United States
                    Environmental Protection
                    Agency
 Environmental Monitoring
 Systems Laboratory
 Research Triangle Park NC 27711
                    Research and Development
EPA/600/S4-86/036 March 1987
SERA         Project Summary
                    Pilot Study  of Sampling  and
                    Analysis for Polynuclear Aromatic
                    Compounds  in  Indoor Air
                    J. C. Chuang, G. A. Mack,
                    J. R. Koetz, and B. A. Petersen
                     The objective of this project was to
                   evaluate and test generic sampling and
                   analysis methodology for characteriza-
                   tion of polynuclear aromatic hydrocar-
                   bon (PAH) concentrations in ambient
                   and indoor air. Sources responsible for
                   variations  in PAH  concentrations  in
                   indoor air were identified from statistical
                   analyses.
                     Preliminary studies of polyurethane
                   foam (PUF) cartridges were conducted
                   to determine whether  PUF cartridges
                   can be used to collect vapor phase
                   PAHs. A modified PUF cleanup proce-
                   dure that combines compression rinsing
                   and Soxhlet extraction  was developed
                   to remove interfering material from the
                   PUF cartridges. It was demonstrated
                   that PAHs can be extracted from the
                   PUF cartridges by Soxhlet extraction,
                   which has previously been used suc-
                   cessfully for polychlorinated biphenyls.
                   Investigation of the retention char-
                   acteristics  of PUF cartridges for PAH
                   showed that, after exposure to about
                   100 m3 of air at -10 C to 0 C, most of
                   the spiked PAHs, except for a few highly
                   volatile and reactive compounds, were
                   retained on the PUF cartridges.
                     Residential air sampling was con-
                   ducted in Columbus, Ohio, during the
                   winter of 1983/84 to determine the
                   effects of the following parameters on
                   PAH concentrations in  air within resi-
                   dences: cigarette smoking,  use  of
                   wood-burning fireplaces, type of home
                   utilities, and ambient air intrusion.  A
                   modified EPA medium volume sampler
                   having a quartz fiber filter to collect
                   paniculate  matter and a PUF cartridge
                   to trap vapors was used.  The deter-
                   mination of PAHs and related com-
pounds was accomplished using on-
column injection fused-silica gas
chromatography with both positive and
negative  chemical ionization/mass
spectrometric detection in the multiple
ion detection mode. The target com-
pounds determined in the air samples
were 15 PAHs, 2 nitro-PAHs, and 2
nitrogen heterocycles. The levels of
PAHs and nitrogen heterocycles de-
tected in outdoor ambient air and indoor
air samples were generally less  than
100 ng/m3 for compounds found at
higher levels and less than 1 ng/m3 for
compounds found at lower levels. Both
nitro-PAHs (2-nitrofluoranthene and 1 -
nitropyrene) were always found at very
low levels (<0.1 ng/m3) with the former
compound always occurring at higher
levels than the latter by a factor of 2 to
10.
  Statistical analyses  of the data in-
volved the generation  of summary
statistics (e.g., means, standard devia-
tions) for the observed concentrations
of target compounds as well as statisti-
cal  modeling  to  assess the separate
influences of each of the four parame-
ters above. The statistical analyses
showed that the type of heat source
and appliances (gas versus electric) and
cigarette smoking had the greatest ef-
fects on the PAH levels found in the air
in residences. The use of woodburning
fireplaces was only weakly correlated
with target compound levels. Outdoor
air intrusion had statistically significant
effects on indoor air PAH levels for only
a few compounds.
  This Project Summary was developed
by EPA's Environmental Monitoring
Systems 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 In-
formation at back).
Introduction
  Polynuclear aromatic hydrocarbons
(PAH) and nitro-substituted polynuclear
aromatic hydrocarbons (NO2-PAH) have
been  identified in organic extracts of
ambient air particulate matter and diesel
exhaust particulate matter. Many of the
identified PAH and N02-PAH are potent
carcinogens and/or mutagens. Therefore,
the determination of PAH and N02-PAH
in ambient and indoor air is of consider-
able importance to the characterization
of air quality.
  The objectives of this project were (a)
to evaluate and test a generic sampling
and analysis methodology to characterize
PAH and N02-PAH  concentrations in
ambient and indoor air, and (b) to identify
sources responsible for variations in the
PAH concentrations in air and estimate
the  quantitative  importance of these
sources.
  A range-finding study  was conducted
to estimate the influence of the following
parameters on the PAH concentrations in
residences:

  • Smoking
  • Woodburning fireplaces
  • Home utilities (gas or electric)
  • Ambient air intrusion.
Eight study homes and two background
homes located in Columbus, Ohio, were
selected for sampling based on the pres-
ence or absence of the above parameters.
Sampling took place during the winter
heating season from  March 12 to 27,
1984. The sample extracts were analyzed
by  positive  chemical  ionization  gas
chromatography/mass spectrometry (PCI
GC/MS) to determine PAH and nitrogen
heterocyclic compounds; and  negative
chemical ionization  (NCI) GC/MS to
determine  N02-PAH.  The target  com-
pounds,  representing three classes of
PAH, were chosen for this study. These
compounds are phenanthrene,  anthra-
cene, fluoranthene, pyrene, cyclopenta
(c,d)pyrene, benzo(a)anthracene,  chry-
sene,  benzofluoranthenes, benzo(e)
pyrene,  benzo(a)pyrene, indeno(1,2,3-
c,d)pyrene, benzo(g,h,i)perylene,  coro-
nene, 1 -nitropyrene, 2-nitrofluroanthene,
quinoline and isoquinoline.
  A statistical analysis was performed on
the analytical data to interpret and to
estimate the influence of the following
parameters on the PAH concentrations in
air within residences:
  • Cigarette smoking
  • Woodburning fireplaces
  • Time of day
  • Location within the residence
  • Home to home variation
  • Day to day variation
  • Nonsystematic variation
  • Gas versus electric utilities
  • Outside air infiltration.
Procedure
  A  modified  EPA  medium  volume
sampler with a rotary carbon vane oilless
vacuum pump was used in the indoor air
sampling; a modified EPA medium volume
sampler with a bypass motor (General
Metal Works, Inc.) was used in the out-
door air sampling.  The  sampler was
equipped with a quartz fiber filter followed
by a polyurethane foam cartridge (PDF).
The sampling head was mounted on  a
tripod to place the inlet 60 inches from
the floor. A heavy wall 1 -inch Tygon tube
led through a window port from the sam-
pling head to the pump, which was located
in a sound-insulated box outside  the
house. Indoor air samples were collected
over three 8-hour periods at different
locations within the house. The sampling
locations and the corresponding sampling
time are: (a) kitchen,  0700-1500 hr; (b)
living room, 1500-2300 hr; and (c) bed-
room, 2300-0700 hr. The sampler flow
rate was approximately 12.5  mVhr,
which collected 100 m3 of air over 8 hr.
During the entire  indoor air sampling, a
single outdoor air sample (24 hr, 300 m3)
was collected  in the  back yard of each
house. Air exchange rates were obtained
by measuring the decay of SF6 injected at
the start of the sampling period, followed
by sampling at one,  three, five, seven,
and eight hours with air-tight  syringes
and subsequent analysis by GC (electron
capture detector).
  The  PUF samples  were Soxhlet ex-
tracted with 800 ml of 5% ether/hexane
and  the filter samples were extracted
with 400  ml of methylene chloride. The
sample extracts were  concentrated by
Kuderna-Danish evaporation and ana-
lyzed by PCI and NCI GC/MS to determine
PAH, nitrogen heterocyclic and NO2-PAH.
Statistical analyses of the analytical data
included the generation of summary sta-
tistics  and statistical modeling to assess
the separate influence of the individual
experimental factors.
Results


  A preliminary study was performed t(
determine how well  PAH are recoverec
from  PUF cartridges, and to determine
how well PAH are retained on the PUF
during eight hours of air sampling (~10C
m3 of air sampled). The results showec
that PAH and their  derivatives can b«
quantitatively recovered from PUF car
tridges  by  Soxhlet extraction with  5%
ether/hexane. Overall,  most PAH anc
their derivatives less  volatile than quino-
line, with the exception of cyclopentc
(c.d)pyrene, were well retained on PUF
cartridges with exposure of 100 m3 of air
It  should be noted  that the ambiem
temperature during this experiment was
-10 C to 0 C, and greater losses of'the
volatile compounds would be expected ai
higher sampling temperature.
  The  masses of particles collected on
the filters ranged from  1.3 to 38.7 mg
indoors and from 3.0 to 22.7 mg outdoors
These particles yielded 0.4 to 28.9 mg
and 0.4 to 8.7 mg of  extractable organic
material indoors and outdoors, respec-
tively. Extractable organic material on the
PUF cartridges ranged from 2 2 to 8.7 mg
indoors and from 1.4 to 10.1 mg outdoors
It  should be  noted  that the  sample
volumes from outdoors are approximately
three times those from indoors.
  Distributions of PAH and their deriva-
tives between filters  and PUF cartridges
were as follows: 3-rmg PAH, 90% on
PUF;  4-ring PAH,  80%  on PUF;  5-ring
PAH, 10% on PUF; and 6- to 7-ring PAH
and 4-ring NO2-PAH, 100% on filters. In
general, these distributions agree with
the  volatilities  of the  individual  PAH
compounds.
  Concentrations of  the individual PAH
measured in homes ranged from 0.00 to
143.58 ng/m3; phenanthrene was most
abundant, followed by fluoranthene and
pyrene  second  and  third, respectively.
The carcinogenic PAH such as benzo(a)-
pyrene and indeno(1,2,3-c,d)pyrene were
present  individually  at concentrations
from 0.00 to 7.35 ng/m3. The concentra-
tions of  the nitrogen heterocyclic com-
pounds, quinoline  and  isoquinoline,
ranged from 0.11  to 63.18 ng/m3.  Be-
cause of the  low recoveries of  these
compounds from PUF, the concentrations
of these compounds reported here are
probably low.  The  NO2-PAH including
1-nitropyrene  and 2-mtrofluoranthene
were detected,  but  the  concentrations
were in the 2 to 185  pg/m3 range which
are much  lower than  their percent
compounds.

