EPA-450/3-76-022
September 1976
PROPERTY OF
DIVISION
OF
NFTFOROLOGY
SO2 OXIDATION IN PLUMES:
A REVIEW AND ASSESSMENT
OF RELEVANT MECHANISTIC
AND RATE STUDIES
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Waste Management
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
-------
EPA-450/3-76-022
SO2 OXIDATION IN PLUMES:
A REVIEW AND ASSESSMENT
OF RELEVANT MECHANISTIC
AND RATE STUDIES
by
\ l.ei\, D.H. DM-VM",. and J.M. Hales
Battelle Pacific Northwest Laboratories
P.O. Box 99
Richland, Washington 99352
Contract No. 68-02-1982
Program Element No. 2AC129
EPA Project Officer: Joseph \. Tikvart
Prepared for
ENVIRONMENT\L PROTECTION AGENCY
Office of Air and \£ aste Management
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
September 1976
-------
This report is issued by the Environmental Protection Agency to report
technical data of interest to a limited number of readers. Copies are
available free of charge to Federal employees, current contractors and
grantees, and nonprofit organizations - in limited quantities - from the
Library Services Office (MD35) , Research Triangle Park, North Carolina
27711; or, for a fee, from the National Technical Information Service,
5285 Port Royal Road, Springfield, Virginia 22161.
This report was furnished to the Environmental Protection Agency by
Battelle Pacific Northwest Laboratories, Richland, Washington 99352 ,
in fulfillment of Contract No. 68-02-1982, Program Element No. 2AC129.
The contents of this report are reproduced herein as received from
Battelle Pacific Northwest Laboratories. The opinions, findings, and
conclusions expressed are those of the author and not necessarily those
of the Environmental Protection Agency. Mention of company or product
names is not to be considered as an endorsement by the Environmental
Protection Agency.
Publication No. EPA-450/3-76-022
11
-------
CONTENTS
Page
List of Figures ii
List of Tables iii
*
Sections
I Conclusions 1
II Introduction 5
III Thermochemistry of S02 Oxidation 9
IV Plume Studies 13
V Homogeneous Studies in the Gas Phase 30
VI Homogeneous Studies in the Aqueous Phase 42
VII Heterogeneous Studies 67
VIII Modelling Applications 77
IX References 80
X Appendix 89
-------
FIGURES
No. Page
1 Equilibrium Conversion SO, , > + H-O , « *- H-SO. , , 11
2 Dewpoint of H2S04, . 12
3 Sulfur Dioxide Decay Rates in Three Relative 19
Humidity Ranges (12/19/68, 1/2/69, 1/3/69)
4 Cross-Plume 03 and SC>2 Variations for Various 25
Downwind Distances
5 Effect of Acidity on Aqueous Phase Oxidation 50
Rate
6 Conversion of SC>2 to S03 in the Presence of 68
Several Catalysts
7 Hypothetical Potential Energy - Reaction 90
Coordinate Diagrams
-------
TABLES
No. Page
1 Equilibrium Constants for the Reaction 9
S°2(g) + 1/202(g)JS03(g)
2 Summary of Studies of S02 Oxidation in Actual 14
Plumes
3 S0_ Oxidation Studies - Colbert Steam Power ig
Plant Plume
4 Pseudo First-Order Rate Constants for Sulfur 27
in the Los Angeles Basin
5 Summary of SO- Oxidation Rates in Plumes 28
6 Estimated Rates of Theoretically Possible 32
Homogeneous Removal Paths for S02 in a
Simulated Polluted Atmosphere
7 S02 Lifetimes for Conversion to H2S04 in 39
Troposphere
8 Estimated Rates of S02 Removal, T £ 300°K 41
9 Temperature Dependence of Equilibrium Constants 44
for the S02-H20 System
10 Summary of Aqueous Phase S02 Mechanisms 45
11 Effect of Acidity on Reaction Velocity 49
12 Effect of Various Solutions on Sulfate 53
Formation
13 Nomenclature and Values Used for Plume SO2 60
Oxidation Calculations
14 Summary of S02 Reactions on Various Particulate 73
Species
111
-------
SECTION I
CONCLUSIONS
The available literature concerning the oxidation of SO- in plumes
does not allow firm quantitative conclusions to be drawn concern-
ing absolute rates or mechanisms. Reported rates cover a wide
range, and few details of the relevant mechanisms are well estab-
lished. Despite these uncertainties, however, one can state with
some confidence several characteristics of in-plume SO- conversion
processes. It is known, for example, that many factors strongly
influence the rate of oxidation of SO-. The evidence collected
thus far indicates that the gas phase oxidation rate increases
with increasing relative humidity. The concentrations of catalysts
(in the case of heterogeneous oxidation) and reactive species (for
homogeneous oxidation) also exert a profound influence on the rate.
In aqueous systems, pH, temperature, and catalyst concentration are
perhaps the most significant factors in determining the oxidation
rate. Furthermore, several complex mechanisms may be involved in
the oxidation under various conditions. Indeed, considering the
magnitude of the problem, it would be extremely fortuitous if a
single rate expression could be applied under all meteorological
conditions and all ambient atmospheres for all plumes generated
by all types and grades of fossil fuels. This, however, is little
comfort to the modeller, whose job it is to reduce the complex
system to a managable number of variables, relate these variables
with a sound mathematical formalism and perform calculations to
yield results which conform qualitatively and (hopefully) quanti-
tatively to the actual situation existing in the atmosphere.
-------
Unfortunately, the development of an entirely general model is
highly impractical. It may, however, be possible to model vari-
ous specific situations with a degree of success by using avail-
able data with appropriate parameterizations, the nature of
which will depend upon the exact requirements of the model.
The oxidation of SC^ has been studied both in the laboratory and
in the field. The results of those studies are divided into
four broad classifications and summarized below.
Plume Studies
1) Measured SO,, oxidation rates range from 0 to 55%/hr.
The dependence of the rate on meteorological variables,
especially relative humidity, is recognized, but there
is no way to control these variables in field experi-
ments.
2) Recent studies indicate lower rates than previously
reported, but as yet only one study has measured sul-
fate formation directly.
3) It is recognized that the rate of oxidation varies
spacially in the plume; the results of one study, how-
ever, indicate a high rate near the stack and a lower
rate downwind, while another suggests the opposite
situation.
4) Two recent studies indicate a correlation between ozone
in the plume and the oxidation of SC, with the oxida-
-------
tion rate increasing downwind as ozone returns to ambi-
ent levels after having been depleted by reaction with
NO in the early plume.
5) Hydrocarbons may promote SO- oxidation via photochemical
smog processes.
Homogeneous Gas Phase Studies
1) Rates are expected to be from 1.6 to 13.3%/hr.
2) Evidence indicates that the reaction of SO- with OH may
be the most significant process in the atmosphere, but
OH levels in plumes are not known precisely.
Aqueous Phase Studies
1) Reported rates range from 0 to 20%/hr.
2) Dissolved manganese and iron are the most efficient
catalysts, and both species are expected to be present
in plumes.
3) Hydration of aerosols, particularly manganese sulfate,
can form an aqueous phase capable of oxidizing SO- at
an appreciable rate.
Heterogeneous Studies
1) Oxidation rates range from 0 to 6%/hr.
2) Lead and iron are efficient catalysts promoting high
reaction rates; vanadium, surprisingly,- appears to be an
inefficient catalyst in plumes.
-------
It cannot be overemphasized that the oxidation of SO- in the
atmosphere is a highly complex process, and that the rate is and
should be highly dependent on the nature of the plume and the
existing meteorological conditions. Because of the extreme vari-
ability of data pertaining to atmospheric SO- oxidation processes,
it is difficult to provide any meaningful recommendation on rate
expressions to be used for practical modeling purposes. It can
be stated with some certainty, however, that insofar as modeling
for scoping purposes is concerned, the results available thus far
do not justify the application of rate expressions any more elab-
orate than the psuedo-first order form 2 _ , ,cn *
dt" * ( 2''
Since S02 oxidation processes are typically highly complex, they
will not be expected to adhere to the above linearized expression
accurately over extensive ranges of conditions. The basically
unsatisfactory state of this field, however, leaves us with lit-
tle reasonable choice but to proceed with this approach. Recog-
nizing these factors, and with a warning against indiscriminate
use, it is suggested that, for general purposes of model calcu-
lations, first order S02 oxidation rates in the range 0.5 to
10%/hr are consistent with the bulk of the literature, and a
reasonable value for many situations is 2%/hr. In the investi-
gation of specific mechanisms, however, it is suggested that
rate data for the specific reactions involved be used whenever
possible.
-------
SECTION II
INTRODUCTION
OBJECTIVE AND SCOPE
This report reviews the scientific literature relating to the
oxidation of sulfur dioxide in power plant plumes, and provides
recommendations on conversion rates to be applied in atmospheric
dispersion models, based on current knowledge in this field.
The importance of these two objectives derives both from the
presently confused state of our knowledge regarding atmospheric
SO- oxidation processes and from the potentially great problems
associated with our rapidly increasing sulfur dioxide emissions.
Recognizing these problems, a program has been initiated to pro-
vide a more complete means for evaluation of the atmospheric im-
pact of increased fossil fuel combustion. The first component
of this program has been a review of the behavior of plumes emit-
ted from large, elevated (primarily fossil fuel combustion) sour-
ces, which has been documented in the report Tall Stacks and the
Atmospheric Environment . From this report it is highly evi-
dent that our present ability to predict the atmospheric impact
of increased fuel utilization is limited by lack of knowledge
regarding in-plume SO- conversion processes. The present report
is intended to help rectify this situation somewhat by providing
-------
a summary of the limited existing knowledge regarding these pro-
cesses. A third and final report will describe a reactive dis-
persion model developed for use in conjunction with recommended
reaction rate parameters to evaluate the atmospheric impact of
power plant plumes, over moderately large scales of time and
distance.
CLASSIFICATION OF CHEMICAL REACTIONS REPORT ORGANIZATION
The reactions by which SO- may be oxidized are generally divided
into two groups identified by the terms 'homogeneous' and 'heter-
ogeneous '. Homogeneous reactions are those in which all react-
ants (including catalysts) are in the same phase (i.e. gas,
liquid, solid); heterogeneous reactions involve-more than one
phase. In the gas phase, S02 oxidation reactions may be either
homogeneous or, if particulate matter is also present, heterogen-
eous. In the aqueous phase, the significant oxidation reactions
are at least quasi-homogeneous; that is, all the reactants are
in solution, even if the aqueous phase consists of a water sheath
around an aerosol particle. Thus, the important oxidation
reactions to be considered in the atmosphere occur via gas phase
homogeneous and heterogeneous paths and by aqueous phase homogen-
eous paths.
(2)
In general terms, the title of a review by Urone and Schroeder
essentially summarizes the status of our understanding of S02
-------
oxidation chemistry: "SO- in the Atmosphere: A Wealth of Mon-
itoring Data but Few Reaction Rate Studies". It is of interest
that despite numerous studies on both the heterogeneous and
homogeneous aspects of S02 oxidation, there is no consensus on
the predominant path of this reaction in the atmosphere in
general, or in plumes in particular. There are more studies
supporting the predominance of heterogeneous paths, but some
recent studies seem to suggest homogeneous rates greater than
two percent per hour, higher than previously observed. In
addition, the difficulties inherent in assessing the importance
of the various paths, especially the aqueous phase reactions,
are compounded by an acute dependence on meteorological factors.
In this report, the oxidation of SO- is reviewed as follows:
Thermochemistry
Plume studies
Homogeneous studies in the gas phase
Homogeneous studies in the aqueous phase
Heterogeneous studies
Modelling applications.
The Appendix contains a brief review of thermochemical kinetics
and definitions of the basic chemical terms which will be used
throughout the report. It is included for the sake of complete-
ness and also as an aid for the non-chemist.
The literature concerning various aspects of the oxidation of
S0 has been reviewed by several authors (Urone and Schroeder
-------
(2), Kellog, et al. (3), Cadle (4), M. Bufalini (5), A. P.
Altshuller and J. Bufalini (6), Harrison, et al. (7), Pierrard
(8)). In general, the available literature pertaining to each
of the topics to be discussed in this report varies widely in
both quantity and quality, making the extraction of relevant
data and conclusions difficult. The aim of this review, then,
is to extract the data which are available from the literature,
however qualified they may be, and also to draw attention to
the areas most lacking in substantive conclusions.
-------
SECTION III
THERMOCHEMISTRY OF S02 OXIDATION
In the gas phase, the production and subsequent reaction of sul-
furic acid (H-SO^ in a plume involves the oxidation of S02 to
SO,, hydration of SO,, condensation of the resulting acid, and
reactions with particulate matter. Equilibrium constants for
the oxidation of SO- are presented in Table 1,
apparent
(9 )
It becomes
Table 1. Equilibrium Constants for the Reaction
-»
S02(g) * 1/2 °2(g) *S0
Temperature
°K °C
300
400
500
600
80
260
440
620
-1 -1/2
Equilibrium constant (Pcn ?_;: Pn ' )
atm-1/2 b°3 b°2 °2
2.069
1.088
2.608
4.892
x 1012
x 108
x 105
x 103
that, under conditions pertinent to gases leaving the stack
(<500 °K) and entering the plume (300-400 °K), the equilibrium
so heavily favors the formation of S03 that in the absence of
kinetic factors, there would essentially be no SO- present.
These kinetic factors, and the ways in which they influence
the oxidation rate, will be addressed in subsequent sections
of this report.
-------
Thermodynamically, the hydration of SO., to gaseous H2SO. also
becomes more favorable as the temperature drops. Figure 1,
from Leppard's review , shows that at temperatures existing
in plumes essentially all the SO- is converted to H^SO, under
equilibrium conditions. This conversion is also achieved rap-
idly: dry sulfur trioxide will form an aerosol almost immedi-
ately upon contact with moist air.
Thermodynamic values for the dewpoint of gaseous H2S04 have been
reviewed by Verhoff and Banchero . Figure 2, again from
Leppard, compares calculated and experimental dewpoint curves.
The rate of condensation is rapid, even at the concentrations
at which S03 or H-SO. is emitted from stacks.
The processes involved in the oxidation of S0_ in the aqueous
phase include absorption of S02 and other gases, hydration and
subsequent dissociation of the dissolved species, and oxidation
of sulfite or bisulfite ions. Because these processes are
highly dependent on thermodynamic factors, the appropriate equil-
ibria will be discussed later in conjunction with the relevant
kinetic factors. In general, however, the oxidation process in
the aqueous phase is thermodynamically favored, as it is in the
gas phase.
10
-------
en
O)
(VI
I
trt
o
c
o
0.8
0.6
0.2
0.0
100
200
300 400
Temperature,C
500
FIGURE 1. EQUILIBRIUM CONVERSION
S03(g) + H20(g) == H2S04(g)(10)
11
-------
100
o
c
(n
T.
Q.
Q_
0.01
Lisle and Sensenbaugh
Verhoff and Banchero
(for 12% water)
90 100 110 120 130
Temperature,C
FIGURE 2. DEW POINT OF H,
140
50 160
(10)
12
-------
SECTION IV
PLUME STUDIES
From a practical point of view, the most obvious or straightfor-
ward way to examine S02 oxidations would be to make measurements
directly in actual plumes. This approach has been carried out by
a number of laboratories, but unfortunately with minimal success.
Experimentally it is essentially impossible to control conditions
in the plume? added to this is the fact that such measurements are
very difficult to carry out. For this reason, results obtained
from such studies tend to conflict, making it difficult to draw
meaningful conclusions about plumes in general or to even compare
the studies to each other. Table 2 presents a summary of the
studies performed to date.
