EPA/600/D-87/211
July 1987
IACP EMISSIONS: TRANSFORATIONS AND. FATE
.by
L. T. Cupitt and L. D. Claxton
Atmospheric Sciences Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
P. B. Shepson and T. E. Kleindienst
Northrop Services, Inc.
Research Triangle Park, NC 27709
EPA Project Officer
L. T. Cupitt
ATMOSPHERIC SCIKNCES RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
-------
NOTICE
This document has been reviewed in accordance with
U.S. Environmental Protection Agency policy and
approved for publication. Mention of trade names
or commercial products does not constitute endorse-
ment or recommendation for use.
-------
.IACP EMISSIONS: TRANSFORMATIONS AND FATE
L. T. Cupitt and L. D. Claxton
U." S. Environmental Protection Agency
Research Triangle Park, NC 27711
P. B. Shepson and T. E. Kleindienst
Northrop Services, Inc.
Research Triangle Park, NC 27709
Diluted emissions from wood stoves and automobiles were irradiated in a
Teflon smog chamber to simulate their photochemical transformation in the
atmosphere. Throughout the experiments, the chemical composition and
physical properties of the gaseous and aerosol-bound complex mixtures were
monitored. The mutagenicity of the gas-phase components and of the aerosol -
bound chemicals were measured both before and after irradiation. Transfor-
mation of the dilute wood smoke occurred readily under all conditions
tested: with and without added NO, with artificial illumination and with
natural sunlight, at outdoor temperatures or at room temperature. As the
photochemical reactions progressed, the volume of the aerosol-bound organics
was seen to increase, suggesting that transformation products were condens-
ing to form additional aerosol-phase materials.
The gas-phase reactants and products were tested for mutagenic activity
by exposing Salmonella typhimurium, strains TA 98 and TA 100, to the
filtered effluent. Filter samples of the starting materials and the
transformed products were collected, extracted, and tested for mutagenic
activity in the same bacterial strains by using a standard plate incorpora-
tion test. The transformations caused a dramatic change in the mutagenic
activity of the gas-phase components. Comparisons of the mutagenic activity
between the gas-phase and aerosol-bound chemicals have been estimated. When
the mutagenicity is expressed in units of revertants per cubic meter of air,
the gas-phase reaction products are found to be the most mutagenic.
-------
Introducti on
The Integrated Air Cancer Project (IACP) has focused on emissions from
residential wood combustion (RWC) and from automobiles. As a part of the
IACP, simulation experiments were conducted to characterize the atmospheric
transformations of these complex mixtures. This research effort focused on
three main questions: (1) Are transformations likely to occur? (2) If
transformations do occur, what changes in chemistry and mutagenicity are
induced? and (3) What is the relationship between gaseous and aerosol-bound
mutagens?
Experimental Methods
A schematic diagram of the 22.7 m^ reaction chamber used for most of
these experiments is shown in Figure 1. The chamber is housed in a truck
trailer and consists of a 7.5 m long cylinder of Teflon film suspended
between two aluminum end plates, each of which is coated with fluorocarbon
paint. Irradiation is accomplished by means of blacklight and sunlamp
fluorescent bulbs which surround the chamber. Oak logs, from the Research
Triangle Park area of North Carolina, were burned in a commercially avail-
able wood stove. A portion of the wood smoke was introduced into a dilution
tunnel and mixed with ambient air to cool the mixture. The diluted emis-
sions were then used to fill the chamber to the desired concentration.* A
few, preliminary experiments have recently been conducted using automobile
emissions. The experimental set up is identical to Figure 1, except that
the wood stove was replaced by a 1980 catalyst-equipped Toyota Corolla
operating at high idle conditions and burning a "super unleaded" grade of
gasoline. For a few wood smoke irradiations, a 9.0 m^ outdoor Teflon smog
chamber was used. The outdoor chamber was at ambient temperatures and used
only natural sunlight.
