EMISSION FACTOR
DOCUMENTATION FOR
AP-42 SECTION 1.5
LIQUEFIED PETROLEUM GAS COMBUSTION
By:
Acurex Environmental Corporation
Research Triangle Park, North Carolina 27709
Edward Aul & Associates, Inc.
Chapel Hill, North Carolina 27514
E. H. Pechan and Associates, Inc.
Rancho Cordova, California 95742
Contract No. 68-DO-00120
Work Assignment No. II-68
EPA Project Officer: Alice C. Gagnon
Office of Air Quality Planning and Standards
Office Of Air And Radiation
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
April 1993
-------
DISCLAIMER
This report has been reviewed by the Office of Air Quality Planning and Standards,
U. S. Environmental Protection Agency, and approved for publication. Mention of trade
names or commercial products does not constitute endorsement or recommendation for
use.
-------
TABLE OF CONTENTS
LIST OF TABLES v
CHAPTER! INTRODUCTION 1-1
CHAPTER2. INDUSTRY DESCRIPTION 2-1
2.1 CHARACTERIZATION OF THE INDUSTRY 2-1
2.2 PROCESS DESCRIPTION 2-2
2.3 EMISSIONS 2-2
2.3.1 Nitrogen Oxides Emissions 2-2
2.3.2 Carbon Monoxide Emissions 2-3
2.3.3 Sulfur Oxides Emissions 2-4
2.3.4 Particulate Matter Emissions 2-4
2.3.5 Organic Compound Emissions 2-5
2.3.6 Trace Element Emissions 2-6
2.3.7 Carbon Dioxide Emissions 2-6
2.3.8 Nitrous Oxide Emissions 2-6
2.3.9 Fugitive Emissions 2-7
2.4 CONTROL TECHNOLOGIES 2-7
REFERENCES 2-10
CHAPTER 3. GENERAL DATA REVIEW AND ANALYSIS PROCEDURES 3-1
3.1 CRITERIA POLLUTANTS 3-1
3.1.1 Literature Search 3-1
3.1.2 Literature Evaluation 3-3
3.2 NON-CRITERIA POLLUTANTS 3-4
-------
3.3 FUGITIVE EMISSIONS 3-4
REFERENCES 3-6
CHAPTER 4. EMISSION FACTOR DEVELOPMENT 4-1
4.1 CRITERIA POLLUTANTS 4-2
4.1.1 Review of Previous Data 4-2
4.1.2 Review of New Data 4-2
4.1.3 Compilation of Emission Factors 4-3
4.2 NON-CRITERIA POLLUTANTS 4-4
4.3 FUGITIVE EMISSIONS 4-4
REFERENCES 4-8
CHAPTERS. AP-42 SECTION 1.5: LIQUIFIED PETROLEUM
GAS COMBUSTION 5-1
TABLE OF CONTENTS (continued)
APPENDIXA. SAMPLE CALCULATIONS A-1
APPENDIX B. MARKED-UP 1982 SECTION 1.5 B-1
IV
-------
LIST OF TABLES
Table
2-1 Comparison of LPG and Natural Gas Sales For 1989 2-9
4-1 Natural Gas Combustion Emission Factors 4-6
4-2 LPG Combustion Emission Factors 4-7
4-3 Fugitive Emission Factors For Various Equipment Types 4-7
-------
1. INTRODUCTION
The document, "Compilation of Air Pollutant Emission Factors" (AP-42), has
been published by the U.S. Environmental Protection Agency (EPA) since 1972.
Supplements to AP-42 have been routinely published to add new emissions source
categories and to update existing emission factors. An emission factor is an average
value which relates the quantity (weight) of a pollutant emitted to a unit of activity of the
source. The uses for the emission factors reported in AP-42 include:
! Estimates of area-wide emissions;
! Emission estimates for a specific facility; and
! Evaluation of emissions relative to ambient air quality.
The EPA routinely updates AP-42 in order to respond to new emission needs of
State and local air pollution control programs, industry, as well as the Agency itself.
Section 1.5 in AP-42, the subject of this Emission Factor Documentation (EFD) report,
pertains to liquefied petroleum gas (LPG) combustion.
Section 1.5, Liquefied Petroleum Gas Combustion, was previously updated in
1982. The purpose of this current revision is to review and update, if possible, the prior
emission factor development for criteria pollutants, and to add non-criteria emission
species if any data exist. This update is part of a larger effort in which all sections of
AP-42 are being revised. This emission factor document is part of the update effort. It
provides background on the decision-making process for emission factor development,
and documents the approach and results of the test data gathering effort.
The present update of AP-42 began with a review of the existing version of
Section 1.5. The previous emission factors developed in 1982 were not based on LPG
test data but were assumed to be the same as the natural gas emission factors on a
1-vi
-------
Btu thermal heat input basis (except for sulfur oxides which were based on fuel sulfur
content). This default approach was adopted because of the lack of LPG data.
An extensive literature review was undertaken to identify and collect emissions
data for LPG combustion for criteria and non-criteria pollutant emissions. The sparse
amount of emissions data obtained were reviewed for quality as outlined in the draft
EPA document, "Technical Procedures For Developing AP-42 Emission Factors And
Preparing AP-42 Sections," (March 6, 1992).
Where data were located, the data reduction used the F-factor method specified
in Reference Method 19 as contained in Appendix A of the Code of Federal Regulation
Title 40 Part 60 (40 CFR). The Ib/million Btu (Ib/MMBtu) emissions rate calculated was
then multiplied by the heating value for propane or butane (in MMBtu/thousand gallons)
to produce an emission factor in terms of Ib pollutant/thousand gallons. Analogous
calculations were performed for metric unit emission factors.
Several new emission factors covering non-criteria pollutants have been added.
These new emission factors pertain to total organic compounds (TOC), air toxics, and
carbon dioxide (C02). Information on control technologies for nitrogen oxides (NOX)
emissions has been revised and updated.