One of the first plume studies directed toward examination of
S02 - sulfate relationship, conducted by Gartrell et al. , of the
(12)
Tennessee Valley Authority (TVA) , has become nearly classic,
in that investigators often attempt to relate theoretical and lab-
oratory experimental results to the TVA results. The TVA program
attempted to measure both S09 and SO, in the plume; however, the
4» J
instrumentation of that period (early 1960) was not well suited
*
to making S0_ (or E^SOJ measurements. Table 3, taken from the
*The inability to make simple, reliable measurements of ambient
f^SO^ levels remains as possibly the most severe limitation to
studies of the fate of S02 in plumes today.
13
-------
Table 2. SUMMARY OF STUDIES OF S02 OXIDATION IN ACTUAL PLUMES.
Study
Species Measured
Gartrell,et at. SO,,, SO,
(1963) (12) 2 3
S02 relative to SFg
SO.
Dennis,et aZ.
(1969) <13>
Coutant.et aZ.
(1972)
Stephens and /-,,-\^02 relative to sub-
McCaldin (1971)U5'micron aerosol
Weber (1970)
(16)
Newman,e± al.
(1975) U7,18,.
S02 relative to C02
32S, 34S
University of Particulate
Utah (1975)(2°) S04 relative to
ambient S02
Davis and. Klauber NO, N02, Oo, S02
(1975) (2j->
Whitby, et al.
(1975) (23)
Aerosol
Comments
Highly variable oxidation
rates. Positive correla-
tion with humidity.
S02 half life from 1.0 to
2.8 hours (mean 1.5 hours).
Laboratory simulation,
various fuels. S02 loss
highly dependent on hum-
idity.
Oxidation rate dependent
on humidity.
Increasing oxidation rate
with increasing humidity.
Higher oxidation rates in
plumes from oil-fired
plants than coal-fired.
No specific humidity dep-
endence noted.
Low oxidation rates, pos-
sibly due to low humidity.
Ozone 'bulge1 noted down-
wind.
Results of single flight
Oxidation rates of 1.5 -
1.8 %/hr inferred.
14
-------
TVA study, presents a brief summary of the oxidation rates
resulting from measurements made in the plume of the Colbert
power plant, near Wilson Dam, Alabama. The results showdft|at
in periods of low humidity, SO- oxidation was quite low, rang-
ing from 1-3 percent per hour. At high relative humidity, a
rate of some 30 percent per hour (55% over 108 minutes) was
observed.
Probably the most significant finding coming out of this study
was the extreme variability of S02 oxidation rates. This
variability may have been the result of changing meteorological
conditions, the difficulties of analysis, or factors unknown.
It is of interest to note that in spite of the tendency of
later workers to attempt to relate back to these studies, it
was never the intent of the authors to consider this a defini-
tive study - in the words of the authors, "These limited data
do not provide a basis for an accura-te estimate of the absolute
rate of SO- oxidation after emission." The fact remains, how-
ever, that as the first major study of its kind, this work
provided a foundation for subsequent studies by pointing out
some of the important variables and giving at least a quali-
tative idea about the nature of S02 - plume chemistry.
15
-------
TABLE 3.
S02 OXIDATION STUDIES COLBERT STEAM PLANT PLUME
(12)
Date
1960
8/2
9/2
10/14
10/26
10/28
5/3
8/19
10/11
Sample
No.
1
2
3
4
1
2
1
2
1
2
1
2
1
2
3
1
2
1
2
Travel from
Time
(min)
5
5
5
15
30
78
12
60
6
84
12
84
13
13
13
108
23
12
96
Point of Emission
Distance
(miles)
"Low Rates"
.25-1
.25-1
.25-1
1-1.5
2-3
8
.5-1.5
5-6
.25-1.25
8-9
.5-1.5
8-9
"High Rates"
1.1
1.1
1.1
8-10
.75-2
.5-1.5
8
Relative
Humidity
in Plume
(Percent)
62
54
45
48
68
70
74
73
S02
Oxidation
(Percent)
0
0
1.20
0
3.70
2.20
2.15
3.23
1.50
2.70
1.10
4.10
13.80
10.00
19.20
55.50
8.00
21.60
32.00
16
-------
A few years after the TVA study, a study was carried out by Den-
nis et al. of the GCA Corporation , in which SCu losses in
stack plumes were determined downwind from a 330-MW coal-fired
power plant. Since it was not feasible to measure S02 directly,
S02 losses were determined relative to SFg, a commonly-used non-
reactive tracer. The results of this program were inconclusive.
The half-life of SCu under the field conditions ranged from 1.0
to 2.8 hours with a mean value of 1.5 hours. There were some dis-
crepencies in this study in extrapolating the field S02/SFg ratios
back to the expected source value for this ratio: Some extrapola-
tions suggested fairly rapid half-lives, such as 0.13 hours. Rela-
tive humidities in this program ranged from 36-53 percent.
As a follow-up to this study a laboratory program was carried out
by Coutant et al., at Battelle-Columbus Laboratories primarily
focused on determining if S02 decay was as rapid as the GCA data
(14 )
suggested within a half mile of the power plant . In this
program, a time-temperature simulation of the stack gases leaving
the furnace was established. The burner was fueled with three
types of coal (at 30 Ib/hr) and two oils (at 3 gal/hr) to assess
the variation of the SCu oxidation rate with fuel composition.
When burning coal, there was about a 10 percent loss of SO-, in
the flue gas within the boiler-economizer section, while there
was essentially no loss of S02 when burning oil. In the plume
region, loss of S02 was very dependent on humidity. First-
17
-------
order rate constants for the loss of SO- varied from 2 x 10
to 13 x 10~3 min"1 (t, ,_ of 1~6 hours).
Stephens and McCaldin carried out a series of measurements
in a power plant plume by assuming that particulate matter in
the submicron size range constitutes a conservative (non-varying)
(22)
tracer, and relating SO- measurements to this quantity. Friend
has taken exception to such a procedure on the grounds that SO-
can be oxidized to sulfate in particulate form, thus invalidating
the assumption that particulate matter is a conservative tracer.
This point notwithstanding, Figure 3 (taken from their study)
gives some appreciation of the nature of the results. Half-
lives for curves A, B, and C are:
Curve RH
A
B
C
30 -
40 -
78 -
40%
55%
80%
144 min
70 min
The results support the generally established concept that hum-
idity exerts a major influence in the oxidation of SO- in the
atmosphere.
Weber has made measurements of SO- decay using C02 as his
conservative tracer. His measurements were not made in an
elevated plume, but at ground-level sampling sites downwind of
a power plant. Using calculated C0-/S02 ratios for the effluent
18
-------
I _
I0~ -
A 12/19/68
01/02/69
O CH/OS/69
76-80% R/H
I I I I 1 I I I I I
40 BO 120 ISO ZOO
Pumt A«e, mn
FIGURE 3. SULFUR DIXOIDE DECAY RATES IN THREE RELATIVE
HUMIDITY RANGES(15)
19
-------
leaving the stack of a power plant in Frankfurt am Main, he then
observed C02/S02 ratios at various distances from the plant. His
results only allow for generalized statements on rates of oxida-
tion. He reports rather short half-lives (20 minutes to 1 hour)
depending on meteorological conditions (which are not well-
defined) . He also reports an increase in AC02/ASO_ with increas-
ing relative humidity, indicative of an increasing oxidation rate
(here, A indicates the difference between the peak and background
levels).
The Brookhaven National Laboratory (BNL) has been carrying out
fairly extensive studies of S02 oxidation in plumes of oil
(18)
and coal-fired power plants. The BNL studies were carried
out with the aid of single engine aircraft outfitted with a high
volume sampler for collecting and measuring sulfate and with S02
absorption scrubbers for applying the BNL isotopic ratio tech-
n g\
nigue . The particulate sulfate catches were too small to be
adequate for accurate analysis. The isotopic ratio technique
depends on measuring the changes in the ratio of the two most
32 34
abundant isotopes of sulfur, S/ S. Since the technique measures
the sulfur originating with the fuel, it can theoretically
discriminate between sources, and thus lends itself to the study
of a specific source.
20
-------
Small deviations from standard isotopic sulfur ratios are expres-
sed by a 'del value1:
5S =
"32 34
S/ S (standard)
- 1
x 1000
S/ S (sample)
If isotopic equilibrium is attained between S02 and SCU so that
34so2 + 32so3 5 32so2 + 34so3
the equilibrium constant can be written in terms of
-------
The authors speculate that a catalytic mechanism may be the domi-
nant pathway for oxidizing S02 in plumes and refer to indigenous
vanadium as a potential catalyst. This conjecture is not to be
discounted, but in the same vein, one might also expect sufficient
Fe203 in the particulate matter from the coal-fired plant to cata-
lyze the oxidation of SO-, and there is growing doubt concerning
the catalytic efficiency of vanadium in plumes. (The reader is
referred to the section on heterogeneous oxidation of S02 for other
comments on vanadium catalysis.)
The BNL investigators attempted to fit the decay to first and
second order kinetics, and based on a second order mechanism
arrive at a rate constant of 1 ppm~ hr , which accounts for the
10 hour half-life referred to earlier. No specific dependence
on humidity was noted in this analysis, although the RH levels
varied from 40-95 percent. In fitting their data to a second
order mechanism the authors do not mean to suggest that the oxi-
dation process arises from a bimolecular reaction of S02, but
that the process involves the reaction of S02 with active part-
icles in the plume through water associated with these particles.
It is their opinion that any homogeneous mechanism is inconsis-
tent with the isotope ratio data.
A study carried out by the University of Utah Research Instit-
ute at the Four Corners power plant in New Mexico in 1974,
also provides evidence for low SO- oxidation rates in coal-
fired power plant plumes. This study was part of an on-going
22
-------
three-year program to measure the degree of conversion of S02
and to evaluate its impact on atmospheric sulfate concentrations
and visibility. S02 conversion rates were obtained by determin-
ing the difference between upwind and downwind concentrations of
particulate ammonium sulfate relative to ambient S0_ and wind
speed. Measurements in these tests were made 13 and 37 miles
downwind from the plant. Conversions of S02 to particulate
sulfate were as follows:
Test 1 0.37%/hr 13 miles downwind RH 36-51 percent
Test 2 0.76%/hr 13 miles downwind RH 26-40 percent
0.45%/hr 37 miles downwind.
S02 concentrations ranged from 0.01-0.06 ppm in these tests.
These results are of interest for several reasons. They are
substantially lower than the oxidation rates reported by TVA,
possibly because of the low relative humidity at the Four
Corners location. Of greater significance, the resultant S02
oxidation rates are derived from direct, positive analysis of
sulfates and not inferred from the loss of S02 relative to
some intrinsic tracer in the plume. It is quite possible that
in this respect the Four Corners tests provide a more accurate
assessment of SO- oxidation rates in plumes.
(21)
A recent study by Davis and Klauber is cited here,
not so much for what it adds to our knowledge on S0_
23
-------
oxidation rates, but because it introduces rather different
thoughts on the interaction of S0_ in plumes. They col-
lected extensive data on NO, N0~/ and 0- and less extensive data
on S02 in the plume of a 1000-MW power plant firing a mixture of
oil and coal. The significant observation in this study was the
"ozone bulge" depicted in Figure 4. In these profiles the plume
boundaries are well characterized by the drop in the ozone
concentration. This is readily accounted for by the rapid
reaction between NO and 0, (NO + 0- -» N02 + 0-) . At 50 km down-
wind, however, one observes an increase in ozone (the "bulge")
coincident with low levels of S02 measured as "plume SO-".
The authors suggest a series of rather questionable reaction
steps to explain the bulge. Of interest to this review, how-
ever, is the fact that these reaction steps, if correct, would
probably lead to H-SO. and aerosol products which would have
some bearing on the rate of oxidation in the atmosphere. The
general reaction sequence suggested in this work is:
°2
OH + M > HS03 -> HS05
N02 + HS04
HSO, °2
b
2NO
2N02 + HS04 <-
°3 .
24
-------
PP6
60
40'
20
0
60
40-
20-
0-
eo
40
20-
o-
1300 EST 23 NOV., 1973
1.6 KM OUT
4.8 KM OUT
11.2 KM OUT
16 KM OUT
2024
KILOi-'ETERS (APPROX.)
OOpptr
60 ppfr
M50-I80OEST 22 JUNE, [974
0, ZPP"
J\l
SO.
1.6 KM
100-
so-
so,
100 -i
50-
0
40 KM
10 KM-
FIGURE 4. CROSSPLUME CK AND SO^ VARIATIONS FOR VARIOUS
DOWNWIND DISTANCES
-------
These experiments are reported here with no further comment.
Some other work by this group is discussed in the section on
homogeneous gas-phase reactions.
(23)
Whitby et al. conducted sampling flights through the plume of
a large power plant near St. Louis, Missouri, as part of the Envi-
ronmental Protection Agency's Project MISTT (Midwest Interstate
Sulfur Transport and Transformation). Using the data from one
particularly good run in 1974, they estimate SO- conversion rates
of 1.5%/hr between 10 and 21 km, and 1.8%/hr between 21 and 32 km
downwind. The relative humidity was 75% and the sulfur flow
appeared to be constant at 4.08 kg/sec.
(24)
Roberts and Friedlander have recently carried out some calcu-
lations which permit estimates of gas-to-particle conversion rates
for sulfur in trajectories in the Los Angeles basin. These analy-
ses are not based specifically on power plant plume trajectories,
but do provide interesting results of pertinence to the oxidation
of sulfur dioxide. The authors used total filter and cascade
impactor aerosol samples from several sites in the Los Angeles
basin. Depending on source and wind direction, SO^/Total S ratios
ranged from 0.88 to 0.98. Estimates of pseudo-first-order rate
constants, which appear to be rather high, are presented in Table
4. The accuracy of these data, like all data of this type, is
strongly dependent on the method of calculation, the measurement
technique, and such complex mechanistic parameters as ozone, free
radicals, and humidity.
26
-------
Table 4. Pseudo First-Order Rate Constants for the
Reaction SO^+1/2 02 k S03 in the Los Angeles
Basin (24).
Date Time of arrival
(1973) at Pasadena
(PST)
k, %/hr
Starting
location
July 10
July 25
July 26
1300
1400
1500
1600
1400
1500
1600
1200
1300
1400
1500
1.2
3.0
9.0
13.0
12.8
8.2
8.8
4.3
5.6
7.6
4.7
El Segundo
Alamitos Bay
El Segundo
Alamitos Bay
27
-------
Table 5 summarizes some of the SO- oxidation rates and SO,, half-
lives reported for measurements made in plumes. The data are
presented in this table in chronological order of publication.
It is interesting to note in this report the general "slowing
down" of S0_ oxidation rates in more recent studies. The signi-
ficance of this observation is not clear, but it may be indica-
tive of the better, or more reliable, experiments that can be
carried out with today's equipment. In essence however, truly
reliable estimates for S02 oxidation rates in plumes are still
wanting. None of the data reported to date can be used without
qualification, although this is not meant to fault any of the
studies. Primarily the results show the difficulty of such
TABLE 5. SUMMARY OF S02 OXIDATION RATES IN PLUMES
SO- Oxidation , .
%/hr rl/2' nrs
Gartrell et al .
Dennis et al-
Weber
Stephens and McCaldin
Coutant et al
University of Utah
Newman et al*
Newman et al
Whitby et al.
0-4
8-55
1.0-2.8
0.3-1
1.2-2.4
0
1-6
0.37-0.76
0
10
1.5-1.8
R. H . , i
45-70
73
36-53
40-80
< 35
40-90
26-50
\, Reference
12 (1963)
12(1963)
13(1969)
16 (1970)
15(1971)
14(1972)
20(1975)
18(1975)
17(1975)
23(1976)
28
-------
studies and the need for considerably more data. The increase
of S0_ oxidation rate with increasing relative humidity is
fairly well accepted; on the other hand, the results do not
provide overwhelming support of this conclusion. Similarly,
the dependence of S02 oxidation rates on heterogeneous cataly-
sis is accepted by many, yet none of the plume experiments has
attempted to link these variables. In summary then, one can
only say at this point that SO,, oxidation rates in plumes may
vary from essentially zero to the order of 50 percent per hour,
A clearer description of actual S02 oxidation processes in
plumes will have to await considerably more experimentation.
