EPA -600/2-75-075
October 1375
Environmental Protection Technology Series
EFFECT OF FUEL SULFUR ON
NOX EMISSIONS FROM PREMIXED FLAMES
Industrial Environmeotal Research Laboraiu..
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development,
U.S. Environmental Protection Agency, have been grouped into
five series. These five broad categories were established to
facilitate further development and application of environmental
technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in
related fields. The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ENVIRONMENTAL PROTECTION
TECHNOLOGY series. This series describes research performed
to develop and demonstrate instrumentation, equipment and
methodology to repair or prevent environmental degradation from
point and non-point sources of pollution. This work provides the
new or improved technology required for the control and treatment
of pollution sources to meet environmental quality standards.
EPA REVIEW NOTICE
This report has been reviewed by the U. S. Environmental Protection
Agency, and approved for publication. Approval does not signify that
the contents necessarily reflect the views and policies of the Agency, nor
does mention of trade names or commercial products constitute endorse-
ment or recommendation for use.
This document is available to the public through the National
Technical Information Service, Springfield, Virginia 22161.
-------
EFFECT OF FUEL SULFUR
ON NOX EMISSIONS
FROM PREMIXED FLAMES
by
J.O. L. Wendt and J.M. Ekmann
University of Arizona
Department of Chemical Engineering
Tucson, Arizona 85721
Grant No. R-802204
ROAPNo. 21ADG-021
Program Element No. 1AB014
EPA Project Officer: W. S. Lanier
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
October 1975
-------
TABLE OF CONTENTS
Page
List of Figures iv
Acknowledgments v
PURPOSE AND SCOPE 1
CONCLUSIONS 2
RECOMMENDATIONS 4
BACKGROUND 5
EXPERIMENTAL APPARATUS 9
COMBUSTION RIG 9
SAMPLING AND ANALYSIS TRAIN 9
RESULTS 14
PREMIXED COMBUSTOR PERFORMANCE 14
PHASE I. EFFECT OF S02 ON THERMAL NOX EMISSIONS 17
PHASE I. EFFECT OF H2S ON THERMAL NOX EMISSIONS 21
PHASE II. EFFECT OF FUEL SULFUR ON FUEL NOx EMISSIONS .... 26
PHASE II. TRIAL 1 31
PHASE II. TRIAL 2 35
MATHEMATICAL MODELING 42
PREMIXED FLAT FLAME MODEL 43
KINETIC MECHANISM 44
CALIBRATION OF FLAT FLAME SIMULATION WITH
DATA OF PEETERS 46
CALIBRATION WITH BASE CASE NO MEASURED 48
EFFECT OF SO? IN FUEL 50
REFERENCES 59
PUBLICATIONS AND PRESENTATIONS RESULTING
FROM GRANT R-802204 61
APPENDICES
A. DESCRIPTION OF COMPUTER PROGRAM REKINET ... 63
B. SAMPLE DATA DECK 69
C. REACTION RATE LIBRARY 73
D. THERMOCHEMICAL LIBRARY 83
ill
-------
FIGURES
Page
FIGURE 1. SCHEMATIC OF APPARATUS 10
FIGURE 2. FLAT FLAME COMBUSTOR 11
FIGURE 3. SCHEMATIC OF ANALYSIS SYSTEM 12
FIGURE 4. THERMAL NO EXHAUST EMISSIONS - BASE CASE 15
FIGURE 5. EXHAUST LEVELS OF NO ACHIEVED RAPIDLY 16
FIGURE 6. S02 INHIBITS NO FORMATION AT ZERO PREHEAT 18
FIGURE 7. SO2 INHIBITS NO FORMATION AT HIGH PREHEAT 19
FIGURE 8. S02 DECREASES RATE OF NO FORMATION NEAR THE FLAME 22
FIGURE 9. EARLY-FORMED NO INHIBITED BY S02 23
FIGURE 10. H2S INHIBITS NO FORMATION AT ZERO PREHEAT. ...... 24
FIGURE 11. H2S INHIBITS NO FORMATION AT HIGH PREHEAT 25
FIGURE 12. EARLY-FORMED NO INHIBITED BY H2S 28
FIGURE 13. CONVERSION OF H2S TO S02 IS MORE RAPID THAN NO
FORMATION-FUEL RICH CONDITIONS 29
FIGURE 14. CONVERSION OF H2S TO S02 IS MORE RAPID THAN NO
FORMATION-FUEL LEAN CONDITIONS 30
FIGURE 15. EFFECT OF S02 ON FUEL NO UNDER FUEL
LEAN CONDITIONS 32
FIGURE 16. EFFECT OF SAMPLING RATE ON FUEL UNDER
FUEL RICH CONDITIONS 33
FIGURE 17. EFFECT OF S02 AND SAMPLING RATE ON NO PROFILES .... 34
FIGURE 18. EFFECT OF S02 ON NO, NOX, PROFILES (FUEL
NITROGEN = NO, 107% STOICHIOMETRIC AIR) 37
FIGURE 19. EFFECT OF S02 ON NO, NOX PROFILES (FUEL
NITROGEN = C2N2, 107% STOICHIOMETRIC AIR) 38
FIGURE 20. EFFECT OF S02, SAMPLING RATE ON NO, NOX, PROFILES
(FUEL NITROGEN = NO, 89% STOICHIOMETRIC AIR) 39
FIGURE 21. EFFECT OF S02, SAMPLING RATE ON NO, NOX PROFILES
(FUEL NITROGEN = C2N2, 89% STOICHIOMETRIC AIR) .... 40
FIGURE 22. S02 ADDITION AFFECTS BOTH OXYGEN ATOM AND
TEMPERATURE PROFILES, BUT LOWERS NO 55
FIGURE 23. UNDER ADIABATIC CONDITIONS S02 DELAYS NO FORMATION . . 57
IV
-------
ACKNOWLEDGEMENTS
The authors would like to acknowledge the help of David W.
Pershing, William K. Taylor and Joannes W. Lee who all contribu-
ted to the successful completion of this research. In addition,
thanks are due to Rhoda Miller and Sue Burnett for their help in
completing secretarial and bookkeeping duties associated with
this project. The help and advice of W. Steven Lanier, who was
the EPA Project Officer, and who contributed much in both the
technical aspects and in the smooth financial operation of this
project, is much appreciated.
-------
PURPOSE AND SCOPE
The objective of this work was to determine the conditions under
which fuel sulfur inhibits the formation of nitrogen oxides in
flames. The importance of this project lies in the need to de-
termine whether fuel desulfurization might have an adverse effect
on nitrogen oxide emissions.
The study consisted of three phases. In Phase I we examined
through controlled laboratory experiments, the effect of fuel
sulfur on nitrogen oxide formation by atmospheric fixation
(Thermal NO). In these experiments fuel sulfur was simulated by
addition of sulfur dioxide and hydrogen sulfide to a nitrogen free
gaseous fuel which was then mixed with air to burn in a premixed
laminar flat flame. The ensuing interactions between sulfur and
thermal NO formation mechanisms were then examined in some detail,
and conclusive results were obtained. In Phase II, we examined,
using the same apparatus, the effect of fuel sulfur on nitrogen
oxide formation by fuel nitrogen conversion (Fuel NO). Fuel
nitrogen compounds were simulated by addition of NO itself and of
cyanogen, C2N2. Results of this phase were not conclusive, and
should be regarded as preliminary. In a contiguous Phase III
effort we focused our attention of sulfur dioxide-thermal NO
interactions and developed a mathematical model that describes
quantitatively the effects measured experimentally in Phase I.
The model involves a computer simulation of a flat flame and is
described in detail. The model was developed so that observed
effects could be interpreted and explained in the light of
fundamental principles. The computer code developed was supplied
to EPA.
-------
CONCLUSIONS
It was found that both sulfur dioxide and hydrogen sulfide, when
added to a gaseous fuel, had a significant inhibition effect on
thermal NOx emissions. The presence of sulfur to make about 6800
ppm S02 in the exhaust lowered NOX emissions by up to 36%. Al-
though it should be noted that these results are valid for pre-
mixed gaseous flames, they do imply that fuel desulfurization may
lead to increased (thermal) NOx emissions from combustion
processes.
It appears to make very little difference whether the fuel sulfur
is introduced into the fuel as S02 or H2S, except under fuel rich
conditions where conversion of H2S to S02 is not rapid. This
indicates that the inhibition of NO formation by fuel sulfur
occurs through mechanisms involving S02.
The data on the effect of fuel sulfur on fuel NO emissions are
inconclusive, due to previously unreported phenomena occurring in
quartz sampling probes under fuel rich conditions. However, pre-
liminary indications are that although the effect of fuel sulfur
on fuel NO emissions is not significant under fuel lean conditions,
it may under fuel rich conditions have a marked influence on the
rate at which fuel NO is formed.
A preliminary analytical model of a premixed flat flame showed
that the inhibition effect of fuel sulfur on "Thermal NO" could
be explained by the homogeneous catalysis of free radical recom-
bination rates by sulfur dioxide. This mechanism lowers the
oxygen atom concentration, when this concentration is above the
-------
equilibrium value, and, for a given radiative heat loss, lowers
NO formation rates. The effect of lower oxygen atom concentra-
tions is greater than the ensuing (coupled) temperature increase.
Under conditions where the temperature was fixed as an independent
variable the theoretically predicted inhibitory effect of S02 on
thermal NOx emissions was even larger. Observations involving
inhibition of NO formation by S02 might therefore be used to
arrive at conclusions concerning the role of superequilibrium
oxygen atoms. This is true even when the primary NO formation
mechanism does not involve oxygen atoms directly, since the
concentrations of other important atoms and free radicals are in-
timately related and coupled to that of the oxygen atom. Since
inhibition was observed under fuel rich and fuel lean conditions
it appears that superequilibrium oxygen atoms and other free rad-
icals play an important role in both regimes. The effect of S02
was especially pronounced on "prompt" NO formation, and this sup-
ports theories that superequilibrium concentrations of oxygen
atoms and other radicals are a factor in the rapid formation of
NO early in the flame. In addition, the experimental results on
the thermal NO showed that "prompt" NO accounted for essentially
all the NOx emission under fuel rich conditions and that it was
not a strong function of.mixture preheat. This implies that NOx
formation mechanisms other than those of Zeldovich are controlling
early in the flame.
-------
RECOMMENDATIONS
Future work should concentrate on three areas in order to iden-
tify the practical aspects of the results of the research re-
ported here. First, the effect of fuel sulfur on NOx emissions
from oil and coal diffusion flames should be investigated, in
order to determine if fuel desulfurization in general will have
an adverse effect on NOx emissions from combustion units of
practical interest. Second, the more fundamental aspects of
fuel sulfur and fuel nitrogen interactions during the combustion
process should be examined further, since an understanding of
these phenomena will aid in the interpretation of new data, and
in the identification of future potential environmental problems
associated with fuel desulfurization. Third, theoretical tools
should be developed in order to allow effects observed from
laboratory scale premixed flame experiments to be extrapolated
with some confidence to oil and coal diffusion flames in practi-
cal combustion units. Laboratory experiments, such as are re-
ported here are relatively fast and inexpensive, and it would be
useful to be able to deduce the correct practical implications
of observed phenomena without having to resort to expensive and
difficult full scale tests.
-------
BACKGROUND
The combustion of many fossil fuels gives rise to emissions of
both sulfur oxides and nitrogen oxides. Sulfur oxide pollution
may be abated by either fuel desulfurization or stack gas scrubb-
ing. The choice of abatement method is usually dictated by
economic considerations. However, in calculating the cost effec-
tiveness of various sulfur oxide abatement strategies it is im-
portant to determine the extent to which the technology used has
an adverse effect on other pollutants, such as NOX. Should sul-
fur compounds in the flame front have an inhibition influence on
the formation of NOx this would indicate that fuel desulfuriza-
tion might require additional NOX abatement methods to be imple-
mented, and that this would involve additional costs which would
not occur with stack gas scrubbing where the sulfur species are
removed after the combustion process.
That sulfur and nitrogen oxides interact at low temperatures is
not newt1'2). This interaction results in the catalysis, by NO,
of the oxidation of SOa to S03. At higher temperatures, under
combustion conditions, the situation is quite different and a
clear distinction should be drawn between low temperature and
high temperature interactions. At high temperatures under com-
bustion conditions, free radicals are produced in superequilib-
rium amounts and this fact has been shown to be important in
explaining high NO production rates (3'"*) where the NO is formed
by atmospheric fixation (Thermal NO). Since it has been
shown(5»6) that sulfur dioxide is an effective catalyst in re-
ducing superequilibrium free radical concentrations, it is
reasonable to expect that sulfur dioxide and possibly other fuel
-------
sulfur compounds, inhibit the formation of NO in flames. In
order to explore this possibility further, it is necessary to
focus on certain fundamental aspects of NO formation mechanisms.
In spite of much research in this area, there is even now no
general agreement on the kinetic mechanisms of thermal NO pro-
duction. Most widely recognized as being important is the
mechanism proposed by Zeldovich^7':
N2 -I- 0 -»• NO + N (1)
N + 02 -»• NO + 0 (2)
with the modification
N + OH + NO + H (3)
and with the free radicals necessary for these reactions being
produced through the combustion process. It should be noted that
the free radicals so produced can have concentrations many fold
in excess of those determined by equilibrium and that the decay
of these radicals towards equilibrium is relatively slow and
occurs downstream from the flame front. Under fuel lean con-
ditions, it is in this region of free radical decay that a sub-
stantial portion of the NO is formed through reactions (1)
through (3). Thompson and Beer(3) have shown that indeed,
superequilibrium concentrations of oxygen atoms are responsible
for high rates of NO formed in their apparatus and their con-
clusion was corroborated by other workers,(4'8) especially as
regards NO formation in the fuel lean regime. In the fuel rich
regime, however, it appears that for hydrocarbon flames, an NO
formation mechanism involving cyanide compounds as intermediates
may be applicable'1*'9), and under these conditions the role of
superequilibrium atom and radical concentrations is unclear. It
is generally recognized, however, that high rates of NO forma-
tion can result from superequilibrium atom concentrations, and
it would therefore appear that catalysts and other impurities,
6
-------
such as SO2, that have been shown to decrease radical concentra-
tions, should tend to lower thermal NO formation rates.
In order to determine the effect of fuel sulfur on thermal NO it
is necessary to devise a well defined laboratory experiment to
answer the following questions:
• Does SOa have an inhibitory effect on thermal NO
emissions?
• Under which conditions is any inhibtion of NO
formation by S02 most significant?
• Does SOa affect the formation of "prompt NO" and
if so what conclusions can be drawn about the role
of superequilibrium atom concentrations and "prompt
NO"?
• With H2S in the fuel, is conversion of the fuel sul-
fur to SOz sufficiently rapid to allow the S02
formed to have the same effect as when added
directly to the fuel?
Literature on flame interactions between species derived from
fuel sulfur and fuel nitrogen in the flame front is quite meager.
Yet the problem of fuel sulfur effects on fuel NO emissions is of
substantial practical interest since most fuels that contain fuel
sulfur contain appreciable amounts of chemically bound nitrogen.
