EPA-600/2-77-008C
JANUARY 1977
:nvironinefiiai Protecicn leclinoioay Series
NOv FORMATION IN CO FLAMES
Industrial Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
-------
RESEARCH REPORTING SERIES
Ftesearch reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into five series. These five broad
calegories 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.
Thu 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
wo'k 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 policy of the Agency, nor does mention ol trade
names or commercial products constitute endorsement or
re :ommendation for use.
Thiu document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/2-77-008c
January 1977
NO FORMATION
X
IN CO FLAMES
by
E. L. Merryman and A. Levy
Battelle-Columbus Laboratories
505 King Avenue
Columbus, Ohio 43201
Contract No. 68-02-0262
ROAPNo. 21ADG-020
Program Element No. 1AB014
EPA Project Officer: W. Steven 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
-------
ABSTRACT
An experimental study was undertaken to determine whether early NO and
NO- could be observed in CO flames. Earlier studies have suggested that
prompt NO is not anticipated in CO flames; also HO- levels might be expected
to be lower in CO flames. Previous studies of the production of NO and NO-
in methane flames, with and without fuel nitrogen, suggest that the early
appearance of NO- comes from the fast reaction of NO + HO- = N0~ + OH.
CO flames containing NO, ammonia, and cyanogen were profiled for NO and N0?
in this study. Effects of flame temperature and of nitrogen- and argon-
"air" were also examined. The results give evidence for prompt NO in CO
flames as well as for early N0?. The prompt NO and early N02 levels are
less than in methane flames, but the same mechanism appears to explain the
formation of NO- in both flame systems. In most instances the quantity of
NO or fuel-nitrogen added to the flames can be accounted for as NO, NO-,
and an unidentified nitrogen species. Analysis of the rates of depletion of
ammonia and cyanogen in the CO flames yield Arrhenius coefficients equiva-
lent to 58 and 41 kcal/mole, respectively.
This report was submitted in partial fulfillment of Contract No.
68-02-0262 by Battelle's Columbus Laboratories under the sponsorship of
the Environmental Protection Agency. Work under this phase was completed
as of June 30, 1975.
iii
-------
CONTENTS
Abstract ill
Figures v
Tables v
Acknowledgments vi
1. Introduction 1
CO flame kinetics 2
2. Experimental Procedures 5
3. Results 6
CO flames - no additives 6
CO flames with additives 9
Summary of results - flames with additives 13
A. Discussion 15
Prompt NO in additive-free CO flames 15
Fuel-N mechanism 16
Postflame equilibrium considerations 21
Kinetics of C N and NH depletion 22
511 fc t J f\ e
Summary 25
References 28
Appendices
A. '.?lame profiles 31
B. Arrhenius plots 45
iv
-------
FIGURES
Number Page
1 Extrapolation of postflame NO data to zero time 10
TABLES
Number Page
1 Nitrogen Oxide Data from CO Flames 7
-------
ACKNOWLEDGMENTS
Special acknowledgment is made to the useful discussions and assistance
of Dr. Robert W. Coutant during the course of this program. The authors
also acknowledge the interest and discussions of Mr. W. S. Lanier, Project
Officer, and Mr. G. B. Martin of the Environmental Protection Agency,
IERL, Research Triangle Park, North Carolina.
vi
-------
SECTION I
INTRODUCTION
It is well recognized today that there are two sources of nitrogen
oxides (NO ), namely that produced by thermal fixation processes via the
X
Zeldovich mechanism, and that produced by oxidation of organic nitrogen
compounds in fuel (fuel-bound nitrogen). Recent studies by the authors and
by other investigators have shown that fuel-bound nitrogen yields high and
varying amounts of NO in combustion product gases (1-7)*. More recently,
the authors have shown that NO and NO- are produced in flames by a mechanism
that differs from both the Zeldovich and the general fuel-bound nitrogen
mechanisms, and that this new mechanism may be important in the development
of other NO mechanisms (8). One of the objectives of the present study
X
has been to determine how well and to what extent this new mechanism lends
itself to CO flames.
The Zeldovich mechanism as used to explain the formation of NO in
post-flame combustion processes is as follows:
0 + N= NO + N (1)
N + 0 = NO + 0 (2)
0+0+M=0_+M (3)
N + OH = NO + H. (4)
More recently, however, Fenimore (9) has indicated that NO can be formed
early and rapidly in hydrocarbon flame systems to produce what he has
termed "prompt" NO. In hydrocarbon flames, he suggests that the early NO
can arise from oxidation of species such as CN and HCN produced in the
flames. It follows from-these same arguments that fuel-nitrogen compounds
* References are listed on page 28.
-------
will yield NH and CN species, i.e., N, NH, NH , CN, HCN, etc., which will
be oxidized to NO in flames.
Ir. the recent flame studies at Battelle-Columbus it has been observed
that NO and NO,, are produced early in methane-air flames, with and without
fuel-bound nitrogen additives, and that this NO may be different from
X
Fenimore's "prompt" NO. In some instances, N02 has been observed in the
visible, flame region prior to the appearance of NO. A mechanism has been
developed which suggests that NO produced in the flame process is scavenged
by HO radicals, producing "early NO ". This early N0_ is then converted
to NO via reactions with excess 0-atoms in the flame- The mechanism used
to expj.ain this phenomenom is as follows,
NH, CN + 0 = NO + OH, CO (5)
NO + H02 - N02 + OH (6)
N02 + 0 - NO + 02. (7)
In order for HO- to scavenge NO in the visible flame this mechanism requires
that Reaction 6 have a fairly high rate constant and the concentrations of
HO- be of the order of 10-50 ppm in the flame. Analysis of conditions in
methane flames has indicated that the oxidation of formaldehyde and CHO
radicals could yield the necessary level of H0_ radicals. A principal
question in the current study concerned whether or not the level of HO
radicals would be high enough to convert NO to N02 in CO flames.
CO FLAME KINETICS
Tie oxidation mechanism of CO in flames is well established to proceed
as folLows,
CO + OH = C02 + H (8)
0 + H2 = OH + H (9)
H + 0 = OH + 0 (10)
0 + H20 = OH + OH (11)
H + H20 = H2 + OH. (12)
-------
One cannot present a CO oxidation mechanism without including oxidation
reactions for hydrogen since it is well known that the direct oxidation of
CO by oxygen
CO + 02 = C02 + 0 (13)
is slow, if it goes at all (10).
In accord with the interests of this study is the role of the hydro-
peroxyl radical. The HO. has often been excluded from oxidation mechanisms,
as noted above, on the basis of its being of minor significance at flame
temperatures. More recent studies, however, indicate that the HO- radical
can play a significant role in hydrocarbon oxidation (11). The current
issue is, how important is the H0_ radical in NO- production from CO flames?
