United States
Environmental Protection
Agency
Air and Energy Engineering
Research Laboratory
Research Triangle Park NC 27711
Research and Development
EPA/600/S2-85/106 Jan. 1986
&ER& Project Summary
Evaluation of the Efficiency of
Industrial Flares: Flare Head
Design and Gas Composition
J.H. Pohl and N. R. Soelberg
This report documents a continua-
tion of Phase 4 of a research program to
quantify emissions from, and efficien-
cies of, industrial flares. Phases 1 (Ex-
perimental Design) and 2 (Design of
Test Facilities) were reported in EPA-
600/2-83-070 (NTIS No. PB83-263723).
Phase 3 (Development of Test Facilities)
and initial work in Phase 4 (Data Collec-
tion) were reported in EPA-600/2-84-095
(NTIS No. PB84-199371). Further data
collection during Phase 4 is reported
here.
Initial results were limited to tests
conducted burning propane/nitrogen
mixtures in pipe flares without pilot
light stabilization. The work reported
here extends the previous results to
other flare head designs and other
gases and includes a limited investiga-
tion of the influence of pilot flames on
flare performance. The following re-
sults were obtained:
• Flare head design influences the
flame stability curve.
• Combustion efficiency can be cor-
related with flame stability for
pressure heads and coanda steam
injection heads.
• For the limited conditions tested,
flame stability and combustion ef-
ficiency of air-assisted heads corre-
lated with the momentum ratio of
air to fuel; the heating value of the
gas had only minor influence.
• Limited data on an air-assisted
flare show that a pilot light im-
proves flame stability.
• The destruction efficiency of com-
pounds depends on the structure
of the compounds.
• For the compounds tested in this
program, the destruction efficiency
of different compounds could be
correlated with the flame stability
curve for each.
This Project Summary was devel-
oped by EPA's Air and Energy Engineer-
ing Research Laboratory, Research Tri-
angle Park, NC, to announce key
findings of the research project that is
fully documented in a separate report
of the same title (see Project Report or-
dering information at back).
Introduction
Industrial flares are commonly used
to safely and economically destroy
large amounts of industrial waste
gases. Since most of the gas flared in
the U.S. is from leaks, purges, and
emergency vents, the amounts and
compositions of flared gases vary
widely and are difficult to measure.
Flare emissions are also difficult to mea-
sure. Most flares are elevated to de-
crease noise and radiation hazards and
to increase dispersion of combustion
products. Probe collection of plume ma-
terial in such situations is impractical.
Remote sensing of flare emissions is an
alternative to direct sampling, but in-
strumentation and techniques for this
purpose are still undeveloped.
To evaluate and control industrial
flare emissions, pilot-scale research is
necessary to obtain direct sampling of
flare emissions. Flare research has been
conducted at Energy and Environmen-
tal Research Corporation (EER) since
1980. A pilot-scale flare test facility was
constructed for the U.S. EPA in 1982.
-------
Previous results (reported in EPA-600/
2-83-070 and -84-095) showed that flare
combustion efficiencies are generally
high (i.e., exceeding 98%), but under
some operating conditions (e.g., excess
steam injection) efficiencies can be low.
The results also showed that, when a
flare flame is stable (i.e., not near blow-
off conditions), combustion is efficient.
However, flares operating with unstable
flames tend to be inefficient. Data on
flare gas exit velocity were correlated
with the gas heating value to describe
the region of flame instability. Thus, for
the conditions tested, the operating
range required for efficient flare per-
formance can be defined.
Caution should be used, however, in
applying these results to situations not
tested. Flare flame stability and com-
bustion efficiency may vary, depending
on the flare head type, gas composition,
and operating conditions. Thus, the re-
search was extended to evaluate the ef-
fects of: (1) flare head type on flare com-
bustion efficiency, and (2) relief gas
composition On flare combustion and
destruction efficiency.
Approach
The program was divided into four
major tasks:
• Task 1—Evaluation of combustion
efficiency for different flare head
types.
• Task 2—Identification of represen-
tative, potentially difficult-to-
destroy compounds.
• Task 3—Evaluation of combustion
and destruction efficiency of se-
lected relief gas mixtures.
• Task 4—Data analysis and report-
ing.
For Task 1, four commercial flare
heads were evaluated: an air-assisted
head, two pressure heads, and a coanda
steam injection head.
Each head was tested on the EER
pilot-scale Flare Test Facility (FTF).
Flame stability and combustion effi-
ciency were measured as functions of:
(1) relief gas and exit velocity, (2) relief
gas heating value, (3) steam assist flow
rate (for the coanda head), (4) air assist
velocity (for the air-assisted head),
(5) relief gas pressure (for the pressure
head), and (6) with and without pilot
flame (for the air-assisted head).
The relief gas for these tests was
propane, mixed with nitrogen to vary
the heating value. Natural gas was used
for the pilot flame.
Tasks 2 and 3 were designed to mea-
sure effects of flare gas composition on
flame pollutant emissions. A wide vari-
ety of industrial compounds are fre-
quently flared in the U.S. Most often,
they are flared in mixtures containing
several different compounds. Each dif-
ferent mixture may exhibit somewhat
different flaring characteristics. Pilot or
large-scale testing of every conceivable
relief gas mixture would be expensive
and unending.