Desulfurization of a fuel does not necessarily lead to a propor-
tional decrease in the fuel nitrogen content.
Since the mechanisms of fuel nitrogen oxidation are presently
quite imperfectly understood, and since the role of superequilib-
rium oxygen atoms in these mechanisms is unclear, speculation on
the effect of fuel sulfur is at this point somewhat premature.
Preliminary experimental results are first required to help
orient our thinking on this question. However, recent work of
Flagan at at. (10) indicates that superequilibrium concentrations
of atoms and free radicals may play a role and this might lead us
7
-------
tentatively to speculate that sulfur dioxide may have an inhibi-
tory effect similar to that hypothesized for thermal NO. More-
over, according to Flagan e.t at. (10) it makes little difference
in what form fuel nitrogen is introduced and so even NO itself
could be considered a fuel nitrogen compound. It is therefore
instructive to simulate fuel nitrogen by both NO and by an
equivalent amount of cyanogen (C2N2). Although ammonia has
often been used as a representative fuel nitrogen compound'8'10',
it is not suitable for this study since it reacts with both S02
and HzS to form solid sulfite and bisulfite salts before com-
bustion is initiated.
In order for laboratory results to be extrapolated to other con-
ditions, it is also necessary to develop theoretical mathematical
models that describe the appropriate kinetic mechanisms, and that
can be used to determine the significance of the results in
other, more practical combustion environments. In particular,
mathematical models will give insight into fundamental questions
such as
• Can the observed effect be explained by catalysis of
atom and radical recombination rates by S02?
• What can be expected under different time temperature
histories?
• What can be expected in real furnace flames?
-------
EXPERIMENTAL APPARATUS
COMBUSTION RIG
A schematic of the premixed combustion rig and supporting equip-
ment used is shown in Figure 1 and a diagram of the burner itself
in Figure 2. Meted amounts of methane (Matheson, C.P.)/ pre-
heated house air, and when applicable, S02 or H2S (Matheson,
C.P.), were allowed to mix, then preheated further before being
fed into a modified Meker burner. The temperature of the gas
mixture entering the burner was controlled. The Meker burner
was modified so that an approximately flat flame could be sup-
ported above the burner grid. The burner was at atmospheric
pressure and enclosed in a pyrex glass chimney.
The combustion rig was designed primarily for a large number of
input/output measurements rather than for detailed in-flame
probing. However, some detailed probing was successfully at-
tempted, and this showed that the flame could be considered
flat to within our experimental error.
Incomplete temperature profiles taken with an uncoated 0.001"
Pt-Pt/10% Rh thermocouple showed that the flame had temperatures
in excess of 2000°K, even with no air preheat. This means that
heat loss to the surroundings was not great, and might dis-
tinguish this flat flame from others(1l•l2).
SAMPLING AND ANALYSIS TRAIN
A schematic of the sampling and analysis train is shown in
Figure 3. The sample was drawn through a 7mm diameter orifice
into a 6mm OD, uncooled, quartz sampling tube. Preliminary
-------
TO ANALYSIS
TRAIN
VOLTMETER
TEMPERA-
TURE
CONTROLLER
-<
//
U
f
—
BURNER
• ASSEMBLY
TO EXTERNAL
POWER SOURCE
I
I
INSULATION I
a i
HEATING TAPES
MIXING
'LENGTH
PREHEAT ASSEMBLY
REGULATOR/
LINE FILTER
ROTAMETER
/VALVE
ASSEMBLY
METHANE
S02/H2S
AMMONIA
LOW PRESSURE AIR
FIGURE 1. SCHEMATIC OF APPARATUS
10
-------
QUARTZ
GLASS
CHIMNEY
GRID
GLASS
BEADS
POROUS
METAL
PLATE
MIXTURE
INLET
COOLING
WATER
COIL
PREMIXED
BURNER
FUEL and
S02/H2S+AIR
SWAGELOK
FITTING
COOLING
WATER
FIGURE 2. FLAT FLAME COMBUSTOR
11
-------
AIR BYPASS
OXYGEN
MONITOR
GAS CHROM
ATOGRAPH
SO?
DETECTOR
TO EXHAUST HOOD
HELIUM
FROM BURNER
*ASSEMBLY
FROM CALIBRA-
TION GAS
SOURCE
NO/NOx
CHEMILUMIN-
ESCENT t
DETECTOR]
IPUMF1
OXYGEN
GO \QQ_
FIGURE 3. SCHEMATIC OF ANALYSIS SYSTEM
12
-------
experiments in which the sampling rate was varied by a factor of
four showed that our data did not depend on sampling rate and
this, together with data from other experimentalists'13) indi-
cates that all reactions were effectively rapidly quenched and
no further reaction took place in the tube. The height of the
sampling tube could be positioned accurately to within 0.03 mm.
Quartz and teflon tubing were used throughout the sampling and
analysis train since stainless steel tubing has been shown to
interfere with NO analysis under rich conditions. A cooled
knockout pot removed moisture in the burned gas sample. The
analysis train had the following features:
NO/NOX analysis by Thermo Electron Chemiluminescence
Analyzer with stainless steel converter
• 02 analysis by Beckman Model 715 (Electrochemical)
02 Monitor
• CO analysis by chromatograph with Porapak Q columh
• S02 analysis by Theta Sensors S02 (Electrochemical)
Monitor
The Chemiluminescence analyzer worked perfectly and showed no
interference by S02, 02 or CO. This confirmed previous re-
sults f1") which showed that S02 does not interfere with (Thermo
Electron) Chemiluminescence measurements of NO.
For the Phase II results, the analysis train had the following
additional features
• Molybdenum converter for Thermo Electron NO/NOX analyzer
• CO analysis by NDIR, Beckman Instruments
• S02 analysis by Thermo Electron Pulsed Fluorescent
S02 Analyzer
13
-------
RESULTS
PREMIXED COMBUSTOR PERFORMANCE
Figure 4 and Figure 5 show results obtained with no S02 or HzS
added to the fuel. These are the base cases showing exhaust NO
emissions as functions of air fuel ratio, air preheat (Figure 4)
and NO concentration within the flame as a function of residence
time from the burner (Figure 5). Figure 4 shows that with no
preheat a maximum of 152 ppm (dry, reduced to 100% stoichio-
metric air NO was obtained at 104% stoichiometric air while with
240°C air preheat the maximum was 232 ppm. There is also a
strong dependence on air fuel ratio. Figure 5 shows that forma-
tion of NO was complete at 6 cm above the burner grid or after
a residence time of'approximately 20 milliseconds and that sam-
pling at that point was truly representative of exhaust NO
emissions. Figure 5 also shows that under fuel rich conditions
all the NO is formed very early in the flame and that this
"prompt NO" was not a strong function of air preheat and that
more "prompt NO" was formed under fuel rich conditions than under
fuel lean conditions. These results agree qualitatively with
those of Fenimore'9'.
The ppm NO measured, under no preheat conditions, is substan-
tially greater than that measured by other workers in flat
flames^12'. This is probably due to the low heat loss rate in
our system, and by the resulting high temperatures. The exis-
tence of temperatures well above 2000°K was confirmed by (in-
complete) temperature measurements(15).
At each point (Figure 5) under fuel lean conditions both NO and
14
-------
Q RUN I
ORUN 3
ARUN 4
ORUN 5
250
240°
PREHEAT
NO
PREHEAT
80
FIGURE 4
90
100
110
120
% STOICHIOMETRIC AIR
THERJ1AL NO EXHAUST EMISSIONS - BASE CASE
15
-------
200
5 '50
I-
c/5
100
Q_
Q.
50
O 80.1% STOICH. AIR
D 101 % STOICH. AIR
117.5% STOICH. AIR
NO PREHEAT
240° PREHEAT
1 i
1
2.0 4.0
10 15 20
TIME (MILLISECONDS)
25
FIGURE 5. EXHAUST LEVELS OF NO ACHIEVED RAPIDLY
-------
NOx were measured by the chemiluminescent analyzer. At most 3
ppm NO2 were observed, and then only under very fuel lean
conditions. We thus did not observe any appreciable early NOa
formation as reported by Merryman and Levy(16).
Additional runs were also made to investigate whether the
addition of a fuel additive, such as S02, would lower NO emis-
sions significantly by virtue of dilution alone. With molecular
N2 as the fuel diluent at zero preheat, 104% stoichiometric air,
it was found that 10% N2 in the fuel led to a reduction of less
than 7 ppm NO in the exhaust. This means that any effect
(larger than this) due to addition of up to 5% SO2 to the fuel
is due to kinetic interactions and not just simple dilution and
temperature reduction.
PHASE I. EFFECT OF S02 ON THERMAL NOX EMISSIONS
The effect of S02 as an additive in the fuel on the exhaust emis-
sions of nitrogen oxide is shown in Figures 6 and 7. In Figure
6 the ppm NO (dry, reduced to stoichiometric) in the exhaust is
shown in the absence of air preheat with and without 4.9 percent
by volume S02 in the fuel. 4.9 percent by volume S02 in methane
leads to approximately 6800 S02 in the exhaust. This sulfur
level is considerably higher than that resulting from typical
fossil fuels, and corresponds roughly to that for coal containing
eight percent sulfur by weight. It can be seen that at approxi-
mately 101% stoichiometric air, 4.9% S02 in the fuel lowers NO
exhaust emissions by 50 ppm or by about 36%. At other air/fuel
ratios the percent reduction is somewhat less as shown on Table
1. At a preheat of 240°C (Figure 7) and at 101.% stoichiometric
air the addition of 4.9% of S02 in the fuel lowers NOx emissions
by about 60 ppm or 30%. Conversely, looking at the effect of
removal of S02 from the fuel one can say in this case, fuel
desulfurization caused increases in thermal NOx emissions of up
to 55%.
Further details are shown in Table 1 in which results from two
17
-------
QRUN I
0 RUN 2
A RUN 6
Q RUN 7
150
0%SO
NO PREHEAT
80
120
90 100 110
% STOICHIOMETRIC AIR
FIGURE 6. S02 INHIBITS NO FORMATION AT ZERO PREHEAT
18
-------
0 RUN 3
4
ORUN s
e
D RUN
240°C PREHEAT
FIGURE 7
120
% STOICH10METRIC AIR
S02 INHIBITS NO FORMATION AT HIGH PREHEAT
19
-------
Table 1. REDUCTION IN THERMAL NOX EMISSIONS BY S02 ADDITION TO FUEL
Percent
Stoichiometric
Air
80
90
101
103
110
g 117
.1
.0
.3
.0
.0
.4
Reduction in NOX
Emissions
2. 5% S02 in Fuel
no preheat
ppm %
7 12.6
—
22.0 14.0
—
—
12.0 11.7
240°C preheat
ppm %
6.0 10.7
— —
21.0 10.7
— —
— —
37.0 27.4
no
4.9%
SO
preheat
ppm %
6.
9.
50.
40.
27.
16.
0
0
0
0
0
0
16.
17.
35.
26.
24.
13.
0
3
7
0
5
9
2 in
240
ppm
13.
15.
60.
68.
60.
22.
6
0
0
0
0
0
0
Fuel
C preheat
%
24
23
30
30
26
21
.6
.1
.9
.4
.4
.6
-------
different flames are presented. (Flame 1 has slightly lower base
case NO emissions). These results clearly show that as the SOz
level in the fuel decreases so does reduction in NOX emissions.
With less than 1% S02 in the fuel any inhibition effect was not
significant.
Figure 8 shows the results of probing within the flame (118.5%
stoichiometric air, 4.9% S02 in the Fuel) and clearly demon-
strates that at both preheats the effect of SC>2 is to quench the
formation of NO fairly early in the flame; This data give in-
sight into a probable kinetic mechanism as described later.
In Figure 9 the effect of 2.5% and 4.9% S02 on the formation of
"prompt NO" is shown, where "prompt NO" is defined in this case
as that formed within 0.3 cm of the burner grid. It should be
noted that our definition of "prompt NO" differs from that of
Fenimore, in that he defined it as the intercept of a linear
extrapolation of the NO concentration profile back to zero
residence time. It can be seen that at both preheats and at all
air fuel ratios, the effect of increasing S02 is to decrease
"prompt NO" formation. This indicates that superequilibrium
concentrations of atoms and free radicals might be important
under all air/fuel ratio conditions.
PHASE I. EFFECT OF HaS ON THERMAL NOX EMISSIONS
In Fossil fuels, sulfur is normally present in the reduced state.
Thus, some experiments were completed with H2S as the fuel
additive, in order to determine whether fuel sulfur in this form
has an effect on thermal NOx formation.
Figures 10 and 11 show the effect of H2S addition at two levels
of mixture preheat. Since H2S is a fuel, addition of this com-
pound changes the air fuel ratio, and this has been taken into
account in labeling the abscissa axis. In Figure 10 it is clear
that both 2.6% and 5.0% H2S in the fuel, with no mixture pre-
heat, inhibit the formation of NOX. NOx emissions were reduced
21
-------
150
x
u
5100
0- 50
CL
0<
O No Preheat
C> No Preheat
Q 240° C Preheat
O 240° C Preheat
FLAME
FRONT
0.0%
SO 2
in fuel'
4.9%
S02
in fuel
0.0% S02
4.9% SO 2
.05
.1
.5 1.0 5.0
TIME (MILLISECONDS)
10.0
FIGURE 8. S02 DECREASES RATE OF NO FORMATION NEAR THE FLAME
-------
Oo% S02
Q 2.5% SO 2
0 4.9%S02
75-
25--
240°C PREHEAT
NO PREHEAT
75--
Q.
Q.
80
FIGURE 9
90
100
no
120
% STOICHIOMETRIC AIR
EARLY-FORMED NO INHIBITED BY SO2
23
-------
O RUN I
DRUM 2
ORUN 14
A RUN 15
NO PREHEAT
0 RUN 16
S3RUN 17
0% H25
.. 5.0%
FIGURE 10,
90 100 110 120
% STOICHIOMETRIC AIR
H2S INHIBITS NO FORMATION AT ZERO PREHEAT
24
-------
O RUN 3
RUN 4
QRUN is
ORUN 19
<2l RUN 20
0 RUN 21
0% H2S
2.6% H2S
5.0% H2S
I
240°C PREHEAT
a.
a.
80
120
90 100 110
% STOICHIOMETRIC AIR
FIGURE 11. H?S INHIBITS NO FORMATION AT HIGH PREHEAT
25
-------
up to 31.6% under fuel lean conditions as shown on Table 2. With
250°C mixture preheat, H2S inhibits NOX formation on the average
by an even greater extent as shown on Figure 11 and Table 2.
L?igure 12 shows the effect of H?S on "prompt NO", and indicates
that under fuel lean conditions the presence of H2S does lower
"prompt NO" formation rates. Under fuel rich conditions, where
conversion of H2S to S02 is not complete, there is little effect
of H2S on "prompt NO".