In hydrocarbon flames, HO- will result from the extraction of hydrogen
from methane, from aldehydes and from other intermediates produced in the
oxidation processes,
CH4 + 02 = CH3 + H02 (14)
CH20 + 02 = CHO + H02 (15)
CHO + 02 = CO + H02. (16)
On the basis that formaldehyde and subsequent CHO radical levels are less
in CO or CO-H- flames, it may well be that early NO- and prompt NO would
also be significantly less in CO or CO-H flames. (Fenimore reports no
prompt NO in CO or H_ flames.)(9)
In the present study, the CO flames were stabilized by adding water
vapor to the cool preflame gases. In water-stabilized CO flames any HO
radicals must form either directly or indirectly from the added water
vapor. One might consider the production of HO radicals by extraction
of hydrogen from the water molecules,
H2° + °2 = H°2 + °H> (17)
This reaction, however, requires about 72.7 kcal of enthalpy at 1000 K.
Equilibrium calculations show that, at most, only a few ppb HO would exist
3
-------
at this temperature. A more likely reaction producing H0_ is obtained
indirectly from the hydrogen atoms derived from Reaction 8. Thus, the
three-body reaction
H + 02 + M = H02 + M, (18)
could occur in the 1 atmosphere CO flames leading to several ppm of H0«.
For example, a flame containing only 3 percent 0~ and a few ppm H at 1400 K
could at equilibrium have as much as 35 ppm H0_ present.
-------
SECTION II
EXPERIMENTAL PROCEDURES
Flat, disc-shaped laminar flow CO flames were probed using the micro-
probing technique described in previous studies (12). The flames were
stabilized at 1 atmosphere pressure on 1-inch and 2-1/2-inch-diameter
stainless steel burners, the smaller burner providing higher flame
temperatures than the larger one. Fuel-rich, stoichiometric and 0_-rich
flames were probed using nitrogen or argon as a diluent. Cyanogen (C_N«),
ammonia (NH-), and nitric oxide (NO) were i
ppm to the cold gas (CO-O.-N^/Ar) mixtures,
ammonia (NH-), and nitric oxide (NO) were added at concentrations of 15-350
A small 1/4-inch-I.D. water-cooled quartz probe with a 100-micron
orifice in the tip was used to withdraw samples from below, in and above
the flames. The pressure in the probe was 2-4 torr. The sample line from
the probe was split to provide simultaneous sampling for a chemiluminescence
instrument and a mass spectrometer. NO, NO., and the fuel-N additives were
analyzed in the chemiluminescence instrument, using stainless steel and
carbon converters. A quadrupole mass spectrometer was used to analyze
for CO, CO , 0_, N_, Ar, and H20. The C-N depletion was also followed on
the mass spectrometer. The conversion of C»N_ and to a lesser extent the
ammonia to NO in the stainless steel converter was not as complete as
desired. Thus, these data are less reliable. Some of these fuel-N profiles
were also generated by the mass spectrometer.
Flame temperatures were recorded for each flame composition used.
Silca-coated Pt - PT + 10 percent Rh thermocouples (2-5 mils dia.) were
used in the temperature probings.
-------
SECTION III
RESULTS
Ths results presented here are drawn primarily from the profiles of
NO, N0_, and fuel-N for the various flame systems. A complete set of these
profiles for all flames probed is presented in Appendix Ao
Taale 1 lists the pertinent data for the 24 flames probed in this
study. In the numbering system used, the odd number flames were O.-rich
flames and the even number flames were fuel-rich or near stoichiometric
flames. The NO and NO data are given in terms of ppm conversion of fuel-
N to NO and NO .
The smaller 1-inch-diameter flames are the hotter flames and are
referred to as the Hi-temp flames; the 2-1/2-inch-diameter flames are the
lower temperature flames and are referred to as the Lo-temp .flames.
Temperature differences between the two flames ranged from 215-260 C. Cold
gas velocities ranged from 13.3-17.2 cm/sec in the Hi-temp flames and from
5.8-7.!) cm/sec in the Lo-temp flames.
Complete oxidation of CO to C02 occurred only in the oxygen-rich
flames., Some CO remained unoxidized in the post flame gases of the stoichio-
metric and fuel-rich flames. Consequently, these latter flames had lower
maximum flame temperatures than did the corresponding 0«-rich flames.
CO FLAMES - NO ADDITIVES
Initially, additive-free CO flames containing argon as a diluent were
probed to determine the extent of N-containing impurities in the original
composition gases. No NO or NO, was observed in these flames, indicating
-------
TABLE 1. NITROGEN OXIDE DATA FROM CO FLAMES
Flame Flame Composition (mole fraction)
No. CO 0, H20 N2
Ar
Max
Temp,
K
Additive
Species
ppro
Maximum Observed
Conversion of
Fuel-N to
NO (ppm)
NO (ppm)
NO, Consumed
in Postflame,
percent
1-Inch-Diameter Burner (High-Temperature Flames)
1
2
3
4
5
6
7
8
9
10
11
12
13
15
16
17
IS
19
20
21
22
190
202
187
199
187
193
183
193
160
168
162
162
160
168
.155
.104
.154
.102
.153
.099
.156
.099
.131
.087
.133
.083
.131
.087
.017
.018
.017
.018
.017
.033
.039
.033
.015
.011
.015
.014
.015
.011
.638
.676
.642
.681
.643
.675
.622
.675
.217
.238
.158
.168
.113
.117
.113
.117
.113
.117
.135
.085
.117
.073
.082
.051
.082
.051
.082
.051
.014
.010
.014
.010
:020
.014
.020
.014
.017
.014
638
676
642
681
643
675
622
675
.694
.734
.690
.741
.694
.734
1609
1516
1609
1516
1609
<1516
<1609
<1516 '
1593
1509
1593
1509
1593
1517
2-1/2-Inch-Diameter Burner
634
667
711
749
..
..
~
.785
.818
.785
.818
.788
.818
1582
1422
1370
1296
1348
1271
1348
1271
1348
1271
None
None
NO
NO
NH
SH*
C2N2
C2N2
NO
NO
NH,
NH,
C2N2
C N
C2N2
35
36
43
1 32
125
165
92
71
65
62
300
350
12.5
32
50
61
53
37
260
295
113
61
56
86
565
690
(Low-Temperature Flames)
None
None
NO
NO
NO
NO
NH
NH,
C N
C0N,
._
48
34
36
64
35
36
122
129
4.3
5.6
34
25
31
49
35
39
125
120
0
4
17
16
3.0
1.6
4.0
2.5
14 *
15
28
14
9
5
4.3
3.8
34
18
28
23
30
12
13
9
0
100
100
100
100
100
100
100
100
ICO
75
72
100
100
65
27
86
100
83
100
83
100
81
100
-------
essentially nitrogen-free cold gas mixtures. Substituting N« for argon in
the additive-free flames resulted in significant nitrogen oxide formation.
These flames, Flames 1, 2, 15, and 16, provide baseline data for the Hi-
and Lo-temp CO flames probed in this study. As seen in the data, all of
the flames with N? as a diluent produced NO and all but one of these flames
produced small but measurable quantities of N0«. The low-temperature
flames, Flames 15 and 16, although at a higher temperature than Shaw and
Thomas' additive-free CO flames (13)(which were <1100 K) yielded about the
same NO levels:
x
Shaw and Thomas 6-12 ppm
This work 9-10 ppm.