Four chambers were used for exposure of Salmonella typhimurium, strains
TA 98 and TA 100, to the various gaseous mixtures. The chambers are 190-L,
rectangular, Teflon-coated containers capable of holding more than 50 test
plates. "Survivor" plates were routinely included for detection of toxicity
effects. The test mixtures were flushed through the bioassay exposure
chambers at 14 L min"*, and exposures of the bacteria to the gaseous
p;tagens was accomplished simply by uncovering the glass Petri dishes
containing the bacteria and permitting the mutagenic materials to dissolve
into the plates. If one assumes that the "dose" is proportional to the
exposure period, a type of dose-response curve can be generated conveniently
by allowing various groups of plates to remain uncovered for differing
periods of time. The details of the mutagenicity testing have been de-
scribed elsewhere.1*3 Particulate samples from each of the test streams
were collected on Teflon-coated glass fiber filters identical to those used
in the ambient IACP field study. Mutagenicity testing of the particulate
extracts was accomplished using the Ames standard plate incorporation test.4
Four distinctive air streams were tested: the clean air used to replenish
the chamber as samples were withdrawn, the ambient air used in the dilution
tunnel, the reactants (i.e., the diluted wood smoke or auto exhaust prior to
irradiation), and the chamber effluent (i.e., the diluted emissions after
-------
In several experiments, two bioassay exposure chambers were used in
series to test the effluent from the reaction chamber. Not all of the
gaseous mutagens entering the first bioassay exposure chamber are removed by
the test plates. Data from the second chamber permits a better estimate of
the total vapor-phase mutagenicity in the test air stream. The calculated
gaseous mutagenicity may still be a lower limit estimate, however, since
non-polar mutagens are not likely to be efficiently removed, even by two
chambers in series.
The experiments were conducted by filling the reaction chamber with the
wood smoke or automotive emissions to the desired concentration ("15 ppm
Carbon), adjusting the N0X concentration, if necessary, and then irradiating
the mixture until an ozone maximum was reached. The lights were then turned
off, and the effluent bioassay exposures were begun. For most of the wood
smoke irradiations, additional NO was added to the system to increase the
extent of reaction and to bring the hydrocarbon to N0X ratio more in line
with those measured in urban and suburban areas.^ Additional NO was not
needed for the experiments involving automotive exhaust. Irradiation of the
outdoor chamber was controlled by removal and re-installation of an opaque
cover. The wood smoke irradiations usually took from one to two hours to
reach the ozone maximum, while the automotive exhaust experiment was
irradiated for about 5 hours. The bioassay exposures typically lasted up to
10 hours. Throughout the bioassay exposure period, the sample withdrawn
from the reaction chamber had to be continuously replaced with clean air.
This meant that the mutagens in the reaction chamber became more and more
dilute with time (*4.5% per hour). An "effective" exposure time was calcu-
lated for the effluent bioassay chamber to account for the effects of
dilution.
Results
Changes in the chemical composition and mutagenic potency were readily
observed for all irradiation conditions. Table I lists the initial and
final concentrations of a variety of chemicals observed during three of the
wood smoke irradiations. Experiment A was conducted with dilute wood smoke
alone, while experiments B and C contained additional N0X. The addition of
N0X to the dilute wood smoke irradiations caused the system to react more
completely and to produce even more mutagenic products than did Experiment
A. Similar experiments conducted iti the outdoor chamber gave essentially
identical results for both chemistry and mutagenicity, despite differences
in the light intensity and distribution and the somewhat lower (by 9x C)
temperatures. Irradiation of the wood smoke mixture caused the aerosol
volume distribution to increase, suggesting that transformation products may
be condensing out on the aerosols. The mutagenicity of both the gas-phase
and the aerosol components were measured before and after irradiation. When
two bioassay exposure chambers were used in series to measure the gas-phase
mutagenicity, the response in the second chamber was about 30% of that in
the first chamber. This implies that the bioassay exposure chamber is
around 70% efficient at removing the mutagens.
Figure 2 shows the comparison of the mutagenicity associated with the
gas-phase and the aerosol-bound organics, both before and after irradiation.
The mutagenicity in Figure 2 is reported in revertants m~3 and is shown for
direct acting mutagens (i.e., without metabolic activation) in two bacterial
-------
strains (TA 98 and TA 100). Prior to irradiation, the bulk of the mutagen-
icity was found 1n the particulate phase. After irradiation, however, the
gaseous transformation products contribute significantly (80 to 99%) to the
total mutagenic burden in the air. It should also be noted that the
mutagenic products proved to be very stable 1n the reaction chambers. As
described in the previous section, dose response curves were derived from
exposures which lasted as long as 10 hours after the lights were turned off.