The foregoing indicates that H2S must be converted to S02 before
inhibition1 of NO is important, and that this occurs rapidly under
fuel lean conditions. This is confirmed in Figures 13 and 14 in
which S02, NO, and 02 concentrations are plotted as functions of
time for zero preheat. At 98% stoichiometric air (Figure 13) H2S
conversion to S02 is essentially complete when NO has attained 70%
of its final value; at 113.5% stoichiometric air the conversion of
H2S to S02 is essentially complete when the NO has attained only
34% of its final value. Since H2S has a greater inhibiting effect
in the fuel lean case, it would appear that inhibition of NO for-
mation occurs through the rapid conversion of H2S to SO2 and by
the subsequent inhibiting effect of S02. Thus, under fuel lean
conditions, a kinetic model simulating fuel sulfur as S02, rather
than as H2S, would be adequate.
PHASE II. EFFECT OF FUEL SULFUR ON FUEL NOX EMISSIONS
In this phase of the research, the effect of the presence of sul-
fur compounds in a gaseous fuel on the formation of nitrogen
oxides arising from fuel nitrogen oxidation was examined. The
problem is important because removal of sulfur from a fossil fuel
does-not necessarily lead to the removal of an equivalent amount
of fuel nitrogen.
Unfortunately, our results are somewhat contradictory and some
further work is required to reconcile some of the discrepancies
discussed below. Experimental difficulties were encountered in
26
-------
Table 2. REDUCTION IN NOX EMISSIONS BY H S ADDITION TO FUEL
Percent
Stoichiometric
Air
Reduction in
2.6
NOX
Emissions
% H2S in Fuel
no preheat
ppm %
80
90
100
103
110
115
5.
8.
28.
17.
15.
15.
0
0
0
0
0
0
11
14
23
11
11
15
.4
.6
.3
.4
.1
.8
24
0°C
ppm
7
10
26
50
44
23
.0
.0
•0
.0
.0
.0
preheat
%
13
15
17
22
19
14
.2
.4
.1
.8
.1
.1
no
ppm
5.
8.
28.
28.
26.
30.
5.0% H
preheat
0
0
0
0
0
0
%
11.4
14.6
23.2
18.9
19.4
31.6
2S in Fuel
240
ppm
13.
20.
42.
75.
63.
48.
°C
0
0
0
0
0
0
preheat
%
24.6
30.8
27.7
34.0
28.0
29.4
-------
O 0% H2S
O 2.6% HaS
Q 5.0% H2S
4-
75"
240° PREHEAT
25
NO PREHEAT
••75
80
FIGURE 12,
90
110
120
% STOICHIOMETRIC AIR
EARLY-FORMED NO INHIBITED BY H2S
28
-------
120
100
g
"o
X80
CM
O
NJ
E 60
o.
6T
o
20
FLAME FRONT
5.0 10.0
TIME (MILLISECONDS)
SO-
6500-
E
CL
O.
CVJ
O
CO
20.0
FIGURE 13. CONVERSION OF H2S TO S02 IS MORE RAPID THAN NO FORMATION-FUEL
RICH CONDITIONS
-------
120
_
g
"o
x 80
CM
JP
60
40
20
O
z
FLAME FRONT
1.0
2.5
6OOO-
5500
5000-
E
a.
CM
O
20.0
5.0 10.0
TIME (MILLISECONDS)
FIGURE 14. CONVERSION OF H2S TO S02 IS MORE RAPID THAN NO FORMATION-FUEL
LEAN CONDITIONS
-------
reproducing exact same fuel nitrogen additive flow rates from one
clay to the next, although within any one experimental run,
additive flow rates were maintained constant. Nitrogen oxide
emission are, of course, very sensitive to the quantity of fuel
nitrogen added to the flame. It should be emphasized, therefore,
that the results from Phase II, are preliminary in nature. They
are reported here because they appeared to indicate the presence
of new, interesting and reproducible phenomena, when both fuel
sulfur and fuel nitrogen are present in a flame. When viewed in
this light, the results are valuable, since they provide impetus
for further investigation.
The research can be divided into two separate experimental trials,
In the first trial the concentration of nitric oxide (NO) was
measured as a function of distance from the burner grid for
various air fuel ratios with and without sulfur dioxide added to
the fuel. The fuel was doped with NO to simulate fuel nitrogen.
In this trial the flame was detached from the burned grid and
this allowed probing well into the flame front.
In the second trial, the combustor was modified to allow greater
ease of operation. The modifications caused the flame to burn
.partially upstream of the grid surface, thus preventing any
probing well into the flame front. This difference in flame be-
havior between the first and second trials, might explain
apparent discrepancies between results from these trials. In the
second trial, NO, NOx, S02 and CO were measured in all runs. A
Molybdenum converter was used to convert NO2 to NO under fuel
rich conditions. The fuel was doped with both NO and with C2N2
to simulate fuel nitrogen.
PHASE II. TRIAL 1
Concentration profiles of NO are shown for three stoichiometric
ratios on Figures 15 through 17. Fuel nitrogen was simulated by
addition of NO in the fuel. Since some NO oxidized to N02 be-
fore reaching the burner grid the inlet values of NO and NOx are
31
-------
1500
Inlet NO
107% Stoichiometric
Inlet NOX Air
1000
E
a
a
500
O No SOz present
3.5% SOa in fuel
i i i i I I I i I I I I
0 5 10
HT. ABOVE GRID, cm (7cm ~ 20 m sec)
FIGURE 15.
EFFECT OF SO2 on FUEL NO UNDER
FUEL LEAN CONDITIONS
32
-------
2000^1 (2150 ppm)
Inlet NO
1500
E
o.
o.
OIQOO
94% Stoichiometric
Air
Inlet NO*
High sample
rate
Low sample rate
II I I I I l I I I I I I
500 -
0 5 10
HT. ABOVE GRID, cm (7 cm ^ 20 m sec)
FIGURE 16. EFFECT OF SAMPLING RATE ON FUEL
UNDER FUEL RICH CONDITIONS
33
-------
2000
89% Stoichiometric
Air
1500
E
a
a
1000
500
High and low sample rate
S02)
High sample
rate
(No SO?)
Low sample
rate
(No S02)
Inlet NO
Inlet NOX
i i i I I I I I
0 5 10
HT. ABOVE Gmtfreitr(7 cmi *v 20m sec)
FIGURE '17. EFFECT OF''s02 AND SAMPLING RATE
ON NO PROFILES
34
-------
those measured leaving the grid in the absence of a flame.
Figure 15 shows results under fuel lean conditions. The pres-
ence of S02 in the fuel had essentially no effect on the NO pro-
file. Figure 16 shows the concentration profile of NO under
fuel rich conditions. A rapid decline in apparent NO concentra-
tions was followed by a very slow increase. This was unex-
pected, and so the sampling rate through the uncooled quartz
probe was varied in order to determine if probe effects were
controlling. Results on Figure 16 indicate that under our con-
ditions reactions in the probe were apparently destroying NO,
but that these reactions do not occur when the sample is with-
drawn far from the flame front. Figure 17 shows NO concentration
profiles under fuel rich conditions, with and without S02 present
(3.5% SO2 in the fuel) and at high and low sampling rates. The
following observations can be made. First, in the absence of
S02, the same basic trends as shown in the previous figure were
observed. Second, S02 markedly affected the apparent NO profile,
although exhaust values did not change significantly. Third,
with SO2 present there appeared to be little effect of sampling
rate. However, it should be noted that neither the presence of
S02 nor sample rate had an effect on exhaust values of NO.
PHASE II: TRIAL 2
Trial 2 was completed after substantial modifications to the
burner had been made. These modifications were necessary be-
cause of wear and tear on the original combustor. It was hoped
to explore the results of Trial 1 in more detail; however, it
became quickly apparent that the combustor performance in Trial
2 was quite different and that this led to significant qualita-
tive differences between results of the two Trials. An important
difference between the two combustors was that in the latter
trial, the stable flame was seated in and below the grid, while be-
fore it was lifted several millimeters above the grid. The results
of the trials should therefore be viewed as results from different
combustors. In Trial 2 both NO and cyanogen were added to the
fuel and both NO and NOX were measured. The effects of SOz and
35
-------
sampling rate were investigated. Results are shown in Figures
18 through 21.
Figure 18 shows concentration profiles for NO addition with and
without S02 addition under fuel lean conditions. Sample rates
were not changed. Essentially no significant effect was observed,
and these data are in agreement with those from Trial 1. The
small decrease in NO due to the presence of S02 is probably due
to thermal fixation effects investigated in Phase 1.
In Figure 19 the concentration profiles are shown for the case
where cyanogen was used to simulate the fuel nitrogen compound.
Conversion of cyanogen to NO was close to 100%. These data agree
with those of DeSoete^17). No effect of sampling rate was
observed.
Figures 20 (NO addition) and 21 (C2N2 addition) show concentration
profiles of NO and NOX under fuel rich conditions. It is
immediately clear that these data differ substantially from those
of Trial 1, in that exhaust values of NO and NOX are achieved much
more rapidly. Indeed, in Trial 2, the rapid decrease of NO to
very low values, in the flame front, was observed only with
difficulty. This was because the flame front extended to behind
the grid. From Figure 20 (NO addition) one can deduce that, in
the flame front region, the apparent values of both NO and NOX did
depend on sampling rate, while with S02 present, they did not.
Exhaust values were essentially unaffected. With C2N2 as the fuel
nitrogen additive, the data (Figure 21) show that sampling rate
did have an effect on NO and NOx in the flame front both with and
without S02. In addition, S02 tended to decrease the exhaust
emissions of NOX by about 400 ppm. This is a new phenomenon, and
should be examined further.
Reproducibility of data shown in Figures 18 through 21 was good,
and many overlapping points have not been shown in order to im-
36
-------
u>
o
I 1000
ex
X
O
o
z
10
.2
I
NOX Inlet value-no flame
NOX No S02
O NOX 3.5% S02
Inlet value -no flame
NO No S02
ONO 3.5% so 2
FIGURE 18
10-' I
DISTANCE FROM BURNER GRID, cm
EFFECT OF SO2 ON NO, NOX, PROFILES, (FUEL NITROGEN = NO, 107%
STOICHIOMETRIC AIR)
10
-------
o
tr
j-
LJ
5
o
X
2000
CO
1000
x
O
z
o"
z
0
lO-2
FIGURE 19.
/
0
/
o
/
NOX No SO2
O NOX 3.5% S02
^ NO No S02
O NO 3.5% S02
10-' I
DISTANCE FROM BURNER GRID, cm
EFFECT OF S02 ON NO, NOX PROFILES (FUEL NITROGEN = C2N2/ 107%
STOICIIIOMETRIC AIR)
10
-------
^-•o--^
vo
X
o
10
_2
Inlet NOx no flame
Inlet NO no flame
ifc
o
o
NOx, High sample rate, No SO2
NOx, Low sample rate, No SOe
NOx, High and low sample rate,
3.5% S02 in fuel
NO, High sample rate, No S02
NO, Low sample rate, No SO2
NO, High and low sample rate,
I
3.5% S02 in fuel
10
.1
I
FIGURE 20,
DISTANCE FROM BURNER GRID, cm
EFFECT OF SO2, SAMPLING RATE ON NO, NOX,.PROFILES, (FUEL NITROGEN
89% STOICHIOMETRIC AIR)
= NO,
10
-------
2000 -
o
a:
h-
UJ
o
o
~ 1000
E
O.
a.
x
O
Low sample rate
NOx High sample rate-
no S02
NOX Low sample rate-
no S02
O NOx High/low sample
. rate-3.5 %S02
-------
prove clarity. In each case S02 concentrations were measured,
and showed that under fuel rich conditions S02 was reduced
slightly in the flame front, and then restored back to its
original value. It should be noted that inlet values of NO and
NOX were measured in the absence of combustion, and therefore in
the absence of water vapor. The actual inlet values of NO and
NOx during combustion would be somewhat higher. The species
concentrations reported are on a dry basis and reduced to
stoichiometric conditions. For the purposes of this calculation
it was assumed that all unburned fuel was in the form of CO. A
logarithmic abscissa scale was used in Figures 18 through 21 in
order to allow an expanded scale in the flame front region.
41
-------
MATHEMATICAL MODELING
In order to model the kinetic mechanisms of the sulfur-nitrogen
oxide interactions experimentally observed it is first necessary
to model the physical environment of a flat flame in which the
reaction chemistry occurs. In a flat flame the physical pro-
cesses of convection and diffusion are important and simple plug
flow models are inadequate. Indeed a substantial amount of back
diffusion into the unburned gases is crucial in allowing a stable
flame front to be maintained, and in allowing ignition to occur.
Simple models that impose a specified time temperature-history on
the kinetics environment can be misleading, especially in the
case examined here, where temperature, free radical concentra-
tions and nitrogen oxide kinetics are intimately coupled. For
example, high superequilibrium concentrations of atoms and free
radicals necessarily lead to significantly lower temperatures
because of the enthalpy of disassociation of oxygen, hydrogren
and water molecules. Thus a substance that catalyzes atom recom-
bination rates and lowers free radical concentrations, will also
raise the flame temperature at that point and this rise in tem-
perature may offset, in some degree, the effect of lower oxygen
atom concentrations as regards NO formation. Thus any reasonable
model describing the observed effects must
a) calculate the resulting temperature from a
heat balance and
b) properly take account of diffusion in the flame
front
Unfortunately, no model of a flat flame is generally available
and so it was necessary to develop a very approximate simulation
42
-------
to be used in this study. It is recognized that substantial im-
provements in such a simulation are desirable; however, our
model is an improvement on others that take no account of dif-
fusion in the flame front. The approach used here was to develop
a simplified premixed flat flame model that takes account of
diffusion, then to calibrate this model against
a) literature data on free radical concentrations and
b) our own base case of NO formation without sulfur
present
It should be noted that the model therefore used only two unknown
parameters, one of which was obtained from the open literature,
the other of which uses our own base case data. This model was
then used to test kinetic mechanisms of sulfur oxide - nitrogen
oxide interactions, and the resulting mechanism was then used to
determine the effect of different environments and different heat
loss rates corresponding to those likely in a furnace.
PREMIXED FLAT FLAME MODEL
The salient features of our preliminary flat flame model is that
the diffusion in the ignition zone is assumed to be such that it
can be simulated by a well stirred stage or pointwise calculation.
This simulation is exact(*8) only when the true profiles are
parabolic and since this is seldom the case, the model should be
considered only an approximate representation of our flat flame.
The ignition zone is then followed by a plug flow calculation.
The volume of the (hypothetical) well stirred stage is determined
by that which allows a certain fraction of a species (designated
"fuel") to be destroyed. The physical assumption is, that for a
given fuel, the correct scaling parameter is a quantity related
to the flame thickness, and that this length can be determined
by the concentration profile of the species designated as "fuel".
The model, at this stage, is largely intuitive, and should be re-
garded merely as a mathematical device to simulate ignition. A
rigorous mathematical justification of this approach is outside
43
-------
the scope of this phase of the project. Here we will merely
demonstrate that' the model does predict atom concentrations of
tho correct, order as those measured (19) and this gives us some
confidence I.hat the model can describe the pheomena of interest
here.
