Volume 22

Number 10

Environmental
October 2020
Pages 1919–2134

Science
Processes & Impacts
rsc.li/espi

ISSN 2050-7887

CRITICAL REVIEW
Frank Wania and Chubashini Shunthirasingham
Passive air sampling for semi-volatile organic chemicals
Environmental
Science
Processes & Impacts

CRITICAL REVIEW

Passive air sampling for semi-volatile organic

Cite this: Environ. Sci.: Processes
chemicals
Impacts, 2020, 22, 1925
Frank Wania * and Chubashini Shunthirasingham †

During passive air sampling, the amount of a chemical taken up in a sorbent from the air without the help of
a pump is quantified and converted into an air concentration. In an equilibrium sampler, this conversion
requires a thermodynamic parameter, the equilibrium sorption coefficient between gas-phase and
sorbent. In a kinetic sampler, a time-averaged air concentration is obtained using a sampling rate, which
is a kinetic parameter. Design requirements for kinetic and equilibrium sampling conflict with each other.
The volatility of semi-volatile organic compounds (SVOCs) varies over five orders of magnitude, which
implies that passive air samplers are inevitably kinetic samplers for less volatile SVOCs and equilibrium
samplers for more volatile SVOCs. Therefore, most currently used passive sampler designs for SVOCs are
a compromise that requires the consideration of both a thermodynamic and a kinetic parameter. Their
quantitative interpretation depends on assumptions that are rarely fulfilled, and on input parameters, that
are often only known with high uncertainty. Kinetic passive air sampling for SVOCs is also challenging
because their typically very low atmospheric concentrations necessitate relatively high sampling rates
that can only be achieved without the use of diffusive barriers. This in turn renders sampling rates
dependent on wind conditions and therefore highly variable. Despite the overall high uncertainty arising
from these challenges, passive air samplers for SVOCs have valuable roles to play in recording (i) spatial
concentration variability at scales ranging from a few centimeters to tens of thousands of kilometers, (ii)
long-term trends, (iii) air contamination in remote and inaccessible locations and (iv) indoor inhalation
exposure. Going forward, thermal desorption of sorbents may lower the detection limits for some
SVOCs to an extent that the use of diffusive barriers in the kinetic sampling of SVOCs becomes feasible,
Received 28th April 2020
Accepted 20th July 2020
which is a prerequisite to decreasing the uncertainty of sampling rates. If the thermally stable sorbent
additionally has a high sorptive capacity, it may be possible to design true kinetic samplers for most
DOI: 10.1039/d0em00194e
SVOCs. In the meantime, the passive air sampling community would benefit from being more
rsc.li/espi transparent by rigorously quantifying and explicitly reporting uncertainty.

Environmental signicance
Over the past 25 years, numerous passive air samplers (PASs) for SVOCs have been introduced, characterised, and tested. More recently, the applications of PASs,
initially focused on recording the variability in atmospheric contamination on a wide range of spatial scales, has expanded to include measuring indoor
inhalation exposure to SVOCs, probing the exchange of SVOCs between the atmosphere and soil and water, mapping the toxicity of airborne contamination, and
monitoring interannual trends in SVOC air concentrations. The simplicity of passive air sampling is deceptive and many users of PASs do not fully appreciate
their strength and limitations and may have misconceptions as to their applicability and reliability. This comprehensive and critical review assembles and
curates the current knowledge on this topic in order to equip anyone to use PAS appropriately and with condence and to guide further development of PASs in
a direction that will overcome their largest shortcomings.

A. Introduction
A.1 Denition of passive air sampling and semi-volatile
organic compounds
Passive air sampling involves the diffusive uptake of a chemical
Department of Physical and Environmental Sciences, University of Toronto vapour in a sorbent over time. The amount of vapour being
Scarborough, 1265 Military Trail, Toronto, Ontario, M1C 1A4, Canada. E-mail: sorbed during a sampler's deployment is determined by both
[email protected]; Tel: +1-416-287-7225
kinetic and thermodynamic factors and can be interpreted in
† Current affiliation and address: Air Quality Processes Research Section,
terms of volumetric air concentrations based on a quantitative
Environment and Climate Change Canada, 4905 Dufferin St, Ontario, M3H 5T4,
Canada. understanding of these factors. In contrast to active air

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1925
Environmental Science: Processes & Impacts Critical Review

samplers that rely on a pump to pass a volume of air through or

past a sorbent, passive air samplers (PASs) have the advantage
of low price, simple operation, and independence of power
sources. On the other hand, they typically allow only for coarse
temporal resolution. PASs have been developed for a variety of
gaseous chemicals, including volatile organic chemicals (VOCs)
such as benzene, toluene, ethylbenzene, and xylene, classical air
pollutants such as sulfur dioxide, various nitrogen oxides,
hydrochloric acid, ozone and ammonia, as well as mercury.1–5
A distinct set of PASs has been developed for semi-volatile
organic chemicals (SVOCs). The term SVOCs comprises
organic molecules that can occur to a signicant extent in both
the gas-phase and condensed phases, which corresponds to the
vapour pressure range of approx. 10
1 to 10
6 Pa. The group of
SVOCs comprises a large number of commercially produced
substances, including industrial chemicals, pesticides, and
additives to consumer products. Prominent examples are the
polychlorinated biphenyls (PCBs), polybrominated diphenyl
ethers (PBDEs), and organochlorine pesticides (OCPs). Fig. 1Number of publications and citations on the topic of passive air
Combustion products such as the polycyclic aromatic hydro- sampling for semivolatile organic compounds, based on an analysis
using Web of Science using the search string TOPIC: (“passive air
carbons (PAHs) also count among the SVOCs. Many SVOCs are sampl*”) and TOPIC: (POPs or SVOCs or PCBs or PAHs or semivolatile
a concern for human and environmental health. The atmo- or semi-volatile) on April 11, 2020.
sphere oen plays an important role in the dispersal of SVOCs
and in facilitating organism exposure to SVOCs. PASs for SVOCs
thus address a need for information on the concentrations of (air) sampling in general.6–11 In other cases, the scope is limited
SVOCs in indoor and outdoor air. to only one type of PAS.12,13 Recently, Salim and Górecki14
thoroughly reviewed the theory and modelling of passive
A.2 Motivation for, and scope of, review sampling, which complements our current effort well, as we do
Over the past 20 years, there has been explosive growth in the not delve as thoroughly into this aspect in this review.
development, characterization, testing and application of While the operation of PASs for SVOCs can be deceptively
passive air sampling techniques of SVOCs. A literature search simple, the kinetic and thermodynamic factors controlling the
suggests that more than 40 papers are published in this area uptake of SVOCs in PASs can be surprisingly complex and are
every year, garnering ca. 1500 citations annually (Fig. 1). not always fully understood, even by the community using
There have been a number of reviews on the passive them. Reasons include the diversity of sampler designs, the
sampling of atmospheric SVOCs, but they are either somewhat large and diverse group of SVOCs being sampled, and the wide
out-of-date and/or have a wider scope by focusing on passive range of environments in which PASs are nding use. Here we

Frank Wania studied Environ- Chubashini Shunthirasingham

mental Science at the University is a chemist in the Air Quality
Bayreuth in Germany and received Research Division at Environ-
his Doctorate in Chemical Engi- ment and Climate Change Can-
neering and Applied Chemistry ada. She obtained her BSc in
from the University of Toronto in Chemistry and Biology as well as
1995. Aer two years as a scientist her MSc and PhD in Environ-
at the Norwegian Institute for Air mental Chemistry from the
Research, and three years as an University of Toronto. Her PhD
independent researcher, he joined work focused on the measure-
the University of Toronto Scar- ment of pesticides in the physical
borough in 1999, where he is environment of developing
currently a professor of environ- countries using XAD-resin based
mental chemistry. A co-developer of the XAD-PAS, the FTS and passive air samplers and on the measurement of Henry's law
a PAS for gaseous elemental mercury, he has been calibrating, constants using the inert gas stripping method. Her research
characterising, evaluating and applying passive air sampling interests are atmospheric chemistry, fate and transport of SVOCs
techniques for the past two decades. and method development and validation for emerging organic
pollutants using gas chromatography mass spectrometry.

1926 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

aim to summarise comprehensively the current state-of- DA

ku ¼ (2)
knowledge on passive sampling of SVOCs and to provide guid- DzVS
ance on the use of PASs for SVOCs to existing and prospective
where D is the molecular diffusivity of the target analyte in the
users.
gas-phase D in m2 per day and A is the geometric surface area of
Section B provides an introduction into the basic principles
the PAS sorbent in m2. The rate constant for loss is additionally
of passive air sampling and in doing so identies the two
proportional to the reciprocal of the equilibrium sorption
central characteristics of a PAS. These are the PAS's uptake
coefficients between the gas-phase and the PAS sorbent KSG in
capacity at equilibrium, i.e. the maximum amount of a target
units of m3 gas per m3 absorbent (or m3 gas per m2 adsorbent):
chemical that the sorbent will take up given a certain concen-
tration in the gas-phase, and the PAS's inherent sampling rate, DA
kl ¼ (3)
i.e. the maximum rate at which a target chemical is taken up in DzVS KSG
the sorbent. Section C then highlights the specic challenges
faced when developing a PAS for SVOCs, namely the balancing The amount of analyte in the PAS sorbent, DmS in mol, taken
of conicting needs (i) for high vs. tightly controlled sampling up during its deployment time, Dt in days, then becomes:20
rates and (ii) of SVOCs of variable volatility. Section D intro-    
DmS DA CS CS
duces, contrasts and critically assesses the approximately 50 ¼ CG
¼ SR CG
(4)
PAS designs for SVOCs that have been presented in the peer- Dt Dz KSG KSG
reviewed literature over the past 25 years. Section E discusses In eqn (4), the ratio CS/KSG designates the gas-phase concen-
the empirical and theoretical means by which the uptake tration in equilibrium with the concentration in the PAS
capacity of different PAS sorbents has been determined. Section sorbent. The term (CG
CS/KSG) is the concentration difference
F presents information on how the kinetics of uptake in PASs is that provides the driving force for the diffusive transport from
approached experimentally and theoretically. Section G reviews bulk air to PAS sorbent. The term DA/Dz designates how fast this
the various ways in which PASs for SVOCs have been applied. diffusive transport can take place and is oen called a sampling
Finally, Section H will conclude with an overall assessment of rate SR in m3 per day. The ratio D/Dz is sometimes termed
the state-of-the-art in passive air sampling of SVOCs and a diffusive mass transfer coefficient in units of m per day. A
provide an outlook for future activities in this area. Section I PAS's foremost characteristics are thus described in one kinetic
provides a glossary for acronyms, abbreviations and variables parameter (SR) and one thermodynamic parameter (KSG)
used in this review. describing the rate of uptake and the maximum uptake capacity
Not covered in this review are substances other than SVOCs, of the PAS, respectively.
namely VOCs and inorganic gases or passive samplers for media
other than air. We also do not consider biological matrices,
such as plants (moss, needles, tree bark, wood, etc.), that are B.2 The uptake curve: kinetic vs. equilibrium sampling
oen used in a similar way as PASs for SVOCs.15 Furthermore, A plot of the change in the amount of a chemical sorbed to a PAS
we do not count organic “lms” forming on glass surfaces with increasing length of deployment is called an uptake curve. A
among the passive air samplers in this review.16–19 generic example of such a curve is shown in Fig. 2. In this curve,
the sorbed amount is divided by CG in order to eliminate the
variability caused by changing analyte concentrations in the
atmosphere. The ratio mS/CG, which has units of m3, is some-
B. Principles of passive air sampling times referred to as the equivalent sampling volume. The
B.1 The equation describing diffusive uptake in a PAS uptake curve goes through three distinct phases. The so-called
The amount of a target analyte in the passive sampling sorbent, linear uptake phase, when the rate of uptake is kinetically
mS in mol, changes over time, t in days, as a chemical is taken controlled, a transitional curvi-linear uptake phase, and the
up from the atmosphere at a rate proportional to the concen- equilibrium phase when the gas-phase and the PAS sorbent have
tration in the gas-phase, CG in mol m
3, and is lost from the reached a state of chemical equilibrium and therefore the sor-
sorbent at a rate proportional to the concentration in the bed amount is thermodynamically controlled. The initial slope
sorbent, CS in mol m
3 (or mol m
2 if it is an adsorbent):20 of the uptake curve in Fig. 2 corresponds to the sampling rate
SR, whereas the maximum amount taken up depends on the
dmS
¼ VS ðku CG
kl CS Þ (1) uptake capacity, which is the product of the equilibrium sorp-
dt
tion coefficient between sorbent and gas-phase KSG and the size
where ku and kl are rate constants for uptake and loss in per day of the passive sampling sorbent (VS or AS).
and VS is the volume of the PAS sorbent in m3. If chemical B.2.1 Kinetically controlled uptake. Initially, the sorbent is
uptake and loss in a PAS is controlled by the air-side resistance, clean, i.e. CS is very small. If the sorbent has a high uptake
i.e. transport within the sorbent is not rate-limiting, the rate capacity, KSG is large. Then, CS/KSG  CG and (CG
CS/KSG)  CG
constants for uptake can be derived by applying Fick's rst law and eqn (4) simplies to:
to the molecular diffusion of the target analyte through a stag-
DmS DA
nant air boundary layer surrounding the sorbent of thickness Dz ¼ CG ¼ SR  CG (5)
Dt Dz
in m:20

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1927
Environmental Science: Processes & Impacts Critical Review

precise value of these times depends on the acceptable devia-

tions from linearity and equilibrium. For example, uptake may
be considered linear as long as CS/KSG is smaller than 25% of CG
and equilibrium may be considered reached when CS/KSG
exceeds 90% of CG (Fig. 2 and 3).
There are two types of PASs. A kinetic sampler seeks to
remain in the kinetically controlled uptake phase during the
entire length of deployment, i.e. tlinear is the maximum
deployment length of a kinetic sampler. A kinetic sampler yields
a gas-phase concentration averaged over the time of deploy-
ment by using a rearrangement of eqn (5):
DmS
CG ¼ (7)
SR  Dt

An equilibrium sampler, on the other hand, seeks to reach

Fig. 2 Generic illustration of the uptake of an SVOC in a passive air
sampler, showing the increase in the sorbed amount mS divided by the equilibrium, whereupon the gas-phase concentration at the
atmospheric concentration CG as a function of deployment time t. The time of retrieval is obtained using a version of eqn (6):
time period of linear uptake tlinear and the time to equilibrium tequilibrium mS
are important characteristics of a passive sampler. The displayed curve CG ¼ (8A)
VS KSG
was calculated with PAS-SIM21 for a compound with log KSG of 6 in
a PUF-PAS and an SR of 3.2 m3 per day. A tequilibrium of 246 days is mS
based on CS/KSG > 0.90CG and a tlinear of 24 days is based on CS/KSG < CG ¼ (8B)
AS KSG
0.25CG. See Section F.5 for more information on the PAS-SIM model.

The tequilibrium is the minimum deployment length of an

The amount taken up in the sampler DmS is then simply the equilibrium sampler.
product of the sampling rate SR, the gas-phase concentration Samplers that are neither kinetic nor equilibrium samplers,
CG and the deployment length Dt. need to derive CG from eqn (4), which is challenging as CG and
B.2.2 Thermodynamically controlled uptake. Once equi- KSG generally vary with time (see Section F.3).
librium is established between sorbent and gas-phase, CS/KSG ¼
CG and (CG
CS/KSG) ¼ 0. If that is the case, eqn (4) yields Dm/Dt
¼ 0, i.e. there is no more net uptake of chemical on the PAS B.3 Design considerations for an equilibrium sampler
sorbent. The amount of chemical in the sorbent then is: Because the establishment of equilibrium is essential to the
functioning of an equilibrium PAS, optimizing the design of
mS ¼ KSGVSCG (6A) such a PAS involves minimizing the time to equilibrium. There
are two means to shorten tequilibrium, namely maximizing the
if the sorbent is an absorbent, i.e. CS ¼ Dm/VS, or
sampling rate SR and minimizing the uptake capacity KSGVS (or
KSGAS) (Fig. 3).
mS ¼ KSGASCG (6B)
A large SR can be achieved by a large A and a small Dz, whereas
if the sorbent is an adsorbent, i.e. CS ¼ Dm/AS. It is important to a small sorption coefficient KSG and a small size of the sorbent VS
stress that AS here refers to the total surface area of the adsor- (or AS) contribute to a small uptake capacity. A small Dz is achieved
bent and not the geometric surface area of the sorbent or of the by not sheltering the sorbent from wind or other air turbulence. If
container holding the sorbent, i.e. one should not confuse A an absorbent is used, it should have a large A and a small VS, i.e.
with AS. The terms KSGVS and KSGAS are expressions of the one should seek to maximise its surface area to volume ratio, e.g.
maximum uptake capacity of a PAS, both with units of m3 of air. by using very thin sheets or lms of the absorbent. If an adsorbent
The use of an air volume to describe the capacity of a sorbent for is used, it should have a large geometric surface area A, but a small
a chemical can be understood if we imagine this to be the total surface area available for adsorption AS.
volume of air that contains the same amount of a chemical at There are, however, important limitations to how small the
equilibrium as the sorbent within the PAS. uptake capacity of an equilibrium sampler could be. The rst is
Aer CS/KSG is no longer negligibly small relative to CG and that the sorbent needs to take up an amount of the target
equilibrium is not yet reached, uptake in the PAS is in the curvi- analyte that is sufficient for reliable quantication. If the KSGVS
linear transition region when both kinetic and thermodynamic is too small, the amount mS may be below the limit of detection
factors govern the rate of uptake. (LOD) of the chosen quantication technique. The second is
Two characteristic times can be used to describe a particular that the target chemical may be lost too easily by volatilization
sampler–analyte combination, namely the length of linear from a PAS with too small an uptake capacity, because then the
uptake (tlinear) and the time to equilibrium (tequilibrium). The chemicals can be lost during retrieval, transport, storage and
the processing of the sorbent prior to analysis.

1928 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

Fig. 3 The length of the linear uptake phase, tlinear, and the time to equilibrium, tequilibrium, depend both on kinetic and thermodynamic factors.
Uptake curves for two PASs with equal sampling rate (SR of 3.2 m3 per day), but divergent uptake capacity (log KSG of 5.85 (blue) vs. 6.15 (yellow))
(A) and two PASs with equal uptake capacity (log KSG of 6), yet different uptake kinetics (SR of 2.0 (green) vs. 7.6 m3 per day (red)) (B). tlinear and
tequilibrium increase with increasing uptake capacity (17 to 34 days, 175 to 347 days) and decreasing uptake rate (13 to 36 days, 150 to 345 days). The
displayed curves were calculated with PAS-SIM21 using the parameterization for the PUF-PAS. See Section F.5 for more information on the PAS-
SIM model.

