epa.gov/tnn/chief/. In this website, go to AP-42 and follow the main menu options to
locate and download the database file.
3.2.2 Results of Data Analysis
3.2.2.1 Source Category Selection and Data Review
An important step in emission factor development is to determine which emission sources are
similar enough to be grouped together and represented by a single emission factor. This is accomplished
by investigating what parameters influence emissions and should be used to establish distinct groups
within the natural gas combustion category. The emission factors for each test contained in the database
were analyzed to determine appropriate groupings.
N(X Emission Factors
Based on the analysis of available NOX data, this category was separated into four general
groups: large wall-fired boilers with a heat input greater than 100 MMBtu/hr, small boilers with a heat
input less than 100 MMBtu/hr, tangential-fired boilers, and residential furnaces. These groups were
further separated into the following subcategories:
Large Wall-Fired Boilers (> 100 MMBtu/hr)
Uncontrolled (pre-NSPS)
Uncontrolled (post-NSPS)
Controlled-Low-NOx burner
Controlled-Flue Gas Recirculation (FGR)
Small Boilers (< 100 MMBtu/hr)
Uncontrolled
Controlled-Low-NOx burner
Controlled-Low-NOx burner/FGR
• Tangential-Fired Boilers
Uncontrolled
Controlled-FGR
B01S04.WPD 3.5
-------
• Residential Furnaces
The designation of pre- and post-NSPS refers to boilers that are subject to 40 CFR 60 Subparts D
and Db. Post-NSPS units are boilers with greater than 250 MMBtu/hr of heat input that commenced
construction, modification, or reconstruction after August 17, 1971, and units with heat input capacities
between 100 and 250 MMBtu/hr that commenced construction, modification, or reconstruction after
June 19, 1984. Analysis of the NOX data showed that uncontrolled wall-fired boilers subject to the NSPS
have considerably lower NOX emissions that those not subject to the NSPS. Such a distinction was not
seen in the data for the tangential boilers and therefore they were not further subcategorized.
The NOX emission factors for the following categories were developed from data received from
ARD: large wall-fired uncontrolled, large wall-fired controlled-low NOX burners, and tangential -fired
uncontrolled. The ARD data were determined to be more representative of these categories than NOX
data taken from compliance and source tests. The ARD data were from all operating utility boilers in the
U.S. and averaged continuously over a three-month period. Since most of the data stored in the database
were from short-term compliance and source tests, and from a much smaller population of boilers, the
ARD data were used for categories where they were available. The NOX emission factors for the
remaining categories, where ARD data were unavailable, were developed from data stored in the
database.
The NOX emission factor for residential furnaces is based on test data from 41 sources.2"3 Since
no new data for NOX from residential furnaces were obtained during this revision, this factor remains
unchanged from the previous version of Section 1.4.
NoO Emission Factors
The emission factors for N2O from large wall-fired boilers is based on test data from five source
tests conducted at three separate locations.4"5 The N2O factor for the large wall-fired boilers with low-
NOX burners is based on two source tests.4"5 Since no new data for N2O were obtained during this
revision, these factors remain unchanged from the previous version of Section 1.4.
CO Emission Factors
Emission factors for CO were not grouped as extensively as the NOX emission factors. For the
wall-fired boiler groups, no clear correlation was observed between boiler type or size and CO emission
levels. CO emission factors for the wall-fired boilers showed wide scatter and average emission factors
developed for the distinct grouping were not consistent with expected values. The EPA believes that
boiler operation plays a more critical role in determining CO emissions than the boiler type. Therefore,
all CO data for wall-fired boilers were averaged to provide a single CO emission factor. For the
tangential-fired boilers, CO emission factors showed less scatter and were strongly dependent on boiler
type. Therefore, CO emission factors for tangential-fired boilers were grouped under the uncontrolled
and controlled-flue gas recirculation categories.
The CO emission factor from residential furnaces is based on test data from 41 sources.2"3 Since
no new data for CO from residential furnaces were obtained during this revision, this factor remains
unchanged from the previous version of Section 1.4.
3.6
-------
Organic Compound Emission Factors
Similar to CO emission factors from wall-fired boilers, organic compound emission factors
(TOC, VOC, methane, formaldehyde, etc.) showed wide scatter and no correlation was observed with
boiler type or size. The EPA believes that the randomness of the organic compound emission factors
from natural gas combustion sources is driven more by individual source operation than source type.
Therefore, the organic compound emission factors for natural gas combustion sources were averaged
across the entire source category to provide single factors for all sources covered by AP-42 Section 1.4.
3.2.2.2 Data Not Included in the Database
Several of the emission factors presented in AP-42 Section 1.4 are not calculated via a simple
averaging procedure in the database. These emission factors include TOC, VOC, PM, CO2, SO2, and
controlled emission factors. The next several sections will discuss the development of these emission
factors.
VOC and TOC Emission Factors
The VOC emission factor for this source category was calculated to correspond with EPA's
definition that VOC comprises total organic compounds excluding methane, ethane, and several
chlorinated and fluorinated compounds.1 Since VOCs cannot be measured directly, VOC emission
factors must be calculated from other organic measurements. Data on hydrocarbon emissions collected
for the revision of AP-42 Section 1.4 included as total hydrocarbons (THC) and non-methane
hydrocarbons (NMHC). Based on an evaluation of the quality and quantity of data available on
hydrocarbons, EPA determined that the NMHC data was the most representative for this source category.
Given the NMHC as the basis for calculating the VOC emission factor, the ethane emission factor was
subtracted and the formaldehyde emission factor added to the NMHC emission factor to provide an
estimate of the VOC emission factor. This calculation is shown below. The TOC emission factor was
estimated by adding the methane and formaldehyde emission factors to the NMHC emission factor. This
calculation is shown below. The data used in these calculations can be found in Table 3.4-1.
VOC = NMHC + Formaldehyde - Ethane
= 8.5 + 0.07 - 3.1
= 5.5 (lb/106 scf)
TOC = NMHC + Formaldehyde + Methane
= 8.5 + 0.07 + 2.3
= 10.9 (lb/106 scf)
3.7
-------
PM Emission Factors
For a limited number of tests, PM measurements were conducted. These PM measurements
included both condensable and filterable PM. As with the organic compounds emitted from natural gas
combustion sources, no correlation between combustion source type and PM emission levels could be
established. Therefore, the PM emission factors presented in AP-42 Section 1.4 are intended to represent
all natural gas combustion sources. To provide a total PM emission factor, the average condensible and
filterable PM fractions were added together. This calculation is shown below. The EPA has assumed
that all condensable and filterable PM resulting from natural gas combustion is less that 1 micrometer
(/mi) in diameter. Therefore, the total PM emission factor also provided an estimate of PM10, PM2 5, and
PMj o emissions from natural gas combustion sources. The EPA believes that these assumptions for PM
size are valid since natural gas does not contain ash and the nucleation of PM from combustion products
ill not yield particles larger than 1 /mi.
