United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park, NC 27711
EPA-453/R-98-001
March 1998
Air
£EPA Basis and Purpose Document
on Specifications for
Hydrogen-Fueled Flares
g S Environmental Protection Agency
Region 5, Library (PL-12J)
77 West Jackson Boulevard, 12th floor
Chicago, IL 60604-3590
-------
-------
Basis and Purpose Document on
Specifications For Hydrogen-Fueled Flares
Emission Standards Division
U.S. Environmental Protection Agency
Office of Air Radiation
Office of Air Quality Planning Standards
Research Triangle Park, North Carolina 27711
March 1998
-------
-------
DISCLAIMER
This Report has been reviewed by the Emission Standards Division
of the Office of Air Quality Planning and Standards, EPA, and
approved for publication. Mention of trade names or commercial
products is not intended to constitute endorsement or
recommendation for use.
-------
ENVIRONMENTAL PROTECTION AGENCY
Basis and Purpose
Document on Specifications for Hydrogen-Fueled Flare
1. The action for which this Basis and Purpose Document was
written, amends the General Control Device Requirements (40
CFR part 60.18) which were issued as a final rule on January
21, 1986, and the Control Device Requirements (40 CFR part
63.11) which were issued as a final rule on March 16, 1994.
The accompanying action adds specifications for hydrogen-
fueled flares to the existing flare specifications for
organic containing vent streams.
2. For additional information contact:
Mr. Robert Rosensteel
Organic Chemicals Group
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Telephone: (919) 541-5608
3. Paper copies of this document may be obtained from:
U.S. Environmental Protection Agency Library (MD-36)
Research Triangle Park, NC 27711
Telephone: (919) 541-2777
National Technical Information Service (NTIS)
5285 Port Royal Road
Springfield, VA 22161
Telephone: (703) 487-4650
in
-------
Electronic copies of this document may be obtained from the
EPA Technology Transfer Network (TTN). The TTN is an
electronic bulletin board system which provides information
and technology exchange in various areas of emissions
control. The TTN can be accessed through the TTN web site:
www.epa.gov/ttn/oarpg/rules.html. For further information,
contact the TTN HELP line at (919) 541-5384, from 1:00 p.m.
to 5:00 p.m., Monday through Friday.
IV
-------
TABLE OF CONTENTS
1.0 INTRODUCTION 1
2.0 BACKGROUND 2
2.1 Existing Flare Requirements 2
2.2 Hydrocarbon Flare Studies Used to Establish the
Specifications for 40 CFR 60.18 and 40 CFR 63.11 . . 5
2.3 DuPont's Hydrogen Flare Status 5
3.0 SUMMARY OF DUPONT PROGRAM 7
3.1 Background on Hydrogen-Fueled Flare Studies .... 7
3.2 Testing Program 8
4.0 SUMMARY OF THE RECOMMENDED SPECIFICATIONS 15
5.0 RATIONALE FOR THE RECOMMENDED SPECIFICATIONS 16
5.1 The Need for Specifications for
Hydrogen-Fueled Flares 16
5.2 The Use of DuPont Test Results as the Basis for
Hydrogen-Fueled Flare Specifications 18
5.3 Selection of the Specifications for
Hydrogen-Fueled Flares 23
6.0 SUMMARY OF ENVIRONMENTAL, ENERGY, AND COST IMPACTS ... 25
6.1 Primary Air Impacts 25
6.2 Other Environmental Impacts 26
6.3 Energy Impacts 26
6.4 Cost and Economic Impacts 26
6.5 Summary of Impacts 27
7.0 SUMMARY 27
-------
LIST OF FIGURES
Figure 1. Hydrogen Volume Fractions at Lift-Off and
Blow-Out Of A 2.9 inch plain pipe flare .
13
Figure 2. Relationship of Ethylene Destruction Efficiencies
to the Ratio of Actual Hydrogen Content
to the Hydrogen Content
14
LIST OF TABLES
TABLE 1. GENERAL CONTROL DEVICE REQUIREMENTS FOR FLARES
CONTAINED IN 40 CFR 60.18. AND 40 CFR 63.11 3
TABLE 2. DEPENDENCE ON TIP VELOCITY OF THE RATIO OF THE PROPOSED
CRITICAL HYDROGEN VOLUME FRACTION TO THE HYDROGEN
VOLUME FRACTION AT BLOW OUT WITH PILOT 15
VI
-------
1.0 INTRODUCTION
The General Control Device Requirements of 40 CFR 60.18 were
issued as a final rule on January 21, 1986 and are applicable to
control devices complying with New Source Performance Standards
(NSPS) promulgated by the Agency under Section 111 of the Clean
Air Act (CAA), and National Emission Standards for Hazardous Air
Pollutants (NESHAP) issued under the authority of Section 112
prior to the CAA Amendments of 1990. The Control Device
Requirements of 40 CFR 63.11 were issued as a final rule on
March 16, 1994 and are applicable to control devices used to
comply with NESHAP issued under the authority of the CAA
Amendments of 1990, for the control of hazardous air pollutants
(HAP). Both of these existing control device requirements
contain specifications defining required operating conditions of
control devices. Specifically, 40 CFR 60.18(b) through (d), and
40 CFR 63.11(b) contain the operating conditions for flares
(i.e., existing flare specifications). Flares operating in
accordance with these specifications destroy volatile organic
compounds (VOC) or volatile hazardous air pollutants (HAP) with a
destruction efficiency of 98 percent or greater. These existing
flare specifications were written for flares combusting organic
emission streams.
E.I. DuPont de Nemours and Company (DuPont) representatives
requested that the EPA either add specific limits for hydrogen-
fueled flares to the existing flare specifications or approve
their hydrogen-fueled flares as alternate means of emission
limitation under 40 CFR 60.484, 40 CFR 61.12(d) and 40 CFR
63.6(g). DuPont subsequently sponsored a testing program to
demonstrate that hydrogen-fueled flares in use at DuPont destroy
emissions with 98 percent or greater efficiency. The test
program demonstrated that these hydrogen-fueled flares achieved
greater than 98 percent destruction efficiency. Further, the EPA
judged the conditions of the test program to be universally
applicable, with certain limitations. This document provides the
-------
background and rationale for the action to add specifications for
hydrogen-fueled flares to the existing flare specifications.
This document is organized as follows. After this
introduction, Section 2.0 provides background of the existing
flare specifications and the studies used to establish them,
along with a description of DuPont's hydrogen-fueled flare
status. Section 3.0 summaries the DuPont program that was
designed to demonstrate that their hydrogen-fueled flares were
equivalent to flares meeting the existing flare specifications.
