£EPA
United States
Environmental Protection
Agency
Engineering Issue
Passive Samplers for Investigations of Air Quality:
Method Description, Implementation, and
Comparison to Alternative Sampling Methods
TABLE OF CONTENTS
1 PURPOSE AND SUMMARY
2 INTRODUCTION
3 PASSIVE SAMPLER BASICS
3.1
3.2
3.3
3.4
3.5
3.6
Theory
Passive Sampler Types
Sorbent Types
Uptake Rates
Sampling Duration
Passive Sampler Geometry and Sorbent
Combinations
3.7 Comparison of Passive Sampling to
Conventional Air Sampling Methods
4 DESIGNING AND IMPLEMENTING A PASSIVE
SAMPLING PROGRAM
4.1 Selecting a Passive Sampler Suited to
Your Investigation
4.2 Placing Passive Samplers Indoors
4.3 Placing Passive Samplers Outdoors
4.4 Instructions for Occupants for Passive
Indoor Air Sampling Events
4.5 Considerations for Other Applications
5 DATA QUALITY OBJECTIVES
5.1 Media Preparation for Field Deployment
5.2 Passive Sampler Deployment—Field
Handling Protocols
5.3 Field Quality Control Samples
5.4 Intermethod Duplicates
6 INTERPRETATION OF PASSIVE SAMPLING
RESULTS
6.1 Measurement Uncertainty and Implications
to Data User
6.2 Other Lines of Evidence
7 CURRENT CHALLENGES, LIMITATIONS, AND
RESEARCH AND DEVELOPMENT NEEDS
1
2
3
3
4
5
6
7
12
12
13
13
13
14
14
15
15
16
16
17
17
18
18
19
19
19
19
20
7.1 Intermethod Comparisons
7.2 Longer-Term Sample Durations
7.3 Additional Compounds
7.4 Challenging Compounds
7.5 Application to Soil Gas
7.6 Sample Duration for Different Exposure
Periods 20
7.7 Triggering Methods for Intermittent Passive
Sampling 20
8 ACRONYMNS AND ABBREVIATIONS 20
9 ACKNOWLEDGMENTS 21
10 REFERENCES 22
The U.S. Environmental Protection Agency (EPA)
Engineering Issue Papers (EIPs) are a series of
technology transfer documents that summarize the
latest available information on selected treatment and
site remediation technologies and related issues. EIPs
are designed to help remedial project managers, on-
scene coordinators, contractors, and other site managers
understand the type of data and site characteristics
needed to evaluate a technology for potential
applicability to their specific sites. Each EIP is
developed in conjunction with a small group of
scientists inside EPA and with outside consultants and
relies on peer-reviewed literature, EPA reports, Web
sources, current ongoing research, and other pertinent
information. As such, this EIP is a technical support
document describing the current state of knowledge on
passive sampler application and performance and does
not represent EPA policy or guidance.
1 PURPOSE AND SUM MARY
The purpose of this EIP is to summarize the "state of
the science" regarding the use of passive air samplers
for investigating subsurface vapor intrusion (VI) to
indoor air. This Paper covers the basics of passive
sampler design, compares passive samplers to
conventional methods of air sampling, and discusses
considerations when implementing a passive sampling
program. The Paper also discusses field sampling and
sample analysis considerations to ensure data quality is
adequate and interpretations based on the passive
sample data are supportable. The reader is expected to
have a basic technical background on the VI exposure
pathway and how to use and interpret indoor air
sampling data in the context of a VI investigation. For
guidance and policy on VI assessment and technical
support documents, please visit:
-------
http://www.epa.gov/oswer/vaporintrusion/basic.html
and
http://www.epa.gov/superfund/sites/npl/Vapor Intru
sion FAQs Feb2012.pdf.
Passive samplers are capable of measuring a large suite
of volatile organic compounds (VOCs) and semi-volatile
organic compounds (SVOCs) in indoor and outdoor air.
Deployment over longer time frames (e.g., weeks or
months) than is possible with conventional canister-
based methods of air sampling.
One key to successful passive sampling is appropriate
calibration of the uptake rate or sampling rate. Uptake
rate is usually determined for a particular sampler and
chemical of interest in experimental chambers with
controls on variables such as the sample duration,
temperature, humidity, velocity, and concentration.
As with any sampling method, one must also control
sources of negative (underestimation) and positive
(overestimation) bias. Potential sources of bias for
passive samplers include starvation, poor retention,
poor recovery, and blank contamination:
• Starvation occurs if the passive sampler
withdraws target compounds from the
surrounding media faster than they are
replenished (essentially scrubbing the VOCs
from the vicinity of the sampler), causing a
negative bias. Starvation can be minimized
either by using low uptake rate samplers (tube
sampler) or by maintaining an adequate face
velocity (i.e., air movement past the sampling
surface).
• Poor retention occurs if a relatively weak
sorbent is used to sample highly volatile
compounds, especially for long sample
durations. This causes a negative bias,
underestimating concentration as VOCs are lost
from the sampler media. Poor retention can be
managed by selecting stronger sorbents.
• Poor recovery can occur when stronger
sorbents are used with strongly sorbed
compounds. Poor recovery underestimates
Passive samplers can yield reliable, time-averaged
sample concentrations with comparable accuracy and
precision to established conventional methods (e.g.,
evacuated canister samples and analysis by U.S.
Environmental Protection Agency [EPA] Method
TO-15 or pumped sorbent tube samples and analysis by
EPA Method TO-17). Passive samplers have the
advantage of greater ease of deployment and
concentrations because the sorbed chemical is
not completely desorbed from the sorbent
during sample analysis. Sorbent selection must
balance retention and recovery.
• Blank contamination can overestimate
concentration and occurs when the sorbents
used in passive sampling are inadvertently
contaminated during sampler preparation,
storage, shipping, or handling. Blank
contamination can be minimized with
appropriate protocols and verified (or detected)
with analysis of trip blanks.
This paper provides information that will be useful in
avoiding such biases and getting the best passive
sampler results possible. As with any VOC sampling
event it is advisable to consult your analytical
laboratory when embarking on a passive air sampling
effort. An experienced analytical chemist is a great
source of information on the most recent passive
sampling analytes, uptake rates, sorbents, sampling
protocols, and quality assurance procedures necessary
to ensure that your passive sampling effort produces
high-quality data.
2 INTRODUCTION
Passive sampling techniques have been used to
monitor VOC and SVOC concentrations for several
decades (e.g., Palmes and Gunnison, 1973). Most of
the early applications were for industrial hygiene
monitoring, but over the past several years there has
been significant progress in using passive sorbent
tsive Samplers for Investigations of Air Quality
-------
samplers for measuring VOCs1 at the lower
concentrations needed to assess human health risk at
sites regulated under the Resource Conservation and
Recovery Act (RCRA); the Comprehensive
Environmental Response, Compensation, and
Liability Act (CERCLA); the Brownfields program;
and various state environmental programs. Passive
samplers have been shown to yield results equivalent
to other established methods for many VOCs
(Begerow et al., 1999; Oury et al., 2006; Mosley et al.,
2008; Cocheo et al., 2009; Lutes et al., 2010; U.S.
Navy, 2013).
Passive samplers have several advantages over more
traditional indoor air sampling techniques (e.g., U.S.
EPA Methods TO-15 and TO-17).
• The sampling protocols are simple, which
reduces the risk of inter-operator error, the cost
of sampling, and the level of training needed for
sampling personnel.
• Passive samplers are small and lightweight
compared with the canisters used in Method
TO-15, so they are discrete to place and less
expensive to ship.
• Passive samplers operate without risk of power
loss, clogging, or leaks that may affect canister
(Method TO-15) or pumped tube (Method
TO-17) samples.
• Passive samplers provide accurate results for a
large range of sampling durations, from daily to
quarterly sampling periods for certain
compounds (U.S. EPA, 2012).
The ability to collect time-weighted average samples
over longer durations than the 8 to 72 hours offered
by the more traditional methods is advantageous
because these longer periods can provide data that are
more representative of the long-term average
exposure concentrations needed for cancer risk
assessment.
A wide variety of organic and inorganic gases can be
monitored using passive samplers, but this paper
focuses primarily on the VOCs most commonly of
concern at VI sites (i.e., chlorinated solvents and
aromatic hydrocarbons).
3 PASSIVE SAMPLER BASICS
A passive sampler is a device that contains a solid
sorbent (usually granular) in an inert container with
openings of known dimensions that allow VOC
vapors to pass through at a fairly constant (and
known) rate. The following sections describe the
basics of passive sampler theory, summarize the
available types of samplers and sorbents, and explain
how uptake rate and sampling duration are
determined. Section 3.6 describes how to consider
these factors when selecting passive sampler/sorbent
combinations and includes a list of commercially
available passive samplers. Section 3.7 compares
passive samplers to other commonly used air
sampling methods (i.e., Summa canisters and active
sorbent tubes). Appendix A provides a more detailed
summary of available research on passive samplers.
3.1 Theory
Passive samplers are deployed for a designated
sampling period and then collected and analyzed by
extracting the VOCs from the sorbent to measure the
total mass of each analyte trapped by the sampler
during the sampling period. If the uptake rate of the
VOC is known, the average concentration over the
sampling period can be calculated as follows:
C = M/(UR x t)
3.1
1 Passive samplers can be used for volatile chemicals including
VOCs, SVOCs, and mercury. Reference to "VOCs" in this
document includes SVOCs and mercury unless otherwise
indicated, although most passive sampling applications are for
VOCs. As explained in Section 3, the applicability of specific
Passive Samplers for Investigations of Air QuaL
passive samplers to specific compounds or substances largely
depends on the availability of measured uptake rates for those
substances with a particular sampler and sorbent.
-------
where
C = time-weighted average air concentration
(^g/m3)
M = mass of VOC retained by passive
sampler (pg)
UR = uptake rate (mL/min, compound-
specific); also called "sampling rate"
T = sampling duration (min)
Mass and sampling duration can be measured with
high levels of accuracy; therefore, the uptake rate, or
sampling rate, is the most critical variable for
accurately determining air concentrations using
passive samplers. Uptake rate has units of
volume/time, but it is not a flow rate; rather, it is
equivalent to the flow rate that would be required for
a pumped adsorptive sample to sorb the same mass
of a target chemical over the same sample duration
when exposed to the same chemical concentration.
The uptake rate is dependent on the geometry of the
sampling device and the diffusion coefficient of the
chemical. The uptake rate is also dependent on the
sorbent characteristics of the passive sampler, which
determine VOC retention as well as desorption
efficiency. Experimentally derived uptake rates for a
range of compounds have been published for most
passive sampler configurations. The selection of a
sampler type for an investigation often is based in
part on the availability and reliability of uptake rates.
Section 3.4 describes several approaches that can be
used to determine uptake rates.
3.2 Passive Sampler Types
Passive samplers can be configured with a porous
diffusive barrier or a nonporous membrane to control
the rate of VOC collection by the adsorbent media.
The porous variety, which allows for movement of air
through the sampler, is the most common design, and
a wide selection is commercially available. The
options of various styles and associated range of
uptake rates allow for flexibility in selecting the
optimal sampler configuration to meet project-
required reporting limits and/or sample durations. In
the case of the nonporous membrane type, referred to
as the permeation passive sampler, VOCs dissolve in
and diffuse through a membrane and are collected on
the adsorbent medium. Although permeation sampler
selection is currently limited, the hydrophobic and
nonporous nature of the membrane can be
advantageous in very high-humidity and high-velocity
environments.
Passive samplers with the porous diffusive barrier can
be categorized as tube, badge, or radial style. Tube-
style (Figure 1) and badge-style (Figure 2) samplers
have a one-dimensional (a.k.a., "axial") diffusion path
to the sorbent bed. The tube style is characterized by
a relatively small cross-sectional surface area (A) and
long diffusive path length (L) compared with the
badge style and therefore has lower uptake rates.
Examples of commercially available tube-style
samplers are the Drager ORSA 5 monitor and the
standard thermal desorption tubes available from
various vendors coupled with a diffusive end cap.
The badge-style passive sampler is characterized by a
larger cross-sectional surface area and shorter
diffusive path length, resulting in relatively higher
uptake rates than tube samplers. The badge face must
be covered with a windscreen to minimize uptake by
advection. This windscreen can be a perforated cover
(e.g., SKC Ultra and 575 and the 521 Organic Vapor
Badge from Assay Technologies) or porous plastic
(e.g., 3MOVM 3500).
L
A J H (j$$&
t
Diffusion cap
1 1
Adsorbent End cap
Figure 1. Tube-style passive sampler
>sive Samplers for Investigations of Air Quality
-------
Diffusive barrier
Adsorbent
Figure 2. Badge-style passive sampler
The radial-style sampler houses the adsorbent
material in a screen mesh tube, and the sorbent tube
is housed in a porous cylinder that serves as the
windscreen and diffusive barrier (Figure 3). Diffusive
sampling occurs parallel to the sampler's radius,
providing a very large surface area and a short
diffusive path. As a result, radial-style samplers
generally exhibit higher uptake rates than tube- or
badge-style samplers. The radial-style sampler is
patented by Fondazione Salvatore Maugeri as the
Radiello sampler.
Cross-sectional view
Adsorbent
Diffusive barrier
Figure 3. Radial-style sampler
The nonporous permeation passive samplers can be
categorized as either the tube or badge style
(Zabiegala and Namiesnik, 2007). The Waterloo
Membrane Sampler (WMS) is a commercially
available tube-style permeation passive sampler
(Figure 4) with uptake rates generally falling between
the range of the nonporous tube and badge-style
diffusive samplers. At the time of this publication, a
commercially available version of the badge type
permeation sampler described by Zabiegala and
Namiesnik (2007) has not been identified.
Adsorbent
Nonporous
Membrane
Figure 4. Permeation (tube-style) sampler
3.3 Sorbent Types
Each sampler type must be paired with an appropriate
sorbent material for the constituents of interest.
There are many different VOCs of interest and many
different sorbents, and each compound has a certain
affinity for each sorbent. The goal is to select a
sorbent that efficiently retains the VOC mass that
enters the sampler but also releases the VOC mass
efficiently during analysis. A negative bias attributable
to poor retention may occur if a sorbent-sorbate
interaction is weak, leading to back-diffusion (or
reverse diffusion). Weakly sorbed VOCs can also be
displaced through competitive adsorption in the
presence of high concentrations of other VOCs or
water vapor. These effects are magnified when the
sample duration is long. Conversely, a negative bias
attributable to poor recovery may occur for
compounds that are very strongly adsorbed by strong
sorbents and are not completely released during
Passive Samplers for Investigations of Air Qua I
-------
thermal desorption or solvent extraction prior to
analysis.
Published studies document the retention of various
VOCs by different sorbents (e.g., Brown and Shirey,
2001; American Society of Testing and Materials
[ASTM], 2009b; U.S. EPA, 1999b). These data were
generated for active (pumped) sampling, but the safe
sampling volume and breakthrough volume
information can provide useful information for
selecting the most appropriate sorbent for passive
sampling applications. In some environmental
settings, many different chemicals with a wide range
of sorptive properties may be present. Therefore,
compromises between retention and recovery for
some of the measured compounds may be required,
along with the advice of an experienced analytical
chemist, to select the appropriate sorbent for use in a
passive sampler. Otherwise, two or more samplers
could be used with sorbents selected for specific
compounds of interest.
Sorbents fall into two general categories—very strong
sorbents requiring solvent extraction and relatively
weaker sorbents amenable to thermal desorption.
Activated charcoal is the sorbent routinely used for
solvent extraction techniques. The large surface area
of activated charcoal allows for efficient adsorption of
many VOCs and limited losses due to back-diffusion
when applied to extended monitoring applications.
The large surface area of charcoal is also beneficial
when sampling environments with high total VOC
concentrations. Carbon disulfide, the typical
extraction solvent in this category, efficiently desorbs
a majority of the adsorbed VOCs.
Thermally desorbable sorbents are characterized by
smaller surface areas to allow for quantitative VOC
recovery at moderate desorption temperatures. A
variety of sorbents with varying adsorptive properties
exist within this category. The three types of thermally
desorbable sorbents commonly used for passive
sorbent samplers and examples of each are
summarized in Table 1. More detailed listings of
sorbents, their physical properties, and selection
considerations are available from sorbent vendors.
Table 1. Thermal Desorbable Sorbents Commonly Used for Passive
Sorbent Samplers
Thermal Desorption Sorbent
Type
Porous Organic Polymers
Graphitized Carbon Blacks
Carbon Molecular Sieves
Example Sorbents
Tenax TA,
Chromosorb 106
Carbopack B,
Carbograph 1TD, Anasorb GCB1 ,
Carbograph 4TD,
Carbopack X, Carbograph 5TD
Carboxen1016
The thermal desorption preparation technique
generally results in analytical sensitivity much greater
than analytical sensitivity using solvent extraction
because only a small fraction of the solvent extract is
injected onto the instrumentation for analysis,
whereas the thermal desorption technique allows for
nearly complete transfer of the adsorbed sample onto
the analytical equipment. However, the advantage of
the improved analytical sensitivity can be offset by the
shorter sample collection periods required to avoid
sorbent saturation and poor retention when using
thermally desorbable sorbents. As sampling periods
are extended, the weaker adsorption characteristics of
these sorbents can result in poor analyte retention and
back-diffusion. Additionally, the lower capacity of
these thermally desorbable sorbents means that care
must be taken when deploying samplers in
environments with potentially high VOC
concentrations.
