x>EPA
EPA > <002
979
Enhanced SOs Emissions
from Staged Combustion
Interagency
Energy/Environment
R&D Program Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development. U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program. These studies relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments of. and development of. control technologies for energy
systems; and integrated assessments of a wide'range of energy-related environ-
mental issues.
EPA REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for publication. Approval does not signify that the contents necessarily reflect
the views and policies of the Government, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service. Springfield. Virginia 22161.
-------
EPA-600/7-79-002
January 1979
Enhanced SOs Emissions from
Staged Combustion
by
Earl L Merryman and Arthur Levy
Battelle-Columbus Laboratories
505 King Avenue
Columbus. Ohio 43201
Grant No. R805330-01-1
Program Element No. EHE624A
EPA Project Officer: W. Steven Lanier
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park. NC 27711
Prepared for
US. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
ABSTRACT
Staged combustion is a recognized, effective means for lowering
emissions. Examination of the staged combustion process suggests however
that the high CO levels produced in the first stage may pump a sufficient
level of oxygen atoms into the second stage to result in increased
(enhanced) 803 formation. Experiments were carried out in a small two-
stage comb us tor which allowed for an examination of S03 formation under
similar single- and two-stage conditions. The experiments show that
although staging can cause enhanced S03 formation, the enhancement is of
short duration, and is dependent on the air /fuel ratio of each stage and
the delay interval between the first and second stages. Kinetic analysis
yields a value of kj = 7.4 x 10 cm mole" sec"1 for the reaction
S<>2 + 0 + M - SOa + M and k£ - 1.5 x 10 cm3 mole'1 sec"1 for the reaction
SO3 + 0 - S02 + 02 (T » 1685 K). The kinetic analysis also shows that en-
hancement of S03 formation in staged combustion can occur. However, the
experimental results suggest that the enhancement may only be a transient
phenomenon dependent on several combustion variables.
iii
-------
CONTENTS
Abstract iii
Figures v
Acknowledgement vi
1. Introduction 1
Thermodynamics and Kinetics 1
Enhanced 803 Emissions 2
2. Experimental 5
Burner System 5
Analytical 5
Flame Conditions 5
3. Results 8
Axial 803 Profiles 8
Radial Profiles 11
4. Discussion 12
General Observations 12
Enhancement 12
Time Delay Effects 12
Air/Fuel Ratio Effects 12
803 Fluctuation iq
Kinetics '.'.'.'.'.'.'.'.'.'.'. 13
V
Reaction 1, S02 + 0 + M —-^* SG3 + M 14
k2
Reaction 2, S03 + 0 * S02 + 02 14
Two-Stage Enhancement * 15
5. Conclusions • 17
References
18
iv
-------
FIGURES
Figure 1. Oxidation of S02 in flames from different fuel gases
/0\
Figure 2. SO. profiles in H S, COS, and CH-SH flames '
Figure 3. Quartz tube, two-staged comb ustor
Figure 4. Oxidation of S02 in single and staged combustion, $1 = 0.95,
$s = $2 * 1-1. At is the delay time before addition of
secondary air
Figure 5. Oxidation of S02 in single and staged combustion, =0.99
Figure 6. Oxidation of SO. in single and staged combustion, $ = 0.90
-------
ACKNOWLEDGMENT
The authors acknowledge the interest and discussion of
Mr. W. S. Lanier, Project Officer, of the Environmental Protection
Agency, IERL, Research Triangle Park, North Carolina, during the course
of this research program.
vi
-------
SECTION 1
INTRODUCTION
Staged combustion has proved to be an effecMve combustion modification
technique for lowering NOX emissions. Whether carried out as a direct two-
step process, i.e., as a fuel-rich' stage followed by a fuel-lean stage,
or by a biased firing process where fuel-rich and fuel-lean zones exist in
the combustor, the process has been quite effective in reducing thermal
NOx, and possibly equally effective in reducing fuel-nitrogen NO^. This
report considers the effect of staging on other combustion pollutants,
specifically on combustion-generated sulfur trioxide.
