United States
              Environmental Protection
              Agency
              Research and Development
 Atmospheric Research and
 Exposure Assessment Laboratory
 Research Triangle Park, NC 27711
EPA/600/SR-94/188  November 1994
EPA      Project  Summary

              Evaluation  of  HVS3  Sampler  for
              Sampling  Polycyclic Aromatic
              Hydrocarbons and
              Polychlorinated  Biphenyls

              J. C. Chuang, S. M. Gordon, J. W. Roberts, W. Han, and M. G. Ruby
               A polyurethane foam (PUF) filter was
             positioned after the cyclone in the High
             Volume Small Surface Sampler (HVS3)
             to determine the penetration of the cy-
             clone by polycyclic aromatic hydrocar-
             bon (PAH) and polychlorinated biphe-
             nyl  (PCB) adsorbed on  house dust.
             Less than 3% of the PAH and less than
             5% of the PCB were found on the PUF
             filter compared with the cyclone catch.
             Therefore, the HVS3 without a PUF fil-
             ter can quantitatively collect PAH and
             PCB in house dust, and it was used to
             collect house dust samples from seven
             houses.
               The collected house dust, founda-
             tion soil, and entryway soil samples
             were analyzed for PAH, PCB, and lead
             (Pb). The house dust, foundation soil,
             and entryway soil samples were spiked
             with a known amount of perdeuterated
             PAH, then extracted with hexane (C8),
             and analyzed by gas chromatography/
             mass spectrometry (GC/MSJ to deter-
             mine PAH. Quantitative recoveries of
             the spiked perdeuterated PAH were ob-
             tained and ranged from 80 to 110%. A
             similar analytical procedure was used
             to determine PCB, except that  10%
             ether in C6 was used as the extracting
             solvent and the extract was fraction-
             ated by silica gel column chromatogra-
             phy prior to GC/MS analysis. Quantita-
             tive recoveries of the spiked 13C-labeled
             PCB were also obtained (73 to 100%).
             The dust and soil samples were ana-
             lyzed for Pb  by energy dispersive X-
             ray fluorescence (XRF).
               The concentration of PAH was higher
             in the house  dust than  in the founda-
 tion soil. The PAH concentrations in
 the entryway soil samples were higher
 than that in the house dust samples in
 three houses, and similar or lower con-
 centrations were observed in the re-
 maining houses. The PAH concentra-
 tions of entryway soil were higher than
 that in  the foundation soil samples in
 all houses. The sum of the concentra-
 tions of all the target PAH in the house
 dust ranged from 6.1 to 26 ug/g (ppm),
 and that in the foundation soil ranged
 from 1.1 to 3.7 ppm. The sum of the
 concentrations of PAH in the entryway
 soil samples varied from 2.9 to 20 ppm.
 The concentration of the sum  of all the
 target PCB in the house dust and soil
 varied from 260 to  760 ng/g (ppb) and
 58 to 240 ppb, respectively. The con-
 centration of Pb in the house dust
 ranged  from 250 to 2250 ppm and 200
 to 4000 ppm, respectively. For all but
 three houses, higher concentrations of
 Pb were found in the foundation soil
 than in  the house dust. By contrast,
 concentrations of PAH and PCB were
 higher in the house dust than in the
 soil samples.
  This report is submitted in fulfillment
 of Contract Number 68-DO-0007, Work
 Assignment No. 35, Task 2, by Battelle
 under the sponsorship of the U.S. En-
vironmental Protection Agency (EPA).
This report covers the period  between
January 3, 1993,  and August 31, 1993.
Work was completed as of August 31,
1993.
  This Project Summary was developed
by EPA's Atmospheric Research  and
Exposure Assessment Laboratory, Fle-
                                                          Printed on Recycled Paper