PM (Total) = PM (Condensable) + PM (Filterable) = PM10 = PM2 5 = VMl 0
= 5.7 + 1.9
= 7.6 (lb/106 scf)
CO. and SO.
As outlined in the Procedures for Preparing Emission Factor Documents,1 emission factors for
CO2 were calculated by mass balance. This approach was also taken for calculating SO2. Since the
carbon and sulfur content in pipeline-quality natural gas is fairly consistent, EPA believes that this is the
best method for calculating CO2 and SO2 emission factors. For CO2, it was assumed that approximately
100 percent of the fuel carbon was converted to CO2. For SO2, a 100 percent conversion of fuel sulfur
was assumed. The CO2 emission factor was based on a carbon weight percent in natural gas of
76 percent and the SO2 emission factor was based on a sulfur content in natural gas of 2,000 grains per
million standard cubic feet.
Selective Non-catalytic Reduction (SNCR) Controlled Emission Factors
Several of the data sources provided emissions data for sources operating with SNCR control.
To evaluate SNCR control efficiency, only tests where NOX measurements were taken upstream and
downstream of the ammonia or urea injection area were considered. This method was chosen to evaluate
SNCR performance while avoiding the effects of boiler performance, with respect to NOX emissions. To
estimate SNCR performance, NOX control efficiency was based on tests conducted upstream and
downstream of the control device. Thirty-three sets of upstream and downstream tests on SNCR
performance were evaluated. The SNCR performance data for wall-fired boilers are presented in
Table 3.2-1 and SNCR performance data for tangential-fired boilers are presented in Table 3.2-2. The
average NOX reduction efficiency achieved by SNCR control on wall-fired and tangential-fired units was
24 percent and 13 percent, respectively. These reduction efficiencies were also put in the footnotes to the
tables presented in Section 1.4 so these reduction efficiencies could be applied to the NOX emission
factor if necessary.
3.8
-------
3.3 Emission Factor Quality Rating System
The quality of the emission factors developed from analysis of the test data was rated using the
following general criteria:
A—Excellent: Developed only from A-rated test data taken from many randomly chosen
facilities in the industry population. The source category is specific enough that variability within the
source category population may be minimized.
B—Above average: Developed only from A-rated test data from a reasonable number of
facilities. Although no specific bias is evident, it is not clear if the facilities tested represent a random
sample of the industries. The source category is specific enough that variability within the source
category population may be minimized.
C—Average: Developed only from A- and B-rated test data from a reasonable number of
facilities. Although no specific bias is evident, it is not clear if the facilities tested represent a random
sample of the industry. The source category is specific enough that variability within the source category
population may be minimized.
D—Below average: The emission factor was developed only from A- and B-rated test data from a
small number of facilities, and there is reason to suspect that these facilities do not represent a random
sample of the industry. There also may be evidence of variability within the source category population.
Limitations on the use of the emission factor are always noted in the emission factor table.
Table 3.2-1. SNCR TEST RESULTS FOR WALL-FIRED BOILERS (NOX)
Database
I.D.
16.1/16.2
16.3/16.4
16.5/16.6
23.1/23.2
23.3/23.4
23.5/23.6
15.1/15.2
15.3/15.6
15.8/15.7
18.2/18.1
18.4/18.3
18.6/18.5
18.7/18.8
18.9/18.10
18.12/18.11
Uncontrolled Emission
Factor (lb/106scf)
1.32E+02
8.14E+01
5.57E+01
1.12E+02
8.20E+01
5.24E+01
1.78E+02
1.08E+02
1.79E+02
1.97E+02
1.03E+02
5.29E+01
1.76E+02
1.01E+02
7.91E+01
Controlled Emission
Factor (lb/106scf)
1.17E+02
6.31E+01
4.53E+01
9.64E+01
5.96E+01
4.10E+01
1.29E+02
9.25E+01
1.51E+02
1.30E+02
7.76E+01
3.08E+01
1.25E+02
7.79E+01
4.81E+01
Percent Reduction
(%)
11
23
19
14
27
22
27
14
16
34
25
42
29
23
39
Average = 24
3.9
-------
Table 3.2-2. SNCR TEST RESULTS FOR TANGENTIAL-FIRED BOILERS (NOX)
Database
ID.
20.2/20.1
20.4/20.3
20.6/20.5
21.1/21.2
21.4/21.3
21.6/21.5
22.2/22.3
22.5/22.6
17.1/17.2
17.10/17.9
17.12/17.11
17.4/17.3
17.6/17.5
17.8/17.7
19.2/19.3
19.6/19.8
19.7/19.8
19.10/19.9
Uncontrolled
Emission Factor
(lb/106 scf)
5.45E+01
8.21E+01
9.08E+01
6.63E+01
9.36E+01
1.05E+02
6.83E+01
4.08E+01
6.70E+01
7.39E+01
8.70E+01
5.42E+01
7.16E+01
8.36E+01
8.38E+01
4.35E+01
4.35E+01
4.79E+01
Controlled
Emission Factor
(lb/106 scf)
4.70E+01
6.87E+01
8.12E+01
5.93E+01
7.77E+01
9.42E+01
5.82E+01
3.53E+01
6.47E+01
5.97E+01
7.34E+01
4.79E+01
4.43E+01
7.64E+01
7.40E+01
4.10E+01
4.10E+01
4.38E+01
Average =
Percent Reduction
(%)
14
16
11
10
17
10
15
13
o
6
19
16
12
38
9
12
6
6
9
13
E—Poor: The emission factor was developed from C- and D-rated test data, and there is reason to
suspect that the facilities tested do not represent a random sample of the industry. There also may be
evidence of variability within the source category population. Limitations on the use of these factors are
always noted, in the emission factor table.
The above criteria for emission factor ratings are defined in and OAQPS document which
provided guidance for preparing emission factor documents. The use of these criteria is somewhat
subjective and depends to an extent upon the individual reviewer. As these criteria were applied to the
emission factors, the term "number of facilities" was interpreted to mean "number of different boilers".
This criteria prevented emission factors generated from multiple tests on a single
boiler from receiving higher emission factor ratings.