Including the Introduction (Chapter 1), this EFD contains five chapters. Chapter
2 provides an overall characterization of LPG usage. The generic types of emissions
are discussed together with general concepts for controlling NOX emissions generated
from LPG combustion. Chapter 3 is a review of the emissions data collection and
review procedures. The sources examined during the literature search are discussed
and the emissions data rating procedures are defined. Chapter 4 details the
development approach for new emission factors. It includes a review of specific data
and the assumptions used to arrive at the final emission factors. Chapter 5 presents
the revised AP-42 Section 1.5. Appendix A provides sample calculations for emission
factor development.
1-vii
-------
2. INDUSTRY DESCRIPTION
This chapter contains a brief characterization of the LPG industry, and discusses
LPG combustion processes, emissions and control options.
2.1 CHARACTERIZATION OF THE INDUSTRY
The term "LPG" in its broadest context refers to a group of refinery byproduct
gases that may include the following compounds alone or in mixture: propane,
propylene, butane, butylene, and isobutane1. Liquefied petroleum gas can be used for
the same domestic, commercial and industrial applications as natural gas. One of the
main LPG markets is in rural areas for domestic cooking and heating. The main
advantage of LPG relative to natural gas is that, under pressure, it is a liquid which
reduces transportation costs and makes long term storage practical. The main
disadvantages are higher fuel costs and potential safety hazards in the event of leaks
arising from the fact that LPG is heavier than air and may settle in explosive pockets in
the absence of local air movement. Liquefied petroleum gas is also used in commercial
and industrial applications as a standby fuel to replace natural gas during emergencies,
or curtailments of baseline fuels. Recently, interest in LPG as a standby fuel has
increased greatly in the South Coast Air Quality Management District (SCAQMD) in
Southern California. The SCAQMD has mandated a phase out of oil as a stationary
source fuel as part of ozone attainment plans. Natural gas is the primary alternative,
but a noninterruptable supply cannot be assured, particularly during winter months.
Accordingly, many boiler operators are installing capability for firing propane, butane or
methanol.
Table 2-1 summarizes LPG sales to the residential, commercial and industrial
sectors for 1989. Sales to the industrial sector are predominantly as chemical
feedstocks.
2-viii
-------
2.2 PROCESS DESCRIPTION
The combustion processes that use LPG as a fuel are very similar to those that
use natural gas. The use of LPG in commercial and industrial applications may require
a vaporizer to provide the burner with the proper mix of air and fuel. An auxiliary
vaporizer is used when the natural vaporization capacity of the LPG storage tank is not
sufficient to meet the demand load of the burner(s) being served. The burner itself will
usually have different fuel injector tips as well as different fuel-air ratio controller
settings than natural gas since the LPG stoichiometric requirements are different than
natural gas requirements. Liquefied petroleum gas is fired as a primary and backup
fuel in small commercial and industrial boilers and space heating equipment. Liquefied
petroleum gas can be used to generate heat and process steam for industrial facilities.
Finally, LPG can be used in most domestic appliances that typically use natural gas.
2.3 EMISSIONS
The pollutants of primary concern from LPG combustion are the criteria
pollutants: NOX, carbon monoxide (CO), particulate matter (PM), sulfur oxides (SOX),
and TOC. As a gaseous fuel, LPG does not produce a large amount of particulate
emissions. The more significant emissions from LPG are the gaseous emissions.
2.3.1 Nitrogen Oxides Emissions
Nitrogen oxides are formed during combustion processes either as a
result of thermal fixation of atmospheric nitrogen (N2) in the combustion air ("thermal
NOX"), or the conversion of chemically-bound nitrogen in the fuel ("fuel NOX"). The term
NOX customarily refers to the composite of nitric oxide (NO) and nitrogen dioxide (N02).
Nitrous oxide (N20) is excluded from this definition of NOX, but is an oxide of increasing
interest as a greenhouse gas. Test data have shown that for most stationary
combustion systems, more than 95 percent of the emitted NOX is in the form of NO.
The qualitative global kinetics of thermal NOX formation show that NOX formation
rates are exponentially dependent on temperature, and proportional to N2 concentration
in the flame, the square root of oxygen (02) concentration in the flame, and the
residence time. Thus, the formation of thermal NOX is affected by four factors: (1) peak
temperature, (2) N2 concentration, (3) 02 concentration (or flame stoichiometry), and (4)
time of exposure at peak temperature. The emission trends due to changes in these
2-ix
-------
factors are fairly consistent for all types of boilers - an increase in flame temperature,
N2 availability, 02 availability, and/or residence time at high temperatures leads to an
increase in thermal NOX production regardless of the boiler type.
Fuel nitrogen conversion is only an important N0x-forming mechanism in oil and
coal-fired combustion systems because of the high nitrogen content of these fuels.
Nearly all NOX formed from LPG combustion is thermal NOX. A number of variables
influence how much thermal NOX is formed with LPG combustion. The combustion
chamber design (particularly the orientation of the heat transfer surfaces) and the
burner aerodynamics affect the peak temperature achieved in the combustion zone and
the mixing rate of oxygen with the LPG. Modifications to the burner hardware or to the
operating procedures can dramatically reduce NOX via changes to peak temperature or
mixing. Low excess air (LEA) firing, flue gas recirculation (FGR), staged combustion
(SC), or some combination thereof may result in NOX reductions of 15 to 75 percent.
Also, load reduction usually decreases NOX production by decreasing the volumetric
heat release in the combustion chamber, thus reducing peak temperatures.
2.3.2 Carbon Monoxide Emissions
The rate of CO emissions from gas-fired combustion equipment depends on the
efficiency of the final burnout of this last remaining product of incomplete combustion.
Carbon monoxide burnout is strongly kinetically controlled. If final burnout is not
achieved quickly at high temperatures, a long residence time in the post-furnace zone
is required to reduce CO to low levels. Carbon monoxide emissions are thus very
sensitive to combustion chamber and burner design and operation. If a unit is operated
improperly or not maintained, the resulting emissions of CO (as well as organic
compounds) may increase by an order of magnitude. Small boilers, heaters, and
furnaces tend to emit more of these pollutants than larger combustors. This is because
small units usually have a higher ratio of heat transfer surface area to flame volume
leading to reduced flame temperature and combustion intensity and, therefore, lower
combustion efficiency than large combustors. Also, these smaller units are usually not
well maintained, resulting in operation at either too lean or too rich fuel/air ratios, thus
increasing CO.