-------
search Triangle Park, NC, to announce
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
  House dust is becoming recognized as
an  increasingly  important  source of
nonoccupational exposure to Pb and pes-
ticides. Other data suggest that PAHs and
PCBs may also be persistent  in the envi-
ronment and collect in soil.  Menzie et ai.
estimate that urban soils may contain from
600 to 3000 ppb of PAH, with the higher
values resulting in areas of heavy traffic
or industrialization, compared to 5 to 100
ppb in background forest soils. With some
of the PAH identified as potential carcino-
gens, the PAH content of house dust may
add significantly to the health risks of very
young children who typically  spend a lot
of time crawling on floors and carpets and
may have a daily dust intake of 0.02 to
0.2 g. There is very little data currently
available on PAH  and PCB concentra-
tions in house dust and its possible ori-
gins. One potential source of PAH and
 PCB in  house dust is the foundation soil
surrounding the house.
   The  HVS3  can collect a reproducible
 sample of surface dust from different sur-
 faces, particularly carpeted  surfaces, with
 a  relatively constant removal  efficiency
 over a wide range of dust loadings. De-
 tailed tests have shown that the HVS3 is
 effective in sampling for Pb and pesti-
 cides.
    The  HVS3 collects a  surface dust
 sample in a cyclone with an  approximate
 5-Hm cut. Previous studies  have shown
 that more than 99% of the Pb present in
 the dust on a rug  is collected  in the cy-
 clone. These same studies  showed that
 more than 97% of five pesticides, repre-
 senting a range of vapor pressures,  in
 rug dust were captured in the cyclone.
 Since the PAH and PCB of most interest
 are also semivolatile organic compounds
 that lie within the range of vapor pres-
 sures of pesticides tested, it was reasoned
 that the HVS3 should be a reliable instru-
 ment  for collecting  these  compounds
 adsorbed on house dust.
    In using the HVS3 to collect these com-
 pounds, one concern is that material may
 be lost by leaving the cyclone either  on
 particles too small to be captured, on par-
 ticles that are blown out  of the cyclone
 catch  cup, or by  being stripped off the
  particles by the air stream. In any of these
 cases, capturing the lost  material would
  require both  a particle filter and a vapor
  adsorbent located behind the cyclone.
  A PDF filter was used in  the  original
High Volume Surface Sampler (HVS2) as
a final gas filter. The PUF filter was pre-
ceded by a quartz-fiber filter.  By contrast,
a PUF filter is used alone, without a par-
ticulate matter filter, in the standard EPA
TO-10 air sampling method  for ambient
pesticides. It is preceded by a quartz-fiber
filter in both the TO-4 sampling  method
for PCB and the TO-13 sampling method
for PAH.
   Camann et al. have shown that a PUF
plug without a quartz-fiber prefilter gives
equivalent recovery of pesticides compared
to a PUF plug with a quartz-fiber prefilter.
This suggests that the PUF  plug is suffi-
ciently effective as a particle  filter, that an
additional filter, such  as the quartz-fiber
filter used with the HVS2, is not neces-
sary.  Therefore, the collection efficiency
of the cyclone in  the HVS3  for PAH and
PCB  can be demonstrated  by capturing
PAH and PCB with a PUF plug  installed
behind the cyclone.

Objectives
   The objectives of this study were to 1)
determine whether the HVS3 can quanti-
tatively collect  PAH and PCB adsorbed
onto house dust and 2)  obtain concentra-
tion profiles for PAH and PCB in house
dust and foundation soil samples from nine
 houses in Seattle, WA, in 1993.