29
-------
SECTION V
HOMOGENEOUS STUDIES IN THE GAS PHASE
Until rather recently it was generally felt that S02 must be
oxidized by heterogeneous (catalytic) paths and that the contri-
bution of homogeneous gas phase reactions was minimal at best.
Subsequently, however, there has been reason to challenge this
assumption. Although one cannot by any means downplay the
potential role of heterogeneous catalysis in S02 oxidation,
several recent studies of S02 oxidation in power plant plumes,
cited in the previous section, show lower rates of oxidation
than were first reported rates well below two percent per
hour. It is quite possible that rates of S02 oxidation derived
from earlier studies were too high because of filter media ef-
fects and the poor analytical techniques available (problems
which persist to this day) as well as the indirect determina-
tion of sulfate formation. Thus, as the reported heterogeneous
rates get lower, the homogeneous reaction paths increase in
relative importance. This trend has approached the point, in
fact, that some investigators argue that homogeneous reaction
paths may be at least as significant as heterogeneous paths.
It is not our objective to argue this point here; the primary
point is that homogeneous reaction rates cannot be arbitrarily
assumed to be negligible compared to heterogeneous rates.
30
-------
Homogeneous gas phase oxidations proceed via second or third
order processes between SO- and molecular or free radical oxidi-
zing agents. Table 6 presents a compilation of homogeneous
reaction paths drawn up by Calvert . Although the merits of
some of these reaction steps may be questionable, they do show
that significant reaction rates can be expected from homogeneous
reactions in a plume. The compilation is of special interest
in that it offers potential support for rates of 1.7-4.7 percent
per hour.
Since the publication of this compilation, Castleman and the
(2 6 ^
group at Brookhaven ' have examined the reaction SG>2 + OE -» HOSOl
in some detail. They report a rate constant of 6 x 10 cm
-1 -1 -1 -1
molec sec (887 ppm min ) for this reaction. If the OH
radical concentration is 10 molecules/cm , SO- conversion rates
of 2 percent per hour would be possible. This conversion rate
is of course very dependent upon the OH concentration in the
plume: if the assumed OH concentration is an order of magnitude
too high (a possibility) then the rate of ~0.23 percent per hour
*
reported in Table 6 would agree with the Castlemen rate data.
In the Castleman study, the reaction S03 + H20 + M + H2S04 + M is
*
A basic dilemma that will become apparent in the following dis-
cussion is the reliability of the rate data relative to the con-
centration data of radical species in the atmosphere. Obviously
this dilemma becomes even more severe in chimney plumes where
the concentrations of radical species are even more in doubt.
31
-------
TABLE 6. ESTIMATED RATES OF THEORETICALLY POSSIBLE
HOMOGENEOUS REMOVAL PATHS FOR SOo IN A
SIMULATED POLLUTED ATMOSPHERE^6'
SO + 1/2 0 + Sunlight - SO
0(3P) + S00 + M - SO. + M
L j
o3 + so2 - so3 + o2
NO + SO - SO + NO
H03 + S02 - S03 + N02
V5 + S°2 - S°3 + "A
c(±>CH2 + S02 . S03 + 2CH20
CH200 ' + S02 - S03 + CH20
CH = CUO + SO - SO + CH 0
HO + SO - HO + SO
- H02SO'
CH302 + S02 - CH30 -f S03
^ CH302S02
HO + SO - HOSO'
CH 0 + SO - CH OSO '
AH298,
kcal/mole
-24
-83
-56
-10
-33
-24
-81
~ -117
~ -85
-19
< -25
-30
< -25
~ -82
~ -73
Approximate
Rate, %
per hr
< 0.021
0.014
0.00
0.00
0.00
0.00
< 0.4-3.0
< 0.4-3.0
0.85
7
~ 0.16
?
~ 0.23
-0.48
Total potential rate of conversion of SO to SO, (or Sulfates)
* 1.7-4.7% per hour.
32
-------
*
also examined. Using fast-flow reaction techniques the pseudo-
bimolecular rate for homogeneous reaction was found to be 9.1 x
10~13 cm3 molec"1 sec"1; the actual termolecular reaction rate at
1.3 Torr and 300°K is thus 2.2 x 10~29 cm6 molec"2 sec"1. (The -
proportionality factor between the two values is, of course, [M]
at the experimental temperature and pressure.) Assuming a four-
center intermediate complex for the formation of H_SO., this
appears to be a very rapid rate.
One of the basic questions in assessing the homogeneous rate of
SO- oxidation in the atmosphere is how fast the reaction can pro-
(27)
ceed in a clean air system. Gerhard and Johnstone , in their
early work in this area, reported an oxidation rate of 0.1 per-
cent per hour in natural sunlight and rates of the order of 0.68
o
percent per hour in U.V.-irradiated systems (2950-3650 A) with NO,.
(28)
Calvert's group considered the photooxidation reaction from
basic principles. Based on new estimates for the triplet SCU
quenching rate with N2/ 0_ , H-0, Ar, He and NO this group esti-
mates that the maximum rate for the homogeneous oxidation is 1.9
(29 )
percent per hour. Friend disagrees with this estimate, believ-
*The intermediate of a reaction is a state in which the two react-
ants are bound in an energetic, short-lived complex. In many cases,
the intermediate is so highly energetic that it flies apart before
the rearrangements necessary for the formation of products can be
accomplished. However, if a third body (molecule or atom) collides
with the complex, it can draw off some of the energy, giving the
complex a longer lifetime and greatly increasing the chance of for-
mation of products; the reaction is then termolecular, and the third
body is designated M. In the atmosphere, of course, M is generally
N or 0 .
33
-------
ing that assumed quantum yields for SO- removal of 0.01-0.001 are
-9
much too high, and are actually closer to 10 . On this basis,
Friend would estimate the homogeneous rate of oxidation to be
near zero.
A number of investigators have considered the oxidation of SO-
and subsequent particle formation in the presence of NO- The
most pertinent considerations are those of Cox and Jaffee and
Klein . Cox considers the oxidation of SO- and of NO- and
their roles in the heteromolecular condensation of H^SO. aerosols.
Using HO- as a possible free radical intermediate, the highest
SO- conversion rate one might expect from his data is 0.1%/hr.
This estimate is based on the reaction
S02 + H02 + S03 + HO
where k = 3 x 10 cm3 molecule' sec and the HO- and SO- con-
8 7
centrations are 5 x 10 molecules/cmj and 30 ppb respectively.
Cox's analysis shows that the rate of formation of H-SO. in urban
air is sufficient for hetermolecular nucleation to form aqueous
sulfuric acid aerosols, while in background areas SO- oxidation
products are removed mainly by condensation on existing aerosol.
Jaffee and Klein irradiated N02 in the presence of SO- at 3660 A, a
wavelength at which SO- is not excited, but NO- is photolyzed to NO
kl
and 0. The rate constant for the reaction SO- + 0 -* SO, is
9 -1 -1
reported to be 1.1 x 10 £ mole sec . When combined with the
34
-------
* ^2 * 3
deactivation processes SO, + 0 + S02 and SO, + M -> SO, + M,
k4
the rate constant k. for overall reaction S0_ + 0 + M -» SO, + M
is given by the expression
*
(The intermediate SO, is assumed to be at its steady state con-
1
centration.) The ratio k2/k, was found to be 0.077 mole L ,
10 2 -2 -1
making k. = 1.4 x 10 £ mole sec under the experimental
-4 -1
conditions where [M]c*10 mole £ (2 Torr). At atmospheric pressure/
M makes only a slightly more significant contribution (assuming
the mechanism is still valid), leading to a value of k. = 9.3 x
9 2 2 1 4 3
10 t mole sec . If we assume some 10 oxygen atoms/cm in a
o
plume, the half-life for the reaction is 1.1 x 10 sec (3.5 years),
implying that the process is negligibly slow.
Direct oxidation of S02 by ozone in the gas phase is also a slow
process. In a power plant plume the process is probably of even
less importance because of the rapid removal of ozone by the NO
(32)
in the plume. Cox and Penkett have considered the oxidation
of SO- in a system containing olefin and ozone. They postulate
that the S02 reacts with an intermediate product (I) resulting
from the reaction between ozone and olefin. The rate of SO, for-
mation is then expressed by RSO = k(S02)(I). The consequences
of these reactions may be greater in terms of long-range S02 trans-
port than in short-term plume chemistry due to the aforementioned
rapid reaction between 0, and NO.
35
-------
(33)
Smith and Urone have performed studies of the photochemical
oxidation of SO- alone and in the presence of NO-, propylene
(C-Hg), and water vapor. The initial SO- concentration throughout
the study was 2 ppm. In air containing only SO-, the initial
photochemical rate (dSO-/dt) was 1.74 x 10 ppm/min (about 0.55%/
hr). The introduction of NO- increased this rate when the SO-:NO-
ratio was 1 or 2, but decreased it when the ratio was 0.6 or less.
When both NO- and propylene were added, the rate increased by some
two orders of magnitude over the rate for the SO^-air system, and
was found to be a function of both the propylene and NO- concen-
trations. The addition of water vapor at 50% relative humidity
to the SO- - NO- system was found to increase the rate about ten-
fold over that for the dry system, but for unknown reasons a simi-
lar addition to the SO- - NO- - propylene system had no effect.
Wilson and Levy ' have also examined the smog process in
irradiated SO- - NO - hydrocarbon systems. This work was directed
at observing the effects of SO- in smog rather than the reverse.
It was generally observed in this work, however, that the decay
of S02 increased in proportion to the reactivity of the hydro-
carbons. No quantitative values came out of these studies.
In a current program under EPA-sponsorship, D. F. Miller is
studying the homogeneous rate of oxidation of SO- in photochem-
ical smog systems. The studies are being conducted in a large
610-cu-ft smog chamber, and while there are admittedly certain
36
-------
limitations to such a device as a basic kinetic tool, a definite
influence of hydrocarbon concentration is being observed in these
studies. Rates of S02 oxidation.,, which appear to be homogeneous
and not especially influenced by the chamber walls, are of the
order of 2-6 percent per hour. Increasing the propylene from 1.6
ppm (as carbon) to about 3 ppm doubled the homogeneous rate.
( 37)
Roberts and Friedlander have recently carried out experi-
ments designed to study the formation of sulfur-containing
aerosols under ambient photochemical smog conditions. These
experiments were carried out in a large (96 m ) Teflon chamber
irradiated with natural sunlight. - Seven olefins were used in
these studies, although most of the experiments dealt with 1-
heptene-S02-NO systems. Based on pseudo-first order deple-
tion of S02, the reaction rates varied from 0-90 percent per
hour, depending on the initial S0_ level. The relative humidity
was quite low (< 40%) , making the extremely high rates even more
surprising, although no explanation of the high rates was
offered. The authors do develop some theoretical kinetic
analyses using a mechanism similar to that of Cox and Peckett
(discussed earlier) and present an expression for estimating
aerosol sulfur content. The experimental results appear con-
sistent with a mechanism for the formation of the ozone-olefin
intermediate (I).
37
-------
(21)
Davis has also reviewed some aspects of homogeneous S02
oxidation kinetics with a slightly different approach from that
of Calvert. Davis considers the following processes and esti-
*
mates SO- lifetimes for the lower troposphere as shown in
Table 7:
0 *
(la) S02 + hu (2400-3400 A) * S02
(lb) SO* + 02 + (S04)
(2) S02 + 0 + M -> S03 + M
(3) S02 + 02(1A) + (S04)
(4) S02 + 03 -» S03 + 02
(5) S02 + NO3 -» S03 + N02
(6) S02 + N205 -* S03 + N204
(7) S02 + H02 » S03 + OH
(8) S02 + OH + M + HS03 + M
(9) S02 + CH302 + S03 + CH30
Reaction 8 is the most probable for the homogeneous oxidation of
S02 in the atmosphere. Thus, the structure and reactions of the
product of reaction 8, HSO,, is very critical to acid aerosol
formation in the atmosphere. It is on this basis that Davis
puts forth the extensive series of reactions on the oxidation of
HSO-. discussed in the section on plume studies.
*The term "lifetimes" is frequently seen in the literature, and
is equivalent to "half-life".
38
-------
Table 7. S02 LIFETIMES FOR CONVERSION TO H2S04 IN TROPOSPHERE
(21)
Reaction
1
2
3
4
5
6
7
8
9
Second Concentration
Species Molecules/cm
__ __
0 (3p) ~ 1 x 104
02('A) ~ 106
03 1 x 1012
NO- ~ 1 x 10
N205 ~ 6 x 10
H02 ~ 5 x 108
OH ~ 5 x 106
CH302 ~ 108
Lifetime days
4 x 107
~ 6 x 104
~108
~105
~ 1016
~109
~ 23
~ 3
100
39
-------
Some of the reactions presented in this discussion are summar-
ized in Table 8. It becomes very apparent when one considers
the role of homogeneous removal paths for S0_ that, although
the specific rate data are possibly good only within a factor
of 10, this uncertainty may not be the most serious problem to
assessing the role of homogeneous reaction paths. The various
concentrations of radical species presented in Tables 7 and 8
are not well established for conditions in the troposphere, and
the existing data concerning them may be even less reliable for
conditions in a power plant plume.
One might summarize this section with the conclusion that homo-
geneous S02 oxidation reactions can play a significant role in
atmospheric reactions under certain meteorological conditions,
but one must still question the significance of these reactions
in the power plant plume. At any rate, it is fairly obvious
that we are not confident of just how to apply homogeneous
reaction steps to mechanisms concerning the oxidation of SO-
in plumes.
40
-------
0)
u
; g
!! <"
;; &
1
3
|
ii
n
.*
i; I
V
;! as
ij
g
O
1
* 3
Psl '""" O
o 2-=
W « Wi
g QJ
fe |2 x c
° cc o u
Cfi !«4 C
CO a. a)
fij ' ^ U
E-> 1 ca u
§ ~£
as 9
Q
" § 1
g " ^
H «-?
£ ?" §'«
W H 1 tj "^
*" " »T 3
c u
5°
|| W *l
05 ij « 8
ii ^j
H [ coo
J as E
2 '' 'J
< ||
ii
|i
I
1
il
Ij
I
I!
,|
]
i
|
|1
J
!
i
!
>
CO <""^ ^ O <" ^O i" * f*» * ( 00 ^H (**J (N ^H "H ^H ^?
jor^ rt ^en cNtN «* JX
CN > o C n o 01 "
t£ 0_0 = = 0 Z S
1 -*
*H r^ ( *
mo-HOO* inom
OO O OOOOO \CONO>O vCO*^ CN OOO O
OO O C -HOCCC-I o O mo
-t CN
2
7
* * * * o *
occ r^rjn^-c1 CNCNCN mg?vc r^cy\ %y «ff
1 1 vC 1 1 1 1 1 III III II 1 CM 1 1
SO OOOOO ^-COO COO OO Ol O O
O -H
xx xxxxx mxxx xxx xx x x x
1 X
. IN .
-
O !
cs
""
_r^
^* O i
0 S ^
+ = i +o i
r-l < PI OOO
O O s en z CN
o co m !/: O
o O = + n + S
ii > ii coon
00 ^^ ct rg c/3
CN < *-s S ZC --^
O -H S T II , II S
w +. -^ + +
CN ^ -g > CN ^
! CNCN CN!
coo e o oo o;
1
f*)
*
w
!"*
o
*~*
^
m
.
CN
«
z
s
1
o
U
S!