2-x
-------
The presence of CO in the exhaust gases of combustion systems results
principally from incomplete fuel combustion. Several conditions can lead to incomplete
combustion. These include:
Insufficient oxygen availability;
Extremely high levels of excess air leading to quenching (more
common with industrial boilers);
! Poor fuel/air mixing;
! Cold wall flame quenching;
! Reduced combustion temperature;
! Decreased combustion gas residence time; and
! Load reduction (i.e., reduced combustion intensity).
Since various combustion modifications for NOX reduction can produce one or more of
the above conditions, the possibility of increased CO emissions is a concern for
environmental, energy efficiency, and operational reasons.
2.3.3 Sulfur Oxide Emissions
Sulfur oxides result from oxidation of residual sulfur compounds in the LPG
either to sulfur dioxide (S02) or sulfur trioxide (S03). With gaseous fuels, essentially all
sulfur is oxidized and emitted with the flue gas. Generally, over 90 percent of SOX is in
the form of S02. The sulfur content of LPG is very low but will vary with supplier
depending on processing prior to distribution.
2.3.4 Particulate Matter Emissions
Particulate matter emissions from LPG combustion are very low and result from
soot, aerosols formed by condensible emitted species, or boiler scale dislodged during
combustion. Combustible emissions are controllable, in part, by modifications to the
combustion conditions. Measurements of condensible emissions are influenced by
sampling conditions which affect whether a species is quantified as vapor or particulate.
2.3.5 Organic Compound Emissions
2-xi
-------
Total organic compounds include volatile organic compounds (VOCs) which
remain in a gaseous state in ambient air, semi-volatile organic compounds and
condensible organic compounds. According to the Federal Register definition (57 FR
3945), VOC has been defined as any organic compound excluding carbon monoxide,
carbon dioxide, carbonic acid, metallic carbides or carbonates, and ammonium
carbonate which participates in atmospheric photochemical reactions. The following
additional compounds have been deemed to be of "negligible photochemical reactivity"
and also are exempt from the definition of VOC: methane, ethane, methyl chloroform,
methylene chloride, and most chlorinated-fluorinated compounds (commonly referred to
as CFCs). Although these compounds are considered "exempt" from most ozone
control programs due to their low photochemical reactivity rates, they are of concern
when developing complete emission inventories which are necessary for the design of
effective ozone control strategies. The term TOC will be considered to include all
organic compounds, i.e., VOCs plus the "exempt" compounds including methane and
ethane, toxic compounds, aldehydes, perchloroethylene, semi-volatiles, and
condensibles (as measured by EPA Reference Methods).
Emissions of VOCs are primarily characterized by the criteria pollutant class of
unburned vapor phase hydrocarbons. Unburned hydrocarbon emissions can include
essentially all vapor phase organic compounds emitted from a combustion source.
These are primarily emissions of aliphatic, oxygenated, and low molecular weight
aromatic compounds which exist in the vapor phase at flue gas temperatures. These
emissions include all alkanes, alkenes, aldehydes, carboxylic acids, and substituted
benzenes (e.g., benzene, toluene, xylene, ethyl benzene, etc.).
The remaining organic emissions are composed largely of compounds emitted
from combustion sources in a condensed phase. These compounds can almost
exclusively be classed into a group known as polycyclic organic matter (POM), and a
subset of compounds called polynuclear aromatic hydrocarbons (PNA or PAH). There
are also the PAH-nitrogen analogs. Information available in the literature on POM
compounds generally pertains to these PAH groups. Polycyclic organic matter
emissions are generally less prevalent from LPG and other gaseous fuel combustion
than volatile organic emissions because of the fuel structure.
2-xii
-------
Formaldehyde is formed and emitted during combustion of hydrocarbon-based
fuels. Formaldehyde is present in the vapor phase of the flue gas. Since formaldehyde
is subject to oxidation and decomposition at the high temperatures encountered during
combustion, large units with efficient combustion (resulting from closely regulated air-
fuel ratios, uniformly high combustion chamber temperatures, and relatively long
retention times) should have lower formaldehyde emission rates than do small, less
efficient combustion units.
2.3.6 Trace Element Emissions
Trace elements are an important class of air toxics compounds. Their
concentration in the flue gas emissions is, however, dominated by the concentration of
contaminants in the fuel unless another source such as process material contains trace
elements. There is a very low concentration of trace elements in LPG due to its origin
(i.e., natural gas wells) or to processing within a refinery. Accordingly, trace element
emissions for LPG are very low compared to oil or coal combustion.
2.3.7 Carbon Dioxide Emissions
Carbon dioxide is the final product of all hydrocarbon combustion. With LPG
combustion, nearly all carbon in the fuel is emitted as C02. Minor amounts, typically
0.01 percent, are emitted as CO, giving a carbon conversion efficiency to the stack of
99.99 percent. Therefore, emission factors for C02 can be better approximated using a
carbon mass balance than by measurement.
2.3.8 Nitrous Oxide Emissions
Nitrous oxide is relatively harmless to human health but is included among
greenhouse gases. Numerous measurements made prior to 1988 suggested that N20
emissions from gas-fired combustion equipment may be comparable to NOX emissions.
In 1988, the earlier sampling protocols proved to be faulty resulting in N20 generation in
the post-test sample processing. Recent N20 emissions data indicate that direct N20
emissions from conventional gas-fired combustion units are an order of magnitude or
more below the measurements made prior to 1988. However, the N20 formation and
reaction mechanisms are still not well understood or well characterized. Emissions can
vary widely from unit to unit, or even for different operating conditions at the same unit.
Additional sampling and research is needed to fully characterize N20 emissions and to
2-xiii
-------
understand the N20 formation mechanism. It has been shown in some cases that N20
increases with decreasing combustion temperature, so some lower-temperature LPG-
fired domestic heating equipment may be suspect.