 Procedures
   The collection  efficiency of the cyclone
 was tested in two houses.  Naturally oc-
 curring PAH and PCB  compounds were
 measured rather than compounds spiked
 onto house dust in known  quantities. It
 was therefore necessary to  collect all the
 dust and vapors that pass through the
 cyclone to determine the amount that is
 not collected by the cyclone. This  was
 done by inserting a PUF filter in the sam-
 pling train behind the cyclone.
   The  added pressure drop of the  PUF
 filter reduced the capacity  of the fan in-
 stalled in the HVS3. To ensure that the
 required volumetric flow rate would be
 achieved on all  types of carpets, the in-
 stalled fan  was backed up  with a blower
 (Cadillac Model HP33P) placed in-line be-
 hind the exhaust of the installed fan. Tests
 with the HVS3, the Cadillac blower, and a
 laminar flow  element demonstrated  this
 arrangement could provide the desired flow
 rates and permitted the development of
  new pressure drop-flow rate curves.
    This modified HVS3, with the PUF filter
  and Cadillac blower, was used to collect
  dust from rugs in two  houses.  The  tests
  were conducted following  the  American
  Society for  Testing Materials (ASTM)
method  and the manufacturer's instruc-
tions at the recommended airflow rates.
The location of these houses and other
relevant information are provided along
with information about the remaining test
houses.
  Foundation  soil  samples  from these
houses  were collected within 2 ft of the
foundation. Approximately  7 g was col-
lected on each side from the top 2 cm of
soil with a stainless  steel spoon. If, for
some reason,  samples could not be col-
lected from one side of the  house, then
extra samples  were collected on the other
sides of the house.  About  10 g  of soil
from each sample was used  to determine
the moisture content of the sample.
  A sample of entryway soil was collected
from the door mat at the primary entrance
of each home. The entryway soil samples
No. 2 and  15 were collected under a tire-
tread doormat with a brush and dust pan.
Samples No. 7, 9, and 14 were collected
at the side or back door. Sample No. 14
was mostly fiber from the  mat.  All the
other samples were  collected from the
front door mats. These samples were ob-
tained by fuming over the entryway mat
and placing it on a clean piece of alumi-
 num foil. The back of the mat was then
 beaten for several minutes before it was
 removed from the foil. The loose  particles
 on the foil were poured  into  a clean
 prelabeled  jar.  This  procedure  was re-
 peated at least three times for each
 entryway mat to collect as much sample
 as possible. After all of the  entryway dirt
 had been  collected at each  home, the lid
 on the sample jar was sealed with Teflon
 tape,  transported back to  Battelle, and
 stored in a -20°C freezer prior to extrac-
 tion.
   Besides the two houses that were used
 to determine collection  efficiency,  the
 HVS3 without a PUF filter and the backup
 Cadillac blower (i.e., in its standard con-
 figuration) was  used to collect dust from
 rugs in seven more houses. The HVS3
 was operated  according  to the ASTM
 method and  the  manufacturer's instruc-
 tions.  The cyclone  was  operated at  its
 designated air flow rate. All PAH  and PCB
 samples were shipped to Battelle with ice
 by  next  day delivery. The Pb  analyses
 were done by Thomas M. Spittler, Region
 I, EPA.
    An aliquot  (200 mg) of each fine portion
 of house dust and foundation soil samples
 was used for PAH analysis. Each aliquot
 of  sample was  spiked with a known
 amount  of perdeuterated  PAH  and  ex-
 tracted by sonicating with two aliquots of
  10 mL C, for 20 minutes. The extract was
         6

-------
 filtered, concentrated, and analyzed by GC/
 MS for target PAH.
   Another  aliquot  (500  mg) of the fine
 portion of house dust and foundation soil
 samples was used for PCB analysis. The
 sample was spiked with a known amount
 of 13C-labeled  PCB  standards  and ex-
 tracted twice by sonication with 10 mL of
 10% ether  in C6. The extract was filtered,
 concentrated, and fractionated on a silica
 gel column. Three eluting solvents, C6, C6/
 dichloromethane (DCM)  1:1, and metha-
 nol were applied to the silica gel column.
 The  target  (CJDCM)  fraction was then
 analyzed by GC/MS for PCB.
   Aliquots of the dust samples were ana-
 lyzed for Pb by energy  dispersive  XRF.
 Calibration of the XRF instrument for house
 dust was performed by using Pb-free sand
 spiked with Pb. Standard soil samples pre-
 pared  by the EPA's Region I Laboratory
 were  used to calibrate the XRF instru-
 ment for soil samples.

