SUMMARY REPORT
ON
SUSPENDED SULFATES AND SULFURIC
ACID AEROSOLS
U. S, ENVIRONMENTAL PROTECTION AGENCY
-------
UKAM
DO HOT QUOTE n* CITE
NOTICE
This 'document fa a preliminary draft. lฃ
has not been formally released by EPA
and should not at this stage be construed
to represent Agency policy. It is being
circulated for comment on its technical*
accuracy and policy implications.
-------
PREFACE '
This document was prepared by a Task Force convened under the
direction of Dr. John F. Finklea, Director, National Environmental
Research Center (NERC). The objective was to review and evaluate the
current knowledge of suspended sulfates and sulfuric-iacid aerosols
in the atmosphere, as related to possible deleterious effects upon
human health and welfare, with a view toward the need for a criteria
document in consonance with the provisions of the Clean Air Act as
amended.
The following members served on the NERC Task Force:
Gerald G. Akland Carl T, Ripberger
Harold A, Bond Gary T. Rochelle
David L, Coffin Eugene Sawicki
Thomas G, Dzubay James R. Smith
Jean G. French G, Wayne Sovocool
J. H, B. Garner Robert K. Stevens
Kenneth T, Knapp Jack Wagman
Robert A, McCormick William E. Wilson
John Moran Eva Wittgenstein
Lawrence Niemeyer
The substance of the document was reviewed by the National Air
Quality Criteria Advisory Committee (NAQCAC) in public sessions on
January, 1973, Members of the NAQCAC were:
Mary 0. Amdur - Howard University
David M. Anderson - Bethlehem Steel Corp,
Anna M, Baetjer - Johns Hopkins University
Samuel S. Epstein - Case Western Reserve University
Arie D, Haagen-Smit - California Institute of Technology
John V, Krutilla *. Resources for the Future, Inc.
Frank J. Massey, Jr. ~ University of California
James McCarrol * University of Washington
Elmer P. Robinson - Washington State University
Eugene P. Odum - University of Georgia
Morton Sterling - Wayne County Michigan Health Department
Arthur C, Stern ซ University of North Carolina
Raymond R. Suskind ~ University of Cincinnati
-------
DRAFT
DO NOT QUOTE OR CITE
Elmer P. Wheeler - Monsanto Company
John T, Wilson - Howard University
A final formal review of the report was conducted by a Task
Force convened under the direction of Dr. Ronald E. Engel of the
Office of Research and Development. Members of the Task Force were:
J. Wesley Clayton, Jr. A. F. Forziati
Robert B. Medz Kenneth L. Bridbord
Thomas L. Gleason Thomas D. Bath
Robert E. McGaughey James R. Smith
Arnold J. Goldberg Lawrence A. Plumlee
Jeannie L. Parrish
Comments and criticisms have been received from industrial and
public interest groups. These have been reviewed and amendments to
the document have been made where deemed appropriate.
-------
DO NOT QUOTE OS CITE
TABLE OF CONTENTS
Page #
PREFACE ,
SUMMARY- -- ...... - ---- - .......... ............. -_. ,-_a-f
I. ENVIRONMENTAL APPRAISAL ......... - - ........ - ................... 1
A. Introduction- T- - , ------ ... ---- - ---------------------------- 1
1. Photooxidation ---- -- ---- - --- __-__,._----..- --- -1
2. Chemical Oxidation ----------------------------------------- - ---- 7
a. 0, + S02--T ----------------------------- - ------------------ ----- 7
b. S02 + N03 (N205) ..... - ......................... ..--.* ........... 7
d. Oxidation Rates- ---- > ---- ------- --- _-, _- _,, ------ , -------- JQ
3. Aerosol Formation-. ------------ - --------------------------- ----- 10
4. Catalytic Reactions ---- - -------- -------------- -- ----------- 15
5. Formation of Other Sulfates --------- --------------------------- 20
6. Summary of S02 Oxidation Mechanism - ----------------------- , 20
1. Direct Photooxidation or the Clean Air Mechanism ---------- - ---- 20
2. Indirect Photooxidation or the Dirty Air Mechanism ----------- --21
3. Air Oxidation in the Condensed Phase or the Clean Water
Mechanism ------ ------- - --------------- ----- ------------- --21
4. Catalytic Oxidation in the Condensed Phase or the Dirty
Water Mechanism ----------------------------------- - -------- --22
B^ Atmospheric Observations-- ------------ ------- ----------- --. ----- 23
C. References-- ------- - ---- - -------- - _-__ ---- ._--- ------ -29
II . SULFATE EMISSIONS ................ - ..................... ---- --39
A. Stationary Sources -------- > ------------------------------------- 39
1. Emission Data --------- ------------ ------------ - ------- - ----- -39
2. Measurement Techniques ------ -. ------- - - < - - ----- --41
B. Mobile Sources ---------------------- - ---- ---------- ^. ---------- 44
1. Emissions ---------------------- ------------ < -------- ------ .--^-, 44
2. Measurement Techniques- ------------ ----------- .--- >-- ---- ----44
C. References ----- : ------------------------ - ---------- ------ --46
III, AMBIENT SULFATE
A. National Air Surveillance Network ? T -.-- --r.---w-^_----^-ซ47
1 . Spatial Patterns ---- ^ - ----------- ----, ------ _--,...- ^-^47
2. Seasonal Patterns-^ ----- "-> -------- ------ -------- ,-- - ------- , 49
3. Empirical Relationship to Sulfur Dioxide-- -- --- -- ซ ---^--50
4. Sample Collection and Analysis-. -------- ---- -^ ----- -------- ---50
B. Review of Measurement Techniques-. ---- --* ------ '- ----------.-55
1. Total Sulfate ---------------- ------- ------- -------- - ---- -*' 55
2. Automated Total Sulfur Monitor-^ -------------------- - --------- >- 55
3. Miscellaneous Sulfates -------------------------------- ---- -
-------
DRAFT
DO NOT QUOTE OR Cut Page
IV. EFFECTS ON HUMAN HEALTH----. ---- - 65
A. Toxicological Appraisal------- -- < 65
1. Current Knowledge----------------- ,----,,.. , 55
2. Deficiencies in Present Information 65
B. Epidemiological Appraisal 70
1. New York Asthma 70
2. Utah Asthma -- 74
3. New York Cardio Pulmonary 74
4. Cincinnati School Children 78
5. Implications and Limitations of Findings' 78
C. References --83
V. OTHER (WELFARE) EFFECTS - 84
A. Ecological Effects 84
1. Vegetation 84
^ * oO J. A ^ซ ^ปซซ~wซwป^wปซป^*ปป^ซ ปซปปซซซปปป Mป^ป^ jj^
B. Sulfate Effects on Materials < 89
C. Weather, Visibility arid Climate 90
1. General Summary --- ---90
2. Effects of Sulfate on the Composition of Rain and Snow --93
3. Effects of S02 and Its Byproducts on Light Transmission
and Visibility in the Atmosphere ซ----, ,, 95
a. Introduction-' , ., --95
b. The Effect of Aerosols on Light Transmission Through Air 96
4. Visibility and Its Relationship to Aerosol Mass
Concentration--- :-, ,-- 98
a. Quantitative Aspects 99
b. Supporting Data- 100
D. References 103
VI. TECHNOLOGY FOR STATIONARY SOURCES -108
A. Sources --~ 108
B. Control Strategy -108
C. Status of Control Technology 110
D. References -- 112
VII. CONCLUSIONS AND RECOMMENDATIONS - 113
A. Conclusions -- ---113
B. Recommendations -114
-------
DRAFT
DO NOT QUOTE OR CITE
SUMMARY
Progress is being made in our efforts to control S02 emissions into
the atmosphere. However, there is now evidence suggesting that sulfuric
acid aerosols and some sulfates may be more potent irritants than
S02. Should this prove to be true, our control strategy should be
examined with a view toward the need for sulfate criteria. However>
it should be emphasized that sulfates found in the ambient atmosphere
have not been well characterized. A working definition, for the
purpose of this document, ismaterial collected on a glass fiber
filter over a 24-hour period and analyzed as water soluble sulfate.
Specific sulfates are not identified.
Biased upon theoretical considerations, the primary urban sources
of sulfate is the atmospheric oxidation of S(>2 to ^SO^ with subsequent
neutralization or exchange reactions giving a variety of sulfates.
Few data are available on the sulfate content emitted directly
from stationary sources, Sulfate from mobile sources has not been
considered significant due to the low sulfur content of refined fuels.
Approximately 95 to 98% of these emissions are in the form of SC^. The
remaining few percent is primarily 803 which is rapidly converted to
sulfuric acid.
The chemical mechanisms for converting S02 to HjSO^ in the atmosphere
are not well known. Two current hypotheses are: 1) catalytic oxidation,
and 2) chemical oxidation by photochemically generated reactants.
The principal mechanisms for removal of sulfur from the atmosphere
are precipitation, and deposition on soil and vegetation. Essentially
all of the SO- in the atmosphere is converted to XSC>4 prior to or during
-a-
-------
DRAFT
DO NOT QUOTE OR CITE
the removal processes. Estimates of residence times range from
approximately one day to one week. Therefore, the effects of sulfur
loading in the atmosphere by man's activities may be significant
several hundred miles from the source regions.
There is a substantial volume of aerometric data on water
soluble sulfates, but little on sulfuric acid or individual sulfate
compounds. In 1970 the national average sulfate concentration at
urban locations was 10.1 yg/nr*. The 24-hour maximum observed w"aฃ 197
pg/m^. The non-urban average was 6,3 yg/m^, A slight seasonal
variation was observed at non-urban sites. No marked long-term trend
has been observed in the United States, despite' the decrease in S02-
Available data indicate that essentially all of the sulfate particles
in the free atmosphere are in the respirable size range.
Most available data have been collected using either impaction or
filtration techniques. Measured values of size and number density are
not obtained. Total sulfate concentrations have been determined largely
by colorimetry, and the acidity by colorimetry or pH techniques. There
is a question concerning the sample integrity using these sampling
techniques. A suitable method for measuring sulfuric acid in the
atmosphere is not available.
Experimental biological studies have indicated that certain of
the paniculate sulfates have a greater biological effect than SC^.
It has been demonstrated that sodium chloride in an atmosphere of S02,
and high relative humidity, enhances the biological response. Large
differences have been found in the degree of biological activity among
-b-
-------
the various sulfate compounds, and as a function of particle size. However,
there are serious gaps in the toxicological information concerning sulfates.
Most of the experiments, performed by one laboratory using the guinea
pig, have been concerned with the effect of acute exposure on pulmonary
function. There is a paucity of information concerning the role of
sulfates (or even S0_) in the production of chronic pulmonary disease. The
relevance of the toxicological studies conducted to date to ambient
pollutant atmospheres is questionable.
Recent Community Health and Environmental Surveillance System
t
(CHESS) program studies indicate that adverse health effects
may be more closely associated with suspended sulfate than with SC^
or total suspended particulate. The CHESS investigators felt there was
substantial evidence that the levels of sulfate necessary to cause
adverse health effects were one to two orders of magnitude lower than
with S02 or total suspended particulate. Laboratory studies have shown
that sulfuric acid and some metallic sulfates are more potent irritants
than SC>2. More information is needed on the irritating effects
relative to specific sulfates and various mixtures found in polluted air,
and the relationship with temperature, particle size and relative
humidity.
Acid rain may adversely effect the pH of soil and fresh water lakes,
and hence the ecology at great distances from emission sources. However,
under certain conditions, the acid rain may be beneficial. Acid rain
also results in the leaching of mineral nutrients from plant surfaces.
The sulfate ion is considered to be the form which is toxic to plants.
S02 entering through the stomata is converted to S03 and S04. SC^.
-c-
-------
00 NOT QUOTE OR CITE
adsorbed by the leaves may be converted to t^SO^ with subsequent damage.
Sulfates accelerate the degradation of certain materials. Rates
of deterioration have not been determined.
The impact of sulfates on weather, visibility and climate evolves
from their chemical and physical properties. Atmospheric sulfate
(and other) particles influence the heat budget, visibility, and
cloud and precipitation processes. They play a role in th.e development
and intensity of the so-called heat island. Particle size and
number density are the important parameters, Significant effects are
known but quantitative data are not available to characterize the
problem.
Although there are many uncertainties concerning the total sulfur
cycle in general and sulfates in the atmosphere in particular, some
conclusions can be drawn: 1) Sulfuric acids and sulfates known to
exist in the atmosphere in sufficient concentrations can have a
deleterious effect upon human health and welfare, Certain of these
compounds appear to be more potent irritants than SC>2. The irritant
response of certain mixtures is probably greater than the sum of the
responses of the individual compounds, and is related to the size
and number density of the sulfates{ 2} Sulfates present in the
atmosphere may have an adverse effect upon the weather, climate, and
visibility; 3) Removal by atmospheric processes may result in adverse
ecological effects at large distances from sources; 4) Sulfates
exit in the atmosphere as a result of natural processes; however, the
-d-
-------
DRAFT
DO NOT QUOTE OR CITE
concentration is significantly influenced by anthropogenic activities;
5) Current knowledge and available data are Inadequate at this time
to establish criteria which might be used as a basis for standards for
control.
It is recommended that particular attention be devoted to the
problem of sulfuric acid and sulfates in the atmosphere in order that
the question of pollutant potential from anthropogenic sources might
be resolved. Study areas should include: 1) the relationship
between adverse health and welfare effects and sulfates in the
atmosphere as a function of particle size, number density, temperature,
humidity, pressure, and chemical composition; 2) the biological
effects of sulfates using a variety of specimens in realistic dynamic
atmospheres as related to the variables listed in (l)j 3) principal
mechanisms, or reactions, and rates for conversion of S02 and I^S to
sulfuric acid and sulfates in the atmosphere; 4) suitable techniques for
i
routine sampling and measurement of sulfuric acid, H2S, and specific
sulfates in the atmosphere, including size distribution and number
density, and implementation of an adequate monitoring program; and
5) potential technology to achieve an adequate reduction in sulfur
emissions from point sources, A balanced research and development
program well coordinated in time and substance will be required. A
summary of the recommended research and development programs follows.
Detailed tracks are included in Section VII. No attempt has been
made to structure the research program in terms of priority of tasks,
It is not envisioned that it would be necessary to implement such a
broad program at a given starting date. The estimated time for completion
-e-
-------
DO NOT QUOTE OR C/TE
applies to each given task. Sequential scheduling would be desirable
for at least a portion of the program. For example, the scope of the
health effects program clearly would be dependent upon measurement
capability and the characterization of sulfates in the atmosphere.
Research Program
1. Modification and expansion of health effects
research to include specifically sulfuric acid
and suspended sulfates in the ambient atmosphere
as related to acute and chronic diseases, and
mortality.
2. Conduct biological experiments to determine
the ecological effects of atmospheric sulfates,
3. Conduct studies to determine the rate and
mechanisms for sulfuric acid and sulfate
formation in urban atmospheres, and
expansion of monitoring program to include
sulfuric acid, ammonium sulfate, and other
sulfates as feasible.
4. Develop reliable field and laboratory
methods for measuring acids and sulfates in
the ambient atmosphere.
5. Develop methods for measuring sulfate
emissions from stationary sources and
determine the characteristics of the
sulfur emitted.
6. Conduct a technology development program
to achieve the required abatement of total S02.
Time Period
for Completion
(Years)
5 to 10
-f-
-------
DRAFT
NOT QUOTE OR CITE
I. ENVIRONMENTAL APPRAISAL
A. Introduction
Of the total sulfur emitted by air pollution sources, approximately
95% is in the form of S02. Therefore, the major source of sulfates
in urban areas is the atmospheric conversion of gaseous S02 to
=
particulate sulfate, 804. Several atmospheric processes are involved
in this conversion, none of which are characterized, However, since the
origin of sulfate is also the fate of S02 some information on SO^,
sulfuric acid, and sulfate formation has been developed in S02 studies,
In the atmosphere, SO^ immediately reacts with H20 to form H2S04.
1. Photooxidation
The photodissociation of S02 (S02 + h^t -* SO + 0) is energetically
o
possible only at wavelengths below 2180 A, which are not present
in sunlight in the lower atmosphere. Therefore, the atmospheric
photochemistry of SO can only involve molecular reactions of
electronically excited sulfur dioxide molecules -- (802)* -- which are
thought to react with oxygen to generate the intermediate 804 species
(Reaction 1-1) . Then in air, 80^ could generate 803 and ozone by
Reaction 1-2.
so2* + o2* soj (i-i)
+
SOJ + 02 -ป - S03 + 03 (1-2)
The major sunlight absorption of 802 occurs witnin tne strong S02
band extending from 3400 A to 2400 A. Initially this absorption results
in the formation of electronically excited singlet state S02 ( S02). A
O o
second very weaJc absorption region of S02 extends'' from 400ฐ A to 3400 A;
-1-
-------
DRAFT
DO NOT QUOTE OR CITE
absorption of light within this region results in the formation of
electronically excited triplet state S02 ( S02). Emission studies1'2
show that both ^(^ and S02 are ultimately formed when SO is excited
within the 2400-3400 A region. However, there is evidence that the
chemically reactive species in the photochemistry of SO. in clean air
is the SO- molecule. The role of S02 is in the generation of
3S02 molecules by intersystem crossing.
In terms of our present knowledge? the triplet sulfur dioxide
molecule would be involved in Reaction 1-1. With the conservation of
electron spin, SO- would react with 0* ( Eg") ground-state molecules
to give an energy-rich, singlet SO. species, Reaction 1-3. The lifetime
of the very simple S04 molecule should be short, since it has few degrees
of freedom in which it might share the excess vibrations! energy. Unless
some vibrational energy is removed in a collision with other molecules
(Reaction 1-8 ), it should dissociate in an early vibration. There are
several alternative dissociation modes which are energetically possible,
Reactions 1-4 to 1-7.
2 (Zg-) S04| (1-3)
3S0 + 0 (3 Zg') (1-4)
2
(3 Eg") (1-5)
(1-6)
SO + 0(3P) (1-7)
S04| + M -ป S04 + M (1-8)
S0 * 0 -v S0 + 0 (1-9)
-2-
-------
DRAFT
DO NOT QUOTE OR CITE
Reactions 1-5 and 1-7 require a spin inversion and would be expected to
be less important than 1-4 and 1-6.
This simple sequence of reactions commonly accepted in the description
of S02 photooxidation probably is not correct in detail. Specifically:
1) 804! never has been identified as a transient in these species; 2) 03
formation never has been shown to accompany S02 photooxidation, although
4
tests have been made; and 3) the pressure dependence, predicted if 3,
4, 8 and 9 were the important reactions, was not observed in the
studies. Therefore, alternative mechanisms need to be considered.
3
The rate constants (at room temperature) for the S02 reactions with
various atmospheric components have been determined by Calvert et al.,
and are given in Table 1-1. Several conclusions can be drawn from the
3
rate constants. Although a chemical reaction between SO- and N , Ar
and other rare gases is very unlikely, the quenching rate of SO by
these gases is significant. The quenching rate constant for 3S02-quenching
by H20 and 03 is about a factor of 10 higher than those of the major
atmospheric gases. For paraffinic hydrocarbons the quenching rate
2
increases with ease of C-H bond rupture. The rate constants for SO-
quenching by olefinic hydrocarbons and NO are near the collision number
(^2x10* liters/mole-sec). Because of the high quenching rate constant
excited S02 molecules may react with olefins even when the olefins are
present at only parts-per-hundred million levels.
It is generally assumed that a fraction of the quenching collisions
with 02 will result in oxidation. Since water is a very effective quencher
the relative humidity is an important parameter. The quenching by oxygen
varies from 23.4% for dry air to 17.5% for air at a 100% relative humidity.
About 25% of the 3S02 species is quenched by water molecules at 100%
relative humidity. Therefore, the photooxidation of S02 might be expected
-3-
-------
TABLE 1-1. SUMMARY OF QUENCHING RATE CONSTANT DATA FOR SULFUR DIOXIDE TRIPLET MOLECULES
WITH VARIOUS ATMOSPHERIC COMPONENTS AND COMMON ATMOSPHERIC CONTAMINANTS AT 25ฐC26
kg liter/mole-sec
Compound -ป--._ x 10"8
Nitrogen 0.85 + 0.10
Oxygen 9.96 + 0.05
Water - 8.9+1.2
Argon 0.52 ฑ 0.05
Helium . 0.68 + 0.07
Xenon 0.71 ฑ 0.11
Carbon Monoxide 0.84+0.04
Carbon Dioxide 1.14+0.07
Nitric Oxide 741+33
Ozone ' 11.0+1.2
* ~
Sulfur Dioxide 3.9 + 0.1 ^
Methane 1.16+0.16 ^
Propane ' 5.11 ฑ 0.58 O
Propylene 850 + 87 'O CD
Cis-2-butene ' 1340 + 98 c
-------
DRAFT
DO NOT QUOTE OR CITE
to depend on relative humidity. It is not known if quenching reactions of
3S02 with water vapor could lead to oxidation of the S02.
Based on the assumption that all 3S02 quenching by 02 results in
oxidation, Calvert et al.3 calculated a theoretical maximum rate for
the homogeneous direct photooxidation of SO in sunlight of about 2% per
hour. Some results of the numerous laboratory studies of the photooxidation
of S02 are tabulated in Table 1-2. The rates and quantum yield of SO
photooxidation in mixtures of S02 and pure air Show large discrepancies.
However most of the rate measurements range from a few hundreds to a
few tenths of a percent per hour. Comparison with Calvert's
theoretical maximum rate and quantum yield suggests that chemical
reaction of excited S02 (Reaction Ir3) occurs in only a few of the
quenching collisions with 02. In polluted atmospheres there are
other possible reactions, homogeneous and heterogeneous, which probably
play a more important role in S02 oxidation. The oxidation rate given
in Table 1-2 varies over two orders of magnitude,
None of these studies shown in Table 1-2 can be considered completely
adequate for two basic reasons. First, it is extremely difficult to make
accurate measurements of light absorption, S02 loss, or SO, formation.