The basic tool used was computer program REKINET which inte-
grates the conservation equations for well stirred and plug flow
reactors both with and without a heat balance. In addition, this
program allows the volume of the well stirred stage to vary until
a specified fraction of a specified species is converted.
The basic approach used to model the flat flame investigated here
consisted of the following steps:
• choose a kinetic mechanism for CHt^/air combustion
• determine which value of percent CHi» consumed (denoted by
X) during pointwise calculation led to measured oxygen
atom concentrations of Peeters and Mahnen(19)
• use this value of x to simulate our flat flame and
calibrate the heat loss parameters in the model until
the predicted and measured base case NO profiles matched
• investigate changes due to SC>2 addition, assuming atom
recombination catalysis with mechanisms and rates pro-
posed by Halstead and Jenkins'5' and Merryman and Levy(20).
No other parameters should be altered in this phase.
KINETIC MECHANISM
The methane air reactions used were those suggested by Waldman
e.t at. (**) in an EPA sponsored investigation of kinetic mechanisms
of methane/air combustion with pollutant formation. A list of
reactions is shown on Table 3. The reactions name ULT36 through
ULT143 denote the reactions numbered 36 through 143 in Table 7.5
of Reference (4). Special consideration is made of catalysis of
atom recombination rates through 02, forming H02 as an inter-
mediate - see reactions ULT101 and ULT85.
44
-------
Table 3. METHANE COMBUSTION MECHANISM
ULT36
ULT77
ULT84
ULT99
ULT101
ULT140
ULT44
ULT46
ULT47
ULT52
ULT59
ULT63
ULT65
ULT66
ULT70
ULT83
ULT85
ULT88
ULT91
ULT98
ULT100
ULT117
ULT125
ULT133
ULT135
ULT143
JOHN 1.1
CHO
CO 2
H20
H
H
N20
CHO
CH20
CHO
CHO
CH3
CH4
CH4
CH4
CO
H
H
OH
OH
H
OH
OH
N
N
N20
CHO
02
+M
+M
+M
+0
+02
+M
+H
+0
+OH
+0
+0
+0
+H
+OH
+OH
+OH
+H02
+H2
+N
+N20
+0
+OH
+NO
+02
+0
+02
+M
=co
=co
=OH
+M =OH
+M =H02
=N2
=CO
=CHO
=CO
=CO
=CH20
=CH3
=CH3
=CH3
=C02
=H2
=OH
=H
=H
=OH
=H
=H20
=N2
=NO
=NO
=CO
=0
+H . +M
+0 +M
+H +M
+M
+M
+0 +M
+H2
+OH
+H20
+OH
+H
+OH
+H2
+H20
+H
+0
+OH
+H20
+NO
+N2
+02
+0
+0
+0
+NO
+H02
+0
45
-------
In addition, for the runs simulating sulfur addition, the atom
recombination catalysis is described by reations shown in Table
4. Those reactions named MERLJ and MERL3A are reactions numbered
1 and 3A by Merryman and Levy(2°) while those labeled JENKI are
from Halstead and Jenkins'5 . Reaction JOHN7J is from Johnston's
review of 0 atom kinetics'21).
These reactions demonstrate the catalysis of 0 atom recombination
by S02 via S03 as an intermediate as well as the catalysis of H
atom and OH radical recombination to form H20, with HS02 as an
intermediate. No adjustment of rate coefficient values from
those suggested by the original authors was made and it was as-
sumed throughout that:
kf/kr ~ Kequil
An important result obtained from this kinetic model is to deter-
mine whether this atom recombination catalysis is sufficient to
account for the drop in NO emissions caused by SOa addition. In
addition, a kinetic calculation of this type allows the separate
effects of temperature profile changes and radical concentration
changes to be investigated. This should lead to greater insight
into the salient features involved.
CALIBRATION OF FLAT FLAME SIMULATION WITH DATA OF PEETERS
In our simulation of a flat flame the ignition zone is simulated
by a well stirred stage or pointwise calculation where the hypo-
thetical volume is determined by that volume which will convert
a certain fraction x» of tne primary fuel. This is followed by
a plug flow heat balanced calculation. 'We settled on a value of
X, of the primary fuel. This is followed by a plug flow heat
balanced calculation. We settled on a value of x by calibrating
our simulation with the data of Peeters and Mahnen(18). Heat
loss in the ignition zone was assumed to be negligible. A value
of
46
-------
Table 4. S02 CATALYZED RECOMBINATION OF ATOMS AND RADICALS
MERL1
MERL3A
JENKI2
JENKI2
S02
SOS
H
HS02
+0
+0
+S02
+OH
+M
+M
=S03
=S02
=HS02
=H20
+M
+02
+M
+S02
47
-------
X = 0.98
was chosen because, as shown in Table 5:
• the maximum CH4 consumption rate was then similar to
that measured
the peak 0 atom mole fraction (0.048) was of the same
order as that measured (0.025) compared to the equilib-
rium 02 mole fraction which was two orders of magnitude
lower
• the temperature of the hypothetical well stirred stage
matched that measured at the maximum CHi» consumption rate.
The simulation did over predict the atom concentration by a fac-
tor of two and also tended to under predict the rate of tempera-
ture increase. Obviously the simulation does not give a true
picture of the flat flame at this stage, and the discrepancies
are probably due to inaccuracies in both the model and the kinet-
ic mechanism. Nevertheless, the simulation was considered suf-
ficiently adequate to investigate the kinetic mechanism appro-
priate to S02 inhibition of nitrogen oxides.
This calculation also demonstrated that the kinetic mechanism of
methane combustion proposed by Waldman ("*) did contain the salient
features observed by Peeters and Mahnen(17). For example, the
predicted formaldehyde, hydroxyl and carbon monoxide profiles
were reasonably close to those measured. This gives both the
kinetic mechanism and the simulation some credence.
CALIBRATION WITH BASE CASE NO MEASURED
The base cases used to test the kinetic model were those with
104% stoichiometric air at both zero and 240°C mixture preheat.
We restricted our investigation to the fuel lean regime because
the dominant NO formation kinetics are there better understood.
Using the value of x = 0.98 determined previously and the kinetic
mechanism for methane combustion shown on Table 3 it was found
48
-------
Table 5. SIMULATION OF FLAME OF PEETERS et al. (1973)
Temperature at
Ignition Zone Exit, °K
Max. Rate of CH^
Consumption, moles/cc sec
Ignition Zone Exit,
Mole Fractions
CO
0
OH
CH20
Max. 0 Atom
Mole Fraction
Simulation
1569
5.53 x 10~5
0.0336
0.0343
0.0173
0.00126
0.0474
Experiment
1550
8.4 x 10~5
0.042
0.011
0.015
0.001
0.025
49
-------
that the NO measurements in our flat flame could be matched by
the simulation with a radiative heat loss coefficient
o = 3.45 x 10-1" cal/sec cm3 °K4
for the case with no preheat, and with
a = 8.0 x 10-14 cal/sec cm3 °K1*
for the case with 240°C mixture preheat.
The discrepancy between these two values indicate shortcomings in
our model. However, since the purpose of our model is to predict
the change in NO due to SO2 addition, it is reasonable to cali-
brate against both the zero and high preheat base cases indi-
vidually. Obviously an improved model should be able to predict
the effect of mixture preheat, without additional calibration.
EFFECT OF SO2 IN FUEL
With 4.9% SOz in the fuel the simulation showed a drop of 49 ppm
NO in the exhaust for the case with no preheat. This compares
with a measured drop of 40 ppm as shown in Table 6. At 240°C
preheat the simulation predicted a drop of 51 ppm NO compared to
a measured reduction of 70 ppm. Given the inaccuracies of the
physical model, and the kinetic rate coefficients, the simulation
predicts the correct effect of SOz addition with remarkable
accuracy, especially for the no preheat case. The discrepancy
between theory and experiment in the high preheat case may be due
to an inaccurate simulation of the heat loss under that condition.
It is clear, therefore, that the S02 catalysis of atoms and free
radicals as described by Reactions MERL7 through JENKI2 on Table 4
can explain the observed inhibition NO formation by SOz-
ANALYSIS OF SIMULATION AND APPLICATION
Calculated profiles of oxygen atom concentration, NO concentration
50
-------
Table 6. EFFECT OF SO2 ON NOX EMITTED: SIMULATION AND MEASURED VALUES
Base case, 104% stoich. air, zero preheat, NO ppm
4.9% SO2 in fuel, NO PPM
Reduction in NO ppm .
Percent reduction in NO
Simulation
148
99
49
33
Measured
150 .
110
40
27
Base case, 104% stoich. air, 240°C preheat, NO ppm
4.9% S02 in fuel, NO ppm
Reduction in NO ppm
Percent reduction in NO
232
181
51
22
232
160
72
30
-------
and temperature are shown on Figure 22 for the zero preheat case.
It is clear that the addition of S02 to the fuel changes both the
oxygen atom and the temperature profiles (assuming that the radia-
tive heat transfer coefficient remains unchanged). In the pres-
sence of SO?, the higher temperature at early times is intimately
coupled with the drop in atom concentration, which is significant.
This indicates that before lO"1* seconds the NO formation rate is
actually slightly higher with S02 than with no S02. However,
during the time when most of the NO is being formed and when
radiative heat loss is important, the drop in 0 atom concentration
dominates, and the resultant NO formed is significantly lower. It
is clear, therefore, that the reason behind the observed effect is
that the presence of S02 catalyzes the recombination of oxygen
atoms, and that the drop in oxygen atom concentration is suffi-
cient to lower the NO formation rate.
There is, however, a qualitative discrepancy between the predicted
profile of NO shown on Figure 22 and that measured (at a different
air/fuel ratio) and shown on Figure 8. In general, the measured
profile showed a more rapid formation of prompt NO than that
predicted. This qualitative discrepancy is probably due to un-
known features in the mechanism of prompt NO formation. It is
felt, however, that this discrepancy is not serious and does not
detract from the point that 0 atom recombination catalysis by S02
can explain the drop in exhaust NO measured, with no adjustment to
known rate coefficients being necessary. Further details of the
results from the model are shown on Table 7 for the no preheat
case and Table 8 for the high preheat case. These tables show the
early formation of superequilibrium concentrations of S02, which
is an intermediate species in the recombination catalysis scheme,
followed by a decline to relatively low values, corresponding to
approximately 1% conversion of S02 to S03. Surprisingly, the
calculations also indicate that the addition of S02 also appears
to hasten the CO burnout rate, although low CO levels were ob-
tained in all cases.
52
-------
Table 7. SIMULATION DETAILS - NO PREHEAT
Mole Fractions
Time (sec)
0% S02 in Fuel
Ignition Zone
Exit
0.11 x 10"5
0.125 x 10~1+
0.482 x 10~4
0.133 x 10-3
0.329 x 10~3
0.780 x 10~3
0.140 x 10~2
0.365 x 10~2
0.996 x 10~2
0.280 x 10"1
0.425 x 10-1
4 .9% S02 in Fuel
Ignition Zone
Exit
0.104 x 10~5
0.118 x ID''1
0.347 x 10-4
0.298 x ID"3
0.723 x 10~3
0.123 x 10~2
0.303 x 10~2
0.835 x ID-2
0.123 x 10"1
0.256 x 10"1
0.372 x 10"1
NO
0.479E-7
0.488E-7
0.720E-7
0.253E-6
0.112E-5
0.404E-5
0.118E-4
0.223E-4
0.537E-4
0.103E-3
0.140E-3
0.145E-3
0.622E-7
0.632E-7
0.883E-7
0.204E-6
0.388E-5
0.106E-4
0.178E-4
0.377E-4
0.701E-4
0.819E-4
0.956E-4
0.978E-4
0
0.458E-2
0.484E-2
0.555E-2
0.465E-2
0.343E-2
0.236E-2
0.158E-2
0.119E-2
0.749E-3
0.454E-3
0.197E-3
0.119E-3,
0.396E-2
0.415E-2
0.466E-2
0.386E-2
0.158E-2
0.994E-3
0.755E-3
0.480E-3
0.282E-3
0.212E-3
0.903E-4
0.457E-4
S03
__
--
—
--
—
—
—
—
—
—
--
—
0.384E-3
0.387E-3
0.325E-3
0.221E-3
0.321E-4
0.136E-4
0.887E-5
0.519E-5
0.425E-5
0.438E-5
0.534E-5
0.650E-5
CO
0.408E-1
0.414E-1
0.364E-1
0.267E-1
0.216E-1
0.183E-1
O.lSlErl
0.129E-1
0.979E-2
0.686E-2
0.338E-2
0.208E-2
0.400E-1
0.406E-1
0.362E-3
0.286E-1
0.154E-1
0.118E-1
0.100E-1
0.704E-2
0.494E-2
0.389E-2
0.180E-2
0.916E-3
Temp.
T°K
1659
1678
1754
1850
1935
2003
2055
2081
2099
2064
1939
1852
1686
1704
1783
1864
2046
2094
2113
2123
2090
2058
1959
1882
53
-------
Table 8. SIMULATION DETAILS - 240°C PREHEAT
Mole Fractions
Time (sec)
0% S02in Fuel
Ignition Zones
Exit
0.4 x 10~6
0.11 x 10-1*
0.4 x 10~3
0.158 x 10~2
0.47 x 10~2
0.16 x 10"1
0.22 x 10"1
4.9% S02 in Fuel
Ignition Zone
Exit
0.5 x 10~6
0.11 x 10-4
0.3 x 10"3
0.179 x 10~2
0.59 x 10~2
0.156 x 10-1
0.216 x 10"1
NO
0.129E-6
0.131E-6
0.196E-6
0.168E-4
0.748E-4
0.164E-3
0.226E-3
0.230E-3
0.169E-6
0.171E-6
0.260E-6
0.128E-4
0.769E-4
0.151E-3
0.180E-3
0.182E-3
0
0.610E-2
0.618E-2
0.721E-2
0.271E-2
0.149E-2
0.891E-3
0.315E-3
0.202E-3
0.539E-2
0.549E-2
0.613E-2
0.220E-2
0.982E-3
0.497E-3
0.131E-3
0.643E-4
SO 3
—
—
—
—
—
—
—
—
0.306E-3
0.307E-3
0.237E-3
0.175E-4
0.528E-5
0.503E-5
0.783E-5
0.106E-4
CO
0.440E-1
0.443E-1
0.394E-1
0.208E-1
0.154E-1
0.111E-1
0.491E-2
0.325E-2
0.432E-1
0.436E-1
0.384E-1
0.191E-1
0.122E-1
0.764E-2
0.251E-2
0.217E-2
Temp.
T°K
1745
1754
1836
2124
2175
2133
1947
1858
1773
1785
1875
2147
2199
2130
1957
1869
54
-------
1000
100
5
QL
CL-
Oxygen Atom x 10-'
10
r4
0% S02 in fuel
4.9%S02 in fuel \
Measured NO \
No preheat, radiation
heat loss
10-3
TIME (SECONDS)
10
r2
FIGURE 22. S02 ADDITION AFFECTS BOTH OXYGEN ATOM
AND TEMPERATURE PROFILES, BUT LOWERS NO
55
2100
2000
1900
1800
LJ
(T
ID
5E
tr.