The dose response curves remained linear over this time period, implying
that the mutagenic vapor-phase products are long lived.
Insertion of an XAD-2 trap into the line between the irradiation
chamber and the bioassay exposure chamber caused the mutagenic response to
decrease by more than 80%. Unfortunately, only about 10% of the removed
mutagenicity was recovered in the XAD-2 extract. Many of the vapor-phase
mutagens removed by XAD-2 may either be unstable on the adsorbent, or may be
lost during the extraction and concentration procedures.
A series of preliminary experiments were also conducted using diluted
exhaust from an idling automobile to fill the simulation chamber. Once
again, the gas-phase transformation products dominated the total mutagenic
burden after irradiation.
Conclusions
Irradiations of complex mixtures involving both wood sr^oke and auto-
mobile exhaust have demonstrated that chemical reactions are probable and
that the gas-phase products which result from these chemical reactions can
constitute the major portion of the total atmospheric mutagenic burden.
Chemical and mutagenic changes were observed for all tested conditions (with
and without added N0X; artificial or natural illumination; controlled indoor
environment or cooler outdoor temperatures), but the mutagenic response
seems to increase with increasing chemical reaction. The mutagenic gas-
phase products have been shown to be quite stable in the simulation cham-
bers, with lifetimes considerably longer than many residential exchange
rates.6 These results suggest that transformation of the gas-phase and
aerosol components of complex mixtures may contribute significantly to the
total burden of mutagens or carcinogens in the environment and should be
considered in assessing risk.
-------
REFERENCES
1. T. E. Kleindienst, P. B. Shepson, E. 0. Edney, L. D. Claxton, L. T.
Cupitt, "Wood smoke: measurement of the mutagenic activities of its
gas- and particulate-phase photooxidation products," Environ. Sci.
Techno!. 20: 493 (1986).
2. L. D. Claxton, "Assessment of bacterial mutagenicity methods for
volatile and semi volatile compounds and mixtures," Environ. Int. 11:
375 (1985).
3. P. B. Shepson, T. E. Kleindienst, E. 0. Edney, G. R. Namie, J. H.
Pittman, L. T. Cupitt, L. D. Claxton, "The mutagenic activity of
irradiated toluene/N0x/H20/air mixtures," Environ. Sci. Technol. 19:
249 (1986).
4. B. N. Ames, J. McCann, E. Yamasaki, "Methods for detecting carcinogens
and mutagens with the Salmonella/mammalian-microsome mutagenicity
test," Mutat. Res. 31: 347 (1975).
5. K. Baugues, "A review of NMQC, NOx and NMOC/NOx ratios measured in 1984
and 1985," EPA-450/4-86-015, U.S. EPA, Research Triangle Park, NC,
1986.
6. R. Zweidinger, S. Stump, D. Dropkin, R. Drago, S. Tejada, "Volatile
organic hydrocarbon and aldehyde composition in Raleigh, NC, during the
-------
Table I. IACP Transformation Study: Gas Phase Concentrations of
Chemicals During Wood Smoke Irradiations (ppb, v/v)
Experiment A
Experiment B
Experiment C
Species
(No NOx Added)
(NOx
Added)
(NOx Added)
Initial
Final
Initial
Fi nal
Initial
Final
Nitric Oxide
75
0
454
0
461
0
N0X
135
64
657
252
576
259
Ozone
0
79
0
467
0
696
CO (ppm)
33.5
32.0
38.0
33.4
38.7
35.5
Ethylene
702
652
537
313
847
439
Benzene
68
62
62
50
102
68
Toluene
27
16
62
15
24
10
Formaldehyde
325
381
269
365
229
383
Acetaldehyde
140
106
88
109
57
75
PAN
0
52
0
174
0
232
HC (ppm-C)
20.6
19.1
16.4
13.2
17.2
15.0
HC/N0X
153
25
30
Figure Captions
Figure 1. Schematic diagram of the experimental apparatus used in the IACP
wood smoke study.