Emission factors for this section were rated in the following manner:
A-Rated Emission factor average based on results of A-rated data from 20 or
more different boilers, or from approved mass balance calculations.
B-Rated Emission factor average based on results of A-rated data from 10 to
19 different boilers.
C-Rated Emission factor average based on results of A-rated data from five to
3.10
-------
nine different boilers.
D-Rated Emission factor average based on results of A-rated data from three to four
different boilers.
E-Rated Emission factor based on less than three A- or B-rated source tests.
In several cases for the revision of AP-42 Section 1.4, the data did not show a strong enough
correlation to boiler type, boiler size, or combustion control to justify the grouping of data by these
parameters. Where data were averaged across these parameters, the resulting emission factors were rated
by the above criteria but subsequently lowered one rating. The decision was made to lower the emission
factor rating in these cases to reflect the lack of certainty in the resulting emission factor.
3.4 Emission Factors
The emission factors for the sources covered in Section 1.4 of the AP-42 document are presented
in Table 3.4-1. This table provides the number of source tests used in calculating the various emission
factors as well as the relative standard deviation associated with each emission factor. This additional
information is intended to provide greater insight to the reader about the background of each emission
factor. For further detail on each emission factor, the database used to generate most of these factors
(except for NOX emission factors generated from ARD data) is provided on the TTN (See Section 3.2.1 of
this document for more details on the database). For NOX emission factors generated from data provided
by the Acid Rain Division, the supporting data is provided in Appendix A.
3.5 Peer Review Process
Part of the development processes of an AP-42 section includes review by a peer group. This
group include individuals from EPA, industry, and environmental organizations. In the peer review
process, EPA gains an extra level of confidence in the final version of a section. Comments received on
the draft version of a section are reviewed to determine if they warrant any changes to the draft version of
the section before it becomes final. Appendix B presents the substantial comments received on the draft
AP-42 Section 1.4 and EPA's responses to those comments.
3.6 References for Section 3
1. Procedures for Preparing Emission Factor Documents, EPA-454/R-95-015, Office of
Air Quality Planning and Standards, U.S. EPA, Research Triangle Park, North Carolina 27711,
September 1997.
2. Muhlbaier, J.L. "Particulate and Gaseous Emissions from Natural Gas Furnaces and Water
Heaters," Journal of the Air Pollution Control Association, December 1981.
3. Evaluation of the Pollutant Emissions from Gas-Fired Forced Air Furnaces: Research Report
No. 1503, American Gas Association Laboratories, Cleveland, OH. May 1975.
4. Nelson, L.P., L.M. Russell, J. J. Watson. "Global Combustion Sources of Nitrous Oxide
Emissions," Research Project 2333-4 Interim Report. Radian Corporation, Sacramento,
California. 1991.
3.11
-------
5. Peer, R.L., E.P. Epner, R.S. Billings. "Characterization of Nitrous Oxide Emission Sources,"
Prepared for U.S. EPA Contract 68-D1-0031. Radian Corporation, Research Triangle Park,
North Carolina. 1995.
3.12
-------
Table 3.4-1. SUMMARY OF EMISSION FACTORS FOR AP-42 SECTION 1.4
Pollutant
2-Methylnaphthalene
3 -Methylchloranthrene
7, 12-Dimethylbenz(a)anthracene
Acenaphthene
Acenaphthylene
Anthracene
Arsenic
Barium
Benz(a)anthracene
Benzene
Benzo(a)pyrene
Benzo(b)fluoranthene
Benzo(g,h,i)perylene
Benzo(k)fluoranthene
Beryllium
Butane
Cadmium
Chromium
Chrysene
CO (Wall-Fired)
CO (Tangential-Uncontrolled)
CO (Tangential-FGR)
Cobalt
Copper
Dibenzo(a,h)anthracene
Dichlorobenzene
Ethane
Fluoranthene
Fluorene
Formaldehyde
Hexane
Indeno(l,2,3-cd)pyrene
Lead
Manganese
Mercury
Methane
Molybdenum
Naphthalene
Number of Tests
4
1
1
1
1
1
2
3
1
17
3
5
1
49
17
7
2
4
1
1
4
1
2
22
2
1
4
2
2
42
2
2
Emission Factor
(lb/106scf)
2.4E-5
<1.8E-6
<1.6E-5
<1.8E-6
<1.8E-6
<2.4E-6
2.0E-4
4.4E-3
<1.8E-6
2.1E-3
<1.2E-6
<1.8E-6
<1.2E-6
<1.8E-6
<1.2E-5
2.1
1.1E-3
1.4E-3
<1.8E-6
84
24
98
8.4E-5
8.5E-4
<1.2E-6
1.2E-3
3.1
3.0E-6
2.8E-6
8.1E-2
1.8
<1.8E-6
4.6E-4
3.8E-4
2.6E-4
2.3
1.1E-3
6.1E-4
Relative Standard
Deviation (%)
72.77%
22.36%
38.85%
172.00%
166.72%
55.69%
124.00%
179.00%
57.00%
63.59%
49.36%
43.77%
14.02%
194.00%
95.61%
77.61%
2.53%
43.50%
118.83%
64.41%
85.19%
3.13
-------
Table 3.4-1. SUMMARY OF EMISSION FACTORS FOR AP-42 SECTION 1.4 (Continued)
Pollutant
Nickel
NMHC
NOx (Small-Unc.)
NOx (Small-Low NOx)
NOx (Small-Low NOx/FGR)
NOx (Large Wall-Fired-Low NOx)
NOx (Large Wall-Fired-FGR)
NOx (Large Wall-Fired Unc. Pre-NSPS)
NOx (Large Wall-Fired Unc. Post-NSPS)
NOx (Tangential -Unc.)
