United States
Environmental Protection
Agency
Environmental Sciences Research EPA-60Q 2-80086
Laboratory V1,iy 1 980
Research Triangle Park NC 27711
Research and Development
Impact of Natural
Hydrocarbons on
Air Quality
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental degradation from point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pollution sources to meet environmental quality standards.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/2-80-086
May 1980
IMPACT OF NATURAL HYDROCARBONS ON AIR QUALITY
by
Joseph J. Bufalini
Gas Kinetics and Photochemistry Branch
Environmental Sciences Research Laboratory
Research Triangle Park, North Carolina 27711
ENVIRONMENTAL SCIENCES RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NORTH CAROLINA 27711
-------
DISCLAIMER
This report has been reviewed by the Environmental Sciences Research
Laboratory, U.S. Environmental Protection Agency, and approved for
publication. Mention of trade names or commercial products does not
constitute endorsement or recommendation for use.
ii
-------
ABSTRACT
The emissions, reactivities, and ozone-forming potential of natural
hydrocarbons are discussed. A review of the data available on emission
levels for natural hydrocarbons indicates that much more information is
needed in order to quantify the absolute emission levels, because emissions
data do not agree with ambient air measurements. These ambient air measure-
Q
ments suggest that the previously-published value of 9 x 10 ton/yr needs to
be lowered to 10 -10 ton/yr. Emissions may be overpredicted by a factor of
15 to 20, as indicated by back calculations using a simple diffusion tra-
jectory model.
Isoprene, when compared to the monoterpenes, is much more efficient in
producing ozone through photooxidation in the presence of NO . This
X
greater ozone production apparently occurs because of the large amount of
carbon consumed in the formation of aerosols for the monoterpenes. Since
rural areas have very low levels of NO , vegetative emissions may in fact
X
act as sinks for ozone rather than as sources. All areas investigated show
very low levels of natural hydrocarbons, suggesting that even if NO were
* X
available, very low levels of ozone would be produced. Air quality is
thus not found to be significantly affected by vegetative emissions.
iii
-------
CONTENTS
Abstract iii
Figures vi
Tables vii
Acknowledgement viii
1. Introduction 1
2. Conclusions 2
3. Recommendations 3
4. HC Emissions 5
5. Reactivities of Natural HC's 17
6. Product Formation 33
7. Source-Receptor Relationships 43
References 47
Appendix 53
-------
FIGURES
Number Page
1 Structure of selected natural HC's ............................ 10
2 Loblolly pine forest vertical gradients, 7/18/77, 1035-1100 14
hr EST [[[
3 Ozone production as a function of HC/NO ...................... 20
A
4 Effect of HC/NO ratio on ozone "wyt"n«" .......... . ............ 21
5 Diurnal variation of HC compounds, Chickatawbut Hill,
July 18, 1976 ............................................... 23
6 Total concentration of the C-0 terpene HC's (ppbC) observed
in the canopy at IBP site. .7 ................................ 2*
7 Total concentration of the C^Q terpene HC's (ppbC) observed
above the canopy at IBP site 25
8 Photooxidation of a-pinene in the presence of NO 33
-------
TABLES
Number Page
1 Emission rates for selected samples at various locations 11
2 Major Emissions from Selected Species 12
3 Estimated U.S. and Global Emissions by Lattitude (gC/yr) 16
4 Rate constants for selected EC's 18
5 Half-life for selected HC's at various ozone concentrations... 27
6 HC' s found near several cities (ppb V/V) 28
7 Summary of biogenic HC concentrations at ambient air 30
8 Reactivity and products by Fourier Transform Spectroscopy 35
(long-path infrared) at 60 min
vii
-------
ACKNOWLEDGMENT
I would like to acknowledge the many helpful discussions with Mr.
Robert Amts concerning the natural hydrocarbon issue. Also, Mr. William
Lonneman was very helpful with the hydrocarbon analyses. Dr. Marcia Dodge
is also acknowledged for her helpful discussions on modeling as is Dr.
Leslie Hull for the product identification for ozonolysis of terpenes.
I'm also indebted to Northrop Services for the use of their data on
the smog chamber studies of natural hydrocarbon and to Ms. A. McElroy for
typing the many iterations of this document.
viii
-------
SECTION 1
INTRODUCTION
The involvement of hydrocarbons (HC's) in the formation of photochemical
smog is well documented (1-4). Also generally accepted is that both the
type and the amount of HC are important in the formation of photochemical
smog. If ozone is taken as the indicator for smog formation, at a favorable
hydrocarbon to _NO ratio (HC/NO ) olefins are found to produce ozone
X X
most quickly, followed by the substituted aromatics and then the slow-
reacting paraffins. However, if reduced visibility (or aerosols) is taken
as the photochemical smog indicator, then the substituted aromatics and
the olefins of higher molecular weight (C, ) are the largest contributors.
Natural HC's, i.e., isoprene and the monoterpenes, are olefinic com-
pounds that are expected to produce both ozone and aerosol. The question
then arises, can natural HC's be a significant source of ozone and visibility
reduction in rural areas? Also, does vegetation contribute to photochemical
air pollution problems that exist in most large metropolitan regions?
In this paper we would like to critically review the relevant literature
concerning natural HC's and their role in pollution problems observed in
urban and rural areas. The HC's to be considered are the monoterpenes
(CnnH,,) and isoprene (C_HQ). Other natural HC's, such as methane (CH.)
J.U J.O jo *»
and stress-evolved ethene (C«H,), will not be considered, since CH, is
thought to be unreactive and C^H, emissions are too small to be significant
(5-7). The specific topics covered are: (1) HC emissions, (2) reactivities
of natural HC's, (3) product formation, and (4) source-receptor relationships.
-------
SECTION 2
CONCLUSIONS
Although additional information is needed in order to firmly establish
the role of natural HC's in ozone formation, the present data suggest
emissions from vegetation are not likely to have significant effects on
air quality. All gas-phase HC analyses and carbon aerosol studies suggest
that natural HC's do not exist at sufficiently high concentrations to
affect air quality in either urban or rural areas. Emissions data do not
agree with ambient air measurements. Those measurements suggest that the
8 6
previously published value of 9 x 10 ton/yr needs to be lowered to 10 -
10 ton/yr.
Isoprene and the monoterpenes produce ozone when photooxidized in the
presence of NO . However, they are not efficient ozone producers since
2t
a large amount of carbon is consumed in the production of aerosols.
Because of the aerosol production, the natural HC's may contribute to
visibility reduction in pristine areas.
-------
SECTION 3
AT
RECOMMENDATIONS
Additional work in the area of vegetative HC emissions should include:
(1) Establishment of better emission values. In-situ vertical profile
measurements are recommended over enclosure techniques. The enclosure
and simulated environmental chamber techniques suffer from possible
pertubation of the normal environment, which results in unusual
emissions. Also, greater understanding and better measurement of the
biomass factor is needed in the enclosure techniques.
(2) Better collection and analyses of gases and aerosols in rural and
urban areas are needed. These improvements should include a measure
14 12
of the C /C ratio for the organics, especially for the fine par-
14 12
ticulate fraction. The C /C ratio in living matter is not the same
as that found in petroleum and coal (8, 9), so one can measure the
natural versus fossil carbon contribution to the atmosphere by
determining this ratio.
(3) Additional smog chamber studies are needed in order to determine the
ultimate fate of the photooxidation products of natural HC's. Previous
studies in this area have met with only limited success and have only
sketchily defined the mechanism for terpene photooxidation. Until
all products are identified, the mechanism will remain this way.
(4) Modeling exercises are needed in order to establish the importance of
natural HC's to urban and rural air quality. Various emission sceneries
could establish the importance of natural versus anthropogenic emissions.
These computations would then be compared to observed air quality data
in order to assess the emission scenerio most compatible with air
quality data.
-------
(5) Finally, we must understand why uninhabited areas (e.g., the Smoky
Mountains) have reduced visibility. The data in the area show little
or no carbon in the fine-particulate fraction but yet reduced visibility
occurs, and was apparently occurring long before the industrialization
of America. Understanding reduced visibility in the mountain areas
is necessary if we are to understand the role of natural HC's to air
quality.
-------
SECTION 4
HC EMISSIONS
Total emission calculations of terpenoid HC's were first made by Went
(10). The actual computation of this often-cited work is based on data
supplied by A.J. Haagen-Smit (through a private communication with Dr. Went).
Haagen-Smit analyzed air that was passed over shrubs of Artemisia tridentata
(Basin sagebrush). After combustion and transformation of the HC to C0~,
-1 -2
Haagen-Smith calculated that approximately 50 kg day km of emissions
o
resulted. From this work, Went extrapolated that 1.75 x 10 ton/yr of
volatile natural HC is liberated worldwide.
The numerous assumptions in Went's calculation include that: the
average number of emission days is 100 per/year; the vegetation covers 2 x
5 2
10 km in the western United States; coniferous and hardwood forest produce
approximately the same amount of volatile organic matter; and, the earth's
7 2
total vegetative area is 10 km . Obviously, Went's number rests upon few
measured parameters. Also, the basis of all Went's calculations, the
Haagen-Smith work, is not given in detail, which makes validation of the
initial number difficult. For example, if the air Haagen-Smith used to
fumigate the vegetation contained significant background HC, then the
estimated emissions level from the number is grossly in error, since the
change in C02 would be small upon combustion. If the change in C02 were not
used in the calculation, then the emission would be too large.
More systematic research of HC emissions was performed by Rasmussen
and Went (11), who studied several locations in the United States and one
in Holland. Ambient air samples, as well as plants with plastic containers,
covered were analyzed. The HC's identified by retention times on the
gas chromatograph (GC) include; isoprene, ot-and 3-pinene, limonene, and
3
myrcene. Additional hydrocarbons, A -carene, myrcene and p-cymene, were
-------
tested for retention times, but Rasmussen and Went were not explicit in
stating if those HC's were detected in the ambient air. The direct measure-
ments of volatile organics in the atmosphere (presumably the summation of
the terpenes and isoprene) showed an average concentration of 10 ppb
(presumably V/V). This value was then assumed to exist in a 2-km column
of air, with a yearly production rate of 270 days. Using an estimated,
18 2
vegetated land area of 10 cm , the total world production was calculated to
8 7
be 4.38 x 10 ton/yr. The investigators also calculated 20-40 x 10 ton/yr
by the enclosed plastic container technique, which makes the two above
methods agree within a factor of roughly two, and suggests that the
earlier estimate of Went is too low.
The work of Arnts et al. (12) that will be described later does not
substantiate the 2-km column height containing 10 ppb terpene. In fact,
the terpene concentration was found to decrease exponentially with height.
At 10 m above a tree canopy, the terpene concentration was determined as less
than 30% of that found just at the canopy height. Considering these
factors, the calculations made by Rasmussen and Went would be reduced by
Q
a factor of over 2000. Thus, the 4.38 x 10 ton/yr would be reduced to
2.2 x 10 ton/yr.
In another study, Ripperton et al. (13,14) found the a- and g-pinene/
5 1 1 5
ozone reactions to be very fast (0.8 x 10 liter mole sec and 0.25 x 10
liter mole" sec ). The authors reasoned that ozone may have destroyed a
portion of the emitted terpenes and concluded, because of the ozone/terpene
reaction, that the original estimates of Rasmussen and Went should be
multiplied by a factor of 2 to 10.
Although these calculations by Ripperton et al. are qualitatively correct,
the two-to-ten fold increase in terpene concentration is difficult to accept.
Since the terpene concentrations were measured within the forest canopy, any
effect from ozone should not be large, and the levels of terpene observed
should be close to the true concentrations emitted without reaction. If
for example, the ozone concentration is approximately 30 ppb in the
forested area (15), then the half-life is approximately one day for isoprene
and a few hours less for ot-pinene. (This calculation assumes a constant
6
-------
reservoir of ozone from a pseudo first-order reaction, though ozone is
actually depleted near the ground by both terpene reaction and dry deposition.)
