&EPA
United States
Environmental Protection
Agency
Environmental Sciences Research
Laboratory
Research Triangle Park NC 27711
EPA-600 3-79-081
September 1979
Research and Development
Photochemistry of
Some Naturally
Emitted Hydrocarbons
-------
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 ECOLOGICAL RESEARCH series. This series
describes research on the effects of pollution on humans, plant and animal spe-
cies, and materials. Problems are assessed for their long- and short-term influ-
ences. Investigations include formation, transport, and pathway studies to deter-
mine the fate of pollutants and their effects. This work provides the technical basis
for setting standards to minimize undesirable changes in living organisms in the
aquatic, terrestrial, and atmospheric environments.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/3-79-081
September 1979
PHOTOCHEMISTRY OF SOME NATURALLY EMITTED HYDROCARBONS
BY
Robert R. Arnts
Bruce W. Gay, Jr.
Atmospheric Chemistry and Physics Division
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 277.11
-------
DISCLAIMER
This report has been reviewed by the Environmental Sciences Research
Laboratory, U.S. Environmental Protection Agency, and approved for publica-
tion. Mention of trade names or commercial products does not constitute
endorsement or recommendation for use.
ii
-------
PREFACE
Some scientists have estimated that the atmospheric loading of non-
methane hydrocarbons by natural sources (i.e. emissions from vegetation and
soils) is significant or even overshadows anthropogenic sources when
totaled on a global basis. The emission of volatile hydrocarbons into the
atmosphere is of concern since these compounds can react photochemically
in the presence of oxides of nitrogen to form ozone and other oxidants.
Elevated oxidant levels in the lower troposphere are undesirable since these
compounds can jeopardize human health and cause damage to vegetation and
materials. Although the absolute 'tons'per year' estimates of natural
hydrocarbon emissions and the relevance of such estimations are still very
much open to question, scientists generally agree that emissions are quite
high. It is therefore of interest to understand what role natural hydro-
carbon emissions play in atmospheric chemistry to better assess their
oxidant forming potential.
Although it is not possible to simulate exact atmospheric conditions,
laboratory smog chamber experiments can provide an opportunity to study the
behavior of gaseous ppllutants in a controlled environment. Scientists have
conducted smog chamber studies to investigate the photochemistry of diluted
auto exhaust as well as the individual hydrocarbon components of auto ex-
haust and gasoline.
In this technical report, similar smog chamber experimental techniques
were employed to characterize the photochemical behavior of volatile
naturally emitted hydrocarbons. Although one cannot predict the ambient
ozone burden from a quantity of naturally emitted hydrocarbons, smog
chamber results can be used to compare the ozone producing ability of these
compounds with that produced by anthropogenic hydrocarbons.
iii
-------
ABSTRACT
Six CiriH,, monoterpenes, a C-_H-c aromatic, and isoprene, all known
10 lo J.U lj
or thought to be emitted to the atmosphere by vegetation, were irradiated in
the presence of NO . The terpenes studied included one acyclic triolefin
X
(myrcene), two monocyclic diol^firis (d-liinonene, terpinolene), and three
3
bicyclic monolefihs (a-piriene, $-pirierie, and A -carerie). The other biogenic
hydrocarbons studied were p-cymerie (p-isopropyl toluene) and isoprene (2-
methyl-1,3-butadiene); Propylene was also studied since this olefin serves
as a point of reference with Othet chamber Studies.
Results showed that the monoterpenes and isoprene promoted the oxidation
of NO to NO arid were themselves consumed at rates Comparable to or greater
than propyierie; p-cymene was decidedly slow in these respects. The mono-
terpenes however did not permit the buildup of ozone due to their rapid
reaction with ozone. The ozone suppression was particularly noticeable at
high carbon/NO ratios. Propylene and isoprane were more efficient in
producing ozone, but exhibited some suppression of ozone at high carbon/NO
X
ratios. Para-cymene produced a uniform concentration of ozone independent of
the carbon/NO ratio; this is due to its low rate of reaction with ozone.
-Ok
Based on this study, speculation on the photochemistry of forest airsheds
is presented. Deciduous forests, isoprene emitters, are expected to con-
tribute more to ozone production relative to the monoterpene producing
coniferous forests. Coniferous forests may in fact function as a sink for
ozone. Reported ambient concentrations of isoprene and terpenic hydro-
carbons in forested areas ace too low to account for more than a few ppb
of ozone even if NO is available.
x
iv
-------
CONTENTS
Preface ill
Abstract iv
Figures vi
Tables viii
Acknowledgement ix
1. Introduction 1
2. Conclusions 7
3. Recommendations 8
4. Experimental 10
5. Results and Discussion 15
Reactivity 15
Product analysis 28
Effect of vegetation on air quality 45
References
Appendices
A. Determination of ozone-isoprene rate constant 67
B. Hydrocarbon-NO irradiation time-concentration profiles
Index * , 70
Figures 73
v
-------
FIGURES
Number Page
1 Schematic of long path infrared photochemical
reaction chamber system 13
2 Effect of hydrocarbon to NO ratio on ozone ._
maximum* • ••
3 Hydrocarbon ozonolysis rate VS. maximum ozone 18
4 Ozone production efficiency relative to carbon
consumption vs C/NO
3t
5 Effect of C/NO ratio on hydrocarbon loss 25
A,
26
6 Effect of C/NO ratio on time to 0 maximum
x 3
7 Effect of C/NO ratio on NO loss 27
x
8 Spectrum of products of isoprene (15 ppm) and NO
(7.5 ppm) 45 minute irradiation
9 Spectra at 216 meters path of photooxidation products
methacrolein, methyl vinyl ketone, and irradiated
isoprene
10 Spectra of alpha-pinene irradiation 34
11 Spectra of beta-pinene irradiation 35
3
12 Spectra of A -carene irradiation 36
13 Spectra of myrcene irradiation 37
14 Spectra of limonene irradiation 38
15 Spectra of terpinolene irradiation • 39
16 Spectra of isoprene irradiation 40
17 Role of monoterpenes in atmospheric chemistry and role
of isoprene in atmospheric chemistry 47
vi
-------
18 Total concentration of the C-- terpene hydrocarbons
(ppbC) observed above the canopy of a loblolly pine
forest 54
19 Total concentration of the C1f) terpene hydrocarbons
(ppbC) observed in the canopy of a loblolly pine _,.
forest
20 Ozone isopleths predicted from Dodge's photochemical s-
model
vii
-------
TABLES
Number Page
1 Natural organics: their occurrence in the biosophere. 6
2 Hydrocarbon reactivity parameters and some selected rate
constants 20
3 Results of irradiated propylene-ct-pinene/NO and NO
systems at low concentration 23
4 Reactivity and products by long path infrared at
60 minutes 41
viii
-------
ACKNOWLEDGEMENTS
We wish to acknowledge the many helpful discussions with Dr. Joseph
J. Bufalini concerning this effort. The GC-MS analyses conducted by Mr. D.
Dropkin were of great aid in confirming suspected products. We would
also like to thank Dr. M. C. Dodge for her computer modeling efforts,
Mr. F. Litten for plotting many of the graphs, Mr. W. A. Lonneman for
his assistance in gas chromatographic analyses, and Mrs. A. McElroy for
typing this document. In addition we also like to thank Mr. J. Saunders
for programming the computer-plotter and to Mr. T. Winfield for his
assistance in conducting some of the irradiations.
We also are indebted to Mr. Frank Mitch of SCM-Glidden Organics and
Mr. I. W. Taylor of International Flavors and Fragances for providing us
with a fine selection of terpenoid compounds.
ix
-------
SECTION 1
INTRODUCTION
Volatile hydrocarbon emissions from natural sources have recently been
examined (1). Identifiable emissions can be roughly divided into two types.
Type I hydrocarbons are those emitted directly from living matter. Type II
are emitted as a result of the destruction of living matter. Hydrocarbons
in type I, which are our main concern in this study, are the terpenoids
(C,nH,,) and isoprene (CCH0) formed in green plants. Type II hydrocarbons
lU J-O _)o
are formed in the microbial decay of organic matter and in forest fires
where large molecular weight organic molecules are broken down to lower
molecular weight compounds.
Estimates of atmospheric loading of biogenic nonmethane hydrocarbons
have been based exclusively on isoprene and terpenoid emission (1-3). Other
sources of natural hydrocarbons, for example stress evolved ethylene (4-6)
from green plants have been discounted as being too small or unimportant
On the basis of background air analysis. Estimates show between 0.5 to
20 times the global atmospheric hydrocarbon emissions are due to natural
sources...assuming that anthropogenic emission estimates are accurate (1,7).
However ambient air analysis in suburban areas show only trace amounts of
isoprene or monoterpenes. Paradoxically in rural areas where one would
expect to measure greater concentrations of natural emissions, ambient air
analyses show a low contribution of isoprene and monoterpenes relative to
the predominate C^-C alkanes, benzene, toluene, and acetone (8). The
possible explanations for this discrepancy are that the emission estimates
are inaccurate for natural and/or anthropogenic hydrocarbons, or natural
hydrocarbon emissions are reacting much faster than anthropogenic emissions
and are removed from the atmosphere thus escaping detection.
Current field and laboratory evidence does not support the hypothesis
that rural ozone is a result of natural hydrocarbons. Field studies have
-------
shown ozone and its precursors can be transported hundreds of kilometers
from urban centers (9-12). The high ozone concentrations found in rural
areas can be explained by the transport of anthropogenic emissions.
Despite the uncertainties associated with natural hydrocarbon emission
estimates, it is of importance to examine the atmospheric chemistry of these
compounds. Isoprene (2-methyl-l,3-butadiene) has been studied in the 1950fs
along with numerous other gasoline and auto exhaust related hydrocarbons (13).
Isoprene is found in gasoline and auto exhaust at trace levels but is not a
significant emission from these sources. When isoprene was irradiated with
ultraviolet radiation in the presence of NO, Schuck and Doyle (13) observed
moderately fast reaction rates. Isoprene reacted one-third as fast as trans-
2-butene and one-fourth as fast a tetramethylethylene. Ozone was produced
in the photooxidation of isoprene at a comparable concentration relative to
other C -Cg olefins at a carbon/NO ratio of 15. Using long path infrared
spectroscopy several products were identified including formaldehyde, carbon
monoxide, peroxyacetylnitrate (PAN), and acrolein, Glasson and Tuesday (14)
classified hydrocarbon reactivity according to their ability to photooxidize
NO to N0_. They found isoprene (diolefins) photooxidized NO faster than C.-C,
paraffins, njonoalkyl benzenes, terminal olefins, and dialkylbenzenes. Hydro-
carbons which reacted faster than isoprene included the cyclohexenes, un-
substitute4 internal olefins, monosubstituted internal olefins, cyclopentenes,
and disubstituted internal olefins.
Few studies have been reported which compare the photochemistry of the
monoterpenes with the commonly studied C^-C.^ auto derived hydrocarbons.
Stephens and Scott (15) were the first to incorporate monoterpenes in
a study of hydrocarbon photochemistry. These investigators carried out
irradiations at hydrocarbon and NO concentrations of 5 ppm (v/v) each.
The rates of hydrocarbon disappearance and formation of PAN and aldehydes
were also measured. They found that the monoterpene alpha-phellandrene
(p-mentha-1, 5-diene) was slightly faster than tetramethylethylene and
six times faster than pinene. The isomer(s) of pinene were not specified
in the study.
-------
More recently Grimsrud, et al. (16) conducted irradiations of a number
of terpenes and isobutene at 10 ppb hydrocarbon, 7 ppb NO concentration.
Rates of reaction of ozone with each of the terpenes were also measured.
The study showed considerable variation in hydrocarbon reactivity which
depended on the number and position of double bonds in the molecule and the
degree of substitution on the olefinic carbons. Darnall, et_ al. (17) derived
rate constants for the reaction of terpenes and hydroxyl radicals from the
data of Grimsrud, et_ al_. (16). Recently a reactivity scale has been compiled
for a large number of hydrocarbons based on reactivity with hydroxyl radicals
(17,18). The importance of the hydrocarbon reaction rate with hydroxyl
radicals in this scheme is the direct relationship of the hydrocarbon's
ability to participate in chain propagating reactions leading to ozone
formation. The hydroxyl radical reaction with terpenes were by far the
fastest of all hydrocarbons studied.
The monoterpenes, based on past studies, can be classified as moderately
reactive to very reactive (15,18). Reactivity in these studies is a measure
of the rate of hydrocarbon loss during photooxidation, the rate at which
nitric oxide is oxidized to nitrogen dioxide in the photooxidation of the
compound, or the rate of hydroxyl radical reaction with the hydrocarbon. The
effective use of hydrocarbon reactivity obtained in the laboratory to describe
ambient air situations depends on the direct relationship between hydrocarbon
reactivity and the photochemical oxidant potential of the hydrocarbon.
Bufalini, et al. (19) recently pointed out that the relationship between
reactivity and oxidant formed is probably valid if the hydrocarbon and reaction
products remain in the gas phase. If a partially oxidized hydrocarbon is
removed from the gas phase system by a physical process such as aerosol
formation, its chain propagating behavior is essentially terminated and
oxidant potential is never realized.
Many aromatics and monoterpene hydrocarbons have been found to produce large
amounts of aerosols in photochemical reactions. The monoterpene hydrocarbons
also produce aerosols in the dark reaction with ozone. Went (20) in 1960
first demonstrated the ability of alpha-pinene and ozone to react and form
light scattering aerosols. He speculated that natural hydrocarbon
emissions of alpha-pinene and other terpenes from forested areas could
-------
after reaction with ambient ozone, form the haze observed in the lustily
forested areas of the Smoky Mountains of western North Carolina/eastern
Tennessee.
A number of studies have demonstrated the ability of monoterpenes,
mostly alpha-pinene, to form aerosols (21-30). Other investigators have
also studied the aerosols formed in the dark reaction of terpenes with
ozone (20, 22, 31) and when terpenes were irradiated with ultraviolet
radiation in the presence of NO (21-30). Using infrared analyses col-
Jv
lected terpene aerosol samples were characterized by the carbonyl (-C = 0)
stretching frequency (22-24). Laboratory samples of alpha-pinene reaction
product aerosols were collected and prepared for gas chromatographic-mass
spectrometric analysis using traditional acid-neutral-base extractions
and derivatization techniques. Although a large number of products were
present in the aerosol as evident in the total ion chromatograms, only two
products were identified (26). Present in the acid fraction were pinonic
acid, a cyclic keto-carboxylic acid. Pinononic acid is a homologous acid
resulting from the oxidative decarboxylation of pinonic acid. Pinonic
acid has also been identified in aerosol samples collected in the forested
area of the Pisgah National Forest thirty miles northeast of Asheville,
North Carolina (30). Unfortunately the compositional make up of the
aerosol at this forested area was not determined. Had the total composition
of the aerosol been identified, this information would have been of great
value in determining the nature of the haze in this region of the country.
More recently, however, Weiss, et al. (32) have shown that the bulk of the
light scattering aerosol in the midwestern and southern United States is
primarily sulfate and not carboneous. The source of the sulfur has not
yet been determined.
The aerosols of terpinolene and limonene formed by photooxidation
in the presence of NO have been examined utilizing direct probe mass
X
spectrometry (25). More than thirty products were observed which included
carboxylic acids, aldehydes, alcohols, peroxides, and the nitrate esters
of acids, alcohols, and peroxides.
The chemical behavior of automotive hydrocarbon emissions of C^-C^
paraffins, olefins, diolefins, and aromatics in smog chamber reactions
4
-------
have been related to molecular structure of the compound. The naturally
emitted hydrocarbons that were studied with the exception of isoprene and
p-cymene are structurally unlike the automotive hydrocarbons. The
naturally emitted monoterpenes are C -.H , isomers consisting of bicyclic
10 ID
olefins, monocyclic diolefins, and acyclic triolefins. An endocyclic
olefin such as alpha-pinene easily lends itself to the formation of a
dioxygenated compound through the oxidative cleavage of the double bond.
The difunctional oxygenated compound which might form is likely to con-
dense to form aerosols due to its low volatility caused by the high polar
character of the molecule. The reaction with ozone as well as other
chemical properties distinquish monoterpenes sufficiently from other
previously studied hydrocarbons to warrant further investigations.