3.4 Uptake Rates
Most passive samplers have a list of published uptake
rates for a given number of compounds (see
Section 3.7 for typical rates), but this list may not
include all of the target compounds needed for a
given assessment. Additionally, the rates may not be
validated for longer-term environmental monitoring
applications. Environmental conditions and sorbent-
sorbate interactions are the two primary factors
affecting the accuracy and the reproducibility of the
uptake rate. As a result, nominal uptake rates are
typically determined under a range of environmental
>sive Samplers for Investigations of Air Quality
-------
conditions, concentrations, and expected sample
exposure periods. Standard methods detailing uptake
rate determination can be found in National Institute
for Occupational Safety and Health (NIOSH)
(Cassinelli et al., 1987), Occupational Safety and
Health Administration (OSHA) (2003, 2008),
International Standards Organization (ISO, 2000,
2003), and European Committee for Standardization
(CEN, 2002a, 2002b, 2004) methods. See Appendix A
for further details.
While uptake rates determined experimentally in the
laboratory are generally considered to be the most
reliable, it is not always practical to conduct these
studies, particularly when validating uptake rates over
extended sampling periods. In the absence of
experimentally determined rates, uptake rates can be
estimated based on known or calculated diffusion
coefficients. Assuming an ideal diffusive sampler, the
uptake rate can be expressed by the following
relationship:
UR = A/L x D
3.2
where
with a percentage of passive samplers can be used to
determine uptake rates based on field conditions (see
Section 5.4).
3.5 Sampling Duration
The minimum sampling duration required to provide
adequate sensitivity for assessing human health risk
can be calculated using a rearranged form of 3.1:
t=(Mmm)/URxG
'RBSL
3.3
where
t = duration of sampling (min)
Mmin = laboratory reporting limit for each
individual VOC in the target list
UR = the compound-specific uptake rate
(mL/min)
= risk-based screening levels (ug/m3)
C
RBSL
UR = uptake rate (mL/min)
A = passive sampler surface area (cm2)
L = diffusive path length—distance between
diffusive barrier and sorbent bed (cm)
D = diffusion coefficient (cm2/min)
A discussion of several estimation methods using
diffusion coefficients can be found in Namiesnik et al.
(1984) and Feigley and Lee (1988). In cases in which
VOCs exhibit non-ideal sorbent interaction, the use
of calculated uptake rates may vary significantly from
experimental rates (Walgraeve et al., 2011), and
concentrations generated with calculated rates should
be noted by the laboratory. In the case of
polydimethylsiloxane-based permeation passive
samplers, uptake rates can be estimated from the
analyte's linear temperature programmed retention
time index (LTPRI) (Seethapathy and Gorecki,
2010a). Uptake rates can also be determined in the
field using intermethod duplicate samples. Methods
such as EPA TO-15 or TO-17 collected concurrently
Passive Samplers for Investigations of Air Qua It
A longer sample duration may be advisable to provide
a reporting limit somewhat lower than the risk-based
screening level.
3.6 Passive Sampler Geometry and
Sorbent Combinations
A summary of the sampler geometries and sorbent
combinations is presented in Figure 5. Each sampler
geometry can be paired with either a charcoal sorbent
requiring solvent extraction or a thermally desorbable
sorbent. For a given sampler geometry, the charcoal-
type sorbent is most suited for longer sampling
durations, which helps offset the lower analytical
sensitivity of the solvent extraction preparation
technique. Conversely, selecting a high uptake rate
sampler geometry coupled with a thermally
desorbable sorbent may be most appropriate for a
short sampling period.
When selecting the preferred sampler and sorbent
combination for a particular set of target analytes and
target concentrations, consider the following:
-------
Whether the target analytes all have calibrated
uptake rates for the compounds of interest. If
calibrated uptake rates are not available, the rate
can be estimated as described in Section 3.4.
However, estimated rates may not be
appropriate for the measurement of key risk
drivers at the site without additional verification
samples (Section 5.4).
The sample duration required to provide
reporting limits that meet the target levels given
a sampler's uptake rate (see Equation 3.3.
Section 3.5). If the calculated duration is longer
than the desired sampling period, selecting a
passive sampler geometry with a higher uptake
rate and/or a TD sorbent to provide a lower
analytical reporting limit may be preferred.
Alternatively, the calculated duration may result
in weakly adsorbed compounds exceeding the
estimated safe sampling volumes (Section 4.3),
and a stronger sorbent may be required.
The suitability of available sorbents for the
compounds of interest. In cases where one
sorbent is not suitable for all of the compounds
of potential interest, it may be preferable to
select the sorbent best suited to the
compound(s) considered likely to pose the most
significant contribution to risk (present in
highest concentrations relative to risk-based
target levels). Alternatively, different sorbents
can be deployed at the same time if all of the
compounds of interest cannot be captured
using a single sorbent.
Table 2 provides a list of passive samplers that were
available on the commercial market when this
document was prepared. The table provides the
sampler name, manufacturer, a brief description (with
the range of uptake rates), references with published
uptake rates and other data, and links to
manufacturers' and other web pages where a great
deal of additional detail can be found. This
information should be helpful to the practitioner who
is selecting and applying passive samplers as described
in the following section but should be reviewed
against current information to ensure that it is up to
date in terms of the available samplers and published
uptake rates and other information.
>sive Samplers for Investigations of Air Quality
-------
r-
S
"
3
01
rt-
0
•"
a
8
I
I
m
/N
X & \
7
3
ID
-1
V)
Ql
3
"^
A
n
o'
CO
o
-1
\/
Sorbent
Solvent extracted
(charcoal type)
Thermally
desorbable
(various)
Sampler Geometries
Tube (Axial)
Drager ORSA 5
WMS™-charcoal
(permeation)
Thermal desorption
tubes (various vendors)
WMS™— thermal
desorption
(permeation)
Badge (Axial)
SKC 575 and Ultra
3M OVM 3500
521 Assay
Technologies
SKC Ultra Series
Radial
Radiello 130
Radiello 145
l\
\
Lower Uptake Rate Higher /
"I /
Figure 5. Selection matrix of passive sorbent and sampler types
3.7 Comparison of Passive Sampling to
Conventional Air Sampling Methods
Table 3 compares passive sampling to conventional
air sampling methods—EPA TO-15 (U.S. EPA,
1999a) and TO-17 (U.S. EPA, 1999b)—in terms of
field ease of use, analytical sensitivity, target
compounds, typical uptake rates, and sample
duration. Each method has advantages and limitations
when assessing indoor air quality to support VI
investigations.
EPA Method TO-15 involves collecting air samples in
specially prepared canisters (e.g., Summa canisters)
and analyzing aliquots of the samples by gas
chromatography/mass spectrometry (GC/MS). EPA
Method TO-17 involves collecting VOCs present in
air by actively pumping air through multibed sorbents
encased in inert tubes, thermally desorbing the VOCs
in the laboratory, and analyzing the samples by
GC/MS. Appendix B describes other U.S. methods
and standards that may be less familiar to American
VI practitioners, as well as international and
European standard methods for passive sampling that
are applicable to indoor air applications for VOCs at
low concentrations.
3.7.1 Ease of Use
Passive samplers generally have fairly simple
sampling protocols, and because they do not require
power, the sampling process is less subject to battery
failure or power failure (as in TO-17) or to the flow
controller clogging with dust (as in TO-15). Passive
samplers also are smaller, so they are less expensive to
ship and can be placed more discretely during the
sampling event.
3.7.2 Analytical Sensitivity
Each method can be configured to achieve similar
reporting limits. In the case of the active and passive
sorbent methods, the reporting limits are a function
of sampling parameters. Selection of the proper
sample volume pumped through the TO-17 sorbent
tube and the appropriate deployment time for passive
samplers provides comparable reporting limits to
TO-15.
Passive Samplers for Investigations of Air Qua I
-------
Table 2. Commercially Available Passive Sampling Devices as of June 2014
Name (manufacturer)
Automatic Thermal
Desorber (AID) Tubes
(Perkin Elmer, Markes,
CAMSCO, Supelco)
OVM 3500
(3M)
Ultra, Ultra II, Ultra III
(SKC)
GABIE (SKC)
575 (SKC)
ORSA (Draeger)
Waterloo Membrane
Sampler (WMS; SiREM
Labs)
Radiello (Fondazione
Salvatore Maugeri)
Description (typical
uptake rates)
Open sorbent-filled tube
with dust cap over one
end during deployment
(0.3-0.6 mL/min)
Badge with film of
activated carbon, film of
porous plastic, and air in
between (20-40
mL/min)
Badge with three
different sorbent
configurations (indoor:
7-11 mL/min; outdoor:
12-1 5 mL/min)
Badge
Badge with screen
(9-1 8 mL/min)
Dual-ended tube
(4-8 mL/min)
PDMS (polydimethyl-
siloxane) membrane
sampler (0.80-1 9
mL/min)
Radial porous plastic
diffusive barrier with
cylinder sorbent insert
(40-80 mL/min [solvent
extraction diffusive body
120];
20-30 mL/min [thermal
desorption
configuration, diffusive
body 120-2])
References
Brown (1 999), MDHS 80
(1995), ISO 1601 7-2
(2003), ASTMD61 96-3
(2009b)
13016200-2(2000),
Type B sampler; MDHS
88(1997)
Cassinellietal. (1987),
Guild etal. (1992)
Ouryetal. (2006),
Langlois (2008)
13016200-2(2000),
Type B sampler; MDHS
88(1997)
May(1989);Begerowet
al. (1999); ISO 16200-2
(2000), Type A sampler;
MDHS 88 (1997)
Seethapathy and
G6recki(2010a,b)
Cocheo et al. (2009);
13016200-2(2000),
Type D sampler; MDHS
88(1997)
Weblinks
http://www.perkinelmer.com/CMSResources/lmaaes/44-
74181 APR lndustrialHvaieneMonitorinabvTD.pdf: http://www.markes.com/:
httD://www.camsco.com/: http://www.siamaaldrich.com/analvtical-
chromatographv/air-monitoring.html
http://www.shop3m.com/70070032142.html
httD://multimedia.3m.com/mws/mediawebserver?mwsld=SSSSSu7zK1fslxtU
MX U4xmSev7qe17zHvTSevTSeSSSSSS-
http://www.skcinc.com/prod/575-001.asp
http://www.skcinc.com/prod/690-1 01 .asp
http://www.skcsouthafrica.co.za/index.php?paae=shop. product detoils&flvpa
ae=&product id=172&cateaorv id=47&option=com virtuemart<emid=60
http://www.skcinc.com/reports.asp
http://www.afcintl.com/industries-
served/industrial.aspx?txtSearch=*&catpaaeindex=8&CreatedBvUser=1&Pro
ductlD=172
http://www.siremlab.com/products/waterloo-membrane-sampler
http://www.radiello.com/english/index en. html
Summa canister sampling using TO-15 (U.S. EPA,
1999a) is considered by many to be a "standard" for
VI studies. Interlaboratory studies of TO-15 suggest
variance in the reported values, ranging from about
30% to 300% (e.g., Lutes et al., 2012; Daugherty et al.,
2004; Pearson, 2005). Studies comparing passive
sampler results to TO-15 results show good
agreement (30% to 90% variance) for chlorinated
solvents and volatile aromatics, but poor agreement
for polar compounds (e.g., Sweitzer et al., 2006;
Odencrantz et al., 2008; Lutes et al., 2010; Allen et al.,
2007; and Mukerjee, 2004).
10
3.7.3 Target Compounds
Passive and conventional air sampling methods differ
in the range of chemicals that can be collected and
analyzed. The conventional EPA Method TO-15 is
limited primarily to compounds that have sufficiently
high vapor pressure to minimize condensation on the
canister surface. This translates to an upper carbon
range limit of approximately C12 or naphthalene.
Using a multibed sorbent tube, the conventional
method TO-17 can extend the compound range well
beyond C12 and can be configured to collect VOCs
>sive Samplers for Investigations of Air Quality
-------
Table 3. Comparison of Passive Sampler Method to Conventional
Ambient Air Methods
Method
Passive
single
sorbent
samplers
TO- 15
(canister
samplers)
TO- 17
(active
multibed
sorbent
samplers)
Field
Ease of
Use
Excellent
Good
Fair
Analytical
Sensitivity
Excellent3
Excellent
Excel lentb
Target
Compounds
Depends on
sorbent
selected
C3-C12
Depends on
sorbents
selected
(-C3-C26)
Sample
Duration
Hours to
weeks
Up to 72
hours
Up to 24
hours
a Depends on sample deployment time, uptake rates, and total VOC levels.
b Depends on sample volume collected and total VOC levels.
in the approximate carbon range of C3 to C26,
providing the greatest level of compositional
information for an air sample. In contrast, passive
sorbent samplers are limited to a single sorbent bed,
thereby limiting the targeted volatility range to the
sorbent type selected.
3.7.4 Sample Duration
Passive samplers can be used for longer sampling
periods than EPA Methods TO-15 and TO-17. The
sample period for a TO-15 6L Summa canister is
largely limited by the minimum flow rate that can be
set by the flow controller. Flow controllers are
available to deliver sampling rates to collect samples
over periods up to 72 hours, but beyond that, the
flow rates can be unreliable. Most canister samples are
collected over 8 to 24 hours. For TO-17, the sample
duration is limited by the minimum pump flow rates
and the safe sampling volumes of the sorbent tube,
based on the published uptake rates for a given
number of compounds. Minimum pumped flow rates
are constrained not only by mechanical considerations
but also by the need to avoid back-diffusion. These
limitations translate to a maximum sample period of 8
to 24 hours for TO-17. By contrast, passive samplers
can be deployed over periods spanning days to weeks,
especially with strong sorbents and/or low uptake
rates to avoid exceeding the retention capacity.
It is important to consider the interaction between the
selected duration of sampling and the time
characteristics of the sources of VOCs detected in the
indoor environment (ISO, 2007). Long-term sampling
provides a robust measure of mean indoor air
concentrations, but long-term sampling leads to a loss
of information with respect to the variation with time
of the VOC concentrations, which may make it
harder to discern the impact of indoor sources. For
example:
• Certain building products such as polyvinyl
chloride, linoleum, cork, parquets, and wooden
furniture containing glues emit VOCs slowly
and for a long duration with minimal short-term
changes.
• Paints and adhesives are characterized by a
continuous, irregular, decaying time profile.
• Intermittent sources such as cooking or
smoking often have a spikey and periodic
pattern over time.
• Cleaning, maintenance, and hobby products
have a spikey, irregular, and variable time
pattern over time.
• VI varies in response to changing building
pressure and ventilation.
Longer sample durations are more comparable to the
exposure durations that are of interest for human
health risk assessment. Detailed studies of indoor air
quality in buildings where VI occurs (e.g., Johnson et
al., 2013; Luo et al., 2012; U.S. EPA, 2012) show that
temporal variability can be more than an order of
magnitude. Passive samplers allow longer sample
durations that provide time-weighted average
concentrations that include concentration peaks and
valleys and therefore can be more representative of a
long-term average concentration than a shorter
duration sample. Thus, even if conventional methods
Passive Samplers for Investigations ofAirQualt
11
-------
(e.g., TO-15, TO-17) may have a slightly higher
accuracy, passive samplers can provide more
representative data for long-term indoor air exposure
assessment.
4 DESIGNING AND IMPLEMENTING A
PASSIVE SAMPLING PROGRAM
This section describes how to design and implement a
passive sampling program that will provide reliable
results. Selecting and deploying passive samplers
involve technical decisions on sampler design and
placement. It is important to establish a good
relationship with your analytical chemist early in the
process to select the correct sampler and method
properly.
4.1 Selecting a Passive Sampler Suited to
Your Investigation
4.1.1 Target Chemicals—What are your compounds
of interest?
Passive sampling requires selecting the appropriate
sampler, sorbent, and sampling duration to meet a
preliminary data quality objective (DQO) and project-
specific goals. The first consideration is to determine
the compounds of interest. For most VI
investigations, a list of compounds most likely to
contribute to inhalation risks can usually be developed
by comparing soil gas or groundwater sample
concentrations to risk-based screening levels (RBSLs)
available from EPA or state regulatory programs (e.g.,
EPA Regional Screening Levels2). The compounds
that exceed the RBSL by the greatest margin will
dominate the risk at a site and should be the primary
focus of sampler selection.
4.1.2 Sampler Selection—How many of the target
compounds have known uptake rates?
It is advantageous to select a passive sampler that has
vendor-supplied uptake rates supported by controlled
chamber tests or at least a considerable body of field-
calibrated uptake rates for as many of the target
compounds as possible. (Table 2 has references for
uptake rates available from passive sampler suppliers.)
For compounds without published uptake rates, an
estimate can be made (Section 3.4) and/or verified by
field calibration (side-by-side sampling using both
passive and active methods in a certain percentage of
locations). If diffusion through air is the uptake
mechanism (as it is for indoor and outdoor sampling),
the diffusion coefficient depends primarily on the
molecular weight
(http://www.epa.gov/athens/learn2model/part-
two/onsite/estdiffusion.html). If permeation through
a membrane is the uptake mechanism, the uptake rate
may be proportional to the partitioning and
permeation constants (Seethapathy and Gorecki,
2010a, 201 Ob).
4.1.3 Reporting Limits and Sampling Duration—
What target reporting limits are needed and
how does that influence the duration of
passive sampling?
Once the target compounds and uptake rates are
known, the sample duration needed to provide a
reporting limit as low or lower than the risk-based
target concentration or screening level can be
calculated for each compound using Equation 3.2.
The sample duration will usually be dictated by the
chemical with the lowest target concentration or
lowest uptake rate. At this stage, two checks should
be applied:
1. The sample duration should be compared
with project goals and practical constraints.