Section 4.0 provides a summary of the hydrogen-fueled flare
specifications that are being added to the existing flare
specifications, and Section 5.0 contains the rationale for these
recommended hydrogen-fueled flare specifications. Section 6.0
provides a summary of the anticipated impacts, and Section 7.0
summarizes this document. In this document, references are noted
by their docket item number in Docket A-97-48. Appendix A to
this document is the index for the relevant portions of Docket A-
97-48.
2.0 BACKGROUND
2.1 Existing Flare Requirements
Flares are commonly used in industry to safely combust VOC
and volatile HAP. Flares can accommodate fluctuations in VOC or
volatile HAP concentrations, flow rate, heating value, and inerts
content. Further, flares are appropriate for continuous and
intermittent flow applications. Some organic emission streams
can be flared without the need for supplemental fuel. However,
the use of supplemental organic fuel such as natural gas to
ensure the complete combustion of emissions is common.
The existing flare specifications contained in 40 CFR 60.18
and 40 CFR 63.11 are based upon experience with waste streams
containing organic substances. These existing flare provisions
-------
require that the waste gas being flared have a minimum heat
content, which is specific to the flare head design, and a
maximum waste gas flow rate. The rules mandate that flares be
designed for, and operated with, no visible emissions, except for
periods not to exceed a total of five minutes during any two
consecutive hours. In addition, the existing flare
specifications require that the flare must be operated with a
flame present at all times. The presence of a flare pilot flame
is to be monitored to ensure that a flame is present at all
times. The minimum net heating value of the gas being combusted
and the maximum exit velocity of steam assisted, air assisted,
and nonassisted flares, as specified in 40 CFR 60.18 and 40 CFR
63.11, are provided in Table 1.
TABLE 1. GENERAL CONTROL DEVICE REQUIREMENTS FOR FLARES
CONTAINED IN 40 CFR 60.18. AND 40 CFR 63.11
Net Heating Value of
Flare Type Combusted Gas, HT Allowable Velocity, V
(megajoules per standard (meters per second)
cubic meter)
Air-Assisted 11.2 i HT V -< Vmaxa
Steam-Assisted 11.2 i HT V -c 18.3
Nonassisted 7.45 < HT V x 18.3
Steam-Assisted or 37.3 -< HT 18.3 < V -< 122
Nonassisted
Alternative for Steam- 11.2 < HT s 37.3
Assisted or
Nonassisted
V « vmax> and,
V < 122
avmax= 8.706+0.7084 (HT)
bLog10 (Vmax) = (Hr + 28.8)731.7
As shown in Table 1, air-assisted flares must operate with
an exit velocity less than the calculated maximum allowable
velocity, Vmax, which is calculated from an equation. Also, an
equation is provided to calculate the maximum exit velocity for
nonassisted and steam-assisted flares, as an alternative to the
-------
allowable velocities listed in the table. With steam-assisted
and nonassisted flares, there are two options: (1) calculate the
maximum allowable velocity from the equation, and verify that the
exit velocity is below the calculated maximum allowable velocity,
or (2) verify that the exit velocity is below the given Vmax
values for the heat content of the stream. Table 1 lists the
allowable velocities for the possible heat contents.
The net heating value of the gas being combusted in a flare,
which the owner/operator is required to calculate for all flare
types, is calculated using Equation 1.
Equation I:
HT = K E CIH±
2=1
where:
HT = Net heating value of the sample, Mega Joules per standard
cubic meter (MJ/scm); where the net enthalpy per mole of
off-gas is based on combustion at 25°C and 760 mm Hg, but
the standard temperature for determining the volume
corresponding to one mole is 20°'C.
K= Constant = 1.740X10-' A \\g-*ole}( MJ
scm j ( kcal)
where: ppmv = parts per million by volume, and
kcal = kilo calories
Concentration of sample component i in ppmv on a wet basis,
as measured for organics by Method 18, 40 CFR part 60,
appendix A, and measured for hydrogen and carbon monoxide by
American Society for Testing and Materials (ASTM) Method
D1946-77 (incorporated by reference as specified in 40 CFR
63.14 and 40 CFR 60.17).
Net heat of combustion of sample component i, kcal/g-mole at
25°C and 760 mm Hg. The heats of combustion may be
determined using ASTM Method D2382-76 (incorporated by
reference as specified in 40 CFR 63.14 and 40 CFR 60.17) if
published values are not available or cannot be calculated.
-------
n = Number of sample components.
2.2 Organically-Fueled Flare Studies Used to Establish the
Existing Specifications for 40 CFR 60.18 and 40 CFR 63.11
The EPA determined the destruction efficiency of flares
combusting volatile organic emissions in the early 1980's and
developed the existing flare specifications as a result of this
work. The testing was conducted with a nominal 8-inch diameter
flare head furnished by a vendor (Docket No. A-97-48, Item No. I-
11-12) and pilot-scale flares (Docket No. A-97-48, Item No. I-II-
5) .
In general, the experiments discussed showed that propane-
in-nitrogen mixtures generate stable flames when the heat content
of the mixture is above 200 Btu/scf. These experiments also
showed that the combustion and destruction efficiencies of flares
with waste streams containing organic substances are high (at
least 98 percent) as long as the flame produced by the flare head
at the given operating conditions was stable, based upon gas heat
content and velocity.
2.3 DuPont's Hydrogen-Fueled Flare Status
DuPont owns and operates six flares which are used to
combust waste gases containing hydrogen (from 13 to 22 volume
percent), inert gases (nitrogen, argon, carbon dioxide, and
steam), oxygen (in some streams), and various combinations of the
hazardous air pollutants (HAP) in the 115 ppm to 5 percent mole
fraction (by volume) concentration range.
These six DuPont flares are nonassisted (pipe) flares, and
each flare is similar in respect to the amount of hydrogen in the
gas stream. The hydrogen and other flammable gas concentrations
are such that gas heating values are in the range of 59 to
120 Btu/scf. Typical exit velocities for DuPont's hydrogen-
-------
fueled flares are on the order of 100 ft/s. All six of DuPont's
hydrogen-fueled flares are equipped with continuous pilots to
ensure flame stability.
As stated in the previous section, the existing flare
specifications are based upon existing data which show that
combustion efficiencies greater than 98 percent are achieved when
specific heat content and velocity requirements are met. These
data also show that combustion efficiency is related to flame
stability. A flame is considered stable when the heating value
is high enough to sustain a flame that is void of separations
between the flare tip and any part of the flame.