The OH concentration is also assumed to be zero due to minimal sunlight
penetration into the canopy. The correction for ozone effects made by
Ripperton et. al. is in fact probably no more than a few percent, if 30 ppb
of 0» and 10 ppb of terpene are involved, which falls well within the
errors contained in the original calculations of Rasmussen and Went.
Rasmussen (16) in a later study measured several terpenes emitted from
various types of deciduous and coniferous vegetation. These HC's were
a, and 3-pinene, camphene, limonene, myrcene, and 3-phellandrene, isoprene
was also found to be emitted from hardwoods. Emission rates were measured
by placing an undamaged portion of the plant sample in a glass bell jar.
The HC buildup in the bell jar was then measured at several levels of
radiation exposure (50 to 1200 ftc) and at two different temperatures (17
and 30-32 C). Coniferous types of vegetation were found to emit terpene
independently of light, while deciduous plants emitted isoprene in a
linear dependence upon light intensity. Alpha-pinene emissions from
conifers were found to increase by a factor of 4 to 7.5 when the temperature
was increased from 17 C to 30-32 C.
Rasmussen further gives the percentage of hardwood and softwood
forest in the United States. From the foliage per acre given by others
(17-19), global natural HC emissions are given. Rasmussen estimates from
his emission rates that a 10-cm foliar canopy would result in a worldwide
natural HC level of 23.4 x 10 metric ton/yr. A 50-cm thick canopy would
result in 117 x 10 metric ton/yr; 75 cm in 185 x 10 metric ton/yr; and
finally 200 cm in 464 x 10 metric ton/yr. The abstract of the paper
suggests that the 75-cm canopy value of 175 x 10 metric ton/yr is the
most reasonable. We have been unable to determine from the manuscript his
method of extrapolating from the individual emission rates of the clipped
branches to worldwide emissions or the validity of this method.
-------
In successive studies, Tyson et al. (19) and Dement et al. (20)
measured emissions from the sagebrush, Salvia mellifera. This shrub is
distributed throughout the coastal region of California and covers approx-
imately 45% of the site studied. The investigators found that the emission
levels of terpene were over an order of magnitude lower than the findings
of Went (10), i.e., 3 kg day"1 km2 versus Went's 50 kg day" km" . In the
Dement et al. study, the same sagebrush was investigated under light and
dark conditions, as well as different relative humidities and temperatures.
These investigators found that the emission rates were dependent upon leaf
temperature, but were independent of light. They also found that relative
humidity was a factor in the emissions. Thus, emissions almost doubled
(3.1 yg/dm versus 5.8 yg/dm ) when the air dew point was 22°C compared to 0°C.
Also determined was that the level of HC an excised branch will liberate is
2 2
more than twice that of an uncut branch (13.1 yg/dm versus 5.8 yg/dm , at
D.P. = 22°C). This latter finding is interesting in that it suggests the
Rasmussen value obtained on cut branches overestimates HC emissions.
The Research Triangle Institute (RTI) in 1974 (21) measured natural
emissions of gaseous organic compounds in the Ohio area. In this report
worldwide emissions were also calculated. The authors suggest that the
often cited work of Went (10) and Rasmussen and Went (11) have serious short-
comgings in that only terpenes and their derivatives are considered; the RTI
report lists a much larger number of organic compounds liberated by veg-
etation. The natural emissions estimate of RTI is based on the carbon
fixation process, assuming that 10% of the available carbon goes to hydro-
carbon emissions. By this estimate, Rasmussen's emission estimates are too
low by a factor from <-wo to ten. The total emissions calculated by RTI
10 9
are 0.4-1.2 x 10 ton/yr, with an average of 8 x 10 ton/yr.
We have difficulty evaluating RTI's estimate, since it considers all
organic compounds including the oxygenates. Also, the original estimate by
g
Went of 1.75 x 10 ton/yr was, as stated earlier, based on the Haagen-
Smit work, in which air was passed over sagebrush and all the organic com-
pounds were analyzed. Therefore any other organics liberated by the sage-
brush would also have been oxidized to yield C0?. Again, the work of
Rasmussen and Went is probably in error since a uniform HC column is assumed.
-------
Furthermore, the 10% emission of the C02 fixation suggested by RTI is not
verified. As stated in the RTI report, "ten percent was an estimate which
appeared to be reasonable." The ten percent may be reasonable, but two
points are highly unlikely: (1) that all the material expired by the plants
is necessarily volatile, and (2) that all the EC's would necessarily par-
ticipate in photochemical smog production, i.e. saturated low-molecular-
weight HC's.
The most extensive work on the measurement of emission rates has been
performed by Zimmerman (22, 23). His measurement technique involves an in-
situ enclosure of a portion of the tree or foliage sample with a Teflon
bag. The bag is partially emptied and then flushed with metered volumes
of air. The organic compounds are collected in canisters and returned to
the laboratory for detailed analyses.
In these studies, Zimmerman not only measured emissions from various
types of vegetation, but also examined the different plant species in
different parts of the United States. The emission rates of some of the
plant species, as given by Zimmerman, are shown in Table 1. Table 2 shows
the major emissions from selected vegetation. In Figure 1, the structures
of selected natural HC's are illustrated. Zimmerman further divides the
emission rates into different biotic regions according to Dasmann (24). He
sub divides these regions into latitudinal sectors for average sunlight and
temperature. Drawn from these assumptions, Table 3 summarizes regional and
total global emissions. The total worldwide emissions level of isoprene
1 / Q
plus the terpenes is 8.3 x 10 pgC/yr (9 x 10 metric ton/yr.). This
Q
number agrees reasonably well with the 4.38 x 10 ton/yr value obtained by
Rasmussen and Went (11).
The agreement of the emissions level for several species measured by
different experimenters appears reasonable. For example, Tingey (25) measured
emissions in an environmental chamber that enabled control of the light
intensity and temperature. With live oak, Tingey found that the emissions are
approximately 30 yg g~ nr~ . For slash-pine, Tingey obtained an average
emission of approximately 5 yg g~ hr~ . Zimmerman gives an almost identical
value of approximately 6 yg g~ hr~ . For loblolly piner Arnts et al. (12)
-------
// \v_ Y iv ^ x > > />-\ ff
ISOPRENE p-CYMENE « - PINENE 0-PINENE d - LIMONENE
MYRCENE CAMPHEIME TERPINOLENE A -CARENE OCIMENE
-CH
a - PHELLANDRENE |8 - PHELLANDRENE a -TERPINENE "Y - TERPINENE
Figure 1. Structure of some selected natural hydrocarbons.
-------
found an emission of 3 yg g hr at approximately 30°C in a North Carolina
forest. The Arnts group measured a vertical profile of ot-pinene over the
forest, and by employing an energy balance equation (i.e., by measuring
temperature, water vapor, carbon dioxide, and net solar radiation), they
were able to calculate a-pinene flux values. The vertical gradient measure-
ments for the a-pinene were difficult and could be subject to several errors.
Assumptions of smooth terrain and constant wind fetch dimensions were also
required.
Zimmerman, using the bag enclosure technique in North Carolina,
obtained a value of 4.8 yg g hr for loblolly pine, which agrees favorably
with the Arnts value. However, a reassessment of the Arnts study suggests
that the proper wind fetch may not have been used. This error arose from
a study of the H_0 and terpene vertical profiles.
TABLE 1. EMISSION RATES FOR SELECTED SAMPLES AT VARIOUS LOCATIONS
Sample Type * Emission Rate
(yg g hr~ )
Raleigh, NC (June 1977)
Oaks
Shortleaf Pine
Loblolly Pine
Red Cedar
Virginia Pine
Pullman, Washington (August - November 1976)
Ponderosa Pine
Hugo Pine
Douglas Fir
Juniper
Spruce
Pine litter
Tampa, Florida (April - May 1977)
Pasture
Laurel Oak
Turkey Oak
Water Oak
Blue Jack Oak
All Oak species (night)
26.1
16.3
4.85
1.14
13.6
2.96
1.78
0.86
3.25
7.26^
132.0
*
288.6
11.2
26.2
27.2
16.5
1.20
« 2
"The pine litter and pasture emissions are given in yg/m .hr. In ojjder to
convert the emissions shown for the vegetation from yg/g.hr to yg/m .hr, the
biomass factors must be known. For the Raleigh, NC in June, the loblolly pine
forest conversion factor is 709 yg/m hr = 1 yg/g hr.
11
-------
TABLE 2. MAJOR EMISSIONS FROM SELECTED SPECIES
Vegetation Type
Major Emissions
Pullman, Washington
Ponderosa Pine
Lombard! Poplar (daytime)
Douglas Fir
Santa Barbara, California
Eucalyptus (daytime)
Manzanita
Chemise
Research Triangle Park, NC
Dogwood
Yellow Poplar
American Sycamore (daytime)
Eastern Red Cedar
Research Triangle Park, NC
Loblolly Pine
Shortleaf Pine
Virginia Pine
Tampa, Florida
All oaks (daytime)
36% A -carene
25% f?-pinene
14% a-pinene
99% Isoprene
24% a-pinene
7% d-limonene
5% g-pinene
40% Isoprene
18% Unknown 1
25% Unknown 2
10% A3-Carene
10% d-limonene
8% Unknown 3
5% Terpinolene
74% Isoprene
52% A -carene
26% a-pinene
35% a-pinene
26% d-limonene
19% g^pinene
11% A -carene
53% a-pinene
17% d-limonene
12% g^pinene
11% A -carene
26% Unknown 4
22% a-pinene
10% g-pinene
8% dglimonene
2% A -carene
90-99% Isoprene
12
-------
TABLE 2. (Continued)
Vegetation Type Major Emissions
Long Leaf Pine 30% a-pinene
30% 3-pinene
8% Propane
Slash Pine 27% a-pinene
19% d-limonene
16% B^pinene
12% A -carene
Sand Pine 44% a-pinene
352 B-pinene
Australian Pine (daytime) 92% Isoprene
Saw Palmetto (daytime) 85% Isoprene
Sabal Palmetto (daytime) 90% Isoprene
Cypress 46% a-pinene
21% A -Carene
Sweet gum 51% B-Phellandrene
20% Isoprene
13% a-pinene
The sum of the emissions is not 100% since many small GC peaks were
present and were not identified. Major unknown emissions are given
with an assigned number.
The a-pinene levels determined by Arnts et al. (see Figure 2) should
decrease linearly with the H^O. However, a-pinene decreases with altitude
above the canopy at a much higher rate. If one assumes that the H-0
measurements are correct, then the rapid decrease in terpene concentration
can be a result of either a-pinene reactions with 0,, or insufficient wind
fetch. Since the diffusion time of a-pinene for such small vertical
distances is short, the terpene could not react significantly with 0^.
One must conclude that the wind fetch was improper for suitable measure-
ments, which indicates the Arnts flux value for a-pinene is too large.
The apparent agreement between the Arnts and Zimmerman studies suffers
from another incompatability as well. Arnts found that 70-80% of the
-------
5
Q
IIP I I I I I jl I I I
C09 a-PINENE
I I I I I I I I I
TEMPERATURE
I I I I I I I
I I I I I
a>
o
O
O
u- 15 -
LU
oc
o
LL
LU
O
CO
LU
U
5 -
31.2
TEMPERATURE,
WATER VAPOR
PRESSURE, mb
A CO2, ppm
150 250 350
a - PINENE, jug/m3 ' WIND SPEED, cm/sec
Figure 2. Loblolly pine forest vertical gradients, 7/18/77, 1035-1100 hrs e.s.t.