NOx (Tangential-FGR)
Pentane
Phenanthrene
PM, Condensible
PM, Filterable
Propane
Pyrene
Selenium
Toluene
Vanadium
Zinc
Number of Tests
5
48
18
5
15
5
4
108
13
62
8
1
4
4
21
1
1
1
11
3
1
Emission Factor
(lb/106scf)
2.1E-3
8.5
104
50
32
136
101
275
192
167
76
2.6
1.7E-5
5.7
1.9
1.6
5.0E-6
<2.4E-5
3.4E-3
2.3E-3
2.9E-2
Relative Standard
Deviation (%)
72.26%
150.26%
51.00%
54.00%
18%
37.00%
25.00%
93.00%
36.00%
37.00%
64.00%
63.82%
69.79%
111.47%
93.00%
71.77%
3.14
-------
4.0 AP-42 Section 1.4
4.1
-------
APPENDIX A
Acid Rain Division Data
-------
UNCONTROLLED NOx EMISSION DATA FOR LARGE PRE-NSPS WALL-FIRED BOILERS
util code utility
3 892 City of Coffeyville Mun. Lght & Pow
814 ENTERGY
44372 TU Electric
19804 City of Vero Beach
1167 Baltimore Gas and Electric Company
22500 Western Resources, Inc.
15474 Central and South West Services
14354 Illinois Power
814 ENTERGY
44372 TU Electric
44372 TU Electric
17718 Southwestern Public Service Co.
814 ENTERGY
16572 Salt River Project Ag. Imp. & Power
14534 City of Pasadena, Water & Power Dep
22500 Western Resources, Inc.
15474 Central and South West Services
44372 TU Electric
3278 Central and South West Services
44372 TU Electric
14063 Oklahoma Gas & Electric Co.
44372 TU Electric
22500 Western Resources, Inc.
14063 Oklahoma Gas & Electric Co.
3278 Central and South West Services
17698 Central and South West Services
44372 TU Electric
44372 TU Electric
plant
Coffeyville
Harvey Couch
Handley
Vero Beach Municipal
Riverside
Murray Gill
Southwestern
Gadsby
Lake Catherine
Parkdale
Lake Creek
Plant X
Harvey Couch
Kyrene
Broadway
Murray Gill
Southwestern
Eagle Mountain
Lon C Hill
Morgan Creek
Mustang
Parkdale
Murray Gill
Muskogee
Lon C Hill
Knox Lee
Mountain Creek
Eagle Mountain
Average heat input
state (MMBtu/hr)
KS
AR
TX
FL
MD
KS
OK
UT
AR
TX
TX
TX
AR
AZ
CA
KS
OK
TX
TX
TX
OK
TX
KS
OK
TX
TX
TX
TX
283
129
518
337
308
251
149
438
506
509
519
348
502
391
145
381
149
736
447
449
726
620
428
826
433
436
704
1051
nox rate-3Q
(Ib/MMBtu)
0.155
0.282
0.403
0.124
0.338
0.211
0.278
0.104
0.24
0.339
0.282
0.347
0.1
0.28
0.097
0.165
0.257
0.509
0.254
0.412
0.302
0.41
0.224
0.303
0.222
0.324
0.237
0.29
nox rate-3Q
(Ib/MMscf)
158
288
411
126
345
215
284
106
245
346
288
354
102
286
99
168
262
519
259
420
308
418
228
309
226
330
242
296
A-l
-------
UNCONTROLLED NOx EMISSION DATA FOR LARGE PRE-NSPS WALL-FIRED BOILERS (CONTINUED)
util code utility
16572 Salt River Project Ag. Imp. & Power
16572 Salt River Project Ag. Imp. & Power
44372 TU Electric
6958 City of Garland
14063 Oklahoma Gas & Electric Co.
44372 TU Electric
44372 TU Electric
44372 TU Electric
22500 Western Resources, Inc.
14063 Oklahoma Gas & Electric Co.
44372 TU Electric
44372 TU Electric
44372 TU Electric
20404 Central and South West Services
44372 TU Electric
17609 Southern California Edison Co.
22500 Western Resources, Inc.
3265 Central Louisiana Electric Co., Inc
13407 Nevada Power Company
44372 TU Electric
44372 TU Electric
20404 Central and South West Services
6616 Fort Pierce Utilities Auth
20391 WestPlains Energy
20391 WestPlains Energy
14063 Oklahoma Gas & Electric Co.
16604 City Public Service
6958 City of Garland
2172 Brazos Electric Power Cooperative,
plant
Agua Fria
Agua Fria
Parkdale
Ray Olinger
Horseshoe Lake
Stryker Creek
Mountain Creek
Permian Basin
Murray Gill
Mustang
Lake Creek
Morgan Creek
North Lake
Paint Creek
Graham
Cool Water
Gordon Evans
Coughlin
Clark
North Lake
Valley
Oak Creek
Henry D King
Arthur Mullergren
Cimarron River
Horseshoe Lake
W B Turtle
C E Newman
North Texas
Average heat input
state (MMBtu/hr)
AZ
AZ
TX
TX
OK
TX
TX
TX
KS
OK
TX
TX
TX
TX
TX
CA
KS
LA
NV
TX
TX
TX
FL
KS
KS
OK
TX
TX
TX
742
752
648
630
842
1050
779
185
315
396
1280
1233
770
169
1579
534
546
488
361
1051
1114
548
189
378
394
1275
851
86
245
nox rate-3Q
(Ib/MMBtu)
0.25
0.25
0.369
0.187
0.189
0.36
0.5
0.26
0.263
0.546
0.28
0.29
0.173
0.137
0.29
0.098
0.225
0.321
0.262
0.24
0.24
0.209
0.198
0.12
0.219
0.137
0.131
0.434
0.299
nox rate-3Q
(Ib/MMscf)
255
255
376
191
193
367
510
265
268
557
286
296
176
140
296
100
230
327
267
245
245
213
202
122
223
140
134
443
305
A-2
-------
UNCONTROLLED NOx EMISSION DATA FOR LARGE PRE-NSPS WALL-FIRED BOILERS (CONTINUED)