In looking at the two-stage process, one can rationalize reduced
emissions at least partially in terms of a reduced concentration of oxygenating
species in the fuel-rich first stage combined with a significantly lower tem-
perature in the oxygen-rich second stage. At first glance then one might
logically expect that inasmuch as the formation of 803 in combustion occurs
by an 0-atom process, 803 emissions from staged combustion would also be
reduced. Data in the literature however suggest the contrary, namely that
the two-stage process might actually cause an increase in the concentration
of sulfur trioxide in the flue gas stream. This obviously, of pronounced,
would be objectionable from an operational point of view if it contributed
to increased corrosion problems in boilers, and from an environmental point
of view if it contributed to increased sulfate emissions. The purpose of
this study was to explore the effect of staging on 803, first to determine
if staging had an effect, especially if it could cause an increase, i.e.,
enhancement, of 803 relative to single stage firing, and second to examine
the effects of staging on 803 formation kinetics.
THERMODYNAMICS AND KINETICS
Equilibrium calculations for ratios of 803 to SO 2 as a function of
temperature and excess air are well established. If equilibrium conditions
were to prevail under staged firing one could logically expect that, as excess
air was reduced, the conversion of S0£ to 803 would also be reduced. Strict
thermodynamic criteria are not applicable to flames however. Although equili-
brium conditions may prevail at high temperatures, as gases cool, reaction
rates decrease and equilibria may be frozen. It is also recognized that
sulfur-containing species may take up a quasi-equilibrium (partial equilibrium)
distribution, quite different from true equilibrium (1). Also, S02 is an
effective agent for the recombination of both hydrogen and oxygen atoms (2)
and might readily influence the effect of first stage processes on the oxida-
tion steps in the second stage.
-------
Klnetically the formation of 803 in an oxygen atom process is best
described by the mechanism (3,4)
S02 + 0 + M = S03 + M (1)
S03 + 0 * S02 + 02 (2)
S03 + H = S02 + OH . (3)
Rate data for Reaction 1 are quite variable; the data suggest a value of
k^ = 1.0 x lO1^ cm** mole""2 sec"1 (5) at room temperature with little or no
activation energy required for the reaction. For Reaction 2, Merryman and
Levy obtained
k2 - 2.8 x 1014 exp (-12,000/RT) cm3 mole-1 sec'1
in the temperature range 1480 to 1550 K, and
k'2 = 6.5 x 1014 exp (-10,800/RT) cm3 mole'1 sec'1,
in the temperature range 825 to 1325 K. The latter k_ value takes into con-
sideration Reaction 3, whereas the former does not. Rate values for Reaction 3
have not been obtained experimentally; however, Fenlmore and Jones (6) estimate
the ratio of k-/k2 or k,/k3 to be of the order of 10* cm3 mole"1.
ENHANCED SO. EMISSIONS
Four sources of data suggest that a potential increase in 803 might
result from staged combustion. The first comes from a study by Hedley (7)
where he showed 803 was considerably in excess of thermodynamic equilibrium.
The second comes from studies of the authors (8) where some of the 802-803
kinetics were developed. The third and fourth, from studies of Dooley and
Whittingham (3) and of Gaydon (9), identify the Importance of CO oxidation
kinetics on 803 kinetics.
At the time of Hedley's studies no special attention was directed to
the use of staged combustion for NOx control. However, in the course of
bringing forth his evidence on the formation of 803 at levels considerably
above equilibrium he describes a one-dimensional controlled mixing experiment
and (although this is not presented as such in his paper) he states: "If
combustion took place under stoichiometric or fuel-rich conditions then no
trioxide formation took place. When less than stoichiometric air was used,
the unburnts in the gases consisted solely of carbon monoxide with 802 but no
803. When the remaining excess air was injected into these gases, the
maximum amount of 803 formed was greater than that formed when this addi-
tional air was included with the initial combustion air, the overall excess
of air being the same in both cases."