-------
  The results of the statistical analyses
showed that cigarette smoking and the
use of gas heat and appliances  in the
home increase  the PAH concentrations
in indoor air. The  use of woodburning
fireplaces did not appear to be as im-
portant  as the other two factors. Com-
parisons of the influences of sampling
locations showed that the  living room
PAH levels were higher than the kitchen
levels, which in turn were higher than
the bedroom levels. The air exchange
rates of homes  were statistically signifi-
cant for only a few target compounds.
Sources of variation not included in the
statistical  model constituted the largest
single source of random variation. Home-
to-home variation was second largest for
most of the compounds, while day-to-day
variation was the smallest.
Conclusions and
Recommendations
  The results of the preliminary study of
the PDF cartridges show that using PUF
as cartridges  as  backup sorbent traps
may not be sufficient to collect highly
volatile PAH and reactive PAH. Additional
studies to resolve  this problem are
required.
  The results of the  statistical analyses
provided information  concerning factors
that should be  addressed in a full scale
PAH study. The following factors were
identified in the analysis:
  •  Cigarette smoking is a  major factor
     affecting PAH levels. Therefore, the
     presence or  absence  of  smoking
     should be included as a  stratification
     variable  in the  full  scale design.
     Sample homes should  include both
     those that have  smokers and those
     that do not.
  •  The PAH  levels are  considerably
     higher in  gas-heated  homes than
     electrically heated ones. Since the
     primary interest  of a full scale study
     is in homes with high  PAH  levels,
     the study should include mostly gas
     homes.  Homes  with electric  heat
     should be included only as controls.
  •  Even though the presence  of a
     woodburning fireplace did not seem
     to be a major factor for most com-
     pounds, it should still be included in
     a full scale study. Fireplace use was
     expected a priori to be an important
     factor, and was found to be signifi-
     cant for several of the compounds.
  •  The PAH concentrations can  vary
     significantly depending on the loca-
     tion within a home. Therefore, the
    sampler location must be carefully
    considered in the full scale study.
  • Home-to-home variations and day-
    to-day variations were  similar in
    magnitude, with home-to-home vari-
    ations somewhat larger. Therefore,
    the sample allocation in a full scale
    study should include more sampling
    of different homes and less repeat
    sampling  to assess accurately the
    magnitude of day-to-day variations.
  Both 2-nitrofluoranthene (2-NF) and 1 -
nitropyrene (1-NP) were found  m the
indoor and outdoor air paniculate sam-
ples. Levels of 2-NF were greater than
levels of 1 -NP in nearly all samples, both
indoor and outdoor. The indoor concen-
trations of 2-NF  were correlated with
smoking, whereas the 1 -NP concentra-
tions were higher in homes with gas heat
and appliances, but the significance of
correlation was  low. Because  of the
greater amount of 2-NF and its mutageni-
city, a future full scale PAH study should
include this compound.
  The compound quinoline may serve as
a marker for the contribution of cigarette
smoke to indoor  PAH levels.  However,
further evaluation is necessary,  with a
sampling  system  that provides better
recoveries for this compound than those
obtained with PUF cartridges.

-------
      J. C. Chuang, S. W. Hannan, andJ. R. Koetzare with Battelle Columbus Division,
        Columbus, OH 43201-2693.
      Nancy K. Wilson is the EPA Project Officer (see below).
      The complete report,  entitled "Pilot Study  of Sampling and Analysis for
        Polynuclear Aromatic Compounds in Indoor Air." (Order No. PB 87-129 S24/
        AS; Cost: $24.95, 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:
             Environmental Monitoring Systems Laboratory
             U.S. Environmental Protection Agency
             Research Triangle Park, NC 27711
umiea aiaies
Environmental Protection
Agency
Information
Cincinnati OH 45268
/FN^
'0 -1
[ APR 1 0'87
"™~" \
i PENALTY
\-OR
-RIV'ATE
/ JSE $300
_* 	 *.
i ivvirAL. sviriu
U.S.POSIAGE
:; 0 ? ? =
Official Business
Penalty for Private Use S300
EPA/600/S4-86/036
f B METER]
6250I09 I
              0000329   PS

              U  S  ENVIR PROTECTION  AGENCY
              REGION  5  LIBRARY
              230  S DEARBORN  STREET
              CHICAGO              IL   60604

-------