2.3.9 Fugitive Emissions
Fugitive emissions are unducted pollutants escaping an industrial process via
leakage, materials handling, inadequate operational control, transfer, storage or
distribution. With adequately maintained LPG equipment, fugitive emissions are
primarily confined to tank loading transfer operations, and these emissions are
controllable. With inadequately maintained equipment, leaks in the distribution system
can occur at valves and flanges.
2.4 CONTROL TECHNOLOGIES
The pollutant specie of most concern for stationary LPG combustion is NOX.
Nitrogen oxides are the only pollutants for which controls have been developed for LPG
firing in internal combustion sources. Volatile organic compound controls on standby
generator engines fueled with LPG are discussed in AP-42 Section 3.4.
The NOX control techniques for LPG have generally been adapted from the
experience with low NOX natural gas firing. However, LPG presents two challenges in
translating natural gas technology, however. First, the N0x-forming potential with LPG
is higher, so that baseline emissions are typically considerably higher for LPG than for
natural gas. This is due primarily to the higher heating values and, hence, higher
combustion intensity of LPG relative to natural gas. As a result, higher percent
reductions may be needed with LPG to comply with an emission standard. In this
respect, the N0x-forming characteristics of LPG are more similar to light oil. Second,
LPG burners are more susceptible to sooting and smoking behavior when modified to
off-optimum operation. Thus, attempts to control LPG NOX by combustion modification
may encounter operational constraints well before the full potential for NOX reduction is
reached.
Until recently, there has been little regulatory incentive to develop or implement
low NOX LPG combustion systems. Accordingly, there has been little developmental
activity reported. Flue gas recirculation, a traditional technique with natural gas, has
been attempted on a small commercial boiler fired with LPG2. NOX emissions from
2-xiv
-------
propane combustion were reduced by approximately 50 percent by recirculating 16
percent of the flue gas2. For a pilot scale unit, NOX emissions from butane combustion
were reduced by approximately 79 percent by recirculating approximately 30 percent of
the flue gas3.
Recent SCAQMD regulations have prompted development of advanced low NOX
techniques for firetube and watertube boilers. The SCAQMD Rule 1146 for existing
boilers and stringent new source review limits for new boilers have created the need for
controlled NOX emissions of 30 to 40 ppm. Numerous boiler operators are planning to
use propane or butane to replace distillate oil as an adjunct or backup to natural gas
pursuant to the fuel oil phaseout. Accordingly, several boiler or burner vendors have
started to develop low NOX burners for LPG to comply with Rule 1146. One
manufacturer of low NOX burners reported a 65 percent reduction of NOX emissions
using a low NOX burner and water injection on a propane-fired boiler located in the
SCAQMD4. The boiler was firing propane as a backup fuel to natural gas. There are
several propane-fired installations in the SCAQMD required to meet a NOX limit of 40
ppm at 3 percent 02. Water or steam injection is used as a trimming technique if
compliance cannot be reached by Iow-N0x burners (LNB) alone. The water is injected
into the primary flame zone usually through the oil guns. Most vendors are now
prepared to warrant 30 or 40 ppm NOX emissions (referenced to 3 percent oxygen) for
existing or new boilers.
2-xv
-------
TABLE 2-1. COMPARISON OF LPG AND NATURAL GAS SALES FOR 19891
Annual sales, billion KW-hr (trillion Btu)
Residential Commercial Industrial
LPG 150(505) 26(89) 480(1627)
Natural Gas 1310(4471) 643(2193) 885(3010)
2-xvi
-------
REFERENCES FOR SECTION 2
1. Gas Facts: 1990 Data, The American Gas Association, Arlington Va. 1991.
2. Nitrous Oxide Reduction with the Weishaupt Flue Gas Recirculation System,
Weishaupt Research and Development Institute, Jan. 1987.
3. Some Examples of Combustion Tests for Putting New Fuels to Practical Use,
Ikebe etal, IHI Engineering Review, Oct. 1976.
4. Phone communication memorandum dated May 5, 1992. Conversation between
B. Lusher of Acurex Environmental, Research Tirnagle Park, N.C., and G.
Constonaine of Hague International.
2-xvii
-------
3. GENERAL DATA REVIEW AND ANALYSIS PROCEDURE
This section summarizes the procedures and criteria used for the literature
search and data evaluation.
3.1 CRITERIA POLLUTANTS
3.1.1 Literature Search
An extensive literature search was conducted to identify sources of criteria and
non-criteria emissions data for LPG combustion. The following sources were searched
for emissions data:
Literature
! Existing AP-42 background files
! Files maintained by the EPA's Emission Standards Division and
Emission Factor and Methodologies Section
! National Technical Information Service (NTIS) documents
! EPA contractor files
! NOX, SOX, and PM symposia
Personal contacts
! Regulatory agencies, primarily the South Coast Air Quality
Management District
! Southern California Gas Company
! Low NOX burner (LNB) manufacturers and boiler manufacturers
! An LPG supplier who advises customers on conversion issues and
regulatory requirements
! An LPG systems contractor conducting propane boiler conversions
in Southern California
3-xviii
-------
The main conclusion drawn from the literature search was that little has been
published regarding emissions from LPG combustion. The SCAQMD is a potential
source of future emissions data for LPG since they are the repository of compliance
tests for numerous propane and butane conversions completed or planned in the Los
Angeles basin. These results were not available within the schedule constraints for this
update. Two source reports were requested from SCAQMD for compliance testing
conducted on two boilers believed to be burning LPG and propane. These requests
were added by SCAQMD to a waiting list. If data are received from SCAQMD or other
sources subsequent to the completion of this update, the data will be added to the
Background File for this section for consideration during the next update.
Two reports containing LPG emissions data were obtained from the Southern
California Gas Company. One report contains emissions data for propane and natural
gas in three different boilers equipped with flue gas recirculation, including one set of
baseline concentrations for natural gas and propane fired in the same boiler.1 The
second source is an article which presents emissions data for natural gas, propane and
butane firing in a pilot-scale boiler equipped with flue gas recirculation2. Emissions
rates are given as a function of flue gas recirculation rates which varied from 0 to 40
percent.