The hydrocarbons shown in Table 1 were chosen for this study because
they were either measured in ambient air, detected as emissions from growing
vegetation or found in essential oils of a number of common plants. As
mentioned earlier propylene is included for comparison since it has been
extensively studied in smog chambers.
This study of naturally emitted hydrocarbons was undertaken to
determine the compounds photochemical reactivity, ozone-oxidant formation and
reaction products under simulated atmospheric conditions. These results
coupled with ambient measurements of natural hydrocarbon concentrations
can be used to assess the influence natural hydrocarbon emissions have
on rural and urban air chemistry.
-------
TABLE 1. NATURAL ORGANICS: THEIR OCCURRENCE IN THE BIOSPHERE
Compound
Structure
Reason for Selection
isoprene
p-Cymene
a-Pinene
3-Pinene
d-Limonene
Myrcene
Measured in ambient air in rural areas
(10, 33); emitted by oak, sycamore,
willow, balsam poplar, aspen, spruce,
and others (3,34,35)
Emitted by California black sage (36)
and from 'disturbed1 eucalyptus foliage
(35); found in the gum terpentines of
scotch pine and loblolly pine (37)
Measured in ambient air in rural areas
(33,38); emitted by numerous pines,
firs, spruce, hemlock, cypress (3)
Measured in ambient air in rural areas
(39); emitted by California black sage
(36); loblolly pine (40), spruce, and
redwood (3)
Measured in ambient rural air (39);
emitted by loblolly pine (40), Calif-
ornia black sage (39); and 'disturbed1
eucalyptus (35), found in the gum ter-
pentines of numberous pines (37) and
the essential oils derived from some
fruits (41,42)
Measured in air above pine needle litter;
emitted by loblolly pine (40), California
black sage (36), and redwood (3); found
in the gum turpentines of some pines(37)
Terpinolene
K>
A -Carene
-CX
No ambient measurements or emissions
reported; however, terpinolene is found
in the essential oils of numerous plants
(43) and in some pine gum turpentines
(37)
Measured in ambient air (39); found in
the gum turpentines of some pines (37)
arid in the essential oils of numerous
tower plants (43)
-------
SECTION 2
CONCLUSION
Laboratory smog chamber investigations show the biogenic hydrocarbons
isoprene and the monoterpenes are moderately to extremely 'reactive' when
irradiated in the presence of oxides of nitrogen. Historically, reactivity
has referred to the rate of hydrocarbon disappearance, the rate of
hydroxyl radical attack on the hydrocarbon, or rate at which nitric oxide
is oxidized to nitrogen dioxide; these parameters are used with the intent
of providing a measure of the hydrocarbons potential to participate with
nitrogen oxides and sunlight to produce ozone. This study demonstrates
that the monoterpenes while reactive by these definitions are inefficient
ozone precursors. This is believed to be due in large part to (1) the
fast reaction of ozone with the terpenes and (2) the efficient formation
of aerosol from the photooxidation products. These effects manifest
themselves in a most pronounced manner at high C/NO ratios (50-200).
X
The few measurements of background ambient NO and detailed non-
X
methane hydrocarbons in the literature show a wide range of values; the
variations in reported values probably reflect real uncertainities in
the analytical techniques and true geophysical fluctuations. Thus
estimates of background NMHC/NO ratios exhibit a wide range assuming
X
reported NMHC concentrations (mostly CL-C,. alkanes, benzene, toluene,
and xylene) ranging from 40 to 100 ppbC within the United States and NO-
concentrations on the order of 0.015 ppb to approximately 0.1 ppb. Never-
theless, this range represents conditions where monoterpenes if present
would not contribute more than 1 or 2 ppb 0 given such an NO poor system.
•j X
-------
SECTION 3
RECOMMENDATIONS
It is uncertain whether the terpenes will behave at atmospheric con-
centrations like they behaved at higher chamber conditions. Laboratory
simulations do however provide evidence for mechanisms via the products
observed and thus provide a basis for atmospheric modelers to predict real
world conditions. Of particular concern in this study is the relationship
of aerosol formation to initial hydrocarbon concentrations. If at atmospheric
concentrations terpenes are not converted to aerosols at the apparently high
conversions inferred from chamber studies then it is possible that the partially
oxidized terpenes will remain in the gas phase and continue to participate in
photochemical processes. The end result may be that at atmospheric concen-
trations terpenes may be more efficiently producing ozone than is indicated
at high terpene concentrations. Whether or not this is true depends in part
upon the vapor pressures of the terpene oxidation products. Only fragmentary
information exists about the complex composition of the terpene aerosol.
However we do know that oxygenated compounds (acids, aldehydes, and ketones)
in the C _ range are solids at room temperature and display low volatility.
Furthermore recent work by other investigators has identified a number of C_
thru GIO dicarboxylic acids in aerosol samples collected in the Los Angeles
basin; little or none of these compounds were detected in the gas phase. It
is therefore likely that the terpenes when present at ppb levels will also
produce aerosol efficiently. Nevertheless there is need for cleaner reaction
chambers so that irradiations may be conducted at atmospheric levels free
from any wall contamination problems.
There is a need for more detailed hydrocarbon analysis of air of both
coniferous and deciduous forested rural areas with accompanying ozone and
accurate nitrogen oxides data. These studies would help to define the
relative importance of natural and anthropogenic hydrocarbon contribution to
-------
ambient ozone. In particular the rural hardwood forested area of northeastern
United States, an isoprene source area, should be examined for isoprene con-
tribution to the non-methane hydrocarbon concentrations since isoprene has the
greatest potential of the biogenic hydrocarbons for promoting ozone formation.
Another approach to assessing the impact of natural hydrocarbon emissions
on ambient ozone could be a field investigation of air masses passing over
sparsely inhabited forested areas. By use of aircraft equipped to collect
integrated samples for hydrocarbon analysis and real time monitoring of ozone
and nitrogen oxides the chemical dynamics of an air mass could be observed as
it passes over forested areas. Such an experiment could be simplified by
examining the hydrocarbon poor plume of a nitrogen uxides emitting fossil
fuel power plant which is dispersing over a forested area. The introduction
of a fresh source of nitrogen oxides to the normally nitrogen oxides poor
rural air mass would allow the use of commercially available monitors which
have traditionally suffered from a lack of sensitivity to low rural levels.
Gaseous tracers such as SFfi could be released concurrently at the power plant
stack to follow the plume. The successful monitoring of a dispersing plume
would help to provide modelers with sufficient information necessary to predict
the impact of a forested area on passing air masses.
It is also recommended that any air samples collected for later terpene
analysis be evaluated for loses due to reaction. In the case of whole air
samples it is recommended that the ozone in the collected air be quenched
immediately upon introduction to the sampling vessel and that the sample
be stored in the dark to prevent photochemical reaction; the introduction
of a small quantity of nitric oxide to a collection bag covered with black
polyethylene has proven to be quite effective.
Lastly, it is recommended that hydrocarbon 'reactivity scales' based
on rate of hydrocarbon loss, nitric oxide loss, hydroxyl radical reaction,
etc. be used with some caution. Such scales have shown terpenes to be
moderately to highly reactive. However, this study has demonstrated that
these compounds, while promoting rapid NO oxidation, suppress ozone formation.
-------
SECTION 4
EXPERIMENTAL
All of the hydrocarbon/NO photooxidation smog chamber-type experiments
2t
with the exception of those carried out in the long path infrared reaction
chamber were conducted in 250 liter (2 mil) FEP Type A Teflon bags. The bags
were made by heat sealing a folded sheet (18.9 cm width by 42.5 cm length)
of the Teflon film using a commercial heat sealer. After two sides of the
bag are sealed a two inch square of pressure sensitive tape, which helps to
make an airtight seal for the inlet-outlet fitting, was placed at a suitable
location on the outside bag surface. A hole punched through the tape bag
surface accommodated a Teflon 0-seal straight thread connector which was
secured with a Teflon nut. Once the connector was in place the third side of
the bag was sealed and the airtight bag was now ready for conditioning before
use in experiments. Conditioning the bags was necessary because past experience
had shown low levels of hydrocarbon outgassing occurred with new bags. The
conditioning process involved filling the bag with 3 ppm ozone in clean air
and irradiating the bag in the smog chamber for 48 hours followed by flushing
with a source of clean air. A clean air source was provided by a laboratory
purification train which removed water, hydrocarbons, and nitrogen oxides.
Hydrocarbons were removed by combustion on a heated catalyst of rhodium on
aluminum (Engelhard) and adsorption on activated carbon. Nitrogen oxides were
removed by chemical absorption on Purafil (Borg Warner) and water vapor by
condensation. The resulting air from the system contained less than 2 ppb
nitrogen oxides and less than 50 ppb carbon as nonmethane hydrocarbons most
of which were propane and ethane. To introduce a controlled amount of hydro-
carbon or NO into the Teflon bag measured quantities of liquid or gas of the
X
compound were injected using a calibrated syringe through a septum in a glass
loop placed in-line between the air train and bag. For some of the less
volatile hydrocarbons the glass loop was gently heated to assure volatization.
10
-------
All hydrocarbons used in this study were of 95% or higher purity and obtained
from SCM Glidden, Aldrich, or Chemical Samples Company. Nitric oxide and
nitrogen dioxide were chemically pure grade obtained from Air Products.
The irradiation chamber in which the Teflon bag was suspended was an
aluminum box of 0.78 m width x 1.35 m length x 1.12 m height. Two banks of
40 watt ultraviolet fluorescent lamps evenly distributed along the two inner
sides of the chamber provided the necessary irradiation. In these experiments
twenty two General Electric F40 BLB filtered blacklamps with energy maxima
at 3660 A and four Westinghouse sunlamps with energy maxima at 3160 A were
used to simulate lower atmospheric solar radiation between 290 and 380 nan-
ometers. The light intensity was measured by the photolysis of NCL in
nitrogen (44). The equation employed for this was/-d(NO )/dt = k (NO.).
Since k, = 1.5 k.,, this first-order dissociation constant is a measure of
d l -1
the light intensity. The k, value was 0.45 min . The irradiation chamber
was equipped with an air conditioner which cooled and circulated air
maintaining a temperature of 25 + 2 C during the irradiation.
Ozonolysis experiments of isoprene and propylene were also carried out in
Teflon bags thermostated at 25 + 2°C in the irradiation chamber but in the
dark. Ozone was produced by exposing purified air flowing through a high
voltage discharge ozonator. At a constant air flow through the ozonator the
amount of ozone produced could be varied by varying the applied voltage.
The C--C- oxygenates, propylene, isoprene, and the terpenes were analyzed
by gas chromatography with flame ionization detection. Propylene, isoprene,
and the oxygenates (acetone, acetaldehyde, and propionaldehyde were separated
on a 46 cm x 3.2 mm diameter stainless steel column packed with 60-80 Mesh
Porapak Q. The terpenes were separated on a 1.83 mX 3.2 mm diameter
stainless steel column packed with 10% SF-96 on 6/80 mesh acid washed
Chromosorb W. Both columns were operated isothermally at 100 C. A
Perkin-Elmer model 900 gas chromatograph was modified with two solenoid
actuated Seiscor gas sampling valves (Seismograph Co.). A sensitivity of 50
ppbC could be obtained on this system using a 5 cc gas sample loop.
Peroxyacetylnitrate (PAN) produced in the irradiations was measured by
electron capture gas chromatograph. The system consisted of a Seiscor sampling
11
-------
valve with a 5 cc gas sample loop and a chromatographic column of 10% Car-
bowax 600 on Gas Chrom Z packed in a glass column 90 cm X 3.2 mm. A sensi-
tivity of 0.5 ppb PAN was obtained using a wide range scandium tritide
electron capture detector (Analog Technology Corporation). Both the column
and detector were operated at room temperature. The system was calibrated
with samples of PAN analyzed by long path infrared.
Ozone produced in the hydrocarbon photooxidation reactions or in the
ozonolysis experiments was monitored with a Bendix (model 8002) instrument
using the gas phase chemiluminescence of the ethylene-ozone reaction. Nitro-
gen oxides were measured using a Bendix (model 8101B) NO analyzer via the
X
gas phase chemiluminescence of the ozone-NO reaction. Both instruments were
calibrated using standard gas phase titration techniques of NO with ozone.
Wet chemical analyses techniques were also employed during the irradiation
experiments. The concentration of formaldehyde produced in an irradiation of
a hydrocarbon/NO mixture was measured using the chromotropic acid (45) method
x
and the Saltzman (46) method used to check the NO- concentration against
measurements made by the chemiluminescent instrument.
The ultraviolet irradiations of terpenes were also carried out in a
photochemical chamber which also served as an absorption cell for in situ
infrared measurements. Figure 1 is a schematic of the system. The chamber
is constructed of borosilicate glass pipe (Corning QVF-PS 12-60). Six sections
of pipe are connected together separated by 1.9 cm aluminum spacers with 0.32
cm Teflon gaskets on either side giving a total length of 9.1 m and an internal
diameter of 0.31 m. The five aluminum spacers have inlet ports in them and
were connected via a manifold to a standard gas handling system. The manifold
distributes gaseous sample uniformally throughout the cell. .The chamber ends
were capped with 3.18 cm thick Plexiglas flat plates using Teflon gaskets.
The internal volume of the chamber was 690 liter. The chamber could be
evacuated to a pressure of less than 1 torr with a large displacement vacuum
pump connected to one end plate with a ball vacuum valve. Samples of volatile
materials or gases were mixed in known volume bulbs and introduced into the
chamber through the glass manifold; terpenes low in volatility were vaporized
into the gas phase by injecting liquid sample with a calibrated syringe through
12
-------
ULTRAVIOLET
Figure 1. Schematic of long path infrared photochemical reaction chamber system.
-------
a septum into a glass line on the gas handling system. The injections were
made while air flowed through the gas handling system into the chamber. The
immediate area of the gas handling system near the septum was gently heated
to assure complete vaporization. Surrounding each of the six chamber sections
were two half-section cylindrically constructed light hanks. Each half section
contained six General Electric F-40 BLB and two Westinghouse Sun ultraviolet
fluorescent lamps. The sixteen lamps around each chamber section were
arranged and controlled to give uniform radiation intensity when all, one
half or one fourth of the lamps were turned on. In these experiments a total
of 48 black and 24 sun lamps were used. The light intensity with these lamps
on as measured by the photolysis of N0_ in nitrogen was kd/NO ) = ^*^1 min
except where otherwise noted.
This chamber used as a photochemical reactor also contained within it an
eight-mirror optical system (47). Thus, the chamber served also as a long-
path multiple reflection infrared absorption cell capable of path lengths in
excess of several hundred meters. The optical path length used in these
experiments was 357 meters. Potassium bromide entrance and exit windows were
used on the cell. The infrared instrument was a rapid scan Fourier Transform
Spectrometer (Digilab FTS-20) that uses a Michelson interferometer with a KBr
beam splitter coated with germanium. Liquid nitrogen cooled detectors were
used to cover the spectral region from 700-3400 cm . A more detailed dis-
cussion of the optical system, interferometer, and spectra data techniques
are given elsewhere (48).
14
-------
SECTION 5
RESULTS AND DISCUSSION
REACTIVITY
Nine hydrocarbons were irradiated with UV radiation in the presence of
NO and their photochemistry studied. Except for propylene all of the selected
JL
hydrocarbons are volatile compounds either identified or suspected of being
emitted by vegetation. Propylene an auto exhaust related hydrocarbon was
included for comparison purposes since its photochemical reactivity has been
well characterized in other studies and will serve as a benchmark to gauge
the reactivity of the other compounds.
Isoprene, p-cymene, alpha-pinene, d-limonene, and propylene were studied
at a number of carbon to NO ratios (ppmC/ppm NO ) ranging from 2 to 200 with
X X
a constant NO concentration of 0.33 ppm. Myrcene, terpinolene, and beta-
X 3
pinene were investigated at carbon/NO ratios of 30 and 200 and A -carene at
2C
a ratio of 30. In an urban atmosphere C/NO ratios can be 6:1 during morning
X
rush hour traffic (personal communication, W. A. Lonneman). As the urban
plume ages, while being transported into suburban and rural areas, the NO
X
is depleted increasing the C/NO ratio. In remote areas of the continental
x
United States C/NO ratios are in the range 400 to 6600. The significance
X
of these ratios with respect to the terpenes will be discussed later. The
choice of experimental conditions of 2 to 200 reflects the wide range of
ratios observed in the extremes of urban and remote atmospheres. The selected
experiments performed at C/NO ratios of 30 and 200 were chosen for two
distinct reasons. The ratio of 30:1 was chosen to approximate the study of
Westberg (39) and represents the best case for ozone production. The ratio
of 200:1 was judged as appropriate to represent a real rural/remote NO poor
X
atmosphere. The results of these Teflon bag smog chamber irradiations are
plotted as concentration versus time profiles in Appendix B (figures 1B-55B).
