United States
                     Environmental Protection
                     Agency
Environmental Monitoring Systems
Laboratory
Research Triangle Park NC 2771
                     Research and Development
EPA/600/S4-85/034 June 19851
S€PA         Project Summary

                    Evaluation  of Passive
                    Sampling  Devices (PSDs)
                     Robert W. Coutant
                      The basic objectives of this study
                    were to evaluate the performance of the
                    EPA passive sampling device (PSD) for
                    sampling of ambient level volatile or-
                    ganic compounds (VOCs); to develop
                    an understanding of the mechanics of
                    passive sampling using reversible ad-
                    sorption; and to apply this understand-
                    ing to development of an improved PSD
                    that is usable for sampling of VOCs over
                    periods of 8 to 24 hours. Laboratory and
                    limited field evaluations of the standard
                    and  modified PSDs  were conducted
                    and a model relating sorbent properties
                    and device design to sampling rates was
                    developed. The results show the stand-
                    ard PSDs to be useful for sampling of
                    VOCs having large retention volumes.
                    Modified PSDs having greatly reduced
                    sampling rates  show  promise for sam-
                    pling compounds having retention vol-
                    umes as low as 5 to 10 L/g over 8- to
                    24-hour sampling periods. The use of
                    Spherocarb as an alternative sorbent to
                    Tenax® GC also was investigated as a
                    means for improving the performance
                    of the PSD. This sorbent was found to
                    be unsuitable because of the high tem-
                    peratures required for desorption. It is
                    recommended  that the model which
                    was developed  be used for developing
                    sampling plans for specific applications,
                    and that more extensive field evaluation
                    of the reduced-rate PSDs be conducted.
                      This report was submitted in fulfill-
                    ment of Contract No. 68-02-3487 by
                    Battelle Columbus Laboratories under
                    the sponsorship of the U.S. Environ-
                    mental Protection Agency. It covers a
                    period from April 15.1982 to October
                    31.1984, and work was completed as
                    of October 31,1984.
                      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 infor-
mation at back).

Introduction
  In recent years, there has been  an
increased awareness of the need for
monitoring individual or personal expo-
sures to pollutants and toxic chemicals.
This awareness has prompted the devel-
opment of a variety of personal sampling
devices including  battery-driven pump
systems, passive systems having  high
specificity for individual compounds, and
generalized passive systems intended for
the collection  of volatile organic com-
pounds. Within this latter category, the
primary commercial emphasis has been
on the use of carbon-based sorbents for
monitoring of the relatively high concen-
trations of contaminants found in indus-
trial workplaces. In two earlier programs
for the Environmental Monitoring Sys-
tems Laboratory,  U.S.  Environmental
Protection  Agency, Research Triangle
Park, North Carolina (RTP),  Battelle's
Columbus Laboratories (BCL) explored
the problems and limitations of using
commercially available passive  devices
for  monitoring ambient  level  organic
chemicals.  The performance  of one  of
these devices under simulated ambient
conditions also was subjected to a de-
tailed evaluation.
  Results of these earlier studies have
shown that  the commerciail^available
carbon-based devices are satisfactory for
ambient monitoring o> selected volatile
organic compounds/under some condi-
tions, but they are by no means complete-
ly general  in their applicability under

-------
realistic ambient conditions. For example,
their  performance  is  impaired  under
conditions of high relative humidity. The
Environmental Monitoring Systems Lab-
oratory therefore has undertaken, under
separate contract  with the Monsanto
Research Corporation (MRC), the devel-
opment of a passive sampler that is not
subject to the same  restrictions as the
commercially available devices. The basic
concept invovled in the development of
this  new  device  has been  to  employ
relatively  hydrophobic  porous polymer
sorbents in order to evolve a system that
is readily subject to thermal desorption
for analysis. While  much of the initial
work with this device has been conducted
using Tenax® GC  as the  sorbent, the
fundamental applications concept is flex-
ible to permit the  use of other porous
polymer sorbents, or even activated car-
bon, as may  be  required  for specific
applications.
  This report addresses the findings of
three Work Assignments (13,21, and 33)
conducted  consecutively  at BCL. The
primary objective was  an  independent
evaluation of the applicability of the MRC
passive sampling device (PSD) with re-
spect to monitoring of volatile organic
compounds (VOCs) under ambient condi-
tions. In the  first Work Assignment,
sampling rates were determined for chlor-
oform, 1,1,1-trichloroethane, carbon tet-
rachloride, trichloroethylene, tetrachloro-
ethylene, benzene, and chlorobenzene. A
general model of passive sampling using
thermally reversible adsorption also was
developed. The second Work Assignment
then was conducted to extend the list of
chemicals  to  include acrylonitrile, 1,1-
dichloroethylene,  trichlorotrifluoroeth-
ane, 1,2-dichloroethane, cr-chlorotoluene,
and hexachlorobutadiene, and to test the
general applicability of the performance
model. In the third Work Assignment, the
general  precepts of the performance
model were applied for the purpose of
modifying  the PSD to enable long-term
(8-24 h) sampling of VOCs. Results of the
first two work assignments and some of
the developmental work performed by
MRC are  summarized in  two  papers
accepted for publication in the January
1985 issue of Analytical Chemistry.