One LNB manufacturer provided emission measurements for a propane-fired
boiler3. The boiler was equipped with a LNB and water injection. At full load the NOX
concentration was 52 ppm; with water injection the NOX concentration decreased to
17.9 ppm. This reduction meets the SCAQMD Rule 1146 standard of 40 ppm N0xat 3
percent 02.
Another LNB manufacturer provided some emissions data for a propane-fired
industrial packaged watertube boiler. However, the process was not identified in the
information provided. Moreover, the NOX emissions were extremely high (279 ppm at 4
percent 02) when compared to the other data obtained. As a result, these data were
excluded from further consideration.
3.1.2 Literature Evaluation
3-xix
-------
The data obtained from the literature search were reviewed to determine their
suitability for use in emission factor calculations. Checklists were employed to
standardize and document this evaluation. The completed checklists were placed in
the background file for this update to AP-42. Data with the following characteristics
were excluded from further consideration:
1. Test series averages reported in units that cannot be converted to the
selected reporting units;
2. Test series representing incompatible test methods (i.e., comparison of
EPA Method 5 front-half with EPA Method 5 front- and back-half);
3. Test series of controlled emissions for which the control device is not
specified;
4. The series in which the boiler design and operating conditions are not
clearly identified and described; and
5. Test series in which it is not clear whether the emissions were baseline or
controlled.
Data sets that were not excluded were assigned a quality rating. The rating
system used was that specified in the draft EPA document, "Technical Procedures For
Developing AP-42 Emission Factors And Preparing AP-42 Sections" (March 6, 1992).
The data were rated as follows:
A: Multiple tests performed on the same source using sound methodology
and reported in enough detail for adequate validation. These test are not
necessarily EPA reference method tests, although such reference
methods are preferred and certainly to be used as a guide.
B: Tests that were performed by a generally sound methodology but lack
enough detail for adequate validation.
C: Tests that were based on an untested or new methodology or that lacked
a significant amount of background data.
D: Tests that were based on a generally unacceptable method but may
provide an order-of-magnitude value for the source.
The following criteria were used to evaluate source test reports for sound
methodology and adequate detail:
3-xx
-------
1. Source operation. The manner in which the source was operated is well
documented in the report. The source was operating within typical
parameters during the test.
2. Sampling procedures. The sampling procedures conformed to generally
acceptable methodology. If actual procedures deviated from accepted
methods, the deviations are well documented. When this occurred, an
evaluation was made of the extent such alternative procedures could
influence the test results.
3. Sampling and process data. Adequate sampling and process data are
documented in the report. Many variations can occur unnoticed and
without warning during testing. Such variations can induce wide
deviations in sampling results. If a large spread between test results
cannot be explained by information contained in the test report, the data
are suspect and were given a lower rating.
4. Analysis and calculations. The test reports contain original raw data
sheets. The nomenclature and equations used were compared to those
(if any) specified by EPA to establish equivalency. The depth of review of
the calculations was dictated by the reviewer's confidence in the ability
and conscientiousness of the tester, which in turn was based on factors
such as consistency of results and completeness of other areas of the test
report.
3.2 NONCRITERIA POLLUTANTS
This update was done in parallel with updates to other sections of AP-42
Chapter 1. As part of these other updates, extensive literature searches were made of
the likely sources of test data on non-criteria pollutant species, e.g. speciated VOC, air
toxics, and N20. Sources searched included EPA emissions data bases and research
project reports, symposia and journals, industry trade associations, and computerized
literature data bases. No noncriteria pollutant data directly related to LPG combustion
were identified during the search. Accordingly, no emission factor development was
possible for noncriteria species. In the absence of test data, order of magnitude
qualitative estimates of LPG emissions may be made using natural gas or distillate oil
test results for comparable equipment designs.
3.3 FUGITIVE EMISSIONS
Fugitive emissions have not historically been covered in Chapter 1 of AP-42.
Chapter 4 contains emission factors for evaporative losses from petroleum storage
facilities and will continue to be the source of such data.
3-xxi
-------
A literature search was conducted to quantify fugitive emissions from leaking
seals and fittings that would be present in the fuel feed system for LPG combustion
sources. These fugitive sources include valves, flanges, relief valves and open-end
lines. The literature evaluation verified the conclusions of previous attempts at
determining emission factors for VOCs in the Synthetic Organic Chemical
Manufacturing Industry (SOCMI). During the establishment of proposed standards for
fugitive VOC emissions from the SOCMI, EPA determined (1) that the only quality
emission factor data were generated during an EPA study of 13 petroleum refineries
and (2) that the highest quality data for leakages from each equipment type resulted
from an EPA study of leak frequency in the SOCMI. Data from these two EPA
references were subjectively rated B quality per the general guidelines previously
described. Because of the lack of sufficient data for calculating new emission factors
for fugitive emissions, the remainder of the references obtained were not used and,
therefore, not rated.
3-xxii
-------
REFERENCES FOR CHAPTER 3
1. Nitrous Oxide Reduction with the Weishaupt Flue Gas Recirculation System,
Weishaupt Research and Development Institute, Jan. 1987.
2. Some Examples of Combustion Tests for Putting New Fuels to Practical Use,
Ikebe etal, IHI Engineering Review, Oct. 1976.
3. Phone communication memorandum dated May 5, 1992. Conversation between
B. Lusher of Acurex Environmental and G. Constonaine of Hague International.
4. Phone communication memorandum dated May 5, 1992. Phone conversation
between B. Lusher of Acurex Environmental and S. Londerville of The Coen
Company.
3-xxiii
-------
4. EMISSION FACTOR DEVELOPMENT
In each AP-42 section, tables of emission factors are presented for pollutants
associated with equipment operations within the subject source category. The reliability
or quality of each of these emission factors is indicated in the tables by an overall
Emission Factor Quality Rating ranging from A (excellent) to E (poor). These ratings
incorporate the results of the quality and quantity evaluations on the data sets used to
calculate the final emission factors, as described in Chapter 3. The overall emission
factor quality ratings are described as follows:
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 so that variability within the source category population may be
minimized.
IB-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. As in the A-rating,
the source category is specific enough so 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.. As in the A-rating,
the source category is specific enough so 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 noted in the emissions factor table.
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
4-xxiv
-------
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.