Because the concentrations of the combustible gases are low,
and because the heating value of hydrogen per unit of volume is
low, the DuPont waste streams have low volumetric heat contents
compared with streams containing volatile organics. Therefore,
DuPont's flares do not meet the existing flare specifications of
40 CFR 60.18 and 40 CFR 63.11. As discussed earlier, these
standards, developed primarily for flares with waste streams
containing organic substances, set minimum heat content limits
and maximum velocity specifications at which the flare can
operate. To bring DuPont's hydrogen-fueled flares into
compliance with the existing flare specifications would require
the heat content of the waste stream be augmented with natural
gas. DuPont estimates that the cost of the natural gas would be
approximately $2.8 million per year in order for their six
hydrogen-fueled flares to meet the existing flare specifications.
In March of 1997, DuPont requested that the EPA either add
specific limits for hydrogen-fueled flares to the existing flare
specifications or approve their hydrogen-fueled flares as
alternate means of emission limitation (Docket No. A-97-48, Item
No. II-D-2). DuPont subsequently sponsored a testing program to
demonstrate that hydrogen-fueled flares in use at DuPont destroy
emissions with greater than 98 percent efficiency. Following is
a summary of DuPont's program.
-------
3.0 SUMMARY OF DUPONT PROGRAM
To support petitions to the EPA for approval of hydrogen-
fueled flares as equivalent in performance to the requirements of
40 CFR 60.18 and 40 CFR 63.11, DuPont initiated a two-phase
program. The first phase was to gather background information on
hydrogen-fueled flare studies, and the second phase was the
testing program.
3.1 Background on Hydrogen-Fueled Flare Studies
The objective of the first phase of DuPont's study was to
assemble available information on the flaring of hydrogen and
hydrocarbon gas mixtures to support DuPont's equivalency claim
for hydrogen-fueled flares. The conclusions of DuPont's Phase I
study (Docket A-97-48, Item II-I-2) were reported as follows.
A large body of data exists on the combustion
efficiency of flares incinerating volatile organic
waste gas mixtures. These data show that the
combustion efficiency is related to flame stability.
Federal regulations are based upon existing data on the
flaring of waste streams containing volatile organic
substances, which demonstrate that combustion
efficiencies greater than 98 percent are achieved when
specific heat content and velocity requirements are
met.
Available information on flaring of hydrogen-based
waste gas mixtures indicates that hydrogen gas mixtures
can be burned over a significantly wider range of
velocities and heat content conditions than organic gas
mixtures.
The results of small scale flare and diffusion flame
experiments can be used to develop stability limits for
flaring of hydrogen-inert gas mixtures. However, there
is a wide range of uncertainty in the stability limits
of lean hydrogen-inert mixtures.
-------
• DuPont's hydrogen-fueled flares appear to be operating
at or within the stability limits established from
small scale studies. However, unless it can be shown
that DuPont's flares exceed the stability limits for
lean hydrogen-inert gas mixtures, it is not believed
that this information is sufficient to successfully
petition the U.S. EPA for an exemption.
• This study was not able to locate any information
relating the stability of hydrogen flames to their
organic destruction efficiency. This is expected to be
significant since a central argument in establishing
the equivalency of hydrogen flares is that high
combustion efficiency is concomitant with flame
stability.
Further, the study concluded that it was not expected that
the previous hydrogen flare studies could be used to petition for
a variance for DuPont's flares. Therefore, the study recommended
that the second phase of the program be implemented.
3.2 Testing Program
The second phase of DuPont's program was composed of a
series of tests intended to demonstrate that the hydrogen-fueled
flares at their facilities were achieving a volatile HAP and VOC
destruction efficiency equal to or greater than that of flares
meeting the existing flare specifications.
Testing Plan
The test program was designed to experimentally establish
the stability limits and destruction efficiency of DuPont's
flares under the range of chemical compositions and operating
conditions at the three DuPont facilities with the six hydrogen-
fueled flares. The testing plan was made up of stability and
destruction efficiency tests using a nominal 3-inch diameter
flare under conditions otherwise representative of the DuPont
flares. The nominal 3-inch diameter flare was chosen to provide
a link with the previous flare combustion efforts (Docket A-97-
-------
48, Item Nos. II-I-3, 4, and 5), because that was the size of
flare used for those studies as well. The specific goals of the
test plan were (1) to quantify the stability envelope (minimum
gas hydrogen content versus exit velocity for flame stability)
for hydrogen/waste gas mixtures having hydrogen concentrations
and velocities in the range of DuPont's flares, and (2) to
determine the destruction efficiencies of a surrogate organic
compound added to the flare gas, at selected combinations of gas
composition and velocity that are known to produce stable flames.
Test Results
As noted above, the tests were designed to determine the
flame stability envelope and the destruction efficiency that a
stable flame at a set velocity is able to achieve. The
experiments were done by establishing a stable flame at the
desired velocity, then slowly decreasing the hydrogen flowrate
and recording the velocity and hydrogen content at flame lift off
and again at blow out. Lift off was defined as the time when a
portion of the flame was permanently separated from the flare
tip. Blow out was defined as complete absence of the flame.
The measurements of the hydrogen volume percent at lift off
and blow out for the piloted and unpiloted nominal 3-inch
(2.9 inch inner diameter) pipe flare are shown in Figure I as a
function of velocity. Because the hydrogen content at lift off
was essentially the same for flares with and without a pilot
burner, a single line was fit to the data sets of lift off
measurements for piloted and unpiloted flares; this is
represented by the upper curve in Figure I and by Equation 2.
The data point in the far upper right corner of the figure is an
unexplained outlier that is inconsistent with all other data
points and was excluded from the linear regression analysis of
the lift off data set. The middle and lower curves in Figure 1
are the blow-out curves without and with a pilot, respectively.
-------
Equation 2:
- 6.0
off
16.3 < u < 122 ft/s, with and without a pilot
where:
lift off = The hydrogen content at lift off of piloted and
unpiloted flames, percent.
u = The tip velocity, ft/s.
Destruction efficiencies were determined at eight different
combinations of tip velocity, hydrogen content, ethylene content
(ethylene was the surrogate for which the destruction efficiency
was determined), the presence or absence of a pilot, and high and
low wind conditions. In all cases, the destruction efficiencies
were greater than 98 percent, at a level of 95 percent
confidence. Further, control efficiencies greater than
98 percent were found at hydrogen contents below the lift-off
curve.