-------
loblolly emissions are due to a-pinene. Zimmerman suggests 35%. The
findings by Arnts are more compatible with the work of Mirov (26) in which
the reported gum turpentine had a composition of 71% a-pinene and 22% ct-
pinene. If we assume the emissions to be proportional to the composition,
the pine should liberate higher levels of a-pinene. Perhaps the disagree-
ment in emission percentages rests on the two measurement techniques;
Rasmussen (2 7) found the emissions from Mango foliage depend on the type
of enclosure. Unenclosed, very little isoprene was liberated and the
isoprene/n-butane ratio was quite small. However, when sheltered, the
isoprene/n-butane ratio was greater, with isoprene increasing some forty-
fold.
RasmusSen (28) recently suggested that the bag enclosure technique may
not be a satisfactory way to measure emissions at least for deciduous-type
vegetation. His unpublished work has shown that when a cylinder is placed
over a tree branch and the emisisons are measured dynamically, the emissions
levels are much lower. He suggests that perhaps the emissions are over-
estimated by as much as a factor of 10 with the bag-enclosure technique.
If the Rasmussen findings prove correct, then the Zimmerman worldwide
Q
emission factor of 9 x 10 ton/yr must be revised downward. The Rasmussen
position is consistent with the observations of Lonneman et al. (29), who
calculated that at best only 5% of the total nonmethane hydrocarbon (TNMHC)
in the Tampa St. Petersburg area could be attributed to natural emissions.
In contrast, Zimmerman (30) calculated that 68% of the TNMHC is due to
natural sources,a percentage based upon accurate anthropogenic emissions.
Of course, the above emission estimates could be in error. If the
Lonneman estimate is correct for the Tampa St. Petersburg area, then the
worldwide emission of natural HC's (isoprene and the terpenes) is not
9 x 10 ton/yr as suggested by Zimmerman, but closer to 10 -10 ton/yr.
This value is smaller by a factor of 10-100 than that originally proposed
by Went, but is nonetheless more compatible with ambient air analyses.
We shall see in a later section that the emission rates must be lower than
those given by Zimmerman, since air quality data do not substantiate high
emission values.
15
-------
TABLE 3. ESTIMATED U.S. AND GLOBAL EMISSIONS BY LATITUDE (gC/yr)
Latitude U.S. Isoprene Terpenes
45-50° 1.8 x 1011 4.6 x 1012
40-45° 5.8 x 1011 1.0 x 1013
35-40° 4.3 x 1012 1.6 x 1013
25-35° 1.5 x 1013 1.9 x 1013
Global (temperate) 9.6 x 1013 2.3 x 1014
Global (tropical) 2.5 x 1014 2.5 x 1014
Total Global 3.5 x 1014 4.8 x 1014
16
-------
SECTION 5
REACTIVITIES OF NATURAL HYDROCARBONS
The suggestion has been made that natural HC's are not observed in
high concentrations in the atmosphere since they, being mostly olefinic
react very quickly with the OH and 0., usually present in the ambient air
(31, 32). Qualitatively this premise is correct. Biogenic HC's, like all
organic compounds (including methane), will participate in OH ractions. If
the compounds are olefinic, they will also participate in 0- reactions.
Rate constants for some selected natural HC's are shown in Table 4.
Went (10) suggested that natural HC's react with ozone in the
atmosphere to produce aerosols since a blue smoke appeared when ozone and
crushed pine or fir needles were placed into a jar. In a later publication,
Went suggested that a large portion of condensation nuclei present in the
atmosphere is a direct result of photochemical reactions similar to those
responsible for the production of photochemical smog. Stephens and Scott
(39) studied the photooxidation of monoterpenes, specifically pinene and
a-phellandrene, when investigating the relative reactivities of a number of
HC's. The found a-phellandrene to be very reactive at a rate slightly
greater than tetramethyl-ethylene. The pinene was about six times slower
in reactivity, but slightly faster than isobutene. The pinene isomer
studied is not given.
Ripperton et al. (14) studied the a-pinene/ozone reaction, in which
they noted the production of aerosols. Although a second-order rate
constant is given (0.24 ppm~ min~ ), the experimenters found that much
more a-pinene reacted than ozone. In two replicate runs, the a-pinene/0.,
consumption ratio was approximately 2/1 and over 3/1, respectively. The
work suggested that a reactive product is produced (free radicals) that
will react further with the a-pinene.
17
-------
TABLE 4. RATE CONSTANTS FOR SELECTED HC's
Compound Reactant Rate Constant (ppm~ min ) Reference
n-Butane
Propylene
Isoprene
Trans-2-Butene
ot-Pinene
3-Pinene
d-Limonene
Myrcene
3-Carene
p-Cymene
a-Phellandrene
g-Phellandrene
o
OH
°3
j
OH
°3
OH
o
3
OH
°3
J
OH
°3
J
OH
°3
J
OH
°3
OH
°3
^i
OH
°3
OH
°3
OH
°3
OH
slow
3.5 x 103
1.6 x 10~2
3.8 x 104
1.9 x 10~2
1.2 x 105
0.39
1.0 x 105
2.1 x 10"1
1.2 x 105
5.8 x 10"2
9.8 x 104
0.96
2.1 x 105
0.18
3.4 x 105
0.18
1.3 x 105
6 x 10'7 *
2.3 x 104
1.7
2.6 x 10"1
1.7 x 105
NBS (33)
NBS (33)
NBS (33)
Arnts (34)
Winer (35)
NBS (33)
Atkinson (36)
Grimsrud (37)
Winer (35)
Grimsrud (37)
Winer (35)
Grimsrud (37)
Winer (35)
Grimsrud (37)
Winer (35)
Grimsrud (37)
Winer (35)
Pate (38)
Winer (37)
Grimsrud (35)
__
Grimsrud (35)
Winer (37)
*
This rate constant is for p-xylene but p-cymene should have a rate
constant very similar to xylene.
Lillian (40) also studied a-pinene reactions, but in the presense of
NO . He found that irradiated a-pinene/NO mixtures produced the same
X X
characteristics as photochemical smog, i.e., the reaction of HC's, the
18
-------
oxidation of NO to NO-, and the production of ozone. Alpha-pinene was also
found to act as both a source and a sink for ozone, and to be a source of
light-scattering aerosols.
The most complete study on the reactivities of monoterpenes is the
work of Grimsrud et al. (37). These investigators studied the reactions
of natural HC's with ozone and with NO . The compounds investigated were-
X
p-menthane, p-cymene, a- and $-pinene, isoprene, 3-carene,a - and 3-
phellandrene,y -terpinene, carvomenthene, limonene, myrcene, cis-ocimene,
terpinolene, and qt-terpinene. Isobutene was also studied as a reference
organic compound.
In general, compounds with high ozonolysis rates also have very high
photooxidation rates. Grimsrud et al. found that all the HC's except
ocimene and terpinolene showed 1:1 stoichiometry with ozone. Ocimene is a
triolefin and ozone is expected to react with more than one double bond.
Terpinolene showed 1:2 stoichiometry due to the presence of two double
bonds, but only at high 0_/HC ratios.
The photooxidation of four C-_ terpenes in the presence of NO was
investigated by Westberg (41). The four terpenes were: limonene, a-pinene,
terpinolene, and ocimene. In each compound the amount of ozone produced
was found to be dependent upon the HC/NO ratio. Westberg's data are
A
shown in Figure 3 for the two levels of NO . The curves show that the
x
optimum ozone varies somewhat for different terpenes. The maximum ozone
is produced between the 13/1 ratio for terpinolene (NO = .05 ppm) and the
ji
28/1 ratio for limonene (NO = 0.20 ppm). Higher ozone levels were observed
-------
300
T
ro
o
200
.0
Q.
Q.
X
<
o
N
O
100
LIMONENE I
LIMONENE II-
NOX = 0.05 ppm
NOV = 0.20 ppm
TERPINOLENE I- NOX = 0.05 ppm
TERPINOLENE II- NOX = 0.20 ppm
a-PINENE- NOX = 0.20 ppm
10
20
HC/NO,
30
40
Figure 3. Ozone prediction as a function of HC/NOX.
-------
ro
/ /
CHAMBER CONDITIONS:
2S°C
48 SO 60 n 80
C/NOX RATIO, PpmC VS ppm NOX: (N0x)o = 0,33 ppm
7 t
Figure 4. Effect of hydrocarbon to NOX ratio on ozone maximum.
-------
Arnts and Gay (34) investigated the reactivities of six C-jgH-g mono-
terpenes, a C-_H.._ aromatic, and a C,-Hg diolefin. The terpenes were
myrcene (acyclic triolefin), d-limonene and terpinolene (monocyclic
diolefins), a- and g-pinene, and A -carene (bicyclic monolefins). The
aromatic studied was p-cymene (p-isopropyl toluene), while the C,. diolefin
was isoprene (2-methyl-l,3-butadiene). Propylene was also studied because
this olefin serves as a reference for the other compounds, and is often
used to test chamber reactivity compound. The monoterpenes were found to
oxidize NO to NO. at a rate equal or faster ttanpropylene. The aromatic
p-cymene was very slow in the NO conversion. The monoterpenes did not
permit the buildup of ozone, which was very noticeable at the higher HC/NO
X
ratio (Figure 4). Propylene and isoprene were found to be more efficient
in producing ozone than the C-n terpenes. These compounds also showed
some ozone suppression at the higher HC/NO ratios. Para-cymene showed
X
almost no HC/NO dependence above a HC/NO ratio of 3 and almost no reaction
X X.
with 0». Ozone suppression is probably caused by the very rapid reaction
of 0. with the monoterpenes.
With the exception of p-cymene, the maximum 0. shown in Figure 4 is
observed to occur between 10 to 20. This result agrees well with Westberg
(41). The probable causes for the peaking of 0- are: (1) at the initial
low HC/NO , the NO acts as an 0_ scavenger and quickly titrates any 0,
X J j
produced with excess NO, and (2) at high HC/NO , the excess HC reacts very
X
quickly with 0_, leaving insufficient NO to generate more 0_.
O X j
Lifetimes of some HC's at ambient air concentrations of OH and 0« are
given in Table 5. As shown in this table, many natural HC's are expected
to oxidize quickly, along with many of the manmade HC's. At night, the
OH radical concentration is expected to drop to zero since these radicals
are light dependent. Also during nighttime hours, ground base temperature
inversions decouple the surface ozone from layers aloft. Ozone concentration
consequently decreases even to zero near the surface from deposition losses
and chemical reactions. The natural HC's emitted during nighttime are thus
expected to persist until the following day. However, isoprene is not emitted
at night and monoterpene emissions are minimal at night, since the temperature
22
-------
21
18
15
12
9
6
3
0
1,2.4-TRIMETHYL BENZENE
d-LIMONENE
mjj-ETHYL TOLUENE
I I
" I
ro
30
25
20
15
10
5
CYCLO PENTANE (ISOPRENE)
9T 2-METHYL PENTANE
I I I
0000 0200
0400 0600
0800
1600
1800 2000
1000 1200 1400
TIME, hour of day
Figure 5. Diurnal variation of hydrocarbon compounds, Chickatawbut Hill, July 18,1975.
2200
2400
-------
ro
60
s"
u
f
§«
CO
oc
30
<20
0000
0300
AUGUST 23,1976
SEPTEMBER 8,1976
FEBRUARY 7,1977
MARCH 17,1977
0900
1200
TIME, hr
1500
1800
2100
2400
Figure 6. Total concentration of the C-JQ terpene hydrocarbons (ppb C) observed in the canopy
at the IBP site at different times of the year.
-------
f\>
en
20
18
00
cc
16
14
a
^ 4
a
0000
0300
0600
0900
1200
TIME, hr
1SOO
SEPTEMBER 8,1976
.
FEBRUARY?, 1977
MARCH 17,1977
1800
2100
2400
Figure 7. Total concentration of the CIQ terpene hydrocarbons (ppb C) observed above the
canopy at the IBP site at different times of the year.
-------
is lower. Table 5 shows that some natural HC's can be expected to survive
the attack of OH and 0., long enough to be observed in the ambient air,
especially p-cymene. However, the amount of natural HC's observed in
ambient air analyses is very low.