The use of these criteria is somewhat subjective and depends to an extent on
the individual reviewer. Details of the rating of each candidate emission factor are
provided in the following paragraphs.
4.1 CRITERIA POLLUTANTS
4.1.1 Review of Previous Data
Previously, criteria pollutant emission factors were assumed to be equivalent to
the natural gas emission factors on a Ib/MMBtu (or kg/MW) basis, except for the SOX
emission factor which was based on fuel sulfur content. There was little quantitative
basis for this assumption, however.
Using the new AP-42 evaluation criteria, the prior criteria pollutant emission
factors were designated as E quality because these factors were developed from
essentially no LPG data and from an erroneous assumption about comparability to
natural gas. Although the SOX emission factor is based on fuel sulfur content, the
relative amounts of S02 and S03 (which make up the SOX) are not based on LPG data
but rather the assumption that all input fuel sulfur is emitted as S02. For this reason,
the emission factor for S02 was designated as E quality.
4.1.2 Review of New Data
The Weishaupt report contains one baseline NOX test for natural gas, and
propane fired in a 8 MW (23 MMBtu/hr) boiler at full load.1 For this test, the ratio of
propane NOX emissions to natural gas NOX emissions was 1.6. An emission factor of
1.3 kg N0x/thousand liters (11.4 Ib N0x/thousand gallons) was calculated for the full
load concentration of N0xfrom propane combustion using the F-factorfrom 40 CFR
Reference Method 19. This value is lower than the NOX emission factor of 1.5 kg
N0x/thousand liters (12.4 Ib N0x/thousand gallons) contained in the 1982 version of
Section 1.5.
An article from IHI Engineering Review presents NOX concentrations from a pilot
scale test furnace for methane, propane, and butane.2 These concentrations indicate
(1) that NOX emissions from propane combustion are 1.3 times greater than those for
natural gas and (2) that NOX emissions from butane combustion are 1.5 times greater
4-xxv
-------
than those for natural gas. Using the calculation procedure from 40 CFR Appendix A
Reference Method 19, the propane emission factor is 1.7 kg N0x/thousand liters (14.6
Ib N0x/thousand gallons) and the butane emission factor is 2.2 kg N0x/thousand liters
(19.0 Ib N0x/thousand gallons). The emission factors from the 1982 version of Section
1.5 are 1.5 kg N0x/thousand liters (12.4 Ib N0x/thousand gallons) for propane and 1.6
kg N0x/thousand liters (13.2 Ib N0x/thousand gallons) for butane.
Data from both sources indicate that NOX emissions from propane and butane
combustion are significantly higher than the NOX emissions generated from natural gas.
The data presented in both of these sources do not specify any protocol for the NOX
emissions data collection. It is assumed that the data were collected using continuous
emission monitoring equipment. No information is presented on the amount of testing
conducted to generate the results. Therefore, both of these data sources were
assigned a C data quality ranking. These sources are useful for comparing propane
and butane emissions to natural gas emissions.
The SOX emission factor was recalculated using the exact procedure that was
used to generate the previous emission factor in 1982. The ratio of m3 gas/thousand
liters liquid was changed based on information in Reference 3. The calculated propane
SOX emission factor (in kg S02/1,000 liters) is 0.012 S, where S is fuel sulfur content
expressed as gr sulfur/100 ft3 gas vapor; the butane emission factor (in kg S02/1,000
liters) is 0.011 S. For example, if the butane sulfur content is 0.18 gr/100 ft3, the
emission factor would be (0.011 xO.18 =) 0.0020kg S02/1,000 liters of butane burned.
Sample calculations for the S02 emission factor and other unit conversions are
contained in Appendix A.
The S02 emission factors have been assigned a rating of E because the
emission factors are based on feed gas sulfur content and actual S02 emissions were
not measured. These emission factors are conservative and assume all of the input
sulfur is emitted as S02.
4.1.3 Compilation of Emission Factors
The development approach for new S02 emission factors is discussed above.
Due to the lack of LPG combustion emissions data, emission factors for other criteria
pollutants in this version were based on the natural gas combustion criteria emission
4-xxvi
-------
factors. The natural gas emission factors were converted to LPG factors on a heating
value basis using the following heating values: 9,340 kcal/m3 (1050 Btu/ft3) for natural
gas; 6,090 kcal/liter (91,500 Btu/gal) for propane; and 6,830 kcal/liter (102,600 Btu/gal)
for butane3. The NOX emission factors calculated from the natural gas emission factors
for propane and butane were multiplied by the average ratio of propane and butane
emissions to natural gas emissions. The average ratio of NOX emissions from propane
and butane to natural gas is 1.5. This ratio is based on a limited amount of data, but
does provide an order of magnitude estimate of the NOX emissions from propane and
butane. The TOC emission factor was calculated by adding the methane and non-
methane VOC emission factor. The base natural gas emission factors and the
corresponding propane and butane emission factors are summarized in Tables 4-1 and
4-2.
4.2 NONCRITERIA POLLUTANTS
No data were available for emissions from LPG combustion for VOC speciation,
air toxics, and N20. In the absence of any data, no emission factors can credibly be
developed. The natural gas and distillate oil emission factors in AP-42 Sections 1-3
and 1-4 can be consulted for order of magnitude values, but there is no defensible
basis to assume the emission factors would be the same for LPG or any other
conventional fuel.
For the C02 emission factor, the assumption was made that all carbon in the fuel
is oxidized to C02 and emitted from the stack. Based on this assumption, the emission
factors are:
For propane: 1,500 kg C02/1,000 (12,500 Ib C02/1,000 gal), and
For butane: 1,760 kg C02/1,000 (14,700 Ib C02/1,000 gal).
4.3 FUGITIVE EMISSIONS
Fugitive emissions were not considered for emission factor development in the
previous version of Section 1.5. Chapter 4 of AP-42 contains emission factors for
fugitive releases for most of the activities associated with LPG storage, distribution and
transfer. Additional data compilation was attempted for fugitive releases during
distribution which were not fully treated in Chapter 4. A total of 10 references were
documented and reviewed during the literature search. This group of 10 documents
4-xxvii
-------
was reduced to one primary reference using the criteria outlined in Chapter 3. The nine
rejected reference documents were not used because of inadequate documentation of
operational conditions which could be generalized to emission factor development, or
lack of relevance to LPG.