The accuracy of the CG derived from an equilibrium sampler capacity, i.e. a large KSG. The actual size of the KSG does not need
depends on the accuracy of the knowledge of the uptake to be known, as long as it is known to be large enough to assure
capacity, in particular the KSG. The sampling rate SR does not negligible loss of target analyte from the sorbent during
need to be known, except that it should be known to be large. deployment.
The strong dependence of KSG on temperature is one of the The accuracy of the CG derived from a kinetic sampler
challenges of using equilibrium samplers for ambient applica- depends on the accuracy of the knowledge of SR, which is DA/
tions, because the uptake capacity of the PAS is changing Dz. The molecular diffusion coefficient D of the target analytes
substantively with changes in ambient temperature. It is in the gas-phase is typically well established or can be easily
necessary to know the temperature dependence of KSG quanti- estimated through relationships between D and molecular size.
tatively and the temperature of equilibration in order to obtain Also, the dependence of D on atmospheric pressure and
reliable CG values from an equilibrium PAS. temperature can be estimated with good accuracy. The diffusive
Temperatures and SVOC air concentrations in the atmosphere area A can generally also be derived easily from the geometric
are highly variable on a number of time scales. As a consequence, dimensions of the sampler.
the amount of an SVOC on a sampler that is in equilibrium with The most challenging aspect of characterizing the SR of
the atmosphere is changing rapidly. In other words, chemical a kinetic sampler is therefore the quantication of the diffusive
equilibrium is a constantly moving target, e.g. being different in distance Dz, because Dz depends on the thickness of the stag-
a cold night than during a warmer day. To illustrate this effect, nant boundary layer surrounding the PAS sorbent and this
Fig. 4 shows simulated uptake curves in a PAS under the hypo-
thetical assumption that air concentrations alternate from high to
low every other week. It demonstrates that the more volatile
chemicals (low log KSG), which are suited for equilibrium sampling
because of their short tequilibrium, are the ones that experience the
most pronounced uctuations in the amount sequestered in the
PAS. For such chemicals, the amount sorbed in the PAS at any
moment in time is reective of the conditions immediately prior to
a sampler's retrieval, but also inuenced to some extent by the
exposure history. This suggests that equilibrium sampling for
SVOCs is likely only suitable for situations when both T and CG are
reasonably constant, which may apply for certain compounds in
indoor environments.

B.4 Design considerations for a kinetic sampler

Fig. 4 Illustrative simulations of uptake curves of three hypothetical
Because the condition CS/KSG  CG is essential to the func- chemicals in response to fluctuating air concentrations CG. Fluctua-
tioning of a kinetic sampler, optimizing the design of a kinetic tions in the amount of a chemical sequestered in a PAS increase with
PAS involves choosing a sorbent with a very high sorptive increasing volatility, i.e. decreasing log KSG.

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1929
Environmental Science: Processes & Impacts Critical Review

thickness is (i) difficult to estimate theoretically and (ii) is difference in the diffusive pathlengths under different wind
strongly inuenced by atmospheric turbulence, i.e. is suscep- conditions will be reduced, or – in other words – the wind speed
tible to being dependent on wind speed. As will be discussed dependence of the SR will be smaller.
next, there are strategies that seek to reduce this dependence of The second strategy involves the addition of a diffusive
Dz on wind speed, but they tend to lower the SR. This needs to barrier that forces the analyte to diffuse a longer distance to
be balanced with the need to sample an amount of chemical reach the sorbent (panel C in Fig. 5). The stagnant boundary
during the deployment period that is sufficient for reliable layer now is no longer adjacent to the sorbent, but next to the
quantication. diffusive barrier. The thickness of that boundary layer
B.4.1 Strategies to reduce the wind speed dependence of continues to be variable with wind speed, but the relative
the sampling rate. The SR of PASs is dependent on wind speed, difference in the total diffusion pathlength, which is the sum of
which can introduce considerable uncertainty in the derived air the thickness of the stagnant layer and the effective diffusion
concentrations. Mechanistically, we can conceptualise this distance through the barrier, become much less variable
wind speed dependence as the effect of atmospheric turbulence between high and low wind speed conditions. Again, the effect
on the thickness of the stagnant boundary layer surrounding is a reduced wind speed dependence of the SR. Obviously, the
the sampler sorbent (panel A in Fig. 5). Under low wind thicker the diffusive barrier, the smaller the difference in the
conditions, this layer is thicker and therefore the diffusive relative diffusion pathlengths and therefore the smaller the
pathlength that the analyte has to travel to reach the sorbent is wind speed dependence of the SR (compare panels C and E in
longer than under high wind conditions. As shown above, in Fig. 5).
PASs, where the SR is controlled by the air-side resistance, the It is of course possible to deploy both strategies at the same
SR is inversely proportional to the diffusive pathlength Dz. In time, i.e. place a sampler with a diffusive barrier within a wind
the design of PASs, two strategies are employed to reduce the shelter. However, depending on the thickness of the boundary
wind speed dependence of SR. layer relative to the effective diffusion distance within the
The rst strategy is wind sheltering that is accomplished by barrier, the further reduction in the wind speed dependence
placing the sorbent in some sort of housing that prevents the afforded by a wind shelter may be marginal and is generally
wind from directly blowing onto the sorbent (panel B in Fig. 5). much smaller than the effect of sheltering a sampler without
At low wind conditions this shelter will have little effect on the a diffusive barrier (compare panels C and D and E and F in
diffusive pathlength, but at high wind speeds the calming effect Fig. 5).
of the shelter will result in a thicker stagnant layer, a longer Fig. 5 seeks to illustrate the effect of these two strategies by
diffusive distance and thus a lower SR. More importantly, the depicting the diffusion distance of an analyte from the bulk

Fig. 5 Illustration of the effect of different sampler configurations on the diffusion distance of an analyte from the ambient atmosphere to the
passive sampling sorbent under high (Dzhigh) and low wind conditions (Dzlow). In particular, the effect of a wind shelter (A vs. B, C vs. D, E vs. F) and
the effect of a thin or a thick diffusive barrier (A and B vs. C and D vs., E and F) is shown. The ratio Dzlow : Dzhigh indicates the extent to which the SR
is influenced by wind speed.

1930 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

atmosphere to the PAS sorbent both at high (Dzhigh) and low A fundamental challenge of a PAS for SVOCs is therefore to
wind speed (Dzlow) conditions. The quotient Dzlow/Dzhigh nd the appropriate balance between an SR that is sufficiently
expresses by how much the diffusion distance and therefore the high to sample chemical amounts that can be reliably quanti-
SR is dependent on wind. The numbers for Dz assigned to ed, yet is also reasonably stable and well dened to allow for
different PAS congurations in Fig. 5 are hypothetical, but a quantitative interpretation of the sampled amounts.
plausible. The schematic of different sampler conguration is Whether a PAS should have a high, but variable SR or a low,
based on a sampler design using a sorbent in cylindrical form, tightly controlled SR depends to some extent on the magnitude
a radial diffusion barrier (such as the Radiello), and a jar as and the spatial variability of the air concentrations of the target
a wind shelter, and is inspired by two passive samplers for chemical. A compound with low, but spatially highly variable air
gaseous mercury22,23 (panels D and E, respectively). concentrations is better sampled with a PAS that samples a lot
Reducing the wind speed dependence of the SR of a PAS of air (has a high SR), even if the volume sampled is quite
comes at a price, namely the reduction of the SR. This is less uncertain. This makes sure that the sampled amounts are above
pronounced in the wind shelter strategy, as it mostly reduces LOD, yet it is still possible to compare PAS-derived concentra-
the SR at high wind speeds, but barely affects the SR under low tions at different locations. On the other hand, a compound
wind conditions. It is more notable with the application of with spatially fairly uniform air concentrations demands a PAS
diffusive barriers, as they increase the diffusion distance under with a well-controlled SR, because only then can differences in
any wind speed regime. Also, the thicker the barrier and the sampled amounts be interpreted as differences in volu-
therefore the more effective the reduction in wind speed metric air concentrations.
dependence, the lower the SR becomes. Therefore, eliminating The concentrations of most SVOCs in the atmosphere are
the effect of wind speed always needs to be balanced with the low and declining. Also, concentration differences of most
desire to sample sufficient amounts of an analyte for reliable SVOCs in the atmosphere are quite large, oen ranging over
quantication. many orders of magnitude. As a result, most PASs for SVOCs
were designed to have high SRs, even if it meant that the SRs are
quite variable and therefore uncertain. For example, the widely
C. The major challenges of passive air used PAS based on polyurethane foam (PUF) disks has an SR
sampling for SVOCs around 4 m3 per day (range: 0.5 to 15 m3 per day),24,25 whereas
the PAS using XAD-resin as a sorbent has an SR around 0.5 to 1
There are a number of reasons why passive air sampling for m3 per day (range 0.4 to 5.5 m3 per day).26–28
SVOCs is uniquely challenging. On the other hand, some SVOCs, such as hexa-
chlorobenzene (HCB) and hexchlorobutadiene (HCBD), have
very long atmospheric residence times and therefore fairly
C.1 Opposing demands for high vs. tightly controlled uptake uniform concentrations in the atmosphere.29–31 Unless
rates a sampling site is in the immediate vicinity of a strong source,
Concentrations of individual SVOCs in the atmosphere tend to concentrations in the global atmosphere may only vary within
be very low, oen in the range of 10
12 g m
3 (i.e. picograms per less than an order of magnitude. A PAS would need a tightly
m3) or even less. Even though the detection limits of modern controlled SR to distinguish such small concentration differ-
analytical methods for SVOCs are extremely low, during active ences with condence. While a PAS for SVOCs with a suffi-
air sampling for SVOCs, sampling volumes of several hundred ciently precise SR may not presently exist, an existing PAS for
m3 or even several thousand m3 are commonly used to allow for gaseous elemental mercury22 demonstrates that it is feasible in
reliable detection and quantication, especially in remote principle for a PAS to elucidate small concentration differ-
regions with background concentrations. Such high sampling ences. Its SR is low relative to those of commonly used PAS for
volumes are not attainable with classical PASs with typical SRs SVOCs (0.135 m3 per day), but this is of little concern as the
on the order of 0.1 m3 per day or less. In most cases, SRs on the analytical technique for quantication is powerful enough to
order of 1 m3 per day or higher are required to collect sufficient detect the analyte even aer sampling over as little as a week at
chemical for quantication during reasonable deployment global background concentrations. One of the main reasons
periods (months). for this is that the entire sampled amount is analyzed (by
Kinetic PASs for substances other than SVOCs, whether combustion of the entire sorbent present in the PAS). In the
batch, tube or radial diffusion samplers, have largely relied on typical PAS for SVOCs, the sorbent is solvent extracted and
diffusive barriers to standardise diffusion distances, and thus only a small amount of the sampled compound is injected
control SRs (see Section B.4). PASs for SVOCs, on the other onto the chromatographic column. Interestingly, SVOCs such
hand, tend not to include diffusion barriers in their design. The as HCB and HCBD, also tend to have relatively high ambient
main reason is that the uptake rates of PAS designs that gas-phase concentrations, precisely because of their long
incorporate diffusion barriers are too low to be compatible with atmospheric residence time. If an analytical method were to be
the very low trace concentrations of SVOCs and the detection used that delivers all of the sorbed amount onto the analytical
limits that can be achieved. Another reason is that the diffusive column (e.g. thermal desorption), even PASs with diffusive
barrier itself may act as sorbents for SVOCs and thus interfere barrier may be able to take up amounts of HCB and HCBD
with sampling. above the LOD.

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1931
Environmental Science: Processes & Impacts Critical Review

C.2 Large diversity of SVOCs vs. limited applicability range had found common use in active air sampling for SVOCs. One
of PASs in terms of volatility attractive feature of adopting a passive water sampler design for
passive air sampling is the possibility of sampling air and water
The term SVOCs comprises a wide variety of compounds whose
with the same device and the prospect of recording the air–
volatility spans several orders of magnitude. For example, the
water equilibrium status. However, caution is necessary to
vapour pressure of a 3-ring PAH is approx. 4 orders of magni-
appreciate the different requirements for air and water
tude higher than that of a 5-ring PAH.32 PCB congeners of
sampling; for example, while a water sampler can be expected to
different degree of chlorination cover a similar volatility range.33
be exposed to a relatively small range of temperatures during
Volatility is, of course, also strongly governed by the tempera-
a deployment period, PASs may be deployed at sites with hugely
ture of deployment. Because the uptake capacity of a sampler is
different ambient temperatures that can change rapidly on the
governed to a large extent by compound volatility, this means
time scale of deployment. Also, because the diffusivity of SVOCs
that a sampler will have widely divergent uptake capacities for
in water is so much slower than in the gas-phase and the
different SVOCs and even the uptake capacity for the same
thickness of boundary layers in air and water varies consider-
compound will be different at different deployment tempera-
ably, different factors may be controlling uptake from water
tures. One consequence is that the characteristic times, tlinear
than from air. Most of the PASs that are adopting sorbents from
and tequilibrium, can range over several orders of magnitude for
water sampling rely on non-porous materials such as polymeric
different SVOCs and for the same SVOC at different ambient
sheets and lms, whereas PASs using sorbents adopted from
temperatures.
active sampling tend to rely on porous materials such as poly-
For example, for the PUF-PAS, the simulation model PAS-
mer foam and resins.
SIM21 estimates tlinear to decrease from 8 months to 8 days,
when the log KPUF–G is decreased from 7 to 5.5. KPUF–G changes
by approximately 1.5 log units between a two-ring and a three- D.1 Samplers involving sorbents enclosed within low-
ring PAHs or for one compound if temperature is changed by density polyethylene
30 K. It is obvious that (i) equilibrium sampling for less volatile
D.1.1 Semipermeable membrane devices (SPMDs). The
SVOCs is not feasible because of excessively long times to
rst PAS for SVOCs that adopted a passive sampler for water
equilibrium, and (ii) it would be difficult to nd a deployment
relied on semipermeable membrane devices (SPMDs), which
period suitable for a range of SVOCs with divergent volatility,
consist of thin-walled low-density polyethylene (LDPE) lay-at
because even within the group of relatively volatile SVOCs,
tubing containing a thin lm of the neutral synthetic lipid tri-
tequilibrium will range over orders of magnitude.
olein.34 Most of the early studies using SPMDs as PASs were
However, also kinetic passive sampling is only applicable to
interested in sampling both air and water.35–37 Ockenden et al.
substances within a certain range of volatility. A chemical that is
calibrated SPMDs as PASs for PCBs38,39 and then applied them to
too volatile will approach equilibrium too quickly and therefore
have a very short tlinear and thus maximum deployment time. A repeatedly record a latitudinal gradient in atmospheric
chemical that is too involatile will be sorbed to particles in the contamination in Western Europe.40–45 Söderström et al.46,47
atmosphere and therefore not be available for uptake in the PAS were other early adopters and explored the wind speed depen-
in gaseous form (Section F.6 will discuss the uptake of particle- dence of uptake in SPMD-based PASs47,48 and were the rst to
sorbed SVOCs in PAS.). The range of applicability is dependent use SPMDs for indoor deployments.49
on the sampler design (uptake capacity and kinetics) and also Cranor et al.50 used a laboratory set-up to determine both the
uptake kinetics and the uptake capacity of SPMDs for 48
environmental conditions (e.g. atmospheric particle concen-
compounds including several PAHs, PBDEs, OCPs, and current-
trations), but spans approx. 4 orders of magnitude in volatility.
Consequently, it will be impossible to design a kinetic PAS and use pesticides (CUPs). Cicenaite et al.51 also used laboratory
to use a common set of deployment characteristics (e.g. experiments to determine these parameters for three
deployment length) that is applicable to all SVOCs all the time. compounds (naphthalene, 2-chlorophenol, 1,4-dichloroben-
Unless one is willing to deploy a range of PASs tailored to zene) as a function of temperature in the range from
16 to
different volatility ranges, the ambition therefore should be to +40  C.
design a sampler that is applicable to as many SVOCs as D.1.2 PASs with other sorbents enclosed in PE membranes.
Several modications of SPMDs have been proposed for air
possible and under as many different environmental conditions
sampling, which substituted triolein with other sorbents (Table
as possible.
1). The rationales for this substitution included a simplied
preparation and clean-up procedure in samplers not containing
D. Description of different types of triolein52,53 and the high cost of triolein.54 Also, SPMDs are
passive air samplers for SVOCs a patented product which can impede its widespread adoption.
In principle, a different sorbent may also increase the uptake
A number of different PASs have been proposed for SVOC capacity of a PAS relative to an SPMD.
sampling (Tables 1, 2, 4, and 5). There have been principally two Wennrich et al.52 and Paschke and Popp55 proposed to
commonly employed strategies for the development of PASs for substitute triolein with polydimethyl siloxane (PDMS or “sili-
SVOCs. One involves the use of materials originally developed cone”) as a sorbent in different congurations (silicone tubing,
for water passive sampling, the other the use of sorbents that silicone spiral rods, PDMS-coated stir-bar), either enclosed in

1932 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

Table 1 Passive air samplers for SVOCs involving sorbents enclosed within low-density polyethylene (LDPE)

Name Passive sampling material Housing Ref.

Semipermeable membrane devices Thin-walled LDPE lay at tubing White wooden box with louvred 34
(SPMDs) containing a thin lm of synthetic sides (Stevenson screen)
lipid triolein (1,2,3-tri[cis-9-
octadecenoyl]glycerol)
Poly(dimethylsiloxane) (PDMS) Thin-walled LDPE lay at tubing Stainless steel box (20 cm diameter, 52
containing either PDMS-coated stir 20 cm high), open at the bottom
bar (VS ¼ 24 ml, A ¼ 2.8 cm2) or with perforated sides
silicone tubing (VS ¼ 250 ml,
A ¼ 12 cm2)
Poly(dimethylsiloxane) (PDMS) LDPE cylinder containing either Cylindrical stainless-steel shelter 55
PDMS-coated stir bar (A ¼ 1.67 cm2) with an opening at the bottom
or spiral silicone rod (A ¼ 5 cm2)
Membrane-enclosed copolymer Thin-walled LDPE lay at tubing Perforated steel box 53
(MECOP) samplers containing crystalline ethylvinyl-
benzene–divinyl-benzene
copolymer
Versatile, easy and rapid Various sorbent materials enclosed Unknown 54
atmospheric monitor (VERAM) in thin-walled LDPE lay at tubing
(a mixture of orisil and activated
carbon recommended for VOCs
sampling)

a PE membrane similar to an SPMD52 or in a cylindrical PE results highly challenging. Some studies essentially concluded
tube.55 The fairly low SRs of these PAS designs (on the order of that an incomplete understanding of the effect of temperature
0.01 m3 per day) are at least partly overcome by an analytical on the uptake characteristics of those types of samplers
technique that relies on thermal desorption of the PDMS. This precluded a quantitative interpretation.58,59 Whereas SPMDs
not only delivers the entire sampled amount onto a gas chro- continue to be used occasionally as PASs,61–65 the PAS designs in
matographic column, but the lack of sample extraction proce- Table 1 can no longer be regarded as representing the state-of-
dures may allow for lower blank levels and thus detection limits. the-art in passive air sampling for SVOCs.
Unfortunately, very large differences in uptake rates observed
between compounds, between indoor and outdoor deploy-
ments, and between different sampler congurations remained D.2 Samplers based on non-porous sorbent materials
unexplained52,55 and no follow-up studies based on these The samplers described in Section D.1 consist of two sorbing
approaches appear to exist. phases, namely the enclosed sorbent and the polyethylene. This
Choi et al.53 enclosed a crystalline ethylvinylbenzene– appears redundant, because even with two sorbents present,
divinylbenzene copolymer in a PE membrane, calling it these PASs oen do not have sufficient uptake capacity for more
membrane-enclosed copolymer (MECOP) samplers. No SRs had volatile SVOCs to remain in the linear uptake phase during
been determined, so rates for SPMDs were applied aer typical lengths of deployment (e.g. ref. 41). Also, the presence of
adjusting for the much smaller surface area of the PE two sorbents makes the understanding and quantication of
membrane in the MECOP relative to an SPMD. Esteve-Turrillas uptake kinetics and uptake capacity unnecessarily complicated.
et al.54 experimented with a number of different sorbents in Accordingly, most PASs rely on a single material able to act as
SPMD-type lay at PE tubing, and recommended a mixture of a sorbent. In particular, a fairly large number of PASs relies on
orisil and activated carbon. a non-porous sorbent material, such as a non-porous polymer
D.1.3 Limitations of samplers involving sorbents enclosed or a solid lipid as a sorbent (Table 2).
within LDPE. The justication for enclosing a sorbent within D.2.1 Samplers based on low-density polyethylene (LDPE).
a PE membrane is the desire to control the diffusion distance Müller et al.66 realised that the sorbent enclosed in the PE of the
through the use of a permeable barrier. As discussed in Section SPMDs may be redundant, as the PE itself can act as a sorbent.
B.4 above, a tightly controlled diffusion distance is a key Bartkow et al.67 showed that simple PE sheets of different size
requirement for a PAS that is not susceptible to large variations and thickness suspended in air can constitute a PAS. In
in the SR as a result of different wind exposure. A problem with particular, they showed that the simultaneous deployment of PE
this approach is that a PE membrane itself is a sorbent for sheets of different thickness can be used to determine whether
SVOCs. Furthermore, capacity and permeability of PE are a chemical is in the linear uptake phase or approaches equi-
functions of temperature.41,56,57 Collectively, this leads to highly librium. The amount of substances with kinetically controlled
complex uptake characteristics of PASs using PE-enclosed uptake is independent of the volume of the PE sheet VS and
sorbents that make a reliable, quantitative interpretation of related to its surface area A, whereas the amount of substances

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1933
Environmental Science: Processes & Impacts Critical Review

Table 2 Passive air samplers for SVOCs based on non-porous sorbent materials

Name Passive sampling material Housing Ref.