In addition, most of the quantitative measurements were made with high
concentrations of S02> Under these conditions, the dominant quenching
reaction is the one involving SO molecules, not .the one involving 02
or H20 molecules. Thus, in a 50% mixture of S02 in ฃ>, 80% of the S02
quenching is by S02 molecules. Therefore, extrapolation of quantum
yield data for high SO concentrations cannot be expected to yield a meaningful
prediction of atmospheric S02 photooxidation rates. Recent studies in
-5-
-------
D0 KฐT QUOTE OR
RESULTS OF STUDIES ON PHOTOOXIDATIOM OF SULFUR DIOXIDE (Room temperature
and atmospheric pressure unless otherwise stated).
Reference Reactants Quantum S02 Oxidation
_ ' (Cone, range) yield, 9 x 10 rate, % hr
5 S02 (56-230 torr) 1.5 0.005
02 (5-200 torr)
6 S02 (5-30 ppm) 1-10 0.1 to
in moist air 0.7
7 S02 (0.2-0.6 ppm) 0.3 6.6 to
in purified air 27
8 S02 (12 + 19 ppm) - 0.6
in moist air
9 S02 (60-100%) 0.5-0.22
.in 02
10,11 S02 (20-100 torr) 17 -
02 (50-390 torr)
12 S02 (.15 to .65 ppm) 0.5 0.04-0.62
in purified air
13 S02 (50-100 torr) 6.4-8.8
02 (50-100 torr)
14 S02 (2-70 torr) .0.4-28, 0.02-0.04
Ov (3-150 torr) 0.3 at 1 ppm at 1 ppm SO,
so2 2
15 SO^ (.1-3 ppm) in - 0.4 to 1
purified air of 10-20*
relative humidity . . .
16 S02 (0.15 ppm) in purified 0.05
air of .35% relative
humiditv _fir.
-------
DRAFT
DO NOT QUOTE OR CITE
clean smog chambers of the SCL photooxidation rate in clean, humid air
have yielded values between 0,1 and 1% per hour (references 15. and 16 ),
2. Chemieal Oxidation
As shown in Figure 1-1, during the photochemical smog processes, the
decay of SO* and the formation of sulfuric acid aerosol do not occur to any
appreciable extent until most of the NO is converted to N02ซ At
this point N03 begins to be formed from the 03 + N02 reaction, and organic
peroxides which earlier reacted rapidly with NO can begin to accumulate.
It seems likely that one or more of these oxidizing agents, rather
than the direct photooxidation of S02, may account for the formation
of SO and sulfuric acid mist,
a. 0_ + SO-. Although energetically possible, this reaction has bee'n
O ฃ
~~~~"~~~~~ 17 '18 '
shown to be slow in the gas phase, * presumably due to spin considerations,
18
It occurs very rapidly, however, in the presence of water droplets, and
might be expected to occur on the surfaces of moist particles.
bป SO- + NO, (N-0,.) . S0_ has been observed to react in the presence of
NO- and 0_. This could be due to reaction with N0_ or N-Or generated by
the reaction of 0_ and NO-. It is possible that compounds of S and N may
be formed since S0_ and NO- are known to form solid compounds. A possible
sequence is shown below:
N02 * ฐ3 ~" N03 + ฐ2 (
(-1-1?)
2NOSH04
-7-
-------
I
oo
I
0.0
1-BUTENE/NO/N02/S02
100
ELAPSED TIME, minutes
Figure 1-1. (Smog Profile.
150
200
4
-------
tmm i
DO NOT QUOTE OR CITL
Some support for this reaction system can be obtained from the observation
18
that peak N02 values are reduced by the addition of S02. In the absence
of S02, the sum of NO + N02 is equal to the initial NOX until the N02 peak
is reached. When S02 is present, however, the sum of NO + N0_ begins to
drop off earlier. The compounds (NO)2S207 and NOHSO have been identified
as products of the reaction of S02 and NO at the 100 ppm concentration
. . 19 ,
level.
c. S02 + ROX. There are a number of indications that oxidation of S02
by OH, HO or oxygenated organic radicals may be important. The most
straight-forward is the dark reaction of propylene-SO -ozone to yield
20 *
sulfuric acid. The specific radical responsible is not known, but
it might well be the peroxyacyl, peroxyacetyl, CH CO(00), in the case
of propylene. This radical is formed in large quantities only in the
reaction of olefins. It reacts rapidly with NO to yield NO so,
2 '
unless the reaction with S02 was very rapid, it would not be expected
to oxidize S02 unless most of the NO had been converted to N02. The case
for the peroxyacetyl is further supported by the observation that the
formation of PAN, thought to be due to the combination of RCO(OO) and
.ii
>
systems, eye irritation (presumably due to PAN) is reduced by the
presence of S02>
The dark reaction of olefin-SO_-ozone has been studied in more
V2~'
detail. It was found that for a series of olefins, the initial rate
of formation of SOj correlated well with the initial rate of the dis-
appearance of ozone (determined in the absence of S02). The oxidation
was attributed to a reactive species formed in the ozone-olefin reaction.
For atmospheric comparison, it was calculated that at ozone and propylene
concentrations of 50 ppb, the oxidation rate of 100 ppb S02 would be 0.4%
hr,'1 .9-
N02, is reduced by the presence of S02, and that with olefin/NOx
-------
DRAFT
DO NOT QUOTE OR CITE
d. Oxidation Rates. As shown in Table 1-3, the oxidation of SO in
photochemical smog can be expressed as a rate of %hr~ for comparison
with the direct photochemical oxidation. The rates would be expected
to depend on the concentration of other reactants. Observed rates vary
from 5 to 80% hr~ . In a study of auto-exhaust in a large smog chamber,
with initial values equivalent to those frequently found in Los Angeles,
-1 23
S02 oxidation rates of 5 to 10% hr were observed.
3. Aerosol Formation
In smoggjr atmospheres, the formation of aerosols or particulate matter
is of concern from both health and welfare considerations. The marked
influence of SO on aerosol formation in photochemical smog simulation
^
experiments is well documented in the literature. These studies are
tabulated in Table 1-4. An example of the effect of S02 on a variety
of photochemical smog systems is shown in Figure 1-2. Many of the studies
in Table 1-3 also include measurements of aerosol formation. Although
the exact mechanism by which SO is converted into aerosol remains
- 2
uncertain, the following general conclusions can be drawn.
(1) In the absence of S02, saturated hydrocarbons produce little, if any,
aerosol. In the absence of S02, aromatic hydrocarbons and olefinic
hydrocarbons with five or more carbon atoms from aerosols, those with four
carbons or less do not, Dienes, terpenes, and cyclo-olefins also form
26 27
aerosols in the absence of 802- '
(2) S02 increases light scattering in studies of individual hydrocarbons.
This is particularly noticeable with olefins where the increase is enormous,
There is a moderate increase with paraffins and a small increase with
aromatics.^
-10-
-------
DO NOT QUOTE'OR CITE
Table 1-3
SUMWVRY OF STUDIES ON PHOTOCHEMICALLY INDUCED OXIDATION OF SULFUR DIOXIDE
- '" . ..:/ ' ,' \ ' . \
. - ' ' . ' ' i
Reference System nn airy | SO- Oxidation
' Rate, % hr
..--"
21 S02 (0.1-1.0 ppm) . 50 to 300
NO, (1.0 ppm) .>-"
2-methyl-2-butene
(0.5 tc 3.0 ppm)
~24 -" S02 (1 ppm) 2S '
N02 (1 ppm)
1 hexene (4.5 ppm)
25 S02(0.05 ppm) . 10
NO (0.03 ppm)
'.. : (0.1 ppm)
cis-pentane
4~" SO (.1-75 ppm) 5-50
NO (.75-1.5 ppm) steady-state
various hydrocarbon9 80 max
shown In Figure 1-2
-< PP"0
-11-
-------
DRAFT
DO NOT QUOTE OR CITE
1000
AUTO EXHAUST
MIXTUREIOLEFINS + AROMATICS)
ISOOCTANE ,
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
m
i
250
. 4
Figure 1-2 Influence of S0ฃ On Photochemical Aersol Formation.
-12-
-------
Table 1-4
SUMMARIES OF STUDIES FOR PHOTOCHEMICAL AEROSOL FORMATION
DRAFT
DO NOT QUOTE OR CiTE
Reference
System
Aerosol Results
27
29,30
2,25
28,31
32
27,33,34
27,31
26,34 I
..3SJL6J....L.
4,20,,28- 7
27
27
26,32
27
-4,20
34,35
Paraffin-NO
Paraffin-SO.
Paraffin-NO -S02
Olefin-NO.
Olefin-NO
Olefin-SO,
Olefin-NO -SO
Alkyne-N0x
Alkyne-N0x-S02
Aroinatic-NO
Aromatic-N'O
Aromatic-N0x-S02
Aromatic-N0x-S02
Hydrocarbon mixture-
NO -SO,
No light scattering
Sulfur-containing products
such as sulfanic acids
found but at ppm con-
centrations no detectable
condensate observed
Very little, if any, light
scattering
Very little light scattering
except for highly branched
and cyclic olefins
Light scattering observed
with 1-Heptene
Oil product obtained at high
cone, but no condensate
observed at ppm cone. Produe
tion of aerosol suppressed :
by olefin
Significant increase in light
scattering, especially for
high carbon and branched
olefins. Identification by
IR and microchemical tests.
Differentiation between or-
ganic and inorganic aerosol.
No light scattering
Very little, if any, light
scattering
No light scattering
Significant Light-scattering
Slight to significant light
scattering
Slight.increase due to S02
Conflicting results as to
] ^whether mixture: resulted in,
a positive or. negative
*; aerosol'-forming: syriergism./
-------
DRAFT
DO NOT QUOTE OR CITE
(3) SO appears to influence the reaction rate parameters in simulated
smog. The degree of influence is dependent on such parameters as the
relative humidity, initial N02/NOX ratio, hydrocarbon/NOx ratio and
hydrocarbon type.18'27'2
(4) There is considerable controversy over the chemical composition of
aerosol formed from the hydrocarbon-NO -SC^ photochemical system. However,
the evidence is compelling that the aerosol formed from C <5 olefins-N02-S02
is predominantly inorganicmainly H2SO.--whereas the higher olefinic
7/20,36
hydrocarbons produce carbonaceous aerosol as well. In the
presence of NH_, NH.SO. aerosol is observed. Wilson et al. report
that the infrared spectra of dried, CH Cl_ extracts, of atmospheric
and smog chamber aerosols, show C-H absorptions around 3 y , C = 0
absoprtions around 6.5 y, C = C and/or N = 0 absorptions around 6.7y
and a number of peaks at longer wavelengths. Gaschromatographic-mass-
spectrometric studies indicated organic acids and nitrate esters were
36
also present. Infrared spectra of aerosols formed from 1-heptene-NO
A
with and without S02 indicated that with S02 the aerosol contained a
4
sulfonic acid.
(5). The disagreement and confusion in the literature about the SO-
effect on aerosol formation is a result of the dependence of the S02
effect on the surface condition of the reaction vessel, the stirring
rate, the average reaction time, the relative humidity, the type of
4 27
hydrocarbon, and the type and number of pre-existing nuclei. '
-14-
-------
DRAFT
4. Catalytic Reactions DO NOT QUOTE OR CITE
The interaction of SO with solid and liquid aerosols may be an
important removal mechanism. The extent of removal is dependent upon
chemical composition of the aerosol, its phase, surface area, and
concentration. The rate of reaction may be controlled by SO- diffusion
to the aerosol surface, attachment to the surface, or diffusion of reactants
and products in the particle or droplet. The theory of mass transport
Jo
to aerosol particles is presented in a monograph by Hidy and Brock,
and the rate of reaction of gases with aerosols is discussed by Cadle and
39 40
Robbins. Also, an estimate has been made by Mottershead for gas
collision rates with aerosols; and it appears that the time required
for one-half of the S0_ gas molecules to collide with Los Angeles smog
(1000 yg/m ) is about 5 seconds.
The surface area of suspended aerosol will influence the extent of
SO- interaction. The surface areas of aerosols collected from Italian
2 ' 41 2
cities were reported by Liberti to be about 6.2 m /g. The total pore
volume was 0.104 mfc/g, of which 80% consisted of pores with radii below
o
42 4,
100 A. Corn et al. ' have made measurements for Pittsburgh. The
samples were degassed at 25ฐC, surface area determined, degassed at 200ฐC
and surface area redetermined. The specific surface area was increased
by degassing at the higher temperatures indicating that the particles
become smaller due to loss of volatile material. Values ranged between
1.4 to 4.5 m /g for samples degassed at 25ฐC, and 2.3 to 8.0 m2/g for
samples degassed at 200ฐC. Seasonal variations were demonstrated. A
few studies of the interaction of S02 with powders at ambient
temperatures have been reported. Adsorption/desorption isotherms of
S0 on silica gel (degassed at 240ฐC) over the temperature range
-15-
-------
of 263-323ฐK, reported by Jones and Ross, 5 exhibited complete reversi-
bility. An ultra-high vacuum, dry system was used. The isosteric heat
of adsorption between S02 surface coverages of 20% to 35% varied
linearly between 9.3 to 7.8 kcal/mole, which implies that SO was physically
adsorbed. Colorimetric heats of adsorption and adsorption isotherms were
"46
measured by Glass and Ross for Fe203, Mn203, V205, MnS04, and
"NiS" all supported on silica gel. The measurements were made between
273ฐ and 423ฐK using an ultra-high vacuum, dry system. The
thermodynamic data for the lower temperature range was not discussed,
but at 273ฐK, for surface coverage of about 0.15, the isoteric heat
of adsorption for each of the catalysts would be greater than 20
kcal/mole, which indicates formation of S02 on the surface, and was
confirmed by qualitative chemical analysis for sulfate.
Assuming that the S02 interaction with the surfaces of
suspended aerosols could be described by a Langmuir adsorption
47
isotherm, Pilot calculated the fraction of S02 that would be
adsorbed as a function of concentrations of suspended aerosols and
gaseous S02. However, the arbitrary value he chose for the "Langmuir
adsorption constant" is orders of magnitude greater than that
calculated from theory, which may result in a gross over-estimate of
adsorbed SO . In a dynamic system, Smith et al showed that Fe_04
and A1203 aerosols irreversibly adsorb SO at near ambient
concentration; their conclusion that multilayer adsorption occurred
for S02 concentrations greater than 2 ppm is incorrect due to mis-*
calculating the S02 monolayer coverage by a factor of 100 too great.
-16-
-------
41
For collected atmospheric aerosols, Liberti et al. were unable to
desorb and detect S02- However, using X-ray photo-electron spectroscopy,
also known as "electron spectroscopy for chemical analysis" (ESCA),
49
Novakov et al. analyzed cascade impactor samples and obtained the
diurnal patterns of sulfite and sulfate; sulfite was predominant for
diameters less than 2.0 microns. ESCA was also used by Hulett et
al to investigate the oxidation state of sulfur on flyash from a
power plant, smoke particles from coal burned in a home fireplace>
and Mn02, Fe-O,, and MgO, and CaO exposed to SO . Sulfite and sulfate
/ o 2
were on the flyash and coal smoke, which also had a reduced sulfur species.
The spectra of the transition metal oxides (Mn02, Fe20j) indicated sulfate
to be the only sulfur species; the alkaline earth oxides (MgO, CaO) had
sulfite as predominant, with sulfate present.
Vannerberg and Sydberger constructed a "pEndiagram" (potential
energy diagram) for the system iron-sulfur-oxygen-hydrogen and concluded
that an iron oxide film in contact with an aqueous solution of S02
2-
will form SO. , which was experimentally confirmed. The mechanism suggested
was that Fe ions in the surface are reduced by S02 to Fe +, producing
H SO.; the Fe ions are then oxidized back to Fe + by atmospheric
2 4
oxygen.
The importance of relative humidity on the rate of oxidation of S02
52
by aerosols was demonstrated by Cheng et al. They stablilized MnS04, MnCl2,
CuS04, CuCl2, NaCl particles by deposition on Teflon beads in a fluidized
bed. By flowing air of various relative humidities through the bed, it was
observed that oxidation rate increased as relative humidity increased for
S02 concentrations ranging from 3 to 18 ppm. Marked increases in conversion
rate were noted when the particle became a liquid droplet. The investigators
concluded that the overall rate of reaction was controlled by the
chemical reaction. On a mass basis, MnS04, MnCl2, and CuS04 exhibited
-17-
-------
DRAFF
DO NOT QUOTE OR CITE
S02 oxidation rates of 12.2, 3.5 and 2.4 times that of NaCl. CuCl2 reacted
by forming H2S04 and CuCl. It was estimated that for natural fog formed in
an industrial environment, the rate of oxidation of S02 would be 2 percent/
hour. ,
Matteson et al. studied the catalytic oxidation of S0_ by aqueous
aerosols of MnSO. (95% relative humidity) and suggested that the
oxidation involved intermediate complexes. In this case the reaction
rate would be controlled by the diffusion of these complexes in the
droplet.
39
Cadle and Robbins investigated the rate of reaction between f^SO^
droplets suspended in N with NHj. The rate of reaction for droplets
composed of concentrated H2SO. was controlled by the diffusion of
reaction products in the droplet. It was estimated that only about
0.1 of the collisions of NH3 with the droplet surfaces resulted in
reaction. The reaction rate for dilute acid droplets was too rapid
for the investigator to measure. It was suggested that the rate
was controlled by gas-phase diffusion, as found by Johnstone and
Williams.54
Junge and Ryan studied the oxidation of sulfur dioxide in
solution and found that essentially no reaction occurred in the
absence of a catalyst. When ferric chloride was used as a catalyst,
oxidation did take place. The final amount of sulfate formed was
only slightly dependent on the concentration of the catalyst, but was
a linear function of sulfur dioxide concentration. Johnstone and
Coughanowr estimated from their study of sulfur dioxide oxidation
in small droplets that, if manganese sulfate were present as
3
1-micrometer crystals, the oxidation rate in fog droplets at 2.7 mg/m
(1 ppm) S02 would be about 1 percent per minute. Both investigators
found that manganese salts were more effective catalysts than iron
-18-
-------
DRAFT
DO NOT QUOTE OR CITE
sr
salts. Bracewell and Gall also measured rates of catalytic
oxidation of sulfur dioxide in droplets and estimated that, in the
presence of ferric or manganous ions, rates of oxidation could be
sufficient to account for the sulfuric acid content of urban fogs,
assuming sulfur dioxide concentrations of 1750 vig/m or about 0.6 ppm.
The oxidation of sulfur dioxide essentially stopped when the pH
of the water droplets decreased to about 2 in Junge and Ryan's
experiments.53..; They suggested that the effect is due, at least in part,
to the low solubility of sulfur dioxide in strongly acidic solutions.
If ammonia was present in the air to neutralize the acid as it was formed,
r o-.
oxidation of sulfur dioxide continued. Van den Heuvel and Mason
found that for given concentrations of ammonia and sulfur dioxide the
mass to sulfate formed was proportional to the product of the surface
area of the drops and the time-of exposure.
A mechanism for the non-^catalyzed oxidation of SO^ in water droplets
CO
has been proposed by Scott and Hobbs, They-assumed that SO desorbed
water to form H2SO , H+, HSOg, and SO^ and that SO^ would oxidize to
SO^ by desorbed oxygen. Since SO" but not HSO~ is oxidized by
" o 3
desorbed oxygen, the reaction would become slower as the droplet became
more acidic. Atmospheric ammonia would be desorbed in the particle
reducing the acidity and promoting the conversion of S0_ to sulfate.
This work has been revised and extended by McKay, ' who argued that
oxidation rate is ten times greater than that deduced by Scott and
-19-
-------
DRAFT
DO NOT QUOTE OR CITE
Hobbs. He also concluded that the non-catalyzed reaction rate between
NH and SO in water droplets should increase as temperature is
3 2
lowered, due to increased solubility of NH^ and S02 more than
compensating for the decrease in the rate constant for sulfite
oxidation.
5. Formation of Other Sulfates
Much less is known about the formation of other sulfates. H_SO,
can be neutralized by ammonia. H-SO can also react with a variety
of metals, metal oxides, or metal salts to yield the corresponding
sulfate. The best known of these reactions is the replacement of
Cl in sea salt to yield Na2S04 and NaHS04.
6, Summary of SO? Oxidation Mechanism
1. Direct Photooxidation or the Clean Air Mechanism
light
S07 -^ SO.
V*
SO, + HO ) H SO,
0 2 2
[S0=] -u [S02] x [light]
S02 is activated by absorption of light. The excited S02 reacts
with oxygen. Sulfate formation will be proportional to the concentration
of S0_ and the light intensity. This reaction sequence is too slow to
be important in urban areas. The light is less than 1 percent per hour.
However, this mechanism may be important in long range transport of SO .
-20-
-------
DO NOT QUOTE OR CITE
2. Indirect Photooxidation or the Dirty Air Mechanism "1"
smog
S07 } SO-
[50=]
During the photochemical smog processes, especially in the case of
olefins, oxygenated free radicals are formed which can oxidize SC^ to
SOj. Although the rate of reaction depends on [802] x [RO^] , the reaction
is fast and the amount of sulfate formed depends more on the concentration
of radicals than on the concentration of S02.
3. Air Oxidation in the Condensed Phase or the Clean Water Mechanism
liquid
S07 - - > H0SO,, SOZ, HSO;, and H
H,0 2333
NH, + HSO: -^ NH"*" + so
3 ^4
liquid H_0
SO" > SO
3 ~ 4
*
SO dissolves in ttater droplets, or water on the surface of
aerosol particles, to form sulfurous acid and its dissociation products.