Procedures

Exposure Studies
  The U.S. EPA passive sampling devices
(PSDs) were exposed  in  triplicate to
controlled compositions of the test chem-
icals at ppbv concentration levels in the
Battelle dosimeter test facility. This fac-
ility consists of a 200 L chamber in which
air velocity,  gas composition,  relative
humidity, and temperature can  be con-
trolled. Associated with this chamber is
appropriate monitoring equipment includ-
ing a capillary GC facility having a triple
detector  (in series electron capture and
photoionization with parallel flame ioniza-
tion) arrangement to provide  both high
sensitivity and selectivity for component
analysis. The devices were handled under
a protective atmosphere of zero-nitrogen
at all times  except  during the actual
exposures to the test atmospheres. Test
compositions in the chamber were mon-
itored both by  periodic  direct analysis
with the GC and by a pair of Tenax® GC
traps that were  used to  sample the
chamber gases in an active mode.  The
PSDs and Tenax traps were analyzed by
thermal  desorption  into the GC. Expo-
sures were conducted at different relative
humidities, different concentrations (in
the low ppbv range), different air veloc-
ities, and for various lengths of time (up to
24 hr).


Results
  Inasmuch as sampling rates for most of
the compounds studies had not previously
been determined for the EPA  PSD, this
parameter was used  as  the  basis for
comparison of the device performance. A
generalized model relating sampling rate
to the device geometry  and  chemical
retention volumes was developed. This
model takes into account the  reversible
nature of the adsorption of volatile organ-
ics on Tenax® GC, and permits  the
prediction of the time weighted average
sampling  rate as a  function of time.
Conversely, it has been  demonstrated
that  the  model can be  used  to derive
Tenax retention volumes from  measure-
ments of the apparent sampling rates
with the  MRC device. More importantly,
the model provides guidelines for design
and  application of passive dosimeters
employing reversible adsorption.
  Apparent sampling rates measured for
the 18 test chemicals agreed  well with
those predicted by the model  (generally
within  about  6 percent), except in the
cases of  carbon tetrachloride and tetra-
chloroethylene. With the latter chemicals,
the apparent rates were only about one-
half those expected. Battelle researchers
previously have noted difficulties with
passive sampling for carbon tetrachloride
and  other authors  have noted similar
problems with active  sampling  for this
chemical. No appreciable effect of relative
humidity was found at humidities as high
as 90 percent. The precision of measured
sampling rates was generally of the order
of 20-25  percent,  with  approximately
one-half of this variation being traceable
to differences between individual devices.
The remainder of this variation can be
associated with analytical error.
  Two approaches to modification of the
PSDs were pursued for  the purpose of
improving their  performance  for  long-
term (8 to 24  hour) sampling. These
consisted of (1) replacement of theTenax®
GC with Spherocarb in order to increase
the retention times of  VOCs  and (2)
alteration  of the diffusion barrier to yield
sampling rates reduced by a factor of 30
in comparison with the standard PSDs.
The use of Spherocarb proved unsatis-
factory because of the high temperatures
required for desorption  of most of the
target compounds. Results obtained in
the laboratory tests of the reduced-rate
PSDs showed promise for the use of
these devices in  sampling of chemicals
having retention  volumes as low as 5 to
lOL/g.
  Limited  field studies of the perform-
ances of both the standard and reduced-
rate PSDs yielded overall precisions com-
parable to those  determined in the lab-
oratory tests. These field studies illus-
trated problems associated with the in-
appropriate use of the standard PSDs for
sampling of very volatile chemicals, and
also  indicated potential  problems with
using the  reduced-rate devices  for very
short  time periods. In the latter  case,
uncertainties in blank levels impact se-
verely on the relatively small amounts of
sample  collected by the  reduced-rate
devices when they are used for times that
are too short. In  any case, these results
emphasize the need to use the model as
an aid to design of a sampling plan for a
particular  application.