UJ
a.
-------
It is instructive to examine the effect of SOz in the fuel on NO
formation rates when there is zero radiative heat loss, i.e.
under adiabatic conditions. This is shown in Figure 23 where it
is apparent that under adiabatic conditions the presence of S02
causes NO to reach its equilibrium value more slowly. This simu-
lation also indicated that SOa causes a change in the time
temperature history, but that the primary effect was due to lower
0 atom concentrations.
In order to separate out kinetic and temperature effects it is
useful to determine the role of S02 under a specified time-tem-
perature history. In this case there is no attempt made to
satisfy the heat balance, but rather it is assumed that the
temperature and heat transfer are controlled by the furnace con-
figuration. A realistic temperature history is one with an ex-
ponential temperature drop from 2100°K to 1050°K in one second.
This time-temperature history is roughly representative of that
felt by a labeled volume of premixed gas and fuel as it combusts
and moves through the convection section of. a furnace. Thus
this simulation can give some indication of what might happen in
a utility boiler, under conditions where fuel and air mixing is
very rapid. Results are shown on Table 9, and indicate that
under such conditions fuel sulfur is likely to inhibit the
formation of NOX.
56
-------
1000
Q.
0.
100
10
10
0%S02 in fuel
V 4.9% S02 in fuel
.3
IO-2 IO-1
TIME (SECONDS)
FIGURE 23. UNDER ADIABATIC CONDITIONS S02 DELAYS NO FORMATION
57
-------
Table 9. FURNACE SIMULATION PREMIXED MIXTURE, SPECIFIED
TEMPERATURE FALLING FROM 2100°K to 1050°K IN
ONE SECOND. APPROXIMATELY 4% EXCESS AIR
NO ppm
With sulfur
Approximate Time Without sulfur (4.9% S02 Fuel)
0.4 m sec 33.6 24.5
1 m sec 63.0 38.8
2.5 m sec 218.0 125.0
ExhaUst 339.0 189.0
58
-------
REFERENCES
1. Cullis, C.F., R.M. Henson, and D.L. Trinun, Proc. Roy. Soc.
(London) A295, 72 (1966) .
2. Wendt, J.O.L., and C.V. Sternling, Comb. & Flame, 21, 387
(1973) .
3. Thompson, D., T.D. Brown and J.M. Beer, Fourteenth Symposium
(International) on Combustion, p.787, The Combustion
Institute, Pittsburgh, Pa. (1973).
4. Waldman, D.H., R.P. Wilson, Jr., and K.L. Maloney, "Kinetic
Mechanism of Methane/Air Combustion with Pollutant
Formation". Environmental Protection Technology Series
Report EPA-650/2/74-045 (1974).
5. Halstead, D.J., and D.R. Jenkins, Trans. Faraday Soc., 65,
3013 (1969) .
6. Durie, R.A., G.M. Johnson, and M.V. Smith, Comb. & Flame,
17_, 197 (1971) .
7. Zeldovich, Y., Acta Physiochim, URSS 21, 577 (1946).
8. Bowman, C.T., Fourteenth Symposium (International) on
Combustion, p.729, The Combustion Institute, (1973).
9. Fenimore, C.P., Thirteenth Symposium (International) on
Combustion, p.373, The Combustion Institute, (1971).
10. Wendt, J.O.L., C.V. Sternling, and M.A. Matovich, Fourteenth
Symposium (International) on Combustion, p.897, The
Combustion Institute, Pittsburgh, Pa. (1973).
11. Flagan, R.C., Galant, S., and Appleton, J.P., Comb. & Flame,
2J2, 299 (1974).
12. Sarofim, A.F., and J.H. Pohl, Fourteenth Symposium
(International) on Combustion, p.739, The Combustion
Institute, Pittsburgh, Pa. (1973).
13. Fristrom, R., and A. Westenberg, "Flame Structure", McGraw-
Hill, Inc. New York, N.Y. (1965).
14. Brown, J.W., D.W. Pershing, J.H. Wasser, and E.E. Berkau,
•interactions of Stack Gas Sulfur and Nitrogen Oxides on
Dry Sorbents" U.S. Environmental Protection Series No.
EPA-650/2-73-029 (1973).
15. Ekmann, J.M., M.S. Thesis, Department of Chemical Engineer-
ing, University of Arizona (1974).
59
-------
16. Merryman, E.L., and A. Levy, "Nitrogen Oxide Formation in
Flames: The Roles of N02 and Fuel Nitrogen". Presented
at Fifteenth Symposium (International) on Combustion,
Tokyo, (August 1974).
17. DeSoete, G. "Formation D'Oxyde Nitrique Dans Les Flammes B-
Cyanogene" Report: Institute Franfais Du Petrole,
Division Applications, No. 21.309 May 1973.
18. Sternling, C.V., and J.O.L. Wendt, "Kinetic Mechanisms
Governing the Fate of Chemically Bound Sulfur and Nitro-
gen in Combustion". Environmental Protection Technology
Series, Report EPA-650/2-74-017 (1972) .
19. Peeters, J., and G. Mahnen, Fourteenth Symposium (Interna-
tional) on Combustion, p.133, The Combustion Institute,
Pittsburgh, Pa. (1973).
20. Merryman, E.L., and A. Levy, Thirteenth Symposium (Interna-
tional) on Combustion, p.427, The Combustion Institute,
Pittsburgh, Pa. (1971). .
21. Johnston, H.S., "Gas Phase Reaction Kinetics of Neutral
Oxygen Species", National Stand. Ref. Data Ser. , NBS 2_0_
(1968).
60
-------
PUBLICATIONS AND PRESENTATIONS RESULTING FROM GRANT R-802204
1) Wendt, J.O.L. and J.M. Ekmann, "Effect of Fuel Sulfur Species
on Nitrogen Oxide Emissions from Premixed Flames" to be
published in Combustion and Flame.
2) Wendt, J.O.L. and J.M. Ekmann, "Catalytic Inhibition of
Nitrogen Oxide Formation by Sulfur Dioxide" presented at
67th Annual Meeting, AIChE, Washington, D.C., December
1974. *
3) Sternling, C.V. and J.O.L. Wendt, "On the Oxidation of Fuel
Nitrogen in a Diffusion Flame", AIChE Journal 20, 81
(1974) .
4) Wendt, J.O.L. and J.M. Ekmann, "Effect of Fuel Sulfur on
Nitrogen Oxide Emission" presented at EPA Stationary
Source Combustion Symposium, Atlanta, Ga., September 24-26
1975. Proceedings to be published.
61
-------
APPENDICES
Page
A. Description of Computer Program REKINET 63
B. Sample Data Deck 69
C. Reaction Rate Library 73
D. Thermochemical Library 83
62
-------
APPENDIX A. DESCRIPTION OF COMPUTER PROGRAM REKINET FOR INTE-
GRATING STIFF DIFFERENTIAL EQUATIONS ARISING IN
KINETICS PROBLEMS
SUMMARY
A computer program, REKINET, for integrating sets of differen-
tial equations arising from kinetics problems is described.
This program, which uses "a cantilevered implicit method" is
especially suitable for systems of "stiff" equations such as re-
sult from problems with great disparity in characteristic re-
action times, as for example, in combustion problems. Use of
free format and internal tables of thermochemical and rate con-
stant data make the program especially easy to use. Either well
mixed stages or plug flow reactors or combinations of these
types can be simulated with specified temperature histories or
under heat balanced conditions. The integration method, TYSON,
can be used independently of the chemical reaction features for
dif ficult-to-integrate problems.
PURPOSE AND SCOPE
Program REKINET integrates the conservation equations for indi-
vidual chemical species for a stirred-tank or plug-flow reactor.
Temperature and pressure may be specified as functions of resi-
dence time in the reactor or a heat-balanced solution may be ob-
tained. Up to 35 species may be handled. The reactions (up to
70 in number) may be unimolecular , biomolecular or thermolecular
provided they can be represented by
where A, B, C, D, E, F represents a molecule or molecular frag-
ment. One or two of the reactants or products can be missing.
If a species is mentioned on both sides of the = sign it is
taken to represent a non-reacting third body. The species name
"M" represents a generalized third body. The program treats all
species as ideal gases; however, liquid phase reactions can be
63
-------
can be simulated by use of a (large) effective pressure.
The method integration used is based on the paper "An Implicit
Integration Procedure for Chemical Kinetics", by T.J. Tyson and
J.R. Kliegel, Paper No. 68-180, AIAA 6th Aerospace Sciences
Meeting, (1968). This method of integration, which we call a
"cantilevered implicit method" is especially suited to the inte-
gration of "stiff" equations such as arise when some of the
species react very much faster than some other species as, for
example, in combustion. No special precautions need be taken
when formulating the equations to eliminate nearly equilibrated
reactions as must be done, for example, when using explicit
(predictor-corrector or Runge Kutta) methods. For reaction
systems which are not "stiff" the program described is somewhat
slower than explicit methods but it will often be found useful
because of its convenient input and its general reliability.
Since it is an implicit method, the program must calculate the
partial derivatives of the rate expressions with respect to
temperatures and the concentrations. This is done "analytically"
under the assumption that the reactions are of integral order as
implied by the way in which they are written and that the re-
action rates can be calculated from
rf = af T exp
f"-Ef/RT|
|-Er/RTj
Cf
If less than three reactants or products are involved, the
corresponding concentration in these equations are replaced by
unity.
OUTPUT GENERATED
After printing out the input data the program calculates, prints
64
-------
the temperature, pressure, gas density and species mole fracr
tions as a function of residence time. A time step of variable
size is used in the integration to reduce calculation time. The
program also prints out the forward and reverse rates for each
of the reactions considered at a number of time steps. This
permits one to readily assess the importance of particular re-
actions and to determine which reactions are at equilibrium.
SAMPLE PROBLEM
A sample problem has been supplied to EPA. This problem is the
base case simulation of the flat flame with the results shown
in Table 7.
PROBLEM SETUP
The user will go through five phases in setting up a problem to
be run on REKINET. First he must specify the type and size of
reactor, whether plug flow, series of well mixed stages, or a
well mixed stage followed by a plug flow reactor. Second, he
must specify the kinetic mechanism, components and reactions
with their rate constants. Third, he must specify amount and
type of output. Fourth, he may wish to alter the normal accuracy
criterion and other integration control parameters. The last
phase is program execution.
This program uses a "free-format with control word" type of data
card. Columns 1-6 of each card contain a control word (for
example: VOLUME, EXECUT....) which functions as a machine in-
struction, directing the setting of a set of data or the use of a
particular option. Card order in the data deck is immaterial
(with certain minor and obvious exception, e.g., the EXECUT card
is the last in each data deck). Below we describe briefly the
function of each type of -data card under the five phases of data
preparation. Precise examples of the format of each type of card
are given in Appendix II in which a sample data deck is exhibited.
65
-------
Phase I. Reactor Configuration and Size
WELLSTIRRED
PLUGFLOW
VOLUME
TIMEIN AND
TIMOUT
FLAMEM
FDTEMP
TEMPST
DILUEN
HTLOSS
HEATBALANCE
PDECAY
TDECAY
Calls for a well stirred reactor calculation
Calls for a plug flow reactor following the
well stirred stages
Determines the reciprocal space velocity) of
the well stirred stage
Sets the range of the independent variable
(time or reciprocal space velocity) for the
plug flow reactor
Selects option to allow volume to be adjusted
so that a given fraction X of a species
designated "fuel" is converted, e.g., the
card "FLAMEM, CH4 , 0.02 " designates
that the volume of the well stirred stage
should be such that the flow of CH4 out is
equal to 2% of the flow of CH^ into the
stage.
Designates .feed temperature, i.e., "FDTEMP,
298.0,..."
Designates first guess at outlet temperature
for WSS
Specifies that an unnamed diluent is present
Determines the factors governing heat loss
Selects a heat balanced case as contrasted with
specified temperature and pressure
Sets the pressure or pressure-time profile
Sets the temperature-time profile
Phase II. Defines the Reacting System in Terms of Species and
Reactions
SPECIE
Defines a chemical species and sets its mole
fraction in the feed and estimate of mole
66
-------
fraction in product
ENTHAL Feeds in heat of formation, and specific
heat data
STOICH Defines an elementary chemical reaction
FRATCO Feeds in Arrhenius rate parameters for the
forward direction of a reaction
RRATCO Feeds in Arrhenius rate parameters for the
reverse direction of a reaction
For certain species and reactions the THERMO cards are not
needed. Internally stored data will be used. A reaction rate
library supplies STOICH, FRATCO, and RRATCO cards for a large
number of reactions.
Phase III. Output Control
NPRINT Determines the frequency of detailed output
for plug flow options. Suggested value 10.
PPRINT Determines the frequency of detailed output
for plug flow options.
TYDBUG Determines degree of debug printout called
for. For normal printout omit this card.
LEVEL Determines extent of printout at .each
iteration.
RITEDATA Determines whether diagnostic printout of free
format interpreting routines is desired.
Phase IV. Integration Control
If not placed in the data deck, standard values for these para-
meters will be used.
SSCONT Sets a parameter controlling step size
STEPIN Gives an initial step size. If a value greater
than 10.00.0 is given the program calculates
step size automatically.
67
-------
MAXIT Gives the maximum allowed number of iterations
WTFACT Gives weighting factor for integration of the
heat balanced WSS reactor.
VOLWTF Gives weighting factor for the case where
reactor volume is adjusted.
Phase V. The Execute Card
EXECUT Turns control over to the machine to solve the
problem and prints out the answer.
General! Comments on Data Cards -
* Columns 1-6 are ordinarily for an indentifier word"*****,
TYDBUG, SSCOUT, STEPIN, TIMEIN, TIMOUT, TDECAY, PDECAY,
FRATCO, RRATCO, STOICH, SPECIE, FLAMEM, PLUGFL, WELLST,
VOLUME, NPRINT, PPRINT, LEVEL, WTFACT, VOLWTF, ENTHAL,
MAXIT, HEATBA, or EXECUT"
• The card is punched free format with fields delimited by
commas (or in the case of STOICH cards by + and = also).
Blanks are not significant. Decimal points need not be
punched. Large (or small numbers) may be represented by a
magnitude multiplied by a power of 10., e.g., 123.4E-7.
• Card order is not significant except for the following
point. The EXECUT card must be the last in the deck for
that run.
• Certain "default" variables are built into the program. If
the user does not insert a card certain cards are "under-
stood" to be present. The user should check the program
listing for the default values used.
• Certain identifer cards will cancel out previously desired
options - e.g., NOWELL stirred, NOPLUG flow, NOHEAT balance,
NOFLAME. Thus for several runs using the same deck, the
pertinent identifier word is applicable. Built in default
values set all options to FALSE at the beginning of the
program.