15
-------
It should be noted that N0_ concentration data acquired after the N02 maximum
is increasingly in error with irradiation time. This error is due to the non-
specificity of the heated carbon converter in the chemiluminescence monitor.
Nitrogen containing species such as peroxyacetylnitrate and nitric acid have
resulted in varying unpredictable conversions to nitric oxide based on con-
verter history (49,50). Wet chemical measurements of NO obtained after the
chemiluminescence NO peaked measured as little as 10% of the instrument
response value as actually NO .
Smog chamber data are examined with respect to ozone producing ability
as shown in Figure 2. This figure clearly illustrates and compares the
varying ability of isoprene, d-limonene, alpha-pinene, p-cymene and pro-
pylene to drive the hydrocarbon-NO photochemical system to ozone generation.
3C
With the exception of p-cymene all the hydrocarbons exhibit maximum ozone
generation at a carbon/NO ratio in the range of 10-20. This ratio range
jb
is in good agreement with the work of Westberg (39) who observed optimum
ratios for terpinolene, d-limonene, and alpha-pinene at 25 using initial
NO concentrations of 55 and 170 ppb. Para-cymene produces a fairly con-
A
sistent 410 ppb ozone as its initial concentration is increased to give
higher carbon/NO ratios. The effect of HC/NO ratio on maximum ozone can
X 2C
be explained by the relative contributions of ozone formation and destruction
processes occuring in the photochemical mechanism. Since the photolysis of
NO- is the only ozone forming step in the mechanism any reaction which
accelerates the conversion of NO to NO will ultimately increase ozone. A
key step in the formation of NO is the reaction of NO by alkoxy and acyl
radicals formed by the hydrocarbon-OH reaction. Consequently as the hydrocarbon
concentration is increased the oxidation reaction of NO to NO is increased.
2
The peaking of ozone production and the subsequent ozone decline after carbon/
NO ratios of 10-20 are reached results from the increasing influence of the
ozone-olefinic reaction. This is especially apparent at high carbon/NO ratios
where the ranking of the compounds according to ozone production are directly
related to the hydrocarbon-ozone reaction rate. Figure 3 a plot of log
kozonolvsis versus ozone maximum at carbon/NO ratio of 200 shows a high
correlation (R = 0.98) between the reaction rates with ozone and the
16
-------
3. 300
10
J f-
7 /
CHAMBER CONDITIONS:
Kd = 0.46 miiT1
A PROPYLENE
D ISOPRENE
O d-LIMONENE
a-PINENE
p-CYMENE
50 60 70 80 90
C/NOX RATIO, pprnC VS ppm [JOX; (NOX)0 - 0.33 ppm
Figure 2. Effect of hydrocarbon to NOX ratio on ozone maximum.
100
D
,^A
7 r
200
-------
106
105
10*
MYRCENE I
TERPINOLENE
d-LIMONENE
• a-PINENE
0-PINEWE
E • \« PROPYLENE
ISOPRENE
<
V)
DC
-------
maximum ozone formation in the photochemical reaction of these hydrocarbons.
The fact that p-cymene an aromatic exhibits an apparent plateau for maximum
ozone after an optimum C/NO ratio is consistent with its observed slow
x
reaction rate with ozone. At carbon/NO ratios of 200 p-cymene reacts
X
so slowly with ozone relative to the ozone forming process that very
little ozone is depleted. In contrast, the other terpenes are reacting
10 to 10 times faster with ozone than is p-cymene (see Table 2).
Another semi-quantitative way to compare the ozone forming potential
of the hydrocarbons being studied is to ratio the ozone produced to the
carbon consumed at the time of maximum ozone concentration. These data
are presented in a log-log format shown in Figure 4. The terpenes d-
limonene and alpha-pinene at the optimum carbon/NO range of 10-20 are
x
producing 1.5 to 3 ppb ozone per 20 ppb carbon consumed. Westberg (39)
observed between 0.6 to 2 ppb 0 per 20 ppbC consumed for these terpenes.
Propylene and p-cymene at the same carbon/NO ratio produce between 2.9
X
to 5.2 ppb ozone per 20 ppb carbon consumed. Using the same experimental
conditions at a C/NO ratio of 10 Dodge (51) predicted by computer modeling,
X
an ozone maximum of 880 ppb at 150 minutes which results in the formation
of 5.3 ppbO_ per 20 ppbC consumed. The model compares favorably with the
irradiation at C/NO of 9.0 where 700 ppb 0. were produced at 140 minutes
X j
with the formation of 4.2 ppbO /20 ppbC. It should be noted that the
photochemical model tends to over predict ozone concentrations by about
20-30% for most HC/NO systems. The reasons for this over prediction is
X
currently under study.
Isoprene is less efficient than propylene in producing ozone on a
per carbon basis yielding 2.4 to 3.5 ppbO.,/20 ppbC.
To verify if propylene and a-pinene have the same relative degree of
photochemical reactivity at lower concentrations as was observed at higher
concentrations, further experiments were performed. Hydrocarbon contamination
from the walls of the Teflon bag used as the reaction chamber becomes
significant over long irradiation periods, which necessitated a series of
low concentration hydrocarbon/NO experiments to observe ozone produced.
X
Table 3 summarizes the results of these low concentration irradiations. As
19
-------
TABLE 2. HYDROCARBON REACTIVITY PARAMETERS AND SOME SELECTED RATE CONSTANTS
co
o
Compound
propylene
r
isoprene
MMMV^H^^H^^^MNkMMOVMMM
p-cymene
-
-------
Table 2. (continued)
Compound
(NO
limonene
^CK
3 -pinene
xrf
terpinolene
-OK
myrcene
)=O~^v
3
A -carene /
-fy
^\_y
C/NOX
)= 0.33 ppm
30
200
30
200
30
200
30
200
30
NA
0_ Maximum,
ppb
309
3
255
125
240
15
322
1.5
250
—
Time to 0_
max , min .
120
10
158
50
100
10
240
10
70
—
Hydrocarbon
loss , ppb
mm
39
111
9.7
45
35
97
63
130
22
—
NO loss
min
—
114
5.1
10
40
44
21
36
8.6
—
Rate Constants
H mol sec
kOH=
ko =
V
3
kOH=
ko =
V
kn
3
V
ko
3
k =
k
°3
i n
9.0 X 10
10
8.8 X 10
6.5 X 1010
3.9 X 105
4.0 X 10 U
1.5 X 1010
2.2 X 104
= 4.4 X 105
6.1 X 106
13.7 X 1010
£-
= 7.6 X 10
5.2 X lO1^
= 7.4 X 104
a
c
d
b
a
c
d
b
e
b
c
b
c
b
a Reference: 52
e Reference: 54
i Reference: 56
b Reference: 16
f Reference: 24
c Reference: 52(derived from Ref. 16) d Reference: 53
g Reference: 55 h Reference: This
work. See
Appendix 1
-------
O P-CYMENE
D PROPYLENE
0 100 200
RATIO C/NOX, (NOX)0 = 0.33 ppm
Figure 4. Ozone production efficiency relative to carbon consumption
vs. C/I\IOX.
22
-------
observed at the higher concentrations, the propylene/NO system produces
X
more ozone than the a-pinene/NO system under similar conditions. These
X
sets of experiments demonstrate that doubling the C/NO ratio corresponds
X
to doubling of carbon which increases the ozone production. This agrees
with the higher concentration experiment where maximum ozone production
occurred as the C/NO ratio approached 30:1.
X
TABLE 3. RESULTS OF IRRADIATED PROPYLENE-a-PINENE/NO
AND NO SYSTEMS AT LOW CONCENTRATION
X
PROPYLENE
C/N°
x
ppbC
ppb
min
ppb
min
7.5
4.6
11
16
88
89
97
184
11.7
19.5
8.7
11.5
29
23
27
37
49
23
48
48
85
50
86
88
ALPHA-PINENE
C/NO [C] .
^e^ T
-el?. JL.
4.6 110
8.4 101
12.2 123
16.5 198
__
__ -~ —
[N0x].
24.0
12.2
9.6
11.5
NO
X
[NO ] .
L XI
9.2
10.6
C°3]60min !
11
16
14
21
BLANK
[03]60min '
6.1
5.5
[%N02]i
27
35
43
48
[% N021.
35
29
[%N°2Wn
57
72
76
91
[»2i««.
73
49
Other indicators used to characterize hydrocarbon reactivity include
the rate of nitric oxide oxidation or time to N0-N02 crossover, rate of
hydrocarbon consumption and related parameters such as time to 50% hydro-
carbon loss. In their reviews of photochemical air pollution Altshuller
23
-------
and Bufalini (57,58) found hydrocarbon reactivity based on rates of nitric
oxide oxidation and hydrocarbon consumption gave similar hydrocarbon
rankings. Recently Wu (59) reported that rates of hydrocarbon consumption
and nitric oxide oxidation correlated to a high degree with hydrocarbon-
OH rate constants.
Hydrocarbon loss rates from this study were plotted as a function of
carbon/NO ratios and shown in Figure 5. All of the terpene compounds dis-
x
played a high degree of reactivity except for p-cymene. This is consistent
with data in Figure 6 in which p-cymene reaches its ozone maximum at an
irradiation time ten times longer than the other terpenes all at the same
initial concentration. However, as seen in Figure 2 p-cymene produces 400
ppb or more ozone even at carbon/NO ratios of 200. Clearly hydrocarbon
X
consumption is not a reliable indicator of hydrocarbon reactivity with
respect to photochemical oxidant production because of the importance of
hydrocarbon ozonolysis.
Nitric oxide loss rates or NO to N00 conversion rates versus carbon/NO
4~ X
ratios are shown in Figure 7. Slow NO instrument response relative to the
j£
fast conversion of NO to N0_ for the highly reactive terpenes at high carbon/
NO ratios prevent accurate determinations of NO loss rates. However, the
low capacity for driving the NO photooxidation of p-cymene compared to the
other compounds is clearly shown in this figure.
On the basis of hydrocarbon-OH reaction a large number of hydrocarbons
have been classified relative to their rate of OH reaction (17, 18). Com-
pounds such as 3-carene, d-limonene, and myrcene are placed in the most
reactive group. Propylene, isoprene, alpha-pinene, beta-pinene and p-cymene
are placed in a group of lower reactivity. Although a scale based on OH reaction
rate may be an indicator of a hydrocarbon atmospheric lifetime and its
ability to oxidize NO it may not reflect the photochemical oxidant potential
of a compound or other important parameters such as aerosol formation, eye
irritation or plant damage. As reported earlier the most reactive compounds
at higher carbon/NO ratios do not allow ozone to build up because of their
X
extremely fast reaction with ozone.
An interesting phenomena was observed in some of the d-limonene irradi-
ations: a double ozone maxima. Figures 36-B thru 38-B show the double peaks
24
-------
C/J
V)
o
_1
z
o
00
cc
o
cc
o
50 -
25 -
50
100 150
C/NOX, (NOX)0 = 0.33 ppm
Figure 5. Effect of C/NOX ratio on hydrocarbon loss.
25
-------
1000 ?-
•I 100 —
x
ee
CO
o
10.0 —
1.0
100
• d-LIMONENE
O ISOPRENE
A PROPYLENE
Oa-PINENE
A p-CYMENE
p-CYMENE
ISOPRENE
»
d-LIMONENE :
a-PINENE
A PROPYLENE —
200
RATIO C/NO.
Figure 6. Effect of C/NOX ratio on time to 63 maximum.
300
26
-------
N)
"-4
o
•ft.
100 150
C/NOX, (NOX)0 = 0.33 ppm
Figure 7. Effect of C/NOX ratio on NO loss (maximum).
ZOO
-------
seen only at C/NO ratios between 45 and 50. This anomaly is reproducible
X
and was also observed in a new bag. New bags were used to verify that the
original experiments were not a result of leaks or contamination from
previous experiments. See figures 37-B and 38-B. Along with the ozone
maxima the NO- is seen to maximize with the first ozone peak, quickly
decrease, and then slowly rise again coincident with the second ozone
maximum. This effect is less pronounced with the new bag.
The double ozone peak might be the result of the fast reaction of
ozone with limonene; since its rate of reaction is two orders of magnitude
faster than with propylene. The reaction rate of OH with limonene is only
one order of magnitude faster than with propylene. Thus in the early stages
of the irradiation the ozone formed is soon suppressed by consumption of
the ozone-limonene reaction. As limonene is further reacted, the ozone
again begins to buildup creating the second ozone maximum. This effect
was tested by photochemical computer modeling the propylene/NO system and
X
adjusting certain rate constants (51). The rate constant for the propylene-
ozone reaction was arbitrarily increased by a factor of ten. This produced
in the modeled data an initial ozone maximum at 55 minutes followed by a
decrease in ozone concentration and a subsequent build up of ozone to a
maximum higher than the first.
PRODUCT ANALYSIS
The primary objective of this study was to assess the photochemical
oxidant potential of naturally emitted hydrocarbons. During the irradiations
product analysis were carried out. A complete product analysis of the pro-
pylene-NO irradiation was not carried out since good carbon balances have
x
been reported by other researchers (13,60). An olefin such as propylene for
example when photooxidized yields formaldehyde, acetaldehyde, peroxyacetyl-
nitrate, formic acid, carbon monoxide, and carbon dioxide.
Isoprene photooxidation studies and product analysis were carried out
by Schuck and Doyle in the 1950's (13). Their analysis of products formed
in the irradiation of an isoprene-NO mixture at 100 minutes accounted for 87%
X
of the reacted isoprene as formaldehyde, acetaldehyde, acrolein, peroxy-
acetylnitrate and carbon monoxide. Products were identified by their
28
-------
characteristic infrared absorption bands using long path infrared spectroscopy.
In our study of the irradiated isoprene-NO system using long path
X
infrared spectroscopy the following reaction products and associated absorption
frequencies were observed: acetaldehyde 2705 cm" , carbon dioxide 2355 cm ,
carbon monoxide 2160 cm , formaldehyde 2780 cm , formic acid 1105 cm" ,
methacrolein 932 cm , methylvinyl ketone 932 cm" , nitric acid 986 cm" ,
—1 —1
ozone 1055 cm and peroxyacetyl nitrate 1162 cm" . Figures 8 shows the
spectrum of the photooxidized mixture of 15 ppm isoprme and 7.5 ppm NO
after 45 minutes of irradiation. Shown in Figure 9 are the infrared spectra
of the pure compound methacrolein and methyl vinyl ketone and the reaction
mixture of photooxidized isoprene. No acrolein was detected in the photo-
oxidation of isoprene either by infrared spectroscopy or gas chromatography.
Mass spectral identification of methacrolein was also obtained for a photo-
oxidized sample of isoprene in addition to methyl vinyl ketone. The product
Schuck and Doyle identified as acrolein was most likely methacrolein.
Methacrolein is formed by reaction of ozone across the 3-4 carbon-carbon
double bond of isoprene.
CH, CH, H
I 3 I 3 I
CH = C - CH = CH9 + 0, -> CH = C - C = 0 + CH.O •
^ £m J £* £f £•
Reaction across the 1-2 carbon double of isoprene results in the formation
of methyl vinyl ketone.
CH, CH
,3 ,3
CH_ = CH - C = CH9 + 0» -> CH, = CH-C =0 + CH202 •
,£ ^ j Z. *• *~
These two oxygenated olefinic products are still very reactive. Their con-
centrations in an irradiation maximize before that of ozone as shown in
Figures 1-B to 9-B. Interestingly pyruvic aldehyde an expected common
product of ozone reaction with methacrolein and methyl vinyl ketone, was
not observed using long path infrared or gas chromatography.
