* "'
United States
Environmental Protection
Agency
' -«
Atmospheric Sciences ^
Research Laboratory ^
Research Triangle Park NC 27711 ',
Research and Development
EPA/600/S3-86/049 Dec. 1986
&EPA Project Summary
Mutagenic Activities of Wood
Smoke Photooxidation
Products
T. E. Kleindienst, P. B. Shepson, and E. O. Edney
The full report presents the results of
experiments designed to evaluate the
mutagenic potential of wood smoke in
the environment. The experiments
were conducted by injecting emissions
from a wood stove into a Teflon smog
chamber and irradiating the diluted
mixture. The mutagenic activity of the
gas phase component was tested by
exposing the bacteria Salmonella ty-
phimurium, strains TA98 and TA100, to
the filtered effluent after the irradiation.
The particulate phase was tested for
mutagenic activity using the plate in-
corporation procedure on the filter ex-
tracts. The data show that the irradi-
ated mixture is more mutagenic than
the direct wood smoke emissions for
TA100 in the gas phase and for TA98 in
both the gas and particulate phases.
Comparison of the mutagenic activities
of the gas and particulate phase com-
ponents indicated that the activity of
the gas phase transformation products
was much greater than the particulate
phase when expressed in revertants
per cubic meter of air.
This Project Summary was devel-
oped by EPA's Atmospheric Sciences
Research Laboratory, Research Triangle
Park, NC, to announce key findings of
the research project that is fully docu-
mented in a separate report of the same
title (see Project Report ordering infor-
mation at back).
Introduction
Recently, there has been increased
concern that human exposure to wood
stove and fireplace emissions may pro-
vide a public health concern. Several
published reports have indicated sub-
stantial increases in the mutagenic ac-
tivity of the particulate phase of wood
smoke after reaction in the dark with 03
and N02 and that the most polar com-
ponents of the extract gave rise to the
majority of the mutagenic activity.
There has been little study of the mu-
tagenic activity of the gas phase emis-
sions from wood stoves. Recent investi-
gations have shown that, whereas low
molecular weight alkenes and aromat-
ics themselves show relatively low mu-
tagenic activity, the photolytically in-
duced reaction products can show
significant activity. Since there are sig-
nificant amounts of low molecular
weight reactive hydrocarbons in wood
smoke, the photooxidation products
might be expected to show substantial
mutagenic activity. In this study we re-
port the results of several experiments
in which the total mutagenic response
in Salmonella typhimurium strains
TA98 and TA100 was measured for di-
lute mixtures of wood smoke irradiated
in the presence of NOX.
Procedures
The experimental apparatus con-
sisted of three major components:
(1) the wood stove and dilution tunnel,
(2) the reaction chamber, and (3) the
exposure chambers. The wood stove/di-
lution tunnel combination allowed di-
lute mixtures of wood smoke to be in-
jected into the reaction chamber. The
chamber was loaded such that the initial
total hydrocarbon concentration was
about 18 ppmC.
The reaction chamber was a 22.7-m3
cylindrical vessel of 0.13-mm Teflon
sealed to fluorocarbon-coated, alu-
minum end plates. The chamber was
surrounded longitudinally with a com-
-------
bination of sun lamps and ultraviolet
black light. In some experiments, nitric
oxide was added to the dilute wood
smoke in the reaction chamber such
that the total NOX concentration was
about 600 ppb. NO was added to in-
crease the hydrocarbon conversion dur-
ing the irradiation such that a sufficient
quantity of reacted material would be
available for the exposure. Experiments
were also performed in which the only
source of NOX was from the combus-
tion.
The experiments were conducted by
loading the reaction chamber with
wood smoke to the desired HC level for
several hours. For the experiments with
additional NOX, NO was added immedi-
ately before the start of irradiation.
When the reaction mixture had been
sufficiently mixed, the lights were
turned on, starting the photochemical
conversion. The irradiation continued
until the ozone and peroxyacetyl nitrate
(PAN) that formed in the system
reached constant levels. The chamber
lights were then turned off, and the ef-
fluent was filtered and allowed to flow
into the exposure chamber, which con-
tained the biological assay. After expo-
sure, the filter was solvent extracted,
and the extract was tested for muta-
genic activity. Three exposure cham-
bers used as controls allowed measure-
ment of the gas phase mutagenic
activity of the clean air, the ambient air
used in the dilution tunnel, and the ini-
tial reactant mixture. Filters were also
collected and extracted for the ambient
air and reactant controls.