Reducing the oxygen content of the Hi-temp flame to near stoichiometric
conditions produced about 4 ppm N0? early in the flame reactions, but the
N02 was subsequently consumed in the postflame region leaving only NO in
the emission gases as shown in the Flame 2 data. Thus, in both 02-rich and
near stoichiometric Hi-temp CO flames without additives only NO is present
in the postflame gases. The stoichiometric flame produced more than twice
as much NO as did the oxygen-rich flame.
In the additive-free Lo-temp flames, 15 and 16, the formation of both
NO and N0_ was observed in small but comparable quantities. The N0» was
formed early in the preflame region and peaked in the early postflame
zone. (In these discussions "flame" refers to "visible flame".) Consump-
tion of NO began in the postflame gases but did not go to completion in
either the 0_-rich or fuel-rich flames. Appreciable amounts of NO are
formed only after the NO- has peaked, as was also observed in the stoichio-
metric Hi-temp flame. Also similar to the Hi-temp flames is the production
of larger quantities of NO in the Lo-temp flames under reduced oxygen
conditions. In contrast to the Hi-temp flames, however, both NO and NO.
are present in the postflame gases of the Lo-temp flames, although
concentrations are low. As one might expect, the rates of formation and
depletion of NO and N0« are somewhat slower in the cooler flames than in
the hotter flames.
-------
Post flame NO - Zeldovich Mechanism
The NO data for additive-free CO and CH, flames are plotted in Figure
1 as a function of time. Data from the CH, flame are included here only
for comparison. The postflame NO values have been extrapolated to zero
time which, as discussed later, may give some indication of the contribu-
tion of non-Zeldovich processes to NO formation in the flames. Extrapolation
of the postflame NO data in both the CO and CH. flames produced positive
intercepts. It is quite obvious, however, that the intercept value is much
higher in the CH, flames, as is also the initial rate of NO formation. The
positive intercepts here suggest the presence of prompt NO in both flame
systems. The presence of prompt NO in CO flames is contrary to Fenimore's
original observations.
CO FLAMES WITH ADDITIVES
The three additives, NO, NH_, or C-N., were selected for their
different bonding characteristics of the N-atoms. Each additive has been
examined in terms of its effect on nitrogen oxide chemistry in flames.
NO Addition
Previous studies of hydrocarbon flames using NO as an additive indi-
cated the presence of a species in the region of the visible flame which
readily oxidized NO to NO-. Nearly quantitative conversion of N0? was
observed. NO was added to the preflame gases of C0-0? flames to determine
if these same conditions also exist in CO flames. The data are presented
in Flames 3, 4, 9, 10, and 17 through 20 in the appendix. The NO added to
these flames ranged from about 34 to 92 ppm.
The profiles for these flames indicate that the NO added to the CO
flames is depleted to various degrees as it approaches the hot visible
flame region; the extent of depletion depends on the flame condition, and
is generally greater in the Lo-temp flames than in the Hi-temp flames.
The rate of consumption of NO is slower in the Lo-temp flames as seen in
comparing the data (Flames 3, 4, 9, 10 with Flames 17 through 20). This,
and the greater extent of depletion in the Lo-temp flames is likely due to
-------
12
10
NO (CH. flame)
NO (CO flame)
10 15 20 25 30
TIME, msec.
Figure 1. Extrapolation of postflame NO data to zero time
10
-------
the lower temperatures and longer residence times encountered in these
£lames.
The important observations in the flames with added NO are that in all
cases the NO is consumed as it approaches the hot visible flame and that
this consumption is followed by a corresponding formation of NO.. The NO-
thus formed reaches a peak value coincident with the minimum in the NO
curve. The N0_ is then consumed to varying degrees in the postflame region
with subsequent reformation of the NO. All of this points to direct
formation of NO. from the added NO. Thus a strong oxidant, presumably HO.,
exists in the CO flames.
The above trends in NO and NO. in the CO flames with added NO are
similar to those observed in hydrocarbon flames with added NO, although
the extent of formation and depletion in each flame varies. Generally, the
hydrocarbon flames convert a larger fraction of the NO to N0? than do the
CO flames. This is likely related to the relative amount of oxidant (H02)
present in each flame system.
NH_ Addition
Flames 5, 6, 11, 12, 21, and 22 show the NO and NO. data from CO
flames with NH. as an additive. The flames contained from 32 to 65 ppm
NH.. All flames containing NH. produce NO and varying levels of N0?. Only
the Lo-temp flames, e.g., Flame 21, show strong or moderate coupling (or
interdependence) between the NO and NO- formation and depletion processes.
Variations in NO. levels appear related to a diluent effect as seen
in the data of Flames 5 and 21. N0? formation is significantly less in
flames containing N. as a diluent as compared to those with Ar as a diluent.
About 10 times more NO. is produced in the cooler Lo-temp flame with Ar
than in the Hi-temp flame with N_ even though each flame had comparable
NH levels. Some NO. remains in the postflame gases of the Lo-temp flame,
but none is left in the hotter flame.
11
-------
Even though NO. levels vary, similar trends are observed in all flames
with NH. - the N0« forms early in each flame with subsequent peaking and
consumption of the N02 in the near postflame zone. In this respect, the
early N0« formation pattern is the same in CO flames containing either NH_ or
NO as an additive; extent of NO and NO,, coupling, however, can be quite
different. Also, in the flames containing NH», the major and often only
nitrogen oxide species in the emission gases is NO; the data indicate almost
100 percent conversion of NH, to NO in the flames. Only flames with Ar as
diluent contained measurable quantities of N0? in flue gas. In these
instances, the N0« concentration amounts to about 12 percent or less of the
total NO , which is representative of many reported N09 levels in flue gases.
£2N2 Addition
Flames 7, 8, 13, 14, 23, and 24 show that the NO and NO. data from CO
flames with C2N. as an additive. The level of C2N2 in the flames ranged
from 122 to 350 ppm.
As with NH , all CO flames containing C~N? as an additive, produce
both NO and N02 in the early flame reactions. However, data from the flames
containing C_N do not show any direct relationships (coupling) between NO
and NO ::ormation and depletion. The lack of apparent coupling between NO
and N0« .formation and depletion is likely due in part to the relatively small
fraction of N0? derived from the C-N in these flames. This is especially
true in fill of Hi-temp flames where only 0.7 to 1.6 percent of the CJfl^ was
converted to N0_ in the early flame reactions. More N0_ forms in the Lo-
temp flames with Ar as a diluent as was also noted with flames with NH,.
Except for Flame 23, the NO is completely consumed in the postflame region
of the CO flames with added C2N2 leaving only NO in the flue gas.
Large quantities of NO are formed from the CjN- in each flame as seen
in the figures. The data indicate that the hotter flames convert nearly
100 percent of the C,,N2 to NO, while the cooler flames less conversion to
NO occuns, but at least 45 percent of the C N appears as NO in every case.
12
-------
SUMMARY OF RESULTS - FLAMES WITH ADDITIVES
NO in CO Flames
All three additives, NO, NH3, and C^, have yielded N02 in the CO
flames. The flame diluent, argon or nitrogen, may have had some effect,
however, on the amount of N0_ observed in these flames. Thus, it was
observed that negligible levels of NO were produced on adding NH- or
C N- to nitrogen-diluted CO flames.