CH H
I3 I
CH = C - C = 0 OH
2 II 1
ORCH + 03 -* CH3-C-C= 0 + CH202-
I 3
CH = CH-C = 0 (not observed) pyruvic aldehyde
29
-------
JOOO
1300
J600
1900
Figure 8. Spectrum of products of isoprene (15 ppm) and NO2 (7.5 ppm).
45 min. irradiation • kdNC>2 - °-7 min.'', 216 meter path length.
-------
METHACKOLEIDI
METHYLVINYL KETQNE
700 cm-J
1000
1300
J600
1900
Figure 9. Spectra at 216 meters path of photooxidation products
methacrolein (8 ppm), methyivinyl ketone (8 ppm), and irradiated
isoprene (15 ppm).
31
-------
In the photooxidation of isoprene, pyruvic aldehyde may photolyze quickly
and be present only at very low concentrations as a transient species.
The photolytic decomposition route of-pyruvic aldehyde is likely since
its lower homologue glyoxal readily decomposes in the UV to CO and
CH20 (61). The carbon and nitrogen balances obtained could account for
only 44% of the reacted carbon and only 29% of the initial nitrogen was
accounted for as nitrogen dioxide, nitric acid and peroxyacetyl nitrate.
An unidentified broad absorption at 840 cm" seen in Figure 8 with
characteristic absorption of an epoxide could if added to the carbon
balance raise it to 47%. The unaccounted carbon in the system has either
formed an organic aerosol which cannot be monitored by the long path infrared
system and/or is removed from the gas phase by dry deposition onto the
reactor wall.
The hydrocarbon p-cymene (p-isopropyl toluene) was one of the compounds
photooxidized in bag irradiations but not in the long path infrared chamber.
Thus no infrared product analysis is available. The p-cymene/NO^ system
would be expected to give similar type products and a. poor carbon balance
like the toluene/NO or other aromatic hydrocarbon/NO system. Studies of
X X
the toluene/NO- irradiated system in the long path infrared chamber resulted
in the identification of gaseous products that accounted for less than 50
percent of the reacted carbon. When 10 ppm toluene and 1 ppm NO- were
irradiated with all lamps at 158 minutes, 1.2 ppm of the initial toluene
and all of the NO- had reacted. The products observed were 0.25 ppm
cresols, 0.24 ppm benzaldehyde, 0.03 ppm peroxybenzoyl nitrate, 0.23 ppm
carbon monoxide, and 0.17 ppm formaldehyde. The infrared spectra did not
indicate the formation of phenol, tolualdehyde, phthalaldehydes or benzyl
alcohol. The infrared absorption of cresols have strong bands at 1162 cm" a
unique absorption band for peroxyacetyl nitrate and absorption at 1105 cm
with structure similar to that of formic acid. Considering all absorption
bands of the cresols and peroxybenzoyl nitrate very little if any peroxy-
acetyl nitrate is formed. This result is unlike reported results of Spicer
(62) who observed 0.3 ppm peroxyacetyl nitrate and obtains a 96% carbon
balance for the toluene/NO system in which about 1 ppm toluene reacted
X
in four hours. Other products observed by Spicer were benzaldehyde, cresols,
phthalaldehyde, tolualdehyde, benzyl alcohol and formaldehyde.
32
-------
Six monoterpenes:a-pinene, 3-pinene, A^-carene, myrcene, d-limonene,
terpinolene, and isoprene were irradiated in the long path infrared chamber.
Each hydrocarbon at an initial concentration of 8 ppm compound and nitrogen
dioxide concentration of 1.2 - 1.5 ppm were irradiated at a k , , = 0.51
_1 d(NO )
min . The series of spectra shown in Figure 10 to 16 are for these com-
pounds. The lower spectra in these figures noted as A are of the system
after 60 minutes of irradiation and contains all the spectra information
pertinent to gas phase species after this period of irradiation. The
products observed in these systems at 60 minutes irradiation time are
listed in Table 4. The spectra shown in Figures 10 to 16 also contain
spectra of each system at 30 minutes (spectrum B) , 10 minutes (spectrum C),
and sample at time equal zero or before irradiation (spectrum D). In all
these figures note the decrease in nitrogen dioxide concentration as observed
by the decrease in absorbance of its infrared band at 1600 cm as irradiation
time increases from upper spectrum D, (t=0 minutes) to lower spectrum A,
(t=60). This spectral evidence reinforces the earlier statements made about
nitrogen dioxide chemiluminescence measurements remaining erroneously higher
after the NO- maximum. Except for beta-pinene where only 0.01 ppm peroxy-
acetylnitrate was produced after 60 minutes of irradiation the infrared
spectrum show vividly its formation. In contrast maximum PAN had formed
after just 10 minutes of irradiation of the myrcene-NO mixture shown in
-1 X
Figure 13 (absorbance at 1162 cm ). The carbon balances in general
were poor for all systems. The products that were obtained do offer some
insight into the fragmentations or lack of fragmentation that takes place
during photooxidation. Only about 9% of the carbon can be accounted for as
gaseous products in the alpha-pinene system (Table 3) . The reduction in the
over all infrared absorption in the C-H region of the spectra (2850-3050 cm )
with time with minimal increase in absorption elsewhere implies that products
are being removed from the gas phase. Processes including aerosol formation
with subsequent wall deposition of aerosol material are probably occurring
in the chamber.
In the ozonolysis of alpha-pinene Spencer (31) observed a white mist
wMrh when analyzed consisted of five oxygen atoms combined with each molecule
of alpha-pinene which resulted in a C^H^Og compound. By classical qual-
itative techniques it was postulated that the compound had one of two structures.
33
-------
/800
2OOO<
2 ZOO
2.500
2700
if 00
3100
Figure 10. Spectra of alpha - pinene irradiation. Irradiation times : A = 60 min.,
B = 30 min., C = 10 min., D = 0 min. 357 meters path length, initial concentration
alpha - pinene = 8 ppm, NO2 = 1.2 ppm.
-------
Oi
7OO
900
1100
1300
1
D
C
I
J
i
t
,
k
r
! j
-.^
'
... J
.-A
I
— i-
i
—
—
•M*
'-
i —
•-
_..
""•
-
:_
i
i-
i
1
— i —
i
-.i...
.L
i
i:
i
— r- -
-f-
-F
!
I
-4—
i
._[_
1
i
"1 —
.j..
J
•—
•*••
•
~
—
-?•'
S
_.
J
.!>
(/
^
.
--
~
,
'/
/
/
i
;~
—
-
:
j.
i
1
F
....
—
•-
_
.
...
...
v
\-
v
t
-
2O0Ocm->
a zoo
ewo
1500
J600
1700
/foo
2500
1700
Figure 11. Spectra of beta - pinene irradiation. Irradiation times : A = 60 mm.,
B = 30 min., C = 10 min., D = 0 min. 357 meters path length, initial concentration
beta - pinene = 8 ppm, N(>2 =1.5 ppm.
-------
CO
700cm-
?OO
1100
1300
2O0Ocm->
zzoo
1500
J600
1700
2500
1700
Figure 12. Spectra of A - carene irradiation. Irradiation times : A = 60 min.,
B = 30 min., C -10 min., D = 0 min. 357 meters path length, initial concentration
A^ - carene = 8 ppm, N02 =1.2 ppm.
-------
/8OO
Figure 13. Spectra of myrcene irradiation. Irradiation times : A = 60 min.,
B = 30 min., C = 10 min., D = 0 min. 357 meters path length, initial concentration
myrcene = 8 ppm, NO2 = 1.3 ppm.
-------
00
2 OOO cm-'
Figure 14. Spectra of limonene irradiation. Irradiation times : A = 60 min.,
B = 30 min., C = 10 min., D = 0 min. 357 meters path length, initial concentration
limonene = 8 ppm, NO2 =1.2 ppm.
-------
10
2 OOO cm-'
2 £.00
zveo
Figure 15. Spectra of terpinolene irradiation. Irradiation times : A = 60min.,
B = 30 min., C = 10 min., D = 0 min. 357 meters path length, initial concentration
terpinolene = 8 ppm, ISIC>2 = 1.3 ppm.
-------
700 *•»-
?OO
1100
1300
1300
1600
1700
I
/800
2 OOO cm-'
z-voo
I
ZSOO
1700
Figure 16. Spectra of isoprene irradiation. Irradiation times : A = 60min.,
B = 30 min., C = 10 min., D = 0 min. 357 meters path length, initial concentration
isoprene = 8 ppm, NO2 = 1-3 ppm.
-------
TABLE 4. REACTIVITY AND PRODUCTS BY LONG PATH INFRARED AT 60 MINUTES
Compound C/NOx %HC H-CO HCOOH CO
ratio reacted
Isoprene 31 68 2.39 0.44 0.77
K10 100 6.8 3.4 4
Myrcene 62 73 0.84 0.16 0.26
d-Limonene 67 91 0.9 0.35 0.46
Terpinolene 62 95 1.2 0.31 0.49
\ f V
K>
a-Pinene 67 72 0.33 0.32 0.39
g-Pinene 53 58 0.94 0.39 0.23
A3-Carene 67 60 0.46 0.39 0.50
*
C02 CH3CHO PAN (CH3)2CO Total
ppmC
products
0.2 2.0 0.32 0 8.44
2.0 1.95 1.2 0 32.8
0.24 2.35 0.70 0.76 9.88
0.49 0.23 0.24 0 3.14
0.69 0.35 0.28 0 3.95
0.6 1.57 0.20 0 5.18
0.4 0.05 0.018 0 2.58
0.9 0.33 0.26 0 3.43
% Carbon
accounted
for
31
44
17
4
5
9
6
7
Initial hydrocarbon cone. = 8 ppm
Products as ppm compound (v/v)
k,(NO_) =0.51 min~
d i
Initial concentrations: 15 ppm isoprene, 7.5 ppm N0_; irradiated
for 45 minutes; additional products detected; 1.0 ppm methacrolein
1 ppm methyl vinyl ketone, 0.5 ppm HNO , 1.1 ppm 0 , 0.5 ppm
(remaining)
NO
-------
Verbenone
Oxozonide
Verbenone peroxide
Ozonide
Other researchers have also collected aerosols formed from the alpha-pinene/
NO irradiated system and obtained infrared absorption spectra (21, 23, 24, 28)
x
The prominent absorption obtained of the aerosols is characterized by the
carbonyl (-C=0) stretching frequencies. Wilson, et^ aJL. (30) obtained similar
infrared absorption spectrum for the dichloromethane extracted fraction of
smog chamber generated alpha-pinene aerosols and aerosols collected in the
pine forested Blue Ridge Mountains of North Carolina. The same researchers
also performed extensive extraction, derivatization, and gas chromatographic-
mass spectral analysis in attempts to identify components comprising the
alpha-pinene/NO aerosol. Hampered by a lack of reference standards these
2C
investigators were able to confirm only one compound in the acid extraction:
pinonic acid. A number of oxygenated species were postulated based on mass
spectral analysis of the derivatives of acid, base, and neutral fractions
shown below.
Postulated Oxygenated Species of Alpha-Pinene (26)
COOH
TENTATIVE
a-PINENE
CHO
42
-------
Somewhat surprising is the fact that nitrogen containing aerosol compounds
were not detected. Nitrogen containing aerosols would be expected in the
alpha-pinene/NOx irradiated system because of the poor nitrogen as well
as carbon balance.
Pinonic acid was identified in collected aerosol sample of the Pisgah
National Forest of western North Carolina but was absent in urban samples
(30) . Due to the semiquantitative nature of the study of the alpha-pinene to
pinonic acid aerosol process it is not possible to estimate what portion of
the forest aerosol samples are due to alpha-pinene oxidation. However, as
noted prevously infrared spectra of the dichloromethane extractable fraction
of forest and chamber samples did compare favorably.
Two monocyclic diolefin terpenes d-limonene and terpinolene gave similar
gaseous products as shown in Table 4. These observed gaseous products re-
present less than 10% of the reacted carbon. Unidentified weak absorption
bands were observed at 855, 1630, and 1650 cm in the d-limonene irradiation
and 853, 1290, 1375, and 1630 cm" in the terpinolene irradiation. In neither
case assuming the most favorable absorption coefficients would the unidentified
absorption account for more than a few percent of the reacted carbon. The
formation of aerosol is the probable cause of the carbon balance deficiency
(25).
3
Two bicyclic mono-olefins A -carene and beta-pinene like the other com-
pounds also have poor carbon balances. Gaseous product identification is
shown in Table 4. Less than 10% of the reacted carbon can be accounted for
as gaseous products. Again aerosol formation probably accounts for much
of the unaccounted for carbon. The infrared analysis of aerosol generated
from beta-pinene and alpha-pinene were carried out by Stephens and Price
(28). The only prominent characteristic absorption was the carbonyl region
around 1700 cm"1 for both terpene aerosols. The only previously reported
O
study of the photochemistry of A -carene was the work of Grimsrud et_ al.
(16) but products were not determined.
Myrcene was different than any of the other CIQ monoterpenes studied
since it is acyclic and possessed the greatest number of olefinic bonds:
three. The amount of myrcene reacted in 60 minutes and analysis of gaseous
43
-------
products are given in Table 4. The high yield of acetaldehyde, peroxyacetyl
nitrate and acetone along with other observed products account for 17% of the
reacted carbon. While this is still a relatively poor carbon balance, it
is still significantly better than the cyclic terpene systems previously
described. The open chain structure of myrcene compared to the cyclic ter-
penes apparently fragments more easily into volatile low molecular weight
oxygenates. The monocyclic diolefins and the bicyclic mono-olefins when
reacted with hydroxyl radicals or with ozone across the double bond form
monofunctional oxygenated cyclic compounds or difunctional ring opened
oxygenated compounds. These compounds are expected to be low in volatility
and condense out as aerosols.
In addition to the long path IR studies a sensitive high resolution
gas chromatographic study was performed to help identify terpene oxonolysis
products. Using the cryogenic trapping procedures and the 3 column FID
system described by Lonneman (8) analyses were made on 5 ppm (v/v) terpene
air samples which were reacted with excess ozone. The chromatographic pro-
cedure will in general separate C -C n hydrocarbons and some of their
oxygenates. Unfortunately, very little of the products could be detected.
Small amounts (less than 1%) of acetaldehyde could be seen. It is probable
that the products condensed to form aerosols or being extremely polar were
unable to pass through the chromatographic columns for detection.
The phenomena of aerosols formation should be considered when attempting
to apply laboratory smog chamber results to real atmospheric conditions. In
• jf;~f~ "'"•"
these reported smog chamber experiments the initial..hydrocarbon concentrations
were 10 to 1000 times higher than any reported natural hydrocarbon concen-
tration in the ambient atmosphere. Since many of the olefins react extremely
fast with ozone an accumulation of ozone is suppressed at high carbon/NO
X
ratios. At these higher hydrocarbon concentrations the concentrations of
oxygenates formed are also higher which result in aerosol formation. Recent
studies have shown there is a threshold hydrocarbon concentration below
which aerosol formation does not occur (23,63,64). Since aerosol formation
represents a chain termininating step in the photochemical mechanism its
effect is the removal of partially oxidized hydrocarbons from the system
before the maximum ozone generating potential is realized. Caution should
44
-------
be excercised in extrapolating the ozone producing ability of a compound at
high concentrations to ambient concentrations where aerosols may not be
formed. At low ambient natural hydrocarbon concentration the oxygenated
hydrocarbons could remain in the gas phase and continue to react in the
photochemical system. However, results of experiments at 100 ppbC (a value
which approaches an upper limit of ambient terpene concentrations) and 10
ppb N0x show propylene to be significantly more efficient than a-pinene
in producing ozone (see Table 3).
EFFECT OF VEGETATION ON AIR QUALITY
A number of the more common trees in the continental United States
have been surveyed with respect to hydrocarbon emissions (3). It is generally
found that the deciduous vegetation (hardwoods) are primarily isoprene
emitters and conifers (softwoods) are basically monoterpene emitters. There
are also several species of vegetation that emit both isoprene and mono-
terpenes. Some emission controlling environmental variables in plants have
been found although the biochemical function of both isoprene and the
monoterpenes is poorly understood. The monoterpenes at or close to the
air plant interface accumulate in the resin ducts of the foliage (65).