For example, if the sample duration is
calculated to be relatively short (e.g., less
than 24 hours), it may be appropriate to
extend the duration to achieve a more
representative time-weighted average
concentration for each analyte. Alternatively,
if the sample duration is relatively long (e.g.,
many weeks or months), it may be preferable
to use a higher uptake rate sampler to
achieve a shorter sample duration. A shorter
sample duration can also potentially be
2 http://www.epa.gov/region9/supeffund/pfg/: contact your
state regulatory office for state-specific RBSLs
12
>sive Samplers for Investigations of Air Quality
-------
achieved using thermal desorption instead of
solvent extraction because the thermal
desorption process results in better
sensitivity.
2. The product of the sample duration and the
sampling rate (UR x t) for each compound
should be compared with the recommended
safe sample volume (SSV) for each chemical
on the sorbent typically used with the passive
sampler. If (UR x t) is greater than the SSV,
it may be prudent to select a stronger sorbent
instead. Before a final selection is made, the
target compound list should be reviewed to
assess whether the most strongly adsorbed
compounds will be recovered well during
thermal desorption or solvent extraction
prior to analysis. This step is similar to the
design considerations forTO-17 sampling
and should be performed with the assistance
of an experienced analytical chemist.
Overlapping deployment periods (e.g., Johnson et al.,
2013; Johnston, 2013) can be used as a precaution
against sorbent media overloading or underloading.
U.S. EPA (2012) has information on the performance
of passive samplers over different sampling durations
for tetrachloroethylene (PCE), toluene, chloroform,
and benzene.
4.2 Placing Passive Samplers Indoors
Three technical factors should be considered in
deploying passive samplers indoors:
• Air velocity: The rate of air flow past passive
sampler devices can result in high bias via
turbulent uptake for most samplers, so sampler
devices should not be placed near areas prone
to high air flow such as areas near windows,
doors, chimneys, and air vents. Areas with
insufficient air circulation, which can result in
low bias via starvation, also should be avoided
to provide a representative atmosphere to the
sampler.
• Humidity: Humidity can affect the uptake rate
of some passive samplers, so high humidity
areas such as laundry rooms and bathrooms
should be avoided.
• Security: The samplers should be placed to
minimize the risk of accidental or unauthorized
physical intervention (e.g., from children and
pets).
4.3 Placing Passive Samplers Outdoors
Six technical factors should be considered when
deploying passive samplers outdoors:
• Air velocity: Windy locations as well as very
protected locations should be avoided for the
reasons described above. A shelter may be used
to avoid excessive wind-speeds, precipitation,
and direct sunlight.
• Precipitation: Moisture can affect the uptake
rate or retention of some passive samplers, so
the samplers should be protected from rain and
snow.
• Temperature: The samplers should be protected
from temperature extremes (e.g., avoid direct
sunlight) that can speed degradation of sorbed
chemicals or release of sorbed chemicals.
• Security: The samplers should be placed to
minimize the risk of accidental or unauthorized
physical intervention (e.g., from children and
pets or from vandalism).
• Plant transpiration: Avoid placing samplers
under or in trees that can evapotranspire VOCs
from the subsurface and create a high bias for
ambient samples (Johnson et al., 2003).
• Location relative to building: Avoid placing
outdoor samplers near windows, doors, or
exhaust fans to minimize the influence of
indoor air concentrations on the outdoor
samplers. For commercial buildings, it can be
instructive to position outdoor air samplers near
the building air intake(s) to assess the outdoor
air quality as it enters the building.
4.4 Instructions for Occupants for Passive
Indoor Air Sampling Events
Controlling the use of or removing consumer
products containing VOCs is common practice when
Passive Samplers for Investigations ofAirQuah
13
-------
sampling indoor air over the typical 8- to 24-hour
sampling periods. It is harder and often impracticable
to control or eliminate the use of consumer products
containing VOCs when using passive sampling
methods for longer sampling durations (days to
weeks). Similarly, it should be expected that building
occupants will continue with normal ventilation,
heating, and other occupancy activities (U.S. EPA,
1990). The Massachusetts Department of
Environmental Protection provides detailed
discussions of the pros and cons of opening and
closing windows and doors, operating mechanical
ventilation systems, and isolating certain parts of the
building during indoor air sampling (MADEP, 2002).
It is important to document the presence of
household chemicals or other potential sources of
VOCs to assist with data interpretation (ISO, 2007).
Certain consumer products that contain VOCs (e.g.,
paints, glues, aerosols) are often used very
infrequently and can be stored in plastic bins outside
the residence for an extended sampling duration.
Additionally, it is helpful if the occupant keeps a log
of the types and dates of operation of heating
systems, air conditioners, and ventilation devices like
window and attic fans. This information will be useful
for interpreting the results (CEN, 2004), particularly
when evaluating sequential sampling events in a
particular building or concurrent sampling events in
adjacent areas.
4.5 Considerations for Other Applications
Several past studies have demonstrated novel ways
that passive samplers can be used in a VI
investigation. Some examples include the following:
• Passive samplers can be used in a quantitative
or semiquantitative manner to measure VOC
vapors in subslab and deeper soil gas. During
subslab or soil deployments, passive samplers
should be protected from direct contact with
soil, and the sampler should be sealed in place
with a seal that is at a depth just above the
sampler, not just at ground surface (McAlary et
al., 2014a,b,c; Hodny et al., 2009; Odencrantz
and O'Neill, 2009).
• Zencak et al. (2007) and Johnson and Dawson
(1999) used passive sampling to collect
atmospheric samples suitable for stable isotopic
analysis.
• Researchers have used overlapping deployment
periods as a precaution against media
overloading or underloading, as a method to
determine the stability of uptake rates and
evaluate back-diffusion over time (U.S. EPA,
2012) and as a method to facilitate temporal
analysis of meteorological factors in VI
(Johnston and Gibson, 2013).
Other studies have combined long-term passive
sampling with triggered sampling. Yao (2007)
combined long-term passive sampling with event-
triggered active sampling for meteorological
conditions when atmospheric pollutant transport was
expected. Crump et al. (2011) tested a strategy in
which a photoionization detector monitoring device
was used to trigger a sorbent-based VOC sample in
an aircraft cabin.
5 DATA QUALITY OBJECTIVES
As with any environmental measurement, DQOs for
passive samplers are an important consideration. To
meet the DQOs, the investigation should include
both field and laboratory quality assurance/quality
control (QA/QC) procedures to ensure that sampling
and analysis procedures do not bias the sample results
(i.e., results do not consistently under- or over-
estimate actual concentrations), results are
reproducible and comparable (i.e., precision—results
can be replicated by following the same procedures),
and results are sufficiently accurate to support
environmental decisions to be made with the data.
Passive samplers typically have similar DQOs as other
indoor air measurement techniques (such as TO-15
and TO-17). This section provides an overview of the
important DQOs and QA/QC procedures for indoor
air sampling with a focus on what is different for
14
>sive Samplers for Investigations of Air Quality
-------
passive samplers. Laboratory QA/QC for passive
sampler analysis is discussed in Appendix C.
5.1 Media Preparation for Field
Deployment
Passive samplers require similar preparation as active
samplers used in Method TO-17. Reusable sorbent
cartridges require cleaning and certification to prevent
chemical carryover from previous sampling events.
Samplers can be "batch-certified" (5 to 10% of a
batch is analyzed without being deployed) or
individually certified (every sampler is analyzed before
and after use). Many of the single-use samplers are
pre-certified by the manufacturer without the need
for additional laboratory or field preparation steps.
After manufacture or cleaning/certification, the
storage time before use is usually limited to a few
months, and shelf-life studies are typically conducted
to establish expiration or "use-by" dates.
Recertification and recleaning should be considered
for any samplers used after their expiration date.
Samplers should be stored and transported to and
from the field protected from heat and in well-sealed,
inert containers to protect from the ingress of VOCs.
The recommended storage condition for various
charcoal and thermal desorption-type axial and radial
samples is generally room temperature (EN 13528-3;
CEN, 2003), but polar compounds (e.g., 2-butanone)
have been reported to exhibit poor stability on
charcoal, so transport on ice is recommended if
quantitative results are required (ISO, 2000; 3M,
1996).
5.2 Passive Sampler Deployment—Field
Handling Protocols
Passive sampling is relatively simple: the sampler is
removed from protective packaging, positioned in the
sampling location, and left for a specified interval,
after which it is returned to the protective packaging.
The date and time of deployment and retrieval must
be recorded, along with the sample identification
number and location. Some samplers such as the
WMS, 3M OVM badge, and SKC 575 badge are
preassembled, and deployment simply requires
removing the outer packaging. Other samplers, such
as Radiello, require transferring sorbent resin or
cartridge to the diffusive body. The SKC Ultra III is
available in both the prepacked or user-filled version.
The Automatic Thermal Desorber (ATD) tube
requires removing the end cap and replacing it with a
diffusion cap. All of these procedures are easily
learned, so training is minimal. In fact, building
owners and occupants can be trained to perform
sample deployment and retrieval (e.g., Jaward et al.,
2004a,b; Johnston, 2013; Johnston and Gibson,
2013). A chain of custody should be completed to
document the transfer of samples from the field to
the laboratory.
5.2.1 Recording Field Conditions
Monitoring of ancillary parameters such as
temperature, relative humidity, and barometric
pressure can improve the accuracy of the reported
concentrations and assist in the interpretation of
passive sampler data. Temperature has weak or
negligible impact on membrane samplers (Zabiegala
and Namiesnik, 2007; Seethapathy and Gorecki,
201 Ob) but can affect the uptake rates of diffusion-
controlled samplers. The uptake rates of diffusion
samplers can be corrected using the relationship of
mass adsorbed to temperature derived using
Maxwell's equation and the ideal gas law as described
in ISO 16017-2 (ISO, 2003). Assuming an ideal
diffusive sampler, the mass adsorbed varies with the
square root of the absolute temperature, translating
into approximately a 0.2% increase per degree Celsius.
Indoor air temperature is usually controlled in
occupied buildings to a relatively narrow range for
comfort, however, so the potential bias in most cases
is negligible.
If the reported concentrations are required to be
expressed as standard ambient temperature and
pressure, both the average barometric pressure and
temperatures are required. These adjustments may not
be necessary or significant in buildings where
temperature, humidity, and air flow are well
Passive Samplers for Investigations ofAirQualt
15
-------
controlled, but ancillary measurements may have
other beneficial uses. For example, U.S. EPA (2012)
highlighted the usefulness of differential temperature
as a predictor of the stack effect, which has an effect
on the differential pressures between the building and
the subsurface. It is particularly useful to record the
temperature, relative humidity, and air velocity when
sampling outdoor air, because of the wider range of
values for these parameters.
A sketch of the site should be prepared, noting the
location of key building features and the locations of
the samplers.
5.2.2 Storage and Transport Requirements to
Laboratory
General requirements for passive sampler storage and
return transport to the laboratory are outlined in
standard EN 13528-3 and include packing in inert,
closed shipping containers and avoiding exposure to
high temperatures. Cold storage is not necessary in
most cases, but some methods recommend or require
it. For example, EPA TO-17 requires transporting
ATD sample tubes back to the laboratory at <4°C,
but ISO 16017-2, EN 13528-2, and ASTM D6196-03
methods describe shipping of thermal desorption
samples to the laboratory at ambient temperatures.
Also, as previously mentioned, transport on ice is
recommended for charcoal-based samplers and polar
compounds.
5.3 Field Quality Control Samples
Field QC samples are used to evaluate potential
sources of measurement error during sample
collection and handling.
5.3.1 Trip Blanks
Trip blanks are a critical requirement for all
adsorptive sampling programs and are used to
quantify any potential blank contamination arising
from sample transport, handling, and storage. These
are samplers that travel to and from the field sampling
site without being opened. An adequate number of
field blanks is usually one per shipment but may vary
16
with the size of the sampling program (e.g., 1 per 10
or 20 investigative samples).
5.3.2 Field Duplicates
Field duplicates are routine QC samples to assess
overall precision by evaluating effects of field
conditions on precision. In the case of passive
samplers, care should be taken to place the duplicate
samplers with sufficient distance (air space) between
them to ensure that they do not influence each other's
uptake rates.
5.4 Intermethod Duplicates
A certain percentage of collocated duplicate samples
using alternative methods (e.g., TO-15 or TO-17) can
assist in evaluating passive sampler measurement
accuracy. Passive sampler uptake rates can vary in
response to changes in field conditions (temperature,
humidity, pressure, wind speed, and sample duration).
Intermethod duplicates collected using methods that
are less sensitive to environmental parameters can be
valuable in identifying and evaluating the significance
of any bias inherent in the passive measurement
(Tolnai et al., 2000). Additionally, these concurrent
measurements can be used to calculate site-specific
uptake rates, which is especially useful for situations
in which either the uptake rate is undetermined or
when field conditions are outside the ranges for
which the uptake rates were validated. The
reproducibility of passive samplers is quite good, so
the number of collocated intermethod duplicate
samples need not be excessive to provide a high
degree of confidence in the accuracy of all passive
samplers collected in similar conditions.
Methods TO-15 and TO-17 samples are not designed
for long sample durations that can be achieved with
passive samplers. To extend the sample collection
period to align with the passive samplers, either
multiple successive sorbent tubes or canisters are
required or modifications to the conventional
sampling protocols must be made.
One approach to extend pumped sorbent tubes or
canister sampling times is to collect intermittent
fsiVe Samplers for Investigations of Air Quality
-------
samples over the desired collection period.
Programmable sample pumps have been used to cycle
the pumps on and off at scheduled times, allowing for
pump operation at reliable flow rates without
exceeding safe sampling volumes for the sorbent tube
(Gordon et al, 1999; Jia et al, 2007). To conduct
intermittent canister sampling, a programmable
solenoid/timer device such as the Nutech 2701 can
be attached to the canister flow controller, turning the
Summa canisters on and off for short periods over
the sampling event (U.S. EPA, 2009a, 2009b).
Canister sampling periods can be also be extended by
attaching a set of 6L canisters to a manifold
connected to a single flow controller. This approach
increases the effective collection volume of the
canister, thereby extending the sampling duration
while operating the mass flow controller in the range
of reliable flow rates. Although the laboratory only
needs to analyze one of the set of canisters, this
approach requires preparation and delivery of a set of
6L cans per selected collocated site (U.S. Navy, 2013;
ESTCP, in press; Johnston and Gibson, 2013;
Johnston, 2013). Some studies have deployed
canisters at the beginning, middle, and end of the
sampling period. This approach is most appropriate
under controlled conditions where temporal
variability is expected to be moderate (Lutes et al.,
2010).
Another approach to verifying performance of
passive samplers is to deploy concurrent passive
samplers at submtervals (U.S. EPA, 2012). This may
require using an alternative passive sampler
geometry/sorbent combination to achieve the
targeted sensitivity. Comparing the average of the
subinterval concentration to the measured average
provides information about the stability of the uptake
rate for the long-term passive sampler. This can be
important if the sampler has not been validated for
prolonged exposures or if the selected sorbent is not
ideal for one or more of the target compounds.
6 INTERPRETATION OF PASSIVE
SAMPLING RESULTS
6.1 Measurement Uncertainty and
Implications to Data User
All results reported by a laboratory have the potential
for bias (high or low) and variability, both of which
contribute to measurement uncertainty, regardless of
the analytical method. Any measurement affected by
bias and variability may still be fit for its intended
purpose as long as the uncertainty is considered in the
interpretation. For example, if a particular compound
has a risk-based screening level of 1 ug/m3 and the
measured concentration is 0.1 ug/m3, then the
measured concentration could be interpreted as lower
than the screening level even if the uncertainty in the
measurement was up to an order of magnitude. The
impact of temporal variability decreases as the sample
duration increases (Steck, 2013), so a passive sampler
deployed over longer durations is likely to exhibit
lower overall measurement uncertainty associated
with temporal variability than conventional methods
deployed over shorter durations.
Intermethod sample results should be reviewed to
assess whether any environmental parameters are
skewing the passive sampler results. As described
above, passive sampler uptake rates vary with wind
speed, humidity, temperature, and other
environmental factors, so when these parameters are
unusual or highly variable, the accuracy of passive
samplers may be influenced (Tolnai et al., 2000).
Where sampling conditions are typical of indoor air
(temperature near 21°C, moderate humidity, and
gentle air circulation) and manufacturers have
published uptake rates for a particular compound,
sampler/sorbent, and sample duration, the accuracy
of the passive sampler results is probably comparable
to other indoor air quality monitoring protocols.
Trip blanks and certification blanks should be
reviewed to assess whether any chemicals are present
in the passive samplers from sources other than the
media being monitored. A blank correction may be
worthwhile, depending on the DQO. Retention and
Passive Samplers for Investigations ofAirQuah
17
-------
recovery should be considered for the weakly and
strongly sorbed compounds, respectively. Several
factors may contribute to the overall uncertainty, but
the dominating factor can often be used to estimate
the total measurement uncertainty. Accuracy and
precision can be tested and documented with an
appropriate number of QA/QC samples, and it also is
valuable to include the professional judgment of an
analytical chemist experienced with passive sampler
use and analysis.
6.2 Other Lines of Evidence
In designing and conducting air quality studies,
collecting multiple lines of evidence simultaneously,
rather than sequentially, offers benefits. Multiple
simultaneous lines of evidence can provide greater
power to evaluate interpretative questions such as
"Were the samples collected under near worst case
conditions for VI?" or "Why did the concentration
vary between sampling period A and sampling period
B?" Matching the durations of sampling or
observations of multiple variables facilitates time
series data analysis. In this section, we briefly describe
other lines of evidence that can be collected over the
same sampling durations as passive samplers to assist
with interpreting the passive sampling data.