68
-------
• The program has built-in libraries of thermochemical con-
stants derived from JANAF tables. Data cards ENTHAL have
precedence over the internally stored data.
tr
• Species names are left adjusted using the first 6 charac-
ters only and filling in on the right with blanks if there
are fewer than 6 characters. The program does this auto-
matically.
• Reactions are written with, + and = signs to delimit
fields; the = sigh separates the reactants from the pro-
ducts. By convention M represents a general gaseous third
body.
General Comments on Conversion to Other Computer Systems - RE-
KINET is written for a CDC 6400 computer, although versions com-
patible with UNIVAC 1110 systems are available. Conversion to
other computer systems require changes in ENCODE and DECODE
statements in sub-programs DESTOIC and DECIP. The CDC computer
system uses words consisting of 10 characters. Conversion to
systems with words consisting of 6 characters is simple since no
names used need be longer than 6 characters. Conversion to sys-
tems with 4 characters per word (such as IBM 360) will be more
difficult. REKINET is documented internally within the program
listing.
APPENDIX B. SAMPLE DATA DECK
*
A computer listing of a sample data deck is shown on the follow-
ing pages. This sample problem was run and the complete listing
of REKINET and the output produced was sent to EPA. The sample
problem is that of simulating nitrogen oxide formation from a
flat flame.
-------
SAMPLE PR08LEH INVOLVING FLAME MODEL AND PLUG FLOW HEAT BALANCED
SIMULATION OF EKMANNS FLAT FLAME MAX NO CONDITION
»»»»**»»*«*»*»»**»****«»****»*«»**»** «^ »****«***•********•****»*
PLUGFLOW
HEAT8ALANCEDOPTION
WELLSTIRREO STAGE
VOLUME, 1000.0,
VOLWTF,1.0,
FLAMEM,CH4,0.02,
SPEC IE, CH«»,0.i826E-<) 2, 0.0913,
SPECIE, 0 2, Q.<«03E-01, 0.190 8,
SPEC IE»N 2, 0.7 18, 0.7 18,
SPECIE, CO, Q.«t26E-01,
SPECIE, H2,0.139E-01,
SPECIE, 0,0.<»772E-02,
SPEClE»CHO,0.292E-Ot»,
SPECIE, H,0.130E-01,
SPECIE, C02,0.i»59E-01,
SPECIE ,H20,0.153E«-OQ,
SPECIE, OH, 0.816E-02,
SPECIE, H02,0.977E-07,
SPECIE, N20,0.916E-07,
SPECIE,CH20,0.<»8*»E-03,
SPECIE, CH3,0.'»77E-0 3,
SPECIE, N,0.615E-09,
SPECIE,NO,0.«»99E-07,
PPRINT,5,
NP«INT,20,
FDTEMP, 298.0,
TEMPST,17<»1».66<+,
ULTRASYSTEMS SET OF METHANE COMBUSTION KINETICS AFTER ADJUSTMENT
STOICH,ULT 36, CHO fM = CO * H + M,
FRATCO,ULT 36, 2. 50E+20, -1.5, 16. 8,
RRATCO,ULT36, 3. 23E*20 ,-1. 5, -11. 251,
STOICH,ULT77,C02*M=CO+0*M,
FR*TCO,ULT77, l.OOE* 15,0. ,100.0,
RRftTCO,ULT77,2.<»785E*-7,1.0,-27.7955,
STOICH,ULT8«», H20+M = OH + H*M,
FRATCO,ULT8«»,3.00E*15,0.0,105.0,
RRATCO,ULT8<»,15.756E+13,0.0,-14.628,
STOIC H,ULT99,H*0+M=OH*W,
FRATCO,ULT99,8.00E+15,0.0,0.0,
RRATC09ULT99,2.'»0'tE*21,-1.0,105.1«»81,
STOICH,ULT101,H + 02*MI=H02*M,
FRATCO,ULT101,1»50£>15,0.0,1.0,
RR H TCO, UL T 1 01 v 1. 7573E»16 t 0. 0*16.11 5«
STOICH,ULT1«»0,N20+M=N2«-0*M,
FR A TCO,ULT1<»0,1. ODE *•!«», 0.0, 50.0,
RRATCO,ULT1*»0,2.351E*12,0.0,12.789, .
FRATCO,ULT<*<*,3.00E + 1Q,1.0,0.0,
70
-------
RR AT CO,ULTi»i», 15. 963E + 10 1 1.0,76.3058*
STOICH,ULT*»6,CH20«-0=CHO + OH,
FRATCO,ULTl»6,2.0E*ll,1.0,«».'tOO,
RRATCO,ULT«»6,7.1865E*9,1.0,30.«i78,
STOICH,ULTW,CHO*OH=CO»H20,
FRATCO,ULTt»7,3.00E«-10,1.0,0.0,
RRATCO,ULT<»7,7.3837E+11, 1.0,91.577,
STOICH,ULT52,CHO*0=CO«-OH,
FRATCO,ULT52,3.00E«-11,1.Q,0.5,
RRATCO,ULT52,7.126E«-li,1.0,7'».895,
STOICH,ULT59,CH3*0=CH20*H,
FRATCO,ULT59,2.00E*12, 0.5, -0.3,
RRATCOfULT59,2.<»97E*13,0.5,66.711,
STOICH,ULT63,CH«»*0=CH3*OH,
FRATCO,ULT63,1.00E+10, 1.0,8.0,
,RRATCO,ULT63,1.63'*E*8, 1.0,8.0,
- STOIC H,ULT65,CH«»*hi=CH3*-H2,
FRATCO,ULT65,5.00E-HO,1.0,10.0,
STOICH,ULT66,CH«» + OH=
FRATCO,ULT66,3.00E+13,0.,5.0,
RRATCO,ULT66,5.08E«-12,0.0,22.182,
STOICH,ULT7Q,CO+OH=C02+H,
FRATCO,ULT70,5.60E+11,0.0,1.080,
RRATCO,ULT70,7.32E*13,0.0,23.<»23,
STOICH,ULT83,H4-OH=H2*0,
FRATCO,ULT83,8.00E*09,1.0,7.0,
RRATCO,ULT83,18.39E + 9,1.0,9.0<»6,
STOICH,ULT85,H*-H02=OH-»-OH,
FRATCO,ULT85,2.50E+H»,0.0,1.9,
RRATCO,ULT85,23.«»3E*12,0.0,58.607,
STOICH,ULT88,OH>H2=H*H20,
FRATCO,ULT88,2.50E>13,0.0,5.200,
RRATCO,ULT88,11.27E*13,0.0,20.336,
STOICH,ULT91,OH*N=H*-NO,
FRATCO,ULT91,6.00E+11,0.5,8.0,
RRATCO,ULT91,16.79«*E + 11,0.5,56.10<»,
STOICH,ULT98,m-N20=OH + N2,
FRATCO,ULT98,8.00E+13,0.0,15.00,
RRATCO,ULT98,35.557E+11,0.0,80.235,
STOICH,ULT100, OH»0=H*02,
FRATCO,ULT100,2.50E+13,0.0,0.0,
RRATCO,ULT100,3.303E+1*»,0.0,16.067,
STOICH,ULT117,OH«-OH=H20*0,
FRATCO,ULT117,6.00E*-12,0.0,1.00,
RRATCO,ULT117,6.218E*13,0.0,18.182,
STOICH,ULT125,N*NO=N2*0,
FRATCO,ULT125,6.31E«-11,0.5,0.0,
RRATCO,ULT125,28.105E«-11,0.5,75. 190,
STOICH,ULT133,N+02=NO+0,
FRATCO,ULT 133, 6. DOE* 09, 1.0,6.300,
RRATCO,ULT133,12.51E*8,1.0,38.198,
STOICH,ULT135,N20*0=NO»-NO,
71
-------
FRATCO,ULT135,1.00E»l
-------
APPENDIX C. REACTION RATE LIBRARY
The following pages show a computer listing of the reaction rate
library which can be used with program REKINET. The reaction
names are mnemonics derived from literature sources. Thus names
LDSA, LDSB, LDSC etc. denote reactions from the Leeds University
Reports Numbered 1, 2, 3 respectively. A complete bibliography
of the reaction rate coefficient literature used in compiling
this library can be found elsewhere (Sternling and Wendt, 1972).
A copy of the reaction rate library, in punched computer card
form has been sent to EPA.
73
-------
LISTING OF REACTION RATE LIBRARY
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
STOICH,
FRATCO,
STOICH,
FRATCO,
STOICH,
FRATCO,
RRATCO, BROK<»,
EQUILK,BROK<»,
ARCUL1,
ARCUL1,
ARCUL1,
8ROK1,
8ROK1,
8RCK2,
8ROK2,
BROK3f
8ROK3,
BROK<»,
8ROK<»,
N02 + S02
6.31EH2,
J.
= NO
L. HENDT JULY 10, 1973.
«-S03,
O.t
0.,
27.0,
35.980,
OH*H2=H20+H,
2.3E13,
0*H2=OH*H,
<*.OE13,
H+02*M=H02»M,
1.0F.15,
l.«483E»15, 0.0,
0.67M, 0.0,
0.0,
0.0,
0.0,
0.0,
5.2,
16.50,
10.2,
-1.3,
PAGE 79
79
STOICH,
FRATCO,
RRATCO,
8ROK«*A,
8ROK«*A,
8ROK<*A,
EQUlLK,BROK«»A, 0
STOICH,
FRATCO,
STOICH,
FRATCO*
RRATCO,
STOICH,
FRATCO,
RRATCO,
EQUILK,
STOICH,
FRATCO,
STOICH,
FRATCO,
STOICH,
FRATCO,
STOICH,
FRATCO,
STOICH,
FRATCO,
STOICH,
FRATCO,
STOICH,
FRATCO,
STOICH,
FRATCO,
STOICH,
FRATCO,
STOICH,
FRATCO,
STCICH,
FRATCO,
STOICH,
FRATCO,
STOICH,
FRATCO,
STOICH,
8ROK5,
8RCK5,
8ROK6,
BROK6,
BROK6,
BRCK7,
BROK7,
8ROK7,
8ROK7,
BROK8,
BROK8,
8ROK9,
BROK9,
8ROK10,
8ROK10,
BROK11,
8ROK11,
BROK12,
BROK12,
8ROK13,
BROK13,
BRO'KI**,
BROK1«»,
8ROK15,
8RO,K15,
BROK16,
8RO-K16,
8RQK17,
3RQK17,
BROK18,
BROK18,
BROK19,
BROK19,
BROK20,
BROK20,
BROK21,
H*02*H20=H02«-H20,
30.0E*15, 0.0, -1.3,
<».'»503E*16, 0.0, <»5.«*71,
.671*1, 0.0, -«»6.771, PAGE
CO+OH=C02«-H,
6.6E11, 0.0, 1.03,
H+H02-OH4-OH,
7.0E13, 0.0, 0.0,
0.6623E13, 0.0, 39.62,
OH*H02=H20»C2,
6.0E12, 0.0* 0.0,
7.702E+13, 0.0, 72.857,
0.7790E-01, 0.0, -72.857, P
0*HO?=OH*02,
6.0E12, 0.0, 0.0,
0+H20=OH*OH,
8.<*E13, 0.0, 18.0,
H*H02=H2*02,
2.3E13, 0.0, 0.0,
H02*H2 = HOOH«-H,
1.66E1, 0.0, 25.0,
HQOH*M = OH»OH«-M,
3.19E17, 0.0, i»7.0.