util code utility
44372 TU Electric
7294 City of Glendale, Public Service De
17609 Southern California Edison Co.
13407 Nevada Power Company
20447 Western Farmers Electric
44372 TU Electric
8901 Houston Lighting & Power Company
11269 Lower Colorado River Authority
18445 Electric Operations
3265 Central Louisiana Electric Co., Inc
1015 City of Austin Electric Utility Dpt
5063 CityofDenton
44372 TU Electric
10620 City of Lake Worth
22500 Western Resources, Inc.
15474 Central and South West Services
44372 TU Electric
16463 Ruston Utilities System
20447 Western Farmers Electric
2172 Brazos Electric Power Cooperative,
11269 Lower Colorado River Authority
203 91 WestPlains Energy
7634 City of Greenville
44372 TU Electric
44372 TU Electric
2442 Bryan Utilities
20404 Central and South West Services
814 ENTERGY
20813 CityofWinfield
plant
Handley
Grayson
Cool Water
Sunrise
Mooreland
North Lake
Webster
Sim Gideon
S O Purdom
Coughlin
Holly Street
Spencer
Morgan Creek
Tom G Smith
Gordon Evans
Southwestern
Valley
Ruston
Mooreland
R W Miller
Sim Gideon
Judson Large
Powerlane Plant
Tradinghouse
Graham
Bryan
Rio Pecos
Lake Catherine
East 12Th St
Average heat input
state (MMBtu/hr)
TX
CA
CA
NV
OK
TX
TX
TX
FL
LA
TX
TX
TX
FL
KS
OK
TX
LA
OK
TX
TX
KS
TX
TX
TX
TX
TX
AR
KS
2383
161
709
379
252
2025
1823
664
272
797
810
354
3671
158
1514
576
3312
42
648
499
659
600
31
3111
2161
139
912
2469
194
nox rate-3Q
(Ib/MMBtu)
0.281
0.06
0.106
0.354
0.323
0.28
0.237
0.202
0.202
0.301
0.157
0.334
0.591
0.234
0.409
0.372
0.25
0.182
0.213
0.175
0.189
0.159
0.136
0.335
0.42
0.211
0.384
0.22
0.261
nox rate-3Q
(Ib/MMscf)
287
61
108
361
329
286
242
206
206
307
160
341
603
239
417
379
255
186
217
179
193
162
139
342
428
215
392
224
266
A-3
-------
UNCONTROLLED NOx EMISSION DATA FOR LARGE PRE-NSPS WALL-FIRED BOILERS (CONTINUED)
util code utility
17568 South Missisippi Elec. Power Assoc.
17568 South Missisippi Elec. Power Assoc.
17568 South Missisippi Elec. Power Assoc.
14063 Oklahoma Gas & Electric Co.
44372 TU Electric
17698 Central and South West Services
18445 Electric Operations
2777 Cajun Electric Power Cooperative
3265 Central Louisiana Electric Co., Inc
44372 TU Electric
44372 TU Electric
17698 Central and South West Services
20404 Central and South West Services
6909 Gainesville Regional Utilities
2777 Cajun Electric Power Cooperative
14063 Oklahoma Gas & Electric Co.
44372 TU Electric
2172 Brazos Electric Power Cooperative,
5063 CityofDenton
44372 TU Electric
44372 TU Electric
20404 Central and South West Services
plant
Moselle
Moselle
Moselle
Seminole
Lake Hubbard
Wilkes
Arvah B Hopkins
Big Cajun 1
Teche
Eagle Mountain
Valley
Wilkes
Paint Creek
Deerhaven
Big Cajun 1
Seminole
Tradinghouse
R W Miller
Spencer
Permian Basin
Lake Hubbard
Fort Phantom
Average heat input
state (MMBtu/hr)
MS
MS
MS
OK
TX
TX
FL
LA
LA
TX
TX
TX
TX
FL
LA
OK
TX
TX
TX
TX
TX
TX
454
486
434
1806
2198
1759
433
925
1758
2021
2276
1653
521
638
657
1870
4972
993
418
3929
2844
966
nox rate-3Q
(Ib/MMBtu)
0.323
0.303
0.28
0.167
0.17
0.299
0.239
0.437
0.22
0.17
0.161
0.263
0.309
0.151
0.347
0.188
0.441
0.36
0.294
0.873
0.214
0.331
nox rate-3Q
(Ib/MMscf)
329
309
286
170
173
305
244
446
224
173
164
268
315
154
354
192
450
367
300
890
218
338
Pre-NSPS Average Nox (Ib/MMscf) = 275
(Ib/MMBtu) = 0.27
A-4
-------
UNCONTROLLED NOX EMISSION DATA FOR LARGE POST-NSPS WALL-FIRED BOILERS (CONTINUED)
util code utility
20447 Western Farmers Electric
14063 Oklahoma Gas & Electric Co.
3278 Central and South West Services
44372 TU Electric
6616 Fort Pierce Utilities Auth
9096 Lafayette Utilities System
44372 TU Electric
6958 City of Garland
18445 Electric Operations
7634 City of Greenville
44372 TU Electric
20404 Central and South West Services
5109 Detroit Edison Company
plant
Mooreland
Seminole
La Palma
Decordova
Henry D King
Doc Bonin
Handley
Ray Olinger
Arvah B Hopkins
Powerlane Plant
Handley
Fort Phantom
Greenwood
Average heat input
state (MMBtu/hr)
OK
OK
TX
TX
FL
LA
TX
TX
FL
TX
TX
TX
MI
721
1773
1005
5148
291
822
2629
1009
1510
162
2577
1189
2483
nox rate-3Q
(Ib/MMBtu)
0.224
0.205
0.272
0.324
0.121
0.252
0.15
0.177
0.187
0.097
0.12
0.122
0.19
nox rate-3Q
(Ib/MMscf)
228
209
277
330
123
257
153
181
191
99
122
124
194
Post-NSPS Average NOx (Ib/MMscf) =192
(lb/MMBtu)= 0.19
A-5
-------
NOx EMISSION DATA FOR WALL-FIRED BOILERS WITH LOW NOx BURNERS
Average heat input NOx rate-3Q NOx rate-3Q
utilcode utility plant state (MMBtu/hr) (Ib/MMBtu) (Ib/MMscf)