"'
0
01
tu
^_
s
u
f-
j ii
a -C
C N
£ S
O
S w
V
3 AJ
t 2
s- o
* *
^
41
-------
SECTION VI
HOMOGENEOUS STUDIES IN THE AQUEOUS PHASE
A consideration of primary importance to the question of aque-
ous-phase S02 oxidation is the dissociation behavior of this
material in water. The fact that S02 does form a hydrate which
dissociates in water to yield an acidic solution has been recog-
nized for many years, but until relatively recently the rate of
this dissociation process and its products have been open to spec-
ulation. A great many authors, for instance, have presumed that
undissociated sulfurous acid, H2SO_, is formed upon dissolution
of S02 in water. Spectroscopic studies , however, indicate
that, while H.,SO, does not exist to any appreciable extent in the
S02~H 0 system, such species as undissociated SO-, HSO ~, SO ~,
pyrosulfite ions (eg. HS_05 ), and hydrates (S02'nH20) do exist
in measurable quantities. Furthermore, S02, HSO,~, and SO ~ are
the most abundant species, and for most practical applications
one can represent the dissociation process in terms of the
equations
S°2(g) 1° S02(aq)
Kl - +
S02(aq) + H2° t HS03 + H
- K2 = +
HS03 + S03 + H (3)
42
-------
Here the K's denote equilibrium constants defined by the follow-
ing expressions:
Pso2
_ [HSO
Kl - - f
[so2]
[HS03~]
Values of the equilibrium constants have been measured experi-
mentally, and are given as functions of temperature in Table 9.
These values can be utilized in conjunction with the above
equations to demonstrate that, under typical environmental con-
ditions, the preponderance of sulfur dioxide dissolved in water
exists in the form of HSCU ions.
The question of the rates of the dissociative processes (2) and
(3) has also been debated actively up until the very recent
past. Early semi-quantitative measurements indicating these
reactions to be extremely rapid were questioned in the sub-
sequent chemical engineering literature, and the quantitative
results of Wang and Himmelblau implied that the ionization
processes were sufficiently slow to be rate influencing under a
number of circumstances.
43
-------
Wang and Himmelblau were apparently unaware, however, that Eigen
(47)
et a.1. had performed a previous definitive measurement of the
rates of ionization of hydrated SO-, which strongly indicates their
later results to be in error. Eigen's measurements indicate that
ionization approaches completion at times of the order of 10
seconds, thus enabling these reactions to be regarded as essentially
instantaneous for this assessment of oxidation processes.
TABLE 9. TEMPERATURE DEPENDENCE OF EQUILIBRIUM
CONSTANTS FOR THE S02~H20 SYSTEM<48)
Temperature
25 °C
15
10
0
-3
1.
1.
2.
3.
3.
Ko
24 M atm'1
83
25
46
97
Kl
.0174 M
.0219
.0247
.0319
.0346
6
7
8
11
12
.3
.9
.9
.4
.4
K
X
X
X
X
X
2
10~ M
io-8
-8
10 B
io-8
-8
10 B
The literature pertaining to the aqueous phase homogeneous ox-
idation of SO- is in some ways more pleasing than that for the
gas phase (a summary is given in Table 10). For instance, there
is general agreement that in the aqueous phase, the oxidation of
S(IV)by o2 must be catalyzed in order to proceed at an appreciable
rate, and almost without exception the suggested catalyst has
44
-------
TABLE 10. SUMMARY OF AQUEOUS PHASE S02 OXIDATION MECHANISMS
AUTHOR
Fuller and Crist
1941
(49)
Bassett and
Parker
1941
(50)
Junge and Ryan
1958
(51)
Van den Heuvel
and Mason
1962
(52)
Espensen and
Taube
1965
(53)
Scott and Hobbs
1967
Beilke and
Georgii
1968
(55)
Foster
1969
(56)
Matteson,
Stober , and
Luther
1969
(57)
McKay
1971
(58)
Cheng , Corn , 4
Frohlinger
1971
(59)
Miller and
DePena
1972
(60)
TYPE OF MECHANISM
Sulfite oxidation by 02
with no catalyst;
Cu2* catalyst; mannitol
inhibitor
Sulfurous acid oxida-
tion by metal salts
and O2
SO2 oxidation catalyzed
by Fe2*, with and with-
out ammonia
S02 oxidation catalyzed
by NH3
S02 oxidation by Ozone
SO2 oxidation catalyzed
by NH3
Washout and rainout of
SO 2 and sulfate aerosols
S02 oxidation catalyzed
by metal salts
SO2 oxidation catalyzed
by metal salts
SO, oxidation catalyzed
by NH3
S02 oxidation catalyzed
by metal salts
SO 2 oxidation catalyzed
by NH3
1
ELEMENTS OF MECHANISM
very long chain reaction
uncatalyzed: formation of
complexes such as [02*503]";
catalyzed: formation of com-
plexes such as I02-Mn(S03)2l~"
and rapid oxidation
2S02 + O2 » 2SO3
S03 + H20 * H2S04 * 2H*+S04"
NH3 + H20 * NH4OH ~ NH4*+OH"
S02 -f 03 + H20 - HS04" *
+ H* + 02
(in acid solution)
S03" + 03 * S04" * O2
(in basic solution)
S02, NH3, C02 in equilibrium
with ions and gas
803" + 1/2 02 * SO4"
2S02 + 2H20 + 02
Mn2+ + SO2 I Mn-SO22+
2Mn-S022+ * °2
2Mn-S032*
Mn-S032* + H20 Mn2* + HS04"
+ H*
HS04" * H* * H^SO,,
ame as Scott and Hobbs
--o t -n o i o2 catalyst-
2H2S04
similar to Scott and Hobbs
RATE COEFFICIENTS
AND/OR EXPRESSIONS
d(O-)
3T-' k
k - .013 + 2.5(Cu2+)
with cu2+ catalyst;
k - .013 + 6.6 (H*)1/2
with acid added
conversion rate
= 1.8 x 10"4 %/min
conversion rate
= 2.5%/min
not determined
dj SO." 1
k = .0017 sec"1
assumed instantaneous
oxidation of all S02
incorporated into
droplet
S02 conversion rate «
.09%/min for Mn,
.15 to 1.5%/min for ?e
for typical input
parameters
d . K tin2*]2
dt - KI Mn J0
*l - 2.4 x 10s M"1 s"1
k - (.013 + 59[H*]2)
sec"1
.032 sec"1 at
neutrality (25°C)
SO2 conversion rate
^.03%/min with Mn2*
levels typical of urban
industrial atmosphere ;
\..33%/min with levels
typical of plume from
coal powered plant
k v .003 sec"1 (2S'C)
COMMENTS
25 °C; rate coefficient
with acid added assumes
second dissociation of
H2S03 equals 5 x 10~6
(too large)
sulfate formation
asymptotic to a certain
pH; maximum formation
varies linearly with
SO2 partial pressure
extrapolation to atmo-
spheric conditions used
to estimate conversion
rate
25"C only; k estimated
from Van den Heuvel
and Mason
calculated S04* concen-
trations from 0.3 to
36.5 mg/l depending on
rain rate and assumed
502 ga£ Pnase concentra-
tion
theoretical study; rates
for Mn and Fe are func-
tions of many factors;
rate far Fe catalyzed
oxidation is pR depend-
ent
negligible S04=
formation for RH <95»;
similar mechanism may be
responsible for catalysis
by other metal salts
considered variation
with temperature; k esti-
mated from Fuller and
Crist; found large nega-
tive temperature rate
correlation
oxidation rate estimated
by extrapolation to
atmospheric conditions
k estimated by fit of
experimental data to
theoretical curves
45
-------
TABLE 10. (CONTINUED)
AUTHOR
Penkett
1972
(61)
Chen & Barren
1972
(62)
Penkett and
Garland
1974
(42)
Srimblecomoe &
Speeding
1974
(63)
Freibera
1974
(64)
TYPE OF MECHANISM
SO2 oxidation by ozone
Sulfite oxidation
catalyzed by cobalt
ions
SO2 oxidation by ozone
S02 oxidation by 02
with trace Fe catalyst
SO 2 oxidation catalyzed
by Fe
ELEMENTS OF MECHANISM
HSO3* + 03 » HSO4" + 02
free radical mechanism;
Co (III) reduced
HS03" + 03 » HSO4" + O2
complex
complex
RATE COEFFICIENTS
AND/OR EXPRESSIONS
d(0,l
dt - k2(03) (HS03~)
kx - 3.32 t .13 x
loSM'is-l
t>02 conversion ra^e
v .21%/min
d'(02)
dt k[co(H2o)63+]l/2
[S03-J3/2
d(SO )
- ^r2 ' k[s°3 ]
k - 4.18 x 10"4 +
+ 1.77 [H+]1/2 sec'1
. dis^jui. MFe(11I),
[S(IV)J
k - 100 M-l s-1
S02 conversion rate
^3. 2 I/day in fog
assuming 28 ug/nH S02
and 10" M Fedll!
° s°/
dt KoKs l"2-'0J'
[Fe3*]/iH-']3
ks = 1st dissociation
constant of H2SO3
COMMENTS
9.6*C; SO2 oxidation
rate extrapolated from
data; much faster than
Scott and Hobbs
could not determine
specific value for k
pH range from 4 to 7
10°C; .1 ppm S02,
.05 ppm O3 in fog water
Possibility of Fe(III)
contamination discussed
Rate increases rapidly
with RH and decreases
by about one order of
magnitude with 5*C
increase in temperature.
45a
-------
been either ammonia or a metal salt or oxide, most likely Fe(III)
or Mn(II). (Some investigations have postulated reaction in the
aqueous phase of a deliquescent aerosol particle; in mechanisms of
this type, the classifications homogeneous and heterogeneous lose
much of their meaning, and these reactions are rather arbitrarily
included in this section of the report.) Aqueous phase oxidation
of SO- by 0, has recently been measured and reported. A very lim-
ited amount of work has been done on photochemically induced oxi-
dation in the aqueous phase, but it is generally discussed as a
side reaction in (and dependent upon) the photoxidation of olefins
and reactive hydrocarbons . Because of the low output of these
species from power plants, it is expected that photochemical oxi-
dation of SO- in the aqueous phase is not a significant source of
sulfates in power plant plumes.
The following discussion of the oxidation of SO- by 0_ in the
aqueous phase will assume (unless otherwise stated) that the
reaction is either
HS03" + 1/2 02 * HS04" or
S0= + 1/2 0 *
It will be clear from the context which reaction is being discus-
sed. This section of the report will consider first some general
work on SO- oxidation in the aqueous phase, and then the specific
effects of ammonia and metal salts. Finally, oxidation by ozone
and photochemical processes will be discussed briefly.
46
-------
In general, it should be reiterated that the aqueous phase
oxidation of SO- is highly dependent on meteorological condi-
tions, since even the fastest mechanism can produce no sulfate
if there is no aqueous phase present. Furthermore, even when
there is an aqueous phase present, scavenging efficiency,
solubility, and diffusion must be considered. For these
reasons, aqueous phase oxidation must be considered somewhat
differently than gas phase oxidation when applied to power
plant plumes.
Although oxidation of sulfite ions and "sulfurous acid" by
oxygen was investigated as long ago as 1898 , much of
this early work cannot be considered quantitatively since,
among other things, there is often no record of purity pre-
cautions applied to regents. These works did show the oxida-
tion to be a long chain reaction, and as such to be quite
sensitive to both positive and negative catalysts (catalysis
and inhibition), but the details of the mechanism are still
uncertain. In his investigation of the effect of inhibitors
on the reaction, Backstr5m found that his data fit the
relation
v = A/(B+m)
(where v is the velocity of oxygen absorption, A and B are
constants, and m is the inhibitor concentration) although
deviations were noted for low values of m.
47
-------
The earliest definitive work on the process was reported by
Fuller and Crist (49' in 1941. In the absence of added catalyst,
their data showed the reaction SO ~ + 1/2 02 + sO/~ to be first
order with rate constant k, = 0.013 + 0.0015 sec . The work was
performed at 25 °C with pure oxygen atmospheres. When the inhibi-
tor mannitol was added, the data were found to fit an expression
similar to Backstrc-m' s, and over a wider range of inhibitor con-
centrations:
d(S0=)
where A is about 10 . The catalytic effect of Cu was also
investigated, and it was found that the data fit an expression
of the form
d(S0
3
dt
(Cu
+
(S03 )
and k, was determined to be 2.5 + 0.33 x 10 M~ sec" . The ex-
j ~~ ~
treme sensitivity of the reaction to catalysis by Cu++ can be
-9
seen by noting that at concentrations greater 5 x 10 M,
k3 (Cu ) > k-,. The effect on pH on the oxidation was also invest-
igated, and as noted in previous (and subsequent) work, the
oxidation rate increased with increasing pH (see Table H and
Figure 5). This is consistent with the assumption that the sul-
fite rather than the bisulfite is the species being oxidized;
the addition of H+ shifts the equilibrium (HSO~ ^ H + S03=) to
48
-------
TABLE 11. EFFECT OF ACIDITY ON REACTION VELOCITY
(49)
Hydrogen
ion
added ,
m/1
0.0032
.0065
.0097
.0130
.0162
Time,
sec.
0
20
40
60
80
0
20
40
60
80
0
20
40
0
20
40
0
20
Total
Sulfite
Concn . ,
m/1
0.0164
.0129
.0096
.0072
.0055
.0160
.0121
.0094
.0078
.0069
.0145
.0119
.0104
.0160
.0139
.0131
.0174
.0164
Sulfite
ion
Concn. ,
' m/1
0.0132
.0097
.0064
.0040
.0023
.0095
.0056
.0029
.0013
.0004
.0048
.0022
.0007
.0030
.0009
.0001
.0012
.0002
Av
pH
5.92
5.77
5.60
5.40
5.15
5.47
5.24
4.96
4.60
4.09
5.00
4.66
4.16
4.66
4.14
3.19
4.17
3.40
= 6.6 = 0.47
kl
1.9(a)
5.4
5.9
6.3
6.3
7.0
6.6
6.9
6.8
7.2
7.6
6.9
6.1
(a) This figure was not used in calculating k (average).
49
-------
100
Tiirc in seconds.
200
FIGURE 5. EFFECT OF ACIDITY
(49)
Curves 1 to 8 refer respectively to 0.02 molar solutions of sodium
sulfite to which 0, 0.0032, 0.0065, 0.0097, 0.0130, 0.0162, 0.0195,
and 0.0325 mole per liter of hydrogen ion have been added as
sulfuric acid.
50
-------
the bisulfite side, decreasing the sulfite concentration. This
effect leads to the expression
d(S03 )
dt
+ k4 (if)
(S03 )
-1/2 -1
where k. is found to be 6.6 + 0.47 M ' sec . This value is
calculated assuming the second dissociation constant of sulfur-
ous acid to be 5 x 10 ; more recent research indicates this
value is actually closer to 6.3 x 10 , leading to a better
1 /2 1
value of k = 59 M x sec .
While the details of the chain mechanism responsible for the
oxidation of SO., are not certain, work by Bassett and Henry
and Bassett and Parker led to the conclusion that the un-
catalyzed reaction proceeds via ionic complexes such as (0-*
2 2
SO,) and (O^'S-O,.) . The manganese ion-catalyzed reation was
2-
postulated to involve an intermediate such as (02"Mn(SO,)2)
which could rapidly undergo self oxidation and reduction. The
intermediates in the case of cobalt-, nickel-, and iron-catalyzed
oxidation were assumed to be similar but less active complexes.
Quantitative rate measurements were not performed in this study.
One of the first attempts to study the role of S02 oxidation in
air chemistry was done by Junge and Ryan . Their work showed
51
-------
that the uncatalyzed reaction produced a negligible amount of
sulfate, and that it was not photosensitive in normal sunlight.
In the presence of a catalyst (Fed,,) the sulfate formation was
found to reach a limiting value after a period of time (typi-
cally'one" to" three hours) , and the final sulfate concentration
depended linearly on the initial S02 concentration for a given
catalyst concentration. Furthermore, it was found that the pH
of the solution dropped during the course of the reaction, and
that very little additional sulfate was formed once the pH
reached 2.2.
Junge and Ryan also did a preliminary study to determine the
effectiveness of various salts (at the same concentration by
weight) as oxidation catalysts, and found manganese to be most
efficient (see Table 12). This study was not,meant to be ex-
haustive, however, and it should be kept in mind that in the
atmosphere, the concentrations of various catalysts are not
equal, and thus a less efficient (but more abundant) catalyst
may actually be more important in sulfate production under true
atmospheric conditions.