The mutagenic activity of the particu-
late phase was measured using the
standard plate incorporation test with
the S. typhimurium strains TA98 and
TA100. The gas phase exposures were
performed using petri dishes containing
the same strains of bacteria seeded in
base agar, which were placed in an ex-
posure chamber that was connected to
the reaction chamber. The air from the
reaction chamber was filtered and
passed over the bioassay plates in the
exposure chamber. During the expo-
sure, gas components in the air mass
that were soluble dissolved into the
agar medium, thus allowing contact
with the biological assay.
Inorganic and organic gas phase spe-
cies were monitored by continuous gas
analyzers and gas chromatography.
Aldehyde concentrations were meas-
ured by derivatization in a solution of
2,4-dinitrophenylhydrazine and sub-
sequent analysis of the formed hy-
drazones by high performance liquid
chromatography (HPLC). The total hy-
drocarbon (HC) concentration was
measured using a total hydrocarbon an-
alyzer with flame ionization detection.
The size distribution of the paniculate
matter in the range 0.01 to 1 |xm was
determined using an electrical aerosol
analyzer, and the total particle concen-
tration was measured using a conden-
sation nuclei counter. The particles col-
lected on the glass fiber filters were
extracted and analyzed for polynuclear
aromatic hydrocarbon (PAH) concentra-
tion by GC/MS.
Results
Four gas phase exposure experi-
ments were performed. In one experi-
ment, dilute wood smoke alone was ir-
radiated and the effluent used in the
exposure. In the other three experi-
ments, 0.5 ppm of NO was added to the
diluted wood smoke in the reaction
chamber. In one of the experiments
with added NOX the exposure was per-
formed using S. typhimurium with
metabolic activation (S9).
For the irradiation with wood smoke
only, 135 ppb of NOX was available (as a
result of the combustion), giving an HC/
NOX>100. As a result, the photolytic
conversion of reactants to products was
extremely rapid, but the extent of con-
version was limited by the low concen-
tration of NOX. The major reactive gas
phase components for wood smoke in-
cluded low molecular weight alkenes,
aromatic compounds, oxygen atom het-
erocyclic compounds, and aldehydes.
The aldehydes were present not only as
reactants in the initial mixture but also
as products formed during photooxida-
tion.
The initial particulate distribution was
in the range of 0.01 to,1 ji-m, with the
maximum in the number distribution
occurring at about 0.1 n,m. After irradia-
tion, the total number of particles in this
range decreased, and the maximum in
the number distribution shifted to
0.2 (im. The volume distribution of the
particles initially present had a maxi-
mum at 0.2 p,m. After irradiation, the
total volume of particles in the range
0.01 to 1 fj,m increased, with the maxi-
mum in the volume distribution occur-
ring at 0.4 urn. This increase probably
resulted from the absorption of gas
phase photooxidation products onto
the surface of particles already present.
The particulate matter which was col-
lected and extracted indicated the pres-
ence of several PAHs (e.g., pyrene, fluo-
ranthene, anthracene, fluorene,
chrysene). The PAHs showed significant
degradation after irradiation, some of
which appears to result from chemical
reactions.
In the experiment where wood smoke
alone was irradiated, the gas phase
component of the product mix showed
mutagenic activities of 36 ± 6 and
3.4 ± 0.9 revertants/h for TA100 and
TA98, respectively. The gas phase start-
ing materials showed no mutagenic ac-
tivity within the experimental error. The
mutagenic activities for the particulate
phase reactants, as well as the irradi-
ated mixture, were weak (0.2 to 0.3 re-
vertants/|Ag) for both strains.
Three of the four irradiations, how-
ever, contained additional NO. Since
NO converts H02 to OH during irradia-
tion, greater conversion of reactants to
products was achieved in these experi-
ments. The gas phase reactant mixtures
(with 0.5 ppm NO) showed no muta-
genic activity above the control levels
for either strain. However, the muta-
genic activity of the irradiated gas phase
products increased to an average of
174 ± 16 revertants/h for TA100 and
30 ± 4 revertants/h for TA98. The muta-
genic activity of all particulate phase ex-
tracts remained at a level of 0.2 to 0.3
revertants/mj except that the extract re-
sulting from the product mixture gave a
mutagenic activity of 0.9 ± 0.3 rever-
tants/n,g withTA98.