2507 City of Burbank - Public Service De Magnolia CA 105 0.108 110
14534 City of Pasadena, Water & Power Dep Broadway CA 145 0.107 109
2507 City of Burbank-Public Service De Olive CA 147 0.082 84
3265 Central Louisiana Electric Co., Inc Rodemacher LA 1737 0.203 207
13998 Ohio Edison Company Edgewater OH 380 0.167 170
Average NOx (Ib/MMscf) = 136
(Ib/MMBtu) = 0.13
A-6
-------
UNCONTROLLED NOx EMISSION DATA FOR TANGENTIAL FIRED BOILERS
util code utility
9726 GPU Generation Corporation
195 Alabama Power Company
12686 Mississippi Power Company
3249 Central Hudson Gas & Electric Corp.
14354 Illinois Power
12686 Mississippi Power Company
17718 Southwestern Public Service Co.
803 Arizona Public Service Company
6452 Florida Power & Light Company
3249 Central Hudson Gas & Electric Corp.
803 Arizona Public Service Company
6452 Florida Power & Light Company
44372 TU Electric
17718 Southwestern Public Service Co.
14354 Illinois Power
17609 Southern California Edison Co.
17698 Central and South West Services
12686 Mississippi Power Company
17718 Southwestern Public Service Co.
24211 Tucson Electric Power Company
17609 Southern California Edison Co.
803 Arizona Public Service Company
17698 Central and South West Services
803 Arizona Public Service Company
803 Arizona Public Service Company
24211 Tucson Electric Power Company
7806 Entergy Corporation
17698 Central and South West Services
plant
Gilbert
Chickasaw
Sweatt
Danskammer
Gadsby
Sweatt
Plant X
Saguaro
Cutler
Danskammer
Saguaro
Cutler
Collin
Plant X
Gadsby
San Bernardino
Lieberman
Jack Watson
Cunningham
Irvington
San Bernardino
Yuma Axis
Lieberman
Ocotillo
Ocotillo
Irvington
R S Nelson
Arsenal Hill
Average heat input
state (MMBtu/hr)
NJ
AL
MS
NY
UT
MS
TX
AZ
FL
NY
AZ
FL
TX
TX
UT
CA
LA
MS
NM
AZ
CA
AZ
LA
AZ
AZ
AZ
LA
LA
505
256
344
205
392
346
605
720
518
381
622
919
753
529
624
395
471
309
502
363
393
343
424
598
561
367
810
505
nox rate-3Q
(Ib/MMBtu)
0.238
0.168
0.335
0.08
0.093
0.325
0.125
0.335
0.083
0.102
0.219
0.079
0.139
0.158
0.08
0.1
0.15
0.197
0.225
0.147
0.103
0.071
0.14
0.147
0.138
0.185
0.161
0.134
nox rate-3Q
(Ib/MMscf)
243
171
342
82
95
332
128
342
85
104
223
81
142
161
82
102
153
201
230
150
105
72
143
150
141
189
164
137
A-7
-------
UNCONTROLLED NOx EMISSION DATA FOR TANGENTIAL FIRED BOILERS (CONTINUED)
util code utility
12686 Mississippi Power Company
1015 City of Austin Electric Utility Dpt
8901 Houston Lighting & Power Company
24211 Tucson Electric Power Company
12686 Mississippi Power Company
17718 Southwestern Public Service Co.
17698 Central and South West Services
17718 Southwestern Public Service Co.
22500 Western Resources, Inc.
17718 Southwestern Public Service Co.
44372 TU Electric
16604 City Public Service
17718 Southwestern Public Service Co.
6958 City of Garland
44372 TU Electric
16604 City Public Service
17718 Southwestern Public Service Co.
9096 Lafayette Utilities System
14063 Oklahoma Gas & Electric Co.
1015 City of Austin Electric Utility Dpt
16604 City Public Service
17718 Southwestern Public Service Co.
16604 City Public Service
11269 Lower Colorado River Authority
8901 Houston Lighting & Power Company
16604 City Public Service
16463 Ruston Utilities System
1015 City of Austin Electric Utility Dpt
17718 Southwestern Public Service Co.
11269 Lower Colorado River Authority
plant
Jack Watson
Holly Street
T H Wharton
Irvington
Jack Watson
Nichols Station
Wilkes
Plant X
Hutchinson
Cunningham
Stryker Creek
V H Braunig
Maddox
Ray Olinger
Mountain Creek
V H Braunig
Nichols Station
Doc Bonin
Horseshoe Lake
Decker Creek
V H Braunig
Jones Station
O W Sommers
Sim Gideon
Greens Bayou
O W Sommers
Ruston
Holly Street
Jones Station
T C Ferguson
Average heat input
state (MMBtu/hr)
MS
TX
TX
AZ
MS
TX
TX
TX
KS
NM
TX
TX
NM
TX
TX
TX
TX
LA
OK
TX
TX
TX
TX
TX
TX
TX
LA
TX
TX
TX
307
512
954
443
490
652
841
859
579
1200
3615
1216
883
332
3481
1069
945
427
590
1574
1945
1422
2138
1603
2094
2525
17
1151
1368
2120
nox rate-3Q
(Ib/MMBtu)
0.149
0.102
0.157
0.202
0.194
0.15
0.151
0.181
0.272
0.208
0.16
0.162
0.154
0.12
0.162
0.179
0.218
0.141
0.081
0.155
0.218
0.249
0.205
0.174
0.113
0.153
0.14
0.178
0.245
0.175
nox rate-3Q
(Ib/MMscf)
152
104
160
206
198
153
154
185
277
212
163
165
157
122
165
183
222
144
83
158
222
254
209
177
115
156
143
182
250
179
A-8
-------
UNCONTROLLED NOx EMISSION DATA FOR TANGENTIAL FIRED BOILERS (CONTINUED)
Average heat input nox rate-3Q nox rate-3Q
utilcode utility plant state (MMBtu/hr) (Ib/MMBtu) (Ib/MMscf)
1015 City of Austin Electric Utility Dpt Decker Creek TX 2504 0.113 115
16687 Savannah Electric and Power Co. Riverside GA 175 0.114 116
17718 Southwestern Public Service Co. Moore County Station TX 345 0.138 141
44372 TU Electric Trinidad TX 1550 0.204 208
Average (Ib/MMscf) =167
(Ib/MMBtu) = 0.16
A-9
-------
APPENDIX B
Reviewer Comments and EPA Responses
-------
List of Addresseses for Draft Section 1.4
Mr. Lawrence C. Bradbury, P.E., J.D.
(provided comments)
Director, Environment & Safety
Atlanta Gas Light Company
P.O. Box 4569
Atlanta, GA 30302-4569
Mr. Ray A. Bradford (provided comments)
Manager Safety
Environmental & Regulatory Compliance
Phillips Petroleum Company
P.O. Box 1967
Houston, TX 77251-1967
Mr. Nicholas J. Bush
Natural Gas Supply Association
1129 20th Street N.W.
Suite 300
Washington, D.C. 20036
Mr. R.E. Cannon
Gas Processors Association
6526 E. 60th Street
Tulsa, OK 74145
Mr. Paul Chu (provided comments)
Electric Power Research Institute
3412 Hillview Avenue
Palo Alto, CA 94303
Dr. A. Kent Evans
Sr. Environmental Planner
Consumers Energy
1945 West Parnall Road
Jackson, WI 49201-8642
Mr. Jeff Glenn
Texas Natural Resource Conservation
Commission
P.O. Box 13087
MC 164
Austin, TX 78711-3087
Mr. Robert Hall
Air Pollution Prevention and Control Division
(MD-65)
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
Mr. Craig S. Harrison
Hunton & Williams
2000 Pennsylvania Avenue, N.W.
Washington, D.C. 20036
Mr. Roy Huntley (provided comments)
U.S. Environmental Protection Agency
Emission Factor and Inventory Group (MD-14)
Research Triangle Park, N.C. 27711
Mr. David G. Lachapelle (provided comments)
U.S. Environmental Protection Agency
Air Pollution Prevention and Control Division
(MD-04)
Research Triangle Park, N.C. 27711
Mr. Bill Maxwell (provided comments)
Office of Air Quality Planning and Standards
(MD-13)
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
Mr. Jim McCarthy
Gas Research Institute
8600 W. Bryn Mawr Avenue
Chicago, IL 60631
Mr. Russ Mosher (provided comments)
American Boiler Manufacturers Association
950 N. Glebe Road
Suite 160
Arlington, VA 22203
Mr. Peter Mussio
Supervisor, Environmental Engineering
Union Gas Limited/Centra Gas Ontario, Inc.