 Results
   The maximum  penetration of the cy-
 clone by any of the PAH is 2.8% (phenan-
 threne) and 1.4%  (chrysene) for any of
 the PAH ranked as probable human car-
 cinogens (B-2) by EPA's Integrated Risk
 Information System. The maximum pen-
 etration by any of the PCB is 4.2%. Based
 on the low percentage losses found in this
 collection efficiency test, a decision was
 made to conduct the remaining validation
 study without the PDF filter in the system.
 In summary, the amount of PAH and PCB
 lost without the PDF filter was  insignifi-
 cant,  in agreement with the results ob-
 tained  earlier with the pesticides (3% loss
 without the PUF filter).
  Among the measured target 2- to 6-ring
 PAR the least abundant PAH found in the
 house  dust and the soil samples was
 cyclopenta(c,d)pyrene. The  low  concen-
 tration  of cyclopenta(c,d)pyrene in house
 dust samples is partly due to the reactivity
 of this compound. We have demonstrated
 in a previous study that cyclopenta(c,d)pyrene
 can oxidize to pyrene dicarboxylic acid anhy-
 dride. The most abundant PAH found in these
 samples were fluoranthene,  pyrene, and
 benzofluoranthenes. The highest PAH con-
 centrations  in house dust samples were
 found in the sample collected from House
 No. 14 (HD14). In this house dust sample,
 the concentration of the well known car-
 cinogen, benzo(a)pyrene (BaP), was 1.7
ppm. The highest PAH concentrations in
foundation soil and entryway soil samples
were all from House No. 13. The  BaP
concentration in this foundation soil and
entryway soil sample was 0.20 ppm and
 1.3 ppm, respectively.  In general, higher
 PAH  concentrations were found  in the
 house dust and  entryway soil  samples
 than in the foundation soil samples. The
 PAH concentrations in the entryway soil
 were higher than that in the house dust
 for samples collected from Houses No.  1,
 2, and 13. The reverse concentration trend
 was observed for Houses No. 3,  7, 10,
 14, and 15. Similar PAH concentrations in
 the entryway soil and house dust were
 observed in House  No. 9. Note that the
 sum of the concentrations of the seven B-
 2 PAH are approximately half of the total
 concentrations of  all 16 target PAH (3- to
 6-ring) in all the house dust and founda-
 tion soil samples.
   The most abundant PCB found in house
 dust samples was  penta-PCB,  and the
 least abundant PCB was the most volatile
 mono-PCB. We did  not detect any nona-
 PCB and deca-PCB in the  house dust
 samples. We would expect that significant
 portions  of the more volatile mono- and
 di-PCB adsorbed onto  house dust will
 evaporate into the air and result in lower
 concentrations in the house dusit. The con-
 centrations of total penta-PCB ranged from
 85 to  620 ppb in house dust samples,
 which  represent 22 to 81%  of the total
 PCB concentrations. Similar PCB concen-
 tration  profiles were observed in founda-
 tion soil  samples collected from all but
 two houses. In these two houses, the high-
 est PCB concentrations were from the to-
 tal octa-PCB. The total penta-PCB con-
 centrations ranged from 16 to 44 ppb in
 the foundation soil samples that represent
 15 to  54% of total  PCB concentrations.
 PCB concentrations were  higher  in the
 house  dust samples than in  the founda-
 tion soil samples.
   The  spiked PAH and PCB were quanti-
 tatively recovered from both  house dust
 and soil samples. The  average PAH re-
 coveries  in house dust and soil samples
 from all nine houses ranged from 93 to 99
 percent and from 91  to 99%, respectively.
 The average recoveries for PCB were from
 81 to  89% in house dust samples and
 from 78 to 86% in soil samples.
  The  dust loading in  the nine houses
 ranged from 1.76 to 36.7 g/m2. More than
 50% of the total dust loading is in the fine
 dust (<150 urn) for all but  one  house
 (HD14). The  fine  dust loading in  these
 houses ranged  from 0.46 to 29.8 g/m2.
The house (HD7) that showed the highest
 loading was occupied by college students
who claimed to clean the house monthly,
 but the house was visibly filthy. The low-
est total  dust  loading was observed in
 House  No. 1  (HD1) and the lowest fine
dust loading was observed in HD14. The
highest loadings for PAH (150 ug/m2), PCB
 (17  |ig/m2), and Pb (1500 jig/m2)  were
 found in HD7, because of the high dust
 levels in this house. By the same token,
 the  lowest loadings for PAH (3.0 ng/m2'
 and for Pb (740 |ig/m2) were found in HD1
 and the lowest loading  for PCB (0.31 jig/
 m2) was from HD14.