Data obtained in a number of cities, including Tampa/St. Petersburg,
Miami, Tulsa, Houston, and Boston, show practically no C-_ natural HC's
and very little isoprene (see Table 6). Similar low concentrations of natural
HC's were obtained by Arnts and Meeks (43) in the Smoky Mountains of
Tennessee and in Rio Blanco, Colorado. Arnts and Meeks found a maximum of
5.7 ppbC isoprene in the Smokies when TNMHC was 102 ppbC; isoprene was found
at 5.6 ppbC in Rio Blanco when TNMHC was 117 ppbC. No a-pinene or other
terpenes could be detected in the Rio Blanco samples. In the Smoky Mountain
samples, p-xylene and a-pinene were not separated. However, their combined
concentration was only a maximum of 2.5 ppbC when the TNMHC was 171 ppbC.
Some diurnal variations of natural HC's are shown in Figure 5,6 and 7.
Figure 5 illustrates the data obtained at Chickatawbut Hill southwest of
Boston. Red oak trees, which are known isoprene emitters, grow in the
area surrounding this site. The isoprene in Figure 5 is shown with cyclo-
pentane since GC separation did not distinguish between these two HC's.
However, since roadway samples show 3/1 ratio between cyclopentane and 2-
methyl-pentane, the assumption is reasonable that the increases in cyclo-
pentane/ isoprene concentration are mostly du.i to increases in isoprene
especially since the increases occur during daylight hours when deciduous-
type vegetation is active.
Figures 6 and 7 show the diurnal variations of the C-n terpenes found
at a loblolly pine forest in North Carolina. The Figure 6 data were taken
within the loblolly forest, while the Figure 7 data were obtained 5 m above
the canopy. These figures contrast with the hardwood forest at Chickatawbut
Hill in that the terpene concentration is highest at night. Noctural
inversions serve to contain natural HC's within the canopy, and pine
emissions are not directly affected by the presence or absence of sunlight.
Also shown in these figures is the large difference in concentration
between summertime and wintertime. Of note is that just 5 m above the
26
-------
canopy the natural HC's dropped to one third of the concentration within
the canopy. On August 23 of the study, the biogenic contribution to the
TNMHC was 40% in the canopy. At 5 m above the canopy, the biogenic con-
tribution is just 12%.
TABLE 5. HALF-LIFE IN HOURS FOR SELECTED HC's AT VARIOUS OZONE CONCENTRATIONS2
Ozone Concentration (ppb)
Compound
n-Butane
Propylene
t-Butene-2
Isoprene
a-Pinene
d-Limonene
Myrcene
3-Pinene
3-Carene
p-Cymene
g-Phellandrene
Toluene
m-xylene
0
165
15.2
5.61
4.81
4.81
2.75
1.72
5.88
4.44
25.1
3.39
63.5
16.5
30
Same
9.31
0.86
4.16
1.32
0.35
0.19
3.26
1.4
Same
1.03
Same
Same
50
Same
7.40
0.55
3.40
0.89
0.22
0.12
2.37
0.99
Same
0.70
Same
Same
80
Same
5.66
0.36
2.96
0.60
0.14
.09
1.75
0.68
Same
0.47
Same
Same
100
Same
4.80
0.29
2.68
0.49
0.12
0.07
1.48
0.56
Same
0.39
Same
Same
150
Same
3.65
0.19
2.22
0.39
.09
0.05
1.08
0.39
Same
0.27
Same
Same
200
Same
2.92
0.14
1.86
0.26
0.07
0.03
0.85
0.29
Same
0.21
Same
Same
a 8
OH concentration = 2 x 10 ppm
A summary of the ambient air measurements is given in Table 7, showing
that the highest values are recorded by Rasmussen and Went (11) and by
Whitby and Coffey (46). High-resolution GC columns were not available at
the time of the Rasmussen and Went (11) study, and many unidentified GC
peaks were assumed to be natural HC's in the Whitby and Coffey report. They
made this assumption largely because the samples were collected in farmland
areas. Indeed, if both of these studies are excluded from the table,
the average isoprene concentration measured in ambient air lies between
0.1 and 10 ppbC, while the predominate monoterpene, a-pinene, lies between
27
-------
TABLE 6. HC's FOUND NEAR SEVERAL CITIES (PPB V/V)
Compound
Ethane
Ethylene
Propane
Acetylene
Isobutane
n-Butane
Propylene
Isopentane
n-Pentane
Isobutylene
T-Butene-2
C-BuT-2-Butadiene
Pentene-1
2-Mebutene-l
T-Pentene-2
C-Pentene-2
2-Mebutene-2
Cyclopentane + Isoprene
2-Mepentane
3-Mepentane
4-Mepentene-2
n-Hexane
Hexene-1
T-Hexene-3
2,4 Dimepentane
Mecyclopentane
CIS-Hexene-2
Unknown
3,3 Dimepentane
Cyclohexane
2 Mehexane
2,3 Dimepentane
3 Mehexane
1C3 Dimecypentane
2,2,4 Trimepentane
1T3 Dimecypentane
n-Heptane
Mecyclohexane
Toluene
Unknown
Unknown
Nonane
Unknown
Sub-
urban
Tampa/St.
Petersburg
(5-14-76)
3.0
3.8
3.1
4.6
2.5
7.5
0.9
9.0
4.0
1.3
0.6
1.3
0.1
0.3
5.1
3.0
1.8
0
1.7
0.1
0.3
1.9
-
1.6
1.3
1.1
4.3
7.2
15.1
16.4
2.8
N. Sub-
urban
Miami
(5-16-76)
5.7
10.5
19.5
11.3
5.5
16.4
22.3
10.2
2.6
2.2
0.7
1.8
2.1
-
2.8
6.5
4.0
0.1
3.1
0.1
0.1
.2 -
1.5
1.5
1.2
6.1
0.6
0.4
0.4
Tulsa
(7-27-78)
11.6
11.5
15.5
8.4
8.1
16.5
4.2
17.3
12.1
1.2
1.6
0.8
0.6
-
-
-
3.6
6.3
4.5
-
6.1
0.41
5.1
-
-
-
1.8
5.0
3.3
5.7
1.02
4.6
2.02
6.12
6.68
16.8
1.9
__
Houston
(4-2-74)
41.8
78.2
63.9
7.7
39.7
71.1
27.9
77.4
42.8
21.0
1.8
2.5
1.3
2.1
9-. 6
1.7
1.6
36.5
*
24.9
0.0
27.9
1.1
12.4
20.5
7.8
3.5
10.2
14.7
13.3
15.8
90.5
6.6
~._
S . Sub-
urban
Boston
(7-31-75)
3.6
INT
6.2
4.2
4.1
9.6
1.4
11.6
6.3
1.9
0.9
1.6
0.6
trace
trace
2.3
6.3
4.2
4.7
0.2
0.5
2.5
0.7
2.3
1.0
1.8
1.0
1.5
0.6
24.2
3.3
«.«.
-------
TABLE 6. (CONTINUED)
Sub-
urban N. Sub- S.Sub-
Tampa/ St. urban urban
Petersburgh Miami Tulsa Houston Boston
Compound (5-14-76) (5-16-76) (7-27-78) (4-2-74) (7-31-75)
Ethylbenzene
P-Xylene
M-Xylene
Unknown
a-Pinene
0-Xylene
Unknown
Isopropyl benzene
N-Decane
g-Pinene
N-Propylbenzene
M-Ethyltoluene
A-Carene
1,3,5 Trimebenzene
0-Ethyltoluene
1,2,4 Trimebenzene
Unknown
D-Limonene
p-Diethylbenzene
1.4
1.2
3.3
0.5
4.1
1.4
0.6
1.4
0.8
0.9
1.4
2.6
0.5
_,.
__
1.0
0.4
__
0.6
1.4
2.59
1.6
5.3
_
__«
3.3
__
2.3
«
3.3
__
1.92
2.1
13.0
28.2
36.7
14.6
__
28.5
11.2
___
2.7
10.6
__ '
4.3
1.0
11.6
4.0
__
5.9
2.3
2.3
7.3
4.8
2.0
1.1
_ _
_«.
0.8
3.1
«_
1.0
0.3
6.0
__
__
*
Measured with 3-methyl pentane
less than 0.1 to 14 ppbC. It is not clear why Cronn and Harsch (51)
3
observed such high concentrations of A -carene in the Smoky Mountains.
From Table 2, it is noted that ponderosa pine, dogwood, eastern red
cedar, loblolly pine, shortleaf pine, Virginia pine, slash pine, and
o
cypress all emit A -carene. However, this area contains; Canada hemlock,
eastern hemlock, pitch pine, white ash, yellow buckeye, sugar maple,
basswood, yellow birch, yellow poplar, American beech, mountain silverbell,
black cherry, mountain laurel, northern red oak, rhododendron, and
3
cucumber tree. We suspect that A -carene has been misidentified. This is
especially suspect since only one aromatic toluene was identified and high
values of A3-carene were observed when the acetylene concentration was
also high. This suggests that the samples were contaminated with auto exhaust,
assuming acetylene as properly identified.
29
-------
TABLE 7. Summary of Blogenlc Hi: Concentrations In Ambient Air
u>
o
Investigator-Sampling
location(s)
Rasnussen & Went (11)
Ozark Plateau, MO
Gray Summit, HI
Highland Biological St.NC
Gray Summit, HI
Rasnussen, et.al. (44)
Elkton, HI
Surrounding
vegetation
Hardwood forest
& prairie
Junipers *
Hardwood forest
Hardwood forest
meadow
Hardwoods &
farmland
Blogenlc
hydrocarbon
Isoprene
a-pinene
p-olnene
limonene
myrcene
totaled
totaled
totaled
laoprene
Avg , Cone .
ppbC
totaled
106
41
50
109
6
Max
ppbC
150
60
120
340
28
Min
ppbC
70
30
10
10
0
TNMHC I % Comments
ppbC Uogenic automotive
Lowest levels ob-
served lust before
aunrlse: hlih levels
observed when
deciduous foliage
changes color in
- the autumn; high
- level of volatiles
observed after grass
was mowed; packed
column GC-FID
98 6 12 79 measurements;
most Intensive study
Lonneman «t.«l. (45)
Chlckatawbut Hill, HA
Whltby & Coffey (46)
Adirondack Mountains ,HY
Oak & other
Coniferous
forest
Deciduous/
coniferous
forest
Isoprene
'terpene
species'
'lighter
species'
'terpene
species'
'lighter
species'
Lonneman, et.al. (29)
St. Petersburgh/Tampa,
Miami and the Everglades,
Florida
Orange groves, gum Isoprene
Cyprus, oak, wax d-llmonene
myrtle, black a-pinene
Willow, persimmon g-pinene
10(1 day)
33
63
12
IB
1:61
0
0
0
34
72
123
17
21
4.5
5.3
29.3
6.5
14
0.1
106
4
(ax)
61
yet performed.cc -
FID high resolution
SCOT column
SCOT column
Chromatographlc peato
not Identified;
authors assumed
biogenlc origin due
mainly to rural
location; packed
column CC-FID.
gc-fld; WCOT columns
-------
TABLE 7. (Continued)
Investigator-Sampling
Jocation(s)
Schjoldager & Wathue (47)
Gjerdrum, Norway
Seila (48)
Jones State Forest,
(38 mi North of Houston,
Holdren, et.al (49)
Moscow Mountain, North-
Central Idaho
Arnts & Meeks (43)
Tulsa (Suburban) OK
Rio Blanco, CO
Smokey (its. , TN
Air Resources Board (50)
Lake Tahoe High School
CA
Cronn - Harsch (51)
Smoky Hts TN
Surrounding
Jfegetation
Coniferous forest
(spruce & farmland)
Loblolly pine
TX)
Coniferous forest
(pine and fir)
Deciduous
Farmland
Sage Brush,
coarse grass,
juniper, pine
Deciduous
Coniferous
Fine
Deciduous
Coniferous
Biogenlc
hydrocarbon
a-pinene
B-plnene
limonene
isoprene
a-pinene
isoprene
a-pinene
B-pinene
3-carene
limonene
a-pinene
B-pinene
3-carene
limonene
Isoprene
Isoprene
Isoprene
a-pinene
a-plnene
B-pinene
Isoprene
a-pinono
6-pinenc
A -carene
Myrcene
Avg. Cone.
ppbC
14.0
9.5
3.6
0.1
1.13
0.86
0.64
0.10
1.23
1.73
1.08
0.10
0.45
2.10
3.84
9.2
3.9
4
3.9
2.5
7.34
1.76
Other natural hydrocarbons could be identified if concentrations exceed
Alpha-pinene is shown
with p-xylene, maximum
concentration
of both is 2.