50 Keil Drive North
Chatham, Ontario N7M 5M1
Natural Resources Defense Council
40 West 20th Street
New York, NY 10011
B-l
-------
Mr. Ted M. Polychronis
Senior Air Quality Engineer
South Coast
Air Quality Management District
Planning & Technology Advancement
21865 Copley Drive
Diamond Bar, CA 91765
Mr. John Pratapas
Gas Research Institute
8600 W. Bryn Mawr Avenue
Chicago, IL 60631
Mr. Ralph Roberson
5400GlenwoodAve.
Suite G-11
Raleigh, NC 27612
Ms. Marise Lada Textor
Unit Manager, Water & Ecology
Chevron Research & Technology Company
P.O. Box 1627
Richmond, CA 94802-1627
Ms. Glenda Smith
American Petroleum Institute
1220 L Street, N.W.
Washington, DC 20005
R.E. Sommerlad (provided comments)
Gas Research Institute (GRI)
8600 West Bryn Mawr Avenue
Chicago, IL 60631-3 5 62
Mr. John Stower (provided comments)
Staff Environmental Analysis
Burns and McDonald Engineering
9400 Ward Parkway
Kansas City, MO 64114
Ms. Lori Traweek
American Gas Association
1515 Wilson Blvd.
Arlington, VA 22209
Summary of Comments
Section 1.4 - Natural Gas Combustion
B-2
-------
Emission Factors
GRI: In Table 1.4-1, small wall-fired and residential furnaces (<100 MMBtu/hr) are grouped
in one category. Previous versions had size ranges at <0.3, 0.3 to <10, 10 to <100, and >100
MMBtu/hr. With the present single grouping of <100 MMBtu/hr, the implication is that NOX,
CO, and N2O emissions are independent of size. Is there data to support this grouping under one
size range?
Response: Based on the available data, EPA determined that boiler size had no clear effect
on NOX and CO emissions for boilers less than 100 MMBtu/hr of heat input. The majority of
boilers that are smaller than 100 MMBtu/hr are package units and emissions appear to be more
dependent on individual boiler operation than boiler size.
Atlanta Gas: The EPA should consider adding a third category to Table 1.4-1 to address
either "other" boilers by heat input or address the "ring retention" type boilers. Atlanta Gas has
only ring retention type, fire tube, water/glycol boilers. In order for Atlanta Gas to use emission
factors versus stack testing on boilers, it would need the previously published emission factors
that used heat input or a new category for ring retention.
Response: The EPA changed the small boiler category to include "other" boiler types. In
addition, a footnote to Table 1.4-1 provides a conversion factor for heat input: to convert from
lb/106 scf to Ib/MMBtu, divide by 1,020.
GRI: In Table 1.4-1, the value of 84 lb/106 scf for CO converts to about 115 ppm, which
seems high. In addition, this value implies there is no variation as a function of size. Previous
versions had additional size categories. Is there a reason for the change?
Response: The data supports a CO emission factor for wall-fired boilers that is not
dependent on size. There were 49 tests conducted on 23 boilers, with an average emission factor
of 84.15 and a relative standard deviation of 124 percent. The EPA analyzed CO emissions
versus boiler size and determined that there is no clear relation between size and CO emissions.
It is true that if CO emissions were averaged across the previous size ranges, the various boiler
size categories would have slightly different CO emission factors, but the overall data set showed
no clear relation to size. Therefore, CO emission factors were not categorized by size for the
wall-fired and small boiler categories. During the next revision of this section, if additional CO
B-3
-------
emission data indicates a stronger correlation between size and CO emission levels, then CO
emission factors would be distinguished by size.
Burns & McDonnell: Footnote "d" is not properly referenced in Table 1.4-2. It should
appear with SO2 in the pollutant column. Also, EPA should stress that since the emission factors
are based on a natural gas heat content of 1,020 Btu/scf, users may need to adjust the emission
factors. If the heat content of their natural gas differs from the 1,020 Btu value, users should
adjust the emission factor by a ratio of the heat rates (actual Btu heat content 71,020 Btu value).
In addition, this same approach applies to the assumed 2,000 grains of sulfur/MMscf for the SO2
emission rates within the Table 1.4-2.
Response: Footnote "d" has been corrected to properly reference the SO2 emission factor.
Footnote "d" of Table 1.4-2 was also amended to provide guidance on adjusting emission factors
for sources firing natural gas with Btu ratings different from 1,020 Btu/scf. A similar approach
was taken with the SO2 emission factor; in this case, EPA provided guidance to adjust the SO2
emission factor at sources where the sulfur content of the natural gas was different from
2,000 grains/MMscf.
U.S. EPA, EFIG: The CO2 emission factor in Table 1.4-2 should be 120,000, not 12,000.
Also, correct footnote "b" calculation.
Response: The emission factor has been corrected to 120,000 Ib/MMscf. The EPA also
corrected an error in footnote "b" regarding the calculation of the CO2 emission factor.
GRI: Table 1.4-2 indicates a conversion of fuel carbon to CO2 of 99.5%. This converts to
about 5,000 ppm of CO and other hydrocarbons. This seems high for commercial boilers.
Typical values of CO are less than 50 ppm and other hydrocarbons are typically below 100 ppm.
These would result in a conversion efficiency of 99.995% rounded down to 99.9%. Is this value
correct?
Response: The assumed fuel carbon conversion as been changed to 99.9%. This adjustment
will not change the CO2 emission factor since it was rounded to two significant figures.
B-4
-------
Unidentified commenter via U.S. EPA, EFIG: Put Chemical Abstract Services (CAS)
numbers with the Hazardous Air Pollutant (HAP) in the tables. The HAP list in the section is
confusing because many of those compounds listed are not listed in section 112(b) of the
1990 Amendments to the Clean Air Act, and they only qualify as HAPs because they are
Polycylic Organic Matter (POMs). It would be more clear to label which compounds are HAPs
and which are HAPs because they are POMs.
Response: CAS numbers were assigned to all pollutants for easy identification. The EPA
also distinguished between HAPs and compounds that are classified as HAPs because they are
POMs.
GRI: GRI suggests adding a footnote to the tables to explain to the casual reader the meaning
of "emission factor rating."