The data used in the development of the existing flare
specifications showed that 98 percent destruction efficiency was
achievable by maintaining the heating value of the flare gas a
critical value above the minimum heating value required for flame
stability. The critical value was found to lie in the range from
approximately 1.1 to 1.3 times the minimum heating value for
flame stability.
A similar analysis was conducted for DuPont's current study
of hydrogen-fueled flares. The reference condition for stability
that was used was the hydrogen content of the flare at lift off.
The measured mean ethylene destruction efficiencies and
destruction efficiencies at the 95 percent confidence level are
shown as a function of the ratio of the actual hydrogen content
of the flare gas to the hydrogen content at lift off at the same
10
-------
tip velocity in Figure 2. As observed in Figure 2, all of the
destruction efficiency at conditions more stable than lift off
(stability ratio greater than 1) are above 99 percent. The
DuPont report concluded that extrapolation of the data to the
left of Figure 2 (i.e., for hydrogen content ratios less than
1.0) suggests that the destruction efficiency would be assured
for values of the stability ratio greater than about 0.95 to
0.97. The report recommended that a conservative criterion for
assuring 98 percent destruction in hydrogen flares is that the
stability ratio, or ratio of the hydrogen content to that at lift
off at the same tip velocity, be equal or greater than 1.0.
The choice of the hydrogen content at lift off as the
critical condition places the ratio of the critical to the
minimum value at blow out with pilot at the values shown in
Table 2, as a function of tip velocity. As shown in Table 2, the
excess hydrogen content for minimum flame stability is seen to
increase from a low of 15 to 17 percent at the higher velocities
tested, to around 30 percent at the lower velocities tested.
These data relate the ratio of hydrogen content to the tip
velocity and indirectly relate the tip velocity to the
destruction efficiency in the following manner. As discussed
previously, the hydrogen ratio is also directly proportional to
the destruction efficiency (that is, as the ratio of hydrogen in
the stream versus hydrogen at lift off increases, the destruction
efficiency of the flame increases). Therefore, it can be
concluded that the tip velocity and destruction efficiency are
inversely proportional.
Following are the general conclusions and recommendations
from the testing program.
• The results of the study generally agreed with the results
of earlier studies on the stability and performance of
hydrogen-fueled flares.
11
-------
The heat input to the flare through the pilot was found to
have a significant effect on blow out at high values of the
ratio of pilot to flare heat input.
The hydrogen content at lift off was not strongly influenced
by pilot type or heat input.
The hydrogen content at lift off was the same for piloted
and unpiloted flares.
Blow out occurred at hydrogen contents approximately 1 mole
percent less than lift off in the absence of a pilot, and
3 mole percent lower in the presence of a pilot.
The destruction efficiency was greater than 98 percent, at a
95 percent confidence level, under all conditions
investigated.
Destruction efficiency increased slightly on increasing the
ethylene content of the flare gas, indicating that the
addition of organics at percent levels does not contribute
to the deterioration of hydrogen-fueled flare performance.
Destruction efficiencies greater than 98 percent were
achieved at hydrogen contents as low as 0.955 times the
hydrogen content at lift off. Destruction efficiencies
greater than 99 percent (95 percent confidence level) were
achieved at a ratio of hydrogen content to hydrogen content
at lift off greater than 1.0.
The combination of velocity and hydrogen volume fraction at
lift off were recommended as the conservative criteria for
greater than 98 percent destruction efficiency.
Not all of DuPont's hydrogen-fueled flares meet the
conservative conditions recommended, at the time of this
testing.
12
-------
£O
x-s 20
0^
CM
X
C 15
.2
o
03
v_
LL
CD 10
£
MMB
—
m-J
"o
5
0
i i i i I 1 1
A
2.9" Plain Pipe Flare
A — . — Lift off, without pilot
A Lift off, with pilot
o Blow out, without pilot 0.99723
• BlOW OUt, With pilot °>9986° D0.99788
D Destruction Efficiency 0 gg631 n A
t9A°^Ip..^ ^ ^
...••" 70.98356
>*'cL ^^s*-
T *** ^*" * -^ T
.i,'*^ ' 0.98732 1
..-""*''' +<^\. * *
Ai""^. " " ^ ^C^^ ^'^ ^° ^'ow ou*
A .-••'"<. - " ^^ \ measurements,
c>.-''>r' ^.s^ \ with pilot
' '* ' ^>C Fit to blow out
2 ^^\. measurements,
^ k without pilot
Fit to lift off
j ; measurements,
with and without pilot
1 I 1 1 t 1 1
x
0 20 40 60 80 100 120 140
Velocity (ft/s)
Figure 1. Hydrogen Volume Fractions at Lift-Off and Blow-Out Of
A 2.9 inch plain pipe flare
(Docket A-97-48, Item No. II-I-l)
13
-------
o
c
-------
TABLE 2. DEPENDENCE ON TIP VELOCITY OF THE RATIO OF THE PROPOSED
CRITICAL HYDROGEN VOLUME FRACTION TO THE HYDROGEN VOLUME FRACTION
AT BLOW OUT WITH PILOT
Tip Velocity Ratio of H2 Content at Lift Off to
(ft/s) the H2 Content at Blow Out with Pilot
17
20
40
60
80
100
120
1
1
1
1
1
1
1
.31
.30
.25
.21
.19
.17
.15
4.0 SUMMARY OF THE RECOMMENDED SPECIFICATIONS
The recommended hydrogen-fueled flare specifications add
requirements for nonassisted flares that combust 8.0 percent (by
volume) or greater of hydrogen in the stream and have a 3-inch or
greater diameter. The recommended hydrogen-fueled flare
specifications present an equation that calculates the maximum
allowable flare tip velocity for a given volume percent of
hydrogen. This equation format is similar to the one used for
air-assisted flares in the existing flare specifications. The
specific equation for the maximum tip velocity for hydrogen-
fueled flares is:
Vmax = (XH2 - Kj)* K2
Where:
vmax = Maximum permitted velocity, m/sec.
K! = Constant, 6.0 volume-percent hydrogen.
K2 = Constant, 3 . 9 (m/sec)/volume-percent hydrogen.
15
-------
XH2 - The volume-percent of hydrogen, on a wet basis, as
calculated by using the American Society for Testing
and Materials (ASTM) Method D1946-77.