-------
This observation and its potential impact on the effects of staging on
803 formation then became closely tied to the effect of CO oxidation kinetics
in the second stage production of 803. The CO effect is best borne out in
Figures 1 and 2 where one notes the highest conversion of S02 to 803 in sulfur-
bearing CO flames (3,8). When one couples (1) the observations- in Figures 1 and
2 with (2) Gaydon's observations of high concentrations of oxygen atoms in CO
flames and with (3) the 0-atom mechanism for 803 formation in combustion,
Hedley's statements on the enhancement of 803 in staged firing appear quite
consistent. Basically then 803 formation is an oxygen atom process, and the
question to be addressed is "what is the effect of staging on the oxygen atom
concentration?" and its corallary, "what is the effect of staging on 803
formation?"
-------
20
O
09
O
3
O
O CO Flam*
• Hj Flame
o CH. Flame (Luminous)
% SOin Flue Gas
Figure 1. Oxidation of S02 in Flames From Different Fuel Gases(3)
900
450
400
5 390
at
&
I 300
a
w
•&
g 290
o.
200
8 wo
no
90
-Flame holder
flame
(PT * 250 torr)
H3SH flame
(PT«625 torr)
H»S flame
(PT « 250 torr)
I I
I
O 200 400 600 800 WOO 1200
Distance Above Flameholder, mils
Figure 2. S03 profiles in H2S, COS, and CH3 SH Flames(8)
4
-------
SECTION II
EXPERIMENTAL
A quartz tube burner which allowed one to establish stable methane-
flames within desired fuel-air ratios was used. Two inlets were pro-
vided above the flame (first stage) for adding air to complete the
combustion process (second stage). 863 was then sampled at various positions
downstream of the secondary air. Figure 3 is a sketch of the burner tube
and the insulated postflame reaction chamber with the various sampling
ports.
Burner System
The primary burner tube was constructed of quartz (13 mm I.D.) and
produced a laminar flow bunsen-type flame (L/D >60). The reaction chamber
surrounding the burner tube was 18 mm I.D., also quartz. This chamber con-
tained several temperature and sampling ports spaced from 3-1/2 to 5-1/2 cm
apart in the early postflame zone increasing to about 12-1/2 cm apart in the
far postflame zone (Figure 3). These spacings provided appropriate time
intervals for collecting the 803. The reaction chamber was externally heated
(Chromel "A" wiring) to control second stage temperatures. Total chamber
length was approximately 80 cm, providing a maximum gas residence time of
about 250 msec.
Secondary air was introduced at positions F and G shown in Figure 3.
The jets at F and 6 were at 6-1/2 cm and 46 cm above the first-stage burner
rim, respectively. Temperatures at the secondary air entrance ports were
generally above 950 C. Temperature profiles in the postflame gases were
obtained with silica-coated Pt-Pt/10 percent Rh thermocouples.
Analytical
Gas samples were removed at various locations above the flame via a
quartz sampling probe. 803 was removed from the gas as H2S04 using a
Goksoyr-Ross type (micro) collection apparatus and was determined
colorimetrically by the barium chloranilate procedure. CO, C0£, 02* and
S02 were also measured, mainly for purposes of confirming and comparing
postflame combustion conditions and sulfur oxide levels with calculated
cold gas compositions.
Flame Conditions
Three flame compositions were probed in detail for S03 profiles in this
study. The compositions were
-------
Sompli
pomt I
J—Teflon
J seal
Top View
of Swirling
Vanes
Figure 3. Quartz Tube, Two-Staged Combustor
-------
Gas
CH,
°2
N?
H2S
Single Stage <|>B
0.087
0.191
0.720
0.0015
Mole Fraction (Cold Gases)
Two-Stage*
=1.1 1st Stage (J>, -0.95* 1st
0.099
0.189
0.710
0.0017
Stage 2"l.l; mole fractions were the same as in
the single stage firing.