Max
ppbC
16.5
19.5
trace
7.7
1.2
7.3
4.6
5.4
0.5 .
2.7
5.7
3.7
0.2
1.16
5.63
5.72
13.4
8.6
17.4
7.9
3.1
20.8
2.0
0.1 ppbC
44 ppbC.
Min
ppbC
10.5
3.0
<0.5
<0.1
<0.1
o.i -
<0.1
<0. 1
<0. 1
0.3 .
<0.1
<0« 1
<0. 1
-------
The automotive contribution shown in Table 7 was obtained by employing
the Lonneman method (52). This method uses the TNMHC/acetylene ratio for
tunnels. With this ratio and knowledge of the acetylene concentration of
the ambient air sample, an automotive contribution can be calculated. The
Cronn and Harsch samples contained very high acetylene concentrations
making the calculated automotive contribution greater than the total
observed non-methane hydrocarbon concentration in most samples. Either the
acetylene or the TNMHC concentration is in error. There is also the
possibility that some of the acetylene could arise from the burning of
wood at camp sites nearby (53). This would invalidate the Lonneman method
for calculating automotive contribution to TNMHC.
As stated earlier, isoprene, a-pinene and most certainly p-cymene have
lifetimes that are long enough for observation if their emissions are high,
even though their reactivity is also high. But p-cymene was not observed
at all, though isoprene and a-pinene were observed at very low levels.
Also, the natural HC contribution to the TNMHC in the Smokies was found to
be very low usually less than 5% (43) . The maximum values for isoprene
and a-pinene obtained for the 7-day study (September 21-27, 1978,) were
5.72 ppbC and 1.17 ppbC, respectively, but were not recorded on the same day.
If the Cronn and Harsch work is considered the natural hydrocarbon contri-
bution is still only ,. 20% of the total hydrocarbon burden. Also the TNMHC
is only ~ 90 ppbC. This is still a low percentage considering the fact
that the area is surrounded with vegetation.
32
-------
SECTION 6
PRODUCT FORMATION
As discussed earlier, a number of studies have shown that natural HC's
can produce aerosols when undergoing ozonolysis or photooxidation in the
presence of N0x. Went, who first observed a blue haze by placing 0 and
pine needles in a bell jar (10), further suggested that the mechanisms for
removal of natural HC's are similar to those responsible for smog formation.
In his dissertation work, Rasmussen studied some terpene reactions with 0_
and NO- (54). Though he was unable to analyze the aerosol material resulting
from reaction, he did point out that the products consisted of compound s
both lighter and heavier than the parent compound.
Isoprene was included as one of the HC's studied by Schuck and Doyle
(2). When 3 ppm isoprene were irradiated with 1 ppm NO , the products ob-
X
served were 1.7 ppm formaldehyde, 0.90 ppm CO, 0.20 ppm peroxyacetylnitrate
(PAN), 2.7 ppm acetaldehyde and 1.2 ppm acrolein. These investigators were
able to account for approximately 87% of the reacted carbon as identifiable
products.
When 5 ppm of both pinene and phellandrene were irradiated with 5 ppm
N0_, Stephens and Scott (39) found that 0.25 and 0.23 ppm PAN was produced.
Pinene also produced 1.3 ppm aldehyde, while phellandrene produced 2.0 ppm.
If formaldehyde is taken as the aldehyde product, then only 4.6 and 6.1%
of the carbon is accounted for, assuming all the initial HC reacts.
Haze formation was studied in New York City and the Blue Ridge
Mountains by Wilson et al. (55). They also synthesized aerosol production
in an environmental smog chamber, and collected the aerosols produced from
the photooxidation of a-pinene by extraction and derivation techniques. Gas
chromatograms and mass spectra of the extracted compounds indicated the
presence of pinonic acid and norpinonic acid.
33
-------
In a later study on haze formation also involving photooxidation of
a-pinene in a smog chamber, Schwartz (56) was able to identify an additional
product, pinononic aldehydes, in the neutral fraction. Pinonic and pinononic
acids were observed in the acid fraction of the extraction, but in the base
fraction, too little sample was available for analyses. The aerosols
collected at the Smoky Mountain site also contained the two acids. The
experimenters concluded from this study that the blue haze is a photochemical
aerosol fueled in part by the terpene emissions from trees. Unfortunately,
since no quantitation of the aerosol composition was attempted, one cannot
quantitate the degree of visibility reduction resulting from terpene
oxidation. One would expect in light of recent findings that the contri-
bution of natural HC's to visibility reduction is minimal ( 57, 58).
Weiss et al. (57) measured sulfates in submicron aerosols at several
locations. In the Ozarks, carbon species, not sulfates, were determined
to dominante the haze-producing aerosols. Stevens (58) made similar ob-
servations in the Smoky Mountains in a study in 1978.
Arnts and Gay (34), in their study on the chemistry of natural HC's,
tried to identify products arising from the photooxidation of the HC's with
NO . They studied isoprene, myrcene, d-limonene, terpinolene, a, and 3-
pinene and A -carene. The major gaseous products observed with the long-path
Fourier transform system-infrared (FTS-IR) are given in Table 8. The pro-
ducts include, CH20, HCOOH, CO, C02, CH3CHO, PAN, and (CH^CO. Methyl
vinyl ketone and methacrolein were also observed as derivation products of
isoprene.
Isoprene showed the highest product accountability (24% of reacted
carbon); the remaining compounds showed gaseous products from 2 to 17% of
the reacted carbon. Several unidentified absorption bands were also
detected, suggesting additional products. Using reasonable absorption
coefficients, however, the amount of material attributable to the unidentified
compounds is not very large. GC analyses performed by Arnts and Gay detected
only a small amount of acetaldehyde, although the GC system was able to
detect hydrocarbons to C. . No other products eluted from the GC column.
34
-------
TABLE 8. REACTIVITY AND PRODUCTS BY FOURIER TRANSFORM SPECTROSCOPY
(LONG-PATH INFRARED) AT 60 MIN,a
CO
Compound HC/NO Ratio
cv
g-Pinene
A-Carene
Isoprene
a-Pinene
Myrcene
d-Limonene
Terpinolene
5.3
6.7
6.2
6.7
6.2
6.7
6.2
%HC Reacted H2
58
60
68
72
73
91
95
0.
0.
2.
0.
0.
0.
1.
CO
94
46
39
33
84
9
2
HCOOH
0.39
0.39
0.44
0.32
0.16
0.35
0.31
CO
0.23
0.50
0.77
0.39
0.26
0.46
0.49
co2
0.4
0.9
0.2
0.6
0.24
0.49
0.69
CH3CHO
0.05
0.33
2.0
1.57
2.35
0.23
0.35
PAN (CH3)2CO
0.018
0.26
0.32
0.2Q
0.70
0.24
0.28
0
0
0
0
0.76
0
0
%ACC
2.1
7.1
24.8
8.9
16.9
4.1
4.9
Initial HC concentration = 8 ppm compound; products in ppm compound.
-------
Saunders et al. (59), in analyzing rainwater in the Washington, DC area,
found a compound identified as 3-methyl-furan. They proposed that the furan
is a by-product from the terpene photooxidation, suggesting an unusual ring
closure mechanism to explain their results. However, since 3-methyl-furan
has not been observed by other investigators irradiating terpenes, its
production from terpene photooxidation is highly unlikely. Also, since most
polluted atmospheres contain high ozone levels, the stability of this com-
pound is very doubtful, since it oxidizes when exposed to oxygen.
The composition of aerosol particles produced from the photooxidation
of terpenes with NO and reactions with ozone was investigated by Schuetzle
and Rasmussen (60). For example, the reaction of limonene with NO in the
X
presence of ultraviolet light produced over 30 aerosol products. These
include aldehydes, alcohols, acids, peroxides, nitro esters of alcohols,
acids and peroxides. The mass spectroscopic analyses also suggested that
dimetric and possible trimeric reaction products were produced upon photo-
oxidation. The formation of aerosols from the reaction of limonene with ozone
was found to be much faster than from the photooxidation reaction. Similar
products were observed with ozonolysis reactions, except for the nitrogen-
containing species. The aerosols were collected on stainless steel impactor
plates that collect particles greater than 0.5 ym. The collection of a
smaller fraction size would have been more useful, since a considerable
portion of the aerosols should be below 0.5 ym. The nephelometer used in
the study suggested that the aerosols account for greater than 50% of the
reacted carbon, a conclusion that qualitatively agrees with the earlier
findings of Arnts and Gay. However, the nephelometer could also be missing
a large amount of material because aerosols greater than 0.1 ym are not
easily detected (i.e., the amount detected is not proportional to the mass).
Cronn et al. (61) surveyed particulate samples collected in southern
California. Compounds that were tentatively identified by high resolution
MS include pinonic acid (C10H16°3+)» norpinonic acid (CgH^O^), a C^E^O^
isomer and a CgH-.O,, isomer. The total contribution of the peaks was only
1 yg/m in the midafternoon sample (1200-1420 PST). The total mass loading
3
during this time period for particles of d <_ 3.5 ym was 204 yg/m . The
36
-------
sum of these tentatively-identified natural organic aerosols was lowest at
nighttime and the evening hours. The above data suggest that the natural
HC contribution to aerosol formation in the Los Angeles area is negligible.
A recent study by Hull (62) has shown pinonaldehyde, nopinone, and
myrtenol as products produced from the ozonolysis of a-pinene. This study did
not specify whether the products were in the gas or solid phase, since all the
material collected was introduced into GC-MS.
Zimmerman et al. (63) have suggested that the oxidation of terpenoids
is an important source of CO in the atmosphere. These investigators cal-
14
culated that between 4.2-13.3 x 10 g/yr of CO can be generated from the
atmospheric photooxidation of isoprene and the terpenes. Their calculation
assumes that 60-80% of the carbon in natural HC's can lead to CO formation.
The hypothesis of Zimmerman et al. has been tested recently in an EPA
study with a-pinene (see Figures 8 and 9). Starting with 11.5 ppmC of a-
pinene, 1.3 ppm CO was found after a 6-hr irradiation period, which corresponds
to 11.5% of the original carbon. The figures show that the photooxidation
of a-pinene produces high levels of aerosols, but does not produce high con-
centrations of CO. The aerosols are also apparently lost to the walls of
the chamber, since the number, volume, and surface area of particles all
decrease after 100 min of irradiation, even though all of the a-pinene does
not react.
Aerosols were collected at the end of the irradiation, usually after
6 hr. The total concentration of the material found on the filter was only
1.9% of the original starting compound (with 71 mg of a-pinene in the chamber
initially, and 1.3 mg of aerosol collected). Obviously, a problem exists
with these experiments in terms of carbon balance, since the material was not
observed in either the gas or aerosol phase. We have seen earlier from
the work of Schuetzle and Rasmussen that over 50% of the organic material
was observed in the aerosol phase at least for limonene and terpinolene. The
aerosol data shown in Figure 9, when reasonable estimates are made with the
aerosol density and aerosol composition, agree very well with the amount of
material collected. These data suggest that a large fraction of organic
37
-------
OJ
CO
1.2 -
100
200
300
TIME, minutes
Figure 8. Data on photooxidation of alpha-pinene in the presence of NOX.