5.0 RATIONALE FOR THE RECOMMENDED SPECIFICATIONS
5.1 The Need for Specifications for Hydrogen-Fueled Flares
As discussed below, hydrogen has a lower heat content than
organics commonly combusted in flares meeting the existing flare
specifications and cannot, therefore, be used to satisfy existing
control requirements. However, since the combustion of hydrogen
is different than the combustion of organics, and the test report
demonstrates a destruction efficiency greater than 98 percent,
the EPA believes that hydrogen-fueled flares meeting the
recommended specifications will achieve a control efficiency of
98 percent or greater. This level of control is equivalent to
the level of control achieved by flares meeting the existing
specifications. In addition to achieving the same destruction
efficiency of VOC or organic HAP, these recommended
specifications have the added advantage of reducing the formation
of secondary pollutants; since the combustion of supplemental
fuel would not be required by hydrogen-fueled flares to meet the
existing flare specifications.
The Heat Content of Hydrogen
The heat content of a substance is a measure of the amount
of energy stored within the bonds between atoms in each molecule
of the substance. Hydrogen is a simple molecule consisting of
two hydrogen atoms held together by weak, hydrogen bonds, thus
resulting in a low heat content. In comparison, organic
chemicals are larger chains (or rings) of carbons with hydrogens
and other atoms attached to them. These molecules are held
together with a combination of ionic, covalent and hydrogen
16
-------
bonds, which contain substantially more energy (i.e., higher heat
content) than the hydrogen bond in the hydrogen molecule.
The Difference in Combustion Between Hydrogen and Organics
The first phenomenon to explain the difference in combustion
between hydrogen and organics is related to the thermodynamics of
the combustion reaction. In order for the hydrogen atom to react
in the combustion/oxidation reaction, the weak hydrogen bond
between the two hydrogen atoms must first be broken. Because
there is less energy holding the hydrogen atoms together, less
energy (heat) is required to separate them. Once the hydrogen
bonds are broken, the hydrogen atoms are free to react in the
combustion reaction.
The second phenomenon explaining the difference in
combustion between hydrogen and organics is due to hydrogen's
upper and lower flammability limits. The flammability limits are
the minimum (lower) and maximum (upper) percentages of the fuel
in a fuel-air mixture that can propagate a self-sustaining flame.
The lower and upper flammability limits of hydrogen are 4.0 and
74.2 percent, respectively, which represents the second widest
range of lower and upper limits of substances typically combusted
in flares (Docket No. A-97-48, Item No. II-I-2) .
The third phenomenon explaining the difference in combustion
between hydrogen and organics is the relative difference in
diffusivity between hydrogen and organics in air. Diffusivity
refers to how easily molecules of one substance mix with
molecules of another. Further, the quicker the fuel and air in a
flare mix, the quicker the combustion reaction occurs. The
measure of how quickly a substance mixes with another substances
is expressed in terms of the diffusivity coefficient. The larger
the diffusivity coefficient, the quicker the mixing. The
diffusivity coefficient for the mixture of hydrogen and air is an
order of magnitude higher than those for the mixture of air and
volatile HAP with readily available diffusivity coefficients.
17
-------
Therefore, hydrogen is more diffuse in air compared to organics
and more quickly enters the flammability range than organics.
5.2 Use of DuPont Test Results as the Basis for Hydrogen-Fueled
Flare Specifications
In selecting the conditions under which the pilot flare
testing was to be conducted and interpreting the results of the
testing, a "conservative" decision was made for each choice.
That is, the condition that would most likely assure that a full-
scale flare would achieve at least as high and possibly higher
destruction efficiency was chosen. This approach was applied to
the selection of flare tip design, flare tip diameter, pilot
burner heat input, and characteristics of the surrogate for HAP
for destruction testing. It was also applied to the evaluation
of stability testing and destruction efficiency results, as well
as the selection of operating limits applying to hydrogen
concentration and tip discharge velocity.
The Selection of the Flare Type
A nonassisted, plain-tip flare was used in the testing
program because all of DuPont's flares are nonassisted. A
nonassisted flare is a flare tip without any auxiliary provision
for enhancing the mixing of air into its flame. The plain-tip
means no tabs or other devices to redistribute flow were added to
the rim of the flare. Because the presence of tabs improves the
stability of the flare by channeling the flare's flow and
improving mixing of fuel and air, it was concluded that the lack
of tabs (i.e., plain tip) would result in the least stable test
conditions.
The Comparison of the Selected Flare with the Existing Flare
Specifications
A 3-inch flare was selected for the emission test since this
was the same size flare used for the testing to establish the
basis for the existing flare specifications in 40 CFR 60.18 and
40 CFR 63.11. Stability tests were conducted using propane to
18
-------
determine if the flare was operating properly and could meet the
existing flare specifications. Test results demonstrated that
this flare was stable when it was expected to be stable and not
stable when it was not expected to be (i.e., as indicated by the
existing flare specifications).
The Size of the Test Flare
Another reason for using the 3-inch flare for these tests is
because a 3-inch flare is small, relative to the size of flares
in industry (as a point of reference, the DuPont flares are 16 to
48 inches in diameter). Research indicates that smaller flares
are less stable than larger flares (Docket No. A-97-48, Item No.
II-I-l, Sec 4, page 6). Specifically, the physical parameter
known as the velocity gradient can be used to predict when a
flame will blow out by plotting the velocity gradient versus the
volume-percent hydrogen. The larger the boundary velocity
gradient, the more unstable the flame. Further, the velocity
gradient is inversely proportional to the diameter of the pipe.
Therefore, at a given velocity, the larger the pipe, the smaller
the boundary velocity, and the more stable the flame. The EPA
concludes that if a stable flame can be maintained with a smaller
flare pipe, then a larger flare would be expected to be stable at
lower hydrogen concentrations and higher velocities. Therefore,
the EPA believes that 3-inch or larger flares that meet these
specifications will have destruction efficiencies as high or
higher than those obtained from the 3-inch pipe flares.
The Selection of the Size of the Pilot Burner
The amount of heat input from the pilots on DuPont's full-
scale hydrogen-fueled flares are in the range from 0.05 to'0.6
percent of the total heat input to the flares. A venturi burner
turned down to approximately one third of its 9,000 Btu/hr
capacity was used for the tests described in this document, and
the heat input was equal to 0.3 to 0.6 percent of the pilot
flare's total heat input during the stability and destruction
efficiency tests. Therefore, the heat input from the pilot
during the tests was comparable to the heat input for the full-
scale flares operated by DuPont.