Solid phase micro-extraction SPME bre made from PDMS Needle housing 104
(SPME)
“Synthetic leaf” Fibreglass cloth (250 mm  500 Stainless-steel frames (0.5 m  120
mm) coated with 0.3, 1.5 or 3 g of 0.5 m) in open-sided container
tristearin (2,3-di(octadecanoyloxy)
propyl octadecanoate)
Polymer-coated glass (POG) Hollow glass cylinder (68 mm o.d., Double bowls made of stainless 92
64 mm i.d. and 70 mm tall) coated steel
inside and outside surfaces with
thin lm (0.5 mm thickness) of
ethylene vinyl acetate
Low-density polyethylene (LDPE) Low-density polyethylene lm Galvanised iron chambers with an 67
lms (30 cm by 30 cm) open bottom and louvers on all
sides
Articial leaf (AL)-PAS Paraffin oil on cellulose paper Petri dish 121
Ethylene vinyl acetate (EVA) Glass bre lter (GFF) or thick None 98
aluminum foil coated with ethylene
vinyl acetate
PDMS-based PAS Silicone disk of 0.5 mm thickness in None 114
a 15 cm Petri dish or a baking paper
sheet (40 cm  60 cm) coated on
both sides with a thin lm (0.001
mm) of silicone (0.5 g Si per m2)
PDMS sheet PDMS sheets of 0.1 cm thickness None 111
(11 cm long  9 cm wide  0.1 cm
thick with total surface area 202
cm2)
PDMS brooch PDMS strip (9 cm long  5.5 cm Personal sampler (aluminum plate 113
wide  0.1 cm thick, surface area to prevent contact with clothing)
was 50 cm2)
PDMS-coated stir bar Stir bar, 2 cm long coated with Double stainless-steel domes 110
1 mm thick PDMS lm

achieving equilibrium is proportional to VS.67 Bartkow et al.68 Because PE-based samplers are commonly employed in
demonstrated the viability of using depuration compounds water passive sampling, PE-based PASs offer, similar to SPMDs,
(DCs) spiked into the PE prior to deployment to account for the possibility of using the same sampler material in air and
differences in wind exposure of PASs based on PE-sheets (see water and therefore the possibility to derive the equilibrium
also Section F.4 below). PE-based PASs have since been used status between them.77–90 This is further discussed in Section
extensively by US-based researchers. G.8.1.
Somewhat surprisingly, there appears little attempt to stan- D.2.2 Samplers based on ethylene vinyl acetate (EVA). A
dardise the deployment of PE-based PAS. A variety of different thin lm of ethylene vinyl acetate (EVA) coated on glass was
sizes of PE sheets, strips, and lay-at tubing appear to be in use, introduced as a means to sample SVOCs from the headspace of
but dimensions and sheet thickness are not always provided. a sample vial.91 Large surface area to volume ratios of the EVA
Furthermore, different housings are being used to shelter the lm were meant to assure quick equilibration with the gas-
PE sorbent. Some use a metal box with louvred sides and an phase. Harner et al.92 proposed to use EVA coated onto glass
open bottom,67 whereas others use a metal cage that is not and placed in a wind-shelter as PAS. Farrar et al.93 determined
further described, but appears to be long and upright.69–71 the uptake and loss kinetics of these so-called POlymer-coated
Lohmann et al.72–75 either use the double bowl shelter of Shoeib Glass samplers (POGs) and used them to measure concentra-
and Harner24 or a version where the bottom bowl is eliminated. tion variability along a vertical gradient in the urban boundary
It is also oen not clear how the PE sheets are being placed in layer94 and within a European-scale network.95 POGs have also
these shelters and photographs suggest that little attention is been used for sampling in water.96 Uptake in EVA-lms was
paid to the precise arrangement (see e.g. photograph in ESI in observed to be somewhat faster than in PE sheets, but equili-
ref. 76). Overall, this suggests that even during outdoor bration times were longer because the uptake capacity of EVA
deployments little effort is made to limit or standardise the was larger than that of PE.97 Instead of glass as a base onto
wind exposure of the PE and a lot of trust is therefore placed in which the EVA is applied, it is also possible to use glass bre
the capability of DCs to correct for variable wind exposure lters (GFFs) or aluminum foil.98 While explicitly designed to
(Section F.4). equilibrate fast (within a number of days of deployment), for

1934 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

less volatile chemicals the time to equilibrium can be quite In 2016, two research groups111,114 independently proposed
long. As a result, the interpretation in terms of volumetric air samplers made from PDMS for sampling of SVOCs from indoor
concentrations has to be different for different SVOCs. air. Okeme et al. cut rectangular PDMS sheets of 0.1 cm thick-
D.2.3 Samplers based on polydimethyl siloxane (PDMS). A ness (A ¼ 202 cm2) and suspended them in air.111 Two 50 day
possible drawback of PE-based samplers is the relatively low calibrations for this sampler have been reported. The rst
diffusivity of SVOCs in PE,99,100 which may compromise the calibration involving 5 phthalates, 3 brominated mono-
commonly made assumption that resistance to uptake in aromatics and 9 PBDE congeners yielded an average SR of 1.7
a sampler material is dominated by diffusion through the air m3 per day,111 whereas a second using more or less the same set
boundary layer. Other polymers, in particular PDMSs, tend to of compounds (3 PBDEs less, two OPFRs more) gave an average
allow for faster diffusion and thus pose lower resistance than value of 6.1 m3 per day.112 Going forward Okeme et al. recom-
PE99,100 and one can therefore be more condent that uptake is mend the average of all of these values, which translates to
controlled by the air-side resistance. In Section D.1.1, we already a surface area-normalised SR of 1.5  1.1 m3 per day per dm2.112
learnt that PDMS had been proposed as the main sorbent in Vorkamp et al. suggested either a silicone disk of 0.5 mm
a PAS for SVOC,52,55 but the enclosure of that PDMS in PE thickness in a 15 cm Petri dish (A ¼ 177 cm2) or a baking sheet
compromises the advantage of the higher permeability of the coated on both sides with a thin lm (0.001 mm) of silicone (A ¼
PDMS. Just like SPMDs and LDPE, PDMS is now routinely used 4800 cm2).114 The latter was specically designed with an
for passive water sampling101–103 and thus could be used for extremely high surface area to volume ratio (A/VS), thus aiming
sampling in multiple environmental media. for a low tequilibrium. Experiments conrmed that time to equi-
One of the rst kinetic PASs based on PDMS was a solid librium decreased with A/VS ratio and the baking sheets reached
phase micro-extraction (SPME) bre which is retracted into its equilibrium within two weeks, even for PCB congeners with an
needle housing. Changing the distance by which the bre is intermediate degree of chlorination.114 SRs increased with
retracted within the needle housing results in a change in the surface area, but less than might be expected – while surface
diffusion distance and therefore in the SR.104 While the SRs are area A for the sheet was 27 times larger than for the disk, the SRs
exceptionally small, around 10
5 m3 per day, all of the chemical for PCB-101 were only higher by an order of magnitude (0.2 m3
sorbed to the PDMS is delivered by thermal desorption onto the per day vs. 1.8 m3 per day), even though the sheet may be ex-
analytical column without any sample preparation steps. If pected to have a smaller air side resistance.
a chemical is too volatile, it will be lost from the bre prior to Okeme et al. presented another PDMS-based PAS design to
desorption; if it is too involatile, it will sorb to the inside surface be worn as a personal sampler. A 0.1 cm thick sheet of PDMS (50
of the needle housing.105 Despite the low SR, the tlinear is quite cm2) is attached to an aluminum plate that serves to prevent
short (less than a day), because the uptake capacity is also small contact of the sorbent with clothing when worn as a “brooch”.113
(the PDMS volume VS is small). In its non-retracted mode, an Being calibrated for a number of phthalates against pumped
exposed SPME bre could also function as an equilibrium personal samplers, they estimated an SR of around 0.4 m3 per
passive air sampler.106 This approach was applied to the day or 0.86  0.29 m3 per day per dm2, when normalised to
measurements of pesticide concentration in a greenhouse.107 surface area. The variability in the SR between individuals and
The amount of target chemical sorbed to the small volume of between different SVOCs was large.
PDMS in an SPME bre may oen be too low for reliable Table 3 compares the SRs for the various PDMS-based PASs
quantication. PDMS-coated stir-bars have a higher sorptive that have been proposed in the last few years for indoor
capacity because of a larger sorbent volume VS, yet can still be deployments. The table lists the ranges of the reported SRs for
analyzed with thermal desorption. Stir bar sorptive extraction different SVOCs. Unfortunately, the variability in the SRs
(SBSE) combined with thermal desorption gas chromatography- between different SVOCs for the same sampler and even
mass spectrometry (GC-MS) has been proposed as a passive air between different calibration experiments for the same sampler
sampling technique and applied to record the concentration is large. For example, the highest SR for a chemical in a PAS is
variability of an insecticide108 and PAHs109 in indoor environ- oen a factor of 2 to 4 times higher than the lowest SR, although
ments. Matsiko et al. showed that the technique is suitable for the theory of PAS uptake predicts that SR for different chemicals
measuring organophosphate ame retardants (OPFRs) at should be similar when the resistance to uptake lies on the air
ambient atmospheric concentrations and reported initial SRs side. Even two very similar calibrations for the same sampler
on the order of 2 to 7 m3 per day.110 These SRs appear abnor- yielded SRs that deviate by more than a factor of 3.111,112 More
mally high, when considering that other PDMS-based samplers disconcerting is that surface area-normalised SRs of different
with much larger surface area than a 2 cm-long stir bar have PDMS samplers deviate by orders of magnitude. They range
much lower SRs (see next paragraph111–114). When SBSE is used from a low of less than 0.1 m3 per day per dm2 in the baking
to extract an analyte from a liquid, the stir bar can be spun sheet114 to more than 100 m3 per day per dm2 for the PDMS-
magnetically to lower the boundary layer thickness and thus coated stir bar.110
speed up the uptake kinetics. During its use in air sampling, the The SR is dependent on the extent of air turbulence and
stir bar is not actually spun, i.e. the fact that the PDMS is coated therefore the thickness of the stagnant air layer surrounding the
on a magnetic stir bar is incidental. Unfortunately, it also PDMS. Exposure to air turbulence could differ between different
means that uptake in the stir bar from the air will be dependent deployments, different calibration studies and for the various
on atmospheric turbulence. sampler designs, especially because none of the designs

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1935
Environmental Science: Processes & Impacts Critical Review

Table 3 Sampling rate, surface area and area-normalised sampling rate for various PDMS-based PAS

Exposed surface Area-normalised SR

Sampler SR in m3 per day area in cm2 in m3 per day per dm2 Ref.

Suspended sheet 0.9 to 4.1 202 0.45 to 2.0 111

Suspended sheet 3.5 to 10 202 1.7 to 5 112
Suspended XAD-coated sheet 1.9 to 4.3 102 1.9 to 4.2 112
Sheet attached to aluminum plate 0.2 to 0.66 50 0.4 to 1.32 113
Disk in Petri dish 0.2 to 0.3 177 0.11 to 0.17 114
Coated baking sheet 1.3 to 3 4800 0.03 to 0.06 114
Coated stir bar 1.8 to 6.7 5a 40 to 140 110
a
Assuming a diameter and length of the stir bar of 0.8 cm and 2 cm, respectively.

discussed here relies on a housing to reduce these effects (see layer, it should eventually be feasible to design a PAS with the
Section B.4) and because indoor environments can be quite desired tlinear and tequilibrium for a particular target compound.
inhomogeneous in terms of the air turbulence they experience. In particular, it should be possible to design a PAS that will
However, taken collectively, it suggests that the experimental reach equilibrium for most SVOCs in a reasonable amount of
evidence presented so far is difficult to reconcile with the time. The PDMS-coated baking sheet114 had some of the highest
theoretical expectations on how PDMS-based samplers should achievable A/VS and therefore is the closest of any design to
behave, in particular with respect to an SR that is proportional a true equilibrium PAS for SVOCs. Unfortunately, the sheet has
to surface area A and largely invariant for different substances of to be quite large for quantiable amounts to accumulate in the
similar molecular size. Interestingly, the inconsistency in SRs PDMS lm, which may make it too obtrusive for some indoor
mirrors somewhat the experience with PDMS-based samplers applications.
that include a diffusive barrier52,55 that were discussed in When comparing the relative merits of PE, PDMS and EVA,
Section D.1.2. we can observe that PE has the lowest KSG among the three
D.2.4 Samplers based on lipid-coated bres. SPMDs had polymers used in these types of PASs.97,122 However, it is quite
been conceived as mimicking an aquatic organism consisting of simple to increase a sampler's uptake capacity by increasing its
a lipid reservoir and a membrane that controls uptake from the VS, e.g. by using thicker sheets. While PDMS and PE are non-
aqueous phase.115 In an analogous way, others have sought to polar polymers, EVA also has polar functional groups, which
design PASs that are inspired by plant foliage. Leaves and nee- may increase the KSG for polar target analytes. The KSG of PDMS
dles have been used for monitoring air concentrations of SVOCs is very well studied and good predictive techniques are available
based on diffusive uptake in the waxy plant cuticle116 and the (see Section E.2 below), while the experimental data set for EVA
idea of a “synthetic or articial leaf” is to standardise this and PE is much thinner and, in the case of PE, relies largely on
sampling medium and eliminate variations between species values derived indirectly from partitioning data between PE and
and individual plants117,118 or even within the same plant over water.
time.119 The approach essentially seeks to represent the plant PDMS, and likely also EVA, have the advantage of higher
cuticle with a layer of wax having a large surface area. Proposed permeability relative to PE,100,123 which in principle should
designs include stearin-coated breglass sheet120 and cellulose facilitate validity of the assumption of air-side controlled
paper coated with paraffin oil and placed in a Petri dish.121 kinetics, which is important when using commonly applied
Müller et al. determined both stearin-gas-phase partition coef- equations for the interpretation of results in the curvi-linear
cients (KSG) for PAHs and the mass transfer coefficients uptake region. On the other hand, PE appears to be a more
controlling uptake in and loss from the synthetic leaf.120 Field robust polymer material, which can withstand the rigors of eld
evaluations gave mixed results with agreement within a factor of deployment better than EVA and possibly also PDMS. Interest-
2 in one location and poorer agreement in another. Stracqua- ingly, PDMS-based PASs have mostly been proposed for indoor
danio et al. exposed their articial leaf for only a few hours use, where the demands on the robustness of a polymer are
during which PAHs were expected to have achieved equilibrium reduced.
between paraffin oil and gas-phase.121 The main limitation of PASs using non-porous polymeric
D.2.5 Advantages and limitations of samplers based on sorbents is their relatively low uptake capacity. When an
non-porous sorbent materials. One advantage of the PASs that application of these PASs targets multiple analytes that vary in
rely on a sheet or lm of a non-porous polymer, such as PE, terms of their KSG, some target analytes will be in the linear
PDMS and EVA, is the possibility to custom-design PASs with uptake phase, while others will be in the equilibrium or in the
specic uptake capacities and SRs through the modication of transitional phase. They are therefore neither kinetic nor
the polymer's surface area-to-volume ratio A/VS. Even if the equilibrium samplers. Deciding whether kinetic or thermody-
experimental evidence so far does not support a direct pro- namic factors or both dominate is not straightforward, as it
portionality between surface area A and SR (Table 3), as one depends on the complex interplay between chemical properties
should expect if the uptake is limited by the air-side boundary (KSG, diffusivity in air and polymer), meteorological factors