-21-
-------
DO NOT QUOTE OR CITE
The SO- is converted to SO^ by desorbed oxygen, but the process slows
down as the droplet becomes more acidic. Absorption of NH increases
the pH and prevents the reaction from proceeding.
4. Catalytic Oxidation in the Condensed Phase or the Dirty Water
Mechanism
S0
heavy metal ions
[S0|] ^ [Catalytic Metal Ions]
The presence of heavy metal ions such as Fe, Cu, V, and Mn in
water sollution will catalyze the conversion of SO to SO^.
None of these mechanisms are understood quantitatively7 or qualitatively
Mechanism #1 is important on a global scale but not on an urban scale.
Mechanism #2, #3, and #4 are probably important on both urban and
global scales. For mechanism #2, #3, and #4, the amount of sulfate formed
is not very dependent on the SO- concentration (once it is above a certain
level), but is dependent on other factors: Type and intensity of
photochemical smog concentration of ammonia, or concentration of metal
catalysts. Therefore, reduction in S02 will not produce a proportional
reduction in sulfate. This result has been observed (reference 60) as
discussed in Section C - Atmospheric Observations. An analysis of the
annual average S02 and sulfate levels in 18 U. S. cities over a five year
period, indicated no significant difference in sulfate between 80 and
3 3
200 mg/m of S02; proportional reduction was obtained below 80 rag/m SO 0
-22-
-------
DRAFT
DO NOT QUOTE OR CITE
This suggests that control of sulfate will require very great controls
of S02 or alternatively control not of S02 but of the reactants which
determined the formation of sulfate from SO?.
B. Atmospheric Observations
A considerable volume of measurements is available from 24-hour
integrated network measurements of SO- and suspended water-soluble
sulfates at urban and nonurban continental and maritime sampling sites
61'
in the United States. Altshuller, has recently evaluated these data
emphasizing the distribution of and relationship between S02 and sulfate.
The main results are summarized in Table 1-5. At eastern urban sites
(east of the Mississippi River), sulfate contributed 10-20% of the
annual average particulate loading while at western urban sites
(west of the Mississippi) sulfate contributed 5-10% of the annual
average particulate matter. At eastern non-urban sites, sulfate
contributed 15-25% of the particulate matter. Although the annual
-2
average SO. to particulate ratios for western non-urban sites were
similar to eastern ratios, sulfate did not decrease as rapidly as total
particulate matter so sulfate contributed 25-35% of the total at
the lowest particulate loadings.
The presence of free H^SO. in aerosols is reputed to be particularly
significant in deleterious health, vegetation and corrosion effects.
While much is known about the concentrations of many inorganic
components in aerosols, very few studies have yet been undertaken on
62
the content of free acids. Junge and Scheich's measurements on
aerosols in Rhein-Main and Ruhr regions, London and in Goteborg
indicated the following: (1) the hydrogen ion concentration
-23-
-------
DRAFT
DO NOT QUOTE OR CITE
TABLE 1-5. SULFUR DIOXIDE AND SULFATE IN THE ATMOSPHERE
61
i .
\
so4'2 (%)
Total Particulate
1
Distribution of
SO, yg/m3
.
Distribution of
Urban i
t
East West
10-20 5-10
'
66a
.
13.5
so2 '
S07 : 4'9
1
22a
l
i
6.4
1
\
3.4
i
Non- urban
East
15-25
West
15-25
, ,
7.9C
i
2.6C
i
l
i
i
a. 1964-1968 average concentration.
b. 1968 average from 4 eastern and 1 western non-urban site.
c. 1965-1968 Javerage concentration. . ,.
-24-
-------
of aerosol particles is either due to weak acids or to strong acids
(presumably H980.) which are buffered by weak bases; (2) most of the
water soluble compounds of the aerosols and particularly the
hydrogen ion content is associated with particles smaller than Oily
radius; (3) correlations of the hydrogen ion concentration with water
soluble material, sulfate concentration and S02 concentration.were
low.
Altshuller, by assuming a11 of the most abundant
analyzed metal cations (Zn, Fe and Pb) existed as sulfates, has shown that
on the average only about 20% of the sulfate could exist as these
metal salts. Likewise if all of the available ammonium were associated
with sulfate, only 20 to 30% of the sulfate would be in the form of
(NH.)_ SO.. He thus concludes that 50% is a reasonable percentage
of the sulfate that can exist as sulfuric acid.
63
Atkins, Cox, and Eggleton have reported measurements of 0,, total
sulfate, and hydrogen ion (assumed due to sulfuric acid),. The sulfuric
acid comprised 20 to 60% of the total sulfate and its concentrations
was highly correlated (0.71)with that of 0_. Schuetzle, Crittenden,
64
and Charlson, in time-resolved, mass spectrometric studies of
collected aerosol, have been able to determine H-SCK, NH HSO ,
(NH )2SO , and NaHSO.. In one hi vol sample, not protected
from ammonia during storage, 12% by weight of the total sulfate
appeared as sulfuric acid.
-25-
-------
nn
DO NOT QUOTE OR CITE
O 14
O
u.
Ul
O
ง
ee
12
10
8
6
4
2
0
100 200 ' 300 .
SULFUR DIOXIDE CONCENTRATION (|jg/m3)
a
30 gฃ
2.
O
20 t
ui
U
10 ฐ
I
. to
400
Figure 1-3. Relationship between sulfur dioxide and sulfate concentrations.
61
-26-
-------
DRAFT
DO NOT QUOTE OR CITE
Field studies in a number of geographical locations suggest
that a relationship exists between SO- and sulfate. As reported
in 1969 Air Quality Criteria for Sulfur Oxides unpublished analysis
of 1962 and 1963 Air Quality Data National Air Sampling Network show
that the correlation coefficient between S02 and suspended sulfate
61
ranges between 0.5 and 0.9. However, Altshuller's analysis
of 5-year results from 10 representative urban sites shows that this
relationship is by no means simple. Plots of sulfate concentration
and the S02 to sulfate ratio as a function of SO- are shown in Figure 1-3.
Sulfate does not increase as rapidly as sulfur dioxide at urban
sites and sulfate concentrations tend to plateau with increasing
sulfur dioxide con-cent rat ion (exclujdLng the highest point which
represents a New York City sampling site). Altshuller concludes that
sulfur dioxide to sulfate ratios approaching 1 to 1 should occur at
very low SO- levels, and that the results in Figure 1-3 indicate that
reductions in sulfur dioxide will not necessarily result in proportional
reductions of sulfate.
In recent years, some studies have been made of the particle size
distribution of suspended atmospheric sulfate. This property
65
determines visibility reduction and is also an important factor in
physiological responses because of its relation to the degree of
66
penetration and retention of particles in lungs. Roesler et al.
reported sulfate size distributions in downtown Chicago and Cincinnati
based on 24-hour samples collected with cascade impactors. They found
values for mass median diameter (MMD), i.e., for equivalent spheres of
unit density, that average 0.3 to 0.4 micrometers in the two eities.
-27-
-------
nn
DO NOT QUOTE OR C(Tฃ
These were within the range of average MMD value for total sulfur
(0.2 to 0.9 micrometer) found by Ludwig and Robinson in the
Los Angeles and San Francisco Bay areas. From 8-hour samples
collected continuously for a week in each of four cities, Wagman
68
et al. found values for average sulfate MMD (i.e., 0.42)
micrometer in Chicago, Cincinnati, and Fairfax, and 0.60 micrometer
in Philadelphia) that were in good agreement with these measurements.
They also found that sulfate particle sizes generally increased with
increasing relative humidity, whereas sulfate concentration was more
closely correlated with absolute humidity. Of particular significance
is the fact that all of these investigations showed that a major
fraction (generally 80 percent or more) or urban atmospheric sulfate
is associated with particles below 2 micrometers in diameter.
Suspended sulfate is therefore largely in the respirable fraction of
particulate matter, and is associated mainly with particles that
cause the most pronounced reduction in visibility.
69
In 1961 Junge discovered a sulfate aerosol layer at an
altitude of 20 kilometers in the stratosphere. Using collection
devices carried aloft in balloons and high-flying U-2 aircraft, he
found that a large percentage of the aerosol particles are ammonium
sulfate. Most of the sulfate particles have a radius of a few
tenths of a micron. The origin of these stratospheric sulfate
particles is not clear. One possibility is that these particles
are first produced in the troposphere and then rise into the
stratosphere, another is that they are formed at their observed
altitudes. This stratospheric sulfate layer has recently been
discussed by Lazrus, et al. and Rosen..
-28-
-------
' DRAFT
DO NOT QUOTE OR CITE
Q. References
1. Rao, T.N., Collier, S.S., and Calvert, J.G., "Primary Photophysical
o
Processes in the Photochemistry of Sulfur Dioxide at 2875 A",
J. Amer. Chem. Soc., Vol 91, pp 1609-1615, 1969.
2. Rao, T.N., Collier, S.S., and Calvert, J.G., "The Quenching
Reactions of the First Excited Singlet Triplet States of Sulfur
Dioxide with Oxygen and Carbon Dioxide", J. Amer. Chem. Soc., Vo'l 91,
pp 1616-1621, 1969.
[
3. Sidebottom, H.W., Badcock, C.C., Jackson, G.E., and Calvert, J.G.,
> . '
"Photooxidation of Sulfur Dioxide", Environ. Sci. Technol., Vol 6,
pp 72-79, 1972.
4. Wilson, Wra. ฃ, Jr., "Aerosol Formation in Photochemical Smog. II
The Role of Sulfur Dioxide", Personal Conimunioatiwt. .'"!'.
5. Hall, T.C. Jr., "Photochemical Studies of Nitrogen Dioxide;
and Sulfur Dioxide", Ph.p. Thesis, University of California at
Los Angles, 1953.
6. Gerhard, E.R. and Johnstone, H.F., "The Photochemical
Oxidation of Sulfur Dioxide to Sulfur Trioxide and Its Effects on
Fog Formation", Ind. Eng. Chem., Vol 47, pp 972-976, 1955.
7. Renzetti, N.A. and Doyle, D.J., "Photochemical Aerosol Formation
in Sulfur Dioxide - Hydrocarbon Systems", Intern. J. Air Pollution,
Vol 2, pp 327-345, 1960.
-29-
-------
8- Uvone, P., Letsep, H., Noyes, C.M., and Parcher, J.F., "Static
Studies of Sulfur Dioxide Reaction in Air", Environ. Sci. Technol.,
Vol 2, pp 611-618, 1968.
9. Billings, C.E., Berger, A.W., Dennis, R., Driscoll, J., Lull, D.,
and Warneck, P., "Study of Reactions of Sulfur in Stack Plumes",
First Annual Report, Contract No. PH-86-67 125, GCA Corporation,
Bedford, Mass., March, 1969.
10. Sethi, D.S., Allen, E.R., and Cadle, R.D., "Investigation of
the Room Temperature Photpoxidation of Sulfur Dioxide", Abstract,
Fifth International Conference on Photochemistry, Yorktown Heights,
New York, September, 1969.
11 Sethi, D.S., "Photooxidation of Sulfur Dioxide", J. Air Pollut.
Cont. Assoc., Vol 21, pj> 418-420, 1971.
12. Cox, R.A. and Penkett, S.A., "The Photooxidation of Sulfur
Dioxide in Sunlight", Atmos. Environ., Vol 4, pp 425-433, 1970.
13. Allen, E.R., McQuigg, R.D., and Cadle, R.D., "The Photooxidation
of Gaseous Sulfur Dioxide in Air", Chemosphere, Vol 1, pp 25-32,
1972.
14. Cox, R.A., "Quantum Yields for the Photooxidation of Sulfur Dioxide
in the First Allowed Absorption Region", Journal of Physical Chemistry,
Vol. 76, pp. 814-820, 1972.
-30-
-------
nn DRAFT
DO NOT QUOTF OR CITE
IS. Clark, W.E., Thesis, 1972, University of Minnesota, Minneapolis,
Minn. 55455.
J6. Kosmand, W. Calspan, Buffalo, N. Y. 1973. Personal Communication.
17. Cadle, R.D. and Magi11, P.L., "Chemistry of Contaminated ,
Atmospheres", in Air Pollution Handbook, P.L. Magill, F.R. Holden,
and C. Ackley (eds.), McGraw-Hill, New York, N.Y., 1956, pp ,3-20 - 3-21.
18. Wilson, Wm. E., Jr., Levy, Arthur and Wimmer, D., "A Study of Sulfu
Dioxide in Photochemical Smog. II. The Effect of S02 on Formation of
Oxidant", Journal of the Air Pollution Control Association, Vol. 22,
pp. 311-326, 1972. .
19. Urone, Paul, Schroeder, W.H., and Miller, S.R., "Reactions of Sulfu:
Dioxide in Air", in Proceedings of the Second Clean Air Congress, pp, 370-
375. :
.20. Groblicki, P.J. and Nebel, G.J., "The Photochemical Formation
of Aerosols in Urban Atmospheres", in Chemical Reactions in Urban
Atmospheres, C.S. Tuesday (ed.), Elsevier Publishing Company, New York,
1971, pp 241-267.
,21, Schuck, E.A. and Doyle, G.J., "Photooxidation of Hydrocarbons
in Mixtures Containing Oxides of Nitrogen and Sulfur Dioxide",
Air Pollution Foundation (Los Angeles) Report 29, October, 1959.
.22. Cox, R.A. and Penkett, S.A., "Oxidation of Atmospheric S02 by
Products of the Ozone - Olefin Reaction", Nature, Vol. 230, pp. 321-322,
1971.
-31-
-------
DRAFT
00 NOT QUOTE OR CITE
23. Wilson, Wm. E., Jr., Miller, D.F., Hopper, D.R., and Levy, Arthur,
"A Study of Sulfur Dioxide in Photochemical Smog. III.", American
Petroleum Institute Project S-ll, Battelle Memorial Institute, Columbus,
Ohio, 1971.
24. Ripperton, L.A., Decker, C.E., and Page, M.W., "Effect of
SCL on Photochemical Oxidant Production", presented at American
Chemical Society Meeting, Division of Water, Air and Waste Chemistry,
.Atlantic City, N.J., September 13-17, 1962.
25. Cox, R.A. and Penkett, S.A., "photo-oxidation of Atmospheric S02",
Nature, Vol. 229, pp. 486-488, 1971.
26. Miller, D.F., Levy, A., and Wilson, W.E. Jr., "Aerosol Reactivity
Study of Hydrocarbons", Report to American Petroleum Institute,
BattelJe Memorial Institute, Columbus, Ohio, September, 1971.
27. Wilson, Wm. E., Jr., Merryman, E.L., and Levy, Arthur, "Aerosol
Formation and Visibility Reduction in Photochemical Smog", American
Petroleum Institute Project S-ll, Battelle Memorial Institute, Columbus,
Ohio, 1969.
28. Wilson, W. E., Jr. and Arthur Levy. A Study of Sulfur Dioxide in
Photoehmicai Smog. Summary Report. American Petroleum Project S-ll,
Battelle Memorial Institute, Columbus, Ohio. 1972.
-32-
-------
DRAFT
DO NOT QUOTE OR CITS
29. Dainten, F.S. and Ivin, K.J., "The Photochemical Formation of
Sulphinic Acids from Sulphur Dioxide and Hydrocarbons", Trans.
Faraday Soc., Vol 46, pp 374-381, 1950.
30. Timmons, R.G., LeFevre, H.F. and Hollinden, G.A., "Reactions
of Sulfur Dioxide of Possible Atmospheric Significance" in Chemical
Reactions in Urban Atmospheres, C.S. Tuesday (ed.), Elsevier Publishing
Company, New York, 1971, pp 159-190.
*
31. Prager, M.J., Stephens, E.R., and Scott, W.E., "Aerosol Formation
from Gaseous Air Pollutants", Ind. and Eng. Chem., Vol 52, pp 521-524,
1960.
32. Wilson, Win. E., Merryman, E.L., Levy, Arthur, and Taliaferro, H.R.,
"Aerosol Formation in Photochemical Smog. I. Effect of Stirring",
Journal of the Air Pollution Control Association, Vol. 21, pp. 128-132,
1971. _________
33. Kopc7vnski, S.I. and Altshuller, A.P., "Photochemical Reactions
of Hydrocarbons with Sulfur Dioxide", Int. J. Air $ Water Poll.,
Vol 6, pp 133-135, 1962.
-33-
-------
DRAFT
DO NOT QUOTE OR CITt
34. Mader, P.P., MacPhee, R.D., Lofberg, R.T., and Larson, G.P.,
"Composition of Organic Portion of Atmospheric Aerosols in the Los
Angeles Area", Ind. Eng. Chem., Vol 44, pp 1352-1355, 1952.
35. Endow, N., Doyle, G.J., and Jones, J.L., "The Nature of Some
Model Photochemical Aerosols", J. Air Poll. Cont. Assoc., Vol 13,
pp 141-147, 1963.
36. Wilson, W.E. Jr., Schwartz, W.E. and Kinzer, G.W., "Haze
Formation - Its Nature and Origin", Coordinating Research Council
and Environmental Protection Agency Joint Project CPA 70-Neg. 172,
Battelle Memorial Institute, Columbus, Ohio, 1972.
~T
'37. Harkins, J. and Nicksic, S.W., "Studies on the Role of Sulfur
Dioxide in Visibility Reduction", J. Air. Poll. Cont. Assoc., Vol 15,
pp 218-221, 1965.
38. Hidy, G.M.> and Brock, J.R., The Dynamics of Aerocolloidal Systems,
New York, Pergamon, 1970.
_q Cadle, R.D., and Robbins, R.C., "Kinetics of Atmospheric Chemical
Reactions Involving Aerosols", Faraday Soc. Disc. 30, pp. 155-161, 1960.
40. Mottershead, C.T., '.'Collision Rates Between Gas Molecules and Aerosol
Particles", in Atmospheric Chemistry and Physics Task Force Assessment-
Project Clean Air, Univ. of California Research Grant AP-00104, APTIC
No. 30845, Sept. 1, 1970, pp. A-1--A-7.
;
_34-
-------
41. Liberti, A. and Devitofraneesco, G., "Evaluation of Sulfur
Compounds in Atmospheric Dust", Proceedings of the Symposium on the
Physicochemical Transformation of Sulfur Compounds in the Atmosphere and
the Formation of Acid Smogs. Organization for Economic Cooperation and
Development, Mainz, Germany, June 1967.
42^-"' "Corn, M., Montgomery, T.L., and Reitz, R.J., "Atmospheric Particula-
lates: Specific Surface Areas and Densities", Science, Vol. 159, pp. 1350
1351, 1968.
43. Corn, M., Montgomery, T.L., and Esmen, N.A., "Suspended Particulate
Matter: Seasonal Variation in Specific Surface Areas and Densities", Envi-
ron. Sci. Technol., Vol. 5, pp. 155-158, 1971. !
44 , Urone, P., Lutsep, H., Hoyes, C.M., and Parcher, J.F., "Static Studies
of Sulfur Dioxide Reactions in Air", Environ. Sciv Technol., Vol. 2,
pp. 611-620, 1968. . : . /:: vl; .';* ' . .1'' V-'"- -W^ / '.' '-'
~45> Jones, W.J., and Ross, R.A., "The Sorption of Sulfur Dioxide on
Silica Gel", J. Chem. Soc. (London) A, Vol. 7, pp. 1021-1026, 1067.
*?- ' ! .
46., Glass, R.W., and Ross, R.A., "Low Surface Coverage Interaction of
Sulfur Dioxide on Selected Transition Metal Catalysts from 273 to 423ฐK",
Can. J. Chem., Vol. 49, pp. 2832-2839, 1971.
47 Pilot, M.J., "Application of Gas-Aerosol Adsorption Data to
the Selection of Air Quality Standards", J. Air Poll. Cont. Assoc.,
Vol 18, pp 751-753, November, 1968.
48~ \Smith, B.M., Wagman, J., and Fish, B.R., "Interaction of Airborne
Particles with Gases", Environ. Sci. Technol., Vol 3, pp 558-562, 1969.
-35-
-------
49, Novakov, T., Mueller, P.K., Alcocer, A.E., and Otvos, J.W.,
"Chemical Composition of Pasadena Aerosol by Particle Size and Time
of Day: III. Chemical States of Nitrogen and Sulfur by Photoelectron
Spectroscopy", AIHL Report No. 119, Air and Industrial Hygiene
Laboratory, California State Department of Publich Health, Berkeley,
California (submitted for publication in J. Coll. and Int. Sci.,
January, 1972).
50. Hulett, L.D., Carlson, T.A., Fish, B.R., and Durham, J.L.,
"Studies of Sulfur Compounds Adsorbed on Smoke Particles and Other
Solids by Photoelectron Spectroscopy in Determination of Air Quality",
Proceedings of the ACS Symposium on Determination of Air Quality held
in Los Angeles, California, April 1-2, 1971, G. Mamantov and
W.D. Shults (eds.), Plenum Press, New York, 1972, pp 179-187.
,'
51 Vannerberg, N. and Sydberger, T., "Reactions Between S02 and
Wet Metal Surfaces", Corros. .Sci., Vol 10, pp 43-49, 1970.
52.. Cheng, R.T., Frohliger, J.O., and Corn, M., "Aerosol-Gas
Interactions", J. Air Poll. Assoc., Vol 21, pp 138-142, March, 1971.
'53. Matteson, M.J., Stober, W., and Luther, H., "Kinetics of the Oxida-r
tion of Sulfur Dioxide by Aerosols of Manganese Sulfate", Ind. Eng. Chem.
Fundam., Vol. 8, pp. 677-687, 1969.
54 Johnstone, H.F., and Williams, G.C., "Absorption of Gases by Liquid
Droplets", Ind. Eng. Chem., Vol. 31, pp. 993-1001, 1939. .