Conclusions and
Recommendations
  Based on the  results of this work, it
concluded that the EPA PSDs offer some
distinct advantages over other available
designs of passive sampling devices in
application to sampling of ambient level
volatile organic compounds. The ability to
thermally  desorb collected samples pro-
vides more than adequate amounts of
sample for use with conventional GC and
GC/MS analytical procedures, and their
independence from high relative humid-
ities yields more flexible field applicability
than the commercially available devices
using  activated  carbon. Finally, the

-------
Tenax® GC can easily be replaced with
other sorbents for customized sampling
applications. The PSDs should not, how-
ever, be utilized without strict attention to
the mechanics of reversible adsorption
and their implications with the respect to
particular sampling requirements. Specif-
ically:

  1.  The standard  EPA  PSDs  can be
      used for sampling of ambient levels
     of VOCs, but careful attention must
     be paid to the retention volumes of
     target  compounds  and the  appro-
     priate sampling period. In general,
     the standard PSDs are  useful for
     chemicals  having  large retention
     volumes (>100 L/g), but can be
      used only for short sampling periods
     a few hours or less) for compounds
     having small retention volumes.

  2.  Reduced rate PSDs having nominal
     sampling rates of the order of 2.5
     cc/min show promise for applica-
     tions requiring the sampling of
     VOCs over extended periods of 8 to
     24  hours. These reduced rate de-
     vices, however, should not be used
     for  compounds  having  retention
     volumes less than about 5-10 L/g.
     Also, the  current  results show
     potential blank problems with the
     use of the reduced rate devices over
     short sampling periods.
  3.  In any case,  the  time-weighted
     average sampling rate for a partic-
     ular sampling  requirement should
     be estimated using the model pre-
     sented in this report and should be
     used as  a guide in designing the
     sampling plan for a particular appli-
     cation. In general, rates significant-
     ly less than about 70-80 percent of
     the  initial rate (Ro)  may indicate
     potential sampling error.
  4.  The model  of  PSD  performance
     presented in this report has been
     shown to represent correctly the
     effects of retention volume, sam-
     pling time,  and air velocity  on the
     effective time-weighted average
     sampling rates of the EPA PSDs.

  5.  The use of  Spherocarb in the EPA
     PSDs may offer some advantages
     for  sampling  of a  few  selected
     VOCs,  but  general use of this
     sorbent is  not recommended be-
     cause of problems generally assoc-
     iated with the high temperatures
     needed for  desorption of this sor-
     bent. Contamination of the sorbent
     by  unpyrolyzed polymer presents
     some special difficulties with clean-
     up and preparation of this sorbent
     for use in the PSDs.
 6.  The EPA PSDs are not affected by
     high humidities, and therefore are
     not subject to the same limitations
     as the  commercially available de-
     vices based on activated carbon.
 7.  Protective shields developed for the
     EPA PSDs on this program provided
     protection against contamination
     during  handling of the devices in
     the field without  significantly af-
     fecting  their sampling rates.
 8.  Blank contamination of  the  EPA
     PSDs has generally  not been  a
     problem, but a few instances of
     such have been observed. In recog-
     nition of the fact that the PSDs may
     not always be handled by  trained
     laboratory personnel in the field, it
     is recommended that a formalized
     procedure and containment system
     be developed for cleanup and pro-
     tection  of the PSDs.
 9.  It is recommended that further field
     testing  of the reduced rate PSDs be
     considered.
10.  Investigation of the application of
     the EPA PSD to sampling  of pol-
     lutants  other than VOCs (e.g., N02,
     volatile polar organics, etc.) is rec-
     ommended to take advantage of the
     capacity for use of different  sor-
     bents with this device.
  R. W. Coutant is with Battelle's Columbus Laboratories, Columbus, OH 43201.
  James D. Mulik is the EPA Project Officer (see below).
  The complete report, entitled "Evaluation of Passive Sampling Devices (PSD's),"
    (Order No.  PB 85-196 418/AS; Cost: $10.00, 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
                                    i, U.S. GOVERNMENT PRINTING OFFICE: 1985-559416/27082

-------
.4  .
      United States                      Center for Environmental Research
      Environmental Protection             Information
      Agency                          Cincinnati OH 45268
      Official Business
      Penalty for Private Use $300
           OC00329   PS

           U S  ENVIR  PROTECTION  AGENCY
            CHICAGO               IL   
-------