 Conclusions and
 Recommendations
   This study has shown that the HVS3
 can  be used without a PUF filter for quan-
 titative  collection  of  PAH  and  PCB
 adsorbed  onto house dust. Less than 3%
 of the PAH and less than 5% of the PCB
 were found to penetrate the cyclone and
 were recovered from the PUF filter. This
 finding agrees with that of a previous study
 in which  it was found that 97% of the
 pesticides in house dust was collected in
 the cyclone.
   Quantitative  recoveries of  spiked
 perdeuterated PAH and 13C-labeled PCB
 were obtained from the house dust and
 soil  samples. The average recoveries of
 spiked PAH in house dust, foundation soil,
 and  entryway soil samples from the nine
 houses sampled ranged  from 93 (perylene-
 d12) to 99% (pyrene-d,0), from 94 (chrysene-
 d,2)  to 99%  (fluorene-d ), and from 91
 (fluorene-d,0) to 96% (perylene-d,12), respec-
 tively. The average  recoveries of spiked
 PCB ranged  from  81 to 89% in house
 dust samples and from 78 to 87% in foun-
 dation soil samples.
   The concentrations of PAH and  PCB in
 the house dust were higher than the lev-
 els in the  foundation soil. The PAH con-
 centrations in the entryway  soil samples
 were higher than that in the house dust
 samples for only three houses, and simi-
 lar or lower concentrations were observed
 for the remaining houses. However, higher
 PAH concentrations  were found  in  the
 entryway soil samples  compared to the
 foundation soil samples for all nine houses.
 There were no  known  current  major in-
 door sources of PAH, such  as smokers,
 wood stoves, or gas cooking  stoves,  in
 any of the houses sampled. The sum of
 the target PAH  concentrations in house
dust samples ranged from 6.1 to 26 ng/g
 (ppm) and in the foundation  soil samples
 ranged from 1.1 to 3.7 ppm. The sum of
all target  PCB  concentrations in  house
dust  samples varied between 260 and 760
ng/g (ppb) and  in  the foundation soil
samples ranged from 58 to 240 ppb. Un-
like PAH and PCB, higher Pb concentra-
tions were found  in  the foundation soil
samples than in the house dust samples
for all but  three  houses. The Pb concen-
trations  in the house dust and the soil
ranged from 250 to 2250 ppm and 200 to

-------
4000  ppm. respectively. There  was  no
correlation among the concentrations of
Pb, PAH, and  PCB in the house dust
samples or in the foundation soil samples.
The house dust and foundation soil con-
centrations for PAH, PCB, and  Pb also
showed no significant correlation with dis-
tance from a freeway.
  To assess and manage exposure from
home soil and house dust, there are sev-
eral important issues that  remain to be
addressed. In future studies, we recom-
mend
(1) Measuring the PAH and PCB in house
    dust, foundation soil, walkway soil, and
   entryway dirt in a sufficiently large,
   representative sample for homes lo-
   cated in older,  large cities in colder
   climates with a history of burning coal.
(2) Conducting a similar study in  a city in
   a  mountain valley with a  history of
   wood burning.
(3) Conducting a similar study in a  city
   with heavy traffic.
(4) Determining if pollutant magnification
   occurs in house dust for PAH and PCB.
(5) Documenting the effect of track-in  and
   dust control techniques on exposure
   to these pollutants in house dust.
References
Menzie, C.  A., B. B.  Potocki,  and  J.
  Santodonato. "Exposure  to Carcino-
  genic PAHs in the Environment," Envir.
  Sci. and Tech., 26:1278-1284,  1992.

Camann, D. E., H. J. Harding, R. G. Lewis.
  "Trapping of Particle-Associated Pesti-
  cides in Indoor Air  by Polyurethane
  Foam and Exploration of Soil  Track-In
  as a Pesticide  Source." In: Proceed-
  ings of  the 5th  Conference on Indoor
  Air  Quality and Climate,  2:621-626,
  Canada Mortgage and  Housing Corp.,
  Ottawa, Ontario, 1990.
  J. C. ChuangandS. M. Gordon are with Battelle, Columbus, OH 43201, andJ. W.
    Roberts, W. Han, andM. G. Ruby are with Engineering Plus, Inc., Seattle, WA
    98112.
  Robert G. Lewis is the EPA Project Officer (see below).
  The complete report, entitled "Evaluation ofHVSS Sampler for Sampling Polycyclic
    Aromatic Hydrocarbons and Polychlorinated Biphenyls,"  (Order No. PB95-
    123931; Cost: $19.50, 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
                                                       BULK RATE
                                                 POSTAGE & FEES PAID
                                                           EPA
                                                     PERMIT No. G-35
  EPA/600/SR-94/188

-------