*
AMMONIA CATALYSIS
Junge and Ryan investigated the enhanced effectiveness of metal
catalysts in the presence of NH3, and concluded that the NH,
It may be argued that promotion of the oxidation of SO- by
ammonia is not catalysis in the true sense of the word; never-
theless, the term will be used for convenience in the following
discussion with a less-than-formal definition.
52
-------
TABLE 12. THE EFFECT OF VARIOUS SOLUTIONS ON SULFATE
FORMATION(51)
Solution of
51 x 104 yg/m3 S02 and
MnCl2 (1 yg/cm3)
CuCl2
FeCl2
CoCl2
NH.OH
4
NaCl
dist. ELO
SO, Concentration
after 3 hr
329
199
167
49
49
4
3
, 3
yg/cm
53
-------
served to neutralize the sulfate formed in the reaction. Its
role in maintaining a high pH and thus a high sulfite concen-
tration is not mentioned, and this is probably a more important
function. A theoretical investigation of the process in fog
droplets estimated 2.9 yg/m of sulfate would be formed in an
3 3
atmosphere initially containing 20 yg/m of S02 and 3 yg/m of
NH, ("clean country air") with a liquid water content of 0.1
g/m . The same fog in air polluted with 500 yg/m of SC>2 and
10 yg/m of NH. initially should form 26.2 yg/m of S04~, or
nearly an order of magnitude more. No calculations were
attempted for falling drops because of uncertainties regarding
concentration behavior of the catalysts in such systems.
Experimental work on the formation of ammonium sulfate in water
\ '
droplets was done by van den Heuvel and Mason . Using water
droplets suspended on a fiber grid and exposed to an airstream
containing controlled amounts of S02 and NH,, they found that
the mass of salt {assumed to be ammonium sulfate) produced in
the droplet is proportional to the surface area of the drop and
to time. Because of certain problems with the technique, how-
ever, their results are only semi-quantitative. Furthermore,
application of their results to the atmosphere requires extra-
polation to much lower gas phase S0_ and NH- concentrations than
were used in the study. Allowing this extrapolation, their pre-
dicted rate dependence leads to a rather large conversion rate
of about 2.5%/min in an atmosphere containing 100 yg S00/m and
54
-------
10 yg NH,/m . Keeping in mind that the data are only semi-
quantitative, it is still apparent that the process is of impor-
tance in the oxidation of S02 in droplets as well as in bulk
solution.
(54 )
A much-quoted theoretical investigation by Scott and Hobbs
assumes equilibrium is maintained between gaseous and dissolved
S02, NH3, and CO^r and between the ions produced in the dissoci-
ation of the dissolved species:
SO.
2(g)
H"
HS0
NH3(g) + H20
H"
NH3'H2°
OH"
Khs [S°2'H2°]/PSO.
= [HS03~][H+]/[S02'H20]
= [S03=][H+]/[HS03"3
Kha -
[NH4+] [OH~]/[NH3'H20]
C02(g) + H2° * C02'H2°
H
HC0
H
+
H20 * H + OH
K
hc
[HC03"] [H]/[C02'H20]
K
2c = [C03 ][H+]/[HC03~]
K
w
[IT] [OH"]
Using the appropriate equilibrium expressions and constants, it
is possible to derive an expression for the sulfite concentra-
tion as a function only of [H ], which can in turn be determined
55
-------
by invoking a charge balance equation. The formation of sulfate
can then be calculated by a simple integration of the assumed
first-order oxidation rate equation
d[SO,=]
- k
-1 -1
The rate constant k was taken to be 0.1 min or 0.0017 sec
based on an analysis of van den Heuvel and Mason's data. The
calculations made under these assumptions did not show the sul-
fate formation to reach an asymptotic limit in one to two hours
as noted by Junge and Ryan, nor did it show the linear depend-
ence on the initial partial pressure of SO-. The work yields
oxidation rates of the order of 2.5 percent per hour and does
suggest sulfate levels of the same order of magnitude as some
measurements 7 , but the corresponding NH, concentrations and
pH's are much higher than measured, and in general rather long
times are required for large sulfate formation. Depletion of
the gas phase concentrations of S02 and NH3 was not considered.
Assuming the same set of reactions and equilibrium constants as
(58 )
Scott and Hobbs, McKay was able to predict much faster for-
mation of sulfate by calculating the first order rate constant
from the expression given by Fuller and Crist:
k = 0.013 + 59[H+]1/2 sec"1 .
At pH 7, this is 20 times as great as the value used by Scott
and Hobbs, and at pH 5, it is 120 times as great. Furthermore,
56
-------
the work indicated that the reaction should proceed faster as
the temperature is lowered, a feature which had been noted by
other investigators and may be attributed to increased gas
solubility at lower temperatures. Some of McKay's data suggest
rates of oxidation of about 13 percent per hour.
There is still no concensus about the "best" value for the ox-
idation rate for this system. Data taken by Miller and de
Pena'6°) fit a curve corresponding to k ~ 0.003 sec" , very
close to the value used by Scott and Hobbs. Some of the Miller and
de Pena curves, however, suggest oxidation of only 0.1 percent
per hour.
The relevance of the entire SO,.,-NH,-water system to the case of
power plant plumes is quite uncertain. Reported S0_ oxidation
rates attributable to ammonia catalysis vary widely, from 0.1 -
13 percent per hour. The NH. concentrations and pH calculated for
the system, regardless of k, are much higher than measurements
indicate. One possible explanation for this involves the sol-
ubility of NH_ in water: if it is actually a good deal less
than the models assume it is, both [NH*] and pH predictions
could be lowered. This would mean lower [SO,"] and thus, for
the Scott and Hobbs-type analysis, the rate would be lower. For
an analysis using McKay's formula for k, however, the lower pH
also means a larger value for k, and for reasonable ambient
levels of SO- and NH.., the rate will actually increase.
57
-------
The other point which must be kept in mind is the time required
for appreciable oxidation. A fog or mist could be in the vic-
inity of a plume for times on the order of hours. Precipitation
falling through a plume, however, does so in a matter of minutes
or less. Thus, in order to contribute significantly to the sul-
fate formation in the aqueous phase, a mechanism must have a
time constant of the same order as the time that the phase is in
contact with the plume.
METAL CATALYSIS
Concurrently with the above work on the S02~NH,-water system,
there have been several important studies on the effect of metal
catalysts in the oxidation of S02. Work by Johnstone and Cough-
anowr' °' indicated that, for high catalyst concentrations, the
oxidation could be assumed to occur within a spherical shell at
the surface of the drop. This shell becomes thicker as the cat-
alyst concentration decreases, or as the S02 concentration at
the surface increases, and for some critical value of either
parameter, of course, the entire drop is involved.
An excellent review and extension of the earlier theoretical
work was given by Foster ', whose interest was in the oxida-
tion of SO- in power plant plumes. After a lengthy analysis of
droplet (especially MnS04) growth, he considers the oxidation
*The time constant for a mechanism may be thought of as the
half-life of the rate-limiting reaction.
58
-------
rates by manganese and iron catalysts. In the case of iron
catalysis, some of the assumptions made are based on the work
f 79}
of Nytzell-de Wilde and Tavernerv . The derivation led to
these rate expressions:
Rate of S02 oxida- 22 4 K C2V 2
tion by Mn catalyst = "'* *iuiv x 1(T% per min.
10"6 GD
Rate of SO- oxida-
tion by Fe2catalyst =
where the symbols are defined in Table 13 , as are typical values
for each parameter. Using these values and assuming values for
+ i
[K ], K. and K., Foster estimates SO- rates of 0.09%/min for Mn
and 0. 15-1. 5%/min for Fe, suggesting that iron oxides are the
most important catalysts for aqueous phase S02 oxidation in
plumes.
Matteson et al. considered the kinetics of the oxidation
mechanism using a- manganese sulfate aerosol catalyst. A theore-
tical analysis of the mechanism (see Table 10) leads to the con-
clusion that the rate is proportional to the square of the initial
~f"4-
aqueous Mn concentration (Foster had assumed a similar depend-
ence in his calculation of the manganese-catalyzed rate). The
rate constant was found experimentally tobe2.4xlO M s
It was also found that almost no sulfate is formed when the re-
lative humidity is less than 95%, probably because the aerosol
59
-------
TABLE 13. NOMENCLATURE AND VALUESUSED FOR PLUME S02
OXIDATION CALCULATIONS * 6 >
Nomenclature Value
General W
G
D
f
0
S
Manganese M.
ni
f .
i
Iron M.
j_
n.
i
f.
Effluent dust burden, g/1 2 x 10"
Effluent SO
3
content, ppm 2-5 x 10
Effluent dilution factor 10"
Fraction of
total sulphur oxidized 10
Droplet suplhate concentrations, mol/1 A 1
rl- U
2
2
Oxide molecular weight, g/mol 2-29 x 10
Number of catalytic ions per molecule 3
Fraction by
-4
weight of dust soluble 2 x 10
2
as above 1«60 x 10
as above 2
-2
as above 10
60
-------
is insufficiently hydrated. The authors feel that a similar
mechanism may be involved v/here other catalysts are concerned,
although no others were used in either the experimental work or
theoretical development.
Further work on MnS04 and on MnCl2, CuS04, and NaCl aerosols
was reported by Cheng et al. (59'. in terms of catalytic effici-
ency, the following order was established: MnSO. > MnCl2 > CuSO.
> NaCl. In an attempt to extrapolate the experimental results to
atmospheric conditions, they assumed a fog of 15 um droplets con-
taining 0.2 g H20/m in which half of the droplets contain cata-
lyst at concentrations capable of oxidizing S02 at the same rate
as that found with 500 yg MnSO3/m . Under these assumptions,
the extrapolation of their experimental "results indicated that
S0_ at the 0.1 ppm level should be oxidized at a rate of about
20%/hr. The level of catalyst here was taken to be typical of
that in a power plant plume, and the resulting rate is about
four times higher than the .09%/min determined theoretically by
Poster.
A kinetic study of the homogeneous oxidation of S02 by cobalt
ions was performed by Chen and Barron (62). Although the value
for the rate constant was not determined, the reaction was
found to be zero order in oxygen, three halves order with re-
spect to sulfite ion concentration, and one half order with
respect to cobalt catalyst concentration. The mechanism proposed
61
-------
to account for these features is a free radical chain adapted
11 " ("7n }
from Backstromv'u':
S03~ + Co(H20)g+ -*1 Co(H20)|+ +'S03"
k2
so3 + o2 -*>.So5
k,
so~ + so," -» soc + -so ~
3 J 5 3
- k4
S05 + S03~ -»* 2S04~
* S03~ + ' S03~ ~" inert products
S03~ + S05~ * inert products
- k5
S05 + ' S05 -* inert products.
It has been mentioned before that many of the studies on S02
oxidation have been done using rather high S02 concentrations,
requiring a large extrapolation in order to apply the results
to the atmosphere. Brimblecoinbe and Spedding ' attempted to
correct this situation by measuring the oxidation of S02 at
concentrations of about 10~ M in aqueous solutions with Fe(III)
at about 10 M acting as the catalyst, using a radiochemical
method of analysis. The sensitivity to the iron catalyst, even
at these low S02 concentrations, is indicated by the fact that
even in the purest water obtainable, the oxidation proceeded at
a measureable rate, presumably due to trace iron at concentra-
8
tions on the order of 10 M. They suggest that a free radical
62
-------
mechanism similar to that above is responsible, with the initi-
ator being FeOOH:
S03= + FeOOH + 3H+ -» Fe 2+ + 2H20 + 'S03~
2+
and the Fe(III) may be regenerated by oxidation of Fe ,
2+ » 3+
S03 + Fe -* S03 + Fe
allowing oxidation of large amounts of S02 by very small amounts
of Fe(III). (The hydrozylated Fe(III) species FeOOH has a
higher oxidation-reduction potential than Fe (0.908 V com-
pared with 0.77 V)(75'80^/ and PH values in the range pH 4
to pH 5 tend to favor formation of the hydroxylated species.
Thus, FeOOH appears to be a better oxidizing agent than Fe
for the production of *S03 radicals).
In another study of the catalytic effect of iron, Freiberg
obtains a rate expression:
d[S04=] _ KTK^[Fe3+][H2S03]2
dt [H+]3
where K is the first dissociation constant for H2S03 and IL, is
a rather complicated function of [0-]/ [Fe J, and several
equilibrium constants; the expression fits the published data of
63
-------
others ' . The strong dependence of the rate in droplets on
relative humidity is seen to be due to the increase in pH (by
dilution) as the humidity increases.
Other Mechanisms
There have been few mechanisms proposed for the oxidation of S02
in aqueous systems other than the catalytic ones mentioned above.
Work on a photochemical process in the aqueous phase was reported
(82)
by Jones and Adelman , but as is typical of such mechanisms,
the presence of reactive hydrocarbons is necessary. It is prob-
ably safe to say that photochemical processes are not important
in the aqueous phase oxidation of SO- in plumes under ordinary
circumstances due to the low concentrations of such hydrocarbons.
The possibility of entrainment of such species must be considered,
but kinetic studies of the aqueous phase photooxidation of S02
are virtually nonexistent. Of possibly greater importance are
reactions with ozone (HSO ~ + 03 -> HSO. + 0.,) . Espenson and
Taube performed tracer experiments which, although they did
not yield rate data, indicated that the reaction is not as simple
as the stoichiometric relation above suggests, since oxygen atoms
are exchanged with the solvent as well as being transferred between
sulfite ions and ozone. Penkett , using ozone in the range of
3 to 5 x 10 M, found the oxidation of bisulfite to be first
order with respect to ozone. The overall reaction was found to
fit a rate expression of the type - d(0_)/dt = k2(03)(HSO ~)
64
-------
and k2 was found to be 3.32 + 0.11 x 10 Ms. This leads to
an oxidation rate of about 0.21%/min in a cloud under typical
conditions (10 t, pH 5, about .1 to 1 g/m liquid water content)
assuming 7 ppb SO,,. The ozone level assumed, 50 ppb, represents
yet another extrapolation of the laboratory data. This rate is
some 70 times faster than that predicted by the analysis of
Scott and Hobbs, although it is of roughly the same order as
indicated by McKay's calculations, depending on the pH. Sub-
sequent work by Penkett and Garland ^42^ measured the rate of
oxidation of S02 by ozone in fogs formed in a chamber, and the
measured rates agree with those calculated in the earlier work.
It may be noted that in the above review, little attempt has
been made to reconcile the findings of various authors. This
stems partially from a feeling of frustration because there is
such a wide spectrum of results reported, and also because so
little of the work which has been done so far has been directly
concerned with the S02 in plumes. Even those studies which
attempt to relate their findings to the case of polluted atmos-
pheres are forced to make rather gross extrapolations from the
data collected in the laboratory, and the diversity of the re-
sults is hardly surprising. On the other hand, if there is one
fact which the literature does show well, it is that the oxida-
tion of SO- in the aqueous phase is highly sensitive to several
parameters, among which are pH, relative humidity, temperature,
65
-------
and catalyst concentration, which leads to the expectation that
the rate really is highly variable. Thus, the conclusions which
do emerge from the works reported are mainly qualitative:
(1) Under favorable meteorological conditions, aqueous
phase oxidation of S02 must be considered, and in some
cases it may be the predominant mechanism.
(2) Power plant plumes are significant sources of the met-
als (especially manganese and iron) which are able to
catalyze the oxidation.
(3) The influence of ammonia on the reaction is of parti-
cular importance, both in the formation of ammonium
sulfate and in its ability to increase the reaction
rate by maintaining a high pH.
*
(4) Ozone is a potentially important aqueous phase oxidizing
agent for SO-, although its importance in the early plume
may be doubtful because of depletion by reaction with NO.