-------
CO
vo
O TOTAL SURFACE AREA x 103
D TOTAL NUMBER x 105/cc
A TOTAL VOLUME x 102 n3/cc
1.0 _
200 300
TIME, minutes
Figure 9. Aerosol data for the photooxidation of alpha-pinene.
400
-------
carbon exists in the gas phase that is not measured, or that the walls of
the chamber are acting as a significant sink for both aerosols and gaseous
products. In light of the few gaseous products identified by FTS-IR tech-
niques employed by Arnts and Gay (34) (see Table 8), the latter explanation
appears most plausible.
The CO produced in this latest EPA study is clearly greater than that
found earlier by Arnts and Gay (11.5% versus 0.6%). However, the EPA study
was conducted over a period of 6 hr, while the Arnts and Gay work was con-
ducted for 60 min, with a-pinene still remaining in the chamber. The increased
yield in CO is probably a result of further oxidation of gaseous products,
and the total aldehydes are also low. The EPA study further strengthens the
position that gaseous oxygenates are not produced in great quantities as
suggested by Ludlam and Bailey (30).
A significant fraction of the aerosols produced from terpene photo-
oxidation is not likely to lead to CO formation in the atmosphere certainly
not 60-80% of the reacted carbon! Duce (64) calculated a fine-particle aerosol
(d < 1pm) lifetime of 4-7 days. In this time frame, more terpenoid aerosols
could be converted to CO. However, this process requires several days, not
the 2-5 hr suggested by Zimmerman et al. Duce further calculates that only
60-140 metric ton/yr of terpenoid is required to explain that POC in the
atmosphere. The 900 metric ton/yr of terpenoids. given by Zimmerman and
coworkers (63) is clearly not compatible with either particle organic carbon
(POC) observed nor the CO produced in smog chambers.
Hanst et al. (65) recently investigated the CO balance in the atmosphere.
Their calculations suggest that the oxidation of methane by OH cannot
possibly give rise to the levels of CO currently present in the atmosphere.
Instead, they hypothesize that the photooxidation of natural HC's accounts
for the CO present in the atmosphere. To this end, experiments were per-
formed for measuring the photodissociation of Cl. and subsequent reaction
of Cl atoms with natural HC's. Their data with isoprene show that, with
73% of the carbon as CO- and CO, only 45% could be identified as CO. With
a-pinene, the amount of CO was even less, with CO and C0_ accounting for
only 30% of the carbon. Only 12% of this reacted carbon could be established
as CO.
40
-------
Although Hanst et al. did show more CO was produced than in either the
EPA study (Figure 8) or the Arnts and Gay work, the amount of CO was still
low, at least for the case of a-pinene. The experimental data shown by
Hanst et al. do not substantiate their conclusions, i.e., that the terpenoids
are responsible for much of the CO in the atmosphere. Furthermore, these
investigators have employed Cl, a fast-reacting free radical. In the
atmosphere, OH radicals and 03 are the reactive species. Possibly, the
high concentration of Cl atoms could have reacted with the aerosols pro-
duced and resulted in more CO than is produced under normal atmospheric con-
ditions. Certainly, the data shown in Figure 8 do not substantiate the
Hanst hypothesis.
Interestingly, the ozone curve in the irradiation of a-pinene shown in
Figure 8 first increases rapidly, later decreases, and then slowly increases
throughout the irradiation. A similar observation was made by Arnts and Gay
(34) with d-limonene at high HC/NO ratios (greater than 40/1). Arnts and
X
Gay explained this double peak in terms of the very rapid reaction between
ozone and the d-limonene, with subsequent photooxidation of the carbonyl
products and NO later producing more ozone. This explanation is plausible for
«*
modeling efforts, since arbitrarily raising the 0, + propylene reaction rate
by a factor of 10 can produce a double ozone peak in smog chambers. But
atmospheric buildup of ozone in the late afternoon is unlikely because:
(1) smog chambers are atypical since they are efficient sources of NO , ozone
^^ 2w
production can continue when NO is below the detection limits of the
X
instrument (66), and (2) in the atmosphere, the solar intensity begins to
decrease late in the afternoon. Therefore, the system will be less photo-
chemically active late in the afternoon.
A validated mechanism for the photooxidation of terpenoids cannot be
written at this time. The main difficulty lies in the proper identification
of the products arising from the photooxidation. In the case of isoprene,
for example, only Schuck and Doyle (2) and Arnts and Gay (34) have investigated
product formation. Schuck and Doyle could account for a large fraction
(approximately 90%) of the original carbon (or even more than 90%, since
acrolein identified by them was probably methacrolein); however, Arnts and
41
-------
Gay (34) could account for only approximately 44% of the reacted carbon.
They suggest that the remaining carbon may be in the aerosol phase.
A simplified mechanism is given in the Appendix for the photooxidation
of the naturally-occurring HC's. Excluded from the mechanism are the
inorganic reactions, since they are fairly well understood (3).
Most of the reactions shown in the Appendix start with the breaking of
carbon-carbon double bonds with ozone. Obviously, the C-Q terpenes can also
react with hydroxyl radicals, and the product distribtuion does not appear
to be significantly different when isoprene is reacted with OH instead of
0_. This situation will probably be true with the monoterpenes as well.
In the case of p-cymene, only hydroxyl radicals are shown to react, because
as shown in Table 4, the reaction with 0_ is slow.
Nitric oxide is shown with most of the reactions, even with 0_ present.
Although NO and 0- react very quickly, a small amount of NO is always present
in a photochemical system (due to N0» photolysis). The nitric oxide is used
primarily to convert the hydroperoxyl and peroxyalkyl radical to hydroxyl
and alkoxyl radicals.
Organic nitrogen-containing compounds aside from PAN, were not shown.
They are obviously produced, however, since RO and RCO_ radicals can be
expected to react with N0~. As discussed earlier, Schuetzle and Rasmussen
have also reported nitrogen-containing products in aerosols when terpenes
were photooxidized in the presence of NO .
We have mentioned that a large fraction of the terpene products is ex-
pected to be in the aerosol phase, since upon reacting with OH or 0-, the
fragmented oxygenated products should have a lower vapor pressure than the
parent compound (the monoterpenes have a boiling point greater than 150°C).
Aerosols are produced by quick condensation of the partially-degraded ter-
penoids on pre-existing particles.
42
-------
SECTION 7
SOURCE-RECEPTOR RELATIONSHIPS
Source-receptor relationships are extremely important when considering
oxidant production from natural HC's in the atmosphere, and serious questions
have been raised that the published emission rates are too high. Certainly
the experimentally-available ambient data do not suggest such rates. However,
even if the rates are high, an unfortunate tendency exists to assume one-to-one
relationships between pollutants from various sources of emissions. For
example, the experimental measurements would suggest that nitric oxide emissions
from power plants will not significantly affect ozone production until the
air mass containing them travels 25-30 km downwind f 67, 68;. The reason is a
very low HC/NO ratio. The excess NO actually decreases the ambient levels
X
of 0« near power plant plumes, which is quite a different case from the
nitric oxide emitted by auto exhaust in the presence of HC's.
The question then can be raised as to the effect of vegetation on a
city's air quality. Figure 4 shows that the maximum ozone is generated at a
HC/NO ratio of approximately 10-20. However, the bulk of the natural HC's
X
is emitted in rural areas where vegetation is so plentiful, and NO sources
X
so few, that the HC/NO ratio is expected to be much higher. These ratios
X
have been measured by EPA and Washington State University scientists (69, 70)
and have been found to be as high as 60-100 (with NO levels near or below
X
the sensitivity of the measuring instruments-usually less than 1-2 ppb V/V).
Also, using fingerprint HC's such as acetylene or the xylenes, the major
portion of the HC burden in rural areas is determined to be from anthropogenic
origins. Therefore, insufficient N0x is available to drive the photochemical
reactions necessary for significant ozone production. The little that is
produced is quickly titrated with excess natural HC's, which also results in
consumption of natural HC's. Evidence for such reactions is shown in Table 5,
where the daylight lifetimes of some natural and manmade HC's are given. The
43
-------
table also shows that isoprene and a-pinene will nonetheless be transported
into urban areas from forested tracts, provided the emissions of these
compounds are large.
In order to assess the importance of emissions from forested areas into
a city, a series of calculations was performed using a dispersion equation
( II) . This equation was derived from the numerical integration of infinite
line sources with respect to upwind distance increments. Its general form is:
dist.
0
where
X = concentration of pollutant
q = source flux
u - wind speed
The calculations were made with the following boundary conditions:
(1) Meteorological stability Class 4
(2) Receptor height 2 m
(3) Emission height 10 m
(4) Inversion height 100 m
2
(5) Source flux (a) 100 yg/m min
(6) Wind speed (U) 5 m/sec
4 2
(7) Source area 10 km
(8) Source receptor 1 km downwind.
The concentration of natural HC's found at a downwind receptor, i.e., a
4 2
city 1 km downwind from a forest of 10 km , was determined to be 184 ppbC.
But this concentration is unrealistically high, since (1) the emission levels
assumed were very high, (2) the area assumed (10 km of forest) was very
large, and (3) the natural HC, probably isoprene within a decidous forest,
44
-------
or a-pinene within a coniferous forest, is a reactive olefin that will undergo
reactions with OH radicals and ozone. However, of note is that when an air
quality simulation model similar to the Empirical Kinetic Modeling Approach
(EKMA) (72) is employed, the 184 ppbC cannot exceed the present air quality
standard of 0.12 ppm. In fact, when one considers that terpenes are not very
efficient ozone producers (in terms of ozone produced per parts per million
carbon) , the amount produced from the 184 ppbC is less than that predicted
by the air quality model which is based on propylene/n-butane mixtures.
If the above calculation is repeated with a-pinene as the natural HC, and
8
0.05 ppm ( 73> and 3 x 10 ppm as the steady-state concentrations of 0, and
OH, respectively, are assumed then the gas phase a-pinene observed in a city
will be 81.4 ppbC. The aerosol loading would contain 102.6 ppbC (60.4 pg/m3),
assuming all the a-pinene reacts to produce aerosols.
3
Total particulate carbon at a concentration of 60.4 yg/m is much too
high. As stated earlier, a recent study (58) in the Smoky Mountains has
3
shown that the total carbon concentration was 3.3 yg/m in the fine-particulate
fraction. The total carbon in both the large and small fractions was only
3
4 yg/m . Even if one assumes that all of the carbon in the area was of
natural sources, the model overpredicts the aerosol burden by a factor of 15.
In fact, part-if not most-of the organic aerosols were probably of anthropogenic
origin, since the TNHC in the area averaged 120 ppbC, with 95% being anthro-
pogenically produced.
The size of the source area in the above calculations is too large,
because few if any areas have 10 km of dense forest, and the largest con-
tribution (50%) of theo-Pinene is received from the last 5 km of forests.
Hence, even though a very large area was assumed, the bulk of the HC load
arises from sources near the receptor.
A box-type photochemical model was also employed in order to determine
the amount of ozone produced from slow emissions, HC's and N0x. The conditions
employed for this modeling effort were:
4 2
(1) Source area of 10 km
(2) Diurnal insolation ( max = 0.52 min )
45
-------
2 2
(3) HC emission rate at 60 yg/m min; NO at 5 yg/m min
X
(4) Propylene as surrogate terpene
(5) 03 initally of 40 ppb V/V
(6) 5% dilution per hour containing 40 ppb of 03
(7) Mixing height of 1.8 km (the summer condition for NC (76)).
Propylene was chosen as the surrogate terpene since, as stated earlier,
a mechanism for the photooxidation of natural HC's cannot be written. In fact,
-2 -1
the use of propylene and a high emission rate of 60 yg/m min is grossly
overestimating the photochemical potential of terpenes because propylene has
been shown to produce more ozone on a parts-per-million-carbon equivalent
than most terpenes (Figure 4).