Equivalence ratios, defined as
4> - [Air/Fuel]/[Air/Fuel] stoich,
are expressed as 8, i» and $2* i.e., g, single stage; ,, first stage; and
4>2, second stage. In all experiments the second stage firing introduced
sufficient air that $2 was comparable to s.
Total cold gas flow rate for each flame was maintained at 205 cc/sec
and the S(>2 level was kept constant at 1500 ppm (after addition of secondary
air in the two-stage firings).
-------
SECTION III
RESULTS
The results obtained from this study are discussed mainly in relation
to the "enhancement" (i.e., the increase) in 803 from two-stage combustion
as compared to similar single-stage firings. The maximum S03 level observed
in each mode of firing is used as a basis for evaluating the enhancement
phenomenon, if any. Since the enhancement effects are observed to be condi-
tional, one should keep in mind the specific experimental parameters leading
to the observed increase in 863 over single-stage firings and the transient
nature of the observed increase. Also, it is clear that the results obtained
from this study are derived from a homogeneous gas-phase system.
Figures 4, 5, and 6 summarize the major findings in this study. It
should be noted that residence time for the two-stage flames in these
figures refers to the time between the introduction of secondary air and
the 803 measurement position. Also, the At on the above figures refers to
the delay time before adding the secondary air. The mid-flame position was
used as zero time for the single-stage firings (and for calculating At).
CO levels for single-stage combustion were about 1100 ppm Just above
the flame and less than 50 ppm at the exit of the reactor tube. CO levels
for first-stage firing (i - 0.95) were 15,000-17,000 ppm, and these dropped
to 135-260 ppm near the second stage exit. Exploratory experiments showed
that the addition of secondary air at about 850 C resulted in slow reaction
rates and produced considerably less 803 than when the air was added at 950 C.
Therefore, the higher temperature was maintained at the secondary air entrance
ports for all two-stage combustion experiments carried out in this study.
The higher temperature is also more in line with single-stage conditions
and produces better burnout.
Axial SOq Profiles
The data in Figure 4 show the effects of two-stage combustion and the
effects of time of addition of secondary air on 803 formation. The two-stage
firings, Curves B and C, show a transient enhancement of 803 production,
relative to single-stage firing, Curve A. The effect, although it does not
appear pronounced here, occurs in the early second stage region and is about
9 percent at the maxima. This is discussed in more detail later. The 803
patterns change as the reactions continue in the downstream gases. No
enhancement of 803 is observed beyond about 90 msec residence time. As seen
in the figure, the flue gases emerging from the reactor after 150-200 msec
contain the same amount of 803 regardless of the firing mode.
8
-------
2210 2050 1968 1918 1710 1635 1578
1515 T°K (Curvet A and E)
I I I I I I I
1865 laid 1835 1795 1735 1660
1505
III III
1765158515451520 U95 1445 1380
T K (Curve B and 0)
o
CO
S
C4
8
•o
•a
o
3.0
2.5
« 2.0
l.S
1.0
0.5
1260 T K (Curve C)
I I I
20 40 60 80 100 120
Mllllsooonda
140
160
180
200
Figure 4. Oxidation of S02 in Single and Staged
Combustion, 4>i = 0.95, 2 = 1.1.
At is the Delay Time Before Addition of
Secondary Air
-------
20
40 60
Milliseconds
80 100 120
Figure 5. Oxidation of S02 in Single and Staged Combustion, 1 » 0.99
Milliseconds
Figure 6. Oxidation of S02 in Single and Staged Combustion, <^^ = 0.90
10
-------
Curves B and C in Figure 4 present 803 data obtained from adding the
secondary air at two different positions in the postflame gases. Comparison
of these two curves shows that altering the distance between the burner head
and the introduction of secondary air produced some changes in the shape of
803 curves, particularly in the 40-150 msec range. Although maximum 803 levels
are nearly the same in either of the two-stage modes of firing, the depletion
of 803 appears to occur more rapidly when the addition of secondary air is
delayed several msec (Curve C). The initial rate of 803 formation is less,
however, when the secondary air is delayed. Curve C has an initial rate of
803 formation of about 5500 ppm/sec while the rate from Curve B is about
7000 ppm/sec. The difference is likely due to temperature effects.