\
55. Junge, C.E., and Ryan, T., "Study of the SC^ Oxidation in Solution
and its Role in Atmospheric Chemistry", Quart. J. Roy, Meteorol, Soc.,
Vol. 84, pp. 46-55, Jan. 1958.
-36-
-------
TJRAFT
PQJQT QUOTE OR CITE
56. Johnstone, H.F., and Coughanowr, D.R., "Absorption of Sulfur Dioxide
from Air. Oxidation in Drops Containing Dissolved Catalysts", Ind. Eng. ..
Chen., Vol. 50, pp. 1169-1172, 1958.
57.; Bracewell, J.M. and Gall, D., "The Catalytic. Oxidation of Sulfur .Di-
oxide ~in Solution at Concentrations Occurring:'in Fog Droplets",. Proceedings
Symposium on the Physico-chemical Transformation of Sulfur Compounds in
the Atmosphere and the Formation of Acid Smogs. Organization for
Economic Cooperation and Development, Mainz, Germany, June 1967.
:58 \ Van Den Heuvel, A^P., and Mason, B.J., "The Formation of Ammonium
Sulfate in Water Droplets Exposed to Gaseous Sulfur Dioxide and Ammonia",
Quart, J. Roy, Meteorol.Soc., Vol. 89, pp. 271-275, April 1963.
. .
59 : Scott, W.D., and Hobbs, P.V., "The Formation of Sulfate in Water
Droplets", J. Atmos. Sci., Vol. 24, pp. 54-57, 1967.
60. i! McKay, H.A.C., "The Atmospheric Oxidation of Sulfur Dioxide in Water
Droplets in Presence of Ammonia", Atmos. Environ., Vol. 5, pp. 7-14, 1971.
61. "Altshuller, A.P., "Atmospheric Sulfur Dioxide and Sulfate"
Distribution of Concentrations at Urban and Nonurban Sites in the,
, United States", submitted for publication in Science, 1972.
-eY^ Junge, C. and Scheich, G., "Determination of the Acid Content
of Aerosol Particles", Atmos. Environ., Vol 5, pp 165-^175, 1971.
63." Atkins, D.H.F., Cox, R.A., and Eggleton, A.E.J., "Photochemical
Ozone and Sulfuric Acid over Southern England", Nature, Vol. 235,
pp. 372-376, 1972.
-37-
-------
DRAFT
DO NOT QUOTE OR CITE
64. Schuetzle, Dennis, A. L. Crittenden, and P. J. Charlson. Appli-
cation of Computer Controlled High Resolution Mass Spectroscopy to the
Analysis of Air Pollutants. Paper Number 72-15, APCA Annual Meeting 1972.
' Middleton, W.E.K., "Vision Through the Atmosphere", University
of Toronto Press, Toronto, 1952, 250 pp.
>
66. Roesler, J.F., Stevenson, H.J.R., and Nader, J.S., "Size Distribution
of Sulfate Aerosols in the Ambient Air", J.Air Pollution Control Assoc.,
Vol. 15, 1965, pp. 576-579.
67. Ludwig, F.L., and Robinson, E., "Size Distribution of Sulfur-Con-
taining Compounds in Urban Aerosols", J. Colloid Sci., Vol. 20, 1965,
pp. 571-584.
/
6g Wagman, J., Lee, R.E., and Axt, C.J., "Influence of Some Atmospheric
Variables on the Concentration'and Particle Size Distribution of Sulfate in
Urban Air", Atmos. Environ., Vol. 1, 1967, pp. 479-498.
59 Junge, C.E., Hanson, C.W. and Manson, J.E., "Stratospheric
Aerosols", J. Meteorol., Vol 18 (1), pp 81-108, 1961.
70 > Lazrus, A.L., Gandr:id, B., and Cadle, R.D., "Chemical Composition
of Air Filtration Samples of the Stratospheric Sulfate Layer",
J. Geophy. Res., Vol 76, pp 8083-8088, 1971.
71 . Rosen, J.M., "Stratospheric Dust and Its Relationship to the
Meteoric Influx", Space Sci. Rev., Vol 9, pp 58-89, 1969.
-38-
-------
DRAFT
ii. SULFATE EMISSIONS D0 NOT QUOTE OR CITE
A. Stationary Sources
1. Emission Data
Very little information is readily available on the sulfate content
of particulate emitted from various stationary sources. That which is
available is sparse and is included as a relatively minor part of
other studies. Furthermore, in many instances, although the analytical
data are valid, there are reservations about the sampling procedures
used and the degree to which the integrity of the particulate has been
maintained. The information which follows is presented as it relates
i
to H?SO., most likely the principal constituent of the total sulfate compounds
and to SO^ and SO, which are the immediate precursors to the formation of
H-SO-, subsequent inorganic and possible organic sulfates, and other
^ 0
sulfur compounds.
Table II-l shows the emission of sulfur without abatement associated
with various source industries as reported by the National Academy of
Engineering.
TABLE II-l
ESTIMATED NATIONWIDE SULFUR EMISSIONS WITHOUT ABATEMENT
Annual (Millions of Tons)
1970 1980 1990
Power plants (coal ง oil) 20 41'1 62-ฐ
Other coal combustions 4.8 4.0 3.1
Other petroleum combustion
Metallic ore smelting 3<4 3*9 4-3
Petroleum refinery 4.0 5.3 7.1
Miscellaneous sources
2.4 4.0 6.5
2.0 2.6 3.4
TOTAL
36.6 60.9 86.4
-39-
-------
DRAFT
00 NOT QUOTE OR CITE
Sulfur, contained as an impurity in fuels, is converted almost
quantitatively (90+%) during the combustion process into S02 and SO,.
About 98% of the effluent sulfur is released as S02 and the remaining,
1 to 2% as 503. Since most combustion effluents contain some water
vapor, the S03 rapidly reacts with the water molecules to form H2S04.
Additional processes which may act to convert SC>2 to 803 are alkali
f21
and ferrous metal sulfate decompositions v * and heterogeneous catalysis
by vanadium pentoxide and nickel oxides which are found in residual fuel
The ASME) reports the H2S04 dew point as approximately 240ฐF for
1 ppm and 270ฐ F for 10 ppm. At these temperatures, the H2S04 ฃ 803 + HJ3
equilibrium is more than 99+% toward H2S04 . The 863 formed will rapidly
convert to H2S04 if cooled and exposed to any water vapor. Walden^ J and
a number of other investigators have shown that generally about 2% of the
emitted sulfur is 803(^804). When this ratio is applied to the combustion
source emission of Table II-l, Table I1-.2 results.
TABLE II
ESTIMATED COMBUSTION SOURCE EMISSIONS OF SO,(H9SOJ WITHOUT ABATEMENT
5 ฃ T-
1970 1980
(ton, annual)
Power plants (coal ฃ oil) 400,000 820,000
Other coal combustion 96,000 80,000
Other petroleum combustion 68,000 78,000
-40-
-------
DRAFT
00 NOT QUOTE OB..CITE
The S03 emissions measured by Walden*- / using Method 6 of the NSPS
in a contact sulfuric acid plant indicate that the 2% relationship may
be valid for sulfuric acid manufacture. Therefore, it is estimated that
the miscellaneous SOj (l^SO^) emissions are at 40,000 tons per year "for 1970.
Very little data on 803 emissions from the remaining classes of sources
are available. Most of the 802 emissions in refineries come from'burning
various fuels and waste streams to produce heat. Therefore, the S03/S02
ratio is assessed at 2% for other combustion sources and 803 emissions
are estimated at 48,000 tons in 1970. The small amount of data on metallic
ore smelting indicates that the SO, levels are near 2% of the 802 levels.
Their contribution is estimated at 80,000 tons annually (1970). This offers
a potential 1970 nationwide emission of approximately 732,000 tons of 803
(H2S04) annually. ;
2. Measurement Techniques
The major portion of the existing analytical methodology for 803
(^804) emissions from stationary sources is based on the dew point of
803 being somewhat higher than the water dew point in the stream. A
typical combustion effluent will have a water dew point of between 100ฐF
and 140ฐF. The dew point for S03 varies from 240ฐF at 1 ppm to 280ฐF at
f4T
20 ppm.v ' The condensation techniques generally cool the effluent gas
to 170 to 200ฐF and collect the aerosol ^804 by impaction or filtration.
Analysis can be by sulfate specific Barium chemistry (colorimetric,
potentiometric, titration with indicator, gravimetric) 802 is not collected
in this technique and thus the two species are readily separated.
-41-
-------
yrCflh I
DO NOT QUOTE OR CITE
Condensation methods depend on the presence of water or water vapor
to form H2SC>4 and are thus restricted to "moist" sample streams .
The second major SC>2 collection technique is absorption which depends
on the solubility of H2S04 in aqueous solution. Environmental Protection
Agency source method 6^')uses 80% isoproponal in water to trap the SO*
from combustion sources. Several investigators have shown that some higher
alcohols inhibit the oxidation of dissolved SC^ to SOj, The combination
of two IPA bubblers followed by a filter is nearly quantiative for SOj.
(-Q-)
The dissolved SC^ is removed by immediate purging with clean air. Gillhanr
indicates that the conversion rate (SC>2 -* 803) is approximately 25%/24
hours. Arthur D. Little was unable to get more than l%/24 hour con-
version in a similar time period. The reason for this difference is not
known.
The greater number of collection schemes proposed and used by
Monsanto Corporation, Shell Research and Development Company, Chemical
Construction Corporation,and EPA methods 6 and 8 are all modifications and
subsets of one or the other of the two basic techniques. The principle
(varient) being that EPA method 6, the Reich test, and the Shell methods
assume SO^ to be in the vapor phase and treat it as a gas during sampling.
EPA NSPS method 8*- ^, the Monsanto method, and the Chemico methods assume
that the SOj is in the condensed form as an aerosol and attempt to
isokinetically sample it as a particulate. The collection schemes vary
from cyclones and filters to midget impingers, midget bubblers, lamp
sulfur absorbers, and condensation coils with sintered glass frets,
Flint t11-1 indicates that 80% IPA filled bubblers have efficiencies of
42-46% and the in line filter at about 18% for essentially quantitative
recovery. Walden has tested the midget impingers and found them
-42-
-------
DRAFT ^
ฃ0 NOT QUOTE OR Cfd
relatively constant at about 90% for the first impinger and 95% maximum
for two in series. The midget bubblers were quite variable between 58%
at 5 liters/min. and 93% at 0.51/min. for two in series. The larger lamp
sulfur absorbers appear to be quite constant at about 95% over the range
(121
of flows. HissinkV ' has indicated that controlled condensation with a
filter for collection is essentially quantitative when the filter tempera-
ture is below 100ฐC. Walden^ ' has shown that the controlled condensation
approach produces substantially lower (slope 0.45) values than the IPA
method in power plants, sulfuric acid plants, and smelting operations.
Where as the S02 values given by the same apparatus give a slope 'of 0.99
and show excellent correlation.
At this time, no other test or verification or comparison data on
any other test method is available for SO or H SO aerosols. ''
-43-
-------
DO NOT QUOTE OR CITE
B. Mobile Sources
1. Emissions
Sulfate from mobile sources has not been considered significant due
to the low levels of sulfur in refined fuels.
From the 1971-2 Bureau of Mines study gasoline contains 0.044 wt%
sulfur for regular grade and 0.039% for premium. This leads to an esti-
mate of 220,000 tons of sulfur per year nationally (as S02) or 0.8% of
the total national emissions.
Locally, there may be a problem in that southern California has
exceptionally high levels. These are the only areas where this is so,
but::the average concentration there is 0.15% in regular and 0.09% in
premium. Assuming the worst case, this could account for about 36,000
tons of sulfur oxides/year as S02.
Sulfur contents of fuels from other sources also are not significant
due to low usage rates and the low levels (2% maximum) in almost all mobile
fuels.
2. Measurement Techniques
Sulfur is normally measured in the fuel by one of several methods:
x-ray ASTM P-2622, lamp ASTM P-1266, and bomb P-129 among others.
Little work has been done on direct sulfate products from mobile sources,
Usually any such analysis has been in conjunction with trace metal analysis
of particulate matter collected on filters. X-ray techniques are most
often used, but occasionally, wet chemicals have been employed. X-ray
fluorescence techniques yield qualitative data in very little time, but
are difficult with leaded fuels due to peak overlaps of lead and sulfur. Lead
13
sulfate compounds have been identified in automotive exhaust particulates.
-44-
-------
DRAFT
00 NOT QUOTE OR CITE
The direct analysis of sulfur on tailpipe emission is usually done
flame photometrically. Extremely high sensitivity is required because
of the very low levels.
-45-
-------
DRAFT
00 NOT OUOTE OR CITE
C.References
1. "Abatement of Sulfur Oxide Emissions from Stationary Combustion
Sources," National Academy of Engineering, COPAC-2.
2. A. D. Thomas, et al, "Applicability of Metal Oxides to the Develop-
ment of New Processes for Removing S02 from Flue Gases," NAPCA
Contract PH 86-68-68, TRACOR Co. (July 1969).
3. H. Juntgen, "Sulfur Balance and Sulfur Trioxide Equilibrium in
Flue Gases," ERDOL KOHLE 16^, 119 (1963).
4. ASME, "Flue and Exhaust Gas Analysis," Report PTC 19.10 (1968).
5. J. Driscoll, et al, "Improved Chemical Methods for Sampling and
Analysis of Gaseous Pollutants from the Combustion of Fossil Fuels,"
APTD 1126, Walden Research Corporation, June 1971 (Contract CPA-22-
69-95).
6. J. Driscoll, et al, "Verification of Improved Chemical Methods for
Sampling and Analysis of Sulfur Oxides Emissions from Stationary
Sources," EPA-R2-72-105, Dec. 1971, Walden Research Corporation
(Contract 68-02-0009).
7. "Method 6 - Determination of Sulfur Dioxide Emissions from Stationary
Sources," Appendix to Part 60, Chapter I, Title 40, Code of Federal
Regulations.
8. "Method 8 - Determination of Sulfuric Acid Mist and Sulfur Dioxide
Emissions from Stationary Sources," Appendix to Part 60, Chapter I,
Title 40, Code of Federal Regulations.
9. E. W. F. Gillham, "The Determination of Oxides of Sulfur in Flue
Gas," J. Soc. Chem. Indus. 65, 370 (1946).
10. "Manual Methods of Sampling and Analysis of Particulate Emissions
from Incinerators," Arthur D. Little Co., Contract EHSD-71-27,
Dec.1972, Appendix H.
11. D. Flint, "A Method for the Determination of Small Concentrations of
S03 in the Presence of Larger Concentrations of S02," J. of Soc. Chem.
Indus. 67, 2 (January 1948).
1.2. M. Hissink, "An Instrument for Determining Sulfur Oxides in Flue Gases,
J. of the Instit of Fuel (London) 36_, 372 (September 1963) .
13. Moran, J. B. and Manary, 0. J., "Effects of Fuel Additives on the
Chemical and Physical Characteristics of Particulate Emissions in
Automotive Exhaust," EPA Contract R2-72-066V .Bdcember 1972.
-46-
-------
DRAFT
DO NOT QUOTE OR C)Tฃ
III. AMBIENT SULFATE CONCENTRATIONS
A. National Air Surveillance Network
The most extensive body of 24-hour integrated sulfate (sulfuric
acid and sulfate salts) measurements collected and analyzed .by a common
procedure has been compiled by the National Air Surveillance Networks
(NASN). Sulfate data have been collected nationwide since 1957 by a
network which has grown to include approximately 250 sites both urban
and non-urban. The operation of the nationwide network depends on the
services of a volunteer operator at each site to collect the sample on
the designated sample date and mail the collected sample back to EPA
for analysis. The schedule for sample collection was one 24-hour integrated
sample collected randomly once every two weeks. Beginning in 1972, samples
were collected every twelfth-day.
1. Spatial Patterns ;
In general the highest concentrations of sulfate in suspended
particulate matter observed from 1957 through 1970 occurred in the;
industrial northeast. The maximum 24-hour average sulfate concentration
observed during this period in over 35 thousand samples was 197 yg/m3
in a particulate sample of 425 yg/m3 at Charleston, West Virginia, on
September 24, 1967. The 1970 nationwide annual average sulfate concentra-
tion at the urban locations was 10.1 yg/m3, and the non-urban average was
7 1 " - - -.._ __ . _ _ . _'
6.3 yg/m . Table III-l presents the urban and non-urban cumulative frequency
distribution of annual sulfate averages. The number of stations included
in these summaries varies from year-to-year depending upon the number
-47-
-------
DRAFT
DO NOT QUOTE OR CITE
Table III-l.
CUMULATIVE FREQUENCY "niSTP.lClJTIOfl OF ANNUAL SULFATE
AVERAGES
YEAR
_j
1957
19r>,3
1959
1 960
1961
1962
1 963
1 9- 22
25
<5.0
, - ^
'40
4-3
50
4H
36
36
uo/m^
-------
DO NOT QUOTE OR CITE
of stations in the network in a given year, the number of stations
collecting the required number of samples, and the number having the
proper distribution of samples throughout the year. For example, Table HI-1
shows that in 1970, 84 percent of the 164 stations (138 stations) had
T
an annual average less than 15 yg/m . The non-urban portion of the
table shows that in every year since 1965, the annual average for non-
3
urban stations was below 15 yg/m , with an annual average for all stations
about 6 yg/m .
2. Seasonal Patterns
Figure III-l is a smoothed plot of the sulfate data showing seasonal urban
and non-urban patterns. The sulfate patterns in the urban atmosphere
prior to 1966 show a high winter -- low summer relationship. However,
the seasonal sulfate pattern after 1965 shows no consistent winter/summer
relationships. The non-urban pattern reveals a slightly higher
summer lower winter cycle.
i
Figure III-2 is a smoothed plot of the sulfate data showing long term
patterns. It appears that there are no marked long term trends in the
sulfate data. This fact can also be seen in Table III-l where the percentage
i
of stations within each cell changes only slightly from year -to-
year, inspite of control techniques which were in general use in the
urban areas since 1967.
-49-
-------
DRAFT
00 NOT QUOTE OR CITE
3. Empirical Relationship to Sulfur Dioxide
Beginning in 1962 ,and continuing to the present ,the National Air
Surveillance Networks have collected simultaneous ambient measurements
of 24-hour sulfur dioxide and suspended sulfate in various cities.
The earlier data (before 1968) demonstrated with a correlation coefficient
above 0.75 that with increasing amounts of S0ป the levels of the suspended
sulfate aerosol increase. However, as the SCL levels diminish and
approach the minimum detectable limit, there are residual amounts of
sulfate present in ambient suspended particulate matter. The more recent
data do not show the same degree of association, which may be attributable
in part to the decrease in SO- concentrations across the country. For
example, correlating individual 24-hour SCL measurements for the NASN
data collected in 1967, 1968, and 1970 resulted in correlation coefficients
above 0.5 in only 35%, 18% and 7% of the sites, respectively.
For comparison purposes Table III-2 is constructed analogously to
Table III-l, which shows the cumulative distribution of annual S02 averages.
This table, by the way of contrast, illustrates the downward shift in S02 .
concentrations. Figures III-3 and III-4 can be compared with Figure III-l and
III-2, respectively. Note that urban S0_ demonstrates a high winter, low
summer pattern, corresponding to the winter heating season and increased
home fuel usage. Seasonality exists at non-urban S02 sites as well as
urban environments, presumably an influence thereof, but the winter to summer
differences are not pronounced.
4. Sample Collection and Analysis
Samples are collected with a high-volume sampler by drawing air
through a glass fiber filter at a rate of from 40 to 60 cubic feet per
-50-
-------
DRAFT
DO NOT QUOTE OR CITE
TABLE III.-2. URBAN CUMULATIVE DISTRIBUTION BY PERCENT
TATAI \ t+y[ '" *ปi ii iuw i **WX nvci UMG j /
YEAR STATIONS < 20 < 40 < 60 <80 <100
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
33
26
33
34
45
47
98
109
97
71
36.4
15,3
12.2
29.4
24.5
21.3
34.7
39.5
54.6
66.2
60.6
30.7
45.5
44.1
48.9
48.9
64.3
73.5
84.5
90.1
69.7
49.9
57.6
64.6
62.2
68.1
76.5
86.4
91.7
98.5
72.7
57.5
69.7
76.4
73.3
74.5
86.7
94.7
96,9
100.
84.8
76.7
100.
100.
77.7
100.
94.8 |
97.4
100.
NONURBAH CUMULATIVE DISTRIBUTION BY PERCENT
YEAR
TOTAL (yg/nr-Annual SO, Averages)
STATIONS <10 <20 <30
1968
1969
1970
1971
5
6
3
9
40.0
50.0
100.
100.
100.
83.0
100.
-51-
-------
DRAFT
DO NOT QUOTE OR CITE
minute. The collected particles are extracted by refluxing with 40 ml of
distilled water in a 125 ml flask of an 8% section of the "exposed"
glass fiber filter. The extract is then cooled and filtered. The
sample and flask are washed with a small amount of water. The filtered
extract and washers are combined and the volume is made up to 50 ml with
water. The 50 ml sample is then mixed and used for the analysis.
The sulfate ion in the filtrate is determined by the methylthymol
blue method with an Auto-Analyzer*. The filtrate is reacted with a reagent
consisting of equal parts of methylthymol blue dye and barium chloride,
kept at a fH of 2.8 to prevent the formation of a chelate complex from
the dye and the barium. Any sulfate ion in the sample reacts with the
barium, leaving an excess of methylthymol blue dye that is proportional
to the amount of sulfate present. The pH is then raised to 12.4 at
which point the barium not removed by the sulfate forms a chelate
complex with the methylthymol blue dye and the excess dye turns yellow.