Task 2 involved the testing of com-
pounds in a laboratory facility. Al-
though laboratory-scale flare flames are
aerodynamically unlike pilot or large-
scale flare flames, laboratory-scale tests
can be used to economically and swiftly
screen compounds to determine com-
parative potential for destruction in
flares. Compounds which demonstrate
flaring difficulties in the laboratory-
scale Flare Screening Facility (FSF) are
candidates for testing on the FTF.
Twenty-one compounds were se-
lected for laboratory-scale testing in the
FSF, representing the following classes:
sulfur compounds, nitrogen com-
pounds, chlorinated compounds, oxy-
genated compounds, aliphatic hydro-
carbons, aromatic hydrocarbons, and
compounds with low heating value. Of
the 21 compounds screened, 6 were se-
lected as candidates for testing on the
FTF. Selection criteria included low de-
struction efficiency, poor ignitability,
and high soot production.
Three of the six compounds, along
with hydrogen sulfide, were tested on
the FTF.
Hydrogen sulfide and ammonia were
tested in mixtures with propane and ni-
trogen. Ethylene oxide and 1, 3-
butadiene were tested diluted with ni-
trogen to vary the heating value. Flame
stability, combustion and destruction
efficiency, soot production, and
byproduct formation from incomplete
combustion were measured for each
compound. All tests were conducted
using the 3-in.* open pipe flame, with-
out pilot flame stabilization.
For Tasks 1 and 3 (conducted on the
FTF), samples were taken at five radial
positions above the flame. These local
samples were analyzed for 02, CO, C02,
hydrocarbon, NOX, and soot concentra-
tion. Where applicable, the samples
were also analyzed for H2S, S02, and
NH3 concentration.
Sampling in the FSF (Task 2) was eas-
ier. In this facility, the flare nozzle was
enclosed in a reaction chamber, which
(*) To convert to metric units, please use the equiv-
alents at the end of this Summary.
isolated the flame from the external en-
vironment. Sampling of the well-mixed
products at the reactor outlet required
only one probe.
Samples were analyzed during tests
on both the FSF and FTF to evaluate air
dilution, mass balances, combustion ef-
ficiency, and destruction efficiency.
SO2/ injected during some of the pilot
tests, was used as a tracer for mass bal-
ances. Mass balances on the FTF were
more difficult because of product loss,
air dilution in the large exposed flame,
and plume concentration gradients. Lo-
cal mass balances were used to accu-
rately evaluate local mass fluxes, local
combustion efficiency, and destruction
efficiency. Local mass fluxes were radi-
ally integrated to calculate overall com-
bustion and destruction efficiencies.
Results
Flare Head Design
The data show that flare head design
influences the flame stability curve (as a
function of gas heating value) as shown
in Figure 1 for the coanda steam injec-
tion head and the pressure heads. The
flame stability of the air-assisted head
was controlled by the ratio of air to fuel
momentum as shown in Figure 2. The
heating value of the gas had little influ-
ence on flame stability for the air-
assisted flare. The combustion effi-
ciency of the pressure and coanda
steam injection heads correlated with
the gas heating value needed to main-
tain flame stability, as shown in Figure
3. For the air-assisted flare, the air to
fuel momentum ratio was used to de-
velop a correlation with combustion ef-
ficiency. Figure 4 shows this relation-
ship, but caution should be used in
applying these data due to the limited
number of observations.
Gas Composition
The relative destruction efficiency of
different gases was determined in the
FSF. Table 1 gives results of these tests.
Six compounds were identified as po-
tentially difficult to destroy:
• 1, 3-butadiene yielded large
amounts of soot.
• Carbon monoxide was difficult to
ignite when pure.
• Ethylene oxide yielded low destruc-
tion efficiency.
• Vinyl chloride yielded low destruc-
tion efficiency.
• Hydrogen cyanide yielded low de-
struction efficiency.
-------
• Ammonia was difficult to ignite
when pure.
The destruction efficiency (DE) of three
of these compounds and H2S was mea-
sured on the FTP. The flame stability
curve depended on the compound as
shown in Figure 5. (H2S and NH3 were
tested as minor constituents in
propane/nitrogen mixtures.) The DE of
the individual compounds depended on
compound type but correlated with the
stability curve for each compound as
shown in Figure 6.
Conclusions
• Flare head design influences the
flame stability curve.
• Combustion efficiency can be cor-
related with flame stability for pres-
sure heads and coanda steam in-
jection heads.
• For the limited conditions tested,
flame stability and combustion effi-
ciency of air-assisted heads corre-
lated with the momentum ratio of
air to fuel; the heating value of the
gas had only minor influence.
• Limited data on an air-assisted flare
show that a pilot light improves
flame stability.
• The destruction efficiency of com-
pounds depends on the structure of
the compounds.
• For the compounds tested in this
program, the destruction efficiency
of different compounds could be
correlated with the flame stability
curve for each.
Conversion Factors
To convert nonmetric units used in
this Summary to their metric equiva-
lents, please use the following factors:
Nonmetric Multiplied by Yields metric
Btu
ft
ft3
in.