Response: Rather than footnote each table with an explanation of emission factor ratings, the
ratings are discussed at the end of the section. In addition, the EPA fully discusses emission
factor ratings in the introduction to AP-42 and in the Emission Factor Documentation for
Section 1.4 (background report).
Phillips: The emissions data suggest that grouping the toxics data into specific categories of
heaters/boilers could provide more accurate emission factors for air toxics. In the database
enclosed with the report, arithmetic averages are used to calculate the criteria pollutant and toxics
emission factors. By using arithmetic averages, the assumption is made that the distribution is
normal. However, Phillips' review of the normality and probability of the toxics data shows non-
random behavior (non-normal distribution). The commenter suggests that, if it has not already
been considered, the toxics data may be grouped by heater/boiler heat input to increase the
accuracy of the resulting emission factors. (The commenter recognizes that small sample sizes
reduce the effectiveness of normality tests.) If this suggested grouping has already been
considered and would not work, EPA should discuss this in the background report. Otherwise,
EPA should consider a new grouping.
Response: The toxic data were analyzed for these source categories to determine if grouping
these data by source type would provide more accurate emission factors. Based on this analysis
and given the limited data available, no clear relation is apparent between these source categories
B-5
-------
and toxic emission levels. Therefore, EPA maintains that the current grouping is the most
appropriate. The background report provides a discussion of this decision.
Phillips: The emission factors for natural gas-fired heaters should be delayed pending the
results of the GRI/API/Radian study of engine emissions. A program for characterizing and
quantifying emissions from reciprocating engines used in oil and gas production is underway and
the data will be available in October 1997. The program will also investigate the emissions from
a 62.5 MMBtu/hr boiler and a heater treater which is representative of small heaters used in the
oil and gas production industry. The list of analytes chosen for this effort includes those reported
in Tables 1.4-2 and 1.4-4. The resulting data meet EPA's criteria for an emission factor rating of
"A." The value of the data justifies a short delay in publishing the revised emission factors.
Response: The EPA is aware of the data that will be available in the GRI/API/Radian study
but that the final report will not be ready for distribution until early 1998. Given the time frame
of the publication of this report, EPA does not want to delay the revision of Section 1.4 of AP-42
to include this data. The EPA understands that the report has data from 1 boiler and that the
inclusion of 1 extra boiler in the database should have little effect on the emission factors in this
revision. However, the emission test data from the boilers tested in the GRI study will be
incorporated in the next revision to Section 1.4.
Phillips: Only emission factors with an emission factor rating of A, B, or C should be
published in the public domain. The use of emission factors based on poor quality data may have
far-reaching, undesirable consequences.
Response: The primary purpose of AP-42 is to provide emission factors for emission
inventories. The EPA provides emission factors for as many sources and as many pollutants as
available resources allow. The factors are rated "A" through "E" to provide the user with an
indication of how good an emission factor is, with an "A" being excellent and "E" being poor.
The criteria that are used to determine a rating for a factor can be found in the document entitled
"Procedures for Preparing Emission Factor Documents, EPA-454/R-95-0150." While the EPA
shares your concern about poor quality emission factors for various reasons, the factor rating is
used to judge whether the factor is appropriate.
B-6
-------
Controls
GRI: Section 1.4 states that low-NOx burners and flue gas recirculation (FGR) are the most
prevalent combustion NOX control techniques being applied to natural gas-fired boilers. GRI
agrees that low-NOx burners are prevalent in all classes and size ranges of boilers. However,
GRI does not agree that FGR is prevalent for boilers with capacities less than 100 MMBtu/hr.
Also, one NOX control system not mentioned is gas reburning. Gas reburning is an attractive
technical and economic alternative to SNCR or SCR. The commenter cited demonstrations of
gas reburn on a tangentially-fired utility boiler, a front-wall boiler, and an opposed-wall boiler.
The tangentially-fired boiler achieved a 55-65% reduction of NOX and the opposed-wall and
front-wall burners achieved 80% and 73% reductions, respectively.
Response: The EPA is aware that FGR technology is most prevalent in boilers with heat
inputs greater than 100 MMBtu/hr, however, EPA has data from boilers with heat inputs less
than 100 MMBtu/hr that employ FGR and low-NOx burners for NOX control. The EPA has
received tests from several boilers with heat inputs less than 100 MMBtu/hr that have FGR and
low-NOx burners. Furthermore, there is a separate category for the boilers for NOX emission
factors. With respect to gas reburning, EPA does not have any data to evaluate the performance
of gas reburning. In the final version of this revision to Section 1.4, gas reburning will be
mentioned as a NOX control technology. However, NOX reduction efficiencies will not be
presented in this revision to Section 1.4 due to the lack of supporting data.
GRI: Section 1.4 states that the addition of low-NOx burners and FGR may reduce
combustion efficiency. This implies that low-NOx burners and FGR are the direct cause of
reduced combustion efficiency. This is not necessarily correct. Incomplete combustion can be
unburned fuel, unburned carbon, and newly formed solid, liquid, or gaseous hydrocarbons. One
of the later species could be CO. GRI suggests the following revision to the paragraph:
"Improperly tuned boilers and boilers operating at off-design levels can result in increased
partially oxidized combustibles (e.g., CO) and thus, decreased combustion efficiency. The
addition of NOX control systems such as low-NOx burners and FGR may also result in increased
CO or other partially oxidized combustibles, and likewise, decreased combustion efficiency."
B-7
-------
Response: The effects of improperly tuned boilers on CO and hydrocarbon emissions
were addressed in the discussions on CO and hydrocarbon emissions. Therefore, OKI's
suggestion will not be added.
GRI: It is also worth mentioning that current NOX control systems can lower NOX
emissions without increases in other emissions such as CO, VOCs, and PM. An example of this
is shown in Section 2, Reference 6.
Response: The EPA will add this to its discussion of NOX control technologies.
GRI: There has been significant testing of minor products of combustion and also
significant development of low-NOx burners and NOX control technology in recent years. If it is
not within EPA's current resources to obtain later information, it would be well to indicate the
data in the tables are from sources with publication dates ranging from 1990-1996.
Response: The background report provides a list of all the references used in this
revision including testing dates. If users wish to evaluate the age of the data, they can download
this document from the TTN.
Emission Data
ABMA and EPRI both provided emission data for natural gas fired boilers. The data
provided by ABMA was used for comparative purposes and was not included for emission factor
development because it did not contain complete testing information. The data provided by
EPRI did contain complete testing information and was used in the development of emission
factors.
B-8
------- |