1936 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

(temperature, wind), sampler characteristics (thickness of One factor contributing to this variability is wind. Laboratory
polymer) and deployment parameters (deployment length). experiments128 and eld studies129,130 have shown that the SR of
Quantitative knowledge of how these factors interact is oen the PUF-PAS is quite strongly dependent on wind speed. This is
lacking. For example, in his critical review of LDPE as a passive because the double bowl is not entirely effective in preventing
sampling material, Lohmann admits that “no data were found winds from impacting on the PUF disk. Chaemfa et al.
on the temperature effect” on the partitioning of SVOCs measured the wind speed within a double bowl housing
between LDPE and the gas-phase, and that the strong temper- exposed to different wind speeds in a wind tunnel.131 The wind
ature dependence of the diffusivity of compounds in LDPE “has speed above the PUF disk was roughly proportional to the
not been addressed in greater detail”.124 outside wind speed and less than a factor of three smaller. The
Nevertheless, various authors have proposed tting methods wind speed below the PUF disk was somewhat lower and
for deriving air concentrations from measurements of SVOCs in increased linearly with outside wind speed at outside wind
the curvi-linear uptake phase of PASs and these are now speeds less than 4 m s
1.131 This is also consistent with
routinely applied. One should, however, be conscious of the fact computational uid dynamics simulations of housings
that if information required in the quantitative interpretation of involving either one or two bowls,132 which indicate the
relatively low capacity PASs is lacking, incomplete or highly dependence of the SR on wind speed and wind angle of inci-
uncertain, this quantitative interpretation will suffer from very dence.133 DCs can be used to account for differences in wind
considerable uncertainties (see Section F.3 below). exposure between deployments (see also Section F.4 below).130
Generally, it appears that PASs based on non-porous poly- Because of the double bowl shelter's inefficiency in blocking
mers have no standardised wind shelter or housing (except wind from impacting on the sorbent, PASs using this design
possibly the original POGs92). In many cases, no housings are tend to also accumulate particle-bound substances (see Section
used at all. This is somewhat surprising, since in all of those F.6).
PASs, it is assumed that the air-side resistance is rate-limiting Differences in SR can also be expected between indoor and
SVOC uptake and therefore the SRs can be expected to be outdoor deployments. Lower SRs indoors than outdoors are
strongly dependent on the air turbulence surrounding the explained by the lower exposure to atmospheric turbulence.134
polymer. While this may be acceptable for true equilibrium Small differences in the dimensions of the double bowl shelter
samplers or for indoor deployments in places with limited do not contribute to statistically signicant differences in the
turbulence, it makes the SR of such samplers in outdoor eld SR.135
deployments highly variable and wind exposure the main D.3.1.2 The sorbent impregnated PUF disks (SIPs) samplers.
source of uncertainty. One attempt to reduce these uncer- Because the PUF has too limited an uptake capacity to sample
tainties is the use of DCs. This is discussed in greater detail in relatively volatile SVOCs kinetically (see Section E.1.3), Shoeib
Section F.4 below. It is noteworthy that many authors propose et al. coated polyurethane foam with ground-up XAD-4
non-porous polymer PASs mostly for use indoors,111–114 where (divinylbenzene–styrene copolymer) powder to obtain so-
wind exposure is generally limited and where temperatures and called sorbent impregnated PUF (SIP).136 Direct side-by-side
concentrations of many SVOCs tend to be reasonably constant, comparisons showed that this indeed increased the uptake
avoiding the challenges described in Section C. capacity of the sampler and increased tlinear.137,138 Use of SIPs
since has focused on neutral peruoroalkyl substances139–146
and organosiloxanes,147–149 because these groups of SVOCs
D.3 Samplers based on porous sorbent materials comprise some relatively volatile substances. However, SIPs are
D.3.1 Samplers based on polyurethane foam (PUF) also being used for SVOCs that had previously been sampled
D.3.1.1 The PUF-PAS by Shoeib and Harner. Polyurethane with PUF-PASs150–152 as well as organophosphate esters138 and
foam (PUF) has been used as a sorbent for SVOCs in active air ionic peruoroalkyl substances.153
sampling for a long time.125 Shoeib and Harner24 rst intro- The idea of adding XAD to a sorbent to increase its uptake
duced this sorbent to PASs for SVOCs, by placing a foam disk in capacity has since been used also for lter paper154 and PDMS
the space between two metal bowls (Fig. 6A). This sampler sheets.112 XAD-coated and uncoated PDMS sheets were found to
design has become the most popular PAS for SVOCs because have similar area-normalised SRs, but the SR of the XAD-coated
PUF is inexpensive, easy to handle, and has a reasonably high sheet showed less variability for different compounds (RSD
capacity. While the SR for this sampler is oen reported to be reduced almost in half from 31 to 17%).112 The calibration
around 3 to 4 m3 per day, the variability in SRs between different experiment was too short to conrm that the XAD-coated sheet
studies and for different compounds in the same study is quite had a longer tlinear.112
high. Reported SRs range from less than 1 to 30 m3 per day.126,127 The rationale for increasing the sampling capacity of one
Melymuk et al. have discussed at length the factors that sorbent by combining it with a high capacity sorbent is not
contribute to this variability.25 Even under indoor deployment always apparent, considering that one simply could use
conditions, the variability in SR of the PUF-PAS for different a sampler based on a high capacity sorbent in the rst place.
SVOCs and in different calibration studies is large; for example, Drawbacks of using combined sorbents include (1) the need for
a range 0.41–11 m3 per day has been reported for single- additional preparation steps that require time and introduce
sheltered PUF-PAS.112 Section F.2 below addresses the issue of the possibility of sorbent contamination, (2) the challenge of
SR variability in detail. reproducing, and maintaining during deployment and

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1937
Environmental Science: Processes & Impacts Critical Review

Fig. 6 Various PASs based on polyurethane foam (PUF) as sorbent. Designs (A) to (F) are based on a PUF disk of the same size. Designs (B), (E) and
(F) are for indoor deployments. Design (C), (F), (H) and (I) are aiming to sample gaseous and particle-bound SVOCs separately with PUF and GFF.
Design (G) measures emissions from a surface. Designs (J), (K) and (L) are used to measure small scale elevation gradients above soil. Design (M)
was used to probe the existence of a sampler side resistance to uptake in PUF. Designs (N) and (O) are miniaturised samplers to be worn by
people or birds. Designs (P) and (Q) rely on the wind to blow through or past the PUF. A is the PUF surface area exposed to the air and VS is the PUF
volume. Sampling rates SR range from 0.01 m3 per day (design (J)) to 100 m3 per day (design (P)). See Table 4 for more detail.

1938 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

handling, the exact composition of the mixed sorbents and original PUF-PAS (Fig. 6C).156 During indoor deployments, the
therefore the uptake capacity of a PAS, and (3) the likelihood of PAS-GP-I had very low SR of around 0.1 and 0.01 m3 per day for
introducing additional complexity to the uptake kinetics, for gaseous and particle-bound PAHs, respectively,155 whereas the
example with respect to a sampler side resistance. modied PUF-PAS had a SR of around 1 and 0.5 m3 per day for
D.3.1.3 Other PASs based on PUF. The design by Shoeib and gaseous and particulate brominated ame retardants,
Harner24 is not the only PAS that is using PUF as a sorbent respectively.156 In 2009, Tao et al. introduced a modied
(Table 4). In particular, several attempts have been made to design, called PAS-GP-II, in which the can holding PUF and
complement the PUF with a GFF meant to collect particle- GFF was made of an open mesh, which in turn was placed in
bound SVOCs. Tao et al. presented a design, called PAS-GP- a cylindrical wind shelter made of PE (Fig. 6I).157 The moti-
I, involving a cylindrical metal can, that has a PUF disk at vation for the re-design was the need to increase the SR, if the
the closed top and a GFF suspended close to the perforated sampler is to be useful in locations where SVOCs are not
bottom (Fig. 6H).155 Abdallah and Harrad adopted essentially greatly elevated. Indeed, the SRs of PAS-GP-II were deter-
the same idea, except that they placed the PUF at the top and mined to be 0.4 and 0.6 m3 per day for gaseous and particulate
the GFF in the middle of the double bowl housing of the PAHs, respectively.

Table 4 Passive air samplers for SVOCs based on polyurethane foam described in the literature. The letters in brackets behind the name refer to
Fig. 6

Name Passive sampling material Housing Ref.

PUF-PAS (A) PUF disk (14 cm diameter, 1.35 cm Stainless steel double bowls 24 and 161
thick, A: 365 cm2, VS: 207 cm3, (diameter: top 30 cm, bottom 24 cm)
density 0.0213 g cm
3, 4.40 g mass, or single cover plate (indoors)
0.567 cm effective thickness)
Modied PUF-PAS (C) PUF disk (14 cm diameter, 1.2 cm Stainless steel double bowls (top 156
thick, A: 360.6 cm2, density 23 cm, 2 L, bottom: 18 cm, 1 L)
0.02 g cm
3) and GFF (12.5 cm
diameter)
PAS-GP-I (H) PUF disk (10 cm diameter  1 cm Stainless steel can (10 cm diameter 155
thick, density 0.024 g cm
3) and  10 cm high) with perforated
GFF (8 cm diameter) bottom
PAS-GP-II (I) PUF disk (11 cm diameter  1.5 cm Mesh cylinder (11 cm diameter, 157
thick, density 0.024 g cm
3) and 23 cm high) inside a PE cylinder
GFF (11 cm diameter)
Directional PAS (Q) 2 PUF chips (5  7  1 cm3 each, Flow duct (20 cm long, 5 cm high, 172
0.024 g cm
3) 5 cm wide) with 2 one-way valves
PAS-V-I (J) PUF disk (3.8 cm diameter, 1.5 cm Stainless steel cylinder (3.8 cm 169
thick, density 0.024 g cm
3) diameter, 2 cm high) with mesh-
covered bottom (400 mesh, 3 cm
diameter)
PAS for vertical gradients (K) PUF disk (6 cm diameter, 1.5 cm Top bowl, bottom ring and ne 171
thick, density 0.024 g cm
3) mesh (4 cm diameter)
PAS-DD (D) PUF disk (14 cm diameter, 1.35 cm Circular at metal plate placed 2 cm 158
thick, A: 365 cm2 36% covered) above the disk
PAS-H (L) PUF disk (3.5 cm thick, 6.2 mm ABS cylinder (5.5 cm high, 6.2 cm 170
diameter, density 0.024 g cm
3, A: inner diameter), 18 2 mm holes in
128 cm2, VS: 106 cm3) top and bottom, rain shield
Miniature bird-borne PAS (O) PUF (0.2 cm-thick, 0.023 g cm
3 3D-printed elliptical double bowl 168
density, A: 4 cm2) and GFF (0.045 cm
thick, 0.19 g cm
3, A: 3.45 cm2)
Alternatively: PDMS (0.1 cm thick,
1.14 g cm
3 density, A: 4 cm2)
Mini-PUF or PUF-cyl (N) PUF cylinder (10 cm long, diameter Protective cover net mesh size 1.0 166
2.2 cm, A: 77 cm2, density mm, placed on helmet or lapel
0.030 g cm
3)
Indoor PUF-PAS (E) PUF disk (14 cm diameter, 1.35 cm Aluminum rigid mesh holder big 164
thick) enough to just hold the PUF disk
Indoor PUF-PAS (F) PUF disk (14 cm diameter, 1.35 cm Stack of GFF (top), PUF disk 163
thick) and GFF (middle) and bowl (bottom)
PUF-PES (G) PUF disk (14 cm diameter, 1.35 cm Inverted glass Petri dish, covered 165
thick) with aluminium
Flow-through sampler (P) Seven P10z PUF disk (10.5 cm Horizontally oriented, 173
diameter, 2.54 cm thickness and aerodynamically shaped, stainless
large porosity, 10 pores per inch) steel ow tube

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1939
Environmental Science: Processes & Impacts Critical Review

Eng et al. presented a PUF-based PAS specically designed to indoor deployments (Fig. 6E).164 The motivation was to take
quantify the dry deposition of SVOCs.158 A PUF disk of the same advantage of a well-characterised PAS sorbent, but have
dimensions as is used in the regular PUF-PAS is placed 2 cm a minimally intrusive and easy-to-set-up housing that still
beneath a circular at cover plate, which serves as a sun- and allows for efficient airow past the PUF and prevents gravita-
rain shelter (Fig. 6D). The more open shelter design of this so- tional particle deposition. Side-by-side sampling with an active
called passive dry deposition (PAS-DD) sampler allows for sampler allowed for the estimation of SRs and to conrm that
more efficient sampling of particles, particularly large particles. the PAS derived air concentrations of gas-phase SVOCs were
However, this open design also results in less wind sheltering accurately ranked when judged by the results of the active
than in the double bowl housing, which should make the SR of sampler.164
the PAS-DD even more dependent on variations in wind speed Herkert et al. placed a PUF-disk in an inverted Petri dish to
than those of the PUF-PAS. The PAS-DD has been used in record the release of SVOC vapours from indoor surfaces and
combination with the PUF-PAS to determine dry deposition called the device a passive emission sampler (PES) (Fig. 6G).165
uxes and velocities of PAHs in two Nepalese cities159 and to Bohlin et al. used a small PUF cylinder (10 cm long, 2.2 cm
study the dry deposition of pesticides.160 diameter) in a small mesh net, attached to a helmet or to
A number of modied versions of the original PUF-PAS24 a person's clothing, as a personal sampler for SVOCs
have been presented specically for use indoors. In order to (Fig. 6N).166 This sampler design could also be used for
speed up uptake indoors, Shoeib et al.161,162 proposed to elimi- stationary sampling indoors. SRs for that sampler based on
nate the double bowl housing for indoor deployments and only comparisons with concentrations obtained with personal active
use a plate to prevent gravitational settling of coarse particles sampling have been reported.166,167
(Fig. 6B). Eliminating the bottom bowl, as done by Harrad and Sorais et al.168 presented a miniaturised version of the
Abdallah,256 amounts to essentially the same thing. The SR of an sampler of Abdallah and Harrad156 that could be attached to
unsheltered sampler indoors was either 2.5 m3 per day161,162 or free-ranging birds, i.e. a personal sampler for birds. It combined
1.5 m3 per day.256 In order to separate particle-bound and a small PUF disk and GFF in an elliptical double polyamide
gaseous SVOCs, Wang et al. placed a GFF for sampling of bowl manufactured with a 3D printer (Fig. 6O).
particles slightly above the standard PUF disk, which itself was Three PUF-based PASs have been presented that sought to
placed at the top of a stainless-steel bowl (Fig. 6F).163 This determine vertical concentration gradients above the
resembles the set-up by Abdallah and Harrad,156 but by not ground.169–171 In all three cases, a PUF disk was placed in
having a cover at the top, the lter will also be sampling larger a cylindrical housing, made either of stainless steel169,171 or
particles subject to gravitational settling. acrylonitrile butadiene styrene (ABS)170 that was just big enough
Dodson et al. designed a small aluminium housing for to accommodate the disk. The housing either had an opening at
a PUF-disk of the size used in the PUF-PAS, specically for the bottom covered with a stainless-steel screen mesh

Table 5 Passive air samplers for SVOCs based on granular sorbent material described in the literature. The letters in brackets behind the name
refer to Fig. 7

Name Passive sampling material Housing Ref.

XAD-PAS (A) Stainless steel mesh cylinder (10 cm Inverted cylindrical steel cans 26
or 20 cm long, 2 cm diameter) lled
with styrene–divinylbenzene
copolymer
High wind version of the XAD-PAS Same as the XAD-PAS Inverted cylindrical steel cans with 196
(B) wind-buffering baffles at the bottom
Activated carbon felt (ACF PAS) (D) Activated carbon felt Double bowls 185
Polyurethane foam disks PUF disk (14 cm diameter, 1.35 cm Double bowls (upper 30 cm 136
impregnated with ground XAD resin thick) impregnated with styrene- diameter and lower 24 cm diameter)
(SIP-disk PAS) (C) divinylbenzene copolymer powder
(0.4 g of XAD-4 resin)
Axial PAS (E) GFF (2.5 cm diameter) coated with Puck holder with spacer ring and 200
2.3 mg of Tenax TA mesh cover, placed under plastic
rain shelter
Tenax TA PAS (H) Stainless steel mesh cartridge lled Custom made PVC shelter 201
with Tenax TA
XAD-pocket (F) Stainless steel mesh pocket (9 cm None 111
long  5.5 cm wide  0.2 cm thick;
A ¼ 105 cm2; volume ¼ 9.9 cm3)
lled with 7 g of XAD-4
(AS ¼ 750 m2 g
1)
XAD-PDMS (G) PDMS sheet coated with XAD-4 None 112
resin

1940 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

(Fig. 6J),169 or had eighteen 2 mm holes drilled into both the top Sampler” (FTS) resembles more that of a pumped active
and the bottom of the housing, whereby the top was protected sampler, because it does not rely on the diffusion of the target
by a rain shield (Fig. 6L).170 SRs, in both cases obtained by analytes to the sorbent. The sampled air volume is derived from
calibration against another PUF-based PAS, were 0.01 to 0.03 m3 a calibration involving the measured wind speed. The FTS is
per day,169 and 1.5 m3 per day,170 respectively. In the design by particularly useful when trying to measure very low air
Wang et al. the top of the PUF disk is covered by a small inverted concentrations in locations where reliable power is difficult to
bowl, which allows for a larger area to be exposed to air and obtain. Examples are the Arctic,176–178 sub-Arctic,179 the Tibetan
therefore an increased SR (1.5 m3 per day) (Fig. 6K).171 plateau,176,180 and Antarctica.181 However, the sampler has also
Whereas most PASs seek to prevent the wind from impacting been used closer to sources, for example to measure pesticide
on the sorbent, two PUF-based samplers were explicitly spray dri.182
designed to take advantage of the wind in the sampling process. Esen introduced a housing for a PUF-based PAS that seeks to
The “directional” PAS seeks to sample only when the wind minimise the inuence of wind.183 A side-by-side comparison
blows from a particular direction.172 The air is guided to pass by with the double-bowl PUF-PAS indicated lower SRs.184
two small PUF chips within a small wind duct aligned with the D.3.2 Activated carbon bre felt sampler. Another
desired wind direction. Hinged polypropylene aps act as one- approach to increase the limited uptake capacity of poly-
way valves that shut down the airow when there is no wind urethane foam is to retain the double bowl housing of the PUF-
from that direction (Fig. 6Q). Calibrated and eld-tested for PAS, but use a different sorbent (Table 5). Oono et al. experi-
PAHs,172 this sampler does not seem to have found use since. mented with activated carbon bre felt to sample uorotelomer
Xiao et al. designed a horizontal cylindrical ow tube that alcohols from the atmosphere (Fig. 7D).185 During a month-long
turns into the wind with the help of ball bearings and wind deployment, uptake remained linear and an equivalent air
baffles.173–175 The wind is blowing the air through a series of PUF volume of 300 m3 during that period suggest a high SR of 10
disks placed in the ow tube and is then recorded with m3 per day.185 Despite such promising results, no follow-up
a battery-operated anemometer at the outow of the tube studies have appeared to be undertaken with this sorbent.
(Fig. 6P). The PUF used in this sampler (P10z) is more porous D.3.3 Samplers based on XAD-resin. The SIP-PAS described
than that used in other PUF-based PAS. The motivation for this in Section D.3.1.2 sought to increase the uptake capacity of the
design was to achieve SRs that greatly exceed those of other PAS. PUF-PAS by impregnating it with XAD-resin (Fig. 7C). Like PUF,
The sampling principle of this so-called “Flow-Through XAD, which is a divinylbenzene–styrene copolymer, also has

Fig. 7 Various PASs based on porous sorbents other than PUF. Designs (A), (B), and (F) are based on XAD-resin, design (D) on activate carbon felt,
and design (H) on Tenax TA. Other designs use resin to enhance the sorptive capacity of other sorbent materials, namely PUF (C), GFF (E) and
PDMS (G). A is the sorbent surface area exposed to the air and m is the resin mass. Sampling rates SR range from 0.05 m3 per day (design (E)) to
10 m3 per day (design (D)). See Table 5 for more detail.