Emission factors for LPG handling and storage are found in Chapter 4 of AP-42.
The VOC emission factors for the fuel feed system selected for inclusion in AP-42 were
taken directly from the Petroleum Refineries study and the 24-unit SOCMI study
included in Reference 4. The factors specified for LPG-fired boilers are the EPA-
approved values for valves, flanges, relief valves and open-ended lines handling gases.
The data used in calculating these values varied widely for different fluids but the
emission factors were derived through a well-described approach developed for the
SOCMI fugitive emissions standards proposed by EPA in January, 1981. The
procedure used estimates of leaking and non-leaking source emission factors from the
refinery data set and applied these factors to the leak frequencies found in the SOCMI
24-unit screening study to yield emission factors for average SOCMI units. The
resultant "average" SOCMI factors were considered valid for this update. Table 4-3
presents the proposed VOC emission factors for LPG feed systems.
4-xxviii
-------
TABLE 4-1. NATURAL GAS COMBUSTION EMISSION FACTORS
Emission factors, kg/106 m3 (lb/106 ft3)
Pollutant
Filterable Particulate Matter
N°x
CO
TOC
SO2
Small industrial boilers3
100(6.2)
2,258(140)
565 (35)
94 (5.8)
9.7 (0.6)
Commercial boilers'3
73 (4.5)
1600(100)
339(21)
94 (5.8)
-
a Heat input of 10 - 100 MMBtu/hr.
b Heat input of 0.3 - <10 MMBtu/hr.
4-xxix
-------
TABLE 4-2. LPG COMBUSTION EMISSION FACTORS
Butane emissions,
kg/1 000 l(lb/1 000 gal)
Pollutant
Filterable PM
NOX
CO
CO2
TOC
Industrial
boilers
0.07 (0.6)
2.5(21)
0.42 (3.6)
1,760(14,700)
0.07 (0.6)
Commercial
boilers
0.06 (0.5)
1.8(15)
0.25 (2.1)
1,760(14,700)
0.07 (0.6)
Propane emissions,
kg/1 000 I (lb/1 000 gal)
Industrial
boilers
0.07 (0.6)
2.3(19)
0.38 (3.2)
1,500(12,500)
0.06 (0.5)
Commercial
boilers
0.05 (0.4)
1.7(14)
0.22(1.9)
1,500(12,500)
0.06 (0.5)
a All of the criteria pollutants were given an emission factor rating of E. All E-rated emission factors are
based on C and D quality data from a small number of facilities not necessarily representative of the
industry.
TABLE 4-3. FUGITIVE EMISSION FACTORS FOR VARIOUS EQUIPMENT TYPES'
Equipment type
Valve
Relief valve
Open-ended line
Flange
Emission
kg/hr/source
0.0056
0.1040
0.0017
0.00083
factor,
Ib/hr/source
0.00254
0.0417
0.000771
0.000376
Factor
rating
D
D
D
D
4-xxx
-------
REFERENCES FOR CHAPTER 4
1. Nitrous Oxide Reduction with the Weishaupt Flue Gas Recirculation System,
Weishaupt Research and Development Institute, Jan. 1987.
2. Some Examples of Combustion Tests for Putting New Fuels to Practical Use,
Ikebe etal, IHI Engineering Review, Oct. 1976.
3. Perry's Chemical Engineers' Handbook (Sixth Edition), Robert H. Perry and Don
W. Green (Editors), McGraw-Hill Book Company, New York, New York, 1984.
4. Fugitive Emission Sources of Organic Compounds - Additional Information on
Emissions, Emission Reductions, and Costs, EPA-450/3-82-010, U.S.
Environmental Protection Agency, RTP, NC, April 1982.
4-xxxi
-------
5. AP-42 SECTION 1.5: LIQUEFIED PETROLEUM GAS COMBUSTION
The revision to Section 1.5 of AP-42 is presented in the following pages as it
would appear in the document. A marked-up copy of the 1982 version of this section is
included in Appendix B.
5-xxxii
-------
APPENDIX A
SAMPLE CALCULATIONS
A-xxxiii
-------
SCX Emission Factor Calculation
(S grains/100 ft3)(lb/7000 grains)(ft3/103 gal)(2 Ib SO2/lb S) = S Ib SO2/103 gal
From 40 CFR Appendix A. Reference Method 19.
ppmtolb/106Btu:
ppm NOX @ 3% O2
ppm NOX x (1.194 x 10'1 Ib/scf/ppm NOX) = Ib/scf
E = Ib/scf (8710 scf/106 Btu)(20.9/(20.9-3% O2) = lb/106 Btu
lb/106Btutolb/103gal:
91,500 Btu/gal Propane
102,600 Btu/gal Butane
1050 Btu/ft3 Natural Gas
Ib pollutant/103 gal = Ib pollutant/106 ft3 natural gas x 1050 Btu/ft3 x 0.915 x 108 Btu/103 gal
Ib pollutant/103 gal = Ib pollutant/106 ft3 natural gas x 1050 Btu/ft3 x 1.026 x 108 Btu/103 gal
A-xxxiv
-------
APPENDIX B
MARKED-UP 1982 SECTION 1.5
B-xxxv
-------
REPORT ON REVISIONS TO
5TH EDITION AP-42
Section 1.5
Liquefied Petroleum Gas Combustion
Prepared for:
Contract No. 68-D2-0160, Work Assignment 50
EPA Work Assignment Officer: Roy Huntley
Office of Air Quality Planning and Standards
Office of Air and Radiation
U. S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
Prepared by:
Eastern Research Group
Post Office Box 2010
Morrisville, North Carolina 27560
December 1996
B-xxxvi
-------
Table of Contents
Page
1.0 INTRODUCTION 1-1
2.0 REVISIONS 2-1
2.1 General Text Changes 2-1
2.2 Greenhouse Gases 2-1
2.2.1 Carbon Dioxide, CO2 2-1
2.2.2 Methane, CH4 2-2
2.2.3 Nitrous Oxide, N2O 2-2
3.0 REFERENCES 3-1
4.0 REVISED SECTION 1.5 4-1
5.0 EMISSION FACTOR DOCUMENTATION, APRIL 1993 5-1
in
-------
1.0 INTRODUCTION
This report supplements the Emission Factor (EMF) Documentation for AP-42 Section 1.5, Liquified Petroleum Gas
Combustion, dated April 1993. The EMF describes the source and rationale for the material in the most recent updates to the 4th
Edition, while this report provides documentation for the updates written in both Supplements A and B to the 5th Edition.