6.2.1 Radon
Radon has been suggested as a potential gauge of the
spatial and temporal variability of VI (Schuver and
Mosley, 2009; U.S. EPA, 2012; Schuver, 2013). Many
well-tested and cost-effective passive radon sampling
devices are available with durations available that can
be matched to passive samplers for VOCs. These
include Alphatrack detectors for 90- to 365-day
deployments, various types of electret ion chambers
for 2- to 365-day periods, and activated charcoal
adsorption canisters for 2 to 7 days. Continuous
radon monitors are also available. See U.S. EPA
(1992, 2012) for details on these passive and
continuous instrumental measurements.
6.2.2 Passive Air Exchange Rate Measurements
Passive air exchange rate measurements can be made
with a technique that combines a small passive
emitter of perfluorocarbon tracers with a passive
sampler (EPA Method IP-4A; U.S. EPA, 1989). By
varying the number of emitters, experiments can be
designed for a wide variety of durations from 1 day to
several months. Air exchange data can assist with the
interpretation of temporal variability in indoor air
measurements. Indoor air concentrations are strongly
influenced by air exchange rate, exhibiting higher
concentrations with lower air exchange rates (all other
factors being equal).
6.2.3 Differential Pressure Measurements
Differential pressure across the floor slab can be
useful to determine the driving force behind VI,
especially if it is supported by wind speed and
barometric pressure monitoring (U.S. EPA, 2012).
Instructions for differential pressure monitoring in
structures can be found in U.S. EPA (1993), and
procedures for calibrating differential pressure
measurement devices are in U.S. EPA (1998).
Although this technique is still normally only applied
in research contexts to unmitigated structures (e.g.,
U.S. EPA, 2012), it is a routine technique in VI
mitigation design and monitoring.
7 CURRENT CHALLENGES, LIMITATIONS,
AND RESEARCH AND DEVELOPMENT
NEEDS
Although at the time of the publication of this
document there was substantial research available on
passive samplers for VOCs in air focused on
industrial hygiene, ambient outdoor air quality, and
compliant building investigations (see Appendix A).
However, several current challenges and limitations to
using passive samplers remain. Some of the more
important challenges for applying passive samplers in
VI investigations are discussed below. Readers
encountering such challenges are encouraged to
identify and consult more recent studies for
advancements addressing these limitations.
18
>sive Samplers for Investigations of Air Quality
-------
7.1 Intermethod Comparisons
A compilation of the results of intermethod duplicate
sampling studies would be helpful for identifying
field-validated uptake rates for a wider range of
chemicals. If measurements of the environmental
conditions (e.g., temperature, pressure, humidity,
wind speed) also were compiled, a statistical analysis
could be conducted to determine whether and how
much these effects bias the results and whether these
biases are significant compared with spatial and
temporal variability. It should also be recognized that
the primary alternative, TO-15, is only suitable for a
portion of the over 100 compounds of potential
concern for VI and shows interlaboratory variability
that may be significant compared with the
intermethod differences (see Section 3.7).
7.2 Longer-Term Sample Durations
Questions remain regarding the limits of applicability
of passive samplers to longer-term sampling
durations. For example, sampling over very long
durations (e.g., 3 months, 6 months, 1 year) may be
desirable to provide time-weighted average samples
inclusive of the short-duration, infrequent intervals of
enhanced VI that can significantly contribute to long-
term average exposure. However, less strongly sorbed
compounds may not be well retained on passive
samplers over such long intervals. In addition, longer
sampling durations provide more time for the uptake
of water to influence performance through
competition for adsorptive sites for some sorbents
(e.g., activated carbon), hydrolytic degradation of
some chemicals (e.g., methyl ethyl ketone, 1,1,1-
trichloroethane), or reactions with ground-level
ozone.
Long-term controlled chamber tests and field studies
would be beneficial to verify sampler performance
over time and could be designed to address these and
other long-term sampling concerns. In addition,
demonstration and validation studies may be needed
as novel sampling strategies are developed for using
passive samplers in VI investigations. For example,
one could represent long-term exposure in
commercial, industrial, or other occupational settings
by collecting samples over multiple workdays and
opening samplers during business hours and closing
them over nights and weekends.
7.3 Additional Compounds
Most of the studies of passive sampler performance
include a specific list of target analytes or provide data
for compounds present at a particular location. As a
result, some of the more common compounds of
interest for VI (e.g., chlorinated ethanes, ethenes, and
methanes and aromatic compounds such as benzene,
toluene, ethylbenzene, and xylenes [BTEX], trimethyl
benzenes fTMBs], and naphthalene) have been
studied much more extensively than others.
Controlled chamber tests for multiple samplers and a
broad list of chemicals of concern for VI would be
very valuable. Field tests of the passive samplers
provide a useful supplement to chamber tests in that
they challenge the samplers with realistically varying
concentration and temperature profiles that are
difficult to produce in chamber tests.
7.4 Challenging Compounds
Some compounds are particularly challenging to
measure using passive samplers, and additional
research is needed to demonstrate the capabilities and
limitations of various passive sampler/sorbent
combinations in measuring such compounds. For
example, vinyl chloride, chloromethane, and possibly
other low boiling point, low molecular weight
compounds tend to be weakly sorbed and poorly
retained, especially with long sample durations, high
uptake rate samplers, or thermally desorbable
sorbents. Vinyl chloride is retained well by Unicarb™
(Wolfenden, 2010), but other VOCs (e.g.,
naphthalene and SVOCs) may be too strongly sorbed
to allow adequate recovery. Thus, in some cases, it
may be necessary to use parallel samplers with strong
and weaker sorbents to cover the range of target
analytes.
Compounds with relatively low risk-based screening
levels (e.g., 1,3-butadiene, chloroform, 1,2-
Passive Samplers for Investigations ofAirQuah
19
-------
dichloroethane [1,2-DCA], hexachlorobenzene,
hexachlorobutadiene, all of the polycyclic aromatic
hydrocarbons [PAHs] and polychlorinated biphenyls,
1,1,2,2-tetrachloroethane [1,1,2,2-PCA], 1,1,2-
trichloroethane [1,1,2-TCA], and vinyl chloride) may
require long sample durations to achieve reporting
limits comparable to the screening levels. In some
cases, that could result in oversaturation of the
sorbent with compounds that may be more abundant
(e.g., limonene, pinene, and other fragrances;
hydrocarbons; aerosols; and other chemicals from
background sources). Controlled chamber tests
including a variety of sorbents and a range of
challenging compounds would help assess the
capabilities and limitations of using longer-term
passive sampler deployments to meet the reporting
limits needed for VI assessments.
7.5 Application to Soil Gas
Passive sampling has been used to assess soil gas
quality for decades, but until recently, the relationship
between the mass sorbed on the sampler and the
concentration in the soil gas has been poorly
understood. Recent advances indicated that
quantitative passive sampling is possible as long as the
uptake rate of the sampler is the rate-limiting step
(i.e., the rate of diffusive delivery of vapors from the
surrounding soil or fill materials is not the rate-
limiting step) (McAlary et al., 2014a, 2014b, 2014c).
The rate of diffusive delivery from the soil depends
on the porosity and moisture content of the soil.
More field testing is needed to better define the range
of soil types and moisture contents that are conducive
to quantitative passive soil vapor concentration
measurements.
7.6 Sample Duration for Different Exposure
Periods
For some compounds of interest for VI, short- or
medium-term average exposure concentrations may
be important for assessing acute or subchronic health
effects, while long-term average concentrations are
important for assessing the risk of chronic health
effects such as cancer. If the chronic and subchronic
20
screening levels are similar, it would be cost-effective
to have a set of sampling tools that could combine a
long-term average concentration estimate with a
measurement of short- or medium-term peak
concentrations. Additional research is needed on how
to best employ passive samplers in conjunction with
traditional (i.e., TO-15, TO-17) methods and specific
field techniques (e.g., electromechanical timers, sensor
controlled sampling) in a sampling system to achieve
such goals.
7.7 Triggering Methods for Intermittent
Passive Sampling
Passive samplers generally have a long shelf life if
stored in the protective containers provided by the
manufacturers, which creates an opportunity to
monitor during selected intervals of suspected high
potential for VI based on meteorological events (e.g.,
rapid decreases in barometric pressure allow soil gas
to expand and can cause a short-term increase in the
volumetric flow of soil gas into overlying buildings).
Research is needed to assess whether this strategy is
effective and, if so, what special design considerations
are necessary (e.g., high uptake rate samplers may be
needed to achieve low enough reporting limits within
a limited sampling period).
8 ACRONYMNS AND ABBREVIATIONS
ANSI American National Standards Institute
Inc.
ASTM American Society for Testing and
Materials
ATD Automated Thermal Desorber
BTEX Benzene, Toluene, Ethylbenzene, and
Xylenes
CEN European Committee for
Standardization
CERCLA Comprehensive Environmental
Response, Compensation, and
Liability Act
DQO Data Quality Objective
EIP Engineering Issue Paper
EPA Environmental Protection Agency
sive Samplers for Investigations of Air Quality
-------
ESTCP Environmental Security Technology
Certification Program
FID Flame lonization Detector
GC Gas Chromatography
ISEA International Safety Equipment
Association
ISO International Standards Organization
LCS Laboratory Control Spike
LTPRI Linear Temperature Programmed
Retention Time Index
MADEP Massachusetts Department of
Environmental Protection
MDHS Methods for the Determination of
Hazardous Substances
MS Mass Spectrometry/Spectrometer
NIOSH National Institute for Occupational
Safety and Health
ORD Office of Research and Development
OSHA Occupational Safety and Health
Administration
PAH Polycyclic Aromatic Hydrocarbon
PCE Tetrachloroethene
PDMS Polydimethylsiloxane Membrane
Sampler
QA Quality Assurance
QC Quality Control
RBSL Risk-Based Screening Level
RCRA Resource Conservation and Recovery
Act
RSD Relative Standard Deviation
SD Standard Deviation
SOP Standard Operating Procedure
SSV Safe Sample Volume
SVOC Semi-volatile Organic Compound
TCE Trichloroethene
TIC Tentatively Identified Compound
TMB Trimethyl Benzene
UR Uptake Rate
VI Vapor Intrusion
VOC Volatile Organic Compound
9 ACKNOWLEDGMENTS
This Engineering Issue was prepared for the U.S.
Environmental Protection Agency (EPA) Office of
Research and Development (ORD), National Risk
Management Research Laboratory (NRMRL),
Engineering Technical Support Center (ETSC) by
RTI International under Contract No. EP-C-11-036.
Doug Grosse served as the EPA Task Order Manager
and technical lead. Robert Trues dale directed the
work for RTI. Additional authors were Heidi Hayes
of Euro fins Air Toxics, Drs. Helen Dawson and
Todd McAlary of Geosyntec Consultants, and Chris
Lutes of CH2M Hill. Helpful direction and comments
were received from Rich Kapusinski of the Office of
Superfund Remediation and Emergency Response
and Dr. Henry Schuver of the EPA Office of
Resource Conservation and Recovery, for which the
authors would like to express their gratitude. The
paper was peer reviewed by John H. Zimmerman of
the EPA National Exposure Research Laboratory and
Greg Swanson of TetraTech, Inc.
As the technology and science advances, interested
parties should further consult the body of literature
and experience that constitutes the state of the art for
passive samplers. For additional information,
interested parties may also contact EPA
ORD/NRMRL/ETSC:
Dr. John McKernan, Director
U.S. EPA Engineering Technical Support Center
26 W. Martin Luther King Drive, Mail Code-190
Cincinnati, OH 45268
513-569-7415
Reference herein to any specific commercial products,
process, or service by trade name, trademark,
manufacturer, or otherwise does not necessarily
constitute or imply its endorsement,
recommendation, or favor by the United States
Government. The views and opinions of the authors
expressed herein do not necessarily state or reflect
those of the United States Government and shall not
be used for advertising or product endorsement
purposes.
Passive Samplers for Investigations ofAirQualt
21
-------
10 REFERENCES
3M. 1996. 3M™ Diffusion Monitors
3500/3510/3520/3530 Storage and Recovery.
Technical Data Bulletin #125. 3M Occupational
Health and Environmental Safety Division. St.
Paul, MN. Available at
http: / /multimedia.3m.com/mws /mediawebserve
r?mwsId=SSSSSu72KlfslxtU48 SOS eev7gel7z
HvTSevTSeSSSSSS—.
Allen, M.K., D. Grande, and L. Pansch. 2007.
Evaluation of passive sampling techniques for
monitoring roadway and neighborhood exposures
to benzene and other mobile source VOCs.
Wisconsin Department of Natural Resources
(WDNR) Publication AM-384 2007. WDNR,
Madison, WI. funded by U.S. EPA Community
Scale Monitoring Grant. Available at
http: / /dnr .wi.gov/topic /AirOuality/documents /
RoadwayStudyFinal.pdf.
ANSI/ISEA (American National Standards Institute
Inc./International Safety Equipment Association).
1998. Air Sampling Devices—Diffusive Types for Gases
and Vapors in Working Environments. ANSI/ISEA
104-1998 (R2009). American National Standards
Institute, Arlington, VA. Available at
http: / /www.safetyequipment.org/c /stdl 04-
1998.cfm.
ASTM (American Society for Testing and Materials).
2009a. D4597-10 Standard Practice for Sampling
Workplace Atmospheres to Collect Gases of Vapors with
Solid Sorbent Diffusive Samplers. ASTM
International, West Conshohocken, PA. Available
at www.astm.org.
ASTM (American Society for Testing and Materials).
2009b. D6196-03 (reapproved 2009) Standard
Practice for Selection ofSorbents, Sampling, and Thermal
Desorption Analysis Procedures for Volatile Organic
Compounds in Air. ASTM International, West
Conshohocken, PA. Available at www.astm.org.
ASTM (American Society for Testing and Materials).
2010. D6306-10 Standard Guide for Placement and
Use of Diffusion Controlled Passive Monitors for Gaseous
Pollutants in Indoor Air. ASTM International, West
Conshohocken, PA. Available at www.astm.org.
ASTM (American Society for Testing and Materials).
201 la. D7758 Neiv Practice for Passive Soil Gas
Sampling in the Vadose Zone for Source Identification,
Spatial Variability Assessment, Monitoring and Vapor
Intrusion Evaluations. ASTM International, West
Conshohocken, PA. Available at www.astm.org.
ASTM (American Society for Testing and Materials).
201 Ib. D7663-11 Standard Practice for Active Soil
Gas Sampling in the Vadose Zone for Vapor Intrusion
Evaluations. ASTM International, West
Conshohocken, PA. Available at www.astm.org.
ASTM (American Society for Testing and Materials).
2013. D6246-08 (reapproved 2013) Standard
Practice for Evaluating the Performance of Diffusive
Samplers. ASTM International, West
Conshohocken, PA. Available at www.astm.org.
Bartkow, M.E., J.N. Huckms, and J.F. Muller. 2004.
Field-based evaluation of semipermeable
membrane devices (SPMDs) as passive air
samplers of polyaromatic hydrocarbons (PAHs).
Atmospheric Environment 38(35) :5983-5990.
Available at
http://www.sciencedirect.com/science/article/pii
/S1352231004006454.
Begerow, J., E. Jermann, T. Keles, and L. Dunemann.
1999. Performance of two different types of
passive samplers for the GC/ECD-FID
determination of environmental VOC levels in air.
Fresenius Journal of Analytical Chemistry 363:399^-03.
Available at
http://link.spnnger.com/article/10.1007%2FsOO
2160051209.
Bohlm, P., FCC. Jones, and B. Strandberg. 2007.
Occupational and indoor air exposure to
persistent organic pollutants: a review of passive
sampling techniques and needs. Journal of
Environmental Monitoring 9:501—509. Available at
http://pubs.rsc.Org/en/content/articlelanding/2
007 /em/b700627f/unauth.
Brown, J., and B. Shirey. 2001. A tool for selecting an
adsorbent for thermal desorption applications. Supelco.
Available at
http://www.sigmaaldrich.com/etc/medialib/doc
s/Supelco/General Information/t402025.Par.OOO
l.File.tmp/t402025.pdf.
22
>sive Samplers for Investigations of Air Quality
-------
Brown, R.H. 1999. Environmental use of diffusive
samplers: evaluation of reliable diffusive uptake
rates for benzene, toluene and xylene. Journal of
Environmental Monitoring 1:115—116. Available at
http://pubs.rsc.org/en/content/articlepdf/1999/
EM/A807686C.
Brumbaugh, W.G., J.D. Petty, T.W. May, and J.N.
Huckins. 2000. A passive integrative sampler for
mercury vapor in air and neutral mercury species
in water. Chemosphere: Global Change Science 2:1—9.
Available at
http://www.sciencedirect.com/science/article/pii
/S1465997299000550.
CalEPA (California Environmental Protection
Agency). 2012. Advisory—Active Soil Gas
Investigation. Department of Toxic Substances
Control, Cypress, CA. Available at
http: / /www.dtsc.ca.gov/SiteCleanup /upload / VI
ActiveSoilGasAdvisory FINAL 043012.pdf.
Cassmelli, M.E., R.D. Hull, J.V. Crable, and A.W.
Teass. 1987. Protocol for the evaluation of
passive monitors, in Proceedings of Diffusive SampRng:
An Alternative to Workplace Air Monitoring. Royal
Society of Chemistry. London, A. Berlin, R.H.