(5) Aerosols of MnSO, and other species may be important
as sources of catalysts, as well as reaction sites
when the aerosols are hydrated.
(6) Because of the highly sensitive nature of the reaction
(or reactions) to existing ambient conditions, typical
rates probably range from about 0.1 to 2.0%/min, al-
though even higher rates may be possible.
66
-------
SECTION VII
HETEROGENEOUS STUDIES
Considerable attention has been given to the use of catalysts,
especially platinum catalysts, for converting S02 to SO., for
sulfuric acid production. Most recently, platinum and some of
the rare earth oxides have been considered further in their
roles as automotive oxidative catalysts. In this instance, as
in the oxidation that occurs in plumes, the motivation was not
a search for improved catalyst efficiency, but rather to under-
stand the role of the catalyst. A review of the extensive lit-
erature concerning the heterogeneous catalysis of S02 oxidation
- f n "5 \
for sulfuric acid production^ J; would not be generally relevent
to the scope of this report since plume SO- concentrations are
orders of magnitude below the levels used in acid processing;
the kinetic concepts involved, however, should be comparable.
Figure 6 taken from Boreskov presents a brief overview of
the effectiveness of a number of catalysts for oxidizing S02 to
SO, in sulfuric acid production. A most pertinent aspect of
Boreskov's efficiency curves is the sharp drop in the conversion
efficiency of the metal oxides at temperatures below 500°C. The
catalysts most pertinent to plume oxidation shown here might be
V0 and F&° ^at much lower temperatures, of course) but as
67
-------
1001
SQO 400 5OO 600
Temperature, C
700
FIGURE 6. CONVERSION OF S02 TO SO, IK THE PRESENCE
OF SEVERAL CATALYSTS"
68
-------
will be brought out later, this drop-off in catalytic conversion
efficiency may raise questions about the significance of hetero-
geneous gas-phase reactions in plumes.
It has generally been observed in fuel-burning processes that 1-
2 percent of the sulfur in the fuel leaves the stacks as SO,,
and both heterogeneous and homogeneous reactions have been sug-
gested to account for this result. Expressions developed by
/OC^ f R £ ^
Calderbank and by Kodles^ ', respectively, indicate the
rate of S02 oxidation to be negligible (< 3 ppm SO, per hour) at
300°K with 100 ppm S02:
P P
moles SO- converted ,, nnn 0~ S0n
r, ± = exp ( J1'OUU + 12.07) ±-
**F \ J-«- . w i i . i
g catalyst-sec RT P_ f
oO A
r = 1.76 x 106 exp (-
Here the P's are in atmospheres, the N's are mole fractions, T is
the absolute temperature (K) and R is 'the gas constant. This low
rate tends to support earlier work ' 8 ' which shows that the conversion
of S02 to SO, can occur homogeneously in the flame front (the reactive
region of the flame) by stationary state processes:
S02 + 0 + M * S03 + M
so3 + o * so2 + o,
A recent study by Novakov, Chang, and Harker^87) presents evidence
for the possible role of carbon (soot) particles as catalysts
for the oxidation of S02 in the atmosphere. This evidence is
-------
derived from electron spectroscopy for chemical analysis (ESCA)
studies of particulate matter from a premixed hydrocarbon flame
with different SO- levels. The ESCA spectrum of graphite parti-
cles exposed to SO- revealed peaks corresponding to sulfate and
sulfide. Experiments varying the level of SO,, added to the
flame as well as the humidity of the air used also showed an
increase in sulfate with increases in each of these parameters.
Although the results of this study show a consistency for carbon
acting as a catalyst, the evidence presented is somewhat limited.
Also, there are no rate data presented in this study from which
one might evaluate the significance of soot in plumes as a
sulfate catalyst.
(88}
Corn and Cheng have also carried out studies on the adsorp-
tion of SO - by activated charcoal, as well as by Fe^O.,, MnO_
and suspended particulate matter. In these experiments, packed
beds of Teflon beads were coated with micron and submicron size
aerosol particles, the beads were packed in a reactor, and pro-
gress of the reaction was followed by continuously monitoring
the effluent S0_. Activated charcoal yielded steady-state rates
of conversion, or adsorption, of 0.013 and 0.021 yg SO^/min/mg
charcoal at SO., concentrations of 8.0 and 14.4 ppm, respective-
£
ly. The experiments do not distinguish, however, whether the
SO- underwent steady-state conversion in the reactor or whether
there was catalyzed oxidation on the surface.
70
-------
In the Fe?0- experiments there was rapid interaction with S0_,
even at zero relative humidity. In the cases of Fe-O.,, MnO- and
suspended particulate matter there was evidence for significant
physical adsorption of SO-. In all cases, humidity increased the
adsorptivity of S02.
Several species were inert to S0_, i.e., did not adsorb SO-;
these were CaCO,, V,,05 and flyash. The V-O- results appear un-
usual since V_05 is a recognized oxidation catalyst for SO-. At
500°C and above, it is presumed to oxidize via the reaction:
so2 + v2o5 * so3 + v2o4.
At room temperature, however, the sorption of SO- was only 1.2%
at 95 percent relative humidity.
A number of other studies have also been conducted on the inter-
action of SO- with the iron oxides Fe-O-, and Fe,0,. In general,
these studies have limited applicability to resolving the prob-
lem of reactions in plumes in that rate information is limited.
Chun and Quon 89 carried out a study designed to measure the
capacity of ferric oxide particles to oxidize SO- in air. The
procedure used involved laying down a film of Fe-0- on a filter
(by combustion of iron carbonyl vapor) and passing a stream of
S0_ in air through the filter-reactor. The authors describe the
reaction as a "capacity-limited heterogeneous reaction", because
the reaction is not catalytic in the true sense of the term.
71
-------
-------
Active sites on the surface of the particles become occupied by
the products of the reaction, and thus are not available for
further reaction. The capacity of the Fe20, particles to oxid-
ize SC>2 in air was found to be 62.6 yg/mg Fe^O-. The rate con-
-3 -1 -1
stant for the heterogeneous reaction was 9.4 x 10 ppm min
18 311
(6.3 x 10 cm mole sec ). SO- concentrations in this study
were varied from 4.7-18.8 ppm and relative humidity was varied
from 50-94 percent. The rate constant did not appear to be cor-
related with S02 or relative humidity.
Urone et alS ' examined the reactions of S0_ in the presence of
hydrocarbons, nitrogen dioxide, moisture, particulate matter and
ultraviolet radiation. Gaseous mixtures did not react in the
dark and S02 oxidation was of the order of 0.1%/hr under ultra-
violet irradiation. In the presence of particulate matter, the
reaction was significant in some instances.
Table 14 presents a summary compilation of SO- conversion rates
drawn from the Urone study. SO- concentrations in this study
were varied from 8-14 ppm; particulate loadings were 16 -
30 mg. The results show high reactivity with the iron, lead and
calcium oxides and low reactivity with NaCl, CaCO,, Al_0, and
V^Oj.. The results compare well qualitatively with those of
(8 8}
Corn and Chengv, who reported high reactivity for Fe20, and
low reactivity for CaCO., and V20,-.
72
-------
TABLE 14. SUMMARY OF S02 REACTIONS ON
VARIOUS PARTICULATE SPECIES(90)
Particulate Species % S02 Reacted/min
CaC03 .004-.07
Cr203 .009
V205 0-.015
NaCl 0-.02
A1203 .04
CaO 1.8-3.0
Al203/CaO 2.4-2.6
PbO 1.9
Pb02 5.9
Fe.,0. 4.2
0. 4.5
73
-------
Smith, Wagman and Fishv ' report sorption studies of S02 on
Fe3°4' ^2^3 anc^ Platinum, using particles ranging in size from
about 0.01 to 0.1 micron diameter. Using a >0S00 technique, they
4.
developed sorption isotherms. Their technique may prove useful in
distinguishing layers of preferential chemical adsorption from
multilayers of physical adsorption. They observe significant
adsorption on all three surfaces, but do not measure oxidation
rates in these studies.
Some basic studies on the adsorption and conversion of S0_ on
CaO and MgO have been carried out as part of efforts to remove
SO- from power plant stack gases by adding limestone or dolomite.
/ q 2 Q "3 \
Two studies from investigators at New York Universityv^ 'yo' have
been concerned with infrared examination of S02 adsorbed on CaO
and on MgO. Sulfites are produced on the surface and subsequent-
ly oxidize to sulfates on heating in air.
Lunsford has also studied the SO~-MgO system, both at high temp-
eratures and at ambient temperatures'^4 . The low temperature study,
although it provides no rate data, has more potential relevance
to plume chemistry. Samples of MgO and Mg(OH)2 were exposed to
S02 at 25 torr, and electron paramagnetic resonance (EPR) and
infrared (IR) spectra were obtained. The results showed that
heterogeneous oxidation occurred on the MgO particles. Sulfite
ions and monodentate sulfite complexes were identified. The
investigators suggest that water adsorbed with the S02 increases
74
-------
the concentration of these ions on the surface, thus promoting
the conversion rate.
A study of the macroscopic properties of S02 adsorbed on mangan-
(95)
ese oxides ' is presented here mainly because of its relevance
to the catalytic role of Mn in aqueous systems discussed in a
previous section. This study looked at the adsorptive capacity
of wet and dry Mn02 and Mn-O,. Essentially, the study shows the
increased capacity for adsorption with increasing surface area
and decreasing specific gravity, both of which result from the
absorption of water.
/ n c \
A study by Happel and Hnatow v ' is of interest here in relation
to some of the preceding studies. A theoretical analysis of the
18 35
use of the isotopic tracers CU and SO- in the catalytic oxi-
dation of SO- over V-0^ catalyst suggests that in conversion
reactions in commercial acid plants and stack gas converters,
V-Oc is considerably more effective than in plumes. The study
concluded that oxygen chemisorption is the limiting mechanistic
step, with SO., desorption being of next importance in the rate
control. A rate equation is derived for conditions near equilib-
rium. It appears, however, that the rate equation is not applic-
able to ambient, possibly non-equilibrium conditions in a plume.
The analysis is useful, however, as a guide to investigators
applying isotopic tracers (e.g., Urone ^ °' and Smith ')
75
-------
In summary, the heterogeneous solid catalyzed gas phase oxida-
tion studies carried out to date are of very limited usefulness
to plume chemistry calculations. First, the literature is essen-
tially void of reaction rate and/or mechanism studies relating
(dry) heterogeneously catalyzed S02 oxidation systems to atmos-
pheric processes. As pointed out in the previous section on
aqueous phase oxidation systems, the most effective catalysis
( QQ\
occurs in solution. The studies of Com and Chengv ' and
Urone^ ' raise an interesting point on the lack of reactivity
of dry particulate vanadium. Because of the recognized activity
-(97 9 8}
of V205 as an acid catalyst, ' it is often stated that high sul-
fate levels in some areas may be attributable to the use of
vanadium-containing fuels. It would appear from these ambient
temperature investigations, however, that if SO, is produced in
the combustion process, or at least near the stack outlet, the
vanadium-containing particulate matter is not responsible for
excessive sulfate formation in plumes or in the atmosphere in
general.
76
-------
SECTION VIII
MODELLING APPLICATIONS
As suggested in the previous sections of this report, the measure-
ment of S02 removal and/or sulfate formation rates, whether in the
field or in the laboratory, presents a variety of problems. While
the details of using these data in a model to predict sulfate for-
mation are likewise formidable, the technique itself is at least
conceptually uncomplicated. The third report to be prepared for
this project will be the more detailed documentation of an actual
model, STRAC (Source-Transport-Receptor Analysis Code), developed
for this purpose.
The technique, in general, must create the model-world analog of
a real-world situation through a suitable mathematical formalism.
The practical limitations dictated by computer capacity and cost,
as well as the more philosophical ones imposed by an imperfect
knowledge of nature, of course, determine how good the analogy is.
A model of a reactive power plant plume must include a means for
calculating the spatial distributions of pollutants, and it is
this portion of the modelling which involves chemical reaction
data. Specifically, the model is defined by the general continu-
ity equation for each of n chemical species (S£>2r SO.,, NO, 0,...):
dc. ^
77
-------
where c. is the concentration and r. the rate of production of
species i in a differential volume element of atmosphere. The
rates r., of course, will in general be functions of temperature,
relative humidity, insolation, and concentrations of all species.
The solutions which yield values for the c-"s are generally found
by finite difference methods, necessitating rather careful con-
sideration of the technique to be applied, a subject beyond the
scope of this report. The inclusion of chemical reaction mechan-
isms, however, is rather simple since chemical rate equations are
naturally expressed as differential equations, the form required
by the continuity equation.
The program which performs such model calculations can be made
extremely versitile by separating the various functions into dif-
ferent subprograms. For example, in a simple case, the main
program may handle the input/output functions and define such
parameters as the initial calculation point and concentrations,
and step size. A replacable segment may then be called to per-
form the finite difference integration, where the necessary
derivatives are calculated in yet another subprogram. This allows
the examination of different models by the simple replacement of
one subprogram, rather than rewriting the entire code. In such a
configuration, of course, one could also replace the finite dif-
ference method without disturbing the rest of the program, if
such a replacement were desirable.
78
-------
The actual inclusion of chemical rate data into a plume model,
then, is not a great problem. The major obstacle to the use of
such models is the difficulty of finding reliable data to use
for input, since the result of any calculation can be no better
than the assumptions and data upon which it was based. As the
preceding sections point out, data available concerning SG>2
oxidation rates are not extensive, especially for heterogeneous
gas phase processes, a fact which severly limits the immediate
use of models as predictive tools. There are perhaps more impor-
tant uses for them as diagnostic tools in the intercomparison of
various mechanisms. It can be hoped that future studies will
improve our understanding of the important S02 oxidation process
and allow more informative diagnostic plume modelling than pre-
sently possible.
79
-------
SECTION IX
REFERENCES
1. Hales, J. M., Tall Stacks and the Atmospheric Environment.
Battelle-Northwest Laboratories Report to Environmental
Protection Agency, Research Triangle Park, N.C. (January,
1976) .
2. Urone, P., and W. M. Schroeder, SC>2 in the Atmosphere:
A Wealth of Monitoring Data, But Few Reaction Rate Studies.
Env. Sci. Tech. 3: 436-445 (1969).
3. Kellogg, W. W., R. D. Cadle, E. R. Allen, A. L. Lazrus,
and G. A. Martell, The Sulfur Cycle. Science. 175:
587-596 (1972).
4. Cadle, R. D., Formation and Chemical Reaction of Atmospheric
Particles. J. Colloid and Interface Science. 39: 25-41
(1972) .
5. Bufalini, M., Oxidation of Sulfur Dioxide in Polluted
AtmospheresA Review. Env. Sci. Tech. 5: 685-700
(1971).
6. Altshuller, A. P., and J. J. Bufalini, Photochemical
Aspects of Air Pollution: A Review. Photochem. Photobiol.
4: 97-146 (1965).
7. Harrison, H., T. V. Larson, and P. V. Hobbs, Oxidation of
Sulfur Dioxide in the Atmosphere: A Review. Proc. Int'1
Conf. on Environmental Sensing and Assessment, Las Vegas
(September, 1975).
8. Pierrard, J. M., Environmental AppraisalParticulate
Matter, Oxides of Sulfur, and Sulfuric Acid. APCA Journal.
19: 632-637 (1969).
9. JANAF Interim Thermochemical Tables, Thermal Laboratory,
The Dow Chemical Company, Midland, Michigan (1960).
10. Leppard, W. M., Sulfate Control Technology Assessment,
Phase 1: Literature Search and Analysis. EPA/460/3-75-
002-a; PB-240 995/IWP (November, 1974) 50 p.
80
-------
11. Verhoff, F. H., and J. T. Banchero, Predicting Dew Points
of Flue Gases. Chem. Eng. Prog. 70: 71-72 (1974).
12. Gartrell, F. E., F. W. Thomas, and S. B. Carpenter, Atmos-
pheric Oxidation of S02 in Coal-Burning Power Plant Plumes.