Curve D in Figure 4 presents a calculated 803 profile for a single-stage
process and is discussed further in the kinetics discussion later in this report.
Figure 5 compares single- and two-stage firing at an overall equivalence
ratio of 1.2. Enhancement effects here, where $]_ is barely substoichiometrie,
are negligible.
The effect of operating the first-stage equivalence ratio at i=0.90 is
shown in Figure 6. No enhancement of 803 is observed at this lower equivalence
ratio. In fact, less 803 is formed in the postflame gases of this two-stage
process than in the single-stage process, at least in the mid-region of the
postflame zone probed.
Radial Profiles
In the course of developing the data for Figures 4-6, a question was
raised as to the effects of mixing patterns and the possible time dependence
of these patterns on the early postflame chemistry. Radial probings were
therefore carried out for 803, ®2» CO, and CO2 at various positions in the
secondary combustion region. In essence, mixing across the tube was com-
pleted within 10 msec after the addition of secondary air.
11
-------
SECTION IV
DISCUSSION
General Observations
Enhancement—
The data presented here tend to confirm Hedley's observation that
staged combustion can enhance 803 formation. However, our results
indicate 1) that the effect may not be a severe effect, 2) the enhancement
is of short duration, the 803 appearing to approach steady state conditions
about as rapidly as in single-stage combustion, and 3) the enhancement
effects and its duration are dependent on the air/fuel ratio of each stage
and the delay interval in the addition of secondary air.
Time Delay Effects—
It is obvious that delaying the addition of secondary air to the point
where the temperature is below that required to produce favorable conditions
(mainly 0-atoms) for S02 oxidation will prevent further 503 formation. It
follows that the formation of 803 would therefore decrease with decreasing
temperature at the secondary air ports. Barrett, et al., have commented on
this effect in their examination of the formation of 803 in a small combustor
using single- and two-stage firing modes (10). They concluded from their
studies that the addition of secondary air at temperatures below about 950 C
would likely produce little or no additional 803. This is not supported in
the present study. Adding the secondary air at about 850 C produced less
803 than when the air was added at higher temperature, and no enhancement of
803 was observed at the lower temperature. Nevertheless, about 10 ppm of
803 was formed with the addition of secondary air at 850 C and the 803
continued to increase slowly to about 15 ppm 803 in the postflame gases.
Thus, the "nonreactive temperature limit" may be somewhat lower than that
observed by Barrett, et al.
Air/Fuel Ratio Effects—
The trends in the 803 data observed at the different equivalence ratios
used in this study (Figures 4-6) can be rationalized to some extent by the
following considerations.
In considering the ultimate effects of different air/fuel ratios on 803
production in two-stage combustion, one might expect an increase in 803
production in the second stage with decreasing air/fuel ratio in the
12
-------
first-stage firing. This can be reasoned on the basis of an increase in
CO concentration with decreasing air/fuel ratio in the first stage, followed
by a greater enhancement of 803 from the CO oxidation chemistry in the
second stage. (No 803 is observed in the first stage at 4>]_ < 1.0). However,
the temperature also influences the chemistry here, limiting the effect of
the air/fuel ratio in the first stage. As the air/fuel ratio reaches well
below stoichiometric, the temperature of the first stage decreases. Also,
a larger amount of secondary air is needed to restore the second stage to
the desired overall equivalence ratio. Further cooling of the gases takes
place with a resulting overall reduction in the rate of CO oxidation, a
lower 0-atom concentration, and hence less 803 formation. The 803 data of
Figures 4 and 6 tend to confirm this line of reasoning.