Brown and K.J. Saunders (eds.). NIOSH Protocol
for the Evaluation of Passive Monitors, 190—202.
CEN (European Committee for Standardization).
1995. 838 Workplace atmospheres—Diffusive samplers
for the determination of gases and vapours—Requirements
and test methods. Comite Europeen de
Normalisation, Brussels, Belgium. Available at
http://shop.bsigroup.com/ProductDetail/?pid=0
00000000000826428.
CEN (European Committee for Standardization).
2002a. EN 13528-1 Ambient air quality—Diffusive
samplers for the determination of concentrations of gases
and vapours—Requirements and test methods—Part 1:
general requirements. European Standard. Available
at http:// www.bsria.co.uk/information-
membership /information-
centre/library/item/ambient-air-quality-diffusive-
samplers-for-the-determination-of-conc-)an-
2002/.
CEN (European Committee for Standardization).
2002b. EN 13528-2 Ambient air quality—
Diffusive samplers for the determination of
concentrations of gases and vapours—
requirements and test methods—Part 2: specific
requirements and test methods. European
Standard. Available at
http://www.bsria.co.uk/information-
membership /information-
centre/library/item/ambient-air-quality-diffusive-
samplers-for-the-determination-of-conc-jan-
CEN (European Committee for Standardization).
2003. EN 13528-3 Ambient air quality. Diffuse
samplers for the determination of concentrations
of gases and vapours, Part 3: guide to selection,
use and maintenance. European Standard.
Available at
http://esearch.cen.eu/esearch/Details.aspx?id=l
0695537.
CEN (European Committee for Standardization).
2004. EN 14412 Indoor air quality. Diffusive
samplers for the determination of concentrations
of gases and vapours. Guide for selection, use and
maintenance. British-Adopted European
Standard. Available at
http://shop.bsigroup.com/ProductDetail/?pid=0
00000000030075975.
CEN (European Committee for Standardization).
2005a. Ambient air quality. EN 14662-4 Standard
method for measurement of benzene
concentrations. Diffusive sampling followed by
thermal desorption and gas chromatography.
British-Adopted European Standard. Available at
http://shop.bsigroup.com/ProductDetail/?pid=0
00000000030093662.
CEN (European Committee for Standardization).
2005b. Ambient air quality. EN 14662-5 Standard
method for measurement of benzene
concentrations. Diffusive sampling followed by
solvent desorption and gas chromatography.
British-Adopted European Standard. Available at
http://shop.bsigroup.com/ProductDetail/?pid=0
00000000030093665.
Passive Samplers for Investigations ofAirQualt
23
-------
Cocheo, C., C. Boaretto, D. Pagani, F. Quaglio, P.
Sacco, L. Zaratin, and D. Cottica. 2009. Field
evaluation of thermal and chemical desorption
BTEX radial diffusive sampler Radiello®
compared with active (pumped) samplers for
ambient air measurements. Journal of 'Environmental
Monitoring 11:297-306. Available at
http://www.ncbi.nlm.nih.gov/pubmed/
19212585.
Crump, D., P. Harrison, and C. Walton. 2011. Aircraft
Cabin Air Sampling Study: Part 1 of the Final Report.
Cranfield University Institute of Environment and
Health, Bedfordshire, UK. Available at
https://dspace.lib.cranfield.ac.uk/bits tream/1826
75305/1 /AircraftCabm
AirSamplingStudyPartlFinalReport%2020110420.
pdf.
Daughtrey, E.H., K.D. Oliver, H.H. Jacumin, and
W.A. McClenny. 2004. Supplement to EPA
Compendium Method TO-15—reduction of
method detection limits to meet vapor intrusion
monitoring needs. Presented at Symposium on
Air Quality Measurement Methods and
Technology, Research Triangle Park, NC. April
20-22. Available at
http: / /www.epa.gov/ttnamtil /files /ambient/airt
ox/TO-15-Supplement.pdf
Dietz, R.N., and E.A. Cote. 1982. Air infiltration
measurements in a home using a convenient
perfluorocarbon tracer technique. Environment
International 8:419^-33. Available at
http://www.ecd.bnl.gov/pubs/BNL30797.pdf.
ESTCP (Environmental Security Technology
Certification Program). In press. Development of
More Cost-Effective Methods for Eong-Term Monitoring
of Subsurface Vapor Intrusion to Indoor Air Using
Quantitative Passive Diffusive-Adsorptive Sampling
•s. ESTCP ER-0830. Washington, DC.
Feigley, C.E. and B.M. Lee. 1988. Determination of
sampling rates of passive samplers for organic
vapors based on estimated diffusion coefficients.
American Industrial Hygiene Association 49(5) :266—
269. Available at
http://www.tandfonline.com/doi/abs/10.1080/
15298668891379729?)ournalCode=aiha20#previe
Gordon, S.M., P.J. Callahan, M.G. Nishioka, M.C.
Brinkman, M.K. O'Rourke, M.D. Lebowitz, and
D.J. Moschandreas. 1999. Residential
environmental measurements in the National
Human Exposure Assessment Survey (NHEXAS)
pilot study in Arizona: preliminary results for
pesticides and VOCs. Journal of'Exposure Analysis
and Environmental Epidemiology 9(5) :456—70.
Available at
http://www.nchh.Org/Portals/0/Contents/
Article0173.pdf.
Gorecki, T., and J. Namiesnik. 2002. Passive
sampling. Trends in Analytical Chemistry 21 (4) :276—
291. Available at
http://www.sciencedirect.com/science/article/
pii/S0165993602004077.
Guild, L.V., K.H. Myrmel, G. Myers, and D.F.
Dietrich. 1992. Bi-level passive monitor
validation: a reliable way of assuring sampling
accuracy for a larger number of related chemical
hazards. Applied Occupational Environmental Hygiene
7(5):310-317. Available at:
http://www.tandfonline.eom/doi/abs/10.1080/l
047322X.1992.10390161.
Harper, M. 2000. Sorbent trapping of volatile organic
compounds from air. Journal of Chromatography A
885:29-151. Available at
http://www.sciencedirect.com/science/article/pii
/S0021967300003630.
Harper, M. and L.V. Guild. 1996. Experience in the
use of the NIOSH diffusive sampler evaluation
protocol. American Industrial Hygiene Association
Journal 57(12):! 115-1123. Available at
http://www.ncbi.nlm.nih.gov/pubmed/8976586.
Hodny, J.W., J.E. Whetzel Jr., and H.S. Anderson II.
2009. Quantitative passive soil gas and air
sampling in vapor intrusion investigations, in
Proceedings of Vapor Intrusion 2009, Air & Waste
Management Association, San Diego. January.
Available at
https: / /www.agisurveys .net/uploads /Ouantitativ
e Passive Sampling for Vl.pdf.
w.
24
>sive Samplers for Investigations of Air Quality
-------
ISO (International Standards Organization). 2000.
ISO 16200-2 Workplace Air Quality—Sampling and
Analysis of Volatile Organic Compounds by Solvent
Desorption/Gas Chromatography—Part 2: Diffusive
Sampling Method. International Standards
Organization. Available at
http: / /www.iso.org/iso /catalogue detail.htmPcsn
umber=30188.
ISO (International Standards Organization). 2003.
ISO 16017-2:2003 Indoor Ambient and Workplace
Air—Sampling and Analysis of Volatile Compounds by
Sorbent Tube / Thermal Desorption/ Capillary Gas
Chromatography—Part 2: Diffusive Sampling.
International Standards Organization. Available at
http: / /www.iso.org/iso /catalogue detail.htmPcsn
umber=29195.
ISO (International Standards Organization). 2007.
ISO 16000-5:2007 Indoor Air—Part 5: Sampling
Strategy for Volatile Organic Compounds (VOCs).
International Standards Organization. Available at
http: / /www.iso.org/iso /catalogue detail.htmPcsn
umber=37388.
Jaward, P.M., N.J. Farrar, T. Harner, A.J. Sweetman,
and K.C. Jones. 2004a. Passive air sampling of
PCBs, PBDEs and organochlorine pesticides
across Europe. Environmental Science & Technology
38(1):34-41. Available at
http://www.ncbi.nlm.nih.gov/pubmed/
1474Q714.
Jaward, P.M., N.J. Farrar, T. Harner, A.J. Sweetman,
and K.C. Jones. 2004b. Passive air sampling of
polycyclic aromatic hydrocarbons and
polychlorinated naphthalenes across Europe.
Environmental Toxicology and Chemistry 23(6):1355—
1364. Available at
http://onlinelibrary.wiley.com/doi/10.1897/03-
420/abstract.
Jia, C., S. Batterman, and C. Godwin. 2007.
Continuous, intermittent and passive sampling of
airborne VOCs. Journal of Environmental Monitoring
9:1220-1230. Available at
http://pubs.rsc.Org/en/content/articlelanding/2
007/em/b708119g/unauth.
Johnson, B.J., and G.A. Dawson. 1999. Collection of
formaldehyde from clean air for carbon isotopic
analysis. Environmental Sdence & Technology
24(6):898-902. Available at
http://pubs.acs.org/doi/pdf/10.1021/es00076aO
18.
Johnson, P.C., C. Holton, E.H. Luo, Y. Guo, P.
Dahlen, K. Gorder, and E. Dettenmaier. 2013.
Multi-year monitoring of a house over a dilute
CHC plume: implications for pathway assessment
using indoor air sampling and forced under-
pressurization tests. 22nd Annual International
Conference on Soil, Water, Energy, and Air.
Association of Environmental Health and
Sciences (AEHS). San Diego. March. Available at
https://iavi.rti.org/WorkshopsAndConferences.c
fmPPageID=documentDetails&AttachID=588.
Johnson, R.L., R.B. Thorns, and J.S. Zogorski. 2003.
Effects of daily precipitation and
evapotranspiration patterns on flow and VOC
transport to groundwater along a watershed flow
path. Environmental S dence e> Technology
37(21):4944-54. Available at
http://www.ncbi.nlm.nih.gov/pubmed/
14620822.
Johnston, J.E. 2013. Vapor intrusion across space and
time: a pilot community study in San Antonio.
Presented at EPA VI Workshop at 23rd Annual
Conference on Soil, Water, Energy, and Air.
Association of Environmental Health and
Sciences (AEHS). March. Available at
https: / /iavi.rti.org/attachments /WorkshopsAnd
Conferences 708 Johnston 3-15-13.pdf.
Johnston, J.E., and J.M. Gibson. 2013.
Spatiotemporal variability of tetrachloroethylene
in residential indoor air due to vapor intrusion: a
longitudinal, community-based study. Journal of
Exposure S dence and Environmental Epidemiology
April. Available at
http://www.nature.com/jes/iournal/vaop/ncurr
ent/full/jes201313a.html.
Langlois, E. 2008. GABIE and Perkin Elmer passive
sampler performance under fluctuating
concentration conditions. The Annals of
Occupational Hygiene 52(4):239-247. Available at
http://annhyg.oxfordjournals.Org/content/52/4/
239.full.pdf.
Passive Samplers for Investigations ofAirQualt
25
-------
Lewis, R. October 3, 2000. Responses by Dr. Robert
Lewis, EPA to questions and comments regarding
Method TO-13A.
http: / /www.epa.gov/ttnamtil /Files /ambient/airt
ox/commentstol 3.pdf.
Luo, E.H., C. Holton, Y. Guo, and P.C. Johnson.
2012. Field and modeling studies of indoor air
source effects on subslab soil gas concentrations.
Presented at 21st Annual International
Conference on Soil, Water, Energy, and Air.
Association of Environmental Health and
Sciences (AEHS). San Diego, CA. March.
Available at
https://iavi.rti.org/WorkshopsAndConferences.c
fm?PageID=documentDetails&AttachID=553.
Lutes, C., B. Cosky, R. Uppencamp, L. Abreu, B.
Schumacher, J. Zimmerman, R. Truesdale, S. Lin,
H. Hayes, and B. Hartman. 2012. Short-term
variability, radon tracer, and long-term passive
sampler performance in the field. Invited
presentation, EPA Workshop: decent Advances to
Vapor Intrusion Application and Implementation at the
22nd Annual International Conference on Soil, Water,
Energy & Air. Association of Environmental
Health and Sciences (AEHS) San Diego, March.
Available at
https: / /iavi.rti.org/attachments /WorkshopsAnd
Conferences 710 Lutes AFTERNOON FINAL.
pdf.
Lutes, C.C., R. Uppencamp, H. Hayes, R. Mosley, and
D. Greenwell. 2010. Long-term integrating
samplers for indoor air and subslab soil gas at VI
sites. Presented at AWMA Specialty Conference:
Vapor Intrusion 2010, September 28—30. Chicago,
IL. Available at
http://events.awma.org/education/Posters/Final
/Lutes RadonPoster.pdf or
http://events.awma.org/education/Final%20Pap
ers/7-Lutes.pdf.
MADEP (Massachusetts Department of
Environmental Protection). 2002. Indoor Air
Sampling and Evaluation Guide. WSC POLICY #02-
430. Office of Research and Standards, Boston,
MA. Available at
http: / /www.mass.gov/dep /cleanup /laws /02-
430.pdf.
May, W.J. 1989. Gas detection by diffusive
samplers—theory and application to the Draeger
range of samplers. Annals of Occupational Hygiene
33(1):69-78. Available at
http://annhyg.oxfordiournals.org/content/33/I /
69.abstract.
McAlary, T.A., X. Wang, A. Unger, H. Groenevelt,
and T. Gorecki. 2014a. Quantitative passive soil
vapor sampling for VOCs—Part 1:
theory. Environmental Science: Processes & Impacts 16:
482^90. DOI:10:1039/C3EM00652B. Available
at http: / /xlink.rsc.org/?doi=C3EM00652B.
McAlary, T.A., H. Groenevelt, S. Seethapathy, P.
Sacco, D. Crump, M. Tuday, B. Schumacher, H.
Hayes, P. Johnson, and T. Gorecki.
2014b. Quantitative passive soil vapor sampling
for VOCs—Part 2: laboratory
experiments. Environ. Sci.: Processes Impacts
16(3):491-500. doi: 10.1039/C3EM00128H:
http://xlink.rsc.org/?doi=C3EM00128H.
McAlary, T.A., H. Groenevelt, P. Nicholson, S.
Seethapathy, P. Sacco, D. Crump, M. Tuday, H.
Hayes, B. Schumacher, P. Johnson, T. Gorecki,
and I. Rivera Duarte. 2014c. Quantitative passive
soil vapor sampling for VOCs—Part 3: field
experiments. Environ. Sci.: Processes Impacts
16(3):501-510. DOI: 10.1039/C3EM00653K:
http: / /xlink.rsc.org/?doi=C3EM00653K.
MDHS (Methods for the Determination of
Hazardous Substances). 1994. Protocol for Assessing
the Performance of a Diffusion Sampler. MDHS 27.
Volatile Organic Compounds in Air. Health and
Safety Executive, Occupational Medicine and
Hygiene Laboratory.
MDHS (Methods for the Determination of
Hazardous Substances). 1995. Laboratory Method
Using Diffusion Solid Sorbent Tubes, Thermal
Desorption and Gas Chromatographj. MDHS 80.
Volatile Organic Compounds in Air. Health and
Safety Executive, Occupational Medicine and
Hygiene Laboratory. Available at
http: / /www.hse.gov.uk/pubns /mdhs /pdfs /mdh
sSO.odf.
26
>sive Samplers for Investigations of Air Quality
-------
MDHS (Methods for the Determination of
Hazardous Substances). 1997. Laboratory Method
Using Diffusive Samplers, Solvent Desorption and Gas
Chromatography. MDHS 88. Volatile Organic
Compounds in Air. Health and Safety Executive,
Occupational Medicine and Hygiene Laboratory.
December. Available at
http: / 7www.hse.gov.uk/pubns /mdhs /pdfs /mdh
s88.pdf.
Mosley, R.B., D. Greenwell, A. Lee, K. Baylor, M.
Plate, and C. Lutes. 2008. Use of integrated
indoor radon and volatile organic compounds
(VOCs) to distinguish soil sources from above-
ground sources. Extended abstract and oral
presentation. AWMA Symposium on Air Quality
Measurement and Technology, Chapel Hill, NC.
November 6.
Mukerjee, S., L.A. Smith, G.A. Norris, M.T. Morandi,
M. Gonzales, C.A. Noble, L.M. Neas, and A.H.
Ozkaynak. 2004. Field method comparison
between passive air samplers and continuous
monitors for VOCs and NO2 in El Paso, Texas.
Journal of the Air e> Waste Management Association
54:307-319. Available at
http: / 7www.ogawausa.com/pdfs /fieldmethod.
pdf.
Namiesnik, J., T. Gorecki and E. Kozlowski. 1984.
Passive dosimeters—an approach to atmospheric
pollutants analysis. The Science of the Total
Environment 38:225-258.
Odencrantz, J. E., and H. O'Neill. 2009. Passive to
active tie-in for soil gas surveys: improved
technique for source-area, spatial variability,
remediation-monitoring, and vapor-intrusion
assessment. Remediation Journal 19:71—83. Available
at
http://onlinelibrary.wiley.com/doi/10.1002/rem.
20204/abstract.
Odencrantz, J.E., H. O'Neill, S.J. Stemmacher, J.D.
Case, and P.C.Johnson. 2008. Residential vapor
intrusion evaluation: long duration passive
sampling vs. short duration active sampling.
Remediation Journal 18(4):49—54. Available at
http://onlinelibrary.wiley.com/doi/10.1002/rem.