Am. Indust. Hygiene J. 24: 113-120 (1963).
13. Dennis, R., C. E. Billings, F. A. Record, P. Warneck, and
M. L. Arin, Measurements of Sulfur Dioxide Losses from
Stack Plumes. APCA 69th Annual Meeting, New York City
(June, 1969).
14. Coutant, R. W., E. L. Merryman, R. E. Barrett, R. D.
Giammar, and A. Levy, A Study of the Fate of S02 in Flue
Gas. Battelle's Columbus Laboratories Final Report to
EPA/API/BCR, Inc./EEI (March, 1972) 12 p.
15. Stephens, N. T., and R. 0. McCaldin, Attenuation of Power
Station Plumes as Determined by Instrumented Aircraft.
Env. Sci. Tech. 5: 615-621 (1971).
16. Weber, E., Contribution to the Residence Time of Sulfur
Dioxide in a Polluted Atmosphere. J. Geophys. Res.
75: 2909-2915 (1970).
17. Newman, L., J. Forrest, and B. Manowitz, The Application of
an Isotopic Ratio Technique to a Study of the Atmospheric
Oxidation of Sulfur Dioxide in the Plume from an Oil-Fired
Power Plant. Atmospheric Environment. 9: 959-968 (1975).
18. Newman, L., J. Forrest, and B. Manowitz, The Application of
an Isotopic Ratio Technique to a Study of the Atmospheric
Oxidation of Sulfur Dioxide in the Plume from a Coal-Fired
Power Plant. Atmospheric Environment. 9: 969-977 (1975).
19. Newman, L., M. Smith, J. Forrest, W. Tucker, and B. Manowitz,
A Tracer Method Using Stable Isotopes of Sulfur. Proc. Amer.
Nucl. Soc. Topical Mtg. on Nucl. Methods in Env. Res.,
Columbia, MO.
20. University of Utah Research Institute, Rate of Conversion
of Sulfur Dioxide in a Power Plant Plume to Particulate
Sulfate. Final Report, Phase I, APS Contract No. 75-00634,
funded by WEST Associated (May, 1975) .
21. Davis, D. D., and G. Klauber, Atmospheric Gas Phase Oxida-
tion Mechanisms for the Molecule S02. Int. J. Chem. Kin.
Symp. No. 1 (1975), Proceedings of the Symp. on Chemical
Kinetics Data for the Upper and Lower Atmosphere, 13 p.
81
-------
22. Friend, J. P., Attenuation of Power Station Plumes as
Determined by Instrumented Aircraft. Env. Sci. Tech.
6: 172-173 (1972).
23. Whitby, K. T., B. C. Cantrell, R. B. Husar, N. V. Gillani,
J. A. Anderson, D. L. Blumenthal, and W. E. Wilson, Jr.,
Aerosol Formation in a Coal-Fired Power Plant Plume.
Presented before the Division of Environmental Chemistry,
American Chemical Society, New York (April 1976) .
24. Roberts, P. T., and S. K. Friedlander, Conversion of S02
to Sulfur Particulate in the Los Angeles Atmosphere.
Env. Health Perspectives. 10: 103-108 (1975).
25. Calvert, J. G., Interactions of Air PollutantsX. The
Effects of SC>2 Addition of the Reactions in the Simulated
Polluted Atmosphere. National Academy of Sciences, Con-
ference on Health Effects of Air Pollutants, Washington,
D. C. (October 3-5, 1973).
26. Castleman, A. W., R. E. Davis, H. R. Munkelwitz, I. N.
Tank, and W. P. Wood, Kinetics of Association Reactions
Pertaining to I^SC^ Aerosol Formation. Int. J. Chem. Kin.
Symp. No. 1 (1975), Proceedings of the Symp. on Chemical
Kinetics Data for the Upper and Lower Atmosphere, 629-640.
27. Gerhard, E. R., and H. F. Johnstone, Photochemical Oxida-
tion of Sulfur Dioxide in Air. Ind. & Eng. Chem. 47:
972-976 (1955).
28. Sidebottom, H. W., et al, Photooxidation of Sulfur Dioxide.
Env. Sci. Tech. 6: 72-79 (1972).
29. Leifer, R., J. P. Friend, and M. Trichon, The Mechanism of
Formation of Stratospheric Aerosols. J. Atmos. Sci.
30: (1973).
30. Cox, R. A., Particle Formation from Homogeneous Reactions
of Sulfur Dioxide and Nitrogen Dioxide. Tellus. 26:
235-240 (1974).
31. Jaffee, S., and F. S. Klein, Photolysis of NO2 in the
Presence of SO2 at 3660A. Trans. Far. Soc. 62: 2150-
2157 (1966).
32. Cox, R. A., and S. A. Penkett, Aerosol Formation from
Sulfur Dioxide in the Presence of Ozone and Olefinic
Hydrocarbons. J. Chem. Soc. Far. Trans. I, Part 9.
68: 1735-1753 (1972).
82
-------
33. Smith, J. P., and P. Urone, Static Studies of Sulfur
Dioxide Reactions, Effects of N02, C-.H,, and H20. Env.
Sci. Tech. 8: 742-746 (1974). J b
34. Wilson, W. E., Jr., and A. Levy, A Study of Sulfur Dioxide
in Photochemical Smog, I. Effect of S02 and Water Vapor
Concentration in the 1-Butene/NO /SO,., System. APCA Journal.
20: 385-390 (1970). X ^
35. Wilson, W. E., A. Levy, and D. B. Wimmer, A Study of Sulfur
Dioxide in Photochemical Smog II. Effect of Sulfur Dioxide
on Oxidant Formation in Photochemical Smog. APCA Journal.
22: 27-32 (1972).
36. Miller, D. F., Exploratory Study of Factors Affecting
Aerosol Formation. EPA-650/3-75-002; Contract No. 68-
02-1217 (November, 1974).
37. Roberts, P. T., and S. K. Friedlander, Photochemical
Aerosol Formation: SC>2, 1-Heptene and NO in Ambient
Air. Env. Sci. Tech. (in press).
38. Cullis, C. F., R. M. Henson, and D. L. Trimm, The Kinetics
of the Homogeneous Gaseous Oxidation of Sulfur Dioxide.
Proc. Roy. Soc. A295: 72-83 (1966).
39. Vohra, K. G., P. V. N. Nair, and T. S. Muraleedharan,
Possible Role of Singlet Oxygen in an Ion-Induced Reaction
Mechanism of Nucleus Formation by Sulfur Dioxide. Aerosol
Science. 3: 225-236 (1972).
40. Mulcahy, M. F. R., J. R. Steven, and J. C. Ward, The Kin-
etics of Reaction Between Oxygen Atoms and Sulfur Dioxide:
An Investigation by Electron Spin Resonance Spectrometry.
J. Phys. Chem. 71: 2124 (1967).
41. Payne, W. A., L. S. Stiet, and D. D. Davis, A Kinetics
Study of the Reactions of H02 with SO- and NO. J. Amer.
Chem. Soc. 95: 7614 (1973).
42. Penkett, S. A., and J. A. Garland, Oxidation of Sulfur
Dioxide in Artificial Fogs by Ozone. Tellus. 26: 284-289
(1974).
43. Armitage, J. W., and C. F. Cullis, Studies of the Reaction
Between Nitrogen Dioxide and Sulfur Dioxide. Combustion
and Flame. 16: 125-130 (1971).
44. Fair, R. W., and B. A. Thrush, Reactions Between Hydrogen
Atoms and Sulfur Dioxide. Trans. Far. Soc. 65: 1550
(1969).
83
-------
45. Falk, M., and P. A. Giguere, On the Nature of Sulfurous
Acid. Can. J. Chem. 36: 1121-1125 (1958).
46. Wang, J. C. , and D. M. Hinunelblau, A Kinetic Study of
Sulfur Dioxide in Aqueous Solution with Radioactive
Tracers. Am. Inst. Chem. Eng. J. 10: 574-580 (1964).
47. Eigen, M., K. Kustin, and G. Maass, The Speed of Hydration
of SOo in Water Solutions. Z. Physik. Chem. Neue Folge.
30: 130 (1961) .
48. Yue, G. K., C. S. Kiang, V. A. Mohnen, and E. Danielsen,-
The Interaction of Atmospheric Sulfur Compounds with Cloud
and Precipitation Elements. State University of New York
(Albany) Interim Report to National Science Foundation,
143 p. (May, 1975).
49. Fuller, E. C., and R. H. Crist, The Rate of Oxidation of
Sulfite Ions by Oxygen. J. Am. Chem. Soc. 63: 644 (1941).
50. Bassett, H., and W. G. Parker, Oxidation of Sulfurous
Acid, J. Chem. Soc. 1951:1540.
51. Junge, C. E., and T. G. Ryan, Study of Sulfur Dioxide
Oxidation in Solution and Its Role in Atmospheric Chem-
istry. Quart. J. Roy. Meteor. Soc. 84: 46 (1958).
52. van den Heuvel, A. P., and B. J. Mason, The Formation of
Ammonium Sulfate in Water Droplets Exposed to Gaseous
Sulfur Dioxide and Ammonia, Quart. J. Roy. Meteor. Soc.
89: 271 (1963).
53. Espenson, J. H., and H. Taube, Tracer Experiments with
Ozone as Oxidizing Agent in Aqueous Solution. Inorg.
Chem. 4: 704 (1965).
54. Scott, W. D., and P. V. Hobbs, The Formation of Sulfate
in Water Droplets. J. Atmos. Sci. 24: 54 (1967).
55. Beilke, S., and H. W. Georgii, Incorporation of Sulfur
Dioxide into Fog and Rain Droplets. Tellus. 20: 435-442
(1968).
56. Foster, P. M., Oxidation of Sulfur Dioxide in Power Station
Plumes. Atmos. Env. 3: 157 (1969).
57. Matteson, M. J., W. Stflber, and H. Luther, Kinetics of
the Oxidation of Sulfur Dioxide by Aerosols of Manganese
Sulfate. Ind. Eng. Chem. 4: 677 (1969).
84
-------
58. McKay, H.A.C., Atmospheric Oxidation of Sulfur Dioxide in
Water Droplets in Presence of Ammonia. Atmos. Env.
5: 7 (1971) .
59. Cheng, R. T., M. Corn, and J. 0. Frohliger, Contribution
to the Reaction Kinetics of Water-Soluble Aerosols and
Sulfur Dioxide in Air at ppm Concentrations. Atmos. Env.
5: 987 (1971).
60. Miller, J. M., and R. G. dePena, Contribution of Scavenged
Sulfur Dioxide to the Sulfate Content of Rain. J. Geophys.
Res. 77: 5905 (1972).
61. Penkett, S. A., Oxidation of Sulfur. Dioxide and Other
Atmospheric Gases by Ozone in Aqueous Solution. Nature
Phys. Sci. 240: 105 (1972).
62. Chen, T., and C. H. Barren, Homogeneous Kinetics of Sulfite
Oxidation. Ind. Eng. Chem. Fundam. 11: 466 (1972).
63. Brimblecombe, P., and D. J. Spedding, Catalytic Oxidation
of Micromolar Aqueous Sulfur Dioxide (I). Oxidation in
Dilute Solutions Containing Iron (III). Atmos. Env.
8: 937 (1974).
64, Freiberg, J., The Mechanism of Iron Catalyzed Oxidation of
S02 in Oxygenated Solutions. Atmos. Env. 9: 661 (1975).
65.. Bigelow, S. L. , Catalytic Action on the Speed of Oxidation
of Sodium Sulfite by Oxygen in the Air. Z.Physik. Chem.
26: 493 (1898).
66. Titoff, A., Contribution to the Understanding of Negative
Catalysis in Homogeneous Systems. Z. Physik. Chem.
45: 641 (1903).
67. Lumiere,A., L. Lumiere, and A. Seyewetz, Changes in
Anhydrous Crystalline Sodium Sulfite by Exposure to Air.
Moniteur Phot. 2: 77 (1908).
68. Milbauer, J., and J. Pazourek, The Oxidation of Sulfites
in Concentrated Solutions. Chem. Listy. 15: 34 (1921).
69. Reinders, W., and S. I. Vies, Reaction Velocity of Oxygen
with Solutions of Some Inorganic Salts (III) . The
Catalytic Oxidation of Sulfites. Rec. Trav. Chim.
44: 249 (1925).
70. Backstrom, H.L.J., The Chain Mechanism in the Autoxidation
of Sodium Sulfite Solutions. Z. Physik. Chem. B25: 122
(1934) .
85
-------
71. Volfkovich, S. I., and A. P. Belopolskii, Oxidation of
Sulfites. J. Applied Chem.(U.S.S.R.). 5: 509 (1932).
72. Haber, F. and 0. H. Wansbrough-Jones, Autoxidation (VI).
Action of Light on Sulfite Solutions in Absence and
Presence of Oxygen. Z . Physik. Chem. B18: 103 (1923).
73. Briner, E., and H. Bierdermann, The Role of Ozone as
Catalyst of Oxidation (V). The Ozonization of Sodium
SulfiteThe Influence of the Dilution of Ozone and the
Concentration of Hydrogen Ion. Helv. Chim. Acta.
16: 548 (1933).
74. Reinders, W., and P. Dingemans, Speed of Oxidation of
Metol by Air and the Effect of Sodium Sulfite. Rec. Trav.
Chim. 53: 239 (1934).
75. Sillen, L. G., Stability Constants of Metal-Ion Complexes,
Section I. Chem. Soc. Special Publ. No. 17 (1964).
76. Bassett, H., and A. J. Henry, Formation of Dithionate by
the Oxidation of Sulfurous Acid and Sulfites. J. Chem.
Soc. 1935: 914.
77. Dana, M. T., D. R. Drewes, D. W. Glover, and J. M. Hales,
Precipitation Scavenging of Fossil-Fuel Effluents: Power
Plant Measurements and Development of a Chemical Kinetics
Scavenging Model. Final Report to EPA, Research Triangle
Park, N.C. (1975).
78. Johnstone, H. F., and D. R. Coughanowr, Absorption of
Sulfur Dioxide from Air. Ind. Eng. Chem. 50: 1169 (1958)
79. Nytzell-de Wilde, F. G., and L. Taverner, Experiments
Relating to the Possible Production of an Oxidizing Acid
Leach Liquor by Auto-Oxidation for the Extraction of
Uranium. Proc. 2nd U.N. Int. Conf. Peaceful Uses at
Energy, 2nd, Geneva. 3: 303 (1958) .
/ _.
80. Bjerrum, J., Stability Constants of Metal Ion Complexes
(Supplement). Chem. Soc. Special Publ. No. 25 (1971).
81. Freiberg, J., Effects of Relative Humidity and Temperature
on Ion-Catalyzed Oxidation of S02 in Atmospheric Aerosols.
Env. Sci. Tech. 8: 731-734 (1974).
86
-------
82. Jones, P. W., and A. H. Adelman, Photosulfoxidation of
Hydrocarbons in the Liquid Phase. Env. Sci. Tech. 6: 933
(1972).
83. Duecker, W. W., and J. R. West, The Manufacture of Sulfuric
Acid. A.C.S. Monograph No. 144, Reinhold (1959).
84. Boreskov, G. K., Mechanism of the Oxidation of Sulfur
Dioxide Gas on Oxide Catalysts. J. Phys. Chem. (U.S.S.R.).
14: 1337-1346 (1940).
85. Calderbank, P. H. , Contact-Process Converter Design. Chem.
Eng. Prog. 49: 585-590 (1953).
86. Kodles, B., et al, Study of the Oxidation of S02 on a
Vanadium Catalyst in the Region of Internal Diffusion.
IV Int. Congress on Catalysis, Proc. Symposium III,
2477-2499 (1968).
87. Novakov, T., S. G. Chang, and A. B. Harker, Sulfates as
Pollution Particulates: Catalytic Formation on Carbon
(Soot) Particles. Science. 186: 259-261 (October 18,
1974) .