H02*H02=HOOH*02,
1.8E12, 0.0, 0.0,
H>HOOH=H20«-OH,
«*.16E1U, 0.0, 9.00,
0*HOOH=OH>H02,
9.0E13, 0.0, 1.00,
OH+HOOH=H2C+H02,
3.6E12, 0.0, 0.0,
CO+02=C02*0,
2.5E12, 0.0, <*8.0,
H2+M=H*H*M,
1.12F13, 0.5, 92.6,
H+OH+M=H20*M,
1.0E19, -1.0, 0.0,
0*0«-M=02 + M,
8.15E18, -1.22, 0.0,
NO+H02=N02*OH,
PAGE 57
74
-------
FRATCO,
STCICH,
FRATCO,
STOICH,
FRATCO,
STOICH,
FRATCO,
STOICH,
FRAYCO,
STOICH,
FRATCOt
STOICH,
FRATCO,
STOICH,
FRATCO,
STOICH,
FRATCO,
STOICH,
FRATCO,
STCICH,
FRATCO,
STOICH,
FRATCO,
STOICH,
FRATCO,
STOICH,
FRATCO,
STCICH,
FRATCO,
STOICH,
FRATCO,
STOICH,
FRATCO,
STOICH,
FRATCO,
STOICH,
FRATCO,
STOICH,
FRATCO,
STOICH,
FRATCO,
STOICH,
FRATCO,
STOICH,
FRATCO,
STOICH,
FRATCO*
STOICH,
FRATCO,
STOICH,
FRATCO,
STOICH,
FRATCO,
STOICH,
BROK21,
BRCK?2,
RROKP2,
I1ROKP3,
BRCK^S,
RROK2<«,
8ROK2<»,
BROK?5,
BROK?5,
BROK26,
BROK26,
BROK27,
BROK27,
BROK28,
BROK28,
3ROK29,
RROK29,
C80W1,
CBOW1,
CBOW2,
CBOW2,
CBOW3,
C80W3,
CBOW«4,
CBOW<«,
CBOW<5,
CeOW«5,
C80W6,
CBOW6,
CBOW7,
C80W7,
CBOW8,
C80W8,
CBOH9,
C80W9,
CBOW10,
CBOW10,
C80W11,
C8CW11,
CBOW12,
CBOW12,
CBOW13,
CBOW13,
CBOW1«»,
CBOW1<»,
DAVI1,
OAVI1,
OAVI2,
DA\/I2,
OAVI3,
OAVI3,
FENJ3,
FENJ3,
GUTM1,
1.0E13,
N02*H=NO*OH,
7.2E1W,
OfNO? = NO«-02,
1.9E13,
Ht-NO*M=HNO»-M,
(*.QE15, 0.0,
H+HNO-H?+NO«
5.0E13,
OH«-HNO=H20*NO,
3.6E13,
0+HNO=OH*NO,
3.0E13,
H02fNO=HNO*02,
1.0E13,
0+NO*M=N02+M,
9.<*E1«»,
H2 f 02 =
2.5E12,
H * 02 =
2.2E1U,
0 *• H2 =
1.7E13,
H «• H20 =
S.«iEll,
0 «• H20 =
C5.8E13,
H «• H *
t.OElfl,
H «• H *
1.5E18,
0*0*
3.0E17,
0*0*
<».OE17,
H * OH *
0.20E20,
H * OH *
0.t«OE20,
H * OH *
-------
FRATCC,
STOICH,
FRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCC,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
GUTM1,
GUTM2,
GUTM2,
HOMERi,
HOMER1,
HOMERI,
HOMER2,
HOMER2,
HOMER2,
JENKI1,
JENKI1,
JENKI1,
JENKI2,
JENKI2,
JENKI2,
JOHN1,
JCHN1,
JOHN1,
JOHN2,
JOHN2,
JOHN2,
JOHNS,
JOHN3,
JOHN3,
JOHNU,
JOHN<»,
JOHN«»,
JOHNS,
JOHN5,
JOHN5,
JOHN6,
JOHK6,
JOHN6,
JOHN7,
JOHN7,
JOHN/,
JOHNS,
JOHNS,
JOHNS,
JOHN9,
JOHN9,
JOHN9,
JOHN10,
JOHN10,
JOHN10,
JOHN11,
JOHN11,
JOHN11,
JOHN12,
JOHN12,
JOHN12,
LANG1,
LANG1,
LANG1,
, -«».59,
N20 * 0 =N2 *
0.ft5El*», 0
H> OH* H20=
1.5E+25,
28.<»3e*25,
60.0
0?,
.0, 2
H20*
-2.6,
-2.6,
H* OH* M= H20* M
7.5E*23,
l.'*21l>E*25,
-2.6,
-2.6
H *S02 *M =HS02 +
7.256E*16,
7.990E+16,
HS02 *OH =H20
0.6789E*1<»,
0.1169EM6,
02*02=0*0«-02,
27.52F18,
11.78E17,
02*ARGON=0«-0*ARGON,
?.51»8E18,
1.276E17,
03*03=0*02*03,
9.938E1U,
16.79E12,
0*03=02*02,
12.0i»6E12,
12.77E12,
03*02=0*02*02,
4.373El«i»
7.388E12,
03*HE=0*02*HE,
i.srgEi't,
5.709E12,
0.0,
0.0,
*S02
0.0,
0.0,
-1.0,
-1.0,
-1.0,
-1.0,
0.0,
0.0,
0.0,
0.0,
0.0,
0.0,
0.0,
0.0,
03*ARGON=0*02*ARGON,
2.«*85E1<4, ~ '
<».198E12,
03*N2=0*02*N2,
3.876£l
-------
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCOt
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STCICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
LANG2,
LANG2,
LANG2,
LANG3,
LANG3,
LANG3,
LANG<*,
LANG'*,
LANGU,
LANG5,
LANG5,
LANG5,
LANG6,
LANG6,
LANG6,
LANG7,
LANG7,
LANG7,
LANG8,
LANG8,
LANGft,
LANG9,
LANG9,
LANG9,
LANG10,
LANG10,
LANG10,
LANG11,
LANG11,
LANG11,
LANG12,
LANG12,
LANG12,
LANG13,
LANG13,
LANG13,
LANG1«»,
LANG1<»,
LANG1«»,
LANG15,
LANG15,
LANG15,
LANG16,
LANG16,
LANG16,
LANG17,
LANG17,
LANG17,
LANG18,
LANG18,
LANG18,
LANG19,
LANG19,
LANG19,
N2 «• H02~"= N0> HNO,
7.9E10, 0.5, 1*1.6,
9.59E11, 0.0, 2.501,
N2 + 0 * M = N20 * M,
1.62E11, 0.0, 3.180,
7.25E12, 0.0, U0,7,
N2 * OH = N20 «• H,
1.18E12, 0.0, 75.8,
3.0E13, 0.0, 10.77,
N2 * 02 = N20 •• 0,
2.«8E1«», 0.0, 107.8,
5.3E1I*, 0.0, 26.7,
N2 * N02 = N20 + NO,
0.0, 8<*.3,
0.0, 50.0,
N *__02 = NO » 0,
r.«»lE13, 0.0, 7.9,
2.95F.12, 0.0, 39.9,
N «• OH = NO * H,
5.3C11, 0.5, 5.62,
9.53E13, 0,0, 55.0,
N20 *• 0 = NO » NO,
6.3E1<», 0.0, 26.7,
1.61E13, 0.0, 6<».5,
N20 + 02 = NO * N02,
6.0E1«», -1.5, 9.9,
2.0E8, 0.0, 3.22,
N + OH = NH «• 0,
1.29E1<», 0.0, 18.0,
1.0E12, 0.5, 0.1,
N » H2 = NH * H,
1.32E15, 0.0, 22.3,
1.0E12, 0.68, 1.9,
N * H20 = NH «• OH,
3.59E15, 0.0, 36.6,
1.6E12, 0.56, 1.5,
NH «• OH = NO * H2,
1.6E12, 0.56, 1.5,
2.22E15, 0.<], 69.6,
NH «• 0 = NO » H,
5.0E11, 0.5, 5.0,
71.0,
2.9,
13.0,
13.0,
7.0,
50.0,
0.7,
0.0,
71.5,
NH * OH = HNO + H,
6. H20,
2.1E12, 0.5,
U.lEli*. 0.5,
77
-------
STOICH, LANG20,
FRATCO, LANG20,
RRATCO, LANG20,
STOICH, LANG21,
FRATCO, LANG21,
RRATCO, LANG21,
STOICH, LOSA 1,
FRATCO, LOSA 1,
RRATCO, LDSA 1,
STOICH, LOSB 1,
FRATCO, LOSB 1,
RRATCO, LOSB 1, 7.
STOICH, LOSB 3,
FRATCO, LOSB 3,
RRATCO, LOSB 3,
STOICH, LOSB 5,
FRATCO, LOSB 5,
RRATCO, LOSB 5,
STOICH, LOSB 7,
FRATCO, LOSB 7,
RRATCO, LDSB 7,
STOICH, LOSC 1,
FRATCO, LOSC 1,
RRATCO, LOSC 1,
STOICH, LOSC 3,
FRATCO, LDSC 3,
RRATCO, LOSC 3,
STOICH, LOSC 5,
FRATCO, LOSC 5,
RRATCO, LDSC 5,
STOICH, LOSC 7,
FRATCO, LOSC 7,
RRATCO, LOSC 7,
STOICH, LOSC 9,
FRATCO, LDSC 9,
RRATCO, LOSC 9,
STOICH, LDSC11,
FRATCO, LOSC11,
RRATCO, LOSC11,
STOICH, LOSO 1,
FRATCO, LDSO 1,
RRATCO, LDSD 1,
STOICH, LDSD 3,
FRATCO, LOSO 3,
RRATCO, LOSD 3,
STOICH, LDSD 5,
FRATCO, LDSO 5,
RRATCCSLOSO 5,
STOICH, LOSO 7,
FRATCO, LOSD 7,
RRATCO, LOSD 7,
STOICH, LDSO 9,
FRATCO, LOSO 9,
RRATCO, LDSO 9,
HNO * H = NO
1.4E13,
9.5E12,
HNO * 0 ~ NO
5.0E11,
9.3E12,
CO* OH=
5.6E*11,
7.29E+13,
H2 +0
1.7 F+13,
3E+12, 0.00,
H2 *OH
2.19 F*13,
8.41 E*13,
H20 »0
5.75 E + 13,
5.75 E*12,
H20 *M
3.4 E+05,
1.17 F*17,
02 »H
2.24 E + 14,
1.3 EH3,
02 *H
1.59 E + 15,
2.4 E*15,
H202 *H
2.34 E«-13,
9.6 E + 12,
H202 *H
3.18 E*14,
5.6 E*13,
H202 *OH
1.00 E+13,
2.8 E*13,
H202 *M
1.17 E+17,
8.4 E*14,
NO *N
3.10 E*13,
1.36 E*14,
02 *N
6.43 E+09,
1.55 E+09,
N +0
3.9 E*15,
****ȣ>** f
N2 *02
»»***£*** ,
*****E+** ,
N20 *0
6.0 E+14,
1.0 F*14,
«• H2,
0.0, 3.0,
0.0, 58.0,
«• OH,
0.5, 0.0,
0.0, 54.5,
C02* H
0.0, 1.08
0.0, 23.
= H
0.00,9.45
7.3
= H20
0.00,5.15
0.00,20.1
= OH
0.00,18.0
0.00, .78
= H
0.00,0.0
0.00,0.0
=0
0.00,16.8
0.00,0.00
f-H
0.00, 1.0
0.00,45.9
= H2
0.00,9.2
0.00,24.0
= H20
0.00,9.0
0.00,77.9
= H20
0.00,1.8
0.00,32.7
= OH
0.00,45.5
0.00,5.3
= N2
0.00, .334
0.00,75.4
= NO
1.00,6.25
1.00,38.64
*M
0.00,*********
0.00,*********
= NO
0.00,*********
0.00,*********
= NO
0.00,26.7
0.00,76.0
0,
410,
*OH
,
»H
t
,
*OH
»
•
*OH
,
,
*OH
,
,
= H02
,
t
»H02
t
,
+ OH
«
,
*H02
,
,
+ OH
t
,
*0
,
,
»0
,
t
= NO
,
•
*NO
*
,
••NO
»
t
78
-------
STOICH,LDS011,
FRATCO,LOS011,
RRATCO,LOS011,
STOICH,LOSE 1«
FRATCO,LOSE 1,
RRATCO,LOSE 1,
STOICH,LOSE 3,
FRATCO,LOSE 3,
RRATCO,LOSE 3,
STOICHtLOSE 5,
FRATCO,LOSE 5,
STOICH,LOSE 6,
FRATCG.LOSE 6,
RRATCO,LOSE 6,
r* T y* T ^> it fcJ
STOICH,
FRATCO»
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCC,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICHt
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO, ncn
STOICH,NEWH9,
f n ft T r* n M c u
MERLl,
MERL1,
MERLl ,
MERLJA,
MERL3A,
MERL1A,
NEWH1,
NEWH1,
NEWHl,
NEWH2,
NEWH?,
NEWH2,
NEWH3,
NEMH3,
NEWH3,
NEWHU,
NEWH5,
NF.WH5,
NEWH6,
NEWH6,
NF.WH6,
NEWH7,
NEWH7,
NEWH7,
NEWH9,
NFWH8,
NEWH8,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
NEWH9,
NEWH9,
NEWH10,
NEWH10,
NEWH10,
NEWH11,
NEWH11,
NEWH11,
NEWH12,
N02
1.1
1.0
N02
1.0
1.0
N02
1.1
*N
E*13,
E«-10,
«-0
E*13,
E*12,
*M
E*16,
••NO
,
*02
1.C5 E*15,
NO
*C
*
+ HV
«-NO
E+09,
E*12,
* 0 *
NO
2.k
k.O
S02
2.UE17,
11.538E*19,
S03 *• 0
2.6E1*,
1«».18E*12,
5,«4E17,
1.5E16,
3.1E15,
7.0E17,
1.989E20,
0.9F15V
= NO
0.00,0.0
0.00,88.0
= NO
0.00,.6
0.00,(*5.5
= NO
0.00,65.0
0.00,1.87
= N02
0.00,****»*»»*,
+02 =N02
0.00,-1.05 ,
0.00,26.9 ,
M = S03 «• M,
0.0, 2.50,
0.0, 83.090,
= S02 «• 02,
0.0, 12.0,
0.0, 1*9.930,
SYMCA-1969-12-60*
0.0,123.6,
0.0, 0.0,
SYMCA-1969-12-601*,
0.0, 110.0,
-1.0, 0.0,
SYMCA-1969-12-60<»,
-2.5, 150.0,
*M
+ N02
<».75«tE17,
1.0E15,
1.82F13,
2.0E16,
02 + M=0«-
3.563E1B,
OH*H=H2*0,
1.UE12,
3.3E12,
OH+0=02*H,
5.5E13,
7.2E1«»,
OH«-H2=H20«-H,
6.2E13,
3.2E1U,
OH*OH=H20*0,
7.7E12,
8.3E13,
CO*OH=C02*H,
0.0, 0.0,
SYMCA-1969-12-60'*,
-1.5, 22*.9,
0.0, 0.0,
SYMCA-1969-12-60«»,
0.0, f>1.0,
O.Q, ?l.<»,
SYMCA-1969-12-60«»,
-1.0, 7<».0,
0.0, 0.0,
SYMCA-1969-12-60<»,
-1.0, 118.0,
0.0, 0.0,
SYHCA-1969-12-60i»,
0.0, 6.0 ,
0.0, 8.00,
SYMCA-1969-12-CO**,
0.0, 1.0,
0.0, 16.9,
SYMCA-1969-12-60«»,
0.0, 6.0,
0.0, 21.1,
SYNCA-1969-12-60«»,
0.0, 1.00,
0.0, 19.1,
SYWCA-1969-12-60<»,
79
-------
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH,
FRATCO,
RRATCO,
STOICH*
FRATCO*
STOICH*
FRATCO,
STOICH,
FRATCO,
STOICH,
FRATCO,
RRATCO*
STOICH*
FRATCO,
NEWH12,
NEWH12,
NEWH13,
NEWH13,
NFWH13,
NEWH1U,
NEWHU,
NEWHld,
NEWH15,
NEWH15,
NEWH15,
NEWH16,
NEWH16,
NEWH16,
NEWH17,
NFWH17,
PATT1,
PATT1,
PATT1,
PATT1,
PYOY16,
PYDY16,
PYDY16,
PYOY29,
PYDY29,
PYOY29,
7.1E12,
NO»0=02*N,
3.2E9,
13.3E9,
NO+N=N2«-0,
1.55E13,
7.0E13,
NO+02=N02*0,
0.18E11,
0.58E11,
NO+NC=N20*-0,
2.6E12,
l.*f2F.l*»,
H*02«-M=H02«-M,
1.3F.15,
CN*CN=C2+N2,
1.6E15,
CN*CN-C2*N2,
1.6E15,
NO +02 »H
3.697E*08,
2.26E+11,
N03 *NO
9.216E*11,
3.9E»11,
0757 7.7,
0.0, 27.25,
SYMCA-1969-12-60U,
1.0, 39.10,
1.0, 7.08,
SYMCA-1969-12-60<»,
0.0, 0.0,
0.0, 75.50,
SYHCA-1969-12-60<*,
0.5, <*7.0,
0.5, 0.0,
SYMCA-1969-12-60d,
0.0, 63.8,
0.0, 28.0,
SYMCA-1969-12-60U,
0.0, 0.0,
JCPSA-1962-36-ll<»6
0.0,
-------
FRATCO,SBOW10,1.0E*i<», O.,0.,
RRATCO,SBOW10,2.i»5E*15, 0.,91.590,
STOICH,SBOW11,CO+OH=C02»H,
FRATCO,SBOW11,3.1E + U, O.,0.59«»,
RRATCO,SBOWll,<».0<»e«-13, fl.,22.92'*,
STOICH,SBOW12,H«-02=0*OH,
FRATCO,S80W12,2.2E*1<», O.,16,<»5<»,
RRATCO,SBOW12,0.167E«-1<», O.,0.38«»,
STOICH»SBOWi3t04-H2-H+OHt
FRATCO,SBOW13,<».OE*1<», 0.,9.365,
RRATCO,SBOW13,1.762E»l<»t Q.,7.335,
STOICH,SBOW11»,0*H2Q=OH+OH ,
FRATCO»SBOWl«»,8,««E»l<»t 0., 18. 05 8,
RRATCO«SBOW1<«,0.82QE«-1(»« 0.,0.8 88,
STOICH,SBOH15,H*H20=H2*OH,
FRATCO,SBOW15,l.OE*i'», 0., 20.196,
RRATCO,SBOW15,0,22E*1«», Q.,5.056,
STOICH,SBOW16,H»OH*H=H20*-H,
FRATCO,SBOW16,2.00E*19, -1.0,0,
RRATCO,SBOW16,3.7qE*20, -1.0,119.630,
STOICHtSBOW18,CHO+M=H+CO*M,
FRATCO,SBOW18,2.0E»13, 0.5,28.512,
RRATCQ,SBOW18,2.58E*13, 0.5,0.<»72,
STOICH, SEMN03, N03 *S02 =S03 *N02,
FRATCO, SEMN03, 0.5325Et-13, 0.0, 2.050,
RRATCO, SEMN03, 0.1365E*!'*, 0.0, 35.530,
STOICH,WILO 1,H2*NC-HNO*H,
FRATCO,WILD 1,1.*»E*13, 0.0,5
-------
FRATCO,WIL013,1.3E+13, 0.0,9.<»,
RRATCO,WIL013,1.5E*li , 0.0,6.95,
STOICH,WIL01«*,H20*0=OH*OH,
FRATCO,WIL01<«,9.2E«-13, 0.0,18.0,
RRATCO,WIL01<«,7.6E*12, 0.0,1.0,
STOICH,WIL015,H2+M-H+H«-M,
FRATCO,WIL015,«».2E*19, -0. 8<»f 103.2,
RRATCO,WIL015,5.0E*18, -1.0,0.0,
STOICH,WIL017,H+OH»M=H20+M,
FRATCO,WIL017,1.8E>22, -1.5,0.0,
PRATCO,WIL017,1.0E + 2'*, -1. 3^, 118. 0,
STOICH,WIL020,0+NO=Nf02,
FRATCO,WILD20,3.2E*09, 1.0,19.1,
RRArCO,WIL020,1.6E*10, 1.0,7.2,
STOICH,WIL022,N«-NO=N2«-0,
FRATCO,WTL022,1.5E+13, 0.0,0.0,
RRATCO«WILD22,6.9E»13, 0.0,75.25,
STOICH»W!L02«i,H«-02 = OH + 0,
FRATCO,WIL02««,9.5E*13, 0.0«l«i.7,
RRATCO,WIL02<«,2.2E*12, 0.0,0.0,
82
-------
APPENDIX D. THERMOCHEMICAL LIBRARY
A listing of the thermochemical library can be found on the fol-
lowing pages. These data are presently stored internally in
REKINET. A punched copy has been sent to EPA for their use.