20181/abstract.
OSHA (Occupational Safety and Health
Administration). 2003. Performance of SKC Ultra
Passive Sampler Containing Carboxen 1016,
Carbotrap Z, or Chromosorb 106 When
Challenged with a Mixture Containing Twenty of
OSHA SLTC's Top Solvent Analytes. Methods
Development Team, Industrial Hygiene
Chemistry Division, Salt Lake Technical Center.
Salt Lake City, UT. Available at
http: / /www.osha.gov/dts /site /methods /studies /
chrom 106 ultra/chrom 106 ultra.html.
OSHA (Occupational Safety and Health
Administration). 2008. Personal Sampling for Air
Contaminants. Chapter 1 of Section 2 of the
OSHA Technical Manual (OTM). Directive
Number: TED 01-00-015 [TED 1-0.15A].
OSHA, Department of Labor, Washington, DC.
Available at
http://www.osha.gov/dts/osta/otm/otm ii/otm
ii l.html#organic vapors gases.
Oury, B., F. Lhuillier, J. Protois, and Y. Morele. 2006.
Behavior of the GABIE, 3M3500, PerkmElmer
Tenax TA, and RADIELLO 145 diffusive
samplers exposed over a long time to a low
concentration of VOCs. Journal of Occupational and
Environmental Hygiene. 547—557. Available at
http://www.ncbi.nlm.nih.gov/pubmed/
16908455.
Palmes, E.D. and A.F. Gunnison. 1973. Personal
monitoring device for gaseous contaminants.
American Industrial Hygiene Association Journal 34:78—
81. Available at
http://www.tandfonline.eom/doi/abs/10.1080/0
002889738506810#.UbHwzfnVApk.
Pearson, C. 2005. Interlaboratory comparison of
ambient air samples. Presented at National Air
Monitoring Strategy Information: QA National
Meeting, San Diego, CA, April 11—14. Available at
http: / /www.epa.gov/ttnamtil /files /ambient/mo
nitorstrat/pearson.pdf.
Radiello. 2013. Radiello Manual: Volatile organic
compounds (VOCs) chemically desorbed with CS22.
Supelco Edition, Sigma-Aldrich, Bellefonte, PA.
Available at
http://www.sigmaaldrich.com/content/dam/sig
ma-aldrich/docs/Supelco/Application Notes/
radiello dl d6.pdf.
Passive Samplers for Investigations ofAirQualt
27
-------
Schuver, H.J., 2013. Radon as a tracer & risk driving
co-contaminant in chemical vapor intrusion.
Presented at Remediation Technology (RemTEC)
2013 Summit, Vapor Intrusion: Characterization,
Attenuation and Mitigation Session, Westminster,
CO, March 6.
Schuver, H.J., and R.B. Mosley. 2009. Investigating
vapor intrusion with confidence & efficiency
(some observations from indoor air-based radon
intrusion studies). Keynote presentation, Air and
Waste Management Association Vapor Intrusion
Specialty Conference, November 27—30, San
Diego.
Seethapathy, S. and T. Gorecki. 2010a.
Polydimethylsiloxane-based permeation passive
air sampler. Part I: calibration constants and their
relation to retention indices of the znalytes. Journal
ofChromatographyA 1218(l):143-55. Available at
http://www.ncbi.nlm.nih.gov/pubmed/
21112594.
Seethapathy, S. and T. Gorecki. 201 Ob.
Polydimethylsiloxane-based permeation passive
air sampler. Part II: effect of temperature and
humidity on the calibration constants. Journal of
Chromatography A 1217(50):7907-13. Available at
http://www.ncbi.nlm.nih.gov/pubmed/
21067763.
Seethapathy, S., T. Gorecki, and X. Li. 2008. Passive
sampling in environmental analysis—review.
JournalofChromatographyA II84:234—253.
Available at
http://www.ncbi.nlm.nih.gov/pubmed/
17719053.
SKC. 2004. EPA IP-6 Method Update. SKC Publication
1661 Rev 1001. SKC Inc., Eighty Four, PA.
Available at
http: / /www.skcinc.com/instructions /IP6A.pdf.
SKC. 2008. Measuring Sub-ppb Levels ofVOCs in Indoor
Air A Performance Comparison of Diffusive Samplers
and Canisters. SKC Technical Note Publication
1720 Rev 1208. SKC Inc., Eighty Four, PA.
Available at
http: / /www.skcinc.com/instructions /1720.pdf.
Steck, D. 2013. Related evidence from radon VI
studies from buildings across the U.S and the
world. Presented at 22nd Annual International
Conference on Soil, Water, Energy & Air.
Association of Environmental Health and
Sciences (AEHS), San Diego, March 2013.
Available at
https: / /iavi.rti.org/attachments /WorkshopsAnd
Conferences/11 Steck 3-16-13.pdf.
Sweitzer, T., M. Schuchardt, and M. Caudill. 2006.
Case study: diffusion tube (passive) sampling of
air toxics across the Chicago urbanized area.
Presented at U.S. EPA 2006 National Air
Monitoring Conference, November 6—9, Las
Vegas, NV. Available at
http: / /www.epa.gov/ttnamtil /files /2006confere
nce/schuchardt.pdf.
Thoma, E.D., M.C. Miller, K.C. Chung, N.L. Parsons,
and B.C. Shine. 2011. Facility fence-line
monitoring using passive samplers. Journal of the
Air e> Waste Management Association 61:834—842.
Available at
http://www.tandfonline.com/doi/pdf/10.3155/
1047-3289.61.8.834.
Tolnai, B., A. Gelencser, C. Gal, and J. Hlavay. 2000.
Evaluation of the reliability of diffusive sampling
in environmental monitoring. Analjtica Chimica
Acta 408:117-122. Available at
http://www.sciencedirect.com/science/article/
pii/S0003267099008788.
U.S. EPA (United States Environmental Protection
Agency). 1989. MethodIP4-A Determination of Air
'Exchange Rate in Indoor Air. Chapter IP-4 in
Compendium of Methods for Indoor Air. EPA 600/4-
90/010. Office of Research and Development,
Research Triangle Park, NC. April. Available at
http: / /www.arb.ca.gov/research /apr/reports /
13056a.pdf.
U.S. EPA (United States Environmental Protection
Agency). 1990. Compendium of Methods for the
Determination of Air Pollutants in Indoor Air.
EPA/600/4-90/010. Office of Research and
Development, Research Triangle Park, NC. See
especially Appendix C-2 "Procedure for
Placement of Stationary Passive Samplers in
Indoor Environments." Available at
http://nepis.epa.gov/Exe/ZyP URL.cgi?Dockey=
P1004G22.txt.
28
>sive Samplers for Investigations of Air Quality
-------
U.S. EPA (United States Environmental Protection
Agency). 1992. Indoor Radon and Radon Decay
Product Measurement Device Protocols. Office of Air
and Radiation. Available at
http: / 7www.epa.gov/radon /pubs /devprotl .html.
U.S. EPA (United States Environmental Protection
Agency). 1993. Radon Reduction Techniques for
Existing Detached Houses Technical Guidance (Third
Edition) for Active SoilDepressuri^ation Systems.
EPA/625/R-93/011. Office of Research and
Development, Research Triangle Park, NC.
Available at
http://www.ndhealth.gov/aq/iaq/Radon/detach
ed houses.pdf.
U.S. EPA (United States Environmental Protection
Agency). 1997. National Human Exposure
Assessment Survey (NHEXAS) Arizona Study Quality
Systems and Implementation Plan for Human Exposure
.Assessment: Use of a Passive Sampling Device for the
Collection of Airborne VOCs at Fixed Indoor and
Outdoor Sites. Standard Operating Procedure SOP-
UA-F-12.0. Office of Research and Development,
Research Triangle Park, NC. June. Available at
http: / /www.epa.gov/heds /23038doc.pdf.
U.S. EPA (United States Environmental Protection
Agency). 1998. Technical Guidance Document:
Compliance Assurance Monitoring, Chapter 4.0
Technical Reference for Monitoring Equipment
and Instruments. Office of Air Quality Planning
and Standards, Research Triangle Park, NC.
August. Available at
http://www.epa.gov/ttn/emc/cam/sec4-5.pdf.
U.S. EPA (United States Environmental Protection
Agency). 1999a. Compendium of Methods for the
Determination of Toxic Organic Compounds in Ambient
Air, 2nd Ed., Compendium Method TO-15,
Determination of Volatile Organic Compounds
(VOCs) in Air Collected in Specially-Prepared
Canisters and Analyzed by Gas Chromatography/
Mass Spectrometry (GC/MS). PA/625/R-
96/01 Ob. Center for Environmental Research
Information, Office of Research and
Development, Cincinnati, OH. Available at
http: / /www.epa.gov/ttnamtil /files /ambient/airt
ox/TO-15r.pdf.
U.S. EPA (United States Environmental Protection
Agency). 1999b. Compendium of Methods for the
Determination of Toxic Organic Compounds in Ambient
Air, 2nd Ed., Compendium Method TO-17,
Determination of Volatile Organic Compounds in
Ambient Air Using Active Sampling onto Sorbent
Tubes. EPA/625/R-96/010b. Available at
http: / /www.epa.gov/ttnamtil /files /ambient/airt
ox/TO-17r.pdf.
U.S. EPA (United States Environmental Protection
Agency). 2003. Determinative Chromatographic
Separations. SW-846 Method 8000C. Office of
Solid Waste and Emergency Response,
Washington, DC. Available at
http: / /www.epa.gov/osw/hazard/testmethods /p
dfs/8000c v3.pdf.
U.S. EPA (United States Environmental Protection
Agency). 2009a. Standard Operating Procedure for the
Collection of Volatile Organic Compounds for the EPA
School Air Toxics Program. U.S. EPA Region 4,
Science and Ecosystem Support Division, Athens,
GA. August. Available at
http: / /www.epa.gov/ttnamtil /files /ambient/airt
ox/2009sat/SOPVOC.pdf.
U.S. EPA (United States Environmental Protection
Agency). 2009b. Standard Operating Procedure for
Measurement of Volatile Organic Compounds Using
Canisters with Passive Air Sampling Kits. U.S. EPA
Region 4, Science and Ecosystem Support
Division, Athens, GA. July. Available at
http://nepis.epa.gov/Exe/ZyP URL.cgi?Dockey=
P1006MM8.txt.
U.S. EPA (United States Environmental Protection
Agency). 2012. Fluctuation of Indoor Radon and VOC
Concentrations Due to Seasonal Variations.
EPA/600/R-12/673. National Exposure
Research Laboratory, Las Vegas, NV. October.
Available at
http: / /www.epa.gov/esd/cmb /pdf/EPA600-R-
09-073.pdf.
U.S. Navy, SPAWAR Systems Center Pacific
Division. 2013. Improved Assessment Strategies for
Vapor Intrusion Passive Samplers and Building Pressure
Control. Technical Report 2018. September.
Passive Samplers for Investigations ofAirQualt
29
-------
Walgraeve, C., K. Demeestere, J. Dewulf, K. Van
Huffel, and H. Van Langenhove. 2011. Diffusive
sampling of 25 volatile organic compounds in
indoor air: uptake rate determination and
application in Flemish homes for the elderly.
.Atmospheric Environment 45: 5828—5836.
Wolfenden, E. 2010. Sorbent-based sampling
methods for volatile and semi-volatile organic
compounds in air. Part 1: sorbent-based air
monitoring options. Journal of Chromatography A.
1217(16):2674-2684. Available at
http://www.sciencedirect.com/science/article/pii
/S0021967309018470#.
Yao, Y. 2007. Sources and occurrence of dacthal in
the Canadian atmosphere. Environmental Sdence
Technology 41:688-694. Available at
http://www.ncbi.nlm.nih.gov/pubmed/
17328171.
Zabiegala, B., and J. Namiesnik. 2007. Use of
permeation passive samplers in air monitoring.
Chapter 4 in Comprehensive Analytical Chemistry
48:85-106.
Zencak, Z., J. Klanova, I. Holoubek, and O.
Gustafsson. 2007. Source apportionment of
atmospheric PAHs in the western Balkans by
natural abundance radiocarbon analysis.
Environmental Science e> Technology 41 (11):3850—
3855. Available at
http://www.ncbi.nlm.nih.gov/pubmed/
17612159.
30
>sive Samplers for Investigations of Air Quality
-------
APPENDIX A:
RESEARCH ON PASSIVE SAMPLERS
A.1 Introduction
Passive samplers have been used extensively for
industrial hygiene monitoring (ASTM, 2009a) and a
variety of environmental sampling applications
(Gorecki and Namiesnik, 2002; Seethapathy et al.,
2008), but the use of passive sampling for vapor
intrusion (VI) investigations involves lower
concentrations (ppb instead of ppm), a wider range of
conditions (e.g., wind, temperature and humidity for
outdoor air samples, which are important for
assessing background vapor concentrations), and
media that have been less extensively tested (e.g., soil
gas monitoring). The application of passive samplers
to conditions not normally encountered in industrial
hygiene monitoring involves some uncertainties,
particularly the accuracy of the uptake rate (Tolnai et
al., 2000; Bohlm et al., 2007). To address these
uncertainties, recent research has been conducted
specifically to address the testing conditions relevant
to VI investigations and compare passive sampler
performance to methods (e.g., TO-15) that are
considered more conventional for VI investigations.
A.2 ESTCP Passive Sampling Research
The Department of Defense Environmental Security
Technology Certification Program (ESTCP) and the
U.S. Navy sponsored passive sampler testing that
included laboratory tests in controlled chambers with
a range of temperature, humidity, velocity,
concentration, and duration for 10 VOCs spanning a
range of properties (vapor pressure, solubility,
Henry's Constant, and sorptive affinity), as well as
field testing of indoor, outdoor, and soil gas samples
(ESTCP, in press, ER0830). This laboratory showed
that the uptake rates often change by statistically
significant (5% level of significance) amounts in
response to the factors tested in the chambers, but
not in a consistent or readily predictable way for all
combinations of different samplers and chemicals.
The indoor air field testing showed that passive
samplers can provide similar accuracy and precision
to Summa canisters and TO-15 up to a week's
duration for many compound/sampler combinations;
however, a low bias was typically observed when the
product of the uptake rate and sample duration was
greater than the safe sample volume for a particular
chemical/sorbent combination, which appeared to be
attributable to poor retention. Outdoor air field
testing showed mostly results below the detection
limit. A breakthrough was achieved in passive soil gas
sampling because mathematical modeling and field
sampling both showed that passive samplers can be
used to quantify soil vapor concentrations provided
the uptake rate of the sampler is less than the supply
rate of vapors from the surrounding materials to
avoid low bias from the starvation effect (McAlary et
al., 2014a, 2014b, 2014c). This finding contradicts
recent documents (ASTM, 201 la; CalEPA, 2012) that
state that passive sampling cannot be used to quantify
soil vapor concentrations.
A.3 U.S. EPA (2012) Passive Sampler
Research
U.S. EPA (2012) evaluated the performance of
Radiello charcoal passive samplers for various VOCs
at a range of sampling durations in a test house with
chloroform and tetrachloroethene (PCE) VI in
Indianapolis. Indoor air VOC concentrations during
the test are provided in Table A-l for chloroform and
PCE along with trichloroethene (TCE) and
petroleum-related chemicals that were predominantly
from outdoor air. Radiello charcoal samplers were
selected for the study because the high sampling rates
of the radial style sampler provided good sensitivity at
these concentrations for the weekly samples.
Additionally, the charcoal sorbent cartridge was
selected over the thermally desorbable cartridge
because of its stronger retention characteristics for
the target VOCs and its higher VOC loading capacity,
both beneficial attributes for long-term sample
exposure. Daily, 2-week, 4-week, 3-month, 6-month,
and 1-year sample durations were compared against
Passive Samplers for Investigations of Air Qua It
31
-------
concentrations for the same periods built up from
weekly measurements, with an accuracy (percent bias)
goal of ±30%.
Table A-1. VOC Concentration Ranges for U.S. EPA (2012) Based
on 371 Weekly Radiellos
Compound
Benzene
Chloroform
Hexane
PCE
Toluene
TCE
Min.
0.36
0.06
0.23
0.08
0.50
0.01
Mean
0.80
0.33
0.68
1.10
1.80
0.12
Max.
2.3
4.0
2.6
22.0
6.0
2.7
SD
0.30
0.42
0.38
2.40
1.10
0.26
Table A-2 and Figure A-1 show the percent bias by
duration and VOC detected in indoor air. Percent
bias varied by chemical with a general increase with
increasing VOC volatility and increasing sample
duration. Toluene and PCE met the performance
criteria for all durations tested, with mean biases well
below 7% except for 21% for PCE at the 1-year
sample duration. Hexane performed within bias limits
for up to 3 months (with 23% bias), and TCE and
benzene performed well up to 1 month (with 19%
and 13% bias, respectively). Chloroform, the most
volatile chemical tested, did not perform well for
sample durations 1 month or longer. Percent biases
were positive for all sampling durations significantly
different from the 1 -week measurements, indicating
that the longer duration samples consistently
underpredicted results built up from weekly samples,
although in many cases this consistent bias was under
30%.