NO, as an additive, produced consistently high N0? levels in the
presence of N?, ranging from 44 to 70 percent of the NO added; conversion
in flames with argon as a diluent varied from 15 to 77 percent, the low of
15 percent occurring in the hotter CO-0 -Ar flame (Flame 9). On the basis
of N0« formation in the flames, the overall tendency for the fuel-N com-
pounds to produce NO is NO > NH, > CJL. With the exception of Flames 9
and 10, all flames show less N0_ formation from the fuel-N compounds, under
near stoichiometric or oxygen-deficient conditions than when excess 0- is
present.
The extensive consumption of N0« in the postflame gases may be indi-
rectly showing that high levels of 0 atoms were present in these CO flames,
if, as shown in previous studies, N0_ produced in the flame is removed by
the 0 atoms in the postflame. It is interesting to note that in all of the
Hi-temp CO flames containing fuel-N additives, except for the CO-0_-Ar
flames with NH3, the N0_ is completely consumed in the near postflame region
leaving only NO in the exhaust gases. Even when N0« is present in the post-
flame gases (Flames 11 and 12), the fraction of NO- in the NO is small,
£f A
about 11 percent in 02-rich flame and about 4-1/2 percent in the stoichio-
metric flame.
Also, data from the lower temperature flames with additives show that
in all of the stoichiometric or fuel-rich flames, the NO is completely
consumed in the postflame region; in the O^-rich flames, from 2 to 12
percent of the total NO may be present as NO- in the postflame gases.
X 4
13
-------
Thus, all CO flames with fuel-N additives produced NO-. With the
exception of the flames with NH_, all of the N0« produced is subsequently
consumed in the postflame region of all Hi-temp flames and in all fuel-
rich Lo-t:emp flames, leaving only NO in the .downstream gases. In the
remaining flames, NO- levels ranging from 2 to 12 percent of the total NO
£ X
was observed in the flue gases. Overall, the CO flames with fuel-N addi-
tives emit little or no N0_ with respect to total NO . Only one flame with
£» X
N as a diluent, Flame 17, showed N0_ in the flue gas. NO is the major
product formed from the fuel-N compounds and may be present at concentra-
tions ranging from about 50 to 100 percent of the fuel-N additive concen-
tration. Higher temperatures appear to favor more complete conversion.
NO in CO Flames - Material Balance
It is apparent from the NO data in Column 10, that, in the ppm fuel-N
range cohered in this study (<400 ppm max), high conversion of the fuel-
nitrogen to NO occurs in each case whether it be NH_ or C_N9 used as the
J £ £.
additive. The Lo-temp flames generally produced less NO than the Hi-temp
flames, although the most noted differences occurred mainly with C?N .
It is equally apparent from some of the flame data that nitrogen-con-
taining species other than NO and N0« are also being formed'in the conversion
process since a material balance between fuel-N input and NO output is
X
frequently not achieved, especially in the lower temperature flames where
residence times are considerably longer than in the hotter flames. More
experimental work is needed in this area to more fully explain the details
of the paths taken in the conversion of fuel-N to nitrogen oxides.
14
-------
SECTION IV
DISCUSSION
PROMPT NO IN ADDITIVE-FREE CO FLAMES
Fenimore (9) has indicated from studies of nitrogen oxide formation in
flames that "prompt NO" is observed in hydrocarbon flames but not in CO or
H2 flames. Pershing, et al (14) have reported similar results from their
studies. In these studies, prompt NO is defined by extrapolating NO-concen-
tration profiles measured in the postflame zone to zero time. As noted in
Figure 1, the present study suggests that prompt NO, although quite low, may
also exist in CO flames.
Two explanations are offered for these observations, namely, a Zeldo-
vich 0-atom process or a Fenimore CN process. Support for a Zeldovich
process to produce prompt NO lies in the high 0-atom levels in CO flames (15)
and the diffusivity of 0 atoms into the flame. The small positive intercept
in Figure 1 may be explained in terms of the excess 0 atoms. ,
A second possible explanation for prompt NO in these CO flames lies in
the fact that the flames in these studies were moist flames. CH radicals
might readily be generated under moist conditions, which then provide a
path for the production and oxidation of HCN. A typical set of reaction
steps are
CO + HO = HCO + OH (19)
HCO + H20 = H2CO + OH (20)
H2CO + 0 - CH2 + 02 (21)
CH2 + 0,H - CH + OH,H2 (22)
CH + N = CN + NH (23)
15
-------
CN.NH + 02 = NO + CO,OH. (24)
A more remote path may also be traced to excited nitrogen in flames (18)
via
CO + N* -» CN + NO. (25)
Young and Morrow (16) report that CN is formed from CO in active nitrogen by
reaction with an excited state of N. as indicated by Reaction 25.
The much higher intercept value in the CH, flame, however, suggests
that muci of the NO here is derived from non-Zeldovich processes. At the
observed flame temperatures (less than 1550 K) and high initial rate of NO
formation, an unrealistically large increase in 0-atom concentration would
be required to produce the NO by the Zeldovich mechanism. Non-Zeldovich
processes are, therefore, postulated to account for the early and rapid NO
formation in the hydrocarbon flames.
Although the Zeldovich process can be considered an important source
of NO in CO flames without additives, data from similar flames with fuel-N
additives suggest that the same type of Non-Zeldovich processes, similar to
those in hydrocarbon flames, also operate and even dominate in the CO
flames to produce nitrogen oxides. The mechanism involved in these processes
and some equilibrium and kinetic considerations are discussed below.
FUEL-N MECHANISM
The reaction mechanism
CN + 02 = NO + CO (26)
CN + 0. = NCO + 0 (27)
NCO + 0 = NO + CO (28)
NH + 02 - NO + OH (29)
NO + H02 = N02 + OH (6)
and
16
-------
N02 + 0,H = NO + 02,OH (30)
has been postulated to account for most of the NO and N0» formation in the
CO flames with additives. These reactions can be used to explain the early
formation of NO in the CO flames, as well as the formation and depletion
of NO . Certain reactions will predominate depending on the flame condi-
tions.
NO Addition
Data from the flames with NO additions have indicated the presence of
an oxidizing species in the region of the visible flame which readily
oxidized NO to NO.. As mentioned earlier, the oxidizing species has been
postulated as the HO. radical. The oxidation step is shown in Reaction 6.
This is a reasonable selection since the reaction of H0« with NO is fast and
has a low activation energy (17) ; the rate constant is given as
i n x n e in13 -1430 ± 180 3 . -1 -1
k = 2.0±0.5xlO exp - - - cm mole sec
The HO- radical must be formed in the CO flames either directly or
indirectly from the hydrogen in the water vapor used to stabilize the flames.
As brought out in the Results section, the extraction of hydrogen from a
water molecule is not a likely process for producing H0?. However, the
reaction sequence
CO + OH = C02 + H (8)
+M
H -I- 02 ".- H02 + M, (18)
is a likely source of H0_ radicals. From the N02 levels observed, it
appears that at least 25 to 35 ppm HO. may exist in these flames with added
£ N Addition
Spectroscopic studies have confirmed the production of significant
quantities of CN radicals from cyanogen molecules added to CO flames (18) .