More difficult to quantitate are the significance of resin blisters which
appear on the woody parts of the tree as a result of wounds, disease, or
insect attack. Studies by Rasmussen (30), Dement e£.al. (66), and Tingey
et al. (67) show the rate of terpene volatilizing from foliage in
environmental chambers depended upon temperature, vapor pressure of the
emitted compound, and in some cases humidity. Tingey et al. (67) found
foliage temperature rather than ambient air temperature to be a more
accurate predictor of terpene emissions. In-situ measurements of a-pinene
at a pine forest reported by Arnts et al. (38) did not show a strong
correlation of hydrocarbon flux with ambient air temperature. The a-pinene
flux in this study was found to correlate directly with temperature and
inversely with windspeed. This suggests a dependence on foliage surface
temperature rather than air temperature.
In contrast with the monoterpenes, isoprene emission are strongly
45
-------
dependent on sunlight. Rasmussen and co-workers (34,35,68,69) have shown
that isoprene emission mostly by deciduous plants occurs predominately in
sunlight and is therefore probably directly linked to photosynthesis.
Unlike the monoterpenes which can continue to volatilize from the resin
ducts in darkness, the much more volatile isoprene does not appear to
be stored by the plant for release at night. Recent studies by Tingey et.
al^ (70) confirm the sunlight dependence of isoprene emissions. Live oaks
were found to saturate with moderate light intensities, the study also
revealed that temperature became a controlling variable at a given light
intensity. The study concluded that after reaching moderate light in-
tensities temperature becomes the rate controlling variable.
Using emission characteristics of vegetation, chemical nature of the
biogenic hydrocarbons, and a consideration of boundary layer micrometeorology
the air quality of forest airsheds can be visualized. Figure 17 illustrates
the influence of a coniferous forest on the day and nighttime air quality.
During the daytime terpenes are volatilized as the temperature increases.
Terpenes emitted into the atmosphere react with species such as 0_ and OH
to principally form aerosols and if sufficient NO is present ozone and
X
additional aerosol products are formed. At night terpenes are emitted at
a lower rate due to cooling biomass; nighttime inversions tend to trap
the terpenes within the boundary layer which may be as low as the tree
canopy crown. In the absence of sunlight to initiate photochemical reaction
the terpenes react with ozone forming principally aerosols. Evidence for
nighttime ozone destruction has been observed at a pine forest site where
ambient ozone and alpha-pinene concentrations have been monitored concurrently
both above and below the forest canopy (40). Although dry deposition cer-
tainly accounts for some ozone destruction, elevated alpha-pinene concen-
trations below the canopy accompanied by suppressed ozone concentrations
during predawn inversion conditions support this hypothesis. The highest
concentrations of monoterpenes reported in the ambient atmosphere were
measured in the forest canopy. The concentrations observed were on the
order of 40-50 ppbC measured in the early morning hours (3-6 AM) but
quickly declined to less than 10 ppbC as the inversion dispersed after
46
-------
DAYTIME
(GENERATION/03 DESTRUCTION
WIND
TERRENES
AEROSOL
.FORMATION
NIGHTTIME
+ TERRENES - » AEROSOL
ft>
Role of monoterpenes in atmospheric chemistry.
DAYTIME
ISOPRENE
WIND
r r r
NIGHTTIME
Figure 17 Role of isoprene in atmospheric chemistry.
47
-------
sunrise. The monoterpene emitting forests may actually destory more ozone
than it creates since terpenes are very efficient at destroying ozone and
create ozone inefficiently.
Unlike coniferous forest an isoprene emitting deciduous forest could
affect air quality in a quite different manner. Since isoprene release is
controlled primarily by sunlight and temperature, emissions occur during
the day. In the presence of NO isoprene participates in the photochemical
X
generation of ozone nearly as efficiently as propylene. As darkness approaches
photochemical processes stop as does isoprene emission. Thus the net effect
of deciduous forest emissions would be elevation of ozone levels if NO is
x
available.
Since there are large regions of the United States where the natural
vegetation is either primarily deciduous or coniferous, then regional
differences in air quality coinciding with these biotypes should be
observed if the emissions are significant. The Northwest and Rocky Mountains
area forests are generally coniferous i.e. ponderosa pine, lodgepole pine,
douglas fir, redwood (71). Large areas of the Southeast are predominately pine
and oak forest. The Northeast including Maryland, West Virginia, Ohio, Illinois,
Indiana, Kentucky, and the western portion of Virginia are primarily deciduous
forested areas. The western forests should contribute primarily monoterpenes,
the Northeastern forests isoprene and the southeast a combination of the two.
The midwest agricultural lands while generally grass and crops have been
found to be isoprene emitters, but their importance relative to forests is
believed to be small (72).
To assess the contribution of biogenic hydrocarbon emissions to
background ozone levels an accurate knowledge of background anthropogenic/
biogenic ambient hydrocarbons and N& concentrations or their emissions is
X
essential. Unfortunately neither is well documented.
Methods of estimating total biogenic hydrocarbon emissions began with
one measurement and a large degree of presumptive extrapolation. Recent
estimates use marginally refined measurements and equally commensurate
imaginative extrapolations. An examination of the literature reviewed
48
-------
by Robinson and Robbins in 1968 (7) reveals some tenuous assumptions which
can impart a great deal of uncertainty to the early emission estimates.
The first estimate by Went (73) using unpublished data of Haagen-Smit
involved extrapolating the emission rate of one sage brush plant to the
vegetated surface of the earth. This estimate assumed (1) emission of
this plant is representative of the same plant growing in the open environ-
ment, (2) a fixed rate of hydrocarbon release (5% volatiles released of
the photosynthates formed) dependent on photosynthesis, (3) hydrocarbon
release of other plants i.e. coniferous and to an arbitrarily lesser degree
deciduous forests, agricultural lands and steppes are a similar proportion
of volatiles. Even accepting the above assumptions extrapolations were
performed using order of magnitude estimates of the land area covered by
these vegetative types. Thus the first biogenic hydrocarbons estimate
was an extremely crude but nevertheless interesting speculation on the
subj ect.
Rasmussen and Went in 1965 (27) increased Went's previous estimate
by employing two additional methods and including isoprene as a biogenic
emission. They assumed a vertical column of 10 ppb organic volatiles
averaged up to 2 km over the world vegetative surfaces. The isoprene:
monoterpenes ratio making up the 10 ppb organic volatiles was not specified.
Consequently this could represent a minimum concentration of 50 ppbC
(assuming 100% isoprene) to a maximum 100 ppbC (assuming '100% monoterpene) .
The average 10 ppb organic volatiles is based on their ground level measure-
ments only and is not supported by vertical profile measurements. Sub-
sequent detailed hydrocarbon measurements by Lonneman (74), Rasmussen (75),
and Holdren (76) analyzing both ground level and aircraft ambient air samples
have never demonstrated maximum concentrations greater than I/10th of the 10 ppb
concentration; typically values are less than l/100th this value levels.
Arnts et^ aJL (38) observed only a few ppbC of a-pinene a few meters above
the canopy of a pine forest in North Carolina during the day. Holdren
reported similar concentrations of monoterpenes in forested areas of the
Northwest (76). Thus it appears that Rasmussen and Went's estimate is
b*="3 on erroneous assumptions. They also arrived at an estimate of
49
-------
biogenic emissions based on emissions of vegetation placed in a plastic
bag. The drawbacks of this technique and its derivatives will be dis-
cussed later.
Robinson and Robbins (7) in their review found that 66% of the
atmospheric non-methane hydrocarbon loading should be of biogenic origin.
The review has been widely cited in discussions of the efficacy of EPA
hydrocarbon control regulations with a large amount of significance
attached to the estimates. While ambituous in its scope the review did
not attempt to critically review the limitations of its sources nor,discuss
the uncertainties inherent in their estimates.
Since the review of Robinson and Robbins other efforts have been
made to quantitate biogenic hydrocarbon emissions. Rasmussen examined
emissions of foliage enclosed in leaf assimilation chambers over a range
of temperatures and illuminations (3) . Data from these experiments were
extrapolated to obtain estimates of global biogenic emissions. Recently
Zimmerman (72) refined the enclosure techniques of Rasmussen to study
more plant species. In these studies the dry weight of the plant parts are
used to reference emission e.g. gm hydrocarbon emitted/gm dry weight
foliage. A scaling factor (gm dry weight foliage/area of land surface
area) derived from data gathered in the International Biological Program
is used to develop regional emission factors and subsequent global emission
estimates. While there is evidence to believe that the enclosure method
tends to overestimate emissions (77), the larger problem probably lies in
the accurate extrapolation of the emission factors to large land areas of
mixed and discontinuous vegetation.
Research Triangle Institute proposed an upper limit for global biogenic
emissions based on estimates of the total carbon fixed by green plants on
the land surface of the world (1). They assumed that perhaps 10% of the
total fixed carbon could represent an 'upper limit1 of volatiles released
by green plants. This estimate, some 20 times greater than Rasmussen and
Went's estimate, represents assumptions based on order of magnitude estimates
of total fixed carbon and an arbitrarily chosen 10% upper limit of total
fixed carbon as volatiles.
50
-------
Emission estimates to date for biogenic and probably to a lesser
but significant degree for anthropogenic emissions are fraught with much
uncertainty. All estimates of reported emissions to date state that the
majority of hydrocarbons present in the well mixed continential air mass
is of biogenic origin and as monoterpenes and isoprene. This has simply
not been shown to be the case in any field study yet performed. Investi-
gators find 'clean' air masses to be predominately photochemically
olefin depleted dilute auto exhaust, i.e. contain C -C alkanes
and aromatics consisting primarily of benzene, toluene, and the
xylenes. Monoterpenes in rural areas rarely exceed 10 ppbC and are
frequently not detected at limits of detection less than 0.5 ppbC. Isoprene
has been observed as high as 30 ppbC in a forested area near Boston (10).
Washington State University (75) typically observed about 7 ppbC of isoprene
in a lightly forested agricultural area of Missouri. In all cases biogenic
hydrocarbons contributed no more than a maximum of 25% to the NMHC of a
clean air mass which had a NMHC concentration of 80 ppbC or greater.
Although arguments have been made that biogenic emissions react much
faster than anthropogenic emissions, analysis of air samples collected at
forests in the tree canopies where reactions would be considered negligible
and emissions should be greatest have not shown a predominance of biogenic
over anthropogenic hydrocarbons. Emission inventories therefore for both
biogenic and anthropogenic hydrocarbons could be inaccurate. It is necessary
to examine the detailed hydrocarbon analysis of a 'clean1 air mass to determine
anthropogenic and biogenic composition to relate their contribution to back-
ground ozone formation. The transport of hydrocarbons and other pollutants
from urban centers has been well documented (10-12). The concentrations of
urban tracer gases such as acetylene and CC1 F can be used in addition to
back trajectory analysis to establish the movement of the air mass. Accurate
NO concentrations in the air mass must also be determined if the photochemical
X
oxidant potential is to be estimated. If attention is not given to detailed
hydrocarbon, NO analysis and air mass movement details background clean air
status is incorrectly given to air masses which have been strongly influenced
by anthropogenic sources. Recently, Whitehead and Severs (78) reported
total non methane hydrocarbon analysis of ambient air samples collected
in a forested area 38 miles north of Houston, Texas, which averaged 8.7
51
-------
ppmC for 35 morning samples. The investigator concluded the TNMHC levels
were produced by vegetation. The data as reported did not show conclusively
the lack of influence by local anthropogenic sources. A detailed hydro-
carbon analysis if performed by the investigator would have shown the
anthropogenic and biogenic hydrocarbon make up of the air. Seila (79) has
taken ambient air samples from the same forested area, conducted detailed
hydrocarbon analyses and found TNMHC concentrations only l/20th those
measured by Whitehead and Severs. Of those air samples only 2% of TNMHC
could be attributed to vegetative sources. This shows the necessity of
conducting detailed hydrocarbon analysis of air samples to determine its
anthropogenic and biogenic composition. A certain gas chromatographic
expertise is needed in attempting detailed hydrocarbon air analysis.
In a paper by Whitby and Coffey (80) the authors propose that hydro-
carbons measured in the ambient air of an Adirondack mountain pine forest
are terpenes and their oxidation products. The analyses carried out by these
investigators are suspect due to the use of a low resolution 6' packed
column of OV-101. This column used to resolve terpenes will also elute
CQ-CI_ auto related aromatic compounds within the same retention times as
the terpenes. These Cg-C..- aromatics compounds are found in a photochemically
aged urban plume. Lonneman (personal communication) reports various lab-
oratories have observed artifact or ghost peaks associated with the use of
this particular column.
During a field study near Elkton, Missouri, (population 30) an area free
from nearby urban centers researchers from Washington State University (75)
conducted detailed hydrocarbon analysis on 80 ambient air samples. The
results of this study indicated an average TNMHC of about 98 ppbC with a
range of 41 to 150 ppbC. The influence of urban hydrocarbons was evident
in samples at the higher TNMHC range. Monoterpenes were not observed which
is consistent with the local deciduous vegetation. Isoprene was observed
quite regularly at about 6 ppbC and had a range of less than 0.4 to 28 ppbC
in all samples. Isoprene represented the single largest hydrocarbon species
52
-------
which on the average contributed only 6% to the measured TNMHC. In an
extreme situation isoprene constituted 24% of the 108 ppbC TNMHC measured
at 7:00 PM CDT with near zero surface winds.
Isoprene in ambient air samples has also been reported by Lonneman
et al. in the 1975 Northeast oxidant study (81). The Chickatabut Hill
sampling site 10 miles south of downtown Boston was surrounded by oak
forest. Detailed hydrocarbon analysis of ambient air samples showed con-
centrations of isoprene ranging from less than 0.5 ppbC during the night
to an average value of 10 ppbC during midday. Occasionally as high as
30 ppbC isoprene were observed. Air samples collected by aircraft over
the Boston area also contained isoprene which ranged from less than 0.5
ppbC to 3.5 ppbC (9).
Monoterpenes have been measured by Holdren, Westberg, and Zimmerman
(76) in coniferous forests of North Central Idaho using gas chromatographic-
mass spectrometric techniques. A total terpene concentration within the
forest canopy of between 0.5 to 16 ppbC was reported over 10 month sampling
period. Outside of the forest canopy, terpenes could not be detected with
a detection limit of less than 0.1 ppbC. Total non-methane hydrocarbon
measurements were not made in this study.
In a joint EPA-Duke University study to measure a-pinene fluxes of a
loblolly pine forest in North Carolina, over 300 ambient air samples were
analyzed (77). Analysis of air samples collected within a few meters
above the forest canopy crown showed a range of a-pinene concentration of
0.6 to 13 ppbC during midday. In other experiments at the same location
3 hour integrated samples were collected over 24 hour periods. Simultaneously
samples were collected 5 meters above the canopy the other at the base
of the canopy. The results of these studies are plotted in Figures 18 and 19.
The figures serve to illustrate two important factors. First the terpene
concentrations are substantially lower during the winter months probably
due to lower volatilization rates at lower temperatures. Secondly noctural
inversions serve to contain the terpenes as evidence by the concentration
differential between above and below the canopy. In Figure 19 looking at
the data of August 23 the 56 ppbC terpenes concentration represents a
substantial contribution to the TNMHC of 40% in the canopy. Measurements
53
-------
Ui
FEBRUARY?, 1971
MARCH 17,1977
0000
0300
0600
0900
1500
1800
2100
1200
TIME, hr
Figure 18. Total concentration of the C-|g terpene hydrocarbons (ppb carbon) observed above the canopy at the IBP
site at different times of the year.
2400
-------
60
o
ca
CC
«t
u
50 —
40
co
cc
o
cc
o
"I 30
Ln
Ui
UJ
a.
cc
<
10
AUGUST 23,1976
SEPTEMBERS, 1976
FEBRUARY?, 1977
MARCH 17,1977
0000
0300
0600
0900
1200
TIME, In
1500
1800
2100
2400
Fiqurc 19. Totjl coiicontration o( tho CIQ terpene hydrocarbons (pph carbon) observed in the canopy at the IBP site
al different limes of the year.
-------
5 meters above the canopy Figure 18 (August 23) resulted in terpenes concen-
trations accounting for only 12% of the TNMHC. During the early morning
periods which are most conducive to biogenic hydrocarbon buildup in the
surface boundary layer, the dissipation of the layer promotes mixing and
the concentrations decrease.