1.055
0.305
0.028
2.54
kJ
m
m3
cm
1200
1100
1000
I
I
<5
-------
1000
100 -
I
I
0.1
8
Without Pilot
O 180-400 Btu/ft3
D 401-900
O 901-1400
A 1401-2350
With Pilot
€ 209-400 Btu/ft3
\
0.2 0.4 0.6 0.8 1.0
Air-Assist to Gas Momentum Ratio
1.2
1.4
Figure 2.
Maximum gas exit velocity for stability versus air-assist to gas momentum ratio for
the air-assisted head G. with and without pilot.
-------
!
u
.4)
C
.o
1
1
100
98
96
QA
90
88,
86
84
I N 12 in. Coanda Steam-Injected
• u !./„„_/ n
I
HeadD
[j 1.5 in. Pressure-Assisted Head £
' Q 3.8 in. Pressure-Assisted Head F
I Q 3 in. Open Pipe Head with Pilot
1
i
Figure 3.
1 2 2.5
Gas Heating Value/Minimum Heating Value for Stability
Combustion efficiency vs. flame stability for steam-injected and pressure-assisted
flare heads.
-------
99.9 ,
Uj
O
u
a
i
1
o
O
99.0
90.0
.0.1
C* Without Pilot
£ With Pilot
S.R. = Air-Assist to Fuel
Stoichiometric Ratio
0.5
5.0
1.0
10.0
1
.c
.§>
I
w
I
o
(J
100.0
0.2 0.4 0.6 0.8
Air-Assist to Gas Momentum Ratio
1.0
Figure 4. Combustion efficiency vs. air-assist to gas momentum ratio for commercial air-
assisted head G.
-------
Table 1. Results of Compound Screening Tests (Gases in Propane/Nitrogen Mixtures)
Compound
Acetylene
Ethylene
Propylene
1, 3-Butadiene
Butane
Propane
Propane
Benzene
Toluene
Chlorobenzene
Carbon Monoxide
Carbon Monoxide
Carbon Monoxide
Acetone
Acetaldehyde
Ethylene Oxide
Carbon Dioxide
Methyl Chloride
Ethylene Dichloride
Vinyl Chloride
Methyl Mercaptan
Acrylonitrile
Hydrogen Cyanide
Ammonia
Ammonia
%
Compound
in Gas
100
100
100
100
100
100
25
1.50
1.50
1.15
100
30
NA
1.43
2.07
1.42
7.58
9.17
1.43
0.11
10.7
1.47
0.013
20
100
%
Nitrogen
in Gas
0
0
0
0
0
0
75
43
43
43
0
44
NA
43
43
43
43
42
43
44
40
43
44
37
0
Heating
Value
Btu/ft3
1475
1580
2300
2780
3321
2350
1763
2370
2381
2350
[C o u 1
1760
1111
2347
2331
2337
2171
2212
2335
2350
2218
2350
2350
1967
}C o u 1
DE"
%
99.99
99.91
99.98
99.93
99.99
99.98
99.97
99.59
99.99
99.49
d No
99.60
NA
99.80
99.99
IS6.92I
NA
99.94
99.70
I9g.79l
99.39
99.99
I85.00I
99.90
d N o
CE"
%
99.97
99.92
99.93
99.93
99.96
98.18
NA»
99.95
99.90
99.95
t 1 g
99.88
99.42
99.96
99.97
99.95
99.93
99.96
99.95
NA
99.82
99.96
NA
NA
t 1 g
Soot
mg/m3
<1.5
<1.5
<1.5
| 75C |
<1.5
<1.5
NA
<1.0
<1.0
<1.0
n i t e\
<1.0
<1.0
<1.0
<1.5
<1.0
<1.0
<1.0
<1.0
<1.0
<1.0
<1.0
<1.0
<1.0
n i t e\
aDE = Destruction Efficiency.
•"CE = Combustion Efficiency.
'Boxes indicate compounds with potential problems.
<
-------
59.39
99.9
2
CD
I
c
o
99.0
90.0
Ethylene Oxide
1,3-Butadiene
Propane
Q Ammonia, 1.5-4.5 Percent in Propane/Nitrogen
Q 1,3-Butadiene in Nitrogen
U Ethylene Oxide in Nitrogen
ff Propane in Ammonia Tests
W Propane in Hydrogen Sulfide Tests
0.0)
0.1
I
•Q
O
.5
s^
O
!
1.0
10.0
1
i
100.0
Figure 6.
1.0 2.0
Heating Value/Minimum Heating Value for Stability
Destruction efficiency of different gases.
J. Pohl and N. Soelberg are with Energy and Environmental Research Corp.,
Irvine. CA 92718.
Bruce A. Tichenor is the EPA Project Officer (see below).
The complete report, entitled "Evaluation of the Efficiency of Industrial Flares:
Flare Head Design and Gas Composition, "(Order No. PB86-100 559/AS; Cost:
$16.95, subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Air and Energy Engineering Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
. S. GOVERNMENT PRINTING OFFICE: 1986/646-116/20747
------- |