The addition of C_N_ to the CO flames should, therefore, greatly increase
the CN radical concentrations, thereby enhancing the effects of the CN
species on nitrogen oxide formation in the flames. As seen in earlier data,
17
-------
the major oxide of nitrogen formed from the C2N is NO with only 0.7 to 5.5
percent of the C2N2 appearing as N02< The NO forms early and rapidly in the
flame and coincides approximately with the C.N? depletion. From the rela-
tively low temperatures encountered in the region of NO formation, it
appears that reactions of low activation energy are involved in the early
production of NO. Reactions 26, 27, and 28 would appear to satisfy the
low activation energy requirement and are postulated to account for the
early NO formation in the flames with cyanogen.
A sequence of reactions such as
C2N2 + °2 ~" 2NCO (31)
NCO + 0 -» NO + CO, (28)
involving the original C2N2 molecules might also be considered. Reaction
31, however, involves stable molecular reactants and would likely be too
slow to contribute significantly to the NO formation especially at low
flame temperatures. A more likely reaction involving cyanogen molecules is
C0N0 + 0 = NCO + CN. (32)
2 2.
This reaction has been postulated as an initial step in the breakdown of
cyanogen molecules in flames (19). Both NCO and CN radicals are produced
in Reaction 32, and would appear to be an excellent source of radicals for
Reactions 26, 27, and 28.
fiasco (20), in his cyanogen studies, has suggested that 85 percent of
the CN radicals react to: form NCO in accordance with Reaction 21, while
only about 15 percent react to form NO in accordance with Reaction 20.
The NCO produced in Reaction 27 is subsequently oxidized to NO via
Reaction 28. Haynes, et al (3) have suggested the reaction
CN + C02 = NCO + CO, (33)
might a;.so be an additional source of NCO leading to NO production in nonhy-
drocarbon flames. Certainly the NCO as well as the CN radical appears to
play an important role in the formation of NO in flames containing the N-C
type fu<>l-N materials.
18
-------
The very high levels of NO observed in the flames and the fact that
very little N02 is formed in the flames containing C.N. indicate that the
oxidizing step as represented by Reaction 6 is of minor importance in
these CO flames. The presence of H02 radicals or other oxidizing species
in the CO flames has been confirmed in the NO addition studies. Apparently,
the oxidant level is suppressed or reduced in the presence of cyanogen.
The reactive CN and NCO species derived from the cyanogen molecules may be
effective in reducing the concentration of the oxidizing species, as for
example, in the case of H07
CN + H02 = NCO + OH (34)
NCO + HO = NO + CO + OH.. (35)
These reactions also lead to NO production which is the dominant nitrogen
oxide in the flames containing cyanogen.
NH3 Additive
Adding NH- to the CO flames enhance the NH level in the flames (21)
thereby promoting high NO levels via Reaction 29. NO- formation occurs in
all the CO flames with added NH.,, probably through Reaction 6. However, as
mentioned previously, the NO- formation is much greater in flames containing
argon as compared to those with N_ as a diluent. One could consider the
effects of N« on the reaction
. NH2 + NO = N2 + H20, (36)
and the alternate reaction
NH2 + H02 = NH + 20H . (37)
with the reasoning that the N2 diluent shifts Reaction 36 to the left thereby
allowing more destruction of HO radicals through Reaction 37 or a similar
reaction. The reduced H02 level would subsequently produce less N02 via
Reaction 6.
The importance of Reactions 36 and 37, however, is uncertain at this
time. One could reason that similar flames with argon as a diluent would
reverse the above trends. While there is some indication of this in the
19
-------
data (Flames 5, 6, 11, and 12), full compliance with the predicted trends
is not a.chieved.
An alternative to the above considerations which might explain the
effects of N- and Ar on NO- formation in CO flames with added NH- is in the
efficiency of N9 and Ar as third bodies ir the reactions
H + 02 + M = HC2 + M (18)
and
H+H+M=H2+M. (38)
It is apparent, to achieve the observed results, that Ar would have to be
a more effective third body than N« in producing H0? radicals. Also, Ar
should be less effective in the recombination Reaction 38. Reported rate
constants for Reaction 18 using Ar or N_ as a third body (22) indicate
little difference in their effectiveness as third bodies. This is also
borne out in general in the data from the CO flames. For example, in the
CO flamas containing added NO (e.g., Flames 17 and 19), one cannot distin-
guish any real differences in the amount of NO- formed using either N« or Ar
as third bodies. Water vapor is a more effective third body than either N-
or Ar in the formation of H09 via Reaction 18 (22). Overall, however, the
data do not indicate water vapor as the controlling factor.
In regard to Reaction 38, reported values (21) indicate a slightly
higher rate of recombination of H-atoms in the presence of N- than Ar. This
might account for some of the observed difference in N0_ formation in the
CO flames with N or Ar as diluents, but again, would not appear to be the
controlling factor. Certainly the situation is more complex in the flames
with added NH» or C-N-. The reason for the reduced N0« formation in the
flames containing NH. and with N- as a diluent is not clear at this time.
Additional flame probings would prove useful here.
20
-------
POSTFLAME EQUILIBRIUM CONSIDERATIONS
If one considers the following equilibria involving NO and NO- in the
postflame gases
NO + 1/2 02 = N02 (39)
and
N02 + 0 = NO + 02, (40)
calculations show that at the lower flame temperatures encountered in these
studies, i.e., about 1250 K, the N02/N0 ratio for Reaction 39, at an 02 mole
fraction of 0.04, is less than 0.01. The ratio would be much less at higher
temperatures or if Reaction 40 is included. Therefore, at equilibrium at
least 99 percent of the NO is NO. The postflame data from the Hi-temp
flames and all fuel-rich Lo-temp flames also show 99 percent or more of the
NO as NO suggesting equilibrium conditions may prevail as represented by
X
equations 39 and/or 40. The appearance of several ppm NO in the 0,,-rich
Lo-temp postflame gases is unexpected from the above equilibrium (and
temperature-time) considerations and indicate other conditions, involving
NO and N0? formation and depletion processes, may prevail here.
The NO in the postflame gases may, under certain conditions of tempera-
ture and 0?-N- levels, be equilibrated with regard to the reaction
1/2 N2 + 1/2 02 = NO. (41)
Calculations from the flames containing N« (0.72 mole fraction) and excess
0? (.04 mole fraction) indicate that about 90 ppm NO could be formed at
1200 K and about 850 ppm NO at 1600 K. These calculated levels are generally
higher than those observed for NO in the postflame probings and indicate
that nonequilibrium conditions, in regard to Reaction 41, generally prevail
here. In the fuel-rich flames, the NO could exceed the equilibrium level
which would then shift Reaction 41 to the left. Overall, however, the data
do not show any large decreases in NO in the postflame gases even at the
high NO levels observed in the flames with C2N2. Therefore, the above equi-
librium is apparently not dominating in the flames, or residence times and
temperatures are not sufficient to establish the equilibrium conditions. ,_^
21
-------
KINETICS OF C N AND NH DEPLETION
The rates of C2N2 depletion and the subsequent formation of NO and NO
in the CO flames have been examined and some apparent activation energies
for the overall depletion processes involving these. N-bearing species have
been determined. For C-N-, the general overall reaction is assumed to be
C2N2 + a 02 '-» NO + products. (42)
For NH.. the reaction is taken as
NH3 + b 02 * NO + products. (43)
The reaction NO + HO. = NO + OH is also considered in the early flame
reactions and is assumed to account for the early. NO formation. The over-
all rate expressions for C N and NH- depletion is expressed as
R-C N = 1/2(RNO + *»<> }' f°r C2N2
and
R-NH = ^0 + \0 ' f°r M3 dePletion'
For Reaction 42 the rate equation becomes
R-C2N2 = 1/2
-------
Flame No.