19
-------
The relatively small proportion of heat input from the
venturi burner compared to the total heat input to the test flare
would not be expected to have a significant effect on either the
stability or destruction efficiency results, because this amount
of heat is insignificant compared to the flare's total heat
content. Also, the use of a pilot burner is consistent with
EPA's flare specification which requires that the pilot flame be
present at all times.
The Selection of Ethylene as the Surrogate for HAP to be Used in
the Testing
For this study, a surrogate for HAP that was more difficult
to destroy than the volatile HAP present in the large scale flare
waste streams, and which could be measured at a concentration of
10 parts per billion by volume and higher was selected. In
general, the difficulty of destruction for organics increases as
the molecular weight decreases, but the limit of detection
decreases as the molecular weight decreases.
In order to compare the relative difficulty to destroy
various species, a linear multiple regression model was used that
calculates a destruction temperature using parameters describing
the molecular structure, autoignition temperature, and residence
time as inputs to the model. The destruction temperatures
obtained are theoretical temperatures for plug flow reactors to
achieve specified destruction allowing a comparison to be made
among various chemical species to estimate relative
destructibility (Docket No. A-97-48, Item No. II-I-14). As a
first step, the destruction temperatures were calculated for all
the chemical species that were identified in DuPont's full-scale
flare waste streams. The next step was to calculate destruction
temperatures for the surrogates for HAP under consideration.
(The results from this analysis are presented in Tables 4-3 and
Table 4-4 of Docket Item II-I-14).
In comparing the model's destruction temperature estimates
for candidate surrogates for HAP present in DuPont's flare
streams, the best choice as a surrogate was methane, but the
detection limit for methane was too high to be accepted for the
20
-------
field study. The next choice was methanol, but not only is the
detection limit high for methanol, it is a HAP and is a liquid at
ambient temperatures, presenting handling difficulties. The next
candidate considered was ethylene which was selected for the
study. Ethylene has an equivalent or higher destruction
temperature than all the organic HAP in the study, except
methanol, and has an acceptable limit of detection. Therefore,
the substance that was the most difficult to destroy but feasible
to use was chosen for the study.
The Criteria for a Stable Flame
The hydrogen content reported when lift off was first
observed was selected as the criterion for a stable flame,
because it was easy and precise to identify. The EPA concluded
that this was a conservative estimate for the stability limit
because destruction efficiencies greater than 98 percent were
noted even for hydrogen contents below the lift off level.
The EPA also concluded that lift off was a conservative
criterion for a stable flame, based on a correlation between the
stability ratio and the destruction efficiency observed in
earlier flare testing conducted in the 1980's (Docket No. A-97-
48, Item No. II-I-5). At that time the destruction efficiencies
were demonstrated to be directly proportional to the ratio of the
flare gas heating value to the minimum heating value for flame
stability (i.e., stability ratio). Regardless of the substance
being combusted, or the flare design, it was observed that the
destruction efficiency plateaued to greater than 98 percent
destruction when the stability ratio was above approximately 1.2.
For this test program, the destruction efficiency versus the
ratio of actual hydrogen to hydrogen at lift off (analogous with
the stability ratio, and referred to as the hydrogen ratio) was
plotted for this test program. The curve of the data was similar
to those obtained from the flare test programs in the 1980's.
Three data points demonstrated that at stability ratios below
1.0, with the lowest stability ratio of 0.955, destruction
efficiencies greater than 98 percent were achieved. Since these
hydrogen-fueled flare specifications require a stability ratio of
21
-------
1.0 or greater, it is assumed that a 98 percent or greater
destruction efficiency will be achieved.
The Operating Parameters Used for Testing the Destruction
Efficiency (i.e.. Hydrogen Content and Flare Tip Velocity)
The destruction efficiency of ethylene for the hydrogen-
fueled flares was tested at high tip velocities (i.e.,
approximately 100 to 120 ft/sec) because this is the velocity
range expected to produce lower destruction efficiencies.
Therefore, if acceptable destruction efficiencies are observed at
high tip velocities, then at least as high or even higher
destruction efficiencies are expected at lower tip velocities.
The expectation to observe decreased destruction efficiency
at high tip velocities is explained by two phenomena. The first
phenomenon is due to the increased fuel flow. The increased
volume of fuel flow entrains more air, and more eddies are formed
at the boundary between the fuel and the air. These eddies tend
to strip off some of the gases' flow, even before the flame is
able to combust the substances, so uncombusted or incompletely
combusted substances may be lost to the ambient air.
Another phenomenon explaining the expectation of decreased
destruction efficiency at increased tip velocities results from
comparisons of stability ratios at different tip velocities. For
this test program the ratio of the hydrogen content at lift off
to the hydrogen content at blow out with a pilot was used as an
analogous ratio to the previously mentioned stability ratio.
Further, the value of hydrogen at blow out was used as the
minimum hydrogen content, since at essentially this level of
hydrogen, the destruction efficiencies were above 98 percent for
tip velocities of 100 and 120 ft/sec. The DuPont test program's
data revealed a trend where the hydrogen ratios were lower at
higher velocities compared to lower tip velocities, 1.15 to 1.17
versus 1.3, respectively. Since the test programs in the 1980's
demonstrated that the destruction efficiency is directly
proportional to the stability ratio, then it could be expected
that the same or higher destruction efficiencies would be
22
-------
experienced at lower tip velocities where the hydrogen ratios are
larger.
5.3 Selection of the Specifications for Hydrogen-Fueled Flares
The recommended specification for hydrogen-fueled flares is
the maximum tip velocity for a given hydrogen content (determined
according to the equation of the line fitting the data from the
stability testing at lift off conditions as seen in Figure 1).
The equation in the recommended specifications comes directly
from the test report. This equation is Equation 2, with the units
changed to metric.
There are safety requirements that must be carefully
considered for all flare installations, and this is the case for
the use of these hydrogen-fueled flare specifications. As an
example, if the discharge velocity is too low under certain
conditions, the flame could propagate back into the process with
potentially catastrophic results. These recommended
specifications only specify a maximum discharge velocity for the
purpose of assuring efficient destruction of pollutants in waste
streams and do not address any aspect of safe operation. The
user of any EPA flare specifications should carefully consider
all features of this application, not just the limitation on
maximum discharge velocity, and implement all necessary measures
to assure a safe operation. Safe operating conditions are always
the responsibility of the owner/operator at each facility to
assure that all applicable safety requirements are adhered to
whether they are company, consensus and/or governmental
requirements.