88. Corn, M., and R. T. Cheng, Interactions of Sulfur Dioxide
with Insoluble Suspended Particulate Matter. APCA Journal.
22: 871-876 (1972) .
89. . Chun, K. C., and J. E. Quon, Capacity of Ferric Oxide
Particles to Oxidize Sulfur Dioxide in Air. Env. Sci.
Tech. 7: 532-538 (1973).
90. Urone, P., et al, Static Studies of Sulfur Dioxide
Reactions in Air. Env. Sci. Tech. 2: 611-618 (1968).
91. Smith, B. M., J. Wagman, and B. R. Fish, Interaction of
Airborne Particles with Gases. Env. Sci. Tech. 3: 558
(1969) .
92. Low, M.J.D., A. J. Goodsel, and N. Takezawa, Reactions
of Gaseous Pollutants with Solids (I). Infrared Study
of the Sorption of S02 on CaO. Env. Sci. Tech. 5: 1191-
1195 (1971).
93. Goodsel, A. J., M.J.D. Low, and N. Takezawa, Reactions of
Gaseous Pollutants with Solids (II). Infrared Study of
Sorption of S02 on MgO. Env. Sci. Tech. 6: 268-273
(1972) .
87
-------
94.. Lin, M. J., and J. H. Lunsford, Photooxidation of Sulfur
Dioxide on the Surface of Magnesium Oxide. J. Phys. Chem.
79: 892 (1975).
95. Li, K., R. R. Rothfus, and A. H. Adeny, Effect of Macro-
scopic Properties of Manganese Oxides on Absorption of
Sulfur Dioxide. Env. Sci. Tech. 2: 619-621 (1968).
96. Happel, J., and M. A. Hnatow, Catalytic Oxidation of
Sulfur Dioxide Using IsotopiC Tracers. PB No. 224-305,
(EPA-650/2-73-020) 33 p. (August, 1973).
97. Opferkuch, R. E., et al, Applicability of Catalytic
Oxidation to the Development of New Processes for Removing
S02 from Flue Gases. Vol. III. (PB 198-810) U. S. Dept.
of Health, Edication, and Welfare, Cincinnati, Ohio
260 p. (August, 1970).
98. Merryman, E. L., and A. Levy, Sulfur Trioxide Flame
Chemistry. XIII Symposium (Int'l) on Combustion, Univ.
of Utah, The Combustion Institute, Pittsburgh, Pa.
p. 427 (August 28, 1970).
88
-------
SECTION X
APPENDIX
THERMODYNAMICS AND KINETICS: A REVIEW
Because of the volume of material covered by this report, little
attempt has been made to rigorously define the more common chem-
ical terms or to discuss kinetic and thermodynamic concepts. It
is hoped, however, that this material will be useful to the non-
chemist. This section is intended to review basic kinetics and
thermodynamics for these readers, and to simultaneously define
some of the more general terms and concepts used throughout the
report, for the sake of completeness. A more detailed discussion
of thermochemical kinetics than the necessarily brief overview
presented here may be found in any textbook on physical chemistry.
It is essential to remember that the primary factor- determining
whether molecular or atomic species will react is their energy;
species having lower energy are favored, and all reacting systems
will tend toward configurations having the lowest possible poten-
tial energy. This behavior can be illustrated in terms of the
bimolecular reaction,
A + B * C + D
whereby reactants A and B are converted to products C and D. Fig-
ure 7 shows how the potential energy of the system varies with a
parameter called the reaction coordinate, which indicates the
89
-------
Exothermic Reaction
Endothermic Reaction
t
A+B
bJ
Reaction Coordinate
C-hD
Reaction Coordinate
FIGURE 7. HYPOTHETICAL POTENTIAL ENERGY
REACTION COORDINATE DIAGRAMS
90
-------
degree of completion of the reaction. Moving from left to right,
the initially separated molecules A and B approach and collide,
increasing the potential energy of the system until a maximum is
reached, and then separate forming products C and D. If the
energy of the products is lower than the energy of the reactants,
the first figure is appropriate and the reaction is termed exo-
thermic; otherwise it is endothermic as shown in the second fig-
ure. The amount of energy gained (or lost) by the system during
the course of the reaction is called the enthalpy, denoted AH is
negative for exothermic reactions; reactions of this type corres-
pond to lower energy of reaction products and thus, are said to
be "favored" energetically.
Because of the presence of the maxima, or "energy barriers" in
the figures, all collisions between A and B do not result in
reactions. If a collision is not sufficiently energetic to
reach the top of the energy barrier, the molecules must separate
and remain as A and B; in fact, even with enough energy input,
there is still a chance that the necessary rearrangements will
not take place to allow the formation of products. The height of
the energy barrier is called the activation energy of the reaction,
labled Ef. The action of chemical species called catalysts can
thus be understood: without reacting themselves, they provide a
means whereby E^ can be lowered, allowing less energetic colli-
sions to pass over the energy barrier. Figure 7 also reveals
that with sufficient energy, C and D should be able to react to
91
-------
form A and B; that is, the reverse reaction;
C + D -* A + B
is possible. Furthermore, the enthalpy of this reaction is
simply - AH, and the activation energy E = E, - AH. (Here the
subscripts f and r denote reference to the forward and reverse
reactions, respectively.)
It should be noted that the discussion above is not limited to
bimolecular reactions. The concepts involved are rather easily
generalized to more complicated systems, and also apply to the
simpler unimolecular reactions. The latter class includes the
extremely important photochemical reactions, which are initiated
by electromagnetic radiation and are involved in many atmospheric
processes.
It must be remembered that the above discussion was a gross simp-
lification of the systems in which we are usually interested, sys-
tems which actually contain large numbers of atoms and molecules.
In particular, atmospheric proceses generally involve reactions of
pollutants at concentrations on the order of parts-per-billion or
more, meaning some 10 or more molecules/cm of reactive species
19 3
and more than 10 molecules/cm of inert species. Thus, the
reactant molecules have a distribution of kinetic energies which
are dependent on the temperature, and in such systems, it is pos-
sible to determine a rate at which each species is converted; the rate is
92
-------
dependent on the amount of material present. For example, in
the reaction A+B+C+D, A and B are converted, and
(d[A]/dt)f = (d[B]/dt)f = -kf[A][B]
where the brackets denote concentrations. Similarly, A and B
are both produced by the reverse reaction, (C + D -> A + B) , so
(d[A]/dt)r = (d[B]/dt)r = +k[C][D]
and the overall rate of change of A and B is
d[A]/dt = d[B]/dt = kr[C][D] - kf[A][B]
In the same way,
d[C]/dt = d[D]/dt = kf[A][B] - kr[C] [D] = -d[A]/dt
The rate constants kf and k can generally be expressed in the
Arrhenius form:
kf = M exp(- Ef/RT)
kr = N exp(- Er/RT)
where M and N are constants with only small temperature depend-
ences, Ef and E are the molar activation energies for the for-
ward and reverse reactions, respectively, and R and T are the gas
constant (1.98 cal/mole-°K) and the absolute temperature. The
exponential term reflects the kinetic energy distribution of the
reacting molecules, that is, the probability that collisions will
be sufficiently energetic to overcome the activation energy
93
-------
requirements. The units of kf and k , and thus M and N, are
1-r -1
(concentration) t . The reaction order r is the total number
(possibly fractional) of atoms or molecules; in the general case
described here, r=2 for both the forward and reverse reactions.
If, however, reactant A is present in such great quantities com-
pared to B that its concentration remains virtually constant, A
may be combined with k so that
(d[B]/dt)f = -kf[A][B] ~ -k
Here, although the reaction is actually second order, a pseudo-
first order rate expression with constant kf can be applied.
Such approximations are often valid in the atmosphere, especially
when a trace substance is oxidized by oxygen, a major constituent.
The half-life of a species is the time required for half of it to
react. In the special case of a first order (or pseudo-first
order) reaction, this time is simply (In 2)/k, where k is the rate
constant. This quantity is often used to compare speeds of com-
peting reactions.
If the reaction is allowed to proceed undisturbed for a long
enough time, a state of dynamic equilibrium will be established
when the rate of the forward reaction is exactly equal to the
rate of the reverse reaction. That is, although both reactions
are still proceeding, there are exactly as many molecules of
each species formed by one reaction as are destroyed by the
other. Thus,
94
-------
kf [A] [B] - kr[C] [D] = 0
In this case we can write an equilibrium expression for the
reaction,
A + B + C + D
and define an equilibrium constant;
«!£- EC]
[A] [B]
From the preceeding discussion, it is essential to realize that
chemical conversion in macroscopic systems can be limited by
either or both kinetic and thermodynamic factors. The kinetic
factors are essentially determined by the energy available to
the system: if only the most energetic collisions are able to
culminate in reaction because of a high activation energy, the
rate will be slow. Kinetic barriers can be overcome either by
providing more energy to the system (usually as heat) or by ap-
plying an appropriate catalyst to lower the activation energy
of the system. The thermodynamic factors are related to the
equilibrium state of the system, and are somewhat more difficult
to manipulate in situations where they are rate limiting. Be-
cause they only influence the activation energy, and not the
equilibrium composition, catalysts are unable to affect the
thermodynamic factors. The equilibrium constant is a function
of temperature and can be determined from the definitions given
previously:
95
-------
This indicates that K decreases with increasing temperature for
an exothermic reaction.
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-450/3-76-022
3. RECIPIENT'S ACCESSION'NO.
4. TITLE AND SUBTITLE
S02 Oxidation in Plumes: A Review and Assessment
of Relevant Mechanistic and Rate Studies
5. REPORT DATE Date of Approval
September 1976
6. PERFORMING ORGANIZATION CODE
7.AUTHORIS)
A. Levy (Battelle, Columbus Laboratories),
D. R. Drewes, and J. M. Hales
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORG \NIZATION NAME AND ADDRESS
Battelle, Pacific Northwest Laboratories
Battelle Boulevard
Richland, Washington 99352
10. PROGRAM ELEMENT NO.
2AC 129
11. CONTRACT/GRANT NO.
68-02-1982
12. SPONSORING AGENCY NAME AND ADDRESS
Office of Air Quality Planning and Standards
Environmental Protection Agency
Research Triangle Park, North Carolina 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
The scientific literature pertaining to the oxidation of SO2 in power plant plumes
is reviewed. Aqueous phase, homogeneous gas phase, and heterogeneous gas phase
mechanisms are considered, as are actual plume studies. The reported rates vary
over a wide range, which is not totally unexpected due to the highly complex
nature of the oxidation process, and some general conclusions can be drawn:
1) Recent plume studies, in general, indicate lower rates than earlier ones, and
also suggest a coincidence between the reappearance of ozone in the plume and
oxidation of S02-
2) Gas phase studies indicate homogeneous reaction of SO2 with OH radicals and
heterogeneous reactions catalyzed by lead and iron to be perhaps the most
significant processes.
3) In the aqueous phase, the reaction is most effectively catalyzed by iron
and manganese, and ammonia plays an important role in promoting the oxidation
by maintaining a high pH.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Electric Power Plants
Reaction Kinetics
Sulfur Dioxide
Mathematical Models
3. DISTRIBUTION STATEMENT
Release Unlimited
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES
100
20. SECURITY CLASS {Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
-------
INSTRUCTIONS
1. REPORT NUMBER
Insert the EPA report number as it appears on the cover of the publication.
2. LEAVE BLANK
3. RECIPIENTS ACCESSION NUMBER
Reserved for use by each report recipient.
4. TITLE AND SUBTITLE
Title should indicate clearly and briefly the subject coverage of the report, and be displayed prominently. Set subtitle, if used, in smaller
type or otherwise subordinate it to main title. When a report is prepared in more than one volume, repeat the primary title, add volume
number and include subtitle for the specific title.
5. REPORT DATE
Each report shall carry a date indicating at least month and year. Indicate the basis on which it was selected (e.g., date of issue, date of
approvcl, date of preparation, etc.).
6. PERFORMING ORGANIZATION CODE
Leave blank.
7. AUTHOR(S)
Give name(s) in conventional order (John R. Doe, J. Robert Doe, etc.). List author's affiliation if it differs from the performing organi-
zation.
8. PERFORMING ORGANIZATION REPORT NUMBER
Insert if performing organization wishes to assign this number.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Give name, street, city, state, and ZIP code. List no more than two levels of an organizational hirearchy.
10. PROGRAM ELEMENT NUMBER
Use the program element number under which the report was prepared. Subordinate numbers may be included in parentheses.
11. CONTRACT/GRANT NUMBER
Insert contract or grant number under which report was prepared.
12. SPONSORING AGENCY NAME AND ADDRESS
Include ZIP code.
13. TYPE OF REPORT AND PERIOD COVERED
Indicate interim final, etc., and if applicable, dates covered.
14. SPONSORING AGENCY CODE
Leave blank.
15. SUPPLEMENTARY NOTES
Enter information not included elsewhere but useful, such as: Prepared in cooperation with, Translation of, Presented at conference of,
To be published in, Supersedes, Supplements, etc.
16. ABSTRACT
Include a brief (200 words or less) factual summary of the most significant information contained in the report. If the report contains a
significant bibliography or literature survey, mention it here.
17. KEY WORDS AND DOCUMENT ANALYSIS
(a) DESCRIPTORS - Select from the Thesaurus of Engineering and Scientific Terms the proper authorized terms that identify the major
concept of the research and are sufficiently specific and precise to be used as index entries for cataloging.
(b) IDENTIFIERS AND OPEN-ENDED TERMS - Use identifiers for project names, code names, equipment designators, etc. Use open-
ended terms written in descriptor form for those subjects for which no descriptor exists.
(c) COSATI FIELD GROUP - Field and group assignments are to be taken from the 1965 COS ATI Subject Category List. Since the ma-
jority of documents are multidisciplinary in nature, the Primary Field/Group assignment(s) will be specific discipline, area of human
endeavor, or type of physical object. The application(s) will be cross-referenced with secondary Field/Group assignments that will follow
the primary posting(s).
18. DISTRIBUTION STATEMENT
Denote releasability to the public or limitation for reasons other than security for example "Release Unlimited." Cite any availability to
the public, with address and price.
19. &20. SECURITY CLASSIFICATION
DO NOT submit classified reports to the National Technical Information service.
21. NUMBER OF PAGES
Insert the total number of pages, including this one and unnumbered pages, but exclude distribution list, if any.
22. PRICE
Insert the pnce set by the National Technical Information Service or the Government Printing Office, if known.
EPA Form 2220-1 (9-73) (Reverse)
-------
33
m
2
m
D
C
>
r~
O
~D
T3
O
o
3-
=. O !
33
O
Tl
C
Z
H
-o
r~
O
03
C
en
Z
m
m
0
H
O
2
^
O
w
c/i
O
m «
2."
CD
Q) CD
Q. "
D. -
»i
v> °
in a.
CD
CD
Q.
CD
Q.
D
O
O
0. CD
O
O CD
£ <
CD CD
»
!Q r3
CD w
_. CD
3 ^
CT_ Q)
C ~~
a.
o
m
Z
m £ m
"° >
>52
coO°
CJ H Tl
en m m
O m
-I 01
m
Z
O
-------
m
D
C
n
3-
33
-. O!
H
NH
o
2
2
o
M
c/1
o
\
CO
9s
6
o
N O 33
TJlS
0 ° 5
O IT 3
d m ~~
CD ^ 3"
_(£ w
CD
W CD
Q. '-
Q. _
C
on Q.
=, °
% Z
CD ,-.
^5
^S
(/>
3-
D
Q. CD
- O
O CD
CD CD
O ->
0)
13
_ CD
^ 2
o J
c
Q.
3
(Q -
o
TD
T)
O
H
C
DO
C
t/1
Z
m
o. o o
>
Q.
3
*£?:
O 3 C?.
O
9^ ?i *
o_
5
o
2o
03 IJ
m
31
n
-<
m _,
>
i m m
O m
-I vi
i>
>D
cn
m
Z
o
------- |