United States
Environmental Protection
Agency
Atmospheric Research and Exposure - "' „
Assessment Laboratory "/ ^
Research Triangle Park NC 27711
Research and Development
EPA/600/S3-89/055 July 1989
Project Summary
Statistical Comparison of
Results of Two Indoor Air Pilot
Studies
Gregory A. Mack, Jerry W. Stockrahm, and Jane C. Chuang
The objective of this study was to
compare results between two pre-
vious indoor air PAH monitoring
studies conducted by EPA in 1984
and 1987. Both of the previous
studies were pilot studies involving
ambient and indoor air monitoring at
a small number of residences in
Columbus, Ohio. The objectives of
these studies were to characterize
ranges of selected PAH, nitro-PAH,
and nitrogen heterocyclic com-
pounds in the indoor air and to
evaluate contributions of various
indoor combustion sources to the
indoor concentration levels of these
compounds. The 1984 study used an
EPA medium volume air sampler
equipped with a quartz fiber filter and
a polyurethane foam (PUF) vapor trap,
while the 1987 study used a new
prototype air sampler equipped with
a quartz fiber filter and an XAD-4
resin vapor trap.
This current study compared the
results between the 1984 and 1987
studies to determine whether the
results are consistent. Where the
results were found to be consistent,
the data from the two studies were
combined to produce more precise
statistical estimates of concentration
level ranges and more precise esti-
mates of the contributions of differ-
ent indoor combustion sources to
indoor PAH levels.
Concentration level ranges were
found to be consistent between the
1984 and 1987 studies for all com-
pounds except quinoline and iso-
quinoline. For quinoline and isoquin-
oline, excessive breakthrough in the
PUF vapor traps used in the 1984
study resulted in lower measured
concentration levels than found in
the 1987 study. The estimated PAH
contributions of indoor combustion
sources were also found to be
consistent between the studies for all
compounds except quinoline and
isoquinoline. The data from the 1984
and 1987 study were therefore
combined. Using the combined data,
cigarette smoking and gas heating
systems were found to be the
greatest contributors to indoor PAH
concentration levels. However, the
data indicate that the effects of gas
heating systems and gas cooking
appliances are not known with as
much certainty as are the effects of
cigarette smoking.
Finally, data from the 1984 study
were found to be consistent with
models previously developed from
the 1987 data for predicting levels of
PAH target compounds from meas-
ured levels of potential PAH marker
compounds.
This Project Summary was devel-
oped by £PA's Atmospheric Research
and Exposure Assessment Laboratory,
Research Triangle Park, NC, to an-
nounce key findings of the research
project that is fully documented in a
separate report of the same title (see
Project Report ordering information at
back).
Introduction
The EPA as part of its indoor air
methods development program, has
-------
previously conducted two separate
range-finding studies of indoor air levels
of polynuclear aromatic hydrocarbon
(PAH) compounds. The first study was
conducted in 1984 and was titled "Pilot
Study of Sampling and Analysis for
Polynuclear Aromatic Compounds m
Indoor Air." The second study was
conducted in 1987 and was titled "Field
Evaluation of Sampling and Analysis for
Organic Pollutants in Indoor Air."Both
studied involved monitoring of PAH, nitro-
PAH, and nitrogen heterocyclic com-
pounds in ambient and indoor air at a
small number of homes in Columbus,
Ohio. The objectives of both studies were
to evaluate sampling and analysis meth-
odology, to characterize typical con-
centration levels of PAH and related
compounds found in the indoor air, and
to investigate the individual contributions
of different indoor combustion sources on
PAH levels in the home.
The objective of this study was to
compare the results of the 1984 and 1987
studies, and to use the data from the
1984 study to evaluate some of the key
findings of the 1987 study.
Ranges of the concentration levels for
the measured compounds were com-
pared across studies. Also, the estimated
pollutant contributions of various indoor
combustion sources were compared.
Whenever the results were consistent,
the data from the two studies were
pooled to produce an updated set of
results based on the combined data from
both studies. Key questions addressed in
this study include:
1. Are the observed ranges of concen-
tration levels of PAH, nitro-PAH, and
nitrogen-heterocyclic compounds
similar across studies? Are there
differences in concentration levels
due to differences in the degree of
vapor trap breakthrough experienced
between the two studies?