The EPA did not think that extrapolating the data outside
the range of values tested to be prudent; therefore, the
hydrogen-fueled flare specifications have been restricted to the
confines of the conditions used for the test program. The
following restrictions are included in the hydrogen-fueled flare
specifications:
23
-------
Nonassisted Flares
The recommended hydrogen-fueled flare specifications are
applicable to nonassisted flares, because it was the only type of
flare tested.
Continuous Flame
The existing flare specifications require the presence of a
continuous flame where reliable ignition is obtained by
continuous pilot burners designed for stability. To ensure that
the pilot is continuously lit, a flame detection device is
required. These recommended hydrogen-fueled flare specifications
incorporate the same requirements for the same reason, to ensure
flame stability.
Minimum Flare Diameter
The testing was conducted on 3-inch flares, therefore this
is the minimum flare diameter for the recommended hydrogen-fueled
flare specifications.
Minimum Hydrogen Content
The minimum hydrogen content in the gas streams tested was
rounded to the nearest whole number, 8.0 volume percent, and set
as the defining minimum hydrogen concentration cutoff for a
hydrogen-fueled flare.
Maximum Tip Velocity
The maximum tip velocity was set at 37.2 m/sec (122 ft/s),
because that was the highest tip velocity tested.
Flame Stabilizers
Flame stabilizers (often called flame holders) are allowed
because stability and destruction efficiency testing was
conducted without them, so if these tabs stabilize the flame even
better mixing, and potentially greater destruction efficiencies
can be achieved.
Minimum Flare Tip Velocity
A minimum flare tip velocity was not listed since evidence
indicates that performance will not be diminished due to lower
tip velocities (See the preceding discussion concerning safety
responsibilities) .
24
-------
6.0 SUMMARY OF ENVIRONMENTAL, ENERGY, AND COST IMPACTS
The impacts discussed in this section are only for six
DuPont flares that are required by current or pending EPA
regulations to meet the existing flare specifications. The EPA
does not have information, and cannot estimate impacts for other
hydrogen-fueled flares in the United States. Therefore, the
following estimates are limited to these six DuPont flares.
6.1 Primary Air Impacts
The recommended flare specifications will reduce emissions
by the same amount (i.e., 98 percent or greater) as emissions
would be reduced by using flares meeting the existing flare
specifications.
6.2 Other Environmental Impacts
The Agency estimates that the recommended hydrogen-fueled
flare specifications will reduce secondary emissions of
pollutants since the combustion of supplemental organic fuel will
no longer be required; therefore, there will be no emissions
resulting from the combustion of a supplemental fuel. It is
estimated that these recommended hydrogen-fueled flare
specifications will reduce annual emissions from the six affected
DuPont flares by 147 megagrams (161 tons per year) of criteria
pollutants (i.e., 124 megagrams (136 tons per year) of carbon
monoxide, and 22.7 megagrams (25 tons per year) of nitrogen
oxides) and 39,900 megagrams (44,000 tons per year) of carbon
dioxide.
In addition to these secondary emission reductions, there
may also be State regulations that require owners/operators to
follow the existing flare specifications, and by allowing the
owners/operators to meet these recommended hydrogen-fueled flare
specifications, there may be further reductions in secondary air
25
-------
emissions. Therefore, these impacts are a minimal estimate of
the potential secondary air emission reductions.
6.3 Energy Impacts
These recommended hydrogen-fueled flare specifications are
expected to decrease the amount of energy used by DuPont's six
hydrogen-fueled flares since the flares will no longer be
required to combust secondary fuel. The expected energy savings
is estimated to be 7.75 X 108 cubic feet of natural gas annually
(7.75 X 1011 Btu/yr) .
6.4 Cost and Economic Impacts
Cost savings will be realized due to the recommended
hydrogen-fueled flare specifications by not requiring the
combustion of supplemental fuel (to comply with the original heat
content requirements), and by not requiring the subsequent
resizing of the existing flares that would result from a
requirement to combust supplemental fuel in order to accommodate
the additional flow of supplemental fuel. The cost of natural
gas as supplemental fuel for the six affected flares is estimated
to be $2.8 million per year. The capital investment to replace a
smaller flare tip with a larger one is estimated to be
approximately $667,000 per flare or $4 million for all six
flares. The total annual savings achieved by allowing hydrogen-
fueled flares that fulfill the recommended specifications are the
sum of the annual fuel cost savings, and the annualization of the
capital savings (calculated to be $280,000 per year). Therefore,
total annual savings for the six affected DuPont flares are
estimated to be $3.08 million per year. Since sources using
these hydrogen-fueled flare specifications will experience
savings, no adverse economic impacts will result from the
recommended hydrogen-fueled flare specifications.
26
-------
6.5 Summary of Impacts
This section discussed the cost savings, emission reduction
of secondary pollutants, and energy savings from only the six
DuPont flares subject to current or pending regulations. The
recommended hydrogen-fueled flare specification have greater
potential to reduce emissions and save money and fuel from
hydrogen-fueled flares that the EPA is currently aware.
7 . 0 SUMMARY
The purpose of this report is to describe the events leading
up to the development of alternative flare specifications for
hydrogen fueled flares, and to illustrate how test data supplied
by DuPont were used to develop the recommended specifications.
The report opens with a description of the existing flare
specifications followed by a summary of the studies used to
establish the criteria for the hydrogen-fueled flare
specifications. The basic objectives of these sections were to
provide the parameters for the existing rules and to establish
that the existing flare specifications are based on the theory
that HAP destruction can be equated with flame stability.
This report provides the basis for the alternative flare
specifications which the EPA is recommending based on a request
made by DuPont. To support their petition, DuPont provided
experimental data as well as an account from a previous studies.
Additionally, this report provides a descriptive summary of the
data provided from DuPont's tests and an analysis of each
parameter utilized in the testing. The EPA believes that the
results of the DuPont hydrogen-fueled tests provide the
appropriate data to support the hydrogen-fueled flare
specifications.
In closing, this report provides a summary of the
recommended hydrogen-fueled flare specifications and an
27
-------
explanation of the rationale used to establish these recommended
specifications. The EPA believes that flares meeting the
recommended hydrogen-fueled flare specifications being added will
achieve a destruction efficiency of at least 98 percent, and will
result in cost savings to those industries that flare waste gases
containing a hydrogen content of at least 8.0 percent.