24
23
22
21
12
11
Fuel
CO
CO
CO
CO
CO
CO
Fuel-N
Additive
C2N2
NH3
NH3
NH3
Apparent
Act. Energy
kcal/mole
36.4
45.2
I
54.9
86,6*
59.1
59.6
Avg.
' 40.8
* 57.9
* Omitted from averaging.
The average overall activation energy obtained for the depletion of the
cyanogen in the flames is in fair to good agreement with the results
obtained by other investigators. DeSoete (5) reports a value of about 40
kcal for the activation energy, while Allielet (18) reports 36 kcal for the
same process. Our average E value of 57.9 kcal/mole for ammonia is about
3.
26 kcal/mole higher than that obtained by DeSoete (5).
Internally, the calculated energy values, as determined from each
flame, are in fair agreement for each of the species C»N2 and NH,. A
considerably higher activation energy is observed, however, for NH.,. This
may reflect the larger energy needed to break the N-H bond of ammonia
(83.7 kcal/mole) as compared to the C-C bond of cyanogen (58.6 kcal/mole).
Further considerations along these lines show that in the case of cyanogen
the breaking of the C-C bond would lead to CN (or NCO) formation. Studies
have indicated that the subsequent reaction CN + 0,, = NCO + 0, requires no
activation energy (19). On the other hand, assuming NH radicals are derived
from the NH_, the reaction NH + 02 = HNO + 0, is reported to have an acti-
vation energy of approximately 13 kcal/mole (19). From these considerations,
it appears that NH, would require more energy to produce NO than would
cyanogen.
23
-------
Thus, the data indicate that the fuel-N species with cyanogen-type
bonding require less activation energy to produce NO thereby allowing more
NO to form at lower flame temperatures than fuel-N species with N-H
bonding similar to NH_. However, one must also consider an alternate
reaction in the conversion of fuel-N materials to N-containing products,
that is, the general reaction
F-N + NO = N2 and/or NO + products.
It appears from our studies, that for F-N = CJJ-, the above reaction
progressively dominates as the maximum flame temperature is lowered, i.e.,
less NO is formed at the lower flame temperatures. This is not the case
for F-N = NH_ where most of the ammonia appears to be converted to NO even
at the lower flame temperature. However, because sufficiently high tempera-
tures and long residence times are generally encountered in the most
commonly used furnaces and burner equipment, the end results will likely be
the same1, for C N or NH , i.e., at comparable fuel-N concentrations, the
total NCi in the flue gases will be about the same for C.N0 or NH0.
x 223
-------
SECTION V
SUMMARY
Water-stabilized CO flames have been probed in the presence and absence
of fuel-N compounds. In the absence of additives, the CO flames produced
a "prompt-NO" in accordance with the Zeldovich process if one assumes a
diffusion of excess 0 atoms into the visible flame region; NO was also
produced in each flame except for the O^-rich high temperature, high velo-
city flame without additives. In every case, the N0? appeared early in the
flame reactions, producing the same early N0» phenomena observed in hydro-
carbon flames. The appearance of early N0_ was confirmed by adding NO to
the cold gases of CO flames and observing the depletion of NO and subsequent
formation of NO in the early flame reactions. As in hydrocarbon flames,
the HO radical is postulated as the species responsible for NO formation in
the CO flames via NO + H02 = NO + OH. The H0_ radicals are derived from
the water vapor used to stabilize the CO flames. The addition of C9N? or
NH_ to the CO flames produced large quantities of NO. On the other hand,
little N0_ was produced from the C^N^ molecules although increased residence
time in the flames increased the N09 production. More NO- was produced in
the presence of NH,, but only when argon was used as a diluent. Using N?
as a diluent appeared to suppress NO formation in the flames with NH., but
did not affect NO formation. The formation and depletion of NO and N09 in
the CO flames can be expressed by the following equations
N0 + co
NCO + 0
NCO + 0 * NO + CO
NH + 02 = NO + OH
NO + H02 = N02 + OH
25
-------
N02 + 0,H - NO + 02,OH.
An altei.-nate reaction producing essentially Inert N-containing products la
suggested in the pyrolysis of fuel-N compounds, especially at lower flame
temperatures
F-N + NO = N_, NO + products.
The importance of each reaction varies from flame to flame, depending on
the ovexall flame conditions. Apparent activation energies representing
overall oxidation of C.N. and NH. to NO and NO in the flames were deter-
mined asj 40.8 kcal/mole and 57.9 kcal/mole, respectively.
Nitrogen oxide formation and depletion processes in the CO flames are
very similar to those in hydrocarbon flames. The major difference appears
to be i:i the amount of transient N02 formed in the flames. Generally, the
CO flames produce less NO^ than do similar hydrocarbon flames. This can be
related to less HO. production in the hydrogen-deficient CO flames. Ulti-
mately, however, the end results are similar to both CO and hydrocarbon
flames; at low ppm fuel-N levels, most of the fuel-N appears as NO in the
postflane gases. Small amounts of NO. are sometimes present in the post-
flame gases of the O.-rich flames.
Tha flame data show high N0« emission levels can result, from the CO
flames if rapid quenching occurs near the flame front. This latter
situation often occurs in open air flames of various geometries. Rapid
quenching must be avoided if N0? levels are to be kept low in the emission
gases.
There are many unanswered questions insofar as a full understanding
of NO chemistry in CO flames is concerned. These studies of the NO
X X
flame chemistry, however, have provided new information in this area as
'_
summarized above. Certainly, the requirement of H-containing compounds to
stabilize the CO flames appears to produce conditions in these flames
similar to those observed in hydrocarbon flames insofar as the nitrogen
oxide chemistry is concerned. In order to get more information on the
26
-------
role of the HO radical in nonhydrocarbon flames, a dry CO-C^N flame or
even a pure C-N or COS flame may be required.
27
-------
REFERENCES
1. Murryman, E. L. and A. Levy. NO Formation in Combustion: Anomolous-
Early N0« Formation. Proceedings of the Third International Clean Air
Congress. Dlisseldorf, Germany, 1973. p. C117.
2. Turner, D. W. and C. W. Siegmund. Staged Combustion and Flue Gas
Recycle: Potential for Minimizing NO from Fuel Oil Combusting.
Presented at the American Flame Research Committee Flame Day.
Chicago, Illinois, September 6-7, 1972.