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1941
Environmental Science: Processes & Impacts Critical Review

a long history of use as a sorbent in the air sampling of organic either up or down, the PAS is expected to distinguish particle-
compounds.186,187 XAD has a higher uptake capacity for SVOCs bound substances settling gravitationally or by other
than PUF188,189 and also the stability of sorbed compounds means.200 SRs for PAHs and oxygenated PAHs in this sampler
during storage of XAD is higher than on PUF.188 In contrast to have been determined in an exposure chamber receiving diesel
the non-porous polymers and polyurethane, XAD is an adsor- exhaust199 and in a eld study above an urban roadway.200 While
bent and its large uptake capacity is a result of a very high the variability in those SRs was large during the rst test,199 the
specic surface area (XAD-2: 350 m2 g
1, XAD-4: 750 m2 g
1). eld study yielded more consistent SRs on the order of 0.05 to
Because it is a powder, a container is required that holds the 0.1 m3 per day.200 The small SR relative to other PASs for SVOCs
XAD-resin but does not obstruct the diffusion of the target is mostly due to the small size of the sampler and is compen-
analytes to the sorbent. sated for by the quantication technique that delivers the
Wania et al. introduced a PAS which consists of a XAD-2 entirety of the sequestered amount onto the gas chromato-
resin-lled stainless-steel mesh cylinder, which is suspended graphic column.200
in a cylindrical stainless-steel housing (Fig. 7A).26 This sampler Sonnette et al. used a Radiello stainless steel mesh cartridge
has been extensively characterised in terms of the SR of lled with Tenax TA as a PAS sorbent.201 During sampling, they
different SVOCs in different climates21,27,28,190 and with respect placed the cartridge into a PVC shelter but did not use a Radi-
to the inuence of wind on the SR.191 Experiments with different ello diffusive body to control the diffusion distance (Fig. 7H).
sampler congurations conrmed that the uptake rate is They also used thermal desorption onto GC/MS for quantica-
proportional to the surface area of the mesh cylinder192 and tion. While some initial indoor and outdoor deployments are
both a short (10 cm) and long (20 cm) version of the sampler described, no attempts to characterise the samplers have been
have been used. The deployment of multiple mesh cylinders in reported.
the same housing does not affect the SR, i.e. there is no evidence D.3.5 Extraction of porous sorbent materials. Porous
of a so-called starvation effect.193 Because very high winds have sampling sorbents, such as PUF and XAD-resin are typically
been shown to increase the SR,194,195 Gong et al. introduced solvent-extracted using either a Soxhlet-apparatus or pressur-
a modied housing with wind baffles that reduce the wind ised liquid extraction.202 Alternatively, microwave- or
speed dependence of the XAD-PAS's SR (Fig. 7B).196 ultrasound-assisted techniques have been proposed for
Because of its high uptake capacity (see also section E below), extracting organic trace compounds from PUF with the aim to
the XAD-PAS is probably the only PAS for SVOCs that can be reduce solvent volumes and extraction time. A recently pre-
regarded as a true kinetic sampler. Even for more volatile SVOCs sented micro-scale version of an ultrasound-assisted extraction
such as the volatile methyl siloxanes (VMS) and the uo- technique was judged simpler, faster, cheaper and more envi-
rotelomer alcohols (FTOHs), linear uptake was observed for ronmentally friendly than other methods.203
periods as long as three months28 and one year.197 The high D.3.6 Advantages and limitations of samplers based on
uptake capacity also explains, why the XAD-based PAS porous sorbent materials. The PASs based on PUF share with
frequently nds use in extremely remote environments, where the non-porous polymer samplers described in Section D.2,
limited site accessibility may demand very long deployment a quite limited uptake capacity and therefore cannot be
periods. Examples are the Tibetan plateau, mountains and considered linear samplers for more volatile SVOCs (e.g. lighter
polar eld camps (for details see Section G.6). PCB congeners during warm deployments204). The same issues
The XAD-PAS is suited for outdoor deployments, whereas its related to the uncertainty of the quantitative interpretation
housing may be too bulky for indoor deployments and in fact (Section D.2.5) apply therefore also to the PUF-PAS. This is
not necessary. XAD-lled stainless steel mesh bags111 sus- discussed in more depth in Section F.3 below.
pended in air have therefore been explored as potential high The kinetic samplers (XAD-PAS, SIP-PAS), on the other hand,
capacity indoor-PAS (Fig. 7F). Okeme et al. used XAD-4, because have the advantage that no knowledge of KSG is required, and
of a higher specic surface area (750 m2 g
1).111 Because the that long deployment times or sampling of relatively volatile
XAD-lled mesh bag had a lower SR than a PDMS-based PAS SVOCs are possible without worrying about approaching equi-
explored at the same time (see Section D.2.3), the authors librium. They are also the only samplers that give air concen-
seemed to favour the PDMS-PAS.111 Later, they also explored the trations that are truly the average over the entire length of
possibility of enhancing the uptake capacity of PDMS by adding deployment. This is not the case when sampling in the curvi-
XAD-4 powder (Fig. 7G).112 linear uptake range while the air concentrations vary with time.
D.3.4 Samplers based on TENAX. Building on the passive A disadvantage of the PASs with porous sorbents is that the
particle sampler by Wagner and Leith,198 Arnoldsson et al. pre- assumption of air-side resistance is not valid. This has been
sented an axial diffusion sampler with a GFF (2.5 cm diameter), shown for both the XAD-PAS and the PUF-PAS, by demon-
either bare or coated with 2.3 mg of Tenax TA, as the sorbent strating a concentration gradient within the sorbent205 and
(Fig. 7E).199 For analysis, the GFF is folded and placed into through a critical evaluation of reported KPUF–G values.206 This
a tube, from which the target compounds are thermally des- also implies that the equations in Section B for interpreting the
orbed and transferred to a GC-MS. The working hypothesis is uptake of SVOCs in the sorbent cannot be applied uncritically.
that the PAS outtted with an uncoated GFF takes up particle- Instead, more complex uptake models are required for a quan-
sorbed compounds, whereas the Tenax-coated GFF addition- titative interpretation of SRs (see Section F.5 below).21,207
ally takes up gas-phase substances. By having the GFF facing

1942 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

Despite the use of wind-sheltering housings, the SRs of PASs CS

KSG ¼ (9)
using the double bowl design (PUF-PAS, SIP-PAS, carbon felt CG
bre-PAS and others) are quite substantially impacted by wind,
especially, but not only, at high wind speeds.128 The wind effect
KSG thus has units of m3 gas per m3 absorbent (or m3 gas per
on the shelter used for the XAD-PAS is much smaller,26,191 2
m adsorbent). As KSG is an equilibrium constant, we can use
especially in the version by Gong et al.196 In the balancing of the
a van't Hoff type equation to relate it to temperature:
conicting needs for a large vs. a tightly controlled SR (Section
C.1), the double bowl design samplers lean towards higher, yet dln KSG DUSG
¼ (10)
more variable SRs and the XAD-PAS leans towards lower, but dT RT 2
less wind-dependent SRs. where T is absolute temperature in K, R is the ideal gas constant,
Among the strategies to account for the effect of wind on SR and DUSG is the internal energy of transfer between sorbent and
is the use of DCs. It is important to realise that the use of DCs gas-phase in J mol
1. Note that we use the internal energy DUSG
requires the assumption of air-side resistance to be valid (see instead of the enthalpy of phase transfer DHSG, because in the
Section F.4 below). DCs therefore have not been used in the PAS community the abundance of chemical in the gas-phase is
XAD-PAS. Even though there is also evidence that this typically expressed in volumetric air concentrations and not in
assumption is violated in the PUF-PAS,205 DCs are routinely partial pressure. If KSG is expressed in units of mol per m3
employed in the PUF-PAS.127,208,209 sorbent per Pa or mol per m2 adsorbent per Pa, the use of DHSG
By normalizing the SR to the sequestered amount of would be appropriate. The two are easily derived from each
a compound whose concentration can be assumed to be other:
uniform in space,194 it is possible to account for differences in
the wind exposure of the XAD-PAS. HCB can oen be used if it DUSG ¼ DHSG
RT (11)
can be assured that there are no HCB sources in the vicinity of
a sampling site.195,210 If DUSG is assumed to be a constant, the integration of eqn
XAD-resin is quite expensive and it can be difficult to handle, (10) yields:
especially when electrostatic effects occur. Coating XAD-resin  
KSG ðTÞ
DUSG 1 1
onto PUF disks when making SIPs in particular, can be labo- ln
¼
(12)
KSG Tref R Tref T
rious and difficult and can easily lead to sorbent contamination.
On the other hand, high capacity samplers do not require DCs,
which is a considerable expense associated when seeking to With eqn (12) it is possible to calculate KSG at a temperature
quantitatively interpret results of PASs with limited uptake T, if DUSG and KSG at a reference temperature are known.
capacity in the curvi-linear uptake regions. This section reviews both the experimental and theoretical
While a surprisingly large number of PASs for SVOCs have studies that have been conducted to determine KSG and DUSG
been introduced, they are for the most part described in for different types of sorbents used in passive air sampling.
a stand-alone paper with no follow-up studies seeking to When referring to the KSG and DUSG for a specic sorbent, we
characterise, test or apply the sampler. Among the PASs dis- substitute S with the sorbent acronym. For example, KPUF–G and
cussed in section D, only the PUF-PAS and XAD-PAS, and to DUPUF–G refer to the sorption coefficients and internal energy of
a smaller extent also the SIP-PAS and PE-based samplers have phase transfer between PUF and the gas phase, respectively.
been thoroughly characterised and are used most extensively.
Therefore, the remainder of the review will largely focus on
those PAS designs. E.1 Experimentally determined uptake capacities of PAS
E.1.1 Studies reporting on KPE–G. While there appear to
have been no experiments yet that sought to measure the
E. The capacity of different PAS equilibrium sorption coefficient between LDPE and the gas-
sorbents for SVOCs phase (KPE–G) and its temperature dependence under
controlled conditions,124 such data have been repeatedly ob-
As discussed in Section B above, knowledge of the KSG at the tained from eld data. In early work, Bartkow et al.67 and Ken-
temperature of deployment is essential for using equilibrium nedy et al.97 derived such sorption coefficients for PAHs from
PASs as well as for interpreting measurements in the curvi- eld data. Khairy and Lohmann73–75 derived KPE–G for PAHs,
linear part of a PAS's uptake curve. Even if a sampler is oper- OCPs, PCBs and PBDEs from deployments in Alexandria, Egypt,
ated in the kinetic range, knowledge of the KSG is useful to and Providence, RI, USA, using concentrations measured in the
estimate tlinear. atmosphere and concentrations in the LDPE-PAS, mathemati-
We dened KSG in Section B.1 as the equilibrium sorption cally adjusted for equilibrium. However, partitioning data
coefficients between the gas-phase and the PAS sorbent, which derived from eld data inevitably apply to the temperature
is the ratio of the concentration of a chemical in the sorbent CS conditions during deployment and need to be adjusted. It
(in units of mol m
3 sorbent or mol m
2 adsorbent) and in the should be possible to derive values of KPE–G through a thermo-
gas-phase CG (in units of mol m
3 gas) if the chemical has dynamic triangle with water–PE equilibrium partition coeffi-
reached equilibrium between the two phases: cients KPE–W and Henry's law constants KG–W, because uptake in

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1943
Environmental Science: Processes & Impacts Critical Review

PE is a bulk phase sorption process and is not inuenced by the a whole range of VOCs, including aliphatics, chlorinated
presence of water.124 Measured values of KPE–W are much more aliphatics, monoaromatic (BTEX, halogenated benzenes,
common, due to the wide-spread use of PE in passive water phenols), ketones, amines, alcohols, and carboxylic acids at
sampling. 20  C using XAD as a stationary phase.221 For a subset of these
E.1.2 Studies reporting on KPDMS–G. Equilibrium sorption substances, they also reported retention volumes at tempera-
coefficients between PDMS and the gas-phase KPDMS–G have tures between 40  C and 90  C. These retention volumes allow
been measured for a fairly large number of organic compounds, for the calculation of equilibrium sorption coefficients between
because of the use of PDMS in SPME bres. Sprunger et al. have XAD and the gas-phase KXAD–G. Shen et al. used a headspace
compiled KPDMS–G values measured for 64 solutes at room technique to directly measure the sorption of 1,2-dichloroben-
temperature by a number of research groups.211 Gilbert et al. zene, 1,2,4,5-tetrachlorobenzene and HCB onto XAD-2 resin,222
showed that polymer–polymer partition coefficients can be used whereas Hayward et al. applied inverse GC involving a column
to derive KSG values, e.g. by combining KPE–G with KPDMS–PE lled with XAD-resin to determine KXAD–G for a selection of 52
partition coefficients in a thermodynamic triangle to calculate diverse organic compounds as a function of temperature.223 A
KPDMS–G values for PCB congeners.122 Recently, Okeme et al.212 limitation of these studies is that they oen focus on VOCs and
and Tromp et al.213 have greatly expanded the set of measured not on the SVOCs that are typically sampled with the XAD-PAS.
KPDMS–G values, either from isothermal retention volumes on This is because the KXAD–G for SVOCs are typically too large to be
a DB-1 column that has PDMS as a stationary phase212 or by amenable to reliable measurements. Sometimes, it is possible
using a calibration involving an exposure vessel with controlled to infer minimum values of KXAD–G from a eld calibration
temperature and wind speed.213 experiment, even if equilibrium was not reached during the
E.1.3 Studies reporting on KPUF–G. Shoeib and Harner longest deployment. Krogseth et al., for example estimated
derived KPUF–G for several lighter PCB congeners by assuming a lower limit of 6.5 for the log KXAD–G of a number of cyclical and
that they had reached equilibrium with laboratory air aer linear VMS.28
extended exposure, and solving eqn (6A) for KSG and using
measured concentrations in air and PUF.24 Zhao et al. reported
KPUF–G for eight aromatic VOCs at 21  C.214 Kamprad and Goss E.2 Prediction techniques for the uptake capacity of
determined KPUF–G for 103 relatively volatile organic different PAS sorbents
compounds with diverse functionalities by measuring the E.2.1 Single-parameter linear free energy relationships.
chromatographic retention volume of analytes on a column Because measurements of KSG are generally only available for
lled with PUF.215 Bidleman et al. determined KPUF–G by equil- a few substances and at a limited number of temperatures,
ibrating PUF with ambient air at different temperatures in the there has been interest in deriving techniques for predicting
eld for HCB, a-hexachlorocyclohexane (a-HCH) and two KSG. Most oen they take the form of single parameter linear
brominated anisoles.216 More recently, Harner and co-workers free energy relationships that seek to linearly regress log KSG
have used dedicated laboratory experiments using a generator with the logarithm of either liquid-state vapour pressure or the
column approach to determine KPUF–G as a function of octanol–air equilibrium partition coefficient (KOA) of
temperature for parent PAHs and alkylated PAHs,217–219 penta- a compound. For example, Pankow et al. regressed the retention
chlorobenzene (PeCBz), HCB, a- and g-HCH,219 four OPFRs, six volumes on XAD resin against vapour pressure.221 Shoeib and
novel brominated ame retardants (BFRs), HCBD and penta- Harner regressed the measured log KPUF–G for eight PCB
chloroanisole (PCA).220 Tromp et al.213 measured KPUF–G for 98 congeners against log KOA,24 and Bartkow et al. regressed the
SVOCs at 20.7  C during a 160 day-long calibration study in an KPE–G of several PAHs against log KOA.67,97 There are many more
exposure chamber. equations of this type reported in the literature (e.g. see
We should note that for a porous sorbent such as PUF, the compilation for KPUF–G in Okeme et al.206). However, it is now
KPUF–G with units of m3 gas per m3 absorbent, as it is reported in well established that such relationships tend to be only valid for
many of the studies mentioned here,213,217–220 can be confusing groups of closely related substances that interact with the
as the volume of the absorbent could refer to solid polyurethane sorbent with the same type of intermolecular interactions.224 As
or to foam. To avoid misunderstandings, it would be preferable such, their use might be defensible for a group of non-polar
to report the KPUF–G with units of m3 gas per g absorbent215 or to organic compounds, but they clearly cannot be expected to
include information on whether the volume of the foam or the provide reliable estimates for substances that engage in polar
polymer was used.214 No work has been reported that explored interactions.220
whether KPUF–G varies with the type or the density of the PUF, or E.2.2 Poly-parameter linear free energy relationships. More
whether variables such as relative humidity have an inuence widely applicable are so-called poly-parameter linear free energy
on the sorptive capacity of PUF. relationships (ppLFER), which are multi-variate linear regres-
Comparing different prediction methods for KPUF–G, Okeme sions of measured sorption constants against a series of solute
et al.206 noted that some experimental KPUF–G are likely erro- descriptors that quantify a compound's ability to undergo
neous, because the compounds did not achieve equilibrium various types of intermolecular interactions such as hydrogen
between all of the exposed PUF and the gas-phase. bonding.225 Kamprad and Goss,215 Sprunger et al.211 and Hay-
E.1.4 Studies reporting on KXAD–G. Pankow et al. measured ward et al.223 used their respective data sets to derive such
and compiled gas chromatographic retention volumes of relationships for KPUF–G, KPDMS–G and KXAD–G, respectively.

1944 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

Whereas Hayward et al. presented such equations for KXAD–G at absorbents and adsorbents, respectively. In order to compare
different temperatures,223 Kamprad and Goss derived a separate the uptake capacity of different commonly used PAS, we esti-
equation for the enthalpy of PUF–gas-phase transfer DHPUF–G.215 mated their KSGVS or KSGAS values for HCB at approximately
Uber et al. presented ppLFERs for KPE–W,226 including one for 20  C (Table 6). Not only is HCB a relatively volatile SVOC, for
LDPE based on the experimental data by Choi et al.227 If which tlinear could be quite short, but experimental KSG values
combined with a ppLFER for air–water partitioning, this should exist for most sorbents relevant for PAS.
allow for the derivation of a predictive equation for KPE–G. The calculation conrms that in the SIP, the addition of 10%
Finally, there are such relationships for solute uptake in other by mass of XAD-4 resin results in a more than 100-fold increase
sorbents from the gas-phase. These sorbents have found or in the overall sorption capacity of the PUF disk for HCB. Overall,
could conceivably nd use in PAS, such as Chromosorb, Tenax this calculation reveals that the sorption capacity of different
and Porapak.228 PAS for one compound can vary by ve orders of magnitude.
While experimental solute descriptors for frequently This difference would be even larger if the SPME-PAS104 were to
measured SVOCs, such as the PCBs and PAHs are readily be included in the comparison, as its VS is tiny.
available, there has in recent years also been a concerted effort The estimation suggests that the XAD-PAS has a theoretical
to determine experimental solute descriptors for other envi- uptake capacity that is four orders of magnitudes higher than that
ronmentally relevant SVOCs.229–231 Even if no solute descriptors of most other PASs for SVOCs. The comparison is somewhat
are available, they can now be predicted quite well for many misleading, as not all of the sorbent in a PAS necessarily partici-
SVOCs using the quantitative structure property relationships pates in the uptake. For example, most of the XAD in the XAD-PAS
implemented in the UFZ LSER website.232 The ppLFER-based is not accessible to the compounds from the atmosphere because
prediction of KPUF–G, KPDMS–G, KXAD–G and KPE–G values is now of a notable transport resistance within the passive sampling
feasible for a wide range of substances.206 medium.205,207 However, this is partly also true for the other
E.2.3 Predictions based on quantum-chemical calculations sorbents.205,206 Nevertheless, the comparison illustrates why the
and statistical thermodynamics. Whereas the single- and poly- XAD-PAS is the closest to a kinetic PAS for SVOCs. It also explains
parameter linear free energy relationships require experi- why the XAD-PAS generally does not require the consideration of
mental data for calibration, prediction techniques based on curvi-linear uptake, even for more volatile SVOCs, and is more
quantum-chemical calculations and statistical thermodynamics suited for long deployment periods than other PAS.
can predict equilibrium sorption properties directly from
molecular structure. In particular, Goss demonstrated that the
commercial soware COSMOtherm can be used to predict F. The sampling rates of different
partitioning into various polymer matrices by treating the
polymer as a “liquid of monomer repeat units that can move
types of PASs
independently from each other.”233 The method was successful F.1 Empirical determinations of sampling rates
in predicting the KPDMS–G at 25  C and the KPUF–G at 15  C and F.1.1 Deriving a sampling rate from a calibration experi-
95  C.233 Loschen and Klamt later improved the approach “by ment. Rearranging eqn (5) we obtain:
incorporation of polymer-specic entropic contributions due to
DmS
free volume effects”.234 PDMS can be represented as liquid SR ¼ (13)
CG  Dt
oligomer (hexamethyldisiloxane) when using COSMOtherm to
predict KPDMS–G,212 but this approach greatly overpredicts the which implies that we can derive the SR of chemical in a PAS by
measured KPDMS–G, with an average difference in excess of 2 dividing the amount of the chemical DmS taken up during the
orders of magnitude. Smaller overpredictions, generally less time period Dt by the product of the gas-phase concentration CG
than an order of magnitude, were achieved when predicting and this time period. It thus requires CG during the time of PAS
KPUF–G for various compounds at different temperature, repre- deployment to be determined by another method, most
senting polyurethane as “a simple oligomeric 1:1 condensed commonly an active sampling method. This empirically derived
pair of 2,4-toluene-diisocyanide and glycerol with ethyl and SRe only corresponds to the inherent SR, if the PAS is operating
acetate end-caps”. Parnis et al.217,218 used COSMOtherm to in the linear uptake period. Quite frequently, SRes are indeed
predict KPUF–G and its temperature dependence for substituted determined with eqn (13) and a single PAS deployment.210,236,237
PAHs. Chen et al. noted that COSMOtherm was able to predict More commonly, however, SRs of SVOCs in PAS are determined
the KXAD–G at different temperatures with a root mean square from the slope of an uptake curve.26–28 Assembling such a curve
error of 0.35,235 when compared with the experimental values by requires the deployment of multiple samplers at the same time
Hayward et al.223 and their staggered retrieval aer different lengths of time Dt.
Being based on multiple PASs deployed for variable lengths of
time, such calibrations tend to give more reliable SRs and allow
E.3 Comparison of uptake capacity of different PASs for HCB for a recording of the uptake curve, which in turn can provide
The overall uptake capacity of a PAS is not only dependent on information on the approach to equilibrium.
the sorption coefficient between the PAS sorbent and the gas- In order to linearise an uptake curve during a calibration that
phase, but also on the amount of the sorbent in the PAS, experiences variations in CG, one either plots the amount found
namely by the products KSGVS and KSGAS for PASs based on in the PASs, mS, against the product of CG and Dt (see Fig. 8) or,