On the other hand, approaching stoichiometric conditions in the first-
stage would increase the flame temperature to near a maximum, resulting in
an increase in the CO oxidation, thus leaving less CO to be oxidized in the
second stage. This could, within limits, lead to less 803 formation relative
to a richer first-stage firing. Data from the present flame probings do not
show any enhancement in 803 formation in a two-stage process at 4i = 0.99
(Figure 5).
The 803 data from this study indicate that optimum enhancement occurs in
the region 0.99 > 4»i > 0.90.
803 Fluctuation—
The data in Figures 4, 5, and 6 show an interesting, as yet unexplainable
but repeatable, discontinuity as 803 approaches its may-tninm. The authors have
observed similar fluctuation in their microprobing of H2S flames, which they
attributed to the oxidation of SO (11). Although a positive explanation is
lacking at this time, the discontinuity may reflect some mixing and/or wall
effects. Assuming that the fluctuation is real, it could mean that the rate
of 803 formation is being reduced momentarily. This reduction may be brought
about by excess CO molecules or H atoms present early in the second stage or
by excess 0-atoms just above the premixed single-stage flame.
Kinetics
The present study provides further confirmation of specific rate constants
for the formation and depletion of 803. Also, analysis of the rate data
further emphasizes the fact that staged combustion can enhance 803 formation.
13
-------
kl
Reaction 1, S02 + 0 + M -*•> SO- + M—
Analysis of the single stage data allow one to calculate a value
for kj_. Assumptions used in the calculation include:
1. There is negligible depletion of 803
via Reaction 2
2. The 0-atoms are equilibrated
3. Diffusion velocity corrections can be ignored.
The inclusion of Assumptions 1 and 2 in calculating ki are partially
validated in using the initial 863 formation rate, and realizing that the 803
formation occurs in the postflame gases, respectively. In the case of item
number three, diffusion velocity corrections are small in the early postflame
gases relative to the high convectional velocity arising from the reactor
design.
Taking these assumptions into account the calculated early rate of
formation of 803 from the data of Curve A, Figure 4, is approximately 7000 ppm
803/sec. At a constant level of 1500 ppm 802 and at a maximum flame tempera-
ture of 2250 K, where the equilibrium 0-atom level equals 430 ppm, one obtains
a value of 7.4 x 10^-^ cm** mole~2 sec~l for k^, if M is given an intermediate
value of 0.5. The results are in fair agreement with the room temperature
value of 1 x 10^5 cm^ mole"2 sec"1 reported by Cullis and Mulcahey (5). If k^
is temperature dependent, the experimental value is likely to be low.
k2
Reaction 2, S03 + 0 -=— S02 + 02 —
k_ is readily evaluated under steady state conditions with Reaction 1,
Evaluating this rate equation at the may*mum in the 803 versus time curve,
Curve A, Figure 4, gives
(S02) (0) (M) = k2 (S03) (0)
or
k2 (S02)(M) '
The latter relationship thus becomes independent of the 0-atom concentration.
SO
Substituting the ratio of 3 from the experimental data at the maximum in the
so2
Curve A and setting M = 0.5 gives
IT = 0.5*(P/RT) = 6'6 X 1()3 cm3 mole~1 (at 1685 K)'
14
-------
This value of k^/k2 agrees well with Fenimore and Jones ' estimate of
10* cm3 mole'1.
yields
Setting k]_ • 1 x 10 cm mole sec in the above expression
i, 1 x 1015 - e ,nll 3 , -1 -1
k« = - ;r » 1.5 x 10 cm mole sec .
6.6 x 10J
Comparing k,, with Cullis and Mulcahy's preferred value of 1.2 x 1012
cm3 mole'1 (5) yields
k2 = 7.0 x 1010 cm3 mole"1 sec'1 at T = 1685 K.