The Dodge (72) photochemical model used in the calculation indicates that
after an irradiation period of 10 hr, the total ozone in the box of 100 x 100
x 1.8 km was 70 ppb. The concentration of ozone remaining after 10 hr is
actually only 30 ppb above that contained in the dilution air, suggesting
that the olefin/NO mixture contributed very little to ozone formation. When
X
the above calculations were repeated with no initial ozone present, and a HC/
NO ratio of 200 (by reducing NO ) the amount of 0~ produced after 10 hr was
X X .3
12 ppb. This later ratio is more comparable to actual rural HC/NO conditions.
x £
One must therefore conclude that even high emission levels of 60 yg/m min
will not produce significant ozone concentrations.
Natural HC's emitted within urban areas could contribute to ozone
formation if their emission rates were high. The HC/NO ratio found in many
X
cities is favorable for 0_ production. But the amount of vegetation in urban
areas is small and natural HC emissions are expected to be low. Data obtained
in urban areas at a number of cities show very little isoprene and practically
no Clf) natural HC's. These data were shown in Table 6, including: Tampa,
Miami, Tulsa, Houston, and Boston. As shown earlier, isoprene and a-pinene
have sufficiently long lifetimes to be observed if emissions were significant
either upwind from a city or within the urban center. As for the more reactive
HC's their emissions must also be low, since they were not observed in any of
the samples (even though the sensitivity of the measurements is 0.2 ppbC).
Also, such compounds would be expected to produce copious quantities of fine-
particulate aerosols, which was not observed (76). Aldehydes were also found
at low concentrations (77). ,,
-------
REFERENCES
L. Haagen-Smit, A.J., Fox, M.M. "Automobile Exhaust and Ozone Formation "
S.A.E. Trans., 63:575, 1955.
2. Schuck, E.A., Doyle G.J. Air Pollution Foundation Report No. 29, San
Marino, CA, October 1959-
3. Leighton, P.A. "Physical Chemistry: Vol. IX, Photochemistry of Air
Pollution," Academic Press, New York, 1961.
4. Altshuller A.P. Bufalini, J.J., "Photochemical Aspects of Air Pollution-
A review" Photochem. Photobiol., 4:97, 1965.
5. Craker, L.E. "Ethylene Production from Ozone Injured Plans," Pollut.
1:299, 1971.
6. Tingey, D.T. Standley, C., Field, R.W., "Stress Ethylene Evolution:
A Measurement Ozone Effects of Plants," Atmos. Environ., 10:969, 1976.
7. Abeles, A.L., Abeles, F.B. "Biochemical Pathway of Stress-Induced
Ethylene," Plant Physiol, 50:496, 1972.
8. Lodge, J.P., Jr., Bien, C.S., Suess, H.E. "The Carbon-14 Contents of
Urban Airborne Particulate Matter," Internat. J. Air Pollut. 2:309, 1960.
9. Currie, L.A., Klouda, G.A. "Advances in the Discrimination of Natural
from Anthropogenic Carbonaceous Pollutants via Isotopes of Carbon,"
Presented at the Division of Environ. Chem., A.C.S., Honolulu, April
1979.
10. Went, F.W. "Organic Matter in the Atmosphere, and Its Possible Relation
to Petroleum Formation," In Proceed. N.A.S., 46(2):212, 1960.
11. Rasmussen, R.A., Went, F.W. Volatile Organic Material of Plant Origin
in the Atmosphere," In: Proceed. N.A.S., 53(1):215, 1965.
12. Arnts, R.R., Seila, R.L., Kuntz, R.L., Mowry F.L., Knoerr K.R., Dudgeon,
A.C. "Measurement of Alpha-Pinene Fluxes from Loblolly Pine Forest," In:
Proceed. 4th Joint Confer, on Sensing of Environ. Pollut., RTP, NC, 1978,
222 pp.
13. Ripperton, L.A., White, 0., Jeffries, H.E. "Gas Phase Ozone-Pinene
Reactions," 147th National ACS Meeting, Div. Water, Air, Waste Chemistry,
Preprint, 7(2): 54-56, 1967.
47
-------
14. Ripperton L.A., Jeffries, H.E., White, 0. "Formation of Aerosols by
Reaction of Ozone with Selected Hydrocarbons," Advances Chem. Series,
112:219, 1972.
15. Spicer, C.W. Joseph, D.W., Sticksel, P.R., Ward, G.F. "Ozone Sources
and Transport in the Northeastern United States," Environ. Sci, Technol.,
13:475, 1979.
16. Rasmussen, R.A. What Do the Hydrocarbons from Trees Contribute to Air
Pollution." J. Air Pollut. Control Assoc., 22:537, 1972.
17. Kittredge J. "Estimation of the Amount of Foliage of Trees and Stands."
J. Forestry, 42:905, 1944.
18. Blow, F.E. "Quantity and Hydrologic Characteristics of Little Under
Upland Oak Forests in Eastern Tennessee." J. Forestry, 53:190, 1958.
19. Tyson, B.J. Dement, W.A., Mooney, H.A., "Volatilization of Terpenes
from Salvia Mellifera." Nature, 252:119, 1974.
20. Dement, W.A., Tyson, B.J., Mooney, H.A. "Mechanism of Monoterpene
Volatilization in Salvia Mellifera." Phytochemistry, 14:2555, 1975.
21. Research Triangle Institute, "Natural Emissions fo Gaseous Organic Com-
pounds and Oxides of Nitrogen in Ohio and Surrounding States," Final
Report,. EPA Contract No. 68-02-1096, EPA, RTF, NC, December 1974.
22. Zimmerman, P.R. "Testing of Hydrocarbon Emissions from Vegetation, Leaf
Litter, and Aquatic Surfaces, and Development of a Methodology for
Compiling Biogenic Emissions Inventories." Final Report, EPA-450/4-79-004,
EPA, RTP, NC, March 1979.
23. Zimmerman, P.R. "Tampa Bay Area Photochemical Oxidant Study." Final
Report, EPA-904/9-77-028, EPA, RTP, NC, February 1979-
24. Dasmann, R.F. "Environmental Conservation." (4th Ed.), Int. Union
for Conservation of Nature and Natural Resources, Moryes, Switzerland
(1976).
25. Tingey, D.T. Ratsch, H.C., Manning, M., Grothaus, L.C., Burns, W.F.,
Peterson, E.W., "Isoprene Emissions Rates from Live Oak," EPA 904/9-78-
004, May 1978.
26. Mirov, N.T. "Composition of Gum Turpentines of Pines." Forest Service
Technical Bulletin No. 1239, U.S. Department of Agriculture, pp. 1-58,
(1961).
27. Rasmussen, R.A. "Isoprene: Identified as a Forest-Type Emission to the
Atmosphere." Environ. Sci. Technol. 4:667, 1970.
28. Rasmussen, R.A., Presented at Symposium on Biogenic Hydrocarbon
Emission Rates, Concentrations and Fates, January 8-9, 1980, EPA, RTP, NC.
48
-------
29. Lonneman, W.A. Seila, R.L., Bufalini, J.J. "Ambient Air Hydrocarbon
Concentrations in Florida," Environ. Sci. Technol., 12:459, 1978.
30. Zimmerman, p.R. "Procedures for Conducting Hydrocarbon Emission Inventories
of Biogenic Sources and Some Results of Recent Investigations." Pre-
sented at the EPA Emission Inventory/Factor Workshop, Raleigh, NC, Sept.
j.y / / *
31. Sculley, R.D. Correspondence on "Ambient Air Hydrocarbon Concentrations
in Florida." Environ. Sci. Technol., 13:234, 1979.
32. Ludlum, K.H., Bailey, B.S. Correspondence on "Ambient Air Hydrocarbon
Concentrations in Florida." Environ. Sci. Technol., 13:235, 1979.
33. National Bureau of Standards, "Reaction Rate and Photochemical Data for
Atmospheric Chemistry - 1977. " R.F. Hampson, D. Garvin, eds., Special
Pub. 513, Washington, DC 20234.
34. Arnts, R.R., Gay, B.W. Jr. "Photochemistry of Naturally Emitted
Hydrocarbons: Isoprene, p-Cymene, and Selected Monoterpenes."
EPA-600/3-79-081, Research Triangle Park, NC, September 1979.
35. Winer, A.M., Lloyd A.C., Darnall, K.R., Pitts, J.N., Jr. "Relative
Rate Constants for the Reaction of the Hydroxyl Radical with Selected
Ketones, Chloroethenes, and Monoterpene Hydrocarbons." J. Phys. Chem.,
80;1635, 1976.
36. Atkinson R., Pitts, J.N., Jr. "Rate Constants for the Reaction of
OH Radicals with Propylene and the Butenes over the Temperature Range
297-425°K." J. Chem. Phys., 63:3591, 1975.
37. Grimsrud, E.P., Westberg, H.H., Rasmussen, R.A., "Atmospheric Reactivity
of Monoterpene Hydrocarbons, NO Photooxidation, and Ozonolysis,"
Int. J. Chem. Kinetics, Symposium No. 1 (1975).
38. Pate, C.T., Atkinson, R. , Pitts, J.N., Jr. "The Gas Phase Reactions of
0- with a Series of Aromatic Compounds (preprint) 1978.
39. Stephens, E.R., Scott, W.E. "Relative Reactivity of Various Hydro-
carbons in Polluted Atmospheres." Proceedings of the American
Petroleum Institute, 42:665, 1962.
40. Lillian D. "Formation and Destruction of Ozone in a Simulated Natural
System (Nitrogen Dioxide + alpha-pinene + hOi." Advances Chem. Ser.,
113:211, 1972.
41. Westberg H.H., Robinson, E. "Natural Hydrocarbons in Photochemical Air
Pollution." Final Report on EPA Grant No. R800670, October 1979.
42. Altshuller, A.P., Kopczynski, S.L., Wilson, D., Lonneman, W., Sutterfield,
F.D. "Photochemical Reactivities of N-Butane and Other Paraffinic
Hydrocarbons." J. Air Pollut. Contr. Assoc., 19:787, 1969.
49
-------
43. Arnts, R.R., Meeks, S.A. "Biogenic Hydrocarbon Contribution to the
Ambient Air of Selected Areas", U.S. EPA, RTF, NC, EPA-600/3-80-023,
January 1980.
44. Rasmussen, R.A., Chatfield, R.B. Holdren, M.W., Robinson, E., "Hydro-
carbon Levels in a Midwest Open-Forested Area." Draft Report submitted
to the Coordinating Research Council, October 1976.
45. Lonneman, W.A., Seila, R.L., Meeks, S.A. "Preliminary Results of
Hydrocarbon and Other Pollutant Measurements Taken During the 1975
Northeast Oxidant Transport Study. "In: Proceedings of Symposium of
1975 Northeast Oxidant Transport Study, EPA-600/3-77-017,February 1977.
46. Whitby, R.A., Coffey, P.E. "Measurement of Terpenes and Other Organics
in an Adirondack Mountain Pine Forest." Journal of Geophysical
Research, 82:5928, 1977.
47. Schjoldager, J., Wathne, B.M. "Preliminary Study of Hydrocarbons in
Forests." Norsk Institute for Luftforskning, P. 1-26, 1978.
48. Seila, R.L. "Non-Urban Hydrocarbon Concentrations in the Ambient Air
North of Houston, Texas." EPA-600/3-79-010, U.S. Environmental Pro-
Tection Agency, RTF, NC., February 1979.
49. Holdren, M.W., Westberg H.H., Zimmerman, P.R. "Analysis of Monoterpene
Hydrocarbons in Rural Atmospheres." J. Geophysical Research, 84:5083, 1979.
50. California Air Resources Board. "Lake Tahoe Communities Hydrocarbon
Analyses." Haagen-Smit Laboratory, El Monte, California, January 1979.