Measured concentrations of monoterpenes in coniferous forests under
typical summer sunlight and surface wind conditions generally range from
0.5 to 10 ppbC. Isoprene measurements under similar conditions have shown
similar concentration ranges.
To know the effect biogenic hydrocarbons have on oxidant formation, a
knowledge of background NO (N0_ + NO) is essential. Unfortunately back-
ground NO is perhaps even more poorly understood than background hydro-
x
carbons. The lack of NO information stems mostly from inadequate
X
analytical instrumentation in addition to true geophysical fluctuations.
Until recently background NO was considered to be on the order of 0.5
X
to 2 ppb, current measurements place clean continential air between 0.015
to 0.1 ppb NO (82-83).
X
On the basis of the preceeding discussion some estimates of background
NMHC/NO ratios can be made. If background non-methane hydrocarbon is
X
accepted to be in the range of 40 to 100 ppbC and NO in the range of
X
0.015 to 0.1 ppb then clean continental air should have a C/NO ratio
X.
range of about 400 to 7000.
Before examining the concentrations at the above ratios with respect
to the oxidant forming potential of such a system, a brief discussion of
the use of the term 'background' is in order. Anthropogenic and biogenic
emissions estimates lie close enough together such that their range of
uncertainty precludes answering the question: On a global basis are biogenic
emissions larger than anthropogenic emissions? Hence we are forced to
rely on our knowledge of atmospheric chemistry and ambient measurements in
'clean' air masses. A truly clean atmosphere in the strictest sense
described as air unaffected by man's activities no longer exists. Since
the advent of the Industrial Revolution, large scale burning of fossil
fuels and the manufacture of synthetic chemicals has introduced pollutants
56
-------
into the atmosphere. The cleanest air today is recognized by the lowest
measured levels of distinctly anthropogenic origin and preferably tro-
pospherically inert compounds i.e. acetylene, CC13F. The background values
for TNMHC and NO cited earlier reflect the lowest levels measured
j£
in a well mixed air mass not necessarily free from man's influence.
The values should be viewed as contemporary background levels and not
necessarily clean air levels.
Since the irradiations of the hydrocarbons in this study were performed
at higher than ambient concentrations, interpolation of results to
ambient levels must be performed with caution. To estimate the effects
background TNMHC and NO have on background ozone the inefficiency of
X
the terpenes and isoprene to generate ozone relative to propylene will be
coupled with the kinetic model of M.C. Dodge.
The model using a 75% n-butane and 25% propylene mix has been
validated with smog chamber data at NMHC ranging from 0.2 to 5 ppmC and
0.05 to 0.6 ppm NO . In the real atmosphere poorly understood ozone removal
X
mechanisms such as dry deposition at the surface boundary layer and des-
truction by suspended particulates become increasingly important. The
photochemical model, although unvalidated at background pollutant concen-
tration levels, serves to place an upper limit on the ozone potential of
such an air mass.
To simulate an extreme situation in which terpenes and isoprene con-
tributed to 25% of the background NMHC, the system was photochemically
modeled at an initial condition of 75% n-butane and 25% propylene. Figure
20 shows the ozone isopleths generated for this system using initial NMHC
concentration ranging from 0 to 200 ppbC and N0x concentrations ranging
from 0 to 0.14 ppb. Using a background NMHC concentration range of 40 to
100 ppbC and NO concentration of 0.015 to 0.10 ppb the model predicts
*^ x
a maximum ozone concentration of from 0.23-0.26 ppb to 1.3-1.9 ppb. Since
the experimental evidence presented earlier in Figure 4 shows propylene to
be rour times more efficient than isoprene and nearly ten times more
efficient than a-pinene at generating ozone from the same quantity of NO^
then the photochemical model estimates which uses propylene are even more
57
-------
0.14
0.12
0.10H \ KW!
0.08
X
o
0.06
0.04
0.02
0 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 0.18 0.20
TNMHC, ppm (C)
Figure 20. Ozone isopleths predicted from Dodge's photochemical model (75 percent
n-butane/25 percent propylene).
58
-------
overpredictive for these compounds.
In summary the experiments that have been conducted and reported in
this report do not support the hypothesis that terpenes and isoprene
contribute significantly to the 30-40 ppb geophysical ozone concentration.
The biogenic hydrocarbons and oxides of nitrogen are simply not present
in high enough concentrations and at the proper proportions to generate
much photochemical ozone.
59
-------
REFERENCES
1. Research Triangle Institute, Natural Emissions of Gaseous Organic Com-
pounds and Oxides of Nitrogen in Ohio and Surrounding States. Final
Report Contract Number 68-02-1096, pp. 1-15, 1974.
2. Rasmussen, K., M. Taheri, and R. Kabel. Sources and Natural Removal
Processes for Atmospheric Pollutants. Center for Air Environmental
Studies, Univ. Park, Pennsylvania, 1974.
3. Rasmussen, R.A. What Do the Hydrocarbons from Trees Contribute to Air
Pollution? J. Air Pollution Control Assoc. 22_, pp. 537-543, 1972.
4. Craker, L.E. Ethylene Production from Ozone Injured Plants. Environ-
mental Pollution i, pp. 299-304, 1971.
5. Tingey, D.T., C. Standley and R. W. Field. Stress Ethylene Evolution:
A Measurement Ozone Effects of Plants. Atmospheric Environment 10, pp.
969-975, 1976.
6. Abeles, A.L. and F.B. Abeles. Biochemical Pathway of Stress-Induced
ethylene. Plant Physiology 50, pp. 496-498, 1972.
7. Robinson, E., and R. C. Robbins. .Sources, Abundance, and Fate of Gaseous
Atmospheric Pollutants. Report SRI project PR-6755, Stanford Res. ,...
Inst., pp. 1-122, 1968.
8. Bufalini, J.J. and W.A. Lonneman. Environmental Science and Technology
(submitted for publication).
9. Bufalini, J.J., W.A. Lonneman. Ozone Formation from Rural Hydrocarbons.
Presented at Coordinated Research Council, New Orleans, May 1978.
10. Lonneman, W.A. Ozone and Hydrocarbon Measurements in Recent Oxidant
Transport Studies. International Conference on Photochemical Oxidant
Pollution and Its Control, EPA/3-77-001a, U.S. EPA, Research Triangle
Park, North Carolina, pp. 211-223, 1977.
11. Wolff, G.T., P.J. Lioy, G.D. Wright, R.E. Meyers, and R.T. Cederwall.
An Investigation of Long-Range Transport of Ozone Across the Midwestern
and Eastern United States. Atmospheric Environment 11, pp. 797-802, 1977.
12. Cleveland, W.S., B. Kleiner, J.E. McRae, J.L. Warner, and R.E. Pasceri.
Geographical Properties of Ozone Concentrations in the Northeastern United
States. Journal of the Air Pollution Control Association 27, pp. 325-328,
1977.
60
-------
13. Schuck, E.A. and G.J. Doyle. Photooxidation of Hydrocarbons in
Mixtures Containing Oxides of Nitrogen and Sulfur Oxides. Report
Number 29, Air Pollution Foundation, San Marino, California, 1959.
14. Glasson, W.A. and C.S. Tuesday. Hydrocarbon Reactivities in the
Atmospheric Photooxidation of Nitric Oxide. Environmental Science and
Technology 4_, pp. 916-924, 1970.
15. Stephens, E.R. and W.E. Scott, Relative Reactivity of Various Hydrocarbons
in Polluted Atmospheres. Proceedings of the American Petroleum Institute
42_, pp. 665-670, 1962.
16. Grimsrud, E.P., H.H. Westberg, R.A. Rasmussen. Atmospheric Reactivity
of Monoterpene Hydrocarbons, NO Photooxidation Tnd Ozonolysis. Inter.
J. Chem. Kinetics, Sym. 1, pp. $83-195, 1975.
17. Darnall, K.R., A.C. Lloyd, A.M. Winer, and J. N. Pitts, Jr. Reactivity
Scale for Atmospheric Hydrocarbons Based on Reaction with Hydroxyl
Radical. Environmental Science and Technology ]_» PP- 692-696, 1976.
18. Pitts, J.N., A.M. Winer, K.R. Darnall, A.C. Lloyd, and G.J. Doyle.
Hydrocarbon Reactivity and the Role of Hydrocarbons, Oxides of Nitrogen,
and Aged Smog in the Production of Photochemical Oxidants. Proceedings
of the International Conference on Photochemical Oxidant Pollution and Its
Control, EPA/3-77-OOla, U.S. E.P.A., Research Triangle Park, NC, pp. 687-
703, 1977.
19. Bufalini, J.J., T.A. Walter, and M.M. Bufalini. Ozone Formation Potential
of Organic Compounds. Environmental Science and Technology 10, pp. 908-
912, 1976.
20. Went, F.W. Blue Hazes in the Atmosphere. Nature 187, pp.641-643, 1960.
21. Groblicki, P.J. and G.J. Nebel. The Photochemical Formation of Aerosols
in Urban Atmospheres in Chemical Reactions in Urban Atmospheres, C.S.
Tuesday, ed., Elsevier, N.Y. pp. 241-267, 1971.
22. Lillian, D. Formation and Destruction of Ozone in a Simulated Natural
System (Nitrogen Dioxide + Alpha-Pinene + hv). In Photochemical Smog
and Ozone Reactions, R.F. Gould, ed., ACS Advances in Chemistry Series
No. 113, pp. 211-218, 1972.
23. Lipeles, M., D. A. Landis, and G.M. Hidy. The Formation of Organic
Aerosols in a Fast Flow Reactor. Advances in Environmental Science.
Volume £. John Wiley, 1978.
24 Ripperton, L.A., H.E. Jeffries, and 0. White. Formation of Aerosols by
Reaction of Ozone with Selected Hydrocarbons. In Photochemical Smog and
Ozone Reactions, R.F. Gould, ed., ACS Advances in Chemistry Series Number
113, pp. 219-231, 1971.
61
-------
25. Schuetzle, D. and R.A. Rasmussen. The Molecular Composition of Secondary
Aerosol Particles Formed from Terrenes. Journal of the Air Pollution Control
Association 28, pp. 236-240, 1978.
26. Schwartz, W. Chemical Characterization of Model Aerosols. U.S. E.P.A.,
EPA-650/3-74-011, 1974.
27. Rasmussen, R.A., F.W. Went. Volatile Organic Material of Plant Origin in
the Atmosphere. Proc. Nat. Acad. Sci. 53_, pp. 215-220, 1965.
28. Stephens, E.R. and M.S. Price, Smog Aerosol: Infrared Spectra. Science
168, pp. 1584-1586, 1970.
29. Went, F.W. On the Nature of Aitken Condensation Nuclei. Tellus XVIII,
pp. 549-555, 1966.
30. Wilson, W.E., W.E. Schwartz, and G.W. Kinzer. Haze Formation - Its
Nature and Origin. Report to EPA CPA 70-Neg 172 and CRC CAPA 6-68,
1972.
31. Spencer, C.C., W.I. Weaver, E.A. Oberright, H.J. Sykes, A.L. Barney,
and A.L. Elder. Ozonization of Organic Compounds. Journal of Organic
Chemistry _5, pp. 610-617, 1940.
32. Rasmussen, R.A. , and M. W. Holdren. Analyses of C- to C1f) Hydrocarbons
in Rural Atmospheres. Paper presented at the 65th meeting of the Air
Pollution Control Association, June 1972.
33. Rasmussen, R.A., and M.W. Holdren. Analyses of C_ to C,~. Hydrocarbons
in Rural Atmospheres. Paper presented at the 65th meeting of the Air
Pollution Control Association, June 1972.
34. Rasmussen, R.A. Isoprene Plant Species List. Special Report, Oregon
Graduate Center, March 1978.
35. Rasmussen, R.A. Isoprene: Identified as a Forest-Type Emission to the
Atmosphere. Environ. Sci. and Technol. A, PP- 667-670, 1970.
36. Tyson, B.J., W. A. Dement, and H.A. Mooney. Volatilization of Terpenes
from Salvia Mellifera. Science 252, pp. 119-120, 1974.
37. Mirov, N.T. Composition of Gum Turpentines of Pines. Forest Service
Technical Bulletin, No. 1239, U.S. Dept. of Agriculture, pp. 1-158, 1961.
38. Arnts, R.R., R.L. Seila, R.L. Kuntz, F.L. Mowry, K.R. Knoerr, A.C. Dudgeon,
Measurement of Alpha-Pinene Fluxes from Loblolly Pine Forest. Conference
Proceedings, 4th Joint Conference on Sensing of Environmental Pollutants,
New Orleans, November 6-11, 1977.
62
-------
38. Coffey, P.E. and H. Westberg. International Conference on Oxidants, 1976-
Analysis of Evidence and Viewpoints: Part IV. The Issue of Natural
Organic Emissions. U.S. E.P.A., EPA-600/3-77-116, pp. 1-50, 1977.
40. Arnts, R.R., R.L. Seila, Lonneman, W.A., and R. L. Kuntz, unpublished
results.
4i. Teranishi, R., T.H. Schultz, W.H. McFadden, R.E. Lundin, and D.R. Black.
Volatiles from Oranges. I. Hydrocarbons. Identified by Infrared, Nuclear
Magnetic Resonance, and Mass Spectra. Journal of Food Science 28, pp.
541-545, 1963. —
42. Finder, A.R., The Chemistry of the Terpenes. John Wiley and Sons., Inc.,
N.Y., 1960.
43. Nicholas, H.J. The Biogenesis of Terpenes in Plants. In the Biogenesis of
Natural Compounds, P. Bernfeld, ed. A Pergamon Press Book, N.Y., pp. 641-
691, 1963.
44. Tuesday, C.S. The Atmospheric Photooxidation of Nitric Oxide and Trans-
Butene-2. In: Chemical Reactions in the Lower and Upper Atmosphere,
Interscience Press, New York, pp. 15-49, 1961.
45. Altshuller, A.P., D.L. Miller, and S.F. Sleva. Determination of Formalde-
hyde in Gas Mixtures by the Chromotropic Acid Method. Anal. Chem. 33,
pp. 621-525, 1961.
46. Saltzman, B.E. Colorimetric Microdetermination of Nitrogen Dioxide in
the Atmosphere. Anal. Chem. 26, pp. 1949-1955, 1954.
47. Hanst, P.L. Spectroscopic Methods for Air Pollution Measurement. In
Advances in Environmental Science and Technology, Pitts, and Metcalf,
Eds., John Wiley and Sons. New York, pp. 91-213, 1971.
48. Hanst, P.L., A.S. Lefohn, and B.W. Gay. Detection of Atmospheric
Pollutants at Parts-per-Billian Levels by Infrared Spectroscopy. Appl.
Spectros. 22_, pp. 188-198, 1973.
49. Winer, A.M., J. W. Peters, J.P. Smith, and J.N. Pitts, Jr. Response of
Commercial Chemiluminescent NO-NO Analyzers to Other Nitrogen-Containing
Compounds. Environmental Science and Technology 8^ pp. 1118-1121, 1974.
50. Spicer, C.W. Photochemical Atmospheric Pollutants derived from Nitrogen
Oxides. Atmospheric Environment 11, pp. 1089-1095, 1977.
5!. Dodge, M.C., Combined Use of Modeling Techniques and Smog Chamber Data
to Derive Ozone-Precursor Relationships. International Conference on
Photochemical Oxidant Pollution and Its Control, Proceedings 2_ pp. 881-
889. EPA-600/3-77-001b. Environmental Protection Agency, RTP, NC.,
1977.
63
-------
52. Winer, A.M., A.C. Lloyd, K.R. Darnall, J.N. Pitts, Jr. Relative Rate
Constants for the Reaction of Hydroxyl Radical with Selected Ketones,
Chloroethenes, and Monoterpene Hydrocarbons. J. of Phys. Chem. 80,
pp. 1635-1639, 1976.
53. Gaffney, J.S., R. Atkinson, and J.N. Pitts, Jr. Relative Rate Constants
for the Reaction of 0( P) Atoms with Selected Olefins, Monoterpenes,
and Unsaturated Aldehydes. Journal of the American Chemical Society
97_, pp. 5049-5051, 1975.