A.4 Research Comparing Passive
Samplers with Other Methods
Summa canister sampling using TO-15 (U.S. EPA,
1999a) is considered by many to be a "standard" for
VI studies; therefore research comparing the
performance of TO-15 to passive sampling methods
is of interest in this document. With respect to
TO-15, the method defines the single laboratory
performance criterion for method TO-15 as a
replicate precision within 25% and an audit accuracy
"within 30percentfor concentrations normally expected in
contaminated ambient air?' In contrast, results of
interlaboratory comparison tests with method TO-15
suggest that the standard for accuracy is not achieved
in practice. A recent TO-15 interlab oratory
comparison administered by the commercial standard
company, ERA, found the following acceptance
ranges for PCE results: 33% to 168% (for a July to
September 2009 study) and 56% to 131% (for an
October to November 2007 study) (Lutes et al.,
2012).
In 2007, a TO-14/TO-15 study conducted by Scott
Specialty Gasses noted that values for toluene
reported by 12 labs varied from 51 to 290% (Lutes et
al., 2012). In a study by Daugherty et al. (2004)
comparing four laboratories, one under contract to
EPA and three typically used by the Colorado
Department of Health, the results for 1,1-DCE in one
of the laboratories were more than double the results
of the three remaining laboratories.
Results for an intercomparison of 10 laboratories
have been presented by Pearson (2005) of the
California Air Resources Board. This test was
conducted with a real-world ambient air sample
collected using a specially designed apparatus that
allowed 14 Summa canisters to be sampled in parallel.
The results show approximately a 2-times variation
across the set of 10 labs for the compounds of
interest to this project. It is also interesting to note
that the apparent outlier labs were not the same for
each compound.
itions of Air Quality
-------
Table A-2. Average Percent Bias from 1-Week Sample Results for Various Radiello Sample Durations (U.S. EPA, 2012)
voc
Chloroform
Benzene
TCE
Hexane
PCE
Toluene
Vapor Pressure
(mm Hg)
197
95
69
151
19
28
Average Percent Bias (Standard Error) by Sample Duration and Chemical
2 Week
12 (2.0)
8.0(1.9)
11(2.0)
3.1(1.4)
-0.10(0.84)
0.40 (0.84)
1 Month
31 (6.3)
13(3.0)
19 (3.4)
11(2.2)
-0.50(1.2)
0.0(1.2)
3 Month
88 (23)
33 (5.5)
34(6.7)
23(3.7)
5.9(2.0)
2.4 (2.3)
6 Month
150 (31)
40 (5.5)
42 (8.4)
32 (4.5)
2.9 (3.3)
-4.8(3.1)
1Year
NA
NA
NA
67(2.4)
21 (0.77)
0.80 (0.05)
Based on comparison with weekly results, positive numbers indicate negative bias (compared sampler underestimates concentration derived from weekly samples).
Highlights indicate cases where average percent bias is >30.
Italics indicate statistically significant differences (P < 0.05) between weekly samples and tested durations.
NA—not enough data for analysis
Bias of Passive, Solvent Extracted Samplers Over Time
0.06-
0.04-
3.32-
™ 3.33-
E
o
o
o 0.35-
3.34-
3 32-
o.oo-
Sample Duration
Two Weefcs
N=175
One Month
N = 84
Three Months
N = 28
Six Months
N=14
One Yeai
0 53 -50 -25 3 25 -CO -40 -23 0 20
% Bias (Positive bias indicates the longer sample under-reported concentration)
Figure A-1. Percent bias of VOCs by sample duration (U.S. EPA, 2012)
Passive Samplers for Investigations of Air Quai
33
-------
Sweitzer et al. (2006) compared Perkin Elmer passive
samplers to TO-15 and on-site GC in ambient
sampling. Agreement for BTEX was within 30% of
the Summa canister result. Odencrantz et al. (2008)
reported comparison between tube-style passive
samplers and found that the results were always
within two times and that the tube-type samplers were
generally lower than the TO-15. Lutes et al. (2010)
reported two field studies comparing thermally and
solvent extracted Radiello samplers with TO-15
samples collected at the beginning, middle, and end of
2-week-long passive sampling periods. The TO-15
results were generally slightly higher. Good agreement
was found for chlorinated solvents and volatile
aromatics but poor for polar compounds.
Thoma et al. (2011) compared Perkin-Elmer tube-
style samplers with reasonable agreement with an
automated GC with r2 = 0.86 for benzene in ambient
air. Allen et al. (2007) compared a tube-type passive
sampler to an automated field GC and generally
found good correlation and the passive
concentrations averaging 55 to 90% of the GC values.
Mukerjee (2004) compared a 3M OVM 3520 badge
sampler with a field GC and found good agreement
with the passive sampler 5 to 10% above the field GC
values.
Because intermethod results can't be expected to
compare better than interlaboratory variability of the
single methods compared (general + 30%), most of
these TO-15 to passive and GC to passive
comparisons are considered to be acceptable.
itions of Air Quality
-------
APPENDIX B:
AVAILABLE PASSIVE SAMPLING METHODS AND APPLICABILITY
B.1 Introduction
The purpose of this section is to provide an overview
of the available international and U.S. methods and
standards that may be less familiar to American vapor
intrusion (VI) practitioners. International and
European standard methods are available for passive
sampling that are applicable to indoor air applications
for VOCs at low concentrations. The ASTM also has
developed passive sampler standard methods. EPA
has written that the "Use of ASTM and ISO
(International Standards Organization) standards in lieu
of EPA. methods is consistent mth Public Law 104-113,
which mandates that federal agencies use private consensus
standards organisations whenever possible to develop
standardised methods" (Lewis, 2000) Note also that
method TO-17 is largely developed from and cites
United Kingdom (MDHS) methods.
B.2 International Methods and Standards
The available international methods that may be
pertinent to VI investigations include the following.
B.2.1 ISO3 Method ISO 16017-2
ISO 16017-2 (ISO, 2003) is applicable "to a wide range
of VOCs including hydrocarbons, halogenated hydrocarbons,
esters, glycol ethers, ketones and alcohols ... applicable to the
measurement of airborne vapours of VOCs in a mass
concentration range of approximately 0.002 mgl m3 to 100
mgl m3 individual organic for an exposure time of8h or 0.3
Ugl m3 to 300 ngi' m3 individual organic for an exposure time
of four weeks" The method is designed for thermal
desorption with GC/flame ionization detector (FID)
"'or other suitable detector" analysis. The method provides
detailed directions for sampling and analysis. The
method is primarily focused on tube-type geometries
3 ISO is a network of national standards bodies. For example, in
the United States, ANSI, the American National Standards
Institute, is a member.
4 European standards are published by the European Committee
for Standardization, an international nonprofit that works
under a framework from the European Union.
and provides uptake rates for a number of single
sorbents in tube-type samplers. However, the method
is open to other sorbents: "Equivalentproducts may be
used if they can be shown to lead to the same result?'
Tabulated QC data are provided for recoveries from
spiked tubes, precision of analysis, and storage
recovery for durations up to 11 months. The
precision and accuracy estimates compare quite
favorably with those of EPA air sampling methods
with overall uncertainty being better than 30%. Table
Cl in the standard provides a detailed cross reference
between sorbent and suitable analytes.
B.2.2 European Standard4 EN 14412 (2004)
EN 14412 (CEN, 2004) provides guidelines for the
selection, use, and maintenance of diffusive samplers
used to "''analysegaseous'pollutants in indoor air including
measurement and planning" It covers use of passive
media both for indoor air quality sampling and as a
personal sampler (defined as "a device attached to a person
that samples air in the breathing spne"*). The standard
provides specific provisions for sampler deployment
(location, etc.) that are discussed in Section 4 of this
EIP and provides an annex table describing the
characteristics of various tube, badge, and radial
diffusion samplers for benzene and other VOCs as
well as formaldehyde.
B.2.3 United Kingdom Method MDHS5 80
MDHS 80 (MDHS, 1995) is designed for passive
sampling and analysis for a wide variety of volatile
organics (depending on the selected sorbent) and is
primarily focused on tube-type samplers that are
thermally desorbed. MDHS 80 is designed for
concentrations from 1 to 1,000 mg/m3 and relatively
shorter exposure times of 30 min to 8 hours. The
5 MDHS — Methods for the Determination of Hazardous
Substances. Unlike the ISO and EN methods that must be
purchased, the MDHS methods are available for free at
http: / 7www.hse.gov.uk/pubns /mdhs /.
Passive Samplers for Investigations ofAirQuah
35
-------
method refers to published method validation with a
Perkin-Elmer diffusive tube. They report a laboratory
determination of precision of 12% for benzene,
toluene, heptane, xylene, and decane. Storage stability
to 11 months is reported. Diffusive uptake rates are
provided for the Perkin-Elmer sorbent tubes for a
variety of sorbents for 14 hydrocarbons, 19
halogenated hydrocarbons, 11 esters and ethers, and
11 other compounds. A temperature correction of the
uptake rate is provided for benzene.
B.2.4 United Kingdom Method MDHS 88
MDHS 88 (MDHS, 1997) is designed for diffusive
sampling, followed by solvent extraction and gas
chromatography analysis. The method does not
specify one sampler geometry, referring instead to
"sorbent separated from ambient air by some form of diffusion
resistance" The method mentions badge, tube, and
radial geometries. Uptake rates are tabulated for four
different major manufacturer samplers covering tube,
badge, and radial geometries. Uptake rates for the
tube and badge samplers are provided for 39
hydrocarbons, 39 halocarbons, 28 esters, 30 alcohols
and glycol ethers, 20 ketones, 13 ethers, and 15
miscellaneous compounds. For the radial geometry
uptake, rates are provided for 12 hydrocarbons, 6
halocarbons, 4 esters, 11 alcohols/glycol ethers, four
ketones, 2 ethers, and one miscellaneous compound.
Specific sampling instructions are provided for each
of these four types of samplers. An extensive
discussion of the various levels of quality for uptake
rate determinations is provided. Carbon disulfide
desorption is emphasized for nonpolar compounds,
and a variety of solvents are discussed for polar
compounds. Both GC-FID and GC-MS analytical
methods are discussed.
B.2.5 European Standards EN13528
EN13528 is a general standards designation
comprising several standards documents. EN13528-1
(CEN, 2002a) is introductory material, primarily
definitions of terms and DQOs that will be used in
the other parts of EN 13528. The German standard
DIN EN 13528-3 (CEN, 2003) has considerable
overlap in content with the other European standards
discussed above. Tube, badge, and radial samplers are
mentioned with both thermal desorption and solvent
extraction. This document does, however, provide
some useful additional guidance on outdoor sample
placement and operator training. EN13528-2 (CEN,
2002b) focuses on ambient air but also is intended to
be relevant to indoor air. This document includes
procedures to qualify passive sampling devices based
on laboratory or field tests. As such, it is not in and of
itself a field sampling method. It notes: "This standard
can encourage the development of new types of diffusive sampler"
in that it provides procedures for sampler evaluation.
Detailed instructions are included on how to evaluate
passive sampler characteristics including desorption
efficiency, effect of air velocity/sampler orientation,
storage after sampling, shelf life before sampling,
blank values, uncertainty, and uptake rate.
B.2.6 European Standards EN 14662-4, EN 14662-5
EN 14662-4 (CEN, 2005a) is specific to benzene by
diffusive sampling with thermal desorption analysis.
The target concentration range is 0.5 to 50 ug/m3
with a typical sampling period of 14 days. Uptake
rates are provided for three sorbents in tube-type
geometries with varying exposure times and estimates
of standard deviation in uptake rate. An extremely
detailed uncertainty propagation analysis is developed
in an appendix leading to a conclusion that the
combined relative uncertainty, including both
sampling and analysis, will be 6.7%. An
interlaboratory comparison of five laboratories is
provided using six different test atmospheres,
including two derived from ambient air. The resulting
interlab oratory uncertainty for these measurements
ranged from 3.1 to 9.0%. It should be noted that
these interlaboratory results are better than those
commonly reported for Summa canister methods,
which typically show a range of a factor of 2 to 5
between the minimum and maximum of the study
acceptance window, or between the lowest and
highest lab in the study (Lutes et al., 2012; Pearson,
2005, Daugherty et al., 2004). EN 14662-5 (CEN,
2005b) is also specific to benzene but is relatively
Passive Sa
itions of Air Quality
-------
unique among the EN methods in that it specifies a
charcoal passive sampler with solvent desorption for
longer-term sampling.
B.3 U.S. Methods and Standards
In the United States, the most comprehensive
guidance published for diffusive sampling is from
ASTM, although EPA has also developed some
methods documents.
B.3.1 ASTM Method D6196-03
ASTM D6196-3 (ASTM, 2009b) covers ambient and
indoor VOC sampling using either pumped or
diffusive methods. Both tube and radial geometries of
passive sampling are specifically discussed. A large
range of VOCs with boiling points from 0 to 400°C is
encompassed. The method, however, suggests that
radial samplers should only be used for compounds
with an equal or lower volatility than benzene. For
axial samplers, it states a range of 0.3 mg/m3 to 300
mg/m3 for exposure times of 4 weeks. Typical
background artifact levels are provided for various
sorbent types. The method discusses the use of two
sorbents in separate devices used in parallel to extend
the range of compounds monitored. Analysis under
this method is by thermal desorption with GC-ECD
(electron capture detector), GC-FID, or GC/MS.
Humidity up to 95% in the sampling area can be
handled with hydrophobic sorbents, but the method
advises caution for humidity greater than 65% with
less hydrophobic, strong sorbents. Options specified
to manage this problem include dry purging, use of a
membrane that excludes water from the diffusive
sampler, or reduction of the time of diffusive
sampling. Diffusive sampling rates (uptake rates) are
tabulated with a number of different sorbents in axial
diffusion tubes for 29 hydrocarbons, 23 halogenated
hydrocarbons, 22 esters/glycol ethers, 9 ethers, 7
alcohols, and 13 miscellaneous compounds based on
an 8-hour exposure period. Sampling rates for axial
tube samplers and 2- and 4-week exposures are
provided for benzene, toluene, ethylbenzene, and
xylenes (BTEX) and a few other hydrocarbons. The
method provides sampling rates for radial diffusive
samplers for 22 compounds and 1-week exposures.
Sampling periods from 1 to 4 weeks are
recommended for indoor and ambient monitoring.
B.3.2 ASTM Method D6306-10
ASTM D6306-10 (ASTM, 2010) focuses on placing
and using diffusive samplers in the indoor
environment. However, information is provided
about practical matters such as how to label samples
without contaminating them and recording site-
specific information about the sample. Monitoring
placements are specified for both personal and area
sampling.
B.3.3 ASTM Method D6246-08
ASTM D6246-08 (ASTM, 2013) provides procedures
for evaluating the performance of diffusive samplers
through chamber experiments. It is focused on short-
duration exposure periods of 4 to 12 hours and thus
appears to be primarily focused on samplers used in
workplace applications.
B.3.4 ANSI/ISEA Standard 104-1998 (R2009)
A passive sampler standard is also available from the
American National Standards Institute
Inc./International Safety Equipment Association
(ANSI/ISEA 104-1998 [R2009]; ANSI/ISEA, 1998).
This standard was developed for occupational
exposures and covers both direct on-site reading
devices and samplers and laboratory analysis. It
focuses on chamber tests of passive sampler
performance over durations of 15 min to 8 hours. It
describes multifactorial tests that can be done to
evaluate sampler performance at various
temperatures, humidities, face velocities, orientations,
and contaminant concentrations (both constant and
transient peaks). Evaluation approaches for storage
time, shelf life, sorbent reuse, and interferences are
also provided.
There are also several U.S. government (and state)
methods available, including U.S. EPA's Procedure
for Placement of Stationary Passive Samplers in
Indoor Environments (Appendix C-2 in U.S. EPA's
Compendium of Methods for the Determination ofA_ir
Passive Samplers for Investigations ofAirQualt
37
-------
Pollutants in Indoor Air [U.S. EPA, 1990]). This
document refers to sampling durations from 3 days to
1 year. Through a cooperative agreement, Arizona
State University and EPA's Office of Research and
Development prepared a standard operating
procedure (SOP) for passive sampling for VOCs at
indoor and outdoor sites for benzene, toluene, and
TCE (U.S. EPA, 1997). The SOP is specific to
activated charcoal with the 3M OVM 3500 sampler. It
provides some detailed guidance for sampler location
and how to screen for and evaluate problems such as
fallen samplers. U.S. EPA Region 4 has also
developed passive sampling SOPs for the analysis of
VOCs for the EPA School Air Toxics Program (U.S.
EPA, 2009a, 2009b).
B. 3.5 Massachusetts Department of Environmental
Protection Indoor Air Sampling and
Evaluation Guide
This MADEP guidance (MADEP, 2002) discusses
applications of passive badges for up to 3 weeks using
a charcoal adsorbent. They indicate that "In general,
MADEP has found good agreement between results from
badges and pumped samples, and good precision?' They
discuss use of passive sampling as part of a "two-
component sampling period involving sub-chronic and chronic
sampling durations" with the passive samplers used for
the chronic sampling component.