The intensity of the yellow color is then determined colorimetrically at 480
millimicrons.
-52-
-------
nn
DO NOT QUOTE OR CITE
604 (UG/CU.METER)
Flqure .Hl-1
URBAN AND NONURBAN SEASONAL PATTERNS
15-
1 "*
I LJ-
NONURBAN
13b7 I 1353 ' 13V3 I 1360 ' 138.1
' 13S3 r 1364 ' 13Scj ' 1366 ' 1367 ' 1353 ' 1363 i 1370
YERR i
2S-
20-
15-
1r\
o-
III-2
URBAN AND NONURBAN LONG TERM PATTERNS
S04 (UG/CU.METER)
NONURBAN
1967 ' 1353 ' 1:363
>? ' 13^0 ' 13V3 ' 1380 '
"Y5S2 I 1363 I ^3*5A I 1355
YERR
-53-
-------
300-1
2CO-
100-
S02 (Uu/CU METER )
DRAFT
DO NOT QUOTE OR CITE
19S4 T 1965 l 1966 ' 1967 I J9S8 ' 195S l 1970 ! 1971 '
YERR
Figure iii-3.SEASONAL. PATTERNS OF SULFUR DIOXIDE CONCENTRATIONS
Fiqure in-4.LONG-TERM VARIATION IN NASN S02 CONCENTRATIONS
200-
1OH
YEAR
-54-
-------
DO
'*"' QUOTE OR CITE
B. Review of Measurement Techniques
Since there is no room for improvement in all our analytical methods,
whenever one speaks of analysis one talks about problems. The two main
problems in the determination of atmospheric sulfate reside in collection
and in analysis. Sulfate in terms of total atmospheric sulfate,
sulfuric acid and other various types of sulfates is discussed,
1. Total Sulfate
Collection. Usually on glass fiber sheets and occasionally with
cascade impactors, etc. Difficulty lies in the report that the
concentration of sulfates in air depends on the sampling Volume.
This result has been explained on one hand as due to the catalytic
oxidation of SC>2 to SO* at the^surface of the collecting area
and on the other hand as due to interferences in the analytical
method arising from other substances also extracted from the glass
fibers. Information is not generally available on the specific
conditions required and the extent to which SC^ could be converted to
SO* on filters. ;
Analysis. A large number of methods of analysis for total
atmospheric sulfate are available, A critical study and
comparison of these methods is not available. Some of the
techniques and their problems are shown in Table 111*3-,'
2. Automated Total Sulfur Monitor ;
Particulate sulfur collected on millipore filters can be analyzed
with X-ray fluorescence. The filter is irradiated in the analyzer
with X-rays from an X-ray tube and photons emitted by sulfur are sensed
with a lithium drifted silicon detector or a crystalline grating
spectrometer. Analysis time is presently 30 minutes per sample if a
complete elemental analysis is performed using a lithium drifted
silicon detector, If only sulfur data are needed, the analysis is
performed in 10 minutes. Up to 30 samples can be automatically analyzed
without operator intervention, An automatic sampling station allows
one to make collection intervals as short as two hours if diurnal
variations are needed,
-55-
-------
DRAFT
DO NOT QUOTE OR CITE
Sulfuric Acid
Collection. Usually on glass fiber filters and occassionally
on paper or membrane filters, in impingers, containing aqueous
solutions or dry impingers in cascade centrepetors, or, by
sonic impaction on copper discs or planchets.
Analysis. Table III-4 gives a fairly concise list of most
of the techniques in use to determine H2S04. Obviously with
the separation of H^SO, from other sulfates, the various
methods described in the sulfate section can be used.
A measuring method for strong acid aerosol has been developed by
Brosset of Sweden and Liberti and coworkers of Italy. Aerosol is collected
in water, and then a titration is performed to distinguish between strong
and weak acid. The method is specific to strong acid, but is not able to
distinguish between various strong acids, nor is it very sensitive. A
method more specific for sulfuric acid is the collection of particulate on
i
a hi-vol filter followed by heating the filter after returning to the
laboratory and detecting the SO- which evolves. This method requires
making the rather questionable assumption that the H^SO. does not react
with anything between the time it reaches the filter and the time of
laboratory analysis.
The most reliable procedure for determing acid mist should be the
direct determination in the field in order to minimize reaction losses
which occur between times of collection and analysis. Reliable commercial
instrumentation is not presently available, and for this reason prototypes
are now being developed by the Field Methods Development Section of Chemistry
ซuf Physics Laboratory. One devide under development uses a moving filter
tape for collection followed by a heater for driving off SO, which in
O
turn is sensed by a flame photometric detector. The
measuring concept is identidal to the laboratory procedure described
-56-
-------
DRAFT
DO NOT QUOTE OR CITE.
above, but reaction losses are minimized by the short (five
minutes) time period between collection and analysis. A second
technique uses a gas phase titration with ammonia to determine
strong acids. The ammonia is measured with a chemiluminescent analyzer.
Qualitative response to H2S04 aerosol has been obtained. Research
leading toward prototype development is being pursued. The method
sensitive only to strong acids, but it does not distinguish between
H2S04 and HNO ,
is
3. Miscellaneous Sulfates
With collection on glass fiber paper and controlled vaporization
of the sample into a computer-controlled, high resolution mass
spectrometer, Compounds such as H2S04> NH4 HS04, (NH4)2 S04 and
NaHSO. have been characterized and assayed. Because of the
temperatures involved in the vaporization the integrity of
the sample is the big problem. Numbers will be readily
obtained, but their accuracy might be a real problem.
.4. General Difficulties
Sulfate
Collection of sulfate from all types of polluted air
without formation of artifact sulfate during collection.
Water-insoluble sulfates not extracted in analytical procedure.
Problem in all collection methods is that the integrity of
the sample must be proven and this integrity must have a
reasonable shelf life.
Problem of field monitors is their expense, maintenance, rapid
instrument obsolescence and field calibration. This problem
is considerably; enhanced with monopollutant analyzers.
H2S04
In collection on glass fiber filters, these fibers may
catalyze the oxidation of SCL to S0_ and can also
neutralize the H-SCK at the surface of the glass fiber sheet.
-57-
-------
nn DRAFT
DO NOT QUOTE OR CITE
In collection of particulate H-SO. in aqueous solutions
and in extraction of particulate H2SO. with water, it
is possible that dissolving basic particles will neutralize
some of the H^SO^. Thus, an apparent F^SCfy is obtained and
not a "true" H2S34. However, it is argued by some that
the apparent H2S04 more closely approximates the effective
H2S04 than does the "true" H2S04. This problem in
collection would be even more serious in monitoring methods
consisting of short-time collection and elution, where the
precision would vary over a wide range dependent on the
amount of.soluble basic material caught with each batch of
H2SCV
In any methods involving heat, artifact formation could
ensue-. Thus, ammonium sulfate would be measured as F^SC^.
Other compounds, such as organic sulfates, sulfonates,
sulfites and sulfones could interfere in this way dependent
on the conditions and the method used.
-58-
-------
DRAFT
Table ni-3 NOT QUOTE OR <3T
TECHNIQUES FOR DETERMINATION OF ATMOSPHERIC SULFATE:
I. Turbidimetry and Nephelometry
A. Methods
1. Ba salt
2. 4-Amino-4'-chlorobiphenyl
3; 2-Aminopyrimidine
B. Advantages
1. Simplicity
C. Disadvantages
1. Possible interference from colloidal organic matter, filter
media and colored material
2. Coprecipitation problem
3. Poor reproducability
II. Colorimetry with Color Decrease
A. Methods i
1. Barium rhodizonate
2. Morin or hydroxyflavone and thorium
B. -
C. Disadvantages
1. Difficulties in precision with a subtractive method
2. Lack of sensitivity
III. Colorimetry
A. Methods
1. Barium chloranilate
a. Cation interference
b. Centrifugation or filtration needed
c. Phosphate, oxalate and bisulfite interfere
d. Thought by some to be one of the best methods
2. Methylthymol blue
a. Interfering heavy metals have to be removed by ion-exchange
chromatograpfcy
b. Filtration of barium sulfate necessary
c. Range 0.3 to 45 yg SC^'/m3 f air
-59-
-------
DRAFT
Table III-S Continued 00 NOT QUOTE OR CITE
d. Interference by oxidation of dye in alkaline solution.
e. Precision at lower concentrations needs more study
since this appears to be a possible weakness
f . Colored blank interferes
3. Reduction to H^S and reaction with N,N-dimethyl-p-phenylenediamine
to give methylene blue
a. Complicated procedure
4. Reduction to f^S and reaction with ferric ion and o-phenanthroline
to give the ferrous o-phenanthroline complex
a. Complicated procedure
b. Nitrite and other H2S precursors interfere,
IV. Quenchofluorimetry
A. Methods
1. Morin or hydroxyflavone with thorium
B. Advantages ;
1. Simplicity |
2. Sensitivity potential high
C. Disadvantages
1. The difficult problem of measuring a decrease in fluorescence
V. Fluorimetry .
A. Methods
1. Reduction to hydrogen sulfide and reaction with a
non-fluorescent mercury derivative of a powerful
f luorogen to give that fluorogen
B. Advantages
1. High order of sensitivity
C. Disadvantages
1. Complexity of procedure
VI. Ring Oven
A. Methods
1. Barium chloride'-- potassium permanganate
-60-
-------
Table ItI-3 Continued
B. Advantages
1. Simplicity
C. Disadvantages
1. Precision depends on technique, and eye of the experiment's
2. Interferences are phosphate, oxalate, sulflte
VII. Polarography
A. Methods
1. Indirect with lead or barium nitrate
a. Filtration or centrifugation necessary with consequent
problems of adsorption and poorer precision
2. Direct square-wave polarography
a. Ion exchange chromatography necessary to remove
interferences
VIII. Atomic Absorption
A. Methods .
1. Barium
2. Lead
3. Calcium
B. Advantages
1. The measurement sensitivities of the metals used to
precipitate sulfate are of value in this method
C. Disadvantages
1. All the usual problems of indirect methods and especially
those involving precipitation with excess metal salt
followed by subsequent measurement of residual cation
2. Usual interferences arid especially the strong interference
from phosphate in the calcium method.
IX. Gas Chromatography
A. Methods
1. Reduction to H_S and measurement of the H2S by gas
chromatography with flame photometric detectors.
-61-
-------
DRAFT
00 NOT QUOTE OR CITE
Table III-3 Continued
X. Potentiometry and especially Selective Ion Electrodes
A. Methods
1. Direct selective sulfate detector
a. I know of no satisfactory detector of this type.
However, the field is advancing rapidly so a
satisfactory electrode could be developed soon.
2. Indirect ion selective electrode for sulfate
a. Lead selective electrode
1. Most potential at moment
2. Method involves titration with a lead salt
3. Interferences would be any other anions that could
precipitate lead. For the method to be workable
the effect of these interfering anions would have
to be cancelled.
4. Reasonable sensitivity
XI. Flame Photometry
A. Methods
1. Similar to IX but essentially aspiration of sulfate
solution into a hydrogen-rich flame and measuring the emission
B. High Sensitivity
C. Disadvantage
1. Other sulfur compounds would contribute to the total sulfate
X-ray Fluorescence
A. Methods
1. Essentially another laboratory method to measure sulfur.
If all the sulfur were present as sulfate, this might be a
good way to measure it.
B. Disadvantages
1. Other sulfur compounds would be measured
2. Expensive instrumentation not available to most labs
C. Advantages
1. Fast method of analysis. Because of sophisticated
instrumentation a good method for laboratory collection
in the field and analysis in a central laboratory
-62-
-------
DRAM
DO NOT QUOTE OR ClTI
Table ;IiII-4
Determination of Atmospheric H^SO.
I. Colorimetry
A. Methods
1. Vanadate
a. Simple procedure
b. Sensitive to 5 yg H2SO.
c. Average error 9.3%
2. Barium chloranilate
i
a. H-SO. separated from other sulfates by elution with isopropanol
b. Filtration necessary
c. Colored material in the extract interferes, as does any
anion which precipitates barium
3. West-Gaecke method
a. Procedure consists of evaporation of H^SO., reduction with
copper and determination of S02 colorimetrically
b. Complex procedure
c. Ammonium sulfate is also measured
4. Other colorimetric and fluorimetric methods described in the
sulfate section could be used once HLSO. is separated from
other sulfates
II. Coulometry
A. Methods
1. As in IA3 but S02 titrated
III. Flame Photometry: As above but SCL measured with flame photometer
IV. Titrimetry
A. Methods
1. Sodium hydroxide residual titration method
2. Tetraborate residual titration method
3. Separation of H2SO. by diffusion at elevated temperatures and
titration with barium perchlorate using thoron I as indicator
V. pH Method: Especially+measurement of pH and correlation of this
with the H ion concentration
VI. Square-wave Polarography: Volatile H2SO. separated by diffusion and
determined. Method more rapid than titration.
-63-
-------
C. References
DRAFT
DO NOT QUOTE OR CITE
1. National Aerometric Data Bank. 1957-1970, Research Triangle
Park, N. C.
2. Lee, R. E., .Jr. and J. Wagman. ''A Sampling Anomaly in the
Determination of Atmospheric Sulfate Concentration", Am. Ind.
Hyg. Ass. J. 27: 266-271. 1966.
3. Wagman, Jr., R. E. Lee, Jr. and C. J. Axt. 1967, "Influence of
Some Atmospheric Variables on the Concentration and Particle
Size Distribution of Sulfate in Urban Air", Atmos. Environ.
1: 479-489, 1967.
4. Brosset, C., "Particle-borne Strong Acid; Occurrence, Effects
and Determination Methods". Presented to the Division of Water,
Air and Waste Chemistry. Amer. Chem. Soc., New York, August, 1972.
5. Liberti, A., M. Possazini and M. Vicedomini, "Determination of
Non-volatile Acidity of Rainwater by a Coulometric Procedure."
Analyst. 97: 352-356, 1972.
-64-
-------
DO NOT QUOTE OR CITE
IV. EFFECTS ON HUMAN HEALTH
A. Toxicological Appraisal
1. Current Knowledge
Chemical studies of polluted air have demonstrated that sulfur
dioxide undergoes oxidation leading ultimately to sulfuric acid and
particulate sulfates. Experimental biological studies have indicated
that certain of the particulate sulfates have greater biological effects
than sulfur dioxide.
Epidemiological evidence is now emerging that suggests that human
respiratory disease also associates better with sulfates than with
sulfur dioxide.
Toxicologic studies have further demonstrated that the incorpora-
tion of particles such as sodium chloride in an atmosphere of S02;
enhance the biologic response. It could be postulated that this effect
is accomplished through accelerated conversion of SO to sulfate.
Furthermore, it has been determined that there are striking differences
(as much as 20 fold or more) in the degree of biological activity among
the various sulfate compounds and between different particle sizes of
1
the same compound.
2. Deficiencies in Present Information
Despite the foregoing, there are serious gaps in the toxicological
information concerning sulfates. Perhaps the most important problem
with this information is its narrowness in that (1) it stems mainly
from one laboratory -- Harvard School of Public Health; C2) it employs
chiefly one animal species -- the guinea pig and (3) it uses only acute
-65-
-------
DRAFT
00 NOT QUOTE OR CITE
exposures and essentially one parameter of effect, pulmonary function.
While no fault can be found with this body of evidence so far as it
goes, it would have much greater impact from the standpoint of quality
criteria development if confirmation from several independent contributors
were involved. Functional data from the guinea pig may mimic the
hypersensitive person (the asthmatic). However, this fact cannot be
solidly determined until there is more information from similar
studies on other animal species. The parameter employed to
evaluate the SO^/sulfate reaction in animals has, with few exceptions,
been the Amdur/Mead method which employs an indwelling pleural catheter
to measure pulmonary flow resistance. It is possible that this
method heightened the acute response to pulmonary irritations.
Recent experiments carried out by Frank and Mcyilton^
indicate the importance of relative humidity in the response of
animals to the SO /sodium chloride aerosol atmosphere. Guinea
pigs were exposed for 1-hour intervals to atmospheres at two
modes of relative humidity"low" 40% and "high" 80%. Atmospheres
were S02 (2620 yg/m3) alone-, sodium chloride aerosol (900-1000 vg/m3)
alone, and combined sodium chloride and S02. Significant changes
in pulmonary flow resistance occurred only in the combined SOo/sodium
chloride aerosol atmosphere at "high" relative humidity. In these
studies, the concentration of gaseous S02 diminished to almost
zero in the combined atmosphere of S02/sodium chloride aerosol
at the "high" relative humidity in contrast to only a 10% reduction
in the same combination at the "low" relative humidity. The authors
-66-
-------
DRAFT
DO NOT QUOTE OR CITE
suggest that the hydration of the particles and the subsequent
uptake of SO proceeds rapidly at "high" relative humidity. These
findings tend to confirm those reported earlier by Amdur concerning
the importance of the particle in the pulmonary response to S02, and
indicate that in the system "high" relative humidity is a prime
requisite for this biological response.
Perhaps the most important deficiency in experimental studies
of sulfates is the paucity of information concerning the role of sulfate
(or even 862) in the production of, or acerbation of chronic pulmonary
disease. While there is an array of epidemiological data to support
such effect, no definitive experimental studies have been reported
which attack this problem except on the most simplistic and non-realisti<
basis. The etiology of such disease is undoubtedly complex, being
probably involved with primary and secondary factors including the
sulfates, smoking, temperature or humidity variations, and the presence
of biological agents such as bacteria and filterable viruses. No
studies of sufficient sophistication to approach this problem have
ever been undertaken.
An additional problem with the toxicological studies is in respect
to their relevance to natural polluted atmospheres. It was pointed out
above that the various sulfates and even various particle sizes in
the same sulfate engender great differences- in response when inhaled
by animals. While this is of great biological interest, there appears
to be so little known regarding the nature of the sulfates in the
atmosphere that the biological information cannot be related to ambient
air.
-67-
-------
DRAFT
QUOTE OR CITE
Finally, while the conversion of S02 and other sulfur-containing"
gases to sulfate is a dynamic phenomenon associates with water vapor,
discrete particles, oxidant, photochemical energy and the like, animal
experimentation has been performed largely on static simplistic gaseous
or particulate atmospheres with no great potential for interaction,
Thus, formation of atmospheric sulfate in urban atmospheres, with
the possible production of reactive complexes and the like, has not
been mimicked in these exposures. Furthermore, the presence of
other atmospheric substances whose interaction potential is at this
time unknown is precluded,
A study was conducted by Hazleton Laboratories on chronic
exposure of Cynomolgus monkeys to sulfuric acid mist and fly ash
2
mixtures. This investigation was part of a long line of research
into the health effects of long-terra exposure of monkeys and guinea
pigs to SCU, H2SO., and fly ash alone and in various binary and
tertiary combinations. The study was comprehensive, including many
pulmonary physiologic, hematologic, biochemical and histopathologic
studies. Three groups of eight Cynomolgus monkeys were exposed to
either (1) 0.99 mg/m3 H2S04 mist plus 0.53 mg/m3 fly ash; or (2) 0.11
3 3
mg/m H2S04 mist plus 0.53 mg/m fly ash; or (3) filtered room air
(controls) . The particle size of the h^SC^ mist was <;1 micron and
fly ash was <5. The fly ash size range was about 2.5 - 3.5 mass
median diameter (HMD). The monkeys were exposed essentially
continuously for 78 weeks with certain tests being conducted on the
animals at no more frequently than weekly intervals.
-68-
-------
DRAFT
DO NOT QUOTE OR CITt
The significant findings of this study were histologic alterations
in the epithelium of the bronchi and bronchioles of the lung characterized
as focal erosions and epithelial hypertrophy and hyperplasia in
monkeys exposed to 0,99 mg/m H SO^ mist and 0.53 mg/m3 fly ash,
These lesions are chronic inflammatory in nature and are adverse health
effects . No abnormal histologic effects were observed in the
3 3
monkeys exposed to 0.11 mg/m H2SO^ mist and 0,53 mg/m fly ash or in
controls.
A small but progressive increase in pulmonary airflow resistance
was observed in monkeys exposed to both levels of pollutants, The
investigators minimized the significance of this effect apparently
because the control values for the monkeys exposed to pollutants were
less than control values for monkeys exposed to normal, filtered
air. The increase in pulmonary air flow resistance in pollutant-exposed
animals (about 25% increase) raised their values to those of the
control monkeys at the beginning of the experiment, The reasons for
these differences are not easily explained. The increase in pulmonary
airflow resistance might be related to the histologic lesions observed
in the lungs of monkeys exposed to the high f^SO levels. This
interpretation of the physiologic observation is debatable, however,
and should not detract anyone from the principal observation of
histopathologic lesions in the lung.
The health effects reported in the Hazleton study at concentrations
of 0.99 mg/m places it among the few experimental studies that have
reported adverse effects at this relatively low level, The results
corroborate other studies which indicates that fr^SO^ mist of small
size is more toxic than sulfur dioxide.
-69-
-------
DRAFT
DO NOT QUOTE OR CITE
B. Epidemiclogical Appraisal
Recent reports emanating from epidemiological studies carried out
as part of the Community Health and Environmental Surveillance System
(CHESS) of EPA indicate that adverse health effects may be more closely
associated with exposure to suspended atmospheric sulfate than to
other pollutants. The levels of suspended sulfates necessary to
cause adverse health effects were lower than the levels of sulfur
dioxide or total suspended particulates. The findings from the
following studies contained in that report support this statement.