3. Haynes, B. S., D. Iveroch, and N. Y. Kirov. The Behavior of Nitrogen
Species in Fuel Rich Hydrocarbon Flames. Presented at the Fifteenth
Symposium (International) on Combustion. Tokyo, August, 1974.
4. Martin, G. B. and E. E. Berkau. An Investigation of the Conversion of
Various Fuel Nitrogen Compounds to Nitrogen Oxides in Oil Combustion.
A'.tChE National Meeting. Atlantic City, New Jersey, August, 1971.
5. DuSoete, G. Overall Reaction Rates of NO and N Formation from Fuel
Nitrogen. Presented at the 15th International Symposium on Combustion.
Tokyo, Japan, August 25-31, 1974.
6. Funimore, C. P. Formation of Nitric Oxide from Fuel Nitrogen in
E'-.hylene Flame. Combustion and Flame 19, 289-296, 1972.
7. Diixbury, J. and N. H. Pratt. A Shock Tube Investigation of the Fuel
Nitrogen - Nitrogen Oxide Problem. Combustion Institute. European
Symposium. Academic Press. New York, New York, 1973. p. 433.
8. Miirryman, E. L. and A. Levy. Nitrogen Oxide Formation in Flames:
The Roles of NO and Fuel Nitrogen. Presented at the Fifteenth
Symposium (International) on Combustion. Tokyo, Japan, August, 1974.
9. Funimore, C. P. Formation of Nitric Oxide in Premixed Hydrocarbon
Flames. Thirteenth Symposium (International) on Combustion. The
Combustion Institute. Pittsburgh, Pennsylvania, 1971. p. 373.
10. Wires, R. A., L. A. Watermeier, and R. A. Strehlow. The Dry Carbon
Monoxide-Oxygen Flame. J. Phys. Chem., 63, 1959. p. 989.
28
-------
11. Feeters, J. and G. Mahnen. Reaction Mechanisms and Rate Constants of
Elementary Steps in Methane-Oxygen Flames. Fourteenth Symposium
(International) on Combustion. The Combustion Institute. Pittsburgh,
Pennsylvania, 1973. p. 133.
12. Merryman, E. L. and A. Levy. Kinetics of Sulfur-Oxides Formation in
Flames: II Low Pressure H S Flames. J. Air Poll. Control Assoc., 17,
800, 1967.
13. Shaw, J. T. and A. C. Thomas. Oxides of Nitrogen in Relation to the
Combustion of Coal. 7th International Conference on Coal. Prague.
June, 1968.
14. Pershing, D. W., G. B. Martin, and E. E. Berkau. Influence of Design
Variables on the Production of Thermal and Fuel NO from Residual Oil
and Coal Combustion. 66th Annual AIChE Meeting. Phildelphia,
Pennsylvania, November 11-15, 1973.
15. Gaydon, A. G. Continuous Spectra in Flames: The Role of Atomic
Oyxgen in Combustion. Proc. Roy. Soc., A, 1944, 183. p. 111.
16. Young, R. A. and W. Morrow. Formation of CN from CO and Its Excitation
in Active Nitrogen. J. Chem. Phys., 60(3), 1005, 1974.
17. Hack, W. and K. Hoyermann. The Reaction of NO + HO * NO- + OH and
OH -I- H_0 -» HO + HO as a HO -Source. Presented at the Symposium
on Chemical Kinetics Data for the Lower and Upper Atmospheres.
Warrenton, Virginia, U.S.A., September 16-18, 1974. p. 107.
18. Doughtery, G. J., M. J. McEwan, and L. F. Phillips. Some Photometric
Observations of Trace Additives in Dry Carbon Monoxide Flames. Comb.
and Flame 21, 253-259, 1973.
19. Ailliet, M. and A. Van Tiggelin. Burning Velocity in Mixtures of
Cyanogen on Hydrogen Cyanide with Oxygen. Bull. Soc. Chem. Beiges,
77, 1968. pp. 433-446.
20. Basco, N. The Reaction of Cyanogen Radicals with Oxygen. Proc. Roy.
Soc., A, 1965, 283. p. 302.
21. Wolfhard, W. G. and W. G. Parker. A New Technique for the Spectroscopic
Examination of Flames at Normal Pressures. Proc. Phys. Soc. (London)
A62, 1949. p. 722.
22. Sternling, C. V. and J. 0. L. "Wendt. Kinetic Mechanisms Governing
the Fate of Chemically Bound Sulfur and Nitrogen in Combustion.
Shell Development Company. Final Report to Environmental Protection
Agency, Contract No. EHSD 71-45, Task 14, August, 1972.
29
-------
23. Sarofim, A. F., G. C. Williams, N., and A. Padia. Control of Emission
of Nitric Oxide and Carbon Monoxide from Small-Scale Combustors.
Proceedings of the 2nd Conference on Natural Gas Research and
Technology. Atlanta, Georgia, June 5-7, 1972.
30
-------
APPENDIX A
FLAME PROFILES
31
-------
APPENDIX A
FLAME PROFILES
a = Flame holder
b = Bottom edge of flame
c = Top edge of flame
Flames with similar CO, CO^, 0^, and temperature profiles
Flame Nos: 1, 3, 5*, 7
2, 4, 6*, 8
9*. 11, 13
10*, 12, 14
19*, 21, 23
20*, 22, 24
* Detailed profiles presented.
32
-------
Flame
15 -
10 -
5 -
Flame # 2
i
§
£
Oi
40
30
20
10
Stoichiometric
(No additive)
0 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50
DISTANCE ABOVE FLAMEHOLDER, CM
33
-------
CM
§
60
50
40
^ 30
20
10
Flame # 3
02-rich
NO
(34 ppm NO added)
NO*
LL
* No additive present.
Flame # 4
i
60
50
40
30
20
10
NO*
Stoic hiome trie
(36 ppm NO added
* No additive present.
0.:>5 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50
DISTANCE ABOVE FLAMEHOLDER, CM
34
-------
o
f.
o
50
40
30
20
10
K;1260 1603 1604
1575
1513 Flame # 5
1 O.-rich
FN
^
CO
A
_NO_
(35 ppm NH. added
0.20
0.15
0.10
0.05
o
o
to
o
10
CM
o
o
2
T°;C
40
30
20
10
1415
1270 1509 1516 1509
1497
1433
Flame
NO
Stolchlometric
(31 ppra NH- added)
co
u
i\.-^> i_u
L>-'i=^c 1 -^J " I
FN
T^ rrrrr ^r Lrr^.-cr= i.-'^cTrrT
0.20
0.10
o
t>
n>
11
o
rt
o
o
o
0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50
DISTANCE ABOVE FLAMEHOLDER, CM
35
-------
Flame # 7
FN (by chemiluminescence)
300
250
NO
e, 200
o
2
i
g
CL,
150
100
FN (by chemiluminescence)
50
NO,
Flame $ 8
Stoichiometric
(165 ppm C,N,added)
0 0.215 0.50 0.75 1.00 1.25
1.50 1.75 2.00 2.25 2.50
DISTANCE ABOVE FLAMEHOLDER, CM
36
-------
i
£
1581 1588
T Ki 1351 1592 L58I
120
100
80
60
40
20
1474
T°K: 1264 1516 1505
1533
Flame # 9 Ii75
-mrri L-Zrr- t= .,:= =r=i
1490
60
50
40
g
20
10
1423
Flame * 10
\a_
11
_L
NO
Stoichiotnetri<
(71 ppm NO added)
CO
- °
NO
2 i
co
//
V °2 V
CO
1.