An experiment was also performed to
determine whether wood smoke emis-
sions (with 0.5 ppm additional NO) or
the irradiated mixture showed substan-
tially different mutagenic activities
when S9 metabolic activation was
added to the assay. Within the experi-
ment error the data showed that adding
S9 did not change the previously ob-
served mutagenic activities (without
S9).
Discussion
The results of these experiments indi-
cate that irradiation of wood smoke and
wood smoke/NOx mixtures can increase
their mutagenic activity over that of the
reactant mixture. The effect is most ap-
parent for the gas phase component.
However, identifying the species which
give rise to the observed mutagenic ac-
tivity is impractical for a number of rea-
sons. Wood smoke emissions represent
a complex mixture; for example, the
gas phase species identified account for
only half of the available carbon. For
most of the reactant compounds, reli-
able photooxidation reaction mecha-
-------
nisms have not been elucidated. Finally,
even for 'an irradiated system in which
the mechanism has been fairly well elu-
cidated and which shows a mutagenic
response (e.g., propylene/NOx), the
identity of the products, giving rise to a
large fraction of the mutagenic activity,
has not been established.
Instead, the data have been analyzed
in a way to evaluate the relative contri-
bution of each phase to the total muta-
genic activity for each strain. The pri-
mary difficulty in this approach is the
calculation of the mass of material de-
positing the assay during the gas phase
exposure. By measuring the total HC
concentration in the exposure chamber
before and after deposition into the
assay occurs, an estimate of the loading
rate of the gas phase species can be
made.
For the experiments with added NOX
about 10% of the gas phase product
concentration appeared to deposit in
the exposure chamber. We estimate an
average molecular weight of 18.5 g as-
sociated with each mole carbon which
deposits, based on the average molecu-
lar weight (per carbon atom) for the spe-
cies which have been identified in the
product mixture. Thus a deposition rate
of 20 >j,g plate"1 h~1 was determined for
the gas phase products. Combining this
number with the mutagenic activity (re-
vertants/h), the mutagenic activity of
the soluble gas phase products was cal-
culated on a mass basis. This calculated
value is an upper limit, however, since
there is evidence that the mutagenic ac-
tivity of the nonsoluble component is
less than that of the soluble component.
A lower limit to the mutagenic activity
of the gas phase species is obtained by
assuming that the mutagenic activity of
the nonsoluble component is zero. On
the other hand, the mutagenic activity
of the paniculate phase is straightfor-
ward since a known dose of the extract
can be added to the biological assay.
Table 1 summarizes the mutagenic ac-
tivities for the two phases.
A comparison of the mutagenic activ-
ities on a volume basis is perhaps more
pertinent since this calculation includes
the relative quantities of material
present in each phase. The results for
this comparison are shown in Table 2,
with the range for the gas phase prod-
ucts corresponding to the lower and
upper limits given in Table 1.
In summary, the study indicates that
the photooxidation products of wood
smoke show far greater mutagenic ac-
tivity (for TA100 and TA98) than the
emissions. In addition, the contnribu-
tion to the mutagenic activity of the gas
phase products are at least as great, if
not significantly greater, than that of the
particulate phase products for both
strains.
Table 1. Comparison of the Gas and Particulate Phase Mutagenic Activity (Revertants/ng)
for Wood Smoke and Irradiated Wood Smoke, Strains TA100 and TA98 Without
Metabolic Activation
TAWO
TA98
Reactants
Products
Gas
0
1.2-8.5
Particulate
0.3
0.3
Gas
0
0.2-1.5
Particulate
0.2
0.9
Table 2. Comparison of the Gas and Particulate Phase Total Mutagenic (Revertants/m3) for
Wood Smoke and Irradiated Wood Smoke, Strains TAWO and TA98 Without
Metabolic Activation
TAWO
TA98
Gas
Particulate
Gas
Particulate
Reactants
Products
0
12,300-90,600
100
180
0
2,130-16,000
80
730
T. E. Kleindienst, P. B. Shepson. and E. 0. Edney are with Northrop Services,
Inc.—Environmental Sciences, Research Triangle Park, NC 27709.
L. T. Cupitt is the EPA Project Officer (see below).
The complete report, entitled "Mutagenic Activities of Wood Smoke Photo-
oxidation Products," (Order No. PB 86-239 837/AS; Cost: $ 11.95, subject to
change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Atmospheric Sciences Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC27711
-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45260
Official Business
Penalty for Private Use $300
EPA/600/S3-86/049
------- |