1. New York Asthma
A study of asthmatics carried out in the Metropolitan New York
area showed that no pollutants produce a measureable effect on
asthma attack rates when temperatures are below freezing. However,
when temperatures rose to 3Q-50ฐF dose related increments in
asthma attack rates were associated with increments in total suspended
particulates and suspended sulfates but not sulfur dioxide. The
threshold level for morbidity excess occurred at sulfate levels
of 7.3 pg/m when minimum temperature exceeded 50ฐF while excess
morbidity threshold increased to 12 yg/m^ sulfate when minimum temperatures
were -2 to 10ฐC (30 to 50ฐF). The threshold for total suspended
particulates in the -2 to 10ฐC (30 to 50ฐF) range was 56 yg/m3. These
findings are shown in Table 1 and Figure 1 and 2. The methods used for
measuring suspended sulfates were described in the introduction of
this report under CHESS-methods for measurement. No firm evidence
could be found to associate elevations in sulfur dioxide (100-180 yg/m )
-70-
-------
DRAFT
00 NOT OUOTF OR CITE
TABLE 1
Temperature Specific Threshold Estimates for the Effect of Selected
Pollutants on Asthma Attack Rates3
Pollutant
Total Suspended
Particulates
Suspended
Sul fates
Minimum
Daily
Temperature
Of
30-50ฐ
30-50ฐ
>50ฐ
Intercept
(Attack
Rate Per
Person)
.237
.240
.245
Estimated
Effects
Threshold
yg/m3
56
11.9
7.3
Slope
.000254
.000371
.005009
Estimated Percent
Excess Risk At
Specified 24-Hour
Exposures
22% at 260 yg/m3
4% at ,35 yg/m3
57% at 35 yg/m3
threshold function could not be calculated for sulfur dioxide or other temperature
ranges involving the tabled pollutants.
-71-
-------
iv:-:-:li::::;::::vif::!i:
!:i:.::.: I:: .;:..:::;:::, :
:i:;-:i:::;!.;..;.!;;i.::i;.::;: '
,'!'.!'111*!1*!.!' . i''. t !'. lit' ' '' '*
liliHJjiuiisiiyiijlililiy
" '
fiH
Daily
ktt^hJ!?*LJ8!l!:nin;H^
ฃ1:!!!!;;!;!:;!;;^^^
^&m^^m
":'''*'
it-fere: Titi
iliilillliiElS
iii^iSliiiyilsiiiiJiiiliiiiiililiiil!!^^
::'.'\: ;l! tj. I';:!:,; 1.1! i| t::!
;!)-i;;::i:;;:;;;;^
-------
^;?^^;::;:i;^;:;::;:;jJLj::::-;:;\^:;:^i^:: ii:::!v:i:;!::;:!:!:;:!i:;. without high sulfate----------
ilu.O-:!:;!;;^
.
!n;:'^;.:ฃ;;:;Kn:pn;n;;;;::;!l;;T^:^
" -- ' -
h-.T.:-: :- ;:ri::-:~:r :-:-;:rr? T-J-rTt:-:::""rv-rrrnrv::::!.;;-::;:>:::::::::"-:::! r:i:::n::r:r:-::;- ,::.-;::::* v:U=
:i|il;iiH^:ji;ii}iii
::;!:i!ฃililiiili liiiilli!]:^!:
r:Tfo
w
:|::-:t: ; i:;':r;
i:;:::!:-.:
::::!:.:!!.!:.:!;::;:::;:::::::'-'::!::i !::5'!:!.I-::::!:;.:;:list:!::::!':::..:::;.!:: ::.::::.:::;: u:;::::
:--:..T:73T7..;..:r::r..rri:j:,.;r";j.:::rr:r:i. :j;|| u:;: < :;i;;;:i:Ti I':;"!'-:. / ':;; ;.../. .:. ;.:!. * . " "~:
::;::r:;:i:ri::,;: '::^':;*|::iMrv'!:"";:',l|"|'!;:;;|r;t:!":'|:v'';-::;::,:|;;:/;;:;.:';/^:;i: "'^.":i";':'""!'.::;:""
^
iS!;iฅ!H;ii=-f;li!i!
llliiiiiiliilHlSiiliiiP
Liilullijiilijiiiiliiglu;!!
t!-!.!-!l.!>i^^ป-M.'l!MH
: ::. :::::!::
._..,.. rซ. t.-, . , ..._..ซ,., . .. . ... . , _ .. .k-*~->ซ* ซ .ซ**. i >._ป. .*.,.,,,-. . . . . i . < , , i k..,T,t.)..*,..-.._,..
:-:i::::::-:r:::::h::;::: :::::-:::;::::i::;::...::::.::;.:::(;:!:.::; :il:ri::!:::.:::::;:::
Iii::;iil;::HilL:::::::i.:LLr^ii;1!!'':!1'':!!1 '.^"'.'^'.''''i\-'-"i :::!::s!!:::;!' j^:L!Liv
:/;
;;:::::::':::: ;:::.!: :;:::: ::::: i;:!'::::i:;::!:v: ;::::'; :;';::!!:; ':!)! ;!!!' ::'!:i:i' :::: ::.:'' ::;i ::,'."'.: :::;!::: ::;.:;::: :.'J| :;:.:: :i:' :.::.:::..
-- !-::; -~i:
.
y--:-'-.-;;:!
Daily"
Asthma
Attack Rate li-ii-i;:
tr'"~::r::T:::;-:7i .-.-.rr rrrr:;:::77r;-r:-;r::7;:--:;i:-::v";r//:.;;:"-1:v:tf::-:i-;::"7-::::; 171;:::::::-|:;p;7::::-: "i::::!:;::'.:::;.:.-:.:
.
I ;
Hi
'.! 111 ;:::;:: I ':
..i....M
:.!:..l..!.: .:::::. .
.:!; ..;::; <::..I::.;::;- ,
.: ..i.::.:::;!;:::::::
:.-:-'--\~:\~:'-:~:^f:ti
M=
ป.,..
ซ
i.:.
i-f
I :'::' .--f
-73-
-------
DRAFT
DO NOT QUOTE OR CITE
with excessive asthma attack rates on either cold or warm days. A
linear regression of asthma attack rate on temperature showed that
temperature alone had little effect on asthma attack rates.
2. Utah Asthma
Results of a study of asthmatics in the Salt Lake Basin showed the
highest morbidity rates were associated with the highest suspended
sulfate levels. However, the most consistently linked pollutant was
total suspended particulates and the least impressive effect dould be
attributed to sulfur dioxide exposure. The threshold for aggravation of
asthma by suspended sulfates was 1.4 yg/m^ on warm days minimum temperature
10 ฐC (Tmin = 50 ฐF) and 17.4 yg/m on cooler days T . = minus 2 to
10ฐC (Tmin = 30 to 50 ฐF) . Increased asthmatic attacks could be related
to 24 hour exposures to modestly elevated levels of suspended particulate
matter (71 yg/m ) during warm weather T . = 10ฐC (Tmin = 50 ?F)
and higher elevations (107 yg/m ) during cooler weather T _^n = minus 2
to 10ฐC (T . =30 to 50ฐF) . These results are shown in Table
mm
2 and Figure 3 and 4 .
3. New York Cardio Pulmonary
In a study of cardio pulmonary patients in the New York metropolitan
area, the strongest and most consistent pollutant effects were found for
suspended sulfates which were linked to a worsening of symptoms,
particularly shortness of breath, cough and increased production of
phlegm. The worsening of symptoms which was attributed to suspended
sulfates always persisted after the correction for the effects
attributable to temperature and other pollutants. Total suspended
particulates were also correlated with aggravation of symptoms. There
was good evidence that annual average suspended sulfate levels of
-74-
-------
TABLE 2
DO NOT QUOTE OR CITE
Temperature Specific Threshold Estimates for the Effect of Selected
Pollutants on Asthma Attack Rates
Pollutant
Sulfur Dioxide
Total Suspended
Particulates
Without High
Sul fates
Total Suspended
Particulates
With High
Sul fates
Suspended
Sul fates
Minimum
. Dai ly
Temperature
Of
>40ฐ .
>50ฐ
30-50ฐ
>50ฐ
30-50ฐ
>50ฐ
30-50ฐ
>50ฐ
Intercept
(Attack
Rate per
Person)
.12
.08
.148
- .098
.171
.110
.160
.098
Estimated
Effects
Threshold
yg/m3
53
23
107
71
26
26
17.4
1.4
Slope
.000198
.000595
.001141
.000882
.000440
.000686
.005500
.003274
Estimated Percent
Excess Risk
Presently Allowable
Once A Year Levels
51*
254%
118%
170%
60%
146%
9%a
62%a
Since no standard exists for suspended sulfates, the level of 20 yg/m which is not
found on 5-10% of days in urban areas was arbitrarily chosen to illustrate the
excess risk.
-75-
-------
T-::
DO NOT QUOTE 0ซ
!i!!;lipp:::-I;!I:i;i!!j!l!!Ji:TT:T!F=.1... JISURE3 -ii:ii!i .jr^'PiliillilljjTililirTiiiiiiiiiiijjij
Effect of Minimum Daily Toin;ปeraturo and i;.:-jn uncled Sul fates on
i;. H . n " _'_ Daily Ar.tiiina Attnck _U.!;:^ __ ' '^^
sliiiMiiiiilEiSniliHiliJilllySiiiiliS^
:"n::i;;i;T;n;;n:T;;:^i;;;!i^sn;;i;:!:!ii;ir^
-~---~
L:^L::li!l;:::'::!.\iili..! :::i:::L:::::-;;J'..|liI:l:!.:i:'J^.:LL!::l:l:l;i:!!iu!li:iii.::;;i::ilj;i:'::l.l:.i:'.:'...!;;:!.:';.:-!'.' ;::iLH:;:;i;Li::::;::!::!i'^iiLlii:::;
;::::: ::i::::.-:..j:;.:i:;: j ;:::::.;:;::::i,;!:;:::::!: ;!-Xj!;;^!:i.ij;-i i:f::!!":!i'!:i^:i|;i':':!::::- i-':'.:.}" ":!;::iI:!i^i^ii^-r4r-7'Tr^^:* ':::;'-'::!
m
::-:r:.:;:.::i. ::, Suspended SlllfatCS ' iii'iiHiijiMiir-'i^-iiiili
- ! _ . it* !' I- : I
7,
~76~
-------
pnT:":fi:ni~7:r:T^Ti-?[-^nTi-j-n;nn! FIGURE 4 !"[":[^F^nn^nr-rrr
|................I. ...I ..}ซ...) I ..ซ, I . | .,, , . . . M, itil- I .... I t .1,1. ป!.!|. I., I!,
i...::^::.;::-;:::::;:::::!::: i:.:!.- 1... :::..::i::|.':!li!:!|..;,|: ..,;:..,.:..|.|-.i'jn.'Hin:M-:i:!":'l"-;!::::! '..- ::;-..l:>::!::.:;;::^.i..
::::J:i:::j Effect of Total Suspended Particulates With and Without a High Sulfatc
iTF:?::":::::::.' ::!:-..!:i':r:""'"":r:-'i:i |::'il ..t...!!. I.ซMI.; ..,. i (if ป ' < nit-' ':
jiiHi:Ji:i:.:;;: '.::|::;:::; Content on Asthma Attacks Occurring Durinq Warm Days |!::;ii;;
::::;::-: ::::: "-i i " ' :!'-'--- =- ::
DarHy
Asthma
Attack Rate
Per'TOO
,.. ...,
'
.;:: :.....;,
wffis
::::;:_::t:|:j::i.
i:; *.'','.''. .*;;.'.: i'
^M^3^^^:-^]]l[
iiijiiiiiijr;-;:;;
Total Suspended
L With High Sulfato"Cbrit>-
: r-'rr-::i' : ..u.ii., : . i , ...
Without High Sulfato Conic
-77-
-------
DRAFT
DO NOT QUOTE OR CITE
10rl2 yg/mฐ are accompanied by morbidity excess which averages about
6% when temperatures are minus 2 to 10ฐC (30 to 50ฐF) and 32% with
temperatures of >10ฐC (>50ฐF).
4. Cincinnati School Children
A study of school children in Cincinnati found that ventilatory
function of children residing in a moderately polluted industrial
valley of Cincinnati was significantly lower than the performance
of children living in a clean area of the metropolitan region.
Differences in suspended sulfate levels were found to he most
closely associated with differences in ventilatory performance.
5. Implications and Limitations of Findings
Although the preceding findings incriminate suspended sulfates
as a factor in the exacerbation of respiratory symptoms there remain
many unanswered questions which prevent the establishment of
environmental quality standards. The control of air pollution requires
more precise estimates of response thresholds and dose response
relationships than are possible from the temperature specific
attack rates used in these analyses. The studies cited were
restricted to a narrow range of pollutant levels and in some
situations focused upon cooler seasons thus making threshold
estimates for warmer days difficult,
Laboratory studies have shown that in terms of comparative toxicity
sulfuric acid and some metallic sulfate compounds such as zinc ammonium
sulfate are more potent irritants than sulfur dioxide,3 Data from
laboratory experiments on guinea pigs indicate that the irritant
actions of certain particulate sulfates are greater than the irritating
effect of S02; and a much smaller quantity of sulfur delivered to
the respiratory system as an irritating particulate will produce.
-78-
-------
UIWI
DO NOT QUOTE OR CITE
an increase in flow resistance. Studies have also shown that the
size of the irritant particulate material affects the degree of
response. This was demonstrated for sulfuric acid and for zinc
ammonium sulfate. Limited data have shown that within the range
of 0.3 to 2.5 MMD the smaller the particle-size the greater the
irritant potency. In considering an evaluation of the irritant
effect, data on concentration alone are not sufficient but
information on particle size must be considered as well.
Recent studies in Budapest on the size distribution of water
soluble particles in the atmosphere found that sulfate particles
were larger in the winter than in the summer. This finding raises
the question whether the aggravation of respiratory symptoms found
in the CHESS studies was a function of temperature with suspended
sulfates or due t,qป the smaller size of the sulfate particle at
' V
higher temperatures.
Relative humidity has been shown to influence the irritant potency
of H2S04 and sulfates,5 The CHESS studies cited in this report failed
to take relative humidity into consideration in the data analyses.
However, subsequent CHESS studies will include information on both
relative humidity and particle size.
In several of the CHESS studies total suspended particulates along
with suspended sulfates were linked more often as irritants than was
sulfur dioxide. This raises the question of the possible significance
of total suspended particulates in the formation of irritating
metallic suspended sulfates.
-79-
-------
DO NOT QUOTE OR CITE
We have raised a series of unstated questions and issues which
must be addressed. Are all sulfates equally biologically reactive?
Are sulfates reactive because of the chemical properties associated
with specific compounds, or because of physical properties such as
particle size, solubility or pH(? Are sulfates equally reactive
in humid and dry air at warm and cold temperatures? These biological
issues must be addressed and satisfactorily resolved. In addition
to replicating previous studies new research activities will be
initiated to attempt to answer these questions.
6. Model for Determination Biological Effects of Sulfates and
Related Compounds
Biological experimentation indicates that the presence of a
particulate (NaCl crystals) and high relative humidity are
necessary to show biological effects of exposure of guinea pigs to
S02. It is postulated that such conditions produce conversion of
SC>2 to 803 and H2S04 and also possibly lead to deeper penetration
in the lung. Metallic sulfates when administered as aerosols
appear to exert greater effect than SC^ and different degrees of
effect according to species of sulfate and particle size.
Recent preliminary epidemiological correlations indicate that
human health effects (chronic respiratory disease) associate more
strongly with ".Sulfate" than with S02. The uncertain!ty here
concerns the nature of the sulfate since present methods of
collection on fiber glass filters might result in artifactual
conversion of SC^ to sulfate or to subsequent loss of ;$ulfate
due to decay of such compounds with time on the high volume filters.
Furthermore no information is available concerning the cation
-80-
-------
DRAFT
DO NOT QUOTE OR CITE
identification in the "sulfate" compounds.
Because of the above facts and their difficulty in interpretation
the following animal exposure model is proposed for the purpose of
defining basic biological effects based on emerging chemical
evidence to provide a basis for the design of epidemiological
studies for definite health effect information on human beings,
A. Construction of a chemical reaction chamber to provide
exposure atmospheres for animal experimentation.
1. This will provide S02 or SC>2 products, i.e. SOj, ^804
or metallic sulfates according to the reactants
introduced and the reaction time permitted prior to
exposure of animals,
2. According to reactants introduced, atmospheres containing
oxidant, NCL, nitrates, 02, HNOj will be combined with
the SOX atmospheres.
3. Use of this chamber permits the possibility of exposure
of animals in atmospheres which are undergoing dynamic
change or conversion simulating polluted ambient
atmospheres.
B, The following sequences of work will be carried out on.
animals.
1. Acute exposure of guinea pigs, Measurement of pulmonary
flow resistance and other pulmonary physiologic
parameters.
-81-
-------
DRAFT
DO A/or
' """""Off
(a) This method will be used for a first effect
screen to evaluate the influence of many variables,
i.e. (a) reaction-time sequence and reactive
mixture, (b) influence of relative and absolute
humidity, temperature, (c) effect of various
catalysts, (d) effect of specific sulfates and the
like.
The reactive and specific chemicals will be chosen on the basis
of data from on-going studies of ambient air so that comparative
toxicities could be determined on various species of chemicals, of .
relevance to ambient pollution,
C. When the comparative toxicities of the various sulfate sub-
stances present in ambient air are determined more definitively,
longer term exposures would be undertaken in other species
of animal isolated cells, etc., to determine the potential
of the sulfate-like substance to interact with other
pollutants, i.e., oxidant, NO., N, N03, etc., and with
biological agents (bacteria viruses, etc,) in inducing
infection or acerbating preexisting disease, altering
immunological response, etc.
D, The data from this study would be subjected to continuous
review by biologists, epidemiologists and atmospheric
chemists to obtain maximum interaction with, and input into,
other studies and to maximize the input from on^going
findings of air monitoring investigation, and to suggest
selection of elements for epidemiclogical correlations.
-82-
-------
G. References
1. Amdur, Mary 0. Toxicology of Decay Products of Sulfur Dioxide
in the Atmosphere. A Review Prepared for the Division of Health
Effects Research. NAPCA, 1969.
2. Health Consequences of Sulfur Oxides: A Report from CHESS 1970^71,
Human Studies Laboratory, National Environmental Research Center^
Environmental Protection Agency, Research Triangle Park, North.
Carolina. In Press.
3. Amdur, Mary. Toxicological Appraisal of Particulate Matter, Oxides
of Sulfur and Sulfuric Acid. Journal of the Air Pollution1 Control
Association, September 1969, Vol. 19, No, 9,
4. Meszaros, E. The Size Distribution of Water Soluble Particles
In the Atmpsphere, Idojares Budapest 75 (5/6): 308-.314, September/
December 1971.
5. McVilton, C. and R. Frank, The Role of Relative Humidity in the
Synergistic Effect of SO Aerosol Mixture on the Lung Science
(In Press).
6. Hazelton Laboratories, Inc. Chronic Exposures of Cynomologus Monkeys
to Sulfuric Acid Mist and Fly Ash Mixtures, Electric Research Council,
Project RP-74, December 1974.
-83-
-------
NOT QUOTE OR CITE
V. OTHER (WELFARE) EFFECTS
A. Ecological Effects
Sulfur is emitted from anthropogenic sources into the
atmosphere chiefly as S02. However, its deposition onto vegetation
and soil occurs not only as sulfur dioxide but also as sulfur
trioxide and sulfuric acid, depending on the environmental
conditions.
1. Vegetation
The sulfates which react upon vegetation are formed as a
result of the chemical reactivity of SO . SO enters the plants
through the stomata where its reaction with the moisture on
the surfaces of the mesophyll cells results in the formation of
*
sulfites and sulfates. A buildup of both of these compounds is
2 3~
injurious to plants. ' S02 which does not enter the
plants may react with moisture on the leaf surface to form sulfuric
acid with leaf damage resulting. The same affects may be caused
by sulfuric acid mist, acid aerosols or sulfate-bearing particulates
4 5
which dissolve in the dew. '
The deposition of SOX and particulates .onto tree bark can
result in the formation of sulfuric acid and this in turn results in
#*
the acidification of the bark. ^ Acid rain is also instrumental in
7-.7
causing pH changes in the bark.*-' It is the changes in pH of tree
bark and the leaf surfaces which influence the growth of bark-living
and leaf-surface organisms rather than the sulfates themselves. The
effects of the sulfates on the organisms has not been studied except
-84-
-------
NOT QUOTE 0.9 CITE
8 9 I'd
in the case of lichens. ' ' The sulfuric acid formed when
S02 is oxidized by lichens results in the degradation of chlorophyll
9
a_ in the lichen algae making photosynthesis impossible,
2. Soil
Sulfur dioxide and sulfur trioxide upon entering the soil are
I
rapidly oxidized by microorganisms to sulfates,, Sulfates may be absorbed
by plants through the roots, utilized and imobilized by microorganisms
or reduced by microorganisms to other oxidized forms of sulfur or
to hydrogen sulfide, Except in those cases where they are attached to
specific metallic ions which are in themselves toxic sulfates are
usually considered beneficial to the soil and soil organisms, Sulfates
are commonly added to agricultural soil as a component of fertilizers.
The change in pH caused by H2SC>4 in rain results in reactions in the
12 T"? k '~A
soil detrimental to plants and soil-living decomposer organisms. ปA ป14
The sulfate which contributes to the formation of acid rain is usually
attributed to the SC>2 in the atmosphere. However, preliminary studies
indicate that photochemical reactions involving organic sulfides
may result in the formation of sulfates which can also be involved
in rainout. The rainout of sulfate and the concentrations are
shown in the map, Figure 1, The high sulfate levels over the
northeastern U. S. could result in rain with increasingly lower pH
18
levels.' Acid rainfall, regardless of its source, introduces a very
significant stress on the ecosystem. The increased acidity of forest
19
soils is of particular importance in changing the soil communities.
-85-
-------
i
I
T
|
I
T
i
i
I
T
i
f
i
i
T
i .nr.
\ ttttซซ
i.. 3m&
i ...++4444.