2. Are the estimated contributions of
gas heating systems, gas cooking
appliances, and cigarette smoking to
the PAH concentrations in indoor air
consistent between the 1984 and
1987 studies? In particular, results
from the 1987 study indicated that
use of electric cooking appliances
contribute more to the PAH levels
than do gas appliances. Do the data
from the 1984 study refute or confirm
this finding?
3. How well can we predict the levels of
various PAH target compounds from
the measured levels of potential
marker compounds such as pyrene,
fluoranthene, and phenanthrene? Are
the prediction models developed
from the 1984 study data consistent
with the prediction models developed
from the 1987 study data?
Procedure
In the 1984 study, a medium volume
EPA sampler was used which had a
quartz fiber filter to collect particulate
matter and a PDF cartridge to trap
vapors. The study investigated the effects
of (1) gas heating systems, (2) cigarette
smoking, and (3) woodburning fireplaces
as possible contributors to indoor levels
of PAH, nitro-PAH, and nitrogen
heterocyclic compounds. Ten sample
homes were selected to have different
characteristics relative to the presence of
gas versus electric heating systems,
presence/absence of woodburning fire-
places in use, and the presence/absence
of cigarette smoking in the home. At
each sample home, 8-hour samples were
taken at each of three indoor locations
within the home: kitchen (7:00 AM to 3.00
pm), living room (3:00 PM to 11:00 PM),
and bedroom (11:00 PM to 7:00 AM. A
24-hour ambient air sample was also
taken outside the home to coincide with
the three 8-hour indoor samples. For
each home, the air exchange rate was
determined by the decay rate of SF6
injected into the house. A questionnaire
was completed by the residents of each
sample home to record residents' activ-
ities during the sampling periods that
might have affected concentrations of
indoor PAH levels.
The 1984 study found that cigarette
smoking and type of heating system have
the greatest effects on the indoor levels
of most PAH, nitrogen heterocyclic com-
pounds, and nitro-PAH compounds. The
use of a woodburning fireplace was only
weakly correlated with the levels of the
target compounds. The study also found
that bedroom levels were much lower
than levels found in the kitchen and living
room.
The 1987 study used a new prototype
indoor air sampler developed by EPA's
Environmental Monitoring Systems Lab-
oratory. This new sampler allows the
pumping unit to be placed inside the
home with the air sampling unit. The
sampler was equipped with a quartz fiber
filter to collect particulate matter and
XAD-4 resin to trap vapors.
The 1987 study investigated the
contributions of (1) gas heating systems,
(2) gas cooking appliances, and (3)
cigarette smoking to the indoor con-
centration levels of PAH and related
compounds. At each sample home, 8-
hour samples were taken at each of two
indoor locations within the home: kitchen
(7:00 AM to 3:00 pm) and living room
(3:00 PM to 11:00 PM). A 16-hout
ambient air sample was also taken
outside the home to coincide with the twc
8-hour indoor samples. At each home
the air exchange rate was determined by
the decay rate of SF6 injected into the
house. A questionnaire was completed by
the residents of each sample home tc
record the number of cigarettes smoked
and various other activities during
sampling that might have affectec
concentrations of indoor PAH compounc
levels.
Results of the 1987 study found thai
the type of heating system and cigarette
smoking had the greatest effects on the
indoor levels of most PAH, nitroger
heterocyclic, and nitro-PAH compounds
And, although not statistically significan
for most compounds, homes with electric
cooking appliances rather than gas
cooking appliances had higher estimatec
levels of most target compounds. This i;
contrary to what one would expect and i;
one of the key findings investigated if
this comparison study.
Results
Except for quinoline and isoquinoline
there was good agreement between thi
concentration ranges for the 1984 am
1987 studies. The discrepancies ii
ranges for quinoline and isoquinoline ar
due to breakthrough of the PUF trap
used in the 1984 study. The 1987 stud
used XAD-4 resin vapor traps which di
not experience the same level c
breakthrough.