51. Cronn D.R., Harsch D.E., "Smoky Mountain Ambient Halocarbon and Hydro-
carbon Monitoring, September 21-26, 1978, Draft Report, U.S. Environmental
Protection Agency, Grant No. R-804033, (1979).
52. Lonneman, W.A., Kopczynski, S.L., Darley, P.E. Sutterfield, F.D.
"Hydrocarbon Composition of Urban Air Pollution," Environ. Sci. Technol.
£, 229 (1974).
53. Sandberg D.V., Pickford S.G., Darley E.F. "Emissions from Slash Burning
and the Influence of Flame Retardant Chemicals." J. Air Pollut. Control
Assoc. 25_ 278 (1975).
54. Rasmussen, R.A. "Terpenes: Their Analysis and Fate in the Atmosphere."
University Microfilms Inc., Ann Arbor, Michigan, 1966.
55. Wilson, W.E., Schwartz, W.E., Kinzer G.W. "Haze Formation - Its
Nature and Origin." EPA, CPA, 70-Neg. 172, Coordinating Research
Council CAPA-6-68, January 1972.
56. Schwartz, W. "Chemical Characterization of Model Aerosols." EPA Report
No. 650/3-74-011, U. S. Environmental Protection Agency, RTF, NC,
Aug. 1974.
50
-------
57. Weiss, R.E., Waggoner, A.P., Charlson, R.J., and Ahlquist, N.C.
"Sulfate Aerosol: Its geographical extent in the Midwestern and
Southern United States." Science, 195:979, 1977.
58. Stevens, R.K. Concentrations and Organic Aerosols in the Great Smoky
Mountains and the Soviet Union." Presented at the EPA Symposium on
Atmospheric Biogenic Hydrocarbons, Emission Rates, Concentrations and
Fates, Research Triangle Park, NC, January 1980.
59. Saunders, R.A., Griffith, J.R., Saalfeld, F.E. "Identification of Some
Organic Smog Components Based on Rain water Analysis." Bio-med. Mass
Spec., 1:192, 1974.
60. Schuetzle, D. , Rasmussen, R.A., "The Molecular Composition of Secondary
Aerosol Particles Formed from Terpenes." J. Air Poll. Control Assoc.,
28:236, 1978.
61. Cronn D.R., Charlson R.J., Knights R.L., Crittenden, A.L. "A Survey
of the Molecular Nature of Primary and Secondary Components of
Particles in Urban Air by High-Resolution Mass Spectrometry." Atmos-
pheric Environ., 11:929, 1977.
62. Hull, L. "Terpene Ozonolysis Products." Paper presented at the EPA
Symposium on Atmospheric Biogenic Hydrocarbons, Emission Rates,
Concentrations, and Fates, Research Triangle Park, NC, January 1980.
63. Zimmerman, P.R., Chatfield, R.B., Fishman, J., Crutzen, P.J., Hanst,
P.L. "Estimates on the Production of CO and H_ from the Oxidation of
Hydrocarbon Emissions from Vegetation." Geophys. Research Letters,
5:679, 1978.
64. Duce, R.A. "Speculations on the Budget of Particulate and Vapor Phase
Non-Methane Organic Carbon in the Global Troposphere." Pure and
Applied Geophysics, 116:244, 1978.
65. Hanst, P.L., Spence, J.W., Edney, E.G. "Carbon Monoxide Production in
Photooxidation of Organic Molecules in the Air." preprint, February
1977.
66. Bufalini, J.J., Walter, T.A., Bufalini, M.M. "Contamination Effects
on Ozone Formation in Smog Chambers." Environ. Sci. Technol, 11:1181, 1977.
67. Westberg, H., Sexton K., Holdren M. "Contribution of the General Motors
Automotive Painting Facility at Janesville, Wisconsin to Ambient Ozone
Levels." Report to GM by WSU, August 1978.
68. Wilson, W.E. "Sulfates in the Atmosphere: A Progress Report on Project
MISTT." Atm. Environ. 12_, 537 (1977).
69. Lonneman, W. "In: Proceedings of Symposium on 1975 Northeast Oxidant
Transport Study, EPA-600/3-77-017, EPA, RTF, NC, Feb. 1977.
51
-------
70. Westberg, H.H., Allwine, K.J., Robinson, E. "Ambient Hydrocarbon and
Ozone Concentrations near a Refinery." EPA-600/7-77-049, U.S.
Environmental Protection Agency, RTF, NC, May 1977.
71. Turner, D.B. "Dispersion Estimate Suggestion No. 4." U.S. Environmental
Protection Agency, RTP, NC, March 4 (1974).
72. Dodge, M.C. "Effect of Selected Parameters on Predictions of a Photo-
chemical Model." EPA-600/3-77/048, U.S. Environmental Protection
Agency, RTP, NC, June 1977.
73. Singh, H.B., Ludwig, F.L., Johnson W.B. "Tropospheric Ozone: Concen-
trations and Variabilities in Clean Remote Atmospheres." Atmospheric
Environ. 12:2185, 1978.
74. Campbell, M.J., Sheppard, J.C., Au, B.F. "Measurement of Hydroxyl Con-
centration in Boundary Layer Air by Monitoring CO Oxidation." J. Geophys.
Res. Letters, 6:175, 1979.
75. Holzworth G.G. "Mixing Heights, Wind Speeds, and Potential for Urban
Air Pollution Throughout the Contiguous United States." EPA
Publication 101, U.S. Environmental Protection Agency, Research Triangle
Park, NC (1972).
76. Grojean, D., Van Cauwenberghe, K., Schmid, J.P., Keley, P.E., Pitts
J.N. Jr., "Identification of C--C1Q Aliphatic Dicarboxylic Acids in
Airborne Particulate Matter." Environ. Sci. Technol. 12:313, 1978.
77. Joshi, S.B. "Houston Field Study (1978) - Formaldehyde and Total
Aldehydes Monitoring Program." Report to EPA, Contract No. 68-02-2566,
Northrop Services, Inc., Research Triangle Park, NC, February 1979.
52
-------
APPENDIX
REACTIONS INVOLVING SOME NATURAL HYDROCARBONS
ISOPRENE
H 03
(1) CH2 = C-C = CH2 »- CH2C(CH3)CHO + H20 + CO + CH20
CH3 (NO) METHACROLEIN FORMALDEHYDE
(2) CH2 - C-C - CH2 » CH3C(0)CHCH2 + CH20 + H20 + CO
CH3 (NO) METHYLVINYLKETONE
(3) CH3-C-C = CH2 --+- CH3C(0)CHO + H20 + CO
(NO) METHYLGLYOXAL
OH 03 00'
(4) CH3-C-C-CN2 » CH3-C-C^2H02+CO
(NO)
0
It
CH3-C-H + C02
ACETALDEHYDE
H OH
(5) CH2 - C-C » CH2 > CH2C(CH3>CHO + CH20
CH3 NO METHACROLEIN
02
H 0
(6) CH2 - C-C = CH2 *" HC-OH + CH20 + H02 + CO
CH3 FORMIC ACID
0 OH 0
(7) CHs-C-H * CH3-C-0-ON02
02 PEROXYACETYLNITRATE (PAN)
N02
53
-------
MYRCENE
(8)
^
03 CH3-C-CH3 + CH2CHC(CH2)CH2CH2COOH
ACETONE 4-METHLENE-5-HEXENOICACID
CH2CHC(CH2)CH2CH2CHO
4-METHYLENE-5-HEXENAL
(9)
03
(CH3)2CCHCH3CH2C(0)CHCH2 + HCOOH
4-METHYL-4-PENTENYLVINYLKETONE
(10)
(CH3)2CCHCH2CH2C(CH2)CHO + CH20
6-METHYL-2-METHYLENE-5-HEPTENAL
a-PINENE
(11)
CH3-C-CH3 + CH20 + CH200 + HOOCCH2CH2COOH
ACETONE SUCCINICACID
54
-------
(12)
CH3
03 B
(NO)
CH3
XT
x^CHO
(NO)
02
COOH
PINONICACID
H02 + CO
PINONONIC ALDEHYDE
PINONONIC ALDEHYDE
(13)
Hf/OH
(NO) (02)
COOH
NORPINONICACID
/3-PINENE
(14)
03
^^MMi
(NO)
OH
FRAGMENTATION
PRODUCTS
NOPINONE
55
-------
(15)
A-CARENE
03
NO
1-ETHANAL-2, 2 DIM-
ETHYL-3-ACETONYL-
CYCLOPROPANE
2, 2-DIIWETHYL-3-ACE-
TONYLCYCLOPROPYL
ACETIC ACID
(16)
LIMONENE
+ H02
+ CHOOH
4-ACETYL-1-METHYLCYCLOHEX-1-ENE
(17)
/ \ / 03
\ / ^ "i
\. /
(18)
3(2-PROPENYL)-6-KETO-HEPTANAL
-i~yj
HC /
HC
6
3-ACETYL-6-KETO-HEPTANAL
56
-------
(19)
(3-PHELLANDRENE
_03_
(NO)
CH20 + HC02H
4-ISOPROPYLCYCLOHEXENONE
(20)
03
(NO)
V
HOOCC(CH2)CH2CH2-CHCHO
2-METHYLENE-5-ISOPROPYL-6-OXO-HEXANOICACID
(21)
P-CYMENE
OH
CH3
2-P-TOLYLPROPANAL
(22)
OH
02
+ CH3COCH3
ACETONE
57
-------
(23)
OH
3-HYDROXY4-ISOPROPYL TOLUENE
OH
02
FRAGMENTATION PRODUCTS
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
REPi
EPA-600/2-80-086
3. RECIPIENT'S ACCESSION-NO.
TITLE AND SUBTITLE
IMPACT OF NATURAL HYDROCARBONS ON AIR QUALITY
5. REPORT DATE
May 1980
6. PERFORMING ORGANIZATION CODE
. AUTHOR(S)
Joseph J. Bufalini
8. PERFORMING ORGANIZATION REPORT NO.
9. P
ERFORMING ORGANIZATION NAME AND ADDRESS
Environmental Sciences Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
10. PROGRAM ELEMENT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Sciences Research Laboratory -RTP, NC
Office of Research and Development
US Environmental Protection Agency
research Triangle Park, North Carolina 27711
13. TYPE OF REPORT AND PERIOD COVERED
in-house
14. SPONSORING AGENCY CODE
EPA/600/09
15. SUPPLEMENTARY NOTES
16. ABSTRACT
The emissions, reactivities, and ozone-forming potential of natural hydro-
carbons are discussed. A review of the data available on emission levels for
natural hydrocarbons indicates that much more information is needed in order to
quantify the absolute emission levels, becasuse emissions data do not agree with
ambient air measurements.,. These ambient air measurements suggest that the previous-
published value of 9 x 10 ton/yr needs to be lowered to 10 -10 ton/yr. Emissions
may be overpredicted by a factor of 15 to 20, as indicated by back calculations
using a simple diffusion trajectory model. Isoprene, when compared to the mono-
terpenes, is much more efficient in producing ozone through phooxidation in
the presence of NO . This greater ozone production apparently occurs because of the
large amount of carbon consumed in the formation of aerosols for the monoterpenes.
Since rural areas have very low levels of NO , vegetative emissions may in fact
act as sinks for ozone rather than as sources. All areas investigated show very
low levels of natural hydrocarbons, suggesting that even if N0x were available,
very low levels of ozone would be produced. Air quality is thus not found to be
significantly affected by vegetative emissions.
17.
KEY WORDS AND DOCUMENT ANALYSIS
a.
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COS AT I Field/Group
Air pollution
Biological productivity
Hydrocarbons
Ozone
13B
08A
07C
07B
8. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
EPA Form 2220-1 (9-73)
19. SECURITY CLASS (7
UNCLASSIFIED
67
20. SECURITY CLASS (Thispage)
UNCLASSIFIED
22. PRICE
59
------- |