I 44+444-
1 444444-
4 444+4 +f
t 4444-44+4
I +++444-
I 444444-
1 4+444-
1 444'
I 444-
I 444f
I 44444^
J + +t
\ >W\
I !i
U ::
1 <*
1 '
t
1 :'
i
t
j :'
i *
t ."
i .
i .
|:
I t. V P
T
\
..i;x!".""i!"".'.*"."!i."."'"rrr"i""""i:rc":::i""."""'
Ittttt
t^tUf*[[[
UttuH I:::::::-:::::::::::::::::::::::::::::::-::::::::-
4444444+ ... ...... .. ............ ...... .. . . ..
44444.4 1
imfi :: ; ::;
fi4++
444
ซ*[[[-^fl
;i: :::::::::::::::: :::::::::::::::::::::::::::::: jHiHHfi
Hll"1 PRHRRH
... RIB" FKPPFRRRRMR RBBHRH
"??. B"pBBRRBRRRBPBflRRHRBBRRPHBRRr
P.BRBRBI!lR!IRP!tBBBBBRR BBR.
PRRRRRRRRRRRERXflP.nRRBRRRRRXRR.IIRRHRI
BBRRPnnซRBBRRPRD*!l)IBRRRP.BPRRRni!BBRRซ
BRRRBBRP.BHBPIRBP.RPRBBBIIBBBRBBPRBRBI
BB!P.RBRRRRBflnR!IRRRRBBBBBII
B*BRBRRBR!lซg*P.!iRR"RBPBB!l)
RR'RaanaP.HI.'IRRI'RRRRRRBHRSr
Riinnn POTHT VftLOXS TW EIC9 tETปt " *
fiY-ontc ....!.... +4++2++++ OOOOJOPOO IMIUII
..... +4+f+4+++ ooiOOOCOO UUMI
I
I
.
II
<-. ปonปrii, T. mri iieTa. B.: osj , B.l..sc-iT (PP'1
c rirt ronpc': JPCO, mt't HP"ปOIซ. E. TIBCF isu B. BCCOMICK
c ซ?rico r.)-siGปiT!D BI R. HKTOCKITT si ซ, m cซpซ BIT^RIS PT n.
1BK01DT1 VI10K BIIGH IPPtniG TO MCI! LITZt
('PAIIBUR' TUCIODSO II RWHMT WfKl 0ซlf|
-------
Figure V-l-Footnote- 804"~ Rainout Map.
Sulfate rainout concentrations indicate that its major sources are local anthropogenic
origins. There is no real difference between continental and marine air masses except
in the urban areas, implying that in rural areas sulfate can be attributed to an oceanic
source with little rainout, or to a uniform inland source. Sulfate in rainout varied
to a great extent depending on the urban center sampled, with the extreme values
occurring in the northeastern industrial areas.
oo
^j
i
CD
CD
-H
,00
o
o
=o
o
-------
DO NOT QUOTE OR CITE
The climax communities of soil habitats are in equilibrium with their
macro- and micro-environments. A dynamic steady-state balance
exists among the constituent species. The introduction of chemical,
physical or biological stresses into the soil upsets the equilibrium
resulting in changes in the microbiota. Changes in the soil chemistry
will result in changes in microbial-activities, these changes in turn
affect the rates of mineralization and nitrification. Decreases
in the rates of these processes will have a direct effect on mineral
levels in the soil and therefore upon the productivity of plant
and animal life.
-88-
-------
DRAFT
DO NOT QUOTE OR CITE
B. Sulfate Effects on Materials
Sulfates accelerate the degradation of many materials.
"20 21 22
Examples are (1) concrete, (2) steel, (3) textiles.
None of these materials appear to have an ambient sulfate level
below which no effect will occur. This is because the rates
of deterioration appear to be functions of sulfate accumulation.
In the case of steel corrosion the presences of sulfates
in the corrosion products has a catalytic effect on oxidation rates.
That is, panels once innoculated with sulfate will continue to
corrode faster than panels not similarly treated.
Many particulate sulfates are very hygroscopic so that material
surfaces contaminated by them will remain moist well above the
dew point temperature. Because many materials are degraded more
rapidly when wet than when dry a. long list of materials are thus
indirectly affected by sulfate contamination,
With the exception of one EPA in-house study, there appears to
have been no attempts to relate rates of deterioration of materials
with levels of suspended particulate sulfates, The in-house study
on atmospheric corrosion of steel could not conclusively relate
corrosion behavior to sulfate levels because of a strong covariance
between S02 and sulfate levels.
-89-
-------
C. Weather, Visibility and Climate - UKAll
00 ป"T ?'.!?TC OR CITE
1. General Summary
The impact of sulfates on weather and climate processes evolves
from their chemical properties and from the physical form (i.e., particles)
in which they may occur in the atmosphere. Through the precipitation
scavenging mechanism of washout and fallout, sulfates contribute to the
composition and acidity of rainfall. As particles, however, their chemical
composition has little significance in relation to their effects on weather
and climate except to the extent that it may bear on the index of refraction
of the particles. On the other hand, their size distribution is an important
parameter relating to their participation in the scavenging mechanisms which
brings their chemical properties into an important function.
Sulfate particles as secondary pollutants are of intrinsic concern
in air pollution control. Any measure designed to abate SO- or "particulate"
primary emissions will, obviously, be ineffective. Therefore it is of par-
ticular concern to determine the extent to which these "secondary" sulfates,
arising from man-made (presumably controlled) sources, affect weather and
23
climate. Probably the most authoritative estimate is that, on a global
basis, man contributes about 50% of the total atmospheric loading of sul-
fate particles (which amounts to the order of 10% of the total particulate
emissions).
Sulfate (and other) particles act on weather and climate in es-
sentially two ways: a) by modification of the radiation fi&ld by the scatter
and absorption of sunlight and by radiating in the infra-red regions of the
spectrum; and, b) by affecting condensation processes as the number and
nature of the cloud droplets depends on the number and size distribution
of the particles. A complete review of this subject is contained in Ref. 24,
the "SMIC" Report. In summary, it can be stated that, on the global scale,
-90-
-------
' DRAFT
00 NOT QUOTE OR CITฃ
the primary effect of particles on the radiation field lies in its conse-
quences on the heat balance of the earth-atmosphere system. This is a
complex, and unresolved, matter involving the albedo (reflectivity to solar
radiation) of the earth's surface and the relative scattering vs absorption
properties of the particles. On the urban or regional scale, particles play
a role in the development and intensity of the so-called heat island. As
particulate layers radiate nearly as black bodies, they cool rapidly at
night, thus abetting the establishment of a mixing layer below and the
intensification of the upper stable lid, confining the pollutants to the
lower (ground-based) mixing layer. Thus, a feed-back mechanism is established
in which the particulate pollutants contribute to the increase in concentra-
tion of all pollutants by effecting a limited capacity for vertical dispersion.
Cloud properties as well as droplet formation and number density
^4
can be affected by particles in a number of ways. There is evidence that
urban precipitation can be modified (i.e., increased) by "...thermal and
aerosol inputs..." However it is probable that the most important considera-
tion is the impact of particle number density and properties, acting as
condensation nuclei, on regional and global cloud cover. A recent study^
indicates that a change in cloud cover is between 30 and 130 times more
effective "... in causing a change in surface temperature than a change in
dustiness by modification of the radiative transfer."
The two major sulfate removal processes are dry deposition, and
washout and raimautt While cleansing the atmosphere, the washout and rainout
processes may have unfortunate side effects.
Sulfate particles forming the nuclei of cloud and precipitation drop-
lets, and those captured by falling precipitation, are a factor in determining
the acidity of precipitation. The sulfites and sulfates formed from,the con-
-91-
-------
DRAFT
DO NOT QUOTE OR CITE
version of absorbed and adsorbed SO-, SO,, and H_S must also be considered
here. How important a determining factor appears to depend on the particular
situation.
26
Based on an analysis of Swedish ^ata, Granat states: "The pH in
precipitation is thus determined by several compounds of different origin.
It is thus not surprising that the relation between one element (sulphate)
and pH has a correlation coefficient which is not significantly different
from zero."
There are indications that it may be an important determining
factor in specific situations. This has been suggested by studies con-
27 28
ducted by Granat and Rpdhe in Sweden and Summers and Hitchon in Canada.
29
Similarly, Gordon has suggested that the damage to Christmas trees in the
Mt. Storm area may be due to high rainfall acidity resulting from the wash-
out of sulfates and sulfur oxides from a nearby power plant plume. The
questions of the role of sulfates in acid rain situations remains to be
resolved. Both the washout-rainout of particles and gases and the chemical
interactions on and within the cloud and precipitation droplets are only
partially understood. Battelle-Northwest Pacific Laboratories is examining
the washout of SO- problem, but much remains to be done. '.
In summary, effects of sulfates on weather and climate processes
are known to exist, but quantitative data are too meager to justify the
identification of 'criteria".
-92-
-------
DO NOT QUOTE OR CITE
2. Effects of Sulfate on The Composition of Rain and Snow
Sulfur dioxide dissolves readily in water to form sulfurous acid, H0SO_.
fc O
This rather weak acid dissociates in two stages to produce H+ ion
H2S03 = H+ + HSO~ Kj = 1.54 x 10"2 m/fc (1-25)
HS03" = H+ + S03 = K2 = 1.02 x 10"7 m/ฃ (1-26)
The magnitudes of Kj and K2 are such that in dilute solutions (such
as those in equilibrium with a typical atmospheric level of S0_,
like 0.1 ppm) the dissociation is extensive and contributes substantially
to the hydrogen ion concentration in the solution. If the S0_~ oxidizes
0
to S04~, then sulfuric acid is formed and the dissociation is essentially
complete.
The Impact of SO- on acidity of rain can not be assessed
quantitatively since the presence of other acids and bases, such as
(XL, NH_, and limestone dust, will affect the pH also. That S02
clearly causes an acidification of rainwater can be shown by the
following calculation. If there were no CaCCL or other basic dust,
no NH and a fixed, normalaamount of C02 at 320 ppm, the rain
is slightly acidic with a pH of 5.6 due to the dissociation
of carbonic acid. Under the same conditions with the addition of
0.5 ppm of S0_ in air, the equilibrium pH of rain is 4.0, neglecting
the oxidation of SO- to H0SOV When S0_ is converted to H.SO.,
224^ 24
the situation is less predictable due to the many paths by which
H-SO. can be trapped in rain. Bolin, et al. have concluded that
the pH of rain near sources is possibly controlled by S02 inter-
actions with rain, but that at greater distances (1000 km) the major
effects seems to be precipitation of scavenged H2SO. aerosol.
-93-
-------
DRAFT
QOJQLQUQILQR CITE
The study of effects of SCL on the composition of rain has
largely been a European endeavor with only few U. S. data available.
The report submitted by Sweden to the United Nations provides a
useful case study that clearly demonstrates the potential problem,
and the rate of change of composition as S0_ emissions increase
on a large scale.
The Scandinavian effort was based on data taken in a precipitation
sampling network established by the late C. G. Roosby in 1955 which
32
was subsequently extended over the whole of Europe. Oden and others
studied these data and have over the years pointed with alarm to
the trends in such quantities as rain pH, and the acidity of
lakes and rivers. The result of this attention was the establishment
of a general study of the deposition of acidic water on Scandinavia
for submission to the U. N.
31
Bolin, et al. present the data in terms of deposition
of S0v~ and H in mg of sulfur or hydrogen ion per square meter
2
per year (mg/m year) in their report. The similarity of the deposition
pattern of sulfate and H in Europe, coupled with models of transport,
31
emission surveys and other considerations led Bolin et al. to
conclude that the observed acidity was due to H-SO. formed from SCL.
24 2
The relevance of the study by Bolin et air* lies mainly in
real and predicted biological effects (including effects on soil
chemistry). It is pertinent at this stage to" emphasize
again the basic reason for interest in rain pff. Two basic
effects seem to occur: _..-..
-94-
-------
M,
DO NOT QUOTF 0" PITT
f* v ** ซ ซyซ i -k.
1) The leaching of substances from soils, particularly calcium,
with attendant changes in productivity. Some forest soils seem
especially sensitive to this effect.
2) The lowering of pH of some lakes and rivers which have low
buffer capacity. The pH is low enough in some bodies of water
to actually become critical to some species of fish.
3 . Effects of_ S0_2 and Its Byproducts oil Light Transmission and
Visibility in the Atmosphere
a. Introduction
The visibility of objects in air is controlled by the ability of
the atmospheric sight path to transmit light. Similarly, the intensity
of sunlight reaching the ground is controlled by atmospheric
transparency. Decreases both in visibility and in intensity of
solar radiation are important effects that are often attributed
partly to sulfate aerosols created when S02 is oxidized. The
purpose here is to examine the role of SCL and sulfate on light
transmission in air. Only the salient optical features regarding
the role of SO- will be discussed.
-95-
-------
__1. __BO NOT QUOTE OR CITE
t>. The Effect of Aerosols on Light Transmission Through Ai:r
The intensity of a light beam us it passes through the atmosphere
(e.g. from the sun or from an object at a distance) is diminished
by the presence of gases and particles. This process is formally
described for a homogeneous medium by the Beer's-Lambert law
where I0 is the intensity of a light bean of a given wavelength prior
to its passage through a distance x of air, I is the intensity after
passage and b is the extenction coefficient. Thฎ quantity b can
be represented by a sum:
b = b +bnn.,+b, n B, (V-2)
scat Rayleigh abs-aerosol + abs-gas
where b is the extinction due to scattering by aerosols particles,
scat
b_ 1 . , is the scattering due to the air molecules themselves,
bahซs oa* is the absฐrption by gases such as NO _ and b ,
aos-gas 2 abs-aerosol
is the absorption by particles such as iron oxide of soot. In
general, the quantity bscat is the largest except in very remote,
background locations where b often becomes smaller than b
scat Rayleigh'
While few measurements exist, b , and b , are thoueht
abs-gas abs-aerosol 6
to be smaller than bsc&t by a factor of 10 or more in most cases.33'54
(The by-products of SO- oxidation
in air are aerosols, and these sulfates are usually not strongly
colored, (e.g., (NH4)2S04 is white). The foeus for this
discussion is the quantity b Rayleigh scattering, b - -
scat Raleigh,
-96-
-------
DRAFT
00 NOT QUOTE OR CITE
is usually much smaller than b in polluted air, and in any
S Ccl L
case is not controllable. There are two notable features of
light scattering that deserve mention.
First, the process of scattering does not consume energy from a
light beam, but merely redirects its path through the air.
Second, the scattering of light by aerosol and gases is
wavelength or color dependent. In typical smog, this leads to
the predominant scattering of blue light, which in turn allows ,
preferential transmission of red light. The result of this color
effect is that distant dark objects appear blue (like dark mountains)
and bright objects appear reddened (the "harvest" moon, sun, clouds,
snow-capped peaks, etc.).
In the presence of aerosols, a dark-colofed object appears
lighter due to the scattering of light by the particles into the
light path. In the case of a light-colored object, light is lost
from the line of sight due to scattering. This reduced contrast
between an object and the horizon sky is a prime factor in causing
visibility degradation. The limit of visibility is reached when
the eye cannot distinguish a contrast between an object and its
background.
-97-
-------
DRAFT
DO NOT QUOTE OR CITE
4.' Visibility and Its Relationship to Aerosol Mass Concentration
In the simplest case of a non-absorbing, homogenous, low
humidity haze, the visual range, (assuming a limiting contrast of 2%)
L , can be related to the extinction coefficient due to scattering,
scat'
Ly = -^ CV-3)
scat
where b is evaluted at the wavelength of maximum sensitivity of
s cat
35
the eye, 550 nm. In turn, b has been empirically and
s ca u
3^
theoretically related to the mass concentration of aerosol.'
Thus, if SO = is related to total aerosol concentration,
a quantitative relationship might exist between S0.= and L . Although
there are almost no data to confirm such a relationship, there is
every reason to believe that SO = aerosols beSave in similar
33
optical fashion to other atmospheric aerosols. Charlson
and others have shown that the product of visual range times mass
concentration has a most probable value of ca. 1.8 g/m :
Lv(X) x Mass (g/m3) =1.8 (g/m2) (V-4)
with 90% of the cases in ordinary urban aerosol falling between
0.9 and 3.6 g/m . The use of this relationship would depend on
such complex factors as the distribution of SO." with particle
size.
-98-
-------
DRAFT
DO NOT QUOTE OR CITE
At higher humiditiessomewhere above 65 - 70% the situation
is much more complex due to the enhancement of extinction by
adsorbed or liquid H-0 associated with the particles and no quantitative
relationship between SO ~ and L is obvious.
A. Quantitative Aspects
In spite of the complexity of the S0_.*SO ~ conversion process
and its many and varied optical effects, it is clear that SO- can
lead to visibility degradation in the atmosphere and^depending. on
local conditions, it can be the dominant causal factor. The
amount of SO ~ produced by the conversion of a typical level
of SO- in polluted air, say 0.1 ppmy can be related in an approximate
way to the visual range at low relative humidity. This amount of SO-
converted to (NH.)2SO. Ca likely end product) yields a concentration
of 550 yg/m , a quantity that is large by urban standards. The
empirical results of a number of workers suggest that this
amount of aerosol would create an extinction coefficient of about 1.2 x
10" m" , or a visual range of ca. 3.3 Km 0\>2 miles). If even 10% of
the 0.1 ppm of SO- (i.e., 0.01 ppm) were so converted, the amount of
aerosol (55yg/m ) is not inconsequential. The problem remains to
relate SO- to the actual amount of aerosol produced. Attempts to
relate SO- to SO ~ in a simple way, or to relate SO- to visual range
have been made, but these have been criticized as being premature.
There is reason to believe that the optical effects of the
products of gas to particle conversion will be systematically enhanced
by the fact that such reactions oftem.proceed via heterogeneous
condensation. Such processes occur in that part of the size
distribution in which the surface area of particles is maximized,
-99-
-------
DO NOT QUOTE OR CITE
which is usually at sizes below 1 pm. Particles in these small size
classes dominate the optical effects while particles larger than ym seldom
contribute significantly.
Thus the typical range of from 5 - 20% sulfate in urban aerosols
(6.3 -25% as (NH4)2S04) should produce at low R.H. at least 6- 25%
of the observed b , and possibly more due to particle size
S C3-L
effects. At high relative humidity, the fact that many sulfates
are deliquescent salts, or worse yet that HปSO. (a possible product
of SCL) is a very hygroscopic liquid, will enhance the extinction
caused by a given concentration of aerosol. In comparison to
hygrophobic substances, like soot or oil fumes, SO ~ compounds
have a larger effect per unit of mass at high RH.
b. Supporting Data
There is a general lack of simultaneous optical, SO^ and SO."
measurements. While either b . measurements or L observations
scat v
would serve, optical measurements have been made in few U. S.
cities on a routine basis.
*g._ _
Rodhe et al report a study of visibility and SO, in rural
Scandinavia which showed a high correlation (0.82) between L and SC.
Based oH"tihe empirical correlations between total aerosol mass
5|
concentration and L reported by Charlson, it was concluded that
perhaps 4 percent of the visibility degradation was due to sulfate.
However, half the days in one of these studies had RH>70% so a
substantial part of the extinction could have been due to H~0 in
or on the aerosol particles.
-100-
-------
nn
DO NOT QUOTE OR CITE
The effects of high relative humidity have posed difficult
problems in estimating the relationship between visibility and
SO ~ mass concentration. Covert et al. developed a method
for studies of b (or L ) as a function of RH. Both laboratory
S CcL C. V
and atmospheric aerosols were studied. The aerosol size
distribution of the laboratory aerosol was tailored to
replicate typical atmospheric cases. The results are summarized
in Figure V-'l which includes (NH.)?SO. an<* HปSO. laboratory
aerosols as well as an atmospheric case. At 90% RH, b
r ' scat
increases in most cases by a factor of two or more. At 95% RH,
(b increases of over fourfold) are noted in both laboratory
S CcL L
and ambient air. It is important to recognize that the optical
response of an atmospheric aerosol to humidity is influenced by its
molecular composition, not by the presence of SO." ion alone;
that is NH4HS04, (NH4)2S04 and H2S04 all behave differently.
-101-
-------
o
N)
I?
II
&
a
o
(A
JQ
a
o
V)
CC
LLJ
O
CO
H-
u
_i
UJ
p
_J
AMBIENT AEROSOL
SEATTLE, WASH.
1300 PST. THURS.28 JAN. 71
H2S04
O
O
o
20
40
60
80
100
RELATIVE HUMIDITY, (percent)
Figure V-i. Light scattering as a function of relative humidity after Covert
et al(37) fฐr (A) H2S04 (B) (NH^ 804 and (C) an atmosoheric
case (ambient aerosol. Seattle, Wash., 1300 PSTJhurs, 28 Jan.
-------
rn DRAFT
W NOT QUOTE OR CITE
D. REFERENCES
1. Air Quality Criteria for Sulfur Oxides. AP-50. Superintendent
of Documents. U. S. Government Printing Office. 1970.
Washington, D. C.
2. Thomas, M. D., R. H. Hendricks, and G. R. Hill. Sulfur Metabolism
of Plants: Effect of Sulfur Dioxide on Vegetation. Ind. Eng.
Chem. 42:(11):2231-2235, 1950.
3. Thomas, M. D., R. H. Hendricks, and G. R. Hill. Some Impurities
in the Air and Their Effects on Plants. In* Air
Pollution. McCabe, L. C. (ed.). McGraw-Hill, New York, p. 41-47.
1952.
4, Middleton, J, T., E. F. Darley, and R. F. Brewer. Damage to
Vegetation from Polluted Atmospheres. J. Air Pollut. Contr.
Ass, 1958.