Except for the compounds quinolin
and isoquinoline, there was no statistic;
evidence that the estimated effects of ga
heating system, gas cooking appliance:
cigarette smoking, and sampling locatio
differed significantly between the tw
studies. However, in the 1987 stud)
homes having electric cooking appliance
were estimated to have higher levels tha
gas appliance homes for a number <
compounds. For the 1984 study, ga
appliance homes had higher averag
levels than electric appliance homes f(
most of the target compounds. Electr
appliances were associated with highi
levels only for the compounds phei
anthrene, fluoranthene, pyrene, an
coronene. However, the difference
between types of cooking appliance
were not statistically significant for any
the compounds except benzo(e)pyrer^
-------
Results from upcoming studies will
provide additional evidence to determine
the true effect, if any, of gas versus
electric cooking appliances.
Because there was no statistical
evidence of a difference in results
between the 1984 and 1987 studies
(except for quinoline and isoquinoline),
the data for the two studies were
combined. The combined study data
provided more precise estimates of the
true contributions of the various different
combustion sources to indoor PAH levels.
The combined data showed that gas
heating systems and cigarette smoking
produce the largest increases in indoor
PAH concentration levels.
Results from the 1987 study showed
that there are significant linear relation-
ships between potential marker com-
pounds and other target PAH com-
pounds. In the 1987 study, correlation
coefficients between potential marker
compounds and most target compounds
generally ranged from 0.30 to 0.80. Data
from the 1984 study were used to assess
the validity of the models developed from
the 1987 data. Results from the com-
parison of models indicate that, except
for a few compounds, there is no
statistical evidence of systematic differ-
ences between the 1984 and 1987 pre-
diction models. However, there was con-
siderable within-study variability among
the 1984 data and 1987 data. The within-
study variability was sufficiently large so
that the between-study variability was not
considered to be statistically significant
for most compounds. The data from the
two studies were pooled to produce a
single new prediction equation for each
compound. Pyrene was found to be the
best overall marker compound.
Recommendations and
Conclusions
The measured concentration ranges
were found to be consistent between the
1984 and 1987 studies. The quinoline
and isoquinoline levels found in the 1984
study were an order of magnitude lower
due to excessive breakthrough in the
PDF vapor traps used in the 1984 study.
From these results we conclude that PUF
is not as effective as XAD-4 resin for
trapping vapors of volatile compounds.
Estimates of the contributions of
different indoor combustion sources to
the indoor PAH concentration levels were
consistent between the two studies for all
compounds except quinoline and iso-
quinoline. From the combined data we
conclude that cigarette smoking and gas
heating systems produce the greatest
increases in indoor PAH concentration
levels. Gas cooking appliances have the
least effect. However, the data show
there is still considerable uncertainty m
the true effects of gas heating systems
and gas cooking appliances.
The 1984 and 1987 data for predicting
PAH target compound levels from
measured levels of potential PAH marker
compounds were found to be consistent
for all compounds except quinoline and
isoquinoline. The combined data indicate
that the levels of fluoranthene, pyrene,
and phenanthrene show promise as
predictors of the levels of the other PAH
target compounds. Pyrene appeared to
be the best marker compound for most of
the target compounds investigated.
The following recommendations are
based on the results of this study:
1. Winter and summer monitoring
studies should be conducted to
obtain a better assessment of the true
effects of gas heating systems and
gas cooking appliances. Cigarette
smoking should be eliminated as a
factor in these studies. The effects of
smoking are already apparent.
2. The air samplers should be placed in
only one location within the home
rather than placed in different loca-
tions for each sampling period. By
using only one sampling location
(living room), we will be able to
determine whether it is, indeed,
period of the day that affects the
indoor PAH concentration levels.
3. A separate statistical study should be
done following the winter and
summer studies to compare the data
collected during the winter and
summer monitoring studies and the
1984 and 1987 studies. More data
are needed to further evaluate pre-
vious findings and improve prediction
models.
-------
Gregory A. Mack, Jerry W. Stockrahm, and Jane C. Chuang are with Battelle
Columbus Division, Columbus, OH 43201-2693.
Nancy K. Wilson is the EPA Project Officer, see below.
The complete report, entitled "Statistical Comparison of Results of Two Indoor Air
Pilot Studies," (Order No. PB 89-207 021/AS; Cost: $21.95, subject to change)
will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Atmospheric Research and Exposure Assessment Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300
EPA/600/S3-89/055
0000529
0 S
PS
PROTECTION AGENCY
------- |