Figure 2. Graphical comparison of the mutagenicity of the gas- and particu-
late-phase components of wood smoke, before and after irradiation.
(The mutagenicity was measured without metabolic activation and is
-------
Experimental Schematic of the Wood Stove,
Reaction Chamber, and Exposure Apparatus
Dilutlon 0-700 cfm
Mir in
Turbine
Dilution Tunnel
Exhaust
Particulate
Filter
Reactants
Biochamber
Gate
Valve
Metal Bellows
Pumps
=3 Ambient
Air In
Particulate
Filter
Wood
Stove
Ambient Air
Biochamber
1.0% NO
in N2
Particulate
Filter
Teflon
Transfer Lines
Clean
Air
Generator
Lights
^Mass
Flow
Controller
Effluent
Bio-
chamber
Aluminum End Plates
Inlet
Manifold
Reaction Chamber
22.7 m3
Samples
and
Exhaust
Mixing
Fan
Clean Air
-------
Gas and Particulate Phase Mutagenicity
of Dilute Wood Smoke + NOx in Air
17500 \-
m 15000
4->
O)
E
u 12500 t-
n
ZD
Q)
CC
10000
en 7500
5000
2500
0
TA 100
G P
Before After
Irradiation Irradiation
K&&3 Gas Phase
Particulate
Phase
TA 98
G
G P
Before After
-------
TECHNICAL PEPOPT DAT'
(Pleese reed on ti e reveue befor :tii
\. REPORT NO- 2.
EPA/600/D-S7/211
3. RECIPIENT'S ACCESS) 0*P NO.
PB8 7
A. TI-^E AND SUBTITLE
IACP EMISSIONS: TRANSFORMATIONS AND.FATE
6. REPORT DATE
July 1987
6. PERFORMING ORG AN 12 AT ION CODE
7 AUTHORIS)
L. Cupitt, L. D. Claxton, P. B. Shepson*
and T. E. Kleindienst *
B. PERFORMING ORGANIZATION REPORT NO.
9, PERFORMING ORGANIZATION NAME AND ADDRESS
ASRL£ HERL
U.S. EPA, RTP, NC 27711
Northrop Services Inc.*
RTP, NC 27709
10. program element no.
A101/C/83/02 5097 (FY-87)
11. CONTRACT/GRANT NO.
12 SPONSORING AGE NC V NAME AND ADDRESS
Atmospheric Sciences Research Laboratory - RTP.NC
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
13. TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
EPA/600/09
is. Supplementary notes
\
16. ABSTRACT
^'•As part of the Integrated Air Cancer Project (IACP), diluted emissions from
wood stoves and automobiles were injected into a Teflon smog chamber and irradiated
to simulate their photochemical transformation in the atmosphere. Changes in the
chemical composition and physical properties of the gaseous and aerosol-bound
complex mixtures were monitored throughout the experiments. The mutagenicity
of the gas-phase components and of the aerosol-bound chemicals were both measured
before and after i rradiation.--\Transformation of the dilute wood smoke occurred
readily under all conditions tested: with and without added NO, with artifical
illumination and with natural sunlight, at outdoor temperatures or at room
temperature. As the photochemical reactions progressed, the volume of the
aerosol-bound organics was seen to increase, suggesting that transformation
products were condensing,-to form additional aerosol-phase materials. The photo-
chemical transformations caused a dramatic change in the mutagenic activity of the
gas-phase components.^Comparisons of the mutagenic activity between the gas-phase
and aerosol-bound chemicals have been estimated. When the mutagenicity is expressed
in units of revertants per cubic meter of air, the gas-phase reaction products are
found to be the most mutagenic. Similar results were obtained from preliminary
experiments using an idling automobile as the pollutant source.
17. • KEY WORDS AND DOCUMENT ANALYSIS
2. DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSati Field/Group
—
IB. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
18 SECURITY CLASS (Thu Rrport)
UNCLASSIFIED
21. NO. I^PAGES
20 SECURITY CLASS (Ttlil paft)
UNCLASSIFIED
22. PRICE
------- |