M«B«^ ^M - ~
'
~
__ _ . - '."
*T" L1'"""1
u.zu
0.15
0.10
0.05
0.25 0.50 0.75 1,.00 1.25 1.50 1.75 2.00 2.25 2.50
DISTANCE ABOVE FLAMK1IOLDER. CM
37
-------
Flame # 11
-------
Flame tf 13
1&7S
§
£
A<
g
ft,
500 -
41)0 -
300
21)0 -
100 -
FN (by cherailuminescence)
A 2
2e~i 1 -I U.
Flame # 14
Stoichiometric
(350 ppm C2N2 added)
100 -
', FN (by chemiluminescence)
/ 'I
2' r. t. 1
0 0.25 0.50 0.75 ,1.00 1.25 1.50 1.75 2.00 2.25 2.50
DISTANCE ABOVE FLAMEHOLDER, CM
39
-------
1503 1558 1573 1582 1583
Flame f> 15
SB
(«
O
z
CM
§
T K: .1400 1545 1564 1582 1581 1564 1546 1533 1501
1 -
132A 1404 1418 1422 1421
TK: 1227 1373 1410 1421 1422
1403 1384 1365 1359
Fuel-rich
(No additive)
0 l»=-
0.12 0.25 0.37 0.50 1.00 1.50 2.00 2.50
DISTANCE ABOVE FLAMEHOLDER, CM
40
-------
if
2
IK
«M
i
§
1253 1350 1370 1360 1359
T°K: 1146 1314 1364 1369 1351 1355 1352 Flame ff 17 1336
30
25
20
15
10
U /
I I I
" -
FN
(A8 ppm NO adde«9
111(11 I I l^x-^ I
20
16
12
1219 1296 1294 1286 1269
T°K! 1097 1292 1296 1288 1288 1269 1252 Flame #18 1225
\r
NO,
(1237)
Fuel-rich
(34 ppm NO added)
0.12
0.25
0.37
FN
0.50 1.00 1.50 2.00 2.50
DISTANCE ABOVE FLAMEHOLDER, CM
41
-------
(M
g
o
Z
1329 1345 1335 1325 1306
T°K: _ 1247 1348 1340 1332 1321
1273 1253 Flame <' 19
15 -
10 -
5 -
1183 1269 1269
1257
T°K: 1128 1220 1271 1264 1258 1240 ' 1218 1201 Flame # 20
40
32
g 24
16
CO
^
N
r
b
\N°2
. \
1176
Fuel-rich
(64 ppm NO added)
C0
co
?
- 0.10
fl
i
o
3
- 0.05
O
o
0.12 0.25 0.37 0.50 1.00 1.50 2.00 2.50
DISTANCE ABOVE FLAMEHOLDER, CM
42
-------
Flame #21
Flame # 22
35
0 0.12 0.25 0.37 0.50 1.00
_l
2.00 2.50
DISTANCE ABOVE FLAMEHOLDER, CM
43
-------
Flame # 23
120
100 -
80 -
60 -
40 -
20 -
Flame # 24
g
§
120
100 -
80 -
60 -
40 -
20 -
Fuel-rich
(129 ppm C2N2 added)
0 0.12 0.25 0.37 0.50 0.62 0.75 1.50 2.25 2.50
DISTANCE ABOVE FLAMEHOLDER, CM
-------
APPENDIX B
ARRHENIUS PLOTS
45
-------
10
2
11
8
6
10
10
8
6
103
Flame #11
= 59.6 kcal/mole
0.70 0.75 0.80 0.85 0.90
in
ts
io
2
12
8
6
4
8
6
4
10
10
8
6
10
Flame # 12
59.1 kcal/mole
0.70 0.75 0.80 0.85 0.90 0.95
103/T
46
-------
V
V
CO
in
CM
0
u
10
Flame # 21
E -86.6 kcal/mole
fl
0.77 0.79 0.81 0.83
0.85 0.87
10
10
2
11
8
6
4
2
10
8
6
Flame # 22
E =54.9 kcal/mole
I
0.80 0.82 0.84 0.86 0.88
3
10 /T
0.90
47
-------
u
u
Flame If 23
E "45.2 kcal/mole
a
10
20
0.76 0.78 0.80 0.82 0.84
3 10
e
u
u
2
20
8
6
Flame # 24
E = 36.4 kcal/mole
2
10
19
0.82 0.86 0.90 0.94 0.98 1.02
103/T
48
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/2-77-008C
2.
3. RECIPIENT'S ACCESSIOttNO.
4. TITLE AND SUBTITLE
NOx Formation in CO Flames
5. REPORT DATE
January 1977
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
E. L. Merryman and A. Levy
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Batte lie-Columbus Laboratories
505 King Avenue
Columbus, Ohio 43201
10. PROGRAM ELEMENT NO.
1AB014; ROAP 21ADG-020
11. CONTRACT/GRANT NO.
68-02-0262
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
Task Final; 5/72-12/75
14. SPONSORING AGENCY CODE
EPA-ORD
is. SUPPLEMENTARY NOTES
8411 Ext 2432 , Mail Drop 65.
prOject officer for this report is W.S. Lanier, 919/549-
16. ABSTRACT
The report gives results of an experimental study to determine if early NO
and NO2 can be observed in CO flames, since prompt NO is not anticipated and since
HO2 levels might be expected to be lower in CO flames. (Previous studies of NO and
NO2 production in methane flames with and without fuel nitrogen suggested that the
early appearance of NO2 results from the fast reaction of NO + HO2 = NO2.+ OH.)
CO flames containing NO, ammonia, and cyanogen were profiled for NO and NO2 in
this study. Effects of flame temperature and of nitrogen- and argon-"air" were also
examined. The results give evidence for prompt NO in CO flames as well as for
early NO2. The prompt NO and early NO2 levels are less than in methane flames,
but the same mechanism appears to explain the formation of NO2 in both flame sys-
tems. In most instances, the quantity of NO or fuel-nitrogen added to the flames can
be accounted for as NO,NO2, and an unidentified nitrogen species. Analysis of the
rates of depletion of ammonia and cyanogen in the CO flames yields Arrhenius coef-
ficients equivalent to 58 and 41 kcal/mole, respectively.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Air Pollution; Nitrogen Oxide (NO)*;
Nitrogen Dioxide; Flames; Kinetics;
Carbon Monoxide*; Ammonia; Cyanogen;
Nitrogen Oxides*; Oxygen; Hydroperoxides
Air Pollution Control;
Stationary Sources;
Prompt NO*; Early NO2;
Flame Kinetics; Fuel-
Nitrogen*; Oxygen
Atoms; Flame Probing;
13B; 07B;
21B; 20K;
07C;
18. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
55
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
49
------- |