Our k2 value is in good agreement with the literature value. As discussed
below, however, the present work suggests a lower activation than that
suggested by Cullis and Mulcahey.
Two-Stage Enhancement —
The preceding analyses for k^ and k£ were derived from the single
stage flames where the calculations were carried out for the oxygen atoms
under equilibrium conditions. Similar calculations were not carried out for
the present two-stage experiments although the effect of temperature and various
equilibria on the 0-atom level in single and two-stage processes is discussed.
Qualitative statements on the application of the kinetics of enhanced 803
formation are given based on the similarity in Curves A and B in Figure 4.
Taking into consideration that the single-stage flames were at a
higher temperature (2250 K) than the two-stage flames (1950 K) , one notes
first off, that in spite of the temperature difference, the initial 803 rates
and the 803 may-Cma were similar. Inasmuch as the mechanism and kinetics will
be the same for 803 formation by either combustion process, the results provide
qualitative evidence that the 0-atom level in the lower temperature staged
process was increased due to the combustion of CO in the second stage.
Expressing k^ in the form (concentrations given in mole fractions)
k,
1
(S02)(0)CM)
calculations show that the 0-atom level in the second stage would have to be
5.5 times the equilibrium level (0£ * 0 + 0 at 1950 K) to produce the same
rate constant, k^ = 7.4 x 10^ cm^ mole~2 sec"1, observed in the single-stage
firing. Since this is not likely to occur, other 0-atom sources are considered
15
-------
Using the reaction sequence involving CO oxidation
CO + OH - CO. + H (4)
H + 0 = OH + 0 (5)
and assuming these reactions are equilibrated giving
(CO) (0 )
it is then possible to qualitatively account for a high 0-atom level in the
second stage combustion process. Taking the available CO, C0£» and 02
flame data and calculating an 0-atom concentration via Equation I, a value
of 1910 ppm is obtained at 1950 K. This is considerably higher than the
calculated 80 ppm 0-atom from 02 £ 0 + 0 equilibrium at 1950 K, suggesting
Reactions 4 and 5 are a good source of 0 atoms.
Since the 02 concentration at the entrance to the second stage
section will be close to that of the single stage process and since Req
decreases slowly with increasing temperatures, it appears from Equation I that
the equilibrated 0-atom levels depend primarily on the CO/CO2 ratio. In the
present studies, the (CO/CO.,)^ gtage to
-------
analysis Implies that, under comparable temperatures, 803 enhancement could be
as much as 100 percent in staged combustion, although it appears to be only a
transient condition.
One final comment is in order relative to the effect of CO concentra-
tion on 803 formation. The increase in 803 formation in the CO and COS flames
noted earlier (Figures 1 and 2) were some 200 percent compared with 803 produced
in methane or hydrogen sulfide systems. In the present study, on the other hand,
experimental observations show an increase of some 10 to 25 percent, or based on
kinetic analysis some 100 percent at best. The difference is probably a function
of CO concentration in the various systems cited. CO levels in the flames of
Figures 1 and 2 were some 5 to 10 times greater than the CO levels resulting
from staged combustion. Hence if we accept the Semenov (12) mechanisms
CO + 0 = CO* (6)
CO* + 02 = C02 + 20 (7)
for pumping oxygen atoms into Reaction 1, or the equilibrated reaction sequence
represented by Equation I, we logically can expect a lesser effect in a staged
combustion process than in a direct CO oxidation process.
CONCLUSIONS
It is concluded that 803 enhancement can take place in staged combustion,
but it appears to be a transient phenomenon and its effect may be less severe
than one would predict from currently accepted kinetic considerations for CO
and 802 oxidation mechanisms.
17
-------
REFERENCES
1. Johnson, G. M., Matthews, C. J., Smith, M. 7., and Williams, D. J.,
Combustion and Flame 15, 211 (1970).
2. Webster, P. and Walsh, A. D., Tenth Symposium (International) on
Combustion, p 463, The Combustion Institute, 1965.