54. Japar, S.M., Ch.H. Wu and H. Niki. Rate Constants for the Gas Phase
Reaction of Ozone with Alpha-Pinene and Terpinolene. Environmental
Letters ]_ (3), pp. 245-249, 1974.
55. Japar, S.M., C.H. Wu, and H. Niki. Rate Constants for the Reaction of
Ozone with Olefins in the Gas Phase. Journal of Physical Chemistry 78,
pp. 2318-2320, 1974.
56. Pate, C.T., R. Atkinson, and J.N. Pitts, Jr. The Gas Phase Reaction of
0- with a Series of Aromatic Compounds (preprint), 1978.
57. Altshuller, A.P., and J.J. Bufalini. Photochemical Aspects of Air
Pollution: A Review. Photochem. Photobiol. _4, pp. 97-147, 1965.
58. Altshuller, A.P. and J.J. Bufalini. Photochemical Aspects of Air
Pollution: A Review. Environmental Science and Technology 5_, pp.
39-64, 1971.
59. Wu, C.H., S.M. Japar, and H. Niki. Relative Reactivities of HO-Hydro-
carbon Reactions from Smog Reactor Studies. J. Environmental Science
Health-Environmental Science and Engineering All(2), pp. 191-200, 1976.
60. Altshuller, A.P., S.L. Kopczynski, W.A. Lonneman, T.L. Becker, and R.
Slater. Chemical Aspects of the Photooxidation of the Propylene-Nitrogen
Oxide System. Environmental Science and Technology 1, pp. 899-914,
1967. ~
61. Calvert, J.G. and J.N. Pitts, Jr. Photochemistry, John Wiley & Sons,
New York, 1966.
62. Spicer, C.W. and P.W. Jones. The Fate of Aromatic Hydrocarbons in
Photochemical Smog Systems: Toluene. Journal of the Air Pollution
Control Association £7_, pp. 1122-1125, 1977.
63. National Research Council. Ozone and Other Photochemical Oxidants, Vol.
1, National Academy of Sciences, Washington, DC, 1976.
64. O'Brien, R.J., J.R. Holmes, and A.H. Bockian. Formation of Photochemical
Aerosol from Hydrocarbons: Chemical Reactivity and Products. Environmental
Science and Technology 9_, pp. 568-576, 1975.
64
-------
65. Hanover, J.W. Factors affecting the release of volatile chemicals by
forest trees. Mitteilungen Der Forstlichen Bundes-Versuchsanstalt
Wien 97_, pp. 624-641, 1972.
65. Dement, W.A., B.J. Tyson, and H.A. Mooney. Mechanisms of Mono-
terpene Volatilization in Salvia Mellifera. Phytochemistry 14, pp.
2555-2557, 1975.
67. Tingey, D.T., M. Manning, H.C. Ratsch, W.F. Burns, L.C. Grothaus, and
R.W. Field. Monoterpene Emissions Rates from Slash Pine. Corvallis
Environmental Research Center, U.S. EPA report: CERL-045, August 1978.
68. Jones, C.A. and R.A. Rasmussen. Production of Isoprene by Leaf Tissue.
Plant Physiology 55_, pp. 982-987, 1975.
69 . Rasmussen, R.A. and C.A. Jones. Emission Isoprene from Leaf Discs
of Hamamelis. Phytochemistry 12, pp. 15-10, 1973.
70. Tingey, D.T., H.C. Ratsch, M. Manning, L.C. Grothaus, W.F. Burns, and
E.W. Peterson, Isoprene Emission from Live Oak. Corvallis Environmental
Research Laboratory, U.S. EPA Report: CERL-040, May 1978.
71. Stoddard C.H., Essentials of Forestry Practice. Ronald Press Co.,
N.Y., pp. 50-61, 1959.
72 . Zimmerman, P.R. Testing of Hydrocarbon Emissions from Vegetation, Leaf
litter and aquatic Surfaces, and Development of a Methodology for com-
piling biogenic emission inventories - Final Report. EPA-450/4-79-004,
March 1979.
73. Went, F.W. Organic Matter in the Atmosphere, and Its Possible Relation
to Petroleum Formation. Proceedings of the National Academy of Sciences
46_, pp. 212-221, 1960.
74. Lonneman, W.A., R. L. Seila, and S.A. Meeks. Preliminary Results of
Hydrocarbon and Other Pollutant Measurements Taken during the 1975
Northeast Oxidant Transport Study. In the Proceedings of Symposium
on 1975 Northeast Oxidant Transport Study. EPA-600/3-77-017, pp. 40-53,
February 1977.
75. Rasmussen, R.A., R.B. Chatfield, M.W. Holdren, and E. Robinson. Hydro-
carbon levels in a midwest open-forested area. Draft Report submitted
to the Coordinating Research Council, October 1976.
76. Holdren, M.W., H.H. Westberg, and P.R. Zimmerman. Analysis of Mono-
terpene Hydrocarbons in Rural Atmospheres. Unpublished manuscript,
Washington State University.
77. Seila, R.L., R.R. Arnts, R.L. Kuntz, F.L. Mowry, K.R. Knoerr, and
A.C. Dudgeon. Application of the energy balance/Bowen ratio technique
to estimate hydrocarbon fluxes from a pine forest and a comparison
65
-------
with a branch enclosure technique. EPA Technical report (manuscript
in preparation).
78. Whitehead, L. and R.K. Severs. Background hydrocarbon levels in east
Texas. In: Proceedings of American Institute of Chemical Engineers
83rd National Meeting, Houston, Texas, March, 1977.
79. Seila, R.L. Non-Urban Hydrocarbon Concentrations in the Ambient Air
North of Houston, Texas. U.S. Environmental Protection Agency, EPA-
600/3-79-010, February 1979.
8D. Whitby, R.A. and P.E. Coffey. Measurement of Terpenes and Other Organics
in an Adirondack Mountain Pine Forest. Journal of Geophysical Research
82^, p. 5928-5934, 1977.
81. Lonneman, W.A., R.L. Seila, and S.A. Meeks. Preliminary Results of
Hydrocarbon and Other Pollutant Measurements Taken during the
1975 Northeast Oxidant Transport Study. Proceedings of Symposium on
1975 Northeast Oxidant Transport Study, EPA-600/3-77-017, pp. 40-53,
1977.
82 . Noxon, J.F. Nitrogen Dioxide in the Stratosphere and Troposphere
Measured by Ground-Based Absorption Spectroscopy. Science 189,
pp. 547-549, 1975.
83. Noxon, J.F. Tropospheric NO-. Journal of Geophysical Research 83,
pp. 3051-3057, 1978. i
84. Dodge, M.C. and R.R. Arnts. A New Mechanism for the Reaction of
Ozone with Olefins. International Journal of Chemical Kinetics XI,
pp. 399-410, 1979.
66
-------
APPENDIX A
Determination of Ozone-Isoprene Rate Constant
A review of recent kinetic rate constant literature reveals the ozone-
isoprene rate constant has not been reported. The measurement of this rate
constant is important since isoprene and ozone co-exist in the atmosphere,
hence their rate of reaction is of importance to better understanding the
atmospheric chemistry of isoprene. A study was undertaken to determine
experimentally the reaction rate constant for the ozone-isoprene reaction
The ozone-isoprene dark reactions were conducted in 130 X 110 cm
heat-sealed 2-mil FEP Type A Teflon film bags. The Teflon bag was first
filled with 330 liters of purified air containing less than 2 ppb NO and less
«cV
than 50 ppb non-methane hydrocarbon comprised mostly of ethane and
propane. A calculated amount of isoprene (99% purity) was injected into the
purified air stream with a calibrated microliter liquid syringe which had
been cooled to deliver a true liquid sample. After the addition of isoprene,
ozone was introduced into the isoprene-air mixture by passing the air stream
filling the bag through a high voltage discharge ozonator.
Ozone decay in the Teflon bag in the absence of isoprene was less than 1%
per hour. Ozone was monitored intermittently over the 45-minute period of the
experiments with a Bendix Model 8002 ozone-ethylene chemiluminescence in-
strument. This instrument was modified to measure up to 10 ppm full scale.
The signal output was logged by an Esterline Angus (Model D-2020) printing
digital volt meter.
Isoprene decay in the bag in the absence of ozone was negligible during
the time-frame of these experiments. Isoprene was measured with a Perkin-Elmer
900 gas chromatograph with flame ionization detection. The sample was drawn
directly from the bag through a gas-sampling loop and injected via a solenoid-
actuated Seiscor sampling valve. Separation was achieved with a 0.45 m X 3 mm
o.d. stainless steel column containing 60-80 mesh Porapak Q and maintained at
67
-------
100°C. The GC output was interfaced to a Perkin-Elmer PEP-1 data system for
peak integration.
For comparative purposes propylene was reacted with ozone to determine
its rate constant. The results of the propylene and isoprene experiments are
presented in Table 1-A.
TABLE 1-A. REACTION OF OZONE WITH ISOPRENE
AND PROPYLENE
Experimental Data - Reaction at 22 + 1 C, 45 minutes
Initial Hydrocarbon Initial Ozone Ratio Initial Ratio Decrease in
Cone. (PPM) Cone. (PPM) Hydrocarbon to Hydrocarbon Cone, to
ozone Decrease in ozone cone.
Isoprene
Propylene
6.84
0.47
0.61
1.60
5.24
5.61
8.65
0.45
2.29
4.01
1.93
1.12
0.71
0.31
15.1
0.2
0.2
0.8
4.7
7.7
27.6
0.93
0.81
0.65
1.13
1.17
1.19
1.18
In the case of isoprene more ozone than isoprene is being consumed. The
stoichiometry of the reaction is nearer to 1:1 when isoprene is in excess and
the reaction non-stoichiometric when ozone is in excess. This phenomena is
probably due to a second attack by ozone on the remaining double bond of the
products methyl vinyl ketone or methacrolein formed in the first ozone reaction.
The effect is most pronounced when ozone is in excess. Calculated as a
simple bimolecular reaction with 1:1 stoichiometry the isoprene-ozone rate
-2 -1 —1
constant is 1.89 X 10 ppm min . This rate is reasonably close to that
of the 1,3-butadiene-ozone rate constant reported by Hanst (57) of 1.2 X
68
-------
—2 —1 -1
10 ppm min
As a check on the experimental methodology, propylene was investigated
to compare results with published values. Although the determination of
second order rate constant is complicated slightly by competing side
-2 -1 -1
reactions a rate of 1.9 X 10 ppm min was obtained. This rate agrees
extremely well with the rate determined by Japar et^ al^. (55) of 1.92 X 10
ppm min . For a more detailed discussion of the propylene-ozone mechani
see Dodge and Arnts (84).
69
-------
APPENDIX B - INDEX OF FIGURES
Number
IB Irradiations of propylene/nitrogen oxides, C/NO = 4.28 73
X
2B Irradiations of propylene/nitrogen oxides, C/NO = 5.09 74
X
3B Irradiations of propylene/nitrogen oxides, C/NO = 7.44 75
X
4B Irradiations of propylene/nitrogen oxides, C/NO = 9.90 76
X
5B Irradiations of propylene/nitrogen oxides, C/NO = 19.3 77
X
6B Irradiations of propylene/nitrogen oxides, C/NO = 30.3 78
7B Irradiations of propylene/nitrogen oxides, C/NO = 42.4 79
8B Irradiations of propylene/nitrogen oxides, C/NO = 85.0 80
9B Irradiations of propylene/nitrogen oxides, C/NO = 215 81
X
10B Irradiation of isoprene/nitrogen oxides, C/NO - 3.53 82
X
11B Irradiation of isoprene/nitrogen oxides, C/NO = 6.25 83
X
12B Irradiation of isoprene/nitrogen oxides, C/NO - 9.07 84
X
13B Irradiation of isoprene/nitrogen oxides, C/NO = 15.8 85
14B Irradiation of isoprene/nitrogen oxides, C/NO = 24.9 86
X
15B Irradiation of isoprene/nitrogen oxides, C/NO =30.5 87
x
16B Irradiation of isoprene/nitrogen oxides, C/NO =58.8 88
x
17B irradiation of isoprene/nitrogen oxides, C/NO = 90.2 89
x
18B Irradiation of isoprene/nitrogen oxides, C/NO = 223 90
x
19B Irradiation of cx-pinene/nitrogen oxides, C/NO = 1.77 91
X
20B Irradiation of ct-pinene/nitrogen oxides, C/NO = 8.10 92
X
70
-------
Number ' page
21B Irradiation of a-pinene/nitrogen oxides, C/NO = 8.70 93
x
22B Irradiation of a-pinene/nitrogen oxides, C/NO = 18.2.. 94
x
23B Irradiation of a-pinene/nitrogen oxides, C/NO = 32.5 95
x
24B Irradiation of a-pinene/nitrogen oxides, C/NO = 43.3 96
x
25B Irradiation of a-pinene/nitrogen oxides, C/NO = 46.1. 97
x
26B Irradiation of a-pinene/nitrogen oxides, C/NO = 65.3 98
x
27B Irradiation of a-pinene/nitrogen oxides, C/NO = 66.5 '99
x
28B Irradiation of a-pinene/nitrogen oxides, C/NO = 106 100
x
29B Irradiation of a-pinene/nitrogen oxides, C/NO = 195 .............. 101
x
30B Irradiation of d-limonene/nitrogen oxides, C/NO = 1.40 ........... 102
X
31B Irradiation of d-limonene/nitrogen oxides, C/NO = 4.50 ........... 103
x
32B Irradiation of d-limonene/nitrogen oxides, C/RO = 6.30 ........... 104
X
33B Irradiation of d-limonene/nitrogen oxides, C/NO = 11.3 ........... 105
X
34B Irradiation of d-limonene/nitrogen oxides, C/NO = 22.5 ........... 106
X
35B Irradiation of d-limonene/nitrogen oxides, C/NO = 27.0 ........... 107
X
36B Irradiation of d-limonene/nitrogen oxides, C/NO = 45.3 ........... 108
X
37B Irradiation of d-limonene/nitrogen oxides, C/NO = 49.3 ........... 109
X
38B Irradiation of d-limonene/nitrogen oxides, C/NO = 50. 9-. ......... 110
.Jt
39B Irradiation of d-limonene/nitrogen oxides, C/NO = 206 ............ Ill
A.
40B Irradiation of p-cymene/nitrogen oxides, C/NOx = 5.10 ............. 112
41B Irradiation of p-cymene/nitrogen oxides, C/NO = 5.21 .......... . •• 113
X
42B Irradiation of p-cymene/nitrogen oxides, C/NO = 9.00 ............. 114
X
43B Irradiation of p-cymene/nitrogen oxides, C/NOx = 15.5 ............. 115
44B Irradiation of p-cymene/nitrogen oxides, C/NOx = 21.2 ............. 116
45B Irradiation of p-cymene/nitrogen oxides, C/NOx = 32.0 ............. 117
46B Irradiation of p-cymene/nitrogen oxides, C/NOx = 52.8 ............. 118
71
-------
Number Pa8e
47B Irradiation of p-cymene/nitrogen oxides, C/NO = 102 119
X
48B Irradiation of p-cymene/nitrogen oxides, C/NO = 196 120
49B Irradiation of terpinolene/nitrogen oxides, C/NO = 28.3 121
SOB Irradiation of terpinolene/nitrogen oxides, C/NO = 213 122
51B Irradiation of myrcene/nitrogen oxides, C/NO = 31.5 .*,.... 123
52B Irradiation of myrcene/nitrogen oxides, C/NO = 203 124
X,
53B Irradiation of 3-pinene/nitrogen oxides, C/NO = 31.0 125
54B Irradiation of 3-pinene/nitrogen oxides, C/NO = 203 126
55B Irradiation of A -carene/nitrogen oxides, C/NO = 34.0 127
72
-------
•C/NCIX = H-2B
I PDD _.
BDD
snn -
nna -
PAN X 10
400
IB.
200 300
TIME/- MINUTES
IRRRD-IRTION OF" FRDFYL.ENE/'NITRnEEN PXIOES
-------
C/NOX s S.OH
iaan _
BOD
eaa
naa
aaa
a
FIGURE: 25.