B.4 Other Applications
Passive samplers have been applied not only to the
VOCs commonly measured by methods TO-15 and
TO-17 (U.S. EPA, 1999a, 1999b) at VI sites, but also
to a wider range of other volatile and semi-volatile
chemicals:
• formaldehyde and other aldehydes in indoor air
(EPA Method IP-6C; U.S. EPA, 1990; SKC,
2004)
• vapor-phase mercury in workplace
atmospheres, OSHA method ID-1405 and
longer duration sampling applications up to
months (Brumbaugh et al., 2000)
• pesticides (Jaward et al., 2004a), polycyclic
aromatic hydrocarbons (PAHs), and other
semi-volatiles (Jaward et al., 2004b, Bartkow et
al., 2004) in ambient air
• nicotine as a marker for tobacco smoking (see
http: / 7www.epa.gov/ncea/ets /pdfs /
etsch3.pdf)
• tracers used for determining air exchange rates
(Dietz and Cote, 1982; EPA Method IP-4A,
U.S. EPA, 1989)
Passive samplers have also been used to measure
VOC vapors in soil gas in both a semiquantitative
(EMFLUX, Petrex, Gore, and Beacon) and
quantitative (McAlary et al., 2014a, 2014b, 2014c)
fashion. These studies have tried to tie in the
relationship between mass and concentration
(Odencrantz and O'Neil, 2009) or calculate starvation
(Hodny et al., 2009). McAlary et al. (2014a, 2014b,
2014c) showed that the key is to use a sampler with
an uptake rate similar to or lower than the diffusive
supply rate of vapors to the face of the sampler so
that the sampler uptake is the rate-limiting step. In
cases where the uptake rate is not the rate-limiting
step, a low bias from the starvation effect is likely. A
detailed description of subsurface applications of
passive sampling technology is beyond the scope of
this paper, but some important aspects of using
passive samplers for soil gas are discussed in ESTCP
(in press).
5 http://www.osha.gov/dts/sltc/methods/inofganic/idl40/
idl40.html
Passive Sa
itions of Air Quality
-------
APPENDIX C:
ANALYTICAL LABORATORY QA/QC
Analytical test methods and performance
requirements for a variety of passive samplers and
VOCs are detailed in ASTM D6196-3 (2009b), EN
13528-2 (CEN, 2002a, 2002b), ISO 16017-2 (2003),
ISO 16200-2 (2000),7 EN 14662-4 (CEN, 2005a) and
EN 14662-5 (CEN, 2005b). These standards provide
a comprehensive overview of the sampling and
analytical performance requirements for passive
samplers; however, the laboratory may determine
specific analytical parameters, detailed protocols for
QC samples, and corresponding QC acceptance
criteria. Therefore, it is important to select a
laboratory that is knowledgeable and experienced
with passive sampling. This appendix provides a
comprehensive summary of analytical and laboratory
quality considerations unique to the analysis and data
interpretation of passive samplers for long-term
environmental measurements, including possible
enhancements to conventional industrial hygiene
methods to meet DQOs of vapor intrusion
investigations.
C.1 Selectivity—Analytical
Instrumentation: GC/FID vs. GC/MS
Although the ISO and EN standards do not specify
instrument configuration details, the U.S. industrial
hygiene methods for workplace monitoring using
passive charcoal samplers rely on an FID for
compound detection. These OSHA methods are
tailored for a single chemical or a limited set of
chemicals based on the solvents or materials used in
the workplace. In these circumstances, in which the
target compound list is limited with little concern for
interfering compounds present at similar levels, a
nonspecific detector like the FID meets the DQO.
However, when deploying charcoal samplers in
indoor or outdoor environments, the samplers are
exposed to a complex mixture of chemicals of which
the target compounds are typically present at low
concentrations. In this situation, the identification and
quantification of the target compounds benefit from
the selectivity of the mass spectrometer (MS).
Although the extraction solvent for charcoal
samplers, carbon disulfide, can interfere with the
detection of compounds during its elution, negative
impacts can be minimized through proper
configuration of the GC/MS parameters. In the case
of thermally desorbable sorbents, the MS is routinely
applied for environmental analysis following EPA
Method TO-17 (U.S. EPA, 1999b). The MS also
provides additional sample characterization by
allowing for tentative compound identification and
concentration estimation of uncalibrated peaks.
C.2 Measurement Range
The useful VOC concentration range measured with a
passive sorbent sampler is determined by both the
sorbent characteristics and the analytical method. The
upper limit is determined by the sorptive capacity of
the sorbent and the linear range of the analytical
instrumentation. Dilution of the sample can extend
the analytical measurement range. For charcoal-based
samples, the solvent extract can be diluted prior to
injection. In the case of thermally desorbed samplers,
dilution is achieved by adjusting instrument split
flows to reduce the mass injected onto the GC/MS.
The lower measurement limit is set by the blank levels
of VOCs from the sampler and/or preparation steps
(C.3) as well as the analytical sensitivity of the
laboratory equipment. Detector noise levels typically
define the lowest possible calibration level.
C.3 Method Background—Sorbent
Artifacts, Solvent Impurities, and
Common Lab Contaminants
The blank values originating from the sorbent
material and laboratory handling and preparation can
be significant for passive sorbent samplers, and it is
7 Although ISO 16200-2 is written for 8-hour workplace
monitoring, this standard provides detail on solvent-extracted
Passive Samplers for Investigations of Air Qualii
passive samplers, including detail on the various types
available and sampling rates.
39
-------
critical for the laboratory to minimize and monitor
these blank levels to ensure DQOs are met. Prior to
using any sorbent or solvent, the batch should be
verified as clean to the required levels. Carbon
disulfide, the extraction solvent routinely used for
charcoal extraction, commonly contains benzene as
an impurity. Although this solvent can be purchased
with a "low benzene" specification, each lot should be
verified prior to use. Additionally, the handling of
sorbents during preparation can also result in elevated
blank levels. To minimize exposure to lab
contaminants, the sorbent transfer from the sampler
or storage vial to a thermal desorption tube or
extraction vial should be conducted in a solvent-free
environment away from any source of VOCs.
C.4 Laboratory QC Elements
Development of laboratory methods requires defining
analytical measurement objectives, frequency, and
measurement performance criteria. NIOSH and
OSHA charcoal sorbent methods, EPA SW-846 8000
series, and EPA Method TO-17 are resources for
applicable quality elements and acceptance criteria for
passive sampler analysis. Laboratory SOPs should
address the minimum QC elements listed below
including not only the frequency and acceptance
criteria, but also corrective action for noncompliance
as well.
C.4.1 Initial Calibration
During initial calibration the analyst determines the
relationship between concentration and detector
response and defines the range in which relationship
is linear. Internal standard calibration techniques are
recommended for solvent extraction and thermal
desorption GC/MS techniques to adjust for
variations in the sample injection volume as well as
variations in the instrument response. Linearity can be
established by generating relative response factors at
each level and evaluating % relative standard
deviation (RSD) against an acceptance criterion (EPA
TO-17) or by evaluating the fit of the response curve
as described in the NIOSH and OSHA methods.
General guidelines for initial calibrations are described
in detail in EPA SW-846 Method 8000C (U.S. EPA,
2003).
C.4.2 Reporting Limit
The generic term reporting limit is used to denote the
lowest concentration that can be reliably measured by
the method. A detailed discussion of the technical
approaches to determine reporting limits is outside
the scope of this discussion and can be found
elsewhere (e.g., OSHA, 2008). Regardless of the
specific procedure used to establish the reporting
limit, in each case, the reporting limit must be
supported by the lowest concentration of the
calibration curve and also must take into account
method background factors (C.3). Artifacts from the
sorbent material, reagents, and preparation steps may
translate into a reporting limit several times higher
than the lowest calibration level.
C.4.3 Sample Preparation QC Samples
A laboratory blank and laboratory control spike (LCS)
should be prepared with each preparation batch to
monitor blank levels and recoveries. As part of the
initial method validation, desorption efficiencies
should be determined for each target compound.
Desorption efficiencies can be verified with each
preparation batch by preparing an LCS spiked with
representative target compounds. Details regarding
desorption efficiency studies and recommended
frequency are outlined in the ISO and EN standards
listed in Appendix B as well as in the NIOSH and
OSHA reference methods.
C.4.4 Sample Duplicates and/or LCS Duplicates
With sorbent samples, duplicate samples cannot easily
be prepared because the entire sorbent bed is
consumed in the preparation step. As such, sample
duplicates are typically a replicate injection of the
sample extract or analysis or the recollected thermal
desorption sample. Although these replicate injections
provide instrument precision, method precision that
measures both preparation and analysis steps can be
measured by preparing and analyzing two LCS
samples in each preparation batch.
40
>sive Samplers for Investigations of Air Quality
-------
C.4.5 Sample Preparation Surrogate Spikes
Although surrogates are not routinely described in
NIOSH or OSHA charcoal sorbent methods, the
addition of surrogates to the sorbent bed prior to the
extraction step provides verification of the extraction
efficiency and can identify factors affecting
measurement accuracy. For example, the commonly
used solvent for extraction, carbon disulfide, can
evaporate during extract storage, thereby
concentrating the sample and altering the target
compound concentration. In lieu of a surrogate or in
addition to a surrogate, an internal standard can be
added to the charcoal bed prior to extraction. This
internal standard procedure is described in the
Radiello user manual (Radiello, 2013) and EN 14662-
5 (CEN, 200 5b). Any change in extract volume due to
evaporation affects both the internal standard and
target compound concentration similarly, minimizing
any bias in the measured compound concentration.
C.5 Laboratory Identification and
Documentation of Analytical
Interferences
The identification and documentation of analytical
interferences can be critical to properly interpret the
reported concentrations for passive sorbent samplers.
Although efforts can be made in the project planning
stage to properly select samplers and parameters for
anticipated field conditions, the conditions of high
humidity and the presence of high VOC
concentrations should be noted in the report.
C. 5.1 Water (High Humidity)
High mass loading of water on sorbents can occur in
cases when samplers are exposed to prolonged
periods of high humidity. Strong sorbents such as
charcoal or molecular sieve resins are most
susceptible to adsorbing large amounts of water,
affecting both the analytical and sampler
performance. In terms of analytical performance,
excess water is evident in the solvent extraction step
as a separate layer in the carbon disulfide extract,
which can result in partitioning of polar compounds
to the water phase. In the case of thermal desorption,
excess moisture can affect both the chromatography
and the MS detector response. Retention times of the
early eluting peak may shift, the baseline of the total
ion chromatogram may be elevated in the initial stages
of the GC run, and the internal standard responses
may show low recovery.
To some degree, the laboratory can employ measures
to mitigate the effects of moisture during preparation,
thereby minimizing the impact on data quality. In the
case of charcoal sorbents, a desorption solvent
modifier such as dimethylformamide can be added to
carbon disulfide to dissolve the water layer (ISO,
2000). When samplers are prepared using thermal
desorption, the laboratory can use dry-purge
techniques to remove excess water prior to analysis
(ISO, 2003). To assist in the evaluation and
interpretation of the test results (Section 5.2.8), the
laboratory should document if the effects of water are
observed during sample preparation and analysis.
C.5.2 High Concentration Target or Non-target
VOC on Trace-Level Measurements
High mass loadings of VOCs measured on passive
sorbent samplers can affect both sampler
performance (C.8) and analytical reporting limits.
Diluting the sample extract or increasing the split of
the thermal desorption unit will keep target VOCs
within the calibration range and/or minimize the
impact of interfering non-target peaks on the
analytical system. Although, in theory, undiluted or
lesser dilutions can be analyzed to provide the lowest
reporting limits for each target VOC, in practice,
contamination of the analytical unit and subsequent
issues with carryover limit the effectiveness of this
approach. High mass loadings of target and non-
target VOCs measured on the samplers should be
documented in the test report.
C.6 Reporting Tentatively Identified
Compounds—Mass Adsorbed on
Sorbent Bed vs. Calculated Air
Concentrations
Using an MS for sample analysis allows the lab to
report tentative identifications and semiquantitative
Passive Samplers for Investigations of Air Qualii
41
-------
mass concentrations for uncalibrated, non-target
compounds. Although tentatively identified
compounds (TICs) can be important for site
characterization and also provide the data user direct
information regarding total mass adsorbed on the
sampler, calculating corresponding air concentrations
can be difficult. In the absence of published uptake
rates for TIC compounds, estimated rates can be
applied to calculate an approximate air concentration.
However, this would exacerbate the uncertainty in
TIC quantitation, which is already approximate
because of the use of generic response factors.
C.7 Concentration Correction
Calculations
C. 7.1 Adjustments for Field Conditions—
Temperature, Pressure, and Face Velocity
The diffusive uptake rate in the field can vary as a
function of site temperature and pressure. To report
mass concentrations at specified conditions, such as
normal ambient temperature and pressure, the
following equation can be used (ASTM, 2009b):
QT = Q298 x (T/298)1'5 x (101/P)
where
QT = uptake rate at the field temperature T (in
Kelvin)
Q298 = reference uptake rate at 25°C (298 K)
and 101 kPa
T = average field temperature in Kelvin
P = average barometric pressure at field site
(kPa)
The laboratory should document whether uptake
rates have been corrected for temperature and
pressure in the test report.
In addition to temperature and pressure adjustments,
conditions of low face velocity commonly
encountered in indoor air environments can require
correction of the sampling rate. In the case of SKC
badges, sampling rates have been determined in the
laboratory under face velocity conditions of <5 cm/s
to apply to samplers deployed in homes. These
indoor air rates are significantly lower than uptake
rates generated under the standard validation
conditions of 10 to 200 cm/s (SKC, 2008).
C.7.2 Blank Correction
The European and ISO standards, as well as the U.S
industrial hygiene methods, generally require
subtraction of the blank tube analyte mass from the
sample analyte mass to calculate the concentration in
the sampled air. Because blank subtraction is not
allowed in EPA environmental monitoring methods
as stated in EPA SW-846 8000C (U.S. EPA, 2003),
labs applying EPA method QC criteria to the analysis
of passive samplers may report sample concentrations
without blank correction. The test report should
clearly outline the procedure followed by the
laboratory.
C.8 Data Qualifiers
1. Denote Level of Validation Associated with
Uptake Rate—Estimated vs. Validated.
Several levels of evaluation of uptake rates are used to
determine sample analyte concentrations that are
outlined in EN 838 (CEN, 1995), MDHS 27 (MDHS,
1994), and NIOSH (Harper and Guild, 1996;
Cassinelli et al., 1987). Uptake rates calibrated by
controlled chamber tests or by an independent
validated method in the field are assigned a partial or
full evaluation level. If calibrated rates are not
available, sampling rates can be calculated using
published or estimated diffusion coefficients and the
geometric constant of the diffusive sampler. In the
case of permeation samplers, uptake rates can be
calculated using the retention time of the VOC on a
nonpolar phase column (Seethapathy and Gorecki,
201 Oa). If concentrations are determined using
calculated uptake rates, the associated results should
be flagged with an appropriate data qualifier.
42
>sive Samplers for Investigations of Air Quality
-------
2. Denote Sample Duration Exceeding
Manufacturer's Recommended Maximum
and/or Situations in Which the Uptake Rate
Times the Sample Duration Exceed the Safe
Sampling Volume for a Given Chemical
Weakly adsorbed chemicals can reverse diffusion
under prolonged exposure. If the manufacturer's
recommended maximum sampling duration is
exceeded for specific sampler/sorbent combination,
then data should be used with caution. In general, low
bias due to back-diffusion can only be confirmed and
quantified by comparing to field verification samples.
If a low bias is indicated, either the passive sampler
result is flagged as an estimated value or the uptake
rate is adjusted for the associated samples to account
for field-specific conditions. This adjustment should
be documented in the test report.
3. Flag Affected Data in Which Interferences
Result in Low Bias
Evidence of excess water adsorption on the sorbent
sampler noted by the laboratory and/or documented
in the field should alert the data user to potential low
bias for specific compounds depending on the
sorbent used and the sampler storage conditions. For
example, polar compounds such as acetone and
methyl ethyl ketone can exhibit low recovery from
charcoal sorbents exacerbated by sample storage at
room temperature (3M, 1996; Harper, 2000).
In addition, interferences noted by the laboratory and
also identified by inspection of the chromatogram can
result in low bias for weakly retained VOCs when
using non-ideal sorbents. Common indoor air
contaminants from household products and activities
such as limonene and pinene can compete for
sorption sites and displace more weakly adsorbed
target VOCs. Additionally, high concentrations of
target VOCs can saturate the sampler and exceed the
capacity of the media. This oversampling is most
common when using thermal desorbable sorbents due
to their relatively low surface area as compared with
charcoal-based sorbents. Comparison of the total
mass adsorbed to the sorptive capacity of the media
can assist in evaluating and interpreting the results.
Although suspect results can be flagged, field
verification samples are useful to quantify the impact
of the interference on the reported concentration.
Corrections for water and coexisting chemicals may
be possible if the passive samples collected for the
study exhibit similar interferences. Corrections are not
recommended if the sampler was overloaded and
exceeded the sorptive capacity. However, in many
such cases a flagging convention (e.g., "J" flags) can
be used to convey to the decision maker the likely
direction of any bias. For example, sampler
overloading will lead to underestimation of target
compounds, and in many cases this will not affect site
management because overloaded samples also tend to
be well above risk-based screening levels.
4. Flag Results in Which Measured Adsorbed
Mass is Less than a Defined Factor Above the
Field Blank
Detections of VOCs in the trip blank are not
unexpected when evaluating down to nanogram levels
on the thermal desorption sorbent samplers. The
general guideline listed in ISO 16017-2 and ASTM
D6196 is that the analyte peak in the field blank
should be less than 10% of the target peak area
measured in the sample. If this criterion is not met,
the associated sample data should be flagged to
indicate high bias in the test report.
Passive Samplers for Investigations of Air Qualii
43
-------
United States
Environmental Protection
Agency
Office of Research and Development
National Risk Management
Research Laboratory
Cincinnati, OH 45268
Official Business
Penalty for Private Use
$300
EPA/600/R-14/434
December 2014
www.epa.gov
PRESORTED STANDARD
POSTAGE & FEES PAID
EPA
PERMIT No. G-35
Recycled/Recyclable
Printed wilh vfge'aHste-based ink on
paper that omtans a minimum of
50% poa-consuffl « liber content
processed ehlonne free
------- |