3. Dooley, A. and Whittingham, G., Trans. Far. Soc., 42, 354 (1946).
4. Levy, A. and Merryman, E. L., Thirteenth Symposium (International)
on Combustion, p 427, The Combustion Institute, 1971.
5. Cullis, C. F. and Mulcahey, M.F.R., Combustion and Flame 18, 225 (1972)
6. Fenimore, C. P. and Jones, G. W., J. Phys. Chem. 69, 3593 (1965).
7. Hedley, A. B., J. Inst. Fuel 39_, 142 (1966).
8. Merryman, E. L. and Levy, A., Proceedings of the Second International
Air Pollution Conference, Paper CP7A, p 361, 1970.
9. Gaydon, A. G., Proc. Roy. Soc. A183, 111 (1944).
10. Barrett, R. E., Hummell, J. D., and Reid, W. T., Trans. A.S.M.E.,
J. Eng. Power, 88, 165 (1966).
11. Levy, A. and Merryman, E. L., Comb, and Flame £, 229 (1965).
12. Semenov, N., Chemical Kinetics and Chain Reactions, Oxford University
Press, 1935.
18
-------
TECHNICAL REPORT DATA
(Please read Iiuouetiont on the revene before completing)
1. REPORT NO.
EPA-600/7-79-002
4. TITLE AND SUBTITLE
Enhanced SO3 Emissions fi
2.
•om Staged Combustion
7. AUTHOR(S)
Earl L. Merry man and Arthur Levy
9. PERFORMING ORGANIZATION NAME AK
Battelle-Columbus Laborat
505 King Avenue
Columbus , Ohio 43201
ID ADDRESS
ories
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
3. RECIPIENT'S ACCESSION- NO.
6. REPORT DATE
January 1979
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO.
EHE624A
11. CONTRACT/GRANT NO.
Grant R805330-01-1
13. TYPE OF REPORT AND PERIOD COVERED
Final; 7/77-4/78
14. SPONSORING AGENCY CODE
EPA/600/13
IB. SUPPLEMENTARY NOTES T£RL-RTP project officer is W. Steven Lanier,
2432.
MD-65, 919/541-
ie. ABSTRAcrThe repOrt gives results of an experimental study to determine if staged
combustion can increase (enhance) the SO3 level in a combustion gas , relative to
that observed in a similar single-stage process. Methane flames doped with H2S
were used to examine the staging effects , emptying a small quartz tube two-stage
combustor. Examination of the staged combustion process suggests that the high CO
levels produced in the first stage may pump a sufficient level of oxygen atoms into
the second stage to increase SO3 formation. The results further suggest that the
enhancement is small and may only be a transient early postflame phenomenon
dependent on several combustion variables. A kinetic analysis of the data yields
values of: kl = 7. 4 x 10 to the 14th power cm to the 6th power mole to the minus 2
sec to the minus 1 for the SO3 formation process , SO2 + O + M = SO3 + M; and k2 =
1. 5 x 10 to the llth power cm cubed mole to the minus 1 sec to the minus 1 for the
depletion process , SO3 + O = SO2 + O2 (T = 1685 K). The kinetic analysis also
shows that enhancement of SO3 formation can occur in staged combustion.
17.
a. DESCRIPTORS
KEY WORDS AND DOCUMENT ANALYSIS
b.lOENTIFIERS/OPEN ENDED TERMS
Air Pollution Carbon Monoxide Air Pollution Control
Sulfur Dioxide Combustion Stationary Sources
Sulfur Trioxide Flue Gases Staged Combustion
Flames SO3 Enhancement
Kinetics Flame Kinetics
Oxygen Oxygen Atoms
18. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (This Report}
Unclassified
20. SECURITY CLASS (Tnifpage)
Unclassified
c. COSATi Field/Group
13B
07B
21B
20K
21. NO. OF PAGES
25
22. PRICE
EPA Form 2220-1 (9-73)
19
------- |