28
-------
APPENDIX
DOCKET INDEX FOR SUBCATEGORY II-I OF DOCKET A-97-48
-------
Docket: A-97-48
Subcategory n-I
References Not Readily Available to the Public Elsewhere
Document
Number
Date Rcvd
in Docket
Commentor. Addressee. Title
or Description, etc.
Date of
Document
n-i-i
12-16-97
Hydrogen Flare Equivalency Demonstration - Final
Report: Prepared by Energy and Environmental
Research Corporation for W.K. Whitcraft of E.I. du
Pont de Nemours & Company. Wilmington, DE.
October 14, 1997.
10-14-97
H-I-2
2-26-98
Demonstration of the Equivalency of Hydrogen Flares -
Task 1 Final Report. Prepared by Energy and
Environmental Research Corporation for W.K.
Whitcraft of E.I. du Pont de Nemours & Company.
Wilmington, DE. September 28, 1993.
9-28-93
H-I-3
2-26-98
Evaluation of the Efficiency of Industrial Flares:
Background- Experimental Design Facility. Prepared
by Joseph, D., J. Lee, C. McKinnon, R. Payne, and H.
Pohl for the U. S. EPA. Research Triangle Park, NC.
Report No. 600/2-83-070, NTIS No. PB83-263723.
August 1983.
8-83
H-I-4
2-26-98
Evaluation of the Efficiency of Industrial Flares: Test
Results. Prepared by Pohl, J. H., R. Payne, and J. Lee
for the U. S. EPA. Research Triangle Park, NC.
Report No. 600/2-84-095, NTIS No. PB84-199371.
May 1984.
5-84
H-I-5
2-26-98
Evaluation of the Efficiency of Industrial Flares: Flare
Head Design and Gas Composition. Prepared by Pohl,
J. H., and H.R. Soelberg for the U. S. EPA. Research
Triangle Park, NC. Report No. 600/2-85-106, NTIS
No. PB86-100559/AS. September 1985.
9-85
H-I-6
2-26-98
Hydrogen Flare Stack Diffusion Flames: Low and High
Flow Instabilities, Burning Rates, Dilution Limits,
Temperatures, and Wind Effects, Report of
Investigations 7457. Prepared by Grumer, J., A.
Stasser, J. M. Gussey, and V.R. Rowe for the U.S.
Department of the Interior, Bureau of Mines. 1970
1970
10
-------
Docket: A-97-48
Subcategory n-I
References Not Readily Available to the Public Elsewhere
Document
Number
Date Rcvd
in Docket
Commentor. Addressee. Title
or Description, etc.
Date of
Document
H-I-7
2-26-98
Observations on Waste Destruction in Liquid Injection
Incinerators. Prepared by Kramlich, J.C. of Energy
and Environmental Research Corporation. Irvine, CA.
Revised February, 1990.
2-90
H-I-8
2-26-98
Evaluation of the Efficiency of Industrial Flares: H2S
Gas Mixtures and Pilot Assisted Flares. Prepared by
Pohl, J. H., Soelberg, E. for the U. S. EPA. Research
Triangle Park, NC. Contract No. 68-02-3661. July
1986.
7-86
H-I-9
2-26-98
An Experimental Analysis of Flame Stability of Open
Air Diffusion Flames. Prepared by Noble, R.K., M.R.
Keller, and R. E. Schwartz for the American Flames
Research Committee. Tulsa, OK. Presented at the
International Symposium on Alternative Fuels and
Hazardous Wastes. OK. October 9-11, 1984.
10-9-84
H-I-10
2-26-98
Evaluation of the Efficiency of Industrial Flares:
Influence of Gas Composition. Prepared by Soelberg,
N.R. for the U.S. EPA. Research Triangle Park, NC.
Contract No. 68-02-3995. 1987.
1987
H-I-11
2-26-98
Combustion Science for Incineration Technology. lost
O. L. Wendt. Twenty-Fifth Symposium (International)
on Combustion/The Combustion Institute, 19947
pp. 277-289.
1994
n-I-12
2-26-98
Flare Efficiency Study. Prepared by Marc McDaniel of
Engineering-Science, Incorporated for the U.S. EPA.
Office of Research and Development, Washington,
DC. Report No. 600/2-83-052. July 1983.
1985
n-I-13
2-26-98
OAQPS Control Cost Manual - January 1990, Fourth
Edition, Chapter 7. EPA 450/3-90-006
1-90
11
-------
Docket: A-97-48
Subcategory n-I
References Not Readily Available to the Public Elsewhere
Document
Number
H-I- 14
Date Rcvd
in Docket
2-26-98
Commentor. Addressee. Title
or Description, etc.
Predictive Model of the Time-Temperature
Requirements for Thermal Destruction of Dilute
Organic Vapors. Prepared by Lee, K., J.L. Hansen, and
D.C. Macauley of the Union Carbide Corporation.
South Charleston, West Virginia.
Date of
Document
12
-------
TECHNICAL REPORT DATA
(Please read Instructions on reverse before completing)
1. REPORT NO. 2.
EPA-453/R-98-001
4. TITLE AND SUBTITLE
Basis and Purpose Document for Hydrogen-Fueled Flares
7. AUTHOR(S)
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Emission Standards Division (Mail Drop 13)
Office of Air Quality Planning and Standards
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
12. SPONSORING AGENCY NAME AND ADDRESS
Director
Office of Air Quality Planning and Standards
Office of Air and Radiation
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
3. RECIPIENT'S ACCESSION NO.
5. REPORT DATE
March 1998
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-D6-0008
13. TYPE OF REPORT AND
14. SPONSORING AGENCY
EPA/200/04
PERIOD COVERED
CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
This document provides the background technical information related to EPA's amendments to the
control device requirements of 40 CFR 60. 18 and 40 CFR 63. 1 1, permitting the use of hydrogen- fueled
flares. Previous hydrogen-fueled flare test results are summarized; and DuPont's test program and data
are discussed.
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS b. IDENTIFIERS/OPEN ENDED TERMS
Air Pollution Hazardous air pollutants
Hazardous air pollutants
Emission reduction
Control devices
18. DISTRIBUTION STATEMENT 19. SECURITY CLASS (Report)
Unclassified
Release Unlimited 20 SECURITY CLASS (page>
Unclassified
c. CASEATE Field/Group
21. NO. OF PAGES
37
22. PRICE
EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDITION IS OBSOLETE
-------
U.S. Environmental Protection Agency
Region 5, Library (PL-12J)
77 West Jackson Boulevard, 12th Floor
Chicago, It 60604-3590
------- |