5, Crowther, C., and H, G. Rustan. The Nature, Distribution arid
Effects Upon Vegetation of Atmospheric Impurities in and Near
an Industrial Town, J. Agric. Sci. 4^25-55. 1911.
6. Grodzinska, K. Acidification of Tree Bark as a Measure of Air
Pollution in Southern Poland. Bull. Acad. Polonaise, Sci., Series
Sci, Biol, 19(3): 189-195, (In English). 1971.
7. Stoxang, Birgitta. Acidification of Bark of Some Deciduous
Trees, Oikos 20;224-230. 1969.
8, Gilbert, 0. L, Further Studies on The Effect of Sulfur
Dioxide on Lichens and Byyophytes, New Phytol. 69:605-627,
1970.
-103-
-------
DO NOT QUOTE OR CITE
9. Rao, D. N. and F. LeBlanc. Effects of Sulfur Dioxide on the
Lichen Algae, with Special Reference to Chlorophyll. Byrologist.
69_(1): 69-75. 1966.
10. Skye, E. Lichens and Air Pollution. A Study of Cryptogamic
Epiphytes and Environment in the Stockholm Region. Acta Phyto-
geographica Suecia. 5_2:1-123. (In English). 1968.
11. Burns, George R. Oxidation of Sulfur in Soils. Tech. Bull. #13.
The Sulfur Institute. Washington, D. C. 1967. JJ(1):9-15.
12, Mrkva, R. and B, Grunda. Einflussvvon Immissionen auf die
Waldboeden und ihre Mikroflora im Gebiet von Suedmaehren.
[Effect of Immission on the Forest Soils and Their Microflora
in the Region of Southern Moravia.] Acta Univ. Agric., Brne
(Fac. Silva.) _38_(3) :247-270. (Typescript Trans.) 1969
13. Oden, S. Nederboesfdens Och Luftens Forsurning - Dess Orsaker,
Foerlopp Och Verkan i Oilka Miljoer. [The Acidification of Air
and Precipitation and Its Consequences on the Natural Environment.]
Swedish Natural Science Research Council. Stockholm. Bull. No. 1.
86 p, (Typescript Trans.) 1969.
14. Gorham, E, Free Acid in British Soils. Nature 181:106. 1958.
15. Engstom, A, Air Pollution Across National Boundaries,
The Impact on the Environment of Sulfur in Air and Precipitation.
Report of the Swedish Preparatory Committee for the U. N.
Conference on Human Environment. Stockholm 96 pp. 1971.
16. Junge, C. E, Sulfur in the Atmosphere. J. Geophys. Res. 65(1):
227^237, 1960.
-104-
-------
DRAFT
DO NOT QUOTE OR CITE
17. Wolaver, T. G., and H. Lieth. The Distribution of Natural
and Anthropogenic Elements and Compounds in Precipitation
Across the U. S, UNC Duplicating Shop. Chapel Hill, N.C.
1972.
18. Pearson, F, J. and D. W. Fisher. Chemical Composition
of Atmospheric Precipitation in the Northeastern United States.
Geol, Surv. Supply Paper 1535-P. 23 pp. 1971.
19. Likens, G, E., F. H, Bormann, and N. M. Johnson. Acid Rain.
Environment 14(2):33-40. 1972.
20. Buehanan, D. R. "The Corrosion or Deterioration of Concrete."
Australian Corrosion Engineering 14: No. (5), 5 May 1970.
21, Preston, R. St. J. and Sanyal, B. "Atmospheric Corrosion
by Nuclei," J. Applied Chemistry 6^:26. JanUasy, 1956.
22. "Unusual Failures of Stockings in Service," Tech. New Bull.,
Nat, Bur, Stand, p. 33, March 22, 1941.
23. Inadvertent Climate Modification. Report of the Study of Man's
Impact on Climate. MIT 201, The Massachusetts Institute of
Technology Press, Cambridge, Massachusetts, 1971.
24. Huff, F, A., and Changnon, S. A., Jr., Climatological Assess-
ment of Urban Effects on Precipitation. Final Report. Illinois
State Water Survey, Urbana, Illinois. 1972.
25. McClintock, M. et al. Research on the Optical State of the
Atmosphere. Report to EPA, Contract No. 68-02-0337, Space
Science and Engineering Center, University of Wisconsin (draft)
1972,
-105-
-------
NOF QUOTE OR CITE
26. Granat, L. On the Relation Between pH and the Chemical Compo-
sition in Atmospheric Precipitation, Report AC-18, International
Meteorological Institute, Stockholm, Sweden. 1972.
27. Granat, L., and H. Rodhe. A Study of Fallout by Precipitation
Around an Oil-fired Power Plant, Report AC-22, International
Meteorological Institute, Stockholm, Sweden. 1972.
28, Summers, P. W., and B. Hitchon. Source and Budget of Sulphate
in Precipitation from Central Alberta, Canada, Presented at the
1971 Annual Meeting, Pacific Northwest International Section,
APCA, Calgary, Canada. 1971.
29. Gordon, C, C., Plantations vs Power Plants. American Christmas
Tree Journal, August 1972, 5-10. 1972.
30. Butcher, S. S. and Charison, R. J. JStf Introduction to Air
Chemistry, Academic Press, N. Y. 1972.
31, Bolin, B. (Chairman), Sweden's Case Study Contribution to the
United Nations Conference on the Human Environment - Air Pollution
Across National Boundaries: The Impact on the Environment of Sulfui
in Air and Precipitation. Royal Ministry for Foreign Affairs,
Kingl, Boktrycheriet, P. A. Norstedt et Soner, 710396, Stockholm.
1971,
32, Oden, S. Regional Aspects of Environmental Disturbance, Vann,
Vol. 3, Saertrykknr, 46, Johansen and Nielsen, Oslo.
33. Charison, R. J., Atmospheric Visibility Related to Aerosol
Mass Concentration A Review," Environ. Sci. Technol. 3_:913-918.
1969.
-106-
-------
nn
DO NOT QUOTE OR CITE
34. Ensor, D. S., Proch. W. M., Pilaf, M. J., and Charlson, R. J.
Influence of Atmospheric Aerosol on Albedo, J. Appl. Meteorol.,
JLO'. 1303-1306. 1971.
35. Horvath, H. and Noll, K. E. The Relationship Between
Atmospheric Light Scattering Coefficient and Visibility. Atmos.
Environ., 3_:543-552. 1969.
36. Rodhe, H., C. Persson, and 0. Akerson. An Investigation into
Regional Transport of Soot and Sulfate Aerosols, Report AC-15,
NDC 551. 510.4 Meteorological Institute. Stockholm University
Stockholm, Sweden. 1972.
37, Covert, D. S., Charlson, R. J., and Ahlquist, N. C. A Study
of Chemical Composition and Humidity to Light Scattering by
Aerosols, J. Applied Meteorol. _!!_: (6)968-976, 1972.
-107-
-------
DRAFT
DO NOT QUOTE OR CITE
VI. TECHNOLOGY FOR STATIONARY SOURCES
& Sources
Ambient sulfates in urban areas result primarily frcm the conversion
of ambient S0? to sulfate, transport of sulfates from outside the
urban area, and primary emissions of sulfates from combustion sources
and industrial processes.
In suburban and rural areas ambient sulfates result primarily from
the transport of ambient sulfate from urban areas.
B. Control Strategy
S09 is emitted from various types of sources. Technology which is
feasible for one source (such as utilities) will not necessarily be
feasible for other sources (such as industrial combustion, industrial
processes, and home furnaces). Our strategy is intended to provide
control technology for each of these sources. In summary, our strategy
is "to develop control technology which may be applied to various sources
for reducing SO emissions by 90-99%."
X
Recent health studies indicate that the ambient sulfate levels must be
controlled to an annual average of less than 8yg/m .
According to Altshuller , in urban areas where the major source of sulfates
is the conversion of S0 to sulfates, ambient SO concentrations must be
3 3
reduced to 20yg/m to reduce sulfate levels to 7.8yg/m .
-108-
-------
DRAFT
DO NOT QUOTE OR CITE
If we could assume that each source of S0~ only contributed to ambient
SO concentrations according to its pro-rata share, then we would assume
that the total percentage S0~ reduction required would be the same as
the percentage reduction required for each source. Our strategy would
be based on the "roll-back" technique and the percentage reductions
necessary for SO would be as indicated in Table 1.
Annual Average
Ambient SO-
Concentration
in an Urban Area
80
200
300
400
Table 1.
% Reduction of S02
Concentration Needed
to Meet Primary
Ambient Air Quality
Standards of
% Reduction of S02
Concentration Needed to
Meet Potential Ambient
Air Quality Standard
of 20yg/m3
0
60
73.3
80
75
90
93.3
95
Yet more recent SO- and sulfates trend data analyzed by Frank indicate
that S0~ reductions were not accompanied by sulfate reductions. Frank
theorized that sulfate levels remained constant because urban areas re-
duced ambient SO- concentration by requiring non-utility combustion
sources to switch from high sulfur fuels to low sulfur fuels. To supply
the low sulfur fuels (gas), utilities switched to higher sulfur fuels (coal).
Because of their high stacks, the SOV emitted by utilities dispersed without
A
exceeding ambient standards, but the converted sulfates were transported
into the urban areas thus preventing a decrease in sulfate levels. This
-109-
-------
DRAFT
DO NOT QUOTE OR CITE
indicates that even greater percentage reductions in SO. will be needed
than those indicated in Table 1.
Recent experience by Ambient Air Quality Regions has illustrated that
the roll-back technique for achieving ambient air quality standards is
inadequate; S07 concentrations in the vicinity of specific sources can
significantly exceed previously predicted levels. The effect upon air
quality of a specific source depends on many variables such as meteorology,
stack height, temperature of stack gases, and population exposed. Thus
control technology must be made available to reduce S0? emissions beyond
those reductions indicated in Table 1.
C Status of Control Technology
The development of technology for control of S0_ emissions is progressing
quite well. Within several years proven technology will be available for
90-95% reduction of S0~ emissions from large point sources. Further
development could reduce emissions from large point sources by as much as
98-99%. However, for area sources, cost effective reductions are limited
to 50-80% unless natural gas replaces the conventional fuels.
The current technology program is targeted at moderate (30-70%) reductions
of ambient concentrations of SO- in urban areas. This will be accomplished
by using stack gas cleaning on large point sources (75-90% abatement) and
cleaner fuels on most other sources (20-60% abatement). To achieve up to
and greater than 95% reductions of ambient concentrations of S02 in urban
areas large point sources will require high efficiency stack gas cleaning
(90-99%) and other sources will require increased use of clean fuels. A
program to provide the technology which will be needed for reducing S09
is outlined in Section VII.
-110-
-------
UKAh I
DO NOT QUOTE OR CITE
The program presented In Section VII is our approach for controlling
sulfates through reducing SO-. Future work concerning atmospheric
reactions, atmospheric trends and health effects may reveal other
approaches which would be used or may be needed to reduce sulfates (such
as controlling ammonia, vandium, or an unknown catalyst). Thus, as the
sulfate problem becomes better defined, the control technology R&D pro-
gram may have to be significantly increased to ensure adequate sulfate
control most economically for the nation.
-Ill-
-------
D.
References
DRAFT
DO NOT QUOTE OR CITE
1. Altshuller, Aubrey P., "Atmospheric Sulfur Dioxide and Sulfate",
Env. Sci. ง Tech., August 1973.
2. Frank, Neil H., "Temporal Relationships of Sulfates, Sulfur
Dioxide and Total Suspended Particulate", Office of Air Quality
Planning ง Standards, Draft Paper, August 1973.
-112-
-------
nn DRAFT
DO NOT QUOTE OR CITE
VII. CONCLUSIONS AND RECOMMENDATIONS
A. Conclusions
There are still many uncertainties concerning the total sulfur
cycles, and the effects of atmospheric sulfur species on human health
and welfare. Particularly the role of suspended sulfates and sulfuric
acid aerosols is not well known. There are reasons to suspect that
control of SO- may not result in a proportionate control of SO..
& ^v
However, a number of conclusions can be drawn from current knowledge
and available data:
(1) Anthropogenic sources contribute approximately one half as
much as nature to the total sulfur content in the atmosphere. The
anthropogenic contribution from urban area sources is much greater than
the natural sources. This ratio is increasing.
(2) Approximately 95% of the sulfur emitted to the atmosphere
from urban sources is in the form of S0~. The primary urban source of
sulfate is the atmospheric oxidation of S0_ to H-SO. with subsequent
neutralization or exchange reactions giving a variety of sulfates.
Essentially all of the S02 in the atmosphere is converted to XSO^
prior to or during the removal processes.
(3) The major portion of the sulfate particles in the atmosphere
are in the respirable range.
(4) Sulfates can have a deleterious effect on human health and
welfare, including the eco-system; (reliable threshold levels have
not been established.) Sulfuric acids and certain sulfates are more
potent irritants than S0_.
(5) Current knowledge and available data are inadequate at this
time to establish criteria which might be used as a basis for standards.
-113-
-------
DRAFT
' no NOT QUOTE OR CITE
B. Recommendations
There are critical gaps in our knowledge concerning the total
sulfur cycle in general and of the acids and sulfates in particular.
Little is known regarding the rates of conversion of SO- and H2S
to sulfate particles in urban and non-urban atmospheres. Suitable
means of distinguishing contributions by anthropogenic and biogenic sources
to sulfates in the atmosphere are not available, and the nature and
efficiency of removal processes are not well known. There is substantial
evidence that sulfuric acid and sulfates may have more detrimental
effects of human health and welfare than S0_, however, there are serious
gaps in our knowledge concerning the biological and ecological effects of these
potential pollutants. There is a paucity of information concerning the
role of sulfates in the production of chronic pulmonary diseases.
Particle size and number density appear to be important parameters,
and the effect of certain sulfates in combinations likely to be found
in the ambient atmosphere may be greater than the sum of the
individual effects--these aspects have not been studied extensively.
Suspended sulfate may have a significant affect upon weather,
visibility and climate but sufficient quantitative data are not
available to characterize the problem. Suitable methods for
measuring sulfuric acid and sulfates in the atmosphere are not available,
and the integrity of much of the data collected to date is in question.
Solutions to these problems must be achieved before meaningful and
rational decisions can be made concerning the need for control, the
-114-
-------
DRAFT
DO NOT QUOTE OR CITE
optimum control strategy to be employed should the need for control
exist, and the nature of the program necessary for monitoring and
enforcement.
To achieve the necessary level of knowledge and tinderstanding
will require a research and development effort well coordinated in
time and substance. The problems are indeed multidisplinary and
milestone decisions may depend heavily upon the results obtained from
the various approved tasks. Parallel efforts will be required and
a certain element of risk must be accepted in initiating the long-ter
projects which may require controlling input from other program
elements. For example, the success of the CHESS studies will depend
heavily upon sampling and measurement capability and the collection
of proper and reliable atmospheric data. The research task below
constitutes the substance of the recommended research program,
-115-
-------
DRAFT
DO NOT QUOTE OR CITE
Task
1. Conduct a Community Health and
Environmental Surveillance System
(CHESS) study designed specifically
to investigate the effects of total
respirable particulate matter and
suspended sulfates in the atmos-
phere on human health. The current
CHESS studies in the New York
metropolitan area, the Southeastern
U. S. and Salt Lake Basin will be
modified to include necessary new
parameters.
2.a. Conduct biological expert-?
ments using dynamic atmosphere with
gases and particulates similar to
urban atmospheres for exposure of
various species of animals, and other
biological models to determine both
direct effects and effects through
interaction with other pollutants and
infectious agents. Use acute, subacute,
and chronic exposures to determine such
factors as influence of particle size,
specific sulfate, and various combinations
on particle deposition, retention, trans-
location, pulmonary clearance rates,
acceleration of infecious states, and
various parameters related to pulmonary
defense (Coordinate with CHESS studies).
b. Generation of sulfuric acid
aerosols and sulfate aerosols for
biological and health effects studies,
Aerosols will be produced by chemical
techniques which simulate their formation
in the atmosphere. Particle size dis-
tribution and chemical composition will be
monitored and adjusted to give the desired
pulmonary dose.
In-House
Man-Years
60
Time
Period
For
Completion
(Years)
Contact
National
Environmental
Research
Center
Same
Same
-116-
-------
DRAFT
DOT iJOTE OR CITE
Task
3. Rate and Mechanism of Sulfuric
Acid and Sulfate Formation,
Additional studies of the Conversion
of SO, tQ suifuric acid in Urban
atmospheres are needed, Emphasis
should be placed on the reactions
of oxygenated radicals with SOj and
catalytic reactions in or on aerosols.
This information is required to deter*
mine if control of urban sulfate
levels can be obtained by Iwering
862 levels or if better control can
be obtained by lowering oxidants or
catalytic particulate matter.
a. Gas-Phase Reactions
b. Gas-Surface Reactions
4. Formation of Sulfate Compounds,
New studies are needed to determine
the extent to which suifuric ฃei
-------
DRAFT
DO NOT QUOTE OR CITE
Task
1 - Investigate sulfate emissions
from stationary sources. The first
major consideration which needs to be
addressed is the question whether
the sulfate is emitted from stationary
sources in the gas phase as H^SC^
vapor or in the particulate phase as
sulfate. The measurement of the actual
amount of H2S04 vapor present in
source emissions will determine the
ultimate control strategy required.
The following R $ D tasks are
recommended.
In-House
Man-Years
Time
Period
For
Completion
(Years)
Contact
National
Environmental
Research
Center
a. Develop an in-situ H2SO.
measurement method.
b. Verify existing manual
compliance test methods for
S03 and H2S04>
c. Develop Raman scattering
and/or fluorescence technique
for analysis of particulate
sulfate in-situ.
d. Conduct a study of parti ~
culate/filter/H2SO / sulfate
interactions , temperature
dependence of sulfate, con-
version on probes and filters,
verifications of true particulate
sulfate (as distinct from con-
dens ible
8, Evaluate current laboratory
analytical methods for H S04 and
sulfates.
9. Develop a reliable method
for total sulfate, and evaluate
collection techniques for
atmospheric sulfates in the
presence of S02 and other
critical pollutants.
0.25
0.25
0.25
0.7
Same
Same
1 Same
0,25
1 Same
1 Same
1 Same
-118-
-------
nn
00 NOT QUOTE OR CITE
Time
Period
For
In-House Completion
Task Man-Years (Years) Contact
10. Develop a reliable ion- -- 1 National*
selective electrode for sulfate, Environmental
Research
11. Develop a reliable -- 1 Center-
ed lection method for H-SO,.
24
12. Determine the various type of --''.. 1 Same
sulfates and their relative pro~
portions in the atmosphere.
13.a. Characterize the various -- 1 Same
types of sulfur compounds in parti-
culates in the atmosphere in urban
and non-urban atmospheres.
b. Determine to what extent -- 3 Same
man-made sulfates contribute to the
acidity of precipitation.
14. Determine the sulfate and -- 1 Same
sulfuric acid composition relative
to size and mass distribution,
15. Collect additional urban and -- 3 Same
non-urban sulfate data including
particle size and number density,
16. Conduct biological experi- 4 5 National
ments to characterize the physio- Environmental
logical and morphological responses of Research
plants to sulfates and to relate Laboratory
these responses to growth and yield
of plants. (To be conducted at the
Corvallis NERC),
17. Conduct biological experi- 6 4 National
ments to determine the effects of . Environmental
acid rainfall of plants,^the soil and Research
on microorganisms in the soil. Laboratory
To determine the changes which occur
in the ecosystem due to acid rainfall
and how these affect mineral cycling,
(This project has already been sub-
mitted and has received funding for
FY-73. There is need for further
funding for FY-74),
Coordinate with Dr. Hooever, Southeastern Environmental Research Laboratory.
-119-
-------
UHAhl
DO NOT QUOTE OR CITE
In-House
Man-Years
Time
Period
For
Completion
(Years)
7- Investigate sulfate emissions
from stationary sources. The first
major consideration which needs to be
addressed is the question whether
the sulfate is emitted from stationary-
sources in the gas phase as H^SC^
vapor or in the particulate phase as
sulfate. The measurement of the actual
amount of f^SO. vapor present in
source emissions will determine the
ultimate control strategy required,
The following R ง D tasks are
recommended.
a. Develop an in-situ H2SO.
measurement method.
b. Verify existing manual
compliance test methods for
S03 and H2S04.
c. Develop Raman scattering
and/or fluorescence technique
for analysis of particulate
sulfate in-situ.
d. Conduct a study of partis
culate/filter/H2SO / sulfate
interactions, temperature
dependence of sulfate, con-
version on probes and filters,
verifications of true particulate
sulfate (as distinct from con-
densible
Contact
National
Environmental
Research
Center
8. Evaluate current laboratory
analytical methods for H_S04 and
sulfates.
9. Develop a reliable method
for total sulfate, and evaluate
collection techniques for
atmospheric sulfates in the
presence of SC>2 and other
critical pollutants.
0,25
0.25
0.25
0.7 Same
Same
1 Same
0,25
1 Same
1 Same
l Same
-118-
-------
DRAFT
DO NOT QUOTE OR CITE
Task
d. Source Control - Industrial
Combustion: Develop and demon*-
strate flue gas cleaning for
industrial combustion SO^ sources,
providing from 70-99% efficiency,
e. Source Control - Industrial
Processes:
(1) Identify and develop tech-
nology for control of industrial
processes (up to 99% removal),
(2) Demonstrate industrial pro-
cess technology (cost shared 50/50
with private industry),
f. Source Control - Area Sources:
Develop and demonstrate package
sorption techniques for SC^ from
area sources providing from
70-99% efficiency.
In-House
Man-Years
Time
Period
For
Completion
(Years)
Contact
2 National
(FY'75-77) Environmental
Research
Center
2 Same
(FY75-79)
(FY77^79) Same
(FY76-78) Same
-121-
------- |