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1945
Environmental Science: Processes & Impacts Critical Review

Table 6 Estimation of the total uptake capacity of different PASs for hexachlorobenzene at 20  C

Name KSG Mass of sorbent AS or VS Uptake capacity

PDMS-PAS111 KPDMS–G ¼ 105.9 (ref. 212) 10
5 m3 8 m3

LDPE-PAS67 KPE–G ¼ 106.8 (ref. 74) 10
6 m3 6 m3
PUF-PAS24 KPUF–G ¼ 107.1 (ref. 213, 216 and 219) 4.4 g of PUF (10
6 m3 g
1) 4.4  10
6 m3 55 m3
SIP-PAS136 KXAD–G ¼ 30 m (ref. 222)a 0.435 g of XAD-4 (750 m2 g
1) 315 m2 9000 m3
XAD-PAS26 KXAD–G ¼ 30 m (ref. 222)a Long: 20 g of XAD-2 7000 m2 200 000 m3
Short: 10 g of XAD-2 (350 m2 g
1) 3500 m2 100 000 m3
a
Adsorption coefficient has units of length (m is metre).

equivalently plots mS/CG, i.e. the “equivalent sampling volume”, atmosphere or for the particle-bound substance. See Section F.6
against the time of deployment. In either case, the slope of the for in-depth discussion of particle uptake in PAS.
resulting curve corresponds to the SR. F.1.2 Outdoor calibration experiments for PAS. Tables are
Different methods can be applied to derive a SR from such used to summarise outdoor calibrations studies that have been
a plot (Fig. 8).190 Each PAS can be treated as a single calibration conducted for the PUF-PAS (Table 7), XAD-PAS (Table 8), LDPE-
experiment yielding an empirical SRe according to eqn (13). The PAS (Table 9), SIP-PAS (Table 10) and SPMD (Table 11). The PUF-
overall SRe is then the average of the SRe obtained for each PAS. PAS has been calibrated for more different SVOCs than any
Alternatively, one performs a least square regression on the other PAS. This includes OCPs, CUPs, PCBs, parent PAHs,
linearised uptake curve and derives the SRe from its slope. alkylated PAHs, PBDEs, polychlorinated dibenzo-p-dioxins and
Finally, it is also possible to derive an SRe for each time period -furans (PCDD/Fs), polychlorinated naphthalenes (PCNs), poly-
between subsequent PAS retrievals, by using the increase in the cyclic musk compounds, BFRs and a number of peruorinated
mS in PASs from one retrieval to the next. The latter approach is alkyl substances (PFAS) (Table 7). The XAD-PAS has been cali-
advisable if the SR can be expected to vary during the deploy- brated for OCPs, CUPs, PCBs, PAHs, neutral PFAS and VMS
ment periods, e.g. if different deployment periods experienced (Table 8). The number of analytes for which empirical SRs have
different temperatures or wind speeds. Generally, a linear been determined in other PASs is much smaller (Tables 9–11).
regression gives the most reliable SRe. Longer deployments tend Air concentrations during the calibrations were determined
to have a stronger impact on the slope of the regression than either with sporadic high volume (HiVol) or more continuous
short deployments, which is oen desirable as the larger mS in low volume (LoVol) samplers. The fraction of time that is
PASs deployed for a longer length of time can oen be quanti- covered by the active sampling ranges from a few percent to
ed more condently. 100%.
For PASs that also take up particles, CG in eqn (13) can be Despite a large number of studies, the PUF-PAS has been
replaced with the concentration of both gaseous and particle- calibrated mostly in temperate climates, with one calibration
bound SVOCs or of only the particle-bound substance, and study in tropical Singapore.238 The outdoor calibration studies
a calibration then will yield SRs for the total SVOCs in the for the XAD-PAS also extend to locations in tropical,27,197

Fig. 8 Illustration of the three different methods of deriving a sampling rate SR from a calibration experiment involving n PASs deployed at the
same time and retrieved after variable deployment periods (n ¼ 6 in the displayed example) as discussed in ref. 190. In each case, the amount
taken up in a PAS mS is plotted against the product of the length of deployment t and the gas-phase concentration during the time of deployment
CG. The SR is derived either (i) as the average of the SRs calculated for each of the n samplers, (ii) from a linear regression involving all n samplers,
or (iii) from the increase in mS between subsequently retrieved samplers. Only the last method allows for SR to be time-variant.

1946 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
 Table 7 Outdoor studies calibrating the PUF-PAS for semi-volatile organic compounds

Sampler Site/type/climate Period SVOCs Active sampling Passive sampling Ref.

PUF-PAS Regina, Saskatchewan Jun. to Jul. and Jul. Dicamba, g-HCH, 2,4-D, Eight 168 hour Hivol 8 samples (duplicates at 10 241
Canada, rural, to Aug. 2003 triallate, bromoxynil, samples, borosilicate m and 30 m during two
Critical Review

temperate MCPA microbre lter and PUF- periods of 28 days)

XAD-PUF sandwich
PUF-PAS Lancaster, UK, semi- Apr. to Jun. 2006 HCB, HCHs, DDTs, Eight 168 hour HiVol 23 samples (1, 2, 3, 4, 6, 8 135
rural temperate chlordanes, PCBs samples using GFF/PUF- weeks) (triplicates during
PBDE PUF rst six weeks, 8 samples at 242
week 8)
PUF-PAS (low & Oct. to Dec. 2007 PCBs, PBDEs Thirteen 168 hour samples 8 triplicated samples (0.5, 243
high density) using GFF/PUF-PUF 1, 2, 3, 4, 6, 8, 12 weeks)
PUF-PAS Košetice, Czech Oct. 2003 to Jan. PeCB, HCB, HCHs, DDTs, Hundred-sixty-eight 24 42 samples (consecutive 129
Republic, sub-urban, 2007 (3 year period) PCBs, PAHs hour HiVol samples using periods of 28 days each)
temperate GFF/PUF-PUF
PUF-PAS Barcelona, Spain, 3 Mar. to Jun. 2005 PCDD/Fs, PCBs, PCNs HiVol samples using quartz 4 samples, 3 months 236

This journal is © The Royal Society of Chemistry 2020

urban/industrial sites, 1 bre lter/PUF
semi-rural site,
Mediterranean
PUF-PAS Iowa, United States, Aug. 2006 to Oct. PCBs Three 8 hour HiVol 3 quadruplicated samples 209
semi-urban, temperate 2006 samples using GFF/XAD-2 (21, 28, 46 days)
PUF-PAS Singapore, urban/ Apr. to Jun. 2008 PAHs, OCPs Twelve 48 hour HiVol 6 duplicated samples (10, 238
industrial, tropical samples using quartz bre 21, 32, 43, 60, 68 days)
lter/PUF
PUF-PAS Toronto, Ontario Feb. to Jun., 2008 PAHs, PCBs, PBDEs, musks Five 600 hour LoVol 10 samples (0, 25, 50, 75, 25
Canada, semi-urban, (ADBI, ATII, HHCB, AHTN) samples using GFF/PUF- 90, 100, 125 days), 3
temperate PUF duplicates
PUF-PAS Toronto, Ontario Mar. to Oct. 2010 PFAS (PFBS, PFOS, PFDS) Thirty-eight 24 hour HiVol 13 samples (7, 21, 28, 42, 137
Canada, semi-urban, VMS not detected samples using GFF/PUF- 56, 84, 112, 140, 168, 197 149
temperate XAD-2-PUF days), 3 duplicates
PUF-PAS Bursa, Turkey, urban, Jun. 2008 to Jun. PCBs, OCPs Thirty 24 hour samples 3 samples (10 days each) 183
modied Mediterranean 2009 using GFF/PUF-PUF
PUF-PAS Yakima Valley, Mar. to Apr. and Chlorpyrifos, azinphos- Five 24 hour active air 30 days 244
Washington USA, rural/ Jun. to Aug. 2011 methyl, chlorpyrifos-oxon, samples using PUF
agricultural, temperate azinphos methyl-oxon
PUF-PAS Brno, Czech Republic, Sep. 2010 PCBs, PeCB, HCB, DDTs, Twelve 168 hour LoVol 12 triplicated samples (1, 2, 245
urban-background, PBDEs, PAHs, PCDD/Fs, samples using GFF/PUF- 3, 4, 5, 6, 7, 8, 9, 10, 11, 12
temperate BFRs (ATE, TBECH, TBCO, PUF weeks)
p-TBX, PBEB, PBT, DPTE,
HBB, EHTBB, BTBPE,
DBDPE)
PUF-PAS Suwon, and Ansan, Feb. 2010 to Feb. PCDD/Fs, PCBs Suwon: six 168 hour HiVol Suwon: 8 samples; (63, 37, 246
Gyeonggi-do, South 2011 samples using GFF/PUF- 73, 79, 56, 62 days)
Korea, urban and PUF (28%)
industrial areas, Anasan: twenty four 168 Anasan: 3 samples (51, 56,
temperate hours HiVol samples using 66 days)
GFF/PUF-PUF
Environmental Science: Processes & Impacts

Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1947

 Environmental Science: Processes & Impacts Critical Review

subtropical210 and polar26 locations. The longest deployment

Ref.

247

248
249

250

251

252
period in most of the PUF-PAS calibrations was three months.
Reecting the higher uptake capacity of the XAD-PAS and
therefore longer tlinear, most of the calibrations involving the
XAD-PAS lasted one year, except for the one for VMS, which

6 duplicated samples (1, 2,

5 triplicated samples (16,
each with 3 to 5 samples

7, 21, 28, 42, 56, 84, 112,

duplication of 3 samples
1 sample of 60 days at 3

(10, 20, 30, 40, 60 days),

4 seasonal calibrations, lasted three months.28

replication of 10 day
F.1.3 Indoor calibration experiments for PASs. Table 12

30, 50, 73, 92 days)

140, 168, 197 days,

3, 4, 6, 8 months)
Passive sampling

summarises the indoor calibrations studies that have been

conducted for PUF-PAS. The PUF-PAS has been calibrated for
PCBs, PAHs, PBDEs and a number of PFAS. Indoor calibrations
samples

have also been performed for the uptake of PCBs in the XAD-
sites

PAS191,192,205 and neutral PFASs239 and phthalates and PBDEs240

in the SIP-PAS.
Twenty-one 24 hour HiVol

Sixteen active air samples

Sixteen active air samples

F.2 Explanations of the variability in empirically determined
Active sampling (2 days),

samples using GFF/PUF-

sampling rates
unknown number of
21 HiVol samples

The calibrations studies detailed in Sections F.1.2 and F.1.3

Active sampling

revealed that SRe vary considerably between chemicals and

between sampling sites and even between different deployment
samples

periods at the same site. This has been well documented for the
XAD

SRe values of PCBs, PBDEs and PAHs in the PUF-PAS.25 Herkert

et al.126 also compiled the ranges of the SRes reported for PCBs
in the PUF-PAS, noting large differences within and between
studies. For a long time, the prevailing assumption in the SVOC-
Levoglucosan, galactosan,

PAS community was that uptake in PASs is air-side controlled,

PAHs, alkylated PAHs,

i.e. the diffusion through air to the sampling sorbent is the rate-
parent and alkylated
dibenzo-thiophenes

limiting step to the uptake of chemical in the sampler. If this

were true, we would expect only minor differences in the SR
between different chemicals at the same place and during the
mannosan

same deployment, as long as the sampler is operating in the

SVOCs

OPFRs

SCCPs
PAHs
PCBs

linear uptake phase. The reason is that the effective thickness of

the air boundary layer surrounding the PAS sorbent (and
therefore the diffusion path length) is the same for all chemicals
and differences in the SR would only arise from differences in
the diffusion coefficients. Diffusivities in air are related to
Mar. to Oct. 2010
Feb. 2013 to Feb.

Sep. 2016 to May

Dec. 2010 to Jan

Jul. to Oct. 2016

molecular size, but the dependence is relatively weak and the

range of molecular size of SVOCs is limited, so we would expect
only slightly faster SRs for smaller SVOCs than larger ones.
Period

2011

2014

2017

Similarly, diffusivity in air is dependent on temperature and we

might expect slightly faster SRs at higher deployment temper-
atures than during deployments at low temperatures.
Larger differences in SRs between different sites are easier to
Alberta Oilsands region,

reconcile with air-side controlled uptake, as the boundary layer

Dalian, China, urban,
suburban, temperate

Guangzhou, China,
urban, subtropical

thickness can be expected to vary strongly with air turbulence,

Toronto, Ontario,
3 locations in the

boreal, industrial
Site/type/climate

and windy sites should have higher SRs for all chemicals than
Bursa, Turkey

those experiencing less turbulence. This has indeed been

temperate

observed, both in the eld126,129,194,195 and during controlled

experiments in the laboratory.26,48,128,191 This is also the reason
for the generally much lower SRs observed during indoor
compared to outdoor deployments. Other factors also affect the
diffusion distance in air, namely the sampler conguration192
Table 7 (Contd. )

and the presence and design of a housing,193 and thereby

contribute to variability in SR between different samplers and
modied

sampler congurations.
PUF-PAS

PUF-PAS

PUF-PAS

PUF-PAS

PUF-PAS
Sampler

There are a number of reasons, why SRes can differ (i)

between different chemicals during the same deployment and

1948 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
 Table 8 Outdoor studies calibrating the XAD-PAS for semi-volatile organic compounds
Critical Review

Sampler Site/type/climate Period SVOCs Active sampling Passive sampling Ref.

XAD-PAS (20 cm sampler) Point Petre, Ontario, May 2000 to May 2001 HCB, a-HCH Data taken from ongoing 7 duplicated samples (1, 2, 26
Canada, background, active monitoring 4, 6, 8, 10, 12 months)
temperate
Burnt Island, Ontario, Jun. 2000 to Jun. 2001 HCB, a-HCH, g-HCH,
Canada, background, dieldrin, trans-nonachlor
temperate
Alert, Nunavut, Canada Aug. 2000 to Aug. 2001 HCB, a-HCH, g-HCH,
background, polar endosulfan I
XAD-PAS (20 cm sampler) San Antonio de Belen, Oct. 2005 to Oct. 2006 HCB, chlorothalonil, Nine 24 hour HiVol 5 duplicate samples (4, 6, 8, 27
Costa Rica, suburban, chlorpyrifos, dieldrin, samples using GFF/PUF- 10 12 months)

This journal is © The Royal Society of Chemistry 2020

tropics pendimethalin, XAD-2-PUF
endosulfans, chlordanes
8:2 FTOH, 10:2 FTOH, 197
MeFOSA, EtFOSA, MeFOSE
XAD-PAS (10 cm sampler) Maun, Botswana, semi- Passive: May 2006 to May HCHs, HCB, Twenty-seven 24 hour 1 duplicated sample (1 210
urban, semi-arid 2007, active: Jul. 2006 to chlorothalonil, chlordanes, HiVol samples using GFF/ year)
Aug. 2007 endosulfan I, endosulfan PUF-XAD-2-PUF
II, endosulfan sulfate
XAD-PAS (10 cm sampler) Egbert, Ontario Canada, Mar. 2006 to Feb. 2007 PAHs, PCBs Twenty-one 336 hour LoVol 5 duplicate samples (56, 21
rural, temperate PFAS (8:2 FTOH, 10:2 samples using PUF-XAD-2- 122, 185, 241, 365 days) 197
FTOH, MeFOSA, EtFOSA, PUF
MeFOSE)
Chlorthalonil, endosulfan 190
I, endosulfan II,
endosulfan sulfate,
dacthal, disulfoton,
atrazine, alachlor,
metolachlor, HCB, HCHs,
chlordanes
XAD-PAS (10 cm sampler) Toronto, Ontario Canada, Mar. to Jun. 2012 VMS (D4, D5, MDM, Forty-two 57 hour LoVol 8 duplicate samples (7, 14, 28
suburban, temperate MD2M, MD3M) samples using SPE (PE 21, 28, 42, 60, 77, 98 days)
cartridge with 120 mg of
ENV + resin)
XAD-PAS (XAD-4 resin) 1 urban site and 1 rural site Jun. 2013 to Jun. 2015 45 PAHs 72 hour HiVol samples Consecutive 3 months 253
in Minnesota, USA, once every 12 days samples
temperate

Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1949

Environmental Science: Processes & Impacts
Environmental Science: Processes & Impacts Critical Review

Table 9 Outdoor studies calibrating the LDPE-PAS for semi-volatile organic compounds

Sampler Site/type/climate Period SVOCs Active sampling Passive sampling Ref.

LDPE (lm Brisbane, Queensland, Apr. 2002 PAHs One 720 hour, one 1440 hour 3 duplicated samples 67
thickness Australia, semi-urban, to Jun. 2003 and one 2160 hour HiVol (30, 50, 90 days)
200 to 100 um) tropics sample using GFF/two
XAD-2 cartridges
LDPE sampler Providence, Rhode Island Nov. to Dec., HCHs, HCB, heptachlors, Six 72 and 96 hour samples 6 duplicated samples 75
USA, semi-urban, temperate 2012 chlordanes, endosulfans, using GFF/PUF-PUF (3, 7, 10, 14, 17, 21 days)
DDTs, PCBs, PBDEs

Table 10 Outdoor studies calibrating the SIP-PAS for semi-volatile organic compounds

Sampler Site/type/climate Period SVOCs Active sampling Passive sampling Ref.