100 200 300
TIMEx MH4UTCS
IRRROIHTION OF PRDFaYL-ENE/NITROHEN OXID-ES
-------
-------
C/NDX = s.an
i ana _
ean _
Baa _
nan -
2DO _
TIMEx MINUTES
FIBURE 4B. IRRROIRTION DR PRDPYL-ENE^NITRDBEN PXIOES
-------
•c/Nnx = i
i nap __
BDQ _
BOO _
CD
a.
a.
nnn _
2QP _
D
FISURE 5B.
PAN X 10
25
125
50 75
T1ME> MINUTES
IRRHDIRTIdN OF" PRDRYL-ENEXNITRaBEN DXIQES
-------
C/NDX s 3O.3
iaaa
•xl
00
Baa
BOO
naa
aaa
PROPYLENE X 0.1
•>«•••
NO
FIBURE: SB.
40 60
TIMEx MINUTES
IRRRDIRTION Or RRDPVUEINEXNITRPBErN DXIDETS
-------
CXNPX = H2.H
SDD —.
v£>
m
CL
o.
snn _
t PP
7B-
20 40
Ttr
FRRRIMFmDN C3R
60
MfNUTES
80
PXfDETS
-------
C/NDX =
SDD
aan
oo
o
CO
Ib
t an
PROPYLENE X 0.01
0
FIGURE 8B.
40 60
TIMEx MINUTES
IRRRD-IRTIDN OF" PRDPYUENEXNtTRaBEN OXIDES
-------
•C/NDX a 2 I S
5PP ,_.
oo
HPP ~
3DP
ffi
0.
CL
HPO -
I PP _
MINUTES
9B. IRRHOIRTIPN PF* PRPPVL-ENE^NITRPBEN PXIDE5
-------
CX1MDX = 3.
snn _
00
NJ
HDCJ _
CO
n.
a.
0
FIGURE: IOB.
METHYL VINYL KETONE X 10
100
400
200 300
TIMETx MINUTES
IRRROIRT10N DF" ISDf^RETNEIXNITRnEEZN dXIOES
-------
C/IXIQX = B.23B
00
tfl
CL
EL
nan
enn _
BDD
nnn _
METHYL VINYL KETONE X 10
METHACROLEIN X 10
0
50
200
100 150
TIMEx MINUTES
FIGURE HB. IRRROIRTIDN DF ISDPRENEXNlTRnBENl DXIOES
50
-------
•C/NDX = S.O7
nnn _
snn _
00
*-
a.
a.
sno _
nan
2DD
METHACROLEIN X 10
METHYL VINYL KETONE
a CT" 40 80 120
TIMEZx- MINUTES
RIBURE 12B. IRRHOIRTION DR ISaRREZNEXNITRDtaEIN QXID-E!
-------
CXNDX = I E.B
HDQ ._
oo
Ul
tn
n.
a.
son —
^ NO,
METHYL VINYL KETONE
PAN
METHACROLEIN
ISOPRENE X 0.1 °3 X 0>1
80 120
TIMEIx MIKIUTEIS
13B. IRRRCHRTICIN DF-
160
-------
•CXNOX = 2H.S
son _.
HDQ _
oo
Os
CO
0.
EL
i nn _
METHYL VINYL KETONE
METHACROLEIN
»***
ISOPRENE X 0.1
FIGURE: 14B.
40 60
TIME,/ MINUTES
IRRROIRTIdN DF- ISaPRETNEZX-NfTRClBErN dXIQES
-------
CXNFJX =
snn
oo
m
a.
o.
HDH3
I
:3 nn
RIEURE:
METHYL VINYL KETONE
N02
METHACROLEIN
ISOPRENE X 0.1
PAN
NO
40 60 ~80
TIMEx MINUTES
!RRH£>!HTinN DF" ISnRREIMEI/'NITRnGEIN OXIOETS
20
-------
CXNDX =
son ._
.00
oo
nna
a.
D-
ISOPRENE X 0.1
200 _
na
a
F^IEURE
METHYL VINYL KETONE
PAN
16B.
60
MINUTES
IRRRD-IRTIdN C3F" ISClRRENE/NITRnBEIN DXIOES
-------
-------
C/NDX = 223
EDO
so
o
nan
to
a.
a.
2DD
i na
METHACROLEIN
METHYL VINYL KETONE X 0.1
0
40 60
TIMEx MINUTES
riBURE 18B' IRRRDIRTIDN OF" ISPRRENE/NITRCIGEN QXID-ES
-------
C/NDX s 1 .77
srna _
\o
HPP —
CD
0-
0.
1 E3P _
FltSURET
TlMETx MINUTES
19B. IRRRI>iRT!aN OF"
_J
OXIOETS
-------
•C/NOX s B.I
i nan
Baa
Baa
vO
CD
it
naa
o
400
200 300
TIMEx MINUTES
riBURE 2°B- IRRRDJRTION Or RL-PHR-PINENE/NITROBEN OXID-ES
-------
•C/NDX = B.V
f DDD _
BOP _
son _
a.
a.
nan _
son
FIGURE:
o 100
21B.
200 300 400
TIME:, MINUTES
IRRROIRTION CJF RL-FHR—FINENE/NITRDGETN nXID-ETS
-------
C/NDX s I B.2
SDD _
HDD _
3DD -
tfl
It
zan
i aa _
22B.
100 200
TIME/' MINUTES
IRRRD-IHTIDN dP" RUPHFI-PINENEXNITRDBEN DXI£>ES
-------
= 32.B
SDD -_
VO
m
CL
a.
HDD -
i DD :
80
100
FIGURE 23B.
40 60
TIMEx MINUTES
1RRRD-IRTIDN DF" RUPHR—PINENEZ/NlTRnEEIN CJXID-ES
-------
C/NPX s H3.3
son _
to
CL
o_
nnn
3DD
200
i an
FIGURE 24B.
PAN X 10
40 60 80
TIMEx MINUTES
IRRHOIRTIDN PF RL.PHR-PINENE/NITRDBEN DXIOES
-------
C/NPX = HE. 1
EDD _
HDD _
CO
CL
CL
i nn _
o
20
-PINENE X 0.1
100
40 60
TIME,' MINUTES
FIBURE 25B- IRRROIRTIDN DF* RL-PHR—RINENE/NITRntaEN OXID-ES
-------
C/NDX = SS.3
vo
00
son ,_
HDD
3DD
to
0.
0.
2DO
i aa
0
20
PAN X 10
a.
80
40 60
TIMEls MINUTES
F-IEURE 26B- tRRHoiRTiaN DF- RL-PHR-PINENEXNITRDBEN OXIOES
-------
•CXNDX s BS.S
snn
VD
CD
Q.
Q.
nan _
3DO _
i an
80
40 60
TtME!^ MINUTES
27B. IRRROIRTIOK! OR RL.f=«HR~f3INE:NE:XNITRntHE:N QXIOEZB
-------
CXNDX = IDB
srnn _
HDO
30D
i-1
o
o
2OD
NO
0
20
40 60 80
TIME/- MINUTES
FIGURE 28B. |RRRE>IRTIDN DF RL-RHR~RINENEX"NtTRDBEN DXID-ES
-------
•c/Nnx = f
I ODD
o
H1
ana
BPD
m
a.
a.
nan
2DD
F-IGURE:
29B.
20
40 60
TIMETx MINUTETS
IRRHOIRTION DF~ RLRHR—RINETNEx-NITRntHEIN OXID-ETS
-------
•CXNOX s I .HO
snn
o
S3
HOHJ
3DD
m
tL
0.
i nn
At 310 Minutes
N02 (SALTZMAN) = 104 ppb
PAN X 10
FIGURE 30B.
TlMEx MINUTES
IRRROIRTIDN DF" O —L-IMDNEZNET/NITRPBErN OXIOES
-------
= H.srn
fcD
CL
0.
HDO
3DD
2C3CJ
At 185 Minutes
N02 (SALTZMAN) = 91 ppb
50
PAN X 10
FIGURE: 3iB.
100 150 200
TIMEZx MINUTES
IRF3HD-IRTION OF O — L-IMnNENE/'NITRClEErN DXIOETEi
-------
•C/NDX = B.3D
snn _
ED
0.
n.
3DD
i an
70
At 305 Minutes
N02 (SALTZMAN) =3.0 ppb
FIGURE:
32B-
140 210
TIMEx MINUTES
DF~ [> — L-IMQNENE/NITRaBEN nXIE>ES
-------
CXNE3X = ! I .3
sna
ffl
£L
EL
3DQ
2IHO _
I an _.
At 165 Minutes
N02 (SALTZMAN) = 14 ppb
a
MINUTES
RIBURE: 33B. IRRROIRTION OR o
axit>Es
-------
C/NCDX = 22.
son _
m
a.
a.
nan _
2QD _
i an _
At 225 Minutes
NO (SALTZMAN) = 0 ppb
FIGURE:
34B.
130 195
TIMEx MINUTES
IRRHDIHTIDN OF" D--L_IMnNENE/NITRaC3E:N OXIDES
-------
-------
C/NOX = HS.3
snn _
At 330 Minutes
N02 (SALTZMAN) = 20 ppb
HDD _
o
00
ffl
CL
EL
200
i nn _
n
d-LIMONENE X 0.1
36Bl
100 200 300
TIMEx MINUTES
IRRRD-IHTinN PF~ D —L-IMdNETNEr/NITRnGEIN DXIOETS
-------
s H3.3
KDD _
o
m
a.
a.
nan —
son
2DO _
an
o
0
FIGURE: 37B
70
N00
1J PAN
140 210 280
TIME/ MINUTEIS
OF £> —LIMnNENEr/NITRDEEIN OXIDES
-------
<:/Nnx = sn.s
i nnn __
ana
m
a.
a.
snn
nan
2OP
0
FIGURE
O3xio
NEW BAG
PAN X 10
400
j
38B.
200 300
TIMEx MINUTES
IRRROIRTIDN DF D- —L-fMONENEXNITRnEEN DXID-ES
-------
= 23C3E
nnn __
ann
snn
m
0-
CL
nnn
F-IBURE:
OQT)
Jy
d-LIMONENE X 0.1
o3 x 100
20 40 60
TIMETx MINUTES
IRRRDIHTltUN OF" £> —
OXIOES
-------
C/NPX = S. I
SPP _
HDP
3PP -
m
a.
a.
I PP _
NO,
p-^CYMENE
PAN
FIBURE
40B.
300•
MINUTES
1RRHDIRTIPN PF" PRRR—CYMENEXNITRPBEN PXIDES
-------
C/NDX = K.2 I
SOD _
to
Ik
nan ~,
3DD „
I PD _
100
200
400
FIGURE 41.
300
MINUTES
IRRRDIHTIDN DF PRRR-CYMENEXMITRDBEN OXIOES
-------
CXNDX = a.n
son
HDQ
it
200
a
FIGURE
NO,
p- CYMENE
PAN
42B'
TIMEx MINUTES
IRRROIHTIDN DP PRRR— CYMENE/NITRDEEN DXIDEB
-------
O'NCJX = I 5T.K
SDO
c.
CL
HDD
2DD
f DP
D
FIKURE:
100 200
T!ME> MINUTES
43B. IRRREMRTJDN ^F FRRR-CYMETNErXNtTRDKErN DXID-ES
-------
CXNDX = 2 I .2
noa _
BOO
BDD
m
it
HDD
2OO
O
FIGURE
50
100 150 200 250
TIMEx MINUTES
44B- IRRRDIRTIDN DF PRRR—CYMENE/NITROC3EN DXIDES
-------
-------
C/NdX s S2.B
son
3DD
00
saa
aa
a
FIGURE
46B.
TIMEx MINUTES
IRRROIRTIDN DP FRRR—CYMENE/NITRnEEN OXIDES
-------
CXNDX = I O2
SOD _.
£0
0.
EL
HDD ™
3DP -
PAN X 10
! DC3
P
FIGURE:
25
47B.
50 75 100
T!ME> MINUTES
|RRRI>!RT!DN QF PRRR—CYHErNErx'NITRDKErN DXJOEEi
-------
C/NDX = I SB
o
I ODD
BDO
BOQ
tO
it
HDD
25
p-CYMENE X 0.1
50 75
TIME,* MINUTES
100
riSURE
48B.
IRRRD-IRTIdN DF~ PRRR—CYMENE/NITRDEEN OXIDES
-------
CXNdX = 2B.3
! nan r_
enn
CL
CL
TERPINOLENE
At 90 Minutes
N02 (SALTZMAN) = 15 ppb
35 70 105
TlMEx MlNLJTEEri
iRRROIRTICJN OF" T
-------
-------
CXNdX = 3 ! .
i nnn ,-.,
m
EL
D.
snn
nan
n
0
70
140 210
TIME:,' MINUTES
FMBURE: SIB. iRRROiRTtnr-4 DF*
280
350
-------
CXNDX = 2D3
I ODD
HOD
BOD
NJ
a.
Q.
nan
2DD
rtBURE
52B.
T!ME> MINUTES
nr MYR-CENEXNITRDEEN
-------
sna
HDD
C/NHX = 3 ! .D
Ui
tn
n.
a.
3 no
25
50 75
TIME:/ MINUTES
53B.
IRRRDIRTiaN DF*
— PINEINEIXMSTRIIISEtN
-------
CXNCJX = 2D3
i pnn ,_
ena _
enn _
a.
a.
to
0 20
54B.
B-PINENE X 0.1
60 80 100
MINUTES
OF" BETH—FMNENE/NlTRniGEN OXIDES
-------
C/'NDX = 3H.D
CD
0-
CL
HDD
3QD
2nd L.
i an
n
0
A" - CARENE X 0.1
20
40 60
TtMETx- MINUTES
F-fELJREI 55B. tRRHOFRTICIN DR D-ELTR-S-CRRENEXNITRDBEM
. J
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1 REPORT NO.
EPA-600/3-79-081
2.
4. TITLE AND SUBTITLE
PHOTOCHEMISTRY OF SOME NATURALLY EMITTED HYDROCARBONS
7. AUTHOR(S)
Robert R. Arnts and Bruce
W . Gay , Jr .
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Environmental Sciences Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Sciences Research Laboratory-RTF, NC
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
3. RECIPIENT'S ACCESSION«NO.
5 REPORT DATE
September 1979
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO.
1AA603A AC-019 (FY-77)
11. CONTRACT/GRANT NO.
13. TYPE OF REPORT AND PERIOD COVERED
Tn hnijf?i=>
14. SPONSORING AGENCY CODE
EPA/ 600/09
15. SUPPLEMENTARY NOTES
16. ABSTRACT
Six C _H , monoterpenes, p-cymene, and isoprene, all known or thought to-be
emitted to the atmosphere by vegetation, were irradiated in the presence of NOX.
The terpenes studied included one acyclic triolef in (myrcene) , two monocyclic
diolefins (d-limonene, terpinolene) , and three bicyclic monolefins (a-pinene, 3-
pinene, and A -carene) . Propylene was also studied since this olefin serves as a
point of reference with other chamber studies.
Results showed that monoterpenes and isoprene promoted the oxidation of NO to
N0_ and were themselves consumed at rates comparable to or greater than propylene;
p-cymene was decidedly slow in these respects. The monoterpenes however did not
permit the buildup of ozone due to their rapid reaction with ozone. The ozone
suppression was particularly noticeable at high carbon/NOx ratios. Propylene and
isoprene were more efficient in producing ozone, but exhibited some suppression of
ozone at high carbon/NO ratios. Para-cymene produced a uniform concentration of
ozone independent of the carbon/NO ratio. Deciduous forests, isoprene emitters, are
expected to contribute more to ozone production relative to the monoterpene producing
coniferous forests. Coniferous forests may in fact function as a sink for ozone.
Reported ambient concentrations of isopraie and terpenic hydrocarbons in forested
areas are too low to account for more than a few ppb of ozone even if NO is
available. x
17.
a. DESCRIPTORS
KEY WORDS AND DOCUMENT ANALYSIS
b.lDENTIFIERS/OPEN ENDED TERMS
* Air pollution
* Terpene hydrocarbons
* Biological productivity
Test chambers
* Photochemical reactions
* Nitrogen oxides
* Ozone
13. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (This Report)
UNCLASSIFIED
20. SECURITY CLASS (This page)
UNCLASSIFIED
c. COSATI Field/Group
13B
07C
08A
14B
07E
07B
21. NO. OF PAGES
138
22. PRICE
EPA Form 2220-1 (9-73)
128
------- |