Each card has the following format:
ENTHAL, [name of species], [AHf], [AGP], [BCP]
[CCP], [DCP].
where AHf is the standard heat of formation of the species and
the specific heat of the species, Cp, has the following depen-
dence on temperature:
C = ACP + BCP'log T + CCP«T + DCP»T2
p e
The coefficients ACP, BCP, CCP and DCP were derived from a
least squares fit through JANAF data. Errors of up to 8% in Cp
are possible, as a result of this, although in general the
errors are much smaller.
83
-------
LIBRARY
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
, .52117^*01
,.197E»C1
,.56979«,E-01
Q»2,.73«700F»Ol
HPn,,300E*02
C, . 1701Hf.F.»0^. .l»7053i,E»Ot
CCL20,-.52eoaOF*T2,-.230<»f>KE*12,.67723<.E»01
CCL"«,-.22a<.aOE*02,-.2fl9<.23F*02,.91<.17i,E*01
C2N2,.738700E»02 ,-.110718F.»02,.l.35U7ofr«.oi
CH2,.95000 OE*02 ,-. 21086 5F»01,.1<»9«,33E»01
OCT 2,197*. JOLH UNIV
,-.60<,309P*00..1f>10f)5F.-02 ,-.293159E-06,
,.623262E*00 ,-.691408E-03, . 12U55 7E-OF. .
,-.195e-02 .^lUE-Ob ,
,-.199938F.-03,.67ai61F_-07 ,
,-.15996<»E-02,. 218320E-06 .
,-.596f>UE-02,.738r<«6F-06 ,
,-.106525E-01,,157862E-05 ,
,-.598n51E-03,-.202233E-06,
,.370539E-02 , -. 8«,0595E-Of>,
ENTHAL,
ENTHAL,CH2CL2,-.228300F*02,-.360501E*02,.856319E»01
ENTHAL, CH3,.319<«OOE»02 ,-.
ENTHAL,
ENTHAL,
ENTHAL, CH30,.350E*')1
ENTHAL, CH302,.670E»01
ENTHAL,CH302H,-.313E*02
ENTHAL, CN,.111000E+03
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
. 195 038E + 01
,.23<»033Et01
,0.0
,0.0
,0.0
,.26781«E»00
,.1<*QE*02
,.lf>QE»02
,.180E»02
,.501157E»01
CO,-.26<«170E*02,.7«08'»1E«-01
COCL,-. 150000F»02,-.l 1.8 362E*01,.217928E»01
C02.-.9',05<,OF*02,-.1<<790<»E + 02,.<.21328E»01
COS,-.330100E*02,-.m3695E»02,.«.39556E»Ol
C.P, 0.0 ,-.209382E*U2,.<»17905E»01
CS,.550000E*0? ,-.19«.268E*01, . 162157E+01
CS2,.27980(lE*02 ,-. 13393OE*02, .<•<»<.77IE*01
CL,.2892?OF*02 ,. f>29998E»00 ,.905357£»00
CL2, 0.0 ,.303189E»nO , .l<»609i»E»01
CL20,.210flOQE+02
CLO,.2Ut923F.»Q2
CL02,.250003E*02
,.119283F.-01
,0.0
,0.0
,0.0
,.116999E-02
-.288039E*00,.2598<»1E-02
,-.226227E-02,-.120776E-OS.
,.705692E-02 ,-.160Q98E-Q5,
.-.123865E-05,
,-.260<,93E-05,
,0.0 ,
,0.0 ,
,0.0 ,
,-.901735E-07,
,-.528c510E-06,
,-.6«»it6'»5E-03,-.<438559E-07 ,
,-. 133861E-02,-.273210E-07,
,-.2628<»2F-02,.2i»7«.69E-0& ,
,-.310f>03E-02,.3U913E-Of> ,
,-.809223E-03,.3793<«9E-07 ,
,-.37«»760E-Q2,.'»<*761',E-06 ,
,-.187911E-02,.331785E-06 ,
,-.171363E-02,.2fl967i»E-06 ,
,-.i»61263E-02,.688258E-06 ,
,-.195125E-02,.?S3567E-06 ,
,-.i,53i,77E-02,.b06019E-Of) ,
,-.31372'*E»01,.197288E*01
,-.167319F.*02,.<492335E»01
H, .521000E*02,.<»9638QE»01 , .788887E-03,-.150707E-05,.297103E-09 ,
H9R ,-.fl7!300E*1l, .110257E»02 ,-,887<»<«5E»00, . 3«»21<,3E-0 2 ,-. 615252E-06 ,
HCL,-.220630F*T2,.123366F*02 ,-.111960c*01,.355006E-02
HCN,.312000F.»02 ,-. <431i«<»9E*01, . 217 006E »01 ,.195531E-02
HCO,-.29COOOE«01,.97q512E*00 ,.109102E»01 ,.3U3380E-02
HCOCL,-.'«01E*02 ,.891E*«1 ,0.0 ,0.0
H2, Q.O ,.107652E»02 , -. 766202E*00 ,. 19<»39i«E-02
H20,-.577qiOF*02,.li.<.689E*02 , -. 1<*5253E + Q I, .6i»i.539E-02
H202,-.32=:?aOE*02,-.230056E*02,.609937E*01 ,-.52560 i«E-02,.llf><,08e-05 ,
H2S,-.<«8ftn30F.*01,.73935<»E»01 , - . 16<»<»i«5E*00 , . 57 635 OF-02 ,-. 11353UE-35,
,-.59«930F.-06,
,-.U569:UE-06.
,-.79n333E-06,
,0.0 ,
.-.163972E-06,
,-.102179£-05,
ENTHAL,
ENTHAL, H2SO<*.,-.19i45<4flE»03,-. 713213F»02,.:
ENTHAL, HNO,.23fltiOOE»02 ,-.2fU905E*01..179998E*01
fNTHAL,HN02.C.-.11J'»30r»C2,-.229669E»02,.bni»5l«E»01
ENTHAL, HN 0?. T, -. 1 fifl'. nOF»0?,-.2?l 77 5E»Q2, . 59i»S l«.E* 0 t
,-.281095E-t)l,.5l523lE-Q5 ,
,.233806E-02 ,-. f>?96fl 7F-06 i
,-.23<»361F-02,.8<.lJR5F-07 ,
,-.2«»807(lF-02,.135233F.-Of> ,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
FNTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
HN03,-.32inoOE»02.-.<»630«flE*02,.10MU2E»n2
HO CL,-.2?COOOE»02.-.'«f2196E»01,.23799 7F»Ot
H02, .50n30nF»ni,-.i«U792E»01,. 21131 3E»01
HS02,-.b87F.*02
HSn3,-.Ul6E*02
,-.151512E-03,-.6i«fi5i»sr-07,
,.12310tE-02 ,-.371379E-06,
NCO,.225F»02
NH,.810000E*02
Nh2,.<.00700E»02
NH3,-.109700E*02,
N2, 0.0
N2H,.6<*OF*02
N2H2,.509003E»02
V
1
,
t
*
,
1
9
t
•
.179E»02
-.713E»02
,i489709E»01
-. li«OF.»02
,129091E»02
.12?3<»2E*02
,207f,9i.E»01
.. 9?9*?'2E*01
.123E»02
-.132358E*02
.0
* •
, .
* •
» ™
t-
t •
f "
,0
, .
.0
197E»02
139I.79E-01
i»07e»01
.121565E»01
,118608F*0 t
719233E*00
.585<»90E»00
.0
362857E*01
,0
* ~
,-
f *
• •
, .
f .
1 •
*0
f •
•
•
.
•
3
f,
7
0
281E-01
3(47^9<«E-0<
17UF-02
5220f.E-02
73b4HF-02
89556F.-02
*
*
*»
,
*
,
t
0
•
.
.
-
-
-
.
5
9
1
•
.
•
300731E-02 .-.
•
0
397769E-02
•
»
0
-
•
•
0
15E-05
01619E-03
15E-07
556300E-06
12?138E-05
150331F-05
5«li»88E-06
0
106278E-05
t
»
,
»
V
»
f
V
*
t
N2H3,.365E»02
N2HU,.227900E»02
N20,.196100E»02
N202,.«»Q7E»02
N203,.198000E*02
N20<«,.271000E*ni
N205,.270000F*Ol
NO,.215101^*02
N02,.79100TE*dl
N03,.
,0.0
,0.0
..765528E-03
,-.171906E-02,.262810E-07
,0.0 ,0.0
,-.'»96116F-02,.:
,.215E»02 . ,0.0
,-.39617<.F»02,.90<«839E»01
,-.l'«2978E*02, .i»2086=E»01
,.190E»02 ,0.0
,-.296303E*02,.818'»?<;E»01
,-.539296E*02,.131631E»02
,-.72933",E*02,.177761E»02 , -. 1851S8E-01, . 25151QE-0 5
,.6659&i»F*01 ,-.392199E-01,.21i»785e-02 ,-.
,-.13920.7E*02,.'»05P5.3E»01
3E»!)2,.110<«52F.»02 ,-. 10753UE-01, . 1«,0360E-0 5
84
-------
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
EHTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
ENTHAL,
, - .. »00 ,
OH, . 9<»3?0 OF* G I , . l<.9«i99E + 02 , -. 15'«237F*0 1 ,
02, 0.0 ,.72<3132E>00 , . 10 S269E *0l ,
S, 0.0 ,-.m2501E»02,.<4<»20Rfie»01 ,
S.C, (1.0 , . 275552E*01 , . U7<«05E »00 ,
SH,.3<«60aOE»02 ,.17711«E*0? , -. l97f,»«i»E*Ql
S0,.16<«nnor»01 ,
515051E-0 3 ,
35U289E-0 2 ,
122101F.-0 J ,
.139737E-01,
663617E-02 ,
<«5 3U2UE-02 ,
. 7<«971.«<:-07 ,
.510573?-BS ,
. 37«67SE-07 .
31691SE-OS ,
-0 7 ,
IE-Q6 ,
7E-Q6 ,
,- .3200<»6E-02,
S2,.308<«OOE»02 ,-. 193i«96E»01,
S20,-.13SOOOE»02,-.13110RE»02,
Sfl,.2'«2QOOE*02 ,-. 20652^*02, . 10912FE + 02 ,
.979512E»00, .109102E»Ol,
-. 1121«7E*02, . 31197<»E»0 1 ,
799262E-06
CHO,-,29QOE*01
CH20,-.2770E»02
. 1979«<4F.-02,
.i«36175E-02,
.136562E-01,
.3U3380E-02,
.5821'»9E-02,
$81790E-06 ,
21120 UE-0 5 ,
. 790933E-06 ,
85
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/2-75-075
3. RECIPIENT'S ACCESSION-NO.
4. TITLE ANDSUBTITLE
Effect of Fuel Sulfur on NOx Emissions from
Premixed Flames
5. REPORT DATE
October 1975
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO
J.O.L. WendtandJ.M. Ekmann
9. PERFORMING ORGANIZATION NAME AND ADDRESS
University of Arizona
Department of Chemical Engineering
Tucson, Arizona 85721
10. PROGRAM ELEMENT NO.
1AB014; ROAP 21ADG-021
11. CONTRACT/GRANT NO.
Grant R-802204
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final; 4/73-7/75
14. SPONSORING AGENCY CODE
5. SUPPLEMENTARY NOTES
6. ABSTRACT,
The report gives results of an investigation of the effect of fuel sulfur com-
pounds on nitrogen oxides (NOx) emissions from premixed gaseous flames. Labora-
tory measurements, using a methane/air flat flame doped with SO2 or H2S, showed
that fuel sulfur inhibits the formation of NOx arising from thermal fixation. This
inhibition was significant at all air/fuel ratios and especially at high air preheats.
The effect of fuel sulfur on formation of NOx arising from fuel nitrogen oxidation
is less clear because of complex reactions between sulfur- and nitrogen-containing
species in both the flame and the sampling probe. A mathematical simulation of a
flat flame was developed that showed that the observed effect of fuel sulfur on
thermal NO' could be explained by a kinetic mechanism involving the catalysis of
atom recombination reactions by SO2. The experimental and theoretical results
may be especially significant from a practical point of view, since they appear to
indicate that fuel desulfurization may lead to increased NOx emissions.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Air Pollution
Reaction Kinetics
Combustion
Fuel
Desulfurization
Pollutants
Interactions
Mathematical Models
Nitrogen Oxides
Sulfur Oxides
Atomic Structure
Air Pollution Control
Stationary Sources
Synergisms
Atom Recombinations
Superequilibrium
Concentrations
13B
07D
21B
2 ID
07A
12A
07B
20H
. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (This Report/
Unclassified
21. NO. OF PAGES
91
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
86
------- |