SIP-PAS Lancaster, UK, semi-rural, May 2009 to PCBs, PBDEs Twenty-ve 336 hour HiVol 22 and more (duplicate or 150
temperate Apr. 2010 samples using GFF/PUF- triplicate); 4, 7, 14, 21, 28,
PUF 35, 42, 112, 182, 266, 350
days
SIP-PAS Toronto, Ontario, Canada, Mar. to Oct. PFAS (neutral & ionic) Thirty-eight 24 hour HiVol 13 samples (7, 21, 28, 42, 56, 137
semi-urban, temperate 2010 VMS (D3, D4, D5, D6, MDM, samples using GFF/PUF- 84, 112, 140, 168, 197 days), 149
MD2M, MD3M) XAD-2-PUF 3 duplicates
SIP-PAS Toronto, Ontario, Canada, Mar. to Oct. OPFRs Twenty-one 24 hour HiVol 7, 21, 28, 42, 56, 84, 112, 140, 250
suburban, temperate 2010 samples using GFF/PUF- 168, 197 days, duplication of
XAD 3 samples

Table 11 Outdoor studies calibrating SPMD-PAS for semi-volatile organic compounds

Sampler Site/type/climate Period SVOCs Active sampling Passive sampling Ref.

SPMD Lancaster, UK, semi-urban, Summer, 1995 PCBs Thirteen 168 hour Six duplicated samples 38
temperate HiVol samples using (15, 29, 45, 59, 71, 88
GFF/PUF-PUF days)
Winter, 1995/1996 Twelve 168 hour HiVol Five duplicated samples
samples using GFF/ (14, 26, 39, 61, 84 days)
PUF-PUF
SPMD Lancaster region, UK, Passive: Nov. to Dec. PCDD/Fs Fieen 72 to 96 hours Two samples (42 to 45 39
urban/industrial/coastal, 1999 HiVol samples using days)
temperate Active: Jan. to Feb. GFF/PUF
1998
SPMD Brisbane, Queensland, Apr. 2002 PAHs One 32 day HiVol One quadruplicated 254
Australia, urban, tropics samples using GFF/XAD sample (32 days)
cartridge
SPMD Guangzhou, Guangdong, Apr. 2001 to Mar. PAHs Fiy-two 24-hour HiVol Four consecutive 3 255
China, urban, subtropics 2002 samples using GFF/ month periods
PUF-PUF
SPMD Weissuhjoch, Switzerland May 2005 to Nov. HCHs, Three 4 months LoVol Three consecutive 60
Zugspitze, Germany 2006 chlordanes, samples using XAD sampling periods (190,
Sonnblick, Austria, endosulfan, cartridge 210, 400 days)
background, alpine dieldrin

(ii) for the same chemical between deployments at different the uptake rate is expected to be proportional to the KSG of the
temperature. One reason is that air-side resistance may not be compound being sampled,20 i.e. less volatile compounds would
controlling (solely) the diffusive exchange between the ambient have a higher SR than more volatile compounds and the same
atmospheric gas-phase and the PAS sorbent. Zhang et al.205 have compound would be taken up faster at low temperatures than at
demonstrated the existence of a sampler-side resistance in high temperatures. Higher SRes for less volatile PCBs have for
a number of uptake experiments, which revealed a strong example been observed for SPMDs.38,135,209
gradient of SVOC concentrations within PUF disks and XAD-resin However, the situation can be even more complex when the
lled cylinders. If the sampler side resistance is rate-controlling, kinetics of the sorption is being taken into account.207 Model

1950 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

Table 12 Indoor studies calibrating the PUF-PAS for semi-volatile organic compounds

Sampler Site SVOCs Active sampling Passive sampling Ref.

PUF- PAS 8 offices and PBDEs LoVol samples using PUF- 17–20 days 161
laboratories, Ontario, MeFOSE, EtFOSE, PUF 162
Canada EtFOSA, MeFOSEA
PUF-PAS Office, Birmingham, PCBs Five 240 hours LoVol Eight samples (10, 20, 30, 134
United Kingdom samples using GFF/PUF- 40, 50 days), 2 replicates
PUF
PUF-PAS 2 laboratories, Iowa, PCBs Three 8 hours HiVol Three quadruplicated 209
United States samples using GFF/XAD-2 samples (21, 28, 46 days)
PUF-PAS (with Office, United Kingdom TBBP-A, a-HBCD, b- Five 240 hours LoVol Eight samples (10, 20, 30, 156 and
GFF for HBCD, g-HBCD, PBDEs samples (triplicate) using 40, 50 days), 2 replicates 256
particles) GFF/PUF-PUF
PUF-PAS (low Alloy factory (4 PAHs Two 168 hour LoVol 4 triplicated samples, 2 257
protection sampling points), samples using GFF/PUF- weeks
chamber) Sweden PUF at each sampling point
PUF-PAS 2 homes in each of PFAS (1MeFOSE, One 480 hours LoVol 7 samples (20 days) 239
Seoul, Ansan/Shiheung, EtFOSE) sample using PUF-XAD-
Daegu, Korea PUF at each site
PUF-PAS 5 living room, Korea 8:2 FTOH, 10:2 FTOH, 504 hour LoVol samples 21 days (unknown sample 258
EtFOSA, MeFOSE, using GFF/PUF-XAD-PUF number)
EtFOSE, MeFOSEA sandwich (unknown
sample number)
PUF-PAS (plug in University office, PCBs Six 720 hour LoVol samples 6 duplicated samples (0.5, 205
cylindrical Toronto, Canada using PUF-XAD-PUF 1, 2, 4, 6, 12 weeks)
housing)
PUF-PAS Lecture room, Brno, PCBs, OCPs, PBDEs, Twelve consecutive LoVol, 12 triplicated samples (1 to 259
Czech Republic novel BFRs, PAHs, QFF and PUF-PUF 12 weeks)
PCDD/Fs
PUF-PAS 2 university rooms, Pentabromobenzene, Continuous LoVol samples, 1st exp.: 0, 7, 14, 21, 28, 35 240
Toronto, Canada Pentabromotoluene, collected at weekly days, 2nd exp.: 7, 28, 49
Phthalates, PBDEs intervals days, duplicate
PUF-PAS (open Computer laboratory, Phthalates, PBDEs, Continuous LoVol samples 0, 10, 20, 30, 40, 50 days, 112
bottom) Toronto, Canada OPFRs collected at 10 day intervals triplicate

simulations indicated that both an increase and decrease in the The trend in (2) is opposite to the effect in (1), but applies to
SR with KSG and with temperature can occur. The SR of chem- SVOCs at the lower volatility spectrum, whereas the effect in (1)
icals that sorb strongly (high KSG) and slowly is predicted to applies to the more volatile SVOCs.
increase with molecular size and temperature, whereas the SR (3) Finally, if a chemical sorbed to the PAS sorbent is
of chemicals that sorb quickly and less strongly (low KSG) show undergoing degradation, it could appear as if the SRe is
the opposite behaviour.207 reduced. Evidence for this has been observed for selected
Finally, there are reasons why empirically determined SRe pesticide (chlorpyrifos, triuralin, pendimethalin) in XAD-PASs
can be compound and temperature-dependent, even if the deployed for long time periods.27,190 There is indeed evidence
assumption of air-side control is valid: that transformation of chlorpyrifos into its oxon-analog can
(1) Because of large differences in sampling capacity for occur when air is pulled through XAD-2 resin.260 When exposing
different SVOCs, more volatile chemicals will enter the curvi- parent PAHs and alkylated PAHs spiked onto PUF-disks to
linear region of uptake earlier and therefore have lower ozone levels equivalent to two months of deployment in a ow
apparent SRs than less volatile SVOCs, which maintain strictly tube, many compounds showed signicant reactive loss, espe-
linear uptake for a longer period of time, i.e. SRe < SR. For the cially at low relative humidity.261 When comparing the amounts
same reason, we might anticipate lower SRes at higher of SVOCs sequestered in four PUF-PASs deployed for four
temperatures than at lower temperatures, because the latter consecutive weekly periods with those in a PUF-PAS deployed
increases sampler capacity and lengthens the linear uptake simultaneously for one full month, much higher levels of
period. several PAHs, PBDEs and other BFRs in the PASs deployed for
(2) If an SVOC is partially particle-bound and the SR for shorter periods were taken as an indication that those
particles is lower than for vapours, we would expect lower substances were degraded during the longer sampling period.262
empirical SRes both for the less volatile SVOCs and at lower This implies that SRes for such degradable compounds would
temperatures. depend on the length of the deployments used during a cali-
bration study.

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1951
Environmental Science: Processes & Impacts Critical Review

F.3 Accounting for deviations from non-linear uptake

For the PASs with relatively small uptake capacity for the more
volatile SVOCs (e.g. those based on PE and PUF), some
researchers try to account for the non-linearity of uptake. If CG
is assumed to be constant, the integrated form of eqn (1) is:20
ku

mS ¼ VS CG 1
e
kl t (14)
kl

In the specic case, that the air boundary layer is controlling

the diffusive exchange between atmospheric gas-phase and PAS
sorbent:
 


SR
t
mS ¼ VS KSG CG 1
e KSG VS (15)
Fig. 9 Simulated uptake curve for a compound with log KPUF–G of 5.5
in a PUF-PAS during 90 days of constant (blue), increasing (red) and
This equation essentially describes the entire uptake curve in decreasing (green) air concentrations.
a PAS. Harner and colleagues139 routinely use a version of this
equation to estimate the effective sampling volume Vair of
chemicals in the PUF-PAS and SIP-PAS in the curvi-linear uptake will underestimate CG and vice versa. Or, stated differently,
phase: a PAS does not provide a true time-averaged concentration, if it
mS   operates in the curvi-linear uptake region.


SR
t
Vair ¼ ¼ KSG VS 1
e KSG VS (16) We illustrate the magnitude of these effects with PAS-SIM21
CG
calculations of a 90 day deployment of a PUF-PAS for a hypo-
thetical SVOC with a log KPUF–G of 5.5, i.e. a compound whose
Importantly, this equation for estimating Vair is imple-
uptake is curvi-linear within the deployment period. Three
mented in the widely used “Template for calculating PUF and
simulations have the same conditions including the same time-
SIP disk sample air volumes”263 and also in the model by Her-
averaged concentration CG during the 90 days. The only differ-
kert et al.,126 which is discussed in detail in Section F.5 below.
ence is that CG is either constant, increases from 0 on day 1 to 2
Lohmann and colleagues take the same approach with the PE-
times the average CG on day 90, or decreases from twice the
based PAS, when calculating the % of equilibrium that is ach-
average CG on day 1 to 0 on day 90. The uptake curves are very
ieved aer a deployment time t using (e.g. ref. 87):
different (Fig. 9) and importantly the amount taken up aer 90
mS days diverges substantially, which can be expressed by effective
% of equilibrium ¼  100%
VS KSG CG sampling volumes Vair of 60 m3 (constant CG), 89 m3 (increasing
 


SR
t CG) or 31 m3 (decreasing CG). Eqn (16) yields a Vair of 65 m3. This
¼ 1
e KSG VS  100% (17)
implies that the assumption of no sampler side resistance
implicit in eqn (16) in this case only leads to a minor error (5
While the use of eqn (14) through (17) may be considered the m3), largely because of the relatively high volatility of
state-of-the-art for calculating the effective sampling volume for a compound with a log KPUF-G of 5.5. On the other hand, the
situations where linear uptake for a compound during deploy- assumption of constant CG, also implicit in eqn (16), gives
ment cannot be assumed, it is important to be aware of the a large error (29 m3).
underlying assumptions. These equations are based on the Increasing or decreasing trends of CG during a PAS deploy-
assumption that the air-side is controlling the kinetics of ment period are common, considering the pronounced seasonal
uptake. While this is likely the case for some passive sampling variability of the air concentrations of many SVOCs, which is due
material, especially those with high permeability, this has either to seasonal source strength (e.g. pesticides elevated during
already been shown to not always be the case for the PUF-PAS.205 application season,264 PAHs elevated during heating season265),
When applying these equations, one also makes the temperature effects266 or oxidation by photooxidants.267 This
assumption that CG stays constant during the deployment issue is not restricted to the PUF-PAS, but applies to all PAS
period of the PAS. If sampling is conducted during the curvi- operating in the curvi-linear uptake phase. While the precise size
linear uptake phase, it means that the rate of uptake is gradu- of the error depends on the particular combination of KSGVS,
ally decreasing during the sampling period, or in other words, deployment time and temporal concentration trend, and will
air concentrations at the beginning of the sampling period have oen be smaller than in the illustrative example of Fig. 9, it is
more impact on the amount of chemical accumulated in the reasonable to assume that in many instances applying eqn (16)
sorbent than those at the end of the sampling period. This and (17) will incur very substantial errors.
implies that the results will only give an unbiased account of CG It is further important to be aware that these equations
if CG is constant. If CG is increasing during deployment, the PAS require quantitative knowledge of both the PAS's uptake

1952 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

capacity KSGVS and inherent SR, which can be challenging to is that wind exposure affects the kinetics of loss of DCs to the
obtain as both are strongly dependent on meteorological same extent as it affects the kinetics of uptake of the target
conditions that also can be variable during the time of compounds.
deployment: Müller et al.120 were the rst to investigate the rate of loss of
– The SVOC's equilibrium partitioning coefficient KSG compounds spiked onto the sorbent of a PAS, specically the
between sampler sorbent and the gas-phase is a strong function stearin-coated bre cloth. A log–log linear relationship between
of the temperature of deployment Td. It is customary to correct the rate of loss and a compound's KSG was interpreted as indi-
KSG to the average temperature of deployment using the van't cating an air-side control on the rate of mass transfer between
Hoff equation and the internal energy of phase transfer between atmospheric gas-phase and the sorbent. Ockenden et al.41 and
sampling sorbent and the gas-phase DUSG. If DUSG is unknown, Booij and van Drooge37 introduced DCs to the use of SPMDs as
the enthalpy of vaporization DHvap is oen used to approximate PAS, arguing that they can be used to account for differences in
DUSG (e.g. ref. 87). The relationship between KSG and T is not exposure conditions. Söderström and Bergqvist48 conrmed
a linear one, so the use of an average Td is problematic if T is that the extent of loss of deuterated PAHs and 13C-labelled PCBs
increasing or decreasing during the deployment period (see also from SPMDs was related to their exposure to different wind
ref. 204). A further complication arises if the temperature of the speeds in a wind tunnel experiment. Farrar et al.93 observed that
PAS sorbent deviates from ambient temperature, because of the loss of different PCB congeners spiked into EVA-coated
radiative effects associated with the sampler housing.193,268,269 glass-PASs (POGS) increased with compound volatility, wind
– The inherent sampling rate SR (m3 per day), i.e. the SR speed, temperature and deployment time. Bartkow et al.68 and
during the linear uptake phase, is dependent on wind speed Moeckel et al.130 showed that this was also the case with LDPE-
(because of the effect on the thickness of the boundary layer) PASs and PUF-PASs. PUF-PASs and PE-based PASs are frequently
and also somewhat on temperature (because of its effect on the used with DCs, largely owing to the short tlinear for many SVOCs
diffusivity). Because SR is most oen deployment-specic, it is in these sorbents.
usually determined from the loss of depuration compounds The concentration of DCs in the ambient gas-phase CG is
spiked onto the sampler prior to deployment, which implies zero. Therefore, eqn (1) simplies to:
that a further set of assumptions needs to be met (see Section dmS
F.4 below). ¼
kl mS (18)
dt
In summary, it is important to keep in mind that even if the
assumptions underlying the use of eqn (14) through (17) were
correct, the interpretation of PAS results within the curvi-linear In integrated form:
region incurs very signicant uncertainties. As such, it is
mS ¼ mS0e
klt (19)
generally preferable to use a sampler that remains in the linear
uptake region. The XAD-PAS has not been used in the curvi- Therefore, from the amount of the DC in the PAS at the
linear uptake region, because it is generally acknowledged beginning of deployment mS0 and at the time of retrieval mS, the
that the sampler-side resistance is not negligible and the loss rate constant kl can be derived:
assumptions underlying the use of depuration compounds are
m 
violated.
ln
S
mS0
kl ¼ (20)
t
F.4 The use of depuration compounds in PAS
As was discussed in Section C.1, PASs for SVOC tend not to have Under air-side controlled kinetics, eqn (3) relates kl with the
a diffusive barrier in order to allow for SRs that are sufficiently inherent SR:20
high for reliable quantication. One of the consequences is that
DA
the SR of the PASs tends to be quite dependent on wind expo- SR ¼ ¼ kl KSG VS (21)
Dz
sure, especially if the sampling sorbent is not protected by
a wind shelter or if the shelter is ineffective in preventing the
wind from impacting the sorbent. Adopting an approach pio- Therefore, the SR for the conditions of the deployment can
neered within the passive water sampling community, some be calculated from the empirical kl of a DC using:
PAS users spike the passive sampling sorbent with compounds m 
S

ln
prior to exposure and derive information on the kinetics of mS0
SR ¼ KSG VS (22)
uptake from the extent of loss of those spiked compounds t
during the deployment. These compounds are referred to as
either depuration compounds (DCs) or performance reference
This implies that the KSG of the DC at the temperature of
compounds. Because it is necessary to distinguish the spiked
deployment needs to be known. SRs estimated with eqn (22) can
compounds from those taken up in the PAS sorbent during
be seriously awed if the actual temperature in the sampler
exposure, DCs are oen isotopically labelled SVOCs. Alterna-
housing deviates from ambient temperature and the latter is
tively, compounds are selected that are known to not be present
used to estimate KSG.268
in the sampled atmosphere. The idea underlying the use of DCs

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1953
Environmental Science: Processes & Impacts Critical Review

A special case occurs if target chemical and depuration the PUF-PAS, because it relies on a sampler-specic empirical
compound are the same, except that the latter is isotopically tting constant. These efforts have a variety of motivations,
labelled. In this case, the KSG in eqn (16) and (22) applies to the including conrming a comprehensive understanding of the
same compound under the same temperature conditions and is processes and factors controlling the uptake in a PAS,207 aiding
therefore identical. Accordingly, if we substitute the right side of the interpretation of calibration studies and monitoring data
eqn (22) for SR, eqn (16) simplies to: (e.g. by estimating whether the PAS is operating in the linear or
  the curvi-linear uptake phase), and informing the design of PAS
mS
Vair ¼ KSG VS 1
(23) campaigns (e.g. in terms of deciding on the appropriate
mS0
deployment length or the selection of DCs that will experience
the desired extent of loss).21,126,190,204
The ratio mS/mS0 should be in the range from approximately These models can also be used for estimating the uptake of
0.2 to 0.8, if the estimated SR should not have too high an a particular target compound during a particular deployment,
uncertainty.48 This implies that the DC should not be lost too e.g. by using data on variable temperatures, external wind
much (more than 80%) or too little (less than 20%). Because the speeds and ambient air concentrations.21,204 If the model aims
extent of loss of a DC depends on KSG, Td, wind speed and to simulate the mass transfer kinetics between atmospheric gas-
deployment length, it requires considerable prior knowledge of phase and PAS sorbent, they can be used to predict deployment-
the deployment conditions to choose the right DC. In practice, specic, inherent SR during the linear uptake phase.21,272 If the
multiple DCs with variable KSG are used, as this increases the model additionally allows for calculation of uptake in the curvi-
likelihood, that at least one of them is lost to the appropriate linear and equilibrium part of the uptake curve, it can be used
extent. In the case of low temperature deployments, DC would to calculate what is either called an apparent SR21,190 or an
have to be considerably more volatile than the target effective sampling volume.126,204 The effective sampling volume
compounds to achieve losses of 20% during deployment. In is simply the product of the apparent SR and the deployment
fact, typically applied DCs spiked onto PUF-PASs deployed at time.
low temperatures oen do not experience sufficient loss.126 In The models vary widely in terms of the processes being
the case of passive water sampling, the information contained considered. The rst such model by Zhang and Wania was the
in DCs being lost less than 20% and more than 80% should not most comprehensive in terms of treating the uptake process as
be ignored, which likely also applies to passive air sampling.270 a sequential diffusion through a stagnant air-side boundary
It is important to keep in mind that the assumption of air- layer and a porous sampling sorbent followed by the reversible
side resistance-controlled uptake is underlying the use of DCs. sorption onto the sorbent.207 The PAS-SIM model21 does not
Even if this assumption may not be valid, DCs can oen give explicitly consider the kinetics of the sorption process, because
a semiquantitative measure of the wind exposure that a PAS numerical values for the parameterization of this process for
sorbent experienced during a particular deployment and make SVOCs on PUF or XAD generally do not exist. The model by
some allowance for how this affects the SR. However, it would Petrich et al.272 and Herkert et al.204 furthermore does not
not be appropriate to use DCs in PAS, where the sampler-side include a description of transport within the sorbent (speci-
resistance is clearly contributing to the exchange kinetics cally the pores of the PUF), i.e. disregards the possibility of the
(XAD-PAS, SIP-PAS, PUF-PAS under some conditions). In any importance of a sampler side resistance and uses as a starting
case, most SVOCs would not be lost appreciably from the point that the “uptake of PCBs on a PUF-PAS can be modelled as
sorbent of the XAD-PAS during normal deployments, because of a function of the air-side mass transfer coefficient”. This is
its very high uptake capacity (Section E.3). somewhat questionable, considering both empirical205 and
Disadvantages of the use of DCs are the considerable cost of theoretical evidence207 to the contrary.
isotopically labelled compounds and restrictions on their use The models have also a number of things in common. They
during indoor deployments, i.e. it may not be possible to get require input of temporally resolved wind speed and tempera-
approval to deliberately release potentially toxic compounds ture data during a PAS's deployment period and calculate time-
into indoor living environments. They can also not be used, variant KSG for the target compound from these temperature
when the PAS extract is to be used for toxicological character- data using the van't Hoff equation. Armitage et al.21 directly
isation with in vitro bioassays (see Section G.5.5) (e.g. ref. 271). estimate KSG at a reference temperature and the internal energy
Furthermore, the oen very considerable uncertainty in the KSG of phase transfer between sorbent and the gas-phase DUSG
of the DCs propagates directly to the estimated SR. using ppLFERs215,223 (see Section E.2.2). Herkert et al.204 estimate
the KOA and DUOA for the PCBs rst and then use the empirical
single parameter linear free energy relationships between KPUF–
F.5 Theoretical estimation of sampling rates 24
G and KOA by Shoeib and Harner (see Section E.2.1) assuming
There have been several efforts to describe the uptake of SVOCs that DUOA equals DUPUF–G. For the PCBs, these two approaches
in PASs using numerical simulation models.21,126,190,204,207,272 give reasonably consistent results.204
Whereas the models by Zhang and Wania207 and Armitage Zhang and Wania207 and Armitage et al.21 estimate the mass
et al.21 could be applied to any sampler with a porous sorbent transfer coefficient through the air-side boundary layer using
and have been parameterised for the XAD-PAS and the PUF-PAS, molecular diffusivity D (estimated using the semi-empirical
the model by Petrich et al.272 and Herkert et al.126,204 is specic to Fuller–Schettler–Giddings equation) and the thickness of that

1954 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

layer Dz. Armitage et al.21 propose a Dz of 1.5 cm for stagnant air using model-derived SRs for the PUF-PAS with those obtained
conditions as it provides SRs for the XAD-PAS consistent with by HiVol sampling in three different studies.
observations.191 They then apply an empirical function to scale For completeness sake, we also note that there have been
the SR of the XAD-PAS with wind speed based on measure- attempts to predict SRs using quantitative structure property
ments.191 This function shows a stronger dependence on wind relationships. Van Mourik et al. linearly regressed empirically
speed below 1 m s
1 than above, which is also consistent with determined SR for the XAD-PAS against molecular mass in order
studies on other PASs using an effectively wind-sheltered to estimate SRs for chlorinated paraffins.277 Empirical SRs in
sorbent.273–276 A similar approach is used to parameterise the SPMDs were regressed against quantum-chemical descriptors
PAS-SIM model21 for the PUF-PAS relying on the relationship by Zhu et al.278,279 However, these are statistical relationships not
between SR and windspeed by Thomas et al.132 based on a mechanistic understanding of the sampling process.
Petrich et al.272 and Herkert et al.,126,204 on the other hand,
estimate the mass transfer coefficient through the air-side
boundary layer using an equation for laminar ow along a at
plate, which includes molecular diffusivity, the kinetic viscosity F.6 Sampling of SVOCs in the particle phase
of air, the length of mass transfer (taken to be the diameter of As was mentioned in Section F.2, sorption of SVOCs to the
the PUF disk), and the wind speed inside the PAS chamber. The atmospheric particle phase contributes to the variability in
latter is obtained from ambient wind speed using an empirical empirically determined SRs. Whether a PAS is taking up
linear relationship determined in wind tunnel experiments.128 particle-bound SVOCs or not appears to be quite strongly
This approach further requires a tted dimensionless constant, dependent on the sampler housing. For example, there is no
the so-called advective mass transfer coefficient g, presumably evidence of the uptake of particle-bound SVOCs in the XAD-PAS;
because the ow along the PUF disk is not really laminar. substances that are predominantly particle-bound in the
Different values of g, all based on tting empirical data of the atmosphere, such as PAHs with ve or more fused rings or more
loss of DCs, have been presented.126,204,272 The g value depends highly brominated PBDEs, tend to have levels below the limit of
on the precise formulation of the laminar ow equation, the detection in XAD-PASs.280 On the other hand, the PUF-PAS is
data set of DC loss rates being used (in particular the range of much more likely to take up particles, because the double bowl
wind speeds encountered during deployment) as well as the housing does not block wind very effectively. Chaemfa et al.
source of the meteorological data being used (e.g. measure- used microscopy to study the entrapment of particles within the
ments at the nearest airport or meteorological station or data PUF disks deployed in the double bowl housings.131 The PUF
estimated from a global weather reanalysis).126 disk mostly sampled the ne particle fraction (<1 mm) and the
In summary, both approaches ultimately rely to some extent size distribution of particles in the PUF was found to diverge
on tting to parameterise the mass transfer through the strongly from that obtained with an aerosol spectrometer. On
boundary layer. PAS-SIM uses empirical data to t the boundary the other hand, Markovic et al. reported that particle numbers
layer thickness under stagnant conditions,21 whereas Herkert and size distributions measured with a particle spectrometer in
et al. use the observed loss of DCs to t g.126 empty double bowls were similar to those measured outside,
Herkert et al. argue that models in combination with readily suggesting that the housing does not discriminate between
available meteorological data can be used to derive SR when DC- different particles.281
derived SRs are not available or are implausible.126 Mapping An extensive evaluation of the PUF-PAS for different particle-
model-estimated SRs on a global scale suggests that wind speed associated SVOCs suggested a poorer performance than for
is the decisive factor responsible for variable SRs, with higher gaseous compounds, which was attributed to “unpredictable
values over the oceans and lower values within forests. This is accumulation behaviour of particles”.245 SRs were reported to be
consistent with the discussion in Section C.1. They also claim to much lower and more variable for particle-bound substances
be able to use their model to estimate effective air sampling than for gas-phase compounds,25,129,245 which contrasts with
volumes Vair for compounds in the curvi-linear uptake other studies that have reported similar SR for particle-bound
regime.126 Incidentally, the PAS-SIM model could be similarly and gaseous polycyclic aromatic compounds.247,257 Because of
combined with meteorological data to estimate apparent SRs or the propensity to sample particles, the PUF-PAS has even been
Vair for specic deployments, while also taking the sampler side used as a sampler for atmospheric trace metals.282,283 However,
resistance into account. It is important to note, however, that SR for different metals differed widely, presumably because of
such estimations of deployment-specic Vair will inevitably their different size distributions and the different SR of particles
require the assumption of constant air concentrations during of different size.
deployment, the implications of which were discussed in Overall, the evidence as to the capability of the PUF-PAS to
Section F.3 and illustrated in Fig. 9. sample particle-bound substances reliably and reproducibly is
Different approaches have been chosen to evaluate a model's still inconclusive.284 The largest study evaluating the perfor-
ability to predict the uptake behaviour of SVOCs in a PAS. PAS- mance of the PUF-PAS for compounds in the particle phase
SIM has been evaluated by comparing theoretically determined concluded that in all assessment endpoints (detection, preci-
uptake curves with those obtained from a year-long calibration sion, ngerprinting performance, and SR consistency and
study of the XAD-PAS for PCBs and PAHs21 and pesticides.190 magnitude) the performance was poorer than for gas-phase
Herkert et al.204 compared concentrations of PCBs obtained compounds.245

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1955
Environmental Science: Processes & Impacts Critical Review

Table 13 Studies using PASs to measure the spatial variability of SVOCs around major suspected emission sources

Sampler Location, number of sites Time frame SVOCs Ref.

SPMD 13 sites around DEZA chemical Sep. 1998, 4 weeks PAHs 36

plant, Valasske Mezirici, Czech
Republic
PUF-PAS 9 sites around chemical factory, Jan. to Jul. 2004, six 28 day periods PCBs, OCPs, PAHs 289
Neratovice, 11 sites around coal
tar and mixed tar oils processing
plant, Valasske Mezirici, Czech
Republic
PUF-PAS 4 sites around municipal solid Mar. to Jun. 2005, 3 months PCDD/Fs, PCBs, PCNs 236
waste incinerator and power
plant, Barcelona, Spain
PUF-PAS 8 sites around a municipal solid Mar. to Jun. and Sep. to Dec. 2010 PCDD/Fs, PCBs, PCNs 285
waste incinerator, Tarragona, and 2011, 3 months
Spain
PUF-PAS 8 sites around a petrochemical Nov. 2014 to Jan. 2015, one 2 PAHs 290
and a chemical complex, month period
Tarragona, Spain
Membrane-enclosed 6 sites around steel complex, Aug. to Sep. 2005, 37 days PAHs 53
copolymer sampler Pohang, Korea
PUF-PAS 15 sites around an iron and steel Sep. 2006 to Jul. 2007, four 3 PCBs 291
making plant, Pohang, Korea month periods
SIP-PAS 12 sites around a municipal Jul. to Sep. 2009, 63 days PFAS 144
wastewater treatment plant and 4 Jun. to Aug. 2009, 55 days
sites close to two municipal solid
waste landll sites, Ontario
Canada
SIP-PAS 1 site at eight wastewater Aug. to Nov 2013, Jan. to Mar. PFAS, VMS, UV-lters 287
treatment plants and 1 reference 2014, two 2 month periods Synthetic musks 288
site each
PUF-PAS 15 to 17 sites around oilsands Nov. 2010 to Jan. 2016, thirtyone, 2 PAHs, alkylated PAHs, 247, 292
region, Fort McMurray, Alberta, month periods dibenzothiophenes and 293
Canada
PUF-PAS 5 sites in oilsands region, Fort Oct. to Nov. 2015, 2 month period PAHs, alkylated, 294
PAS-DD McMurray, Alberta, Canada nitrated and oxygenated
PAHs
PUF-PAS 5 sites around a secondary copper Apr. to Oct. 2011, two months PCDD/Fs, dlPCBs, PCNs 295
and aluminum metallurgical
facilities, China
PUF-PAS 13 sites around a municipal solid May to Aug. 2012, Oct. 2012 to Jan. PCDD/Fs, PCBs 286
waste incinerator, China 2013, two 3 month periods
PE-PAS 23 sites close to natural gas Feb. 2014, one 3 to 4 week period PAHs 69
extraction wells, Carroll County,
Ohio, USA
PUF-PAS 23 sites near ship-breaking Feb. 2013, one 1 week sample (7 to PAHs, PCBs, DDTs, 296
activities, Chittagong, Bangladesh 9 days) HCB, SCCPs
PUF-PAS 7 sites in an e-waste dismantling Sep. 2009 to Aug. 2010, four 3 Dechlorane plus 297
region, Taizhou, China month periods
PUF-PAS 5 sites around hazardous waste Summer 2014 to autumn 2016, HCHs 298
landll, Sabiñánigo, Aragon, twenty-three consecutive 1 month
Spain periods
PUF-PAS Municipal solid waste landll, Jun. and Oct. 2012, Jan. and Apr. PAHs, PCBs, OCPs 299
Novi Sad, Serbia 2013, four 1 month periods
PUF-PAS 5 sites within a chemical Sep. to Oct. 2016, one 1 month SCCPs 300
manufacturing plant, China period
PUF-PAS 12 sites around a coal-red power 2014/2015, four seasonal samples PAHs 301
plant on south-west coast of India
XAD-PAS 18 or 20 sites in different waste 2013/2014, 2 to 3 month periods PBDEs, PCBs 302 and
handling facilities, Norway 303
PUF-PAS 16 sites in urban area and Jan. to Mar. 2013 and Sep. to Nov. >1000 micro-pollutants, 304 and
vehicular waste processing areas, 2015, two 2 month periods incl. PCBs, PBDEs, 305
Northern Vietnam PAHs, BFRs methylated
PAHs

1956 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

Instead of looking at the uptake of particle-bound exceptions. PAHs are most frequently measured, but OCPs and
substances by the sorbent meant for gaseous compounds, PCBs are also commonly targeted. Occasionally, more unusual
others have sought to add a GFF to a PAS to explicitly sample target substances have been studied, such as short-chain chlo-
particle-bound substances.155–157,168,199 This is discussed in more rinated paraffins (SCCPs) in the Yangtze River Delta360 or three
detail in Section D.3.1.3. Yet others have proposed PASs solely monosaccharides in the province of Guangdong.251 While many
dedicated to particle-bound substances158 or to use different studies only sampled for one or two periods, some seek to
sampler orientation and sorbents to distinguish total atmo- observe temporal patterns by sampling during multiple
spheric load from gaseous contributions.200 seasons. The number of sampling sites varies widely and ranges
from as little as three to more than 30.361–364
G. Applications of PASs for SVOCs G.1.3 National PAS network studies. A number of studies
involved networks of sampling sites across a country (Table 16).
G.1 PAS networks The scale of these studies varies obviously with the size of the
One of the most common uses of PASs is the recording of the country, ranging from as little as ve sampling sites in Lux-
spatial variability in time-averaged air concentrations of SVOCs. embourg385,386 to more than 90 sites within China.387,388 A
This can take the form of networks of samplers. These networks number of these studies were conducted in countries with no
vary widely in scale. prior measurements of SVOCs in the atmosphere and the PAS
G.1.1 Local and urban PAS network studies. Among the network served as an initial reconnaissance of atmospheric
PAS network studies with the smallest scale are those that seek SVOC contamination. Examples are Kuwait,389 Costa
to characterise the inuence of a known or suspected point Rica,280,390,391 Botswana,210 Ghana392,393 and Azerbaijan.394 Some
source on local atmospheric SVOC concentrations (Table 13). studies focus on a particular type of sampling environment, e.g.
The use of PAS networks for mapping concentration variability on cities239,395 or national parks,396 whereas most national scale
close to sources makes intuitive sense, because it is in those studies include a mix of remote, rural and urban sampling
areas that substantial heterogeneity of air concentrations can be locations. Whereas most of these studies analyzed classical
expected to occur. The facilities that have been studied this way SVOCs, such as PCBs and OCPs, some had a focus on chemicals,
include chemical plants, steel plants, power plants, petro- for which the national database of atmospheric measurements
chemical extraction areas, waste incinerators, sewage treatment may have been inadequate. Examples are studies on specic
plants, and landlls. The target chemicals are oen focused on CUPs in Canada,397,398 on dechlorane plus in China399 and SCCPs
the SVOCs that are expected to be emitted by the studied facility, and MCCPs in Australia.277 These studies typically relied on
for example combustion-related SVOCs such as PCDD/Fs, PCBs, PUF-PASs and XAD-PASs, with deployment periods ranging
PCNs from municipal waste incinerators,236,285,286 volatile PFAS, from 1 month to 1 year.
VMS and musk compounds from waste water treatment G.1.4 International regional network studies. Table 17
plants144,287,288 or parent PAHs and alkyl-PAHs in the oilsands compiles information on studies with PAS sampling sites across
region of Alberta, Canada247 and in the vicinity of natural gas an international region. Most of these have been done in parts
extraction wells.69 of Europe, but there are also two in Western Africa421,422 and
With more than 30 studies, networks of PASs across urban a number in East Asia.423,424 One network, relying on SPMDs, has
areas are very common (Table 14). This is not surprising as been measuring SVOC concentrations at 10 to 12 sites along
urban areas with their multitude of potential sources and large a transect from Southern UK to Northern Norway from 1994 to
gradients in land cover and human activities are likely to 2008, with an unusually long deployment period of 2
experience large heterogeneity in SVOC air concentrations. A years.40,42–45 Most of the other studies of this type used PUF-PASs
diverse set of urban areas around the world has been studied, and sampling periods ranging from 1 to 3 months. High
including major metropolitan areas in the USA, such as Phila- capacity sorbents, i.e. XAD-PAS and SIP-PAS, nd use when
delphia306 and Chicago,307,308 a number of Chinese cities, a year-long deployment is used.150,423 The studies using PE-PASs
including Beijing,309 Shanghai,310 Tianjin,311–313 and Harbin,314 around the Laurentian Great Lakes also could be classied as
as well as major cities in developing countries such as international regional studies, but they are discussed separately
Bangkok,46 Manila,315 Alexandria,73,74 Karachi and Lahore.316 in Section G.8.1, because they include paired air and water
However, some PAS networks have also been used to measure passive samplers.
air concentration variability in less well-known, mid-size cities, G.1.5 Continental PAS network studies. A few international
such as Manizales, Colombia,317,318 Ulsan, South Korea319,320 and studies were sufficiently large in scope to categorise them as
Aliaga, Turkey.321 PUF-PASs are most commonly used, but continental in scale (Table 18). An early network of XAD-PASs
SPMDs, XAD-PASs and PE-PASs have also been employed. ranged from the High Canadian Arctic to Central America and
G.1.2 Regional PAS network studies. Studies involving from the Pacic to the Atlantic coast of North America.29,432,434 A
networks of PASs in a region are also very common and have number of studies were Pan-European in scale95,434–436 or
been conducted around the world (Table 15). A region can be an covered large parts of East Asia.143,437–439 Two studies reported
administrative unit, such as a Chinese province251,346–348 or an data on African networks440,441 and a regional Global Atmo-
Italian region349–351 or it could be a country's coastal area352–355 or spheric Passive Sampling (GAPS) effort31,442,443 focussed on the
an entire or partial watershed.356–359 These types of studies have group of Latin American and Caribbean countries. While an
almost exclusively relied upon the PUF-PAS, with a few Australian national study277,414 could also be classied as

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1957
Environmental Science: Processes & Impacts Critical Review

Table 14 Studies using PASs to measure the spatial variability of SVOCs in the urban atmosphere

Sampler Location, no. of sites Time frame SVOCs