Section 1.5 of AP-42 was reviewed by internal peer reviewers to identify technical inadequacies and areas where state-of-the-
art technological advances need to be incorporated. Based on this review, text has been updated or modified to address any technical
inadequacies or provide clarification. Additionally, emission factors were checked for accuracy with information in the EMF
Document and new emission factors generated if recent test data were available.
If discrepancies were found when checking the factors with the information in the EMF Document, the appropriate reference
materials were then checked. In some cases, the factors could not be verified with the information in the EMF Document or from the
reference materials, in which case the factors were not changed.
Four sections follow this introduction. Section 2 of this report documents the revisions and the basis for the changes.
Section 3 presents the references for the changes documented in this report. Section 4 presents the revised AP-42 Section 1.5, and
Section 5 contains the EMF documentation dated April 1993.
1-1
-------
2.0 REVISIONS
This section documents the revisions made to Section 1.5 of the 5th Edition of AP-42.
2.1 General Text Changes
Text was added concerning firing practices based on information in the EMF Document. Also, at the request of EPA, metric
units were removed.
2.2 Greenhouse Gases
2.2.1 Carbon Dioxide, CO2
The CO2 factors in Table 1.5-1 were recalculated assuming 99.5 percent conversion of fuel carbon to CO2 instead of the
100 percent used in earlier AP-42 Version 5. Assuming 99.5 percent conversion of fuel carbon content to CO2 during combustion, the
CO2 emission factors in Table 1 were computed.1"3
Table 1. Default CO2 Emission Factors for Liquid Fuels
Quality Rating: B
2-1
-------
Fuel Type
LPG-Propane
LPG-Butane
%ca
80.7
80.7
Density*
(Ib/gal)
4.24
4.84
New Emission
Factor
(lb/1000 gal)a
12,500
14,300
AP-42 EF
(lb/1000 gal)
12,500
14,700
AP-42 Rating
E
E
a Reference 4. "B" factor assigned because of the varying compositions of LPG gas.
b References 5-6.
0 The following equation was used to develop the emission factor equation for LPG:
44 Ib C02
12 Ib C
x 0.995 x 4.24
Ib
gal
100%
x 80.7% x 1000 = 12,500
Ib C02
1000 gal
2-2
-------
Where: 0.995 = fraction of fuel oxidized during combustion (Reference 1-3);
44 = molecular weight of CO2;
12 = molecular weight of carbon; and
4.24 Ib/gal = density of LPG-Propane (AP-42 Appendix A).
2.2.2 Methane, CH4
Table 1.5-2 lists total organic compound (TOC) emission factors but does not contain a factor for CH4. One factor for
commercial/institutional data was obtained, but source test data was unavailable for verification. This factor was therefore assigned an
"E" rating and appears in Table 2:
Table 2. CH4 Emission Factors for LPG Combustion3
(Ib CH4/1000 gal)
Combustion Category
Commercial/institutional boilers
EF Rating
E
EF
0.25
a Reference 7.
2.2.3 Nitrous Oxide, N2O
Only one value for N2O emissions from LPG combustion (Reference 2) was available. The source data was not listed, so the
factor was given an "E" rating.
Table 3. N?O Emission Factors for LPG Combustion3
-------
(Ib N2O/1000 gal)
Process
Industrial Boilers
EF Rating
E
EF
0.9
Reference 2.
2-4
-------
3.0 REFERENCES
1. G. Marland and R.M. Rotty, "Carbon Dioxide Emissions from Fossil Fuels: A procedure for Estimation and Results for 1951-
1981," DOE/NBB-0036 TR-003, Carbon Dioxide Research Division, Office of Energy Research, U.S. Department of Energy,
Oak Ridge, TN, 1983.
2. A. Rosland, Greenhouse Gas Emissions In Norway: Inventories And Estimation Methods, Oslo: Ministry of Environment,
1993.
3. Sector-Specific Issues And Reporting Methodologies Supporting The General Guidelines For The Voluntary Reporting Of
Greenhouse Gases Under Section 1605(b) Of The Energy Policy Act Of 1992, DOE/PO-0028, Volume 2 of 3, U.S. Department
of Energy, 1994.
4. K. Dahlberg et al., Emissions OfN2O, CO, CO2, COS, and CS2 From Stationary Combustion Sources. Institute Vatten-
Luftvaardsforsk, Sweden, [Publ.] B, IVL B 891, 1988. 23 pp.
5. R. H. Perry and D. Green, Perry's Chemical Engineers* Handbook, Sixth ed., New York: McGraw Hill, 1984.
6. Compilation Of Air Pollutant Emission Factors, Volume I: Stationary Point And Area Sources, U. S. Environmental Protection
Agency, AP-42. Fifth Edition. Research Triangle Park, NC, 1995.
7. Ortech Corporation, Inventory Methods Manual For Estimating Canadian Emissions Of Greenhouse Gases, Prepared for
Environment Canada, 1994.
3-1
-------
4.0 RE VISED SECTION 1.5
This section contains the revised Section 1.5 of AP-42, 5th Edition. The electronic version can be located on the EPA TTN at
http://www.epa.gov/ttn/chief/ap42cl.html
4-1
-------
5.0 EMISSION FACTOR DOCUMENTATION, APRIL 1993
This section contains the Emission Factor Documentation for Section 1.5, dated April 1993. The electronic version can be
located on the EPA TTN at http://www.epa.gov/ttn/chief/ap42cl.html. The zipped file on the TTN contains this (1996) background
report as well as the 1993 Emission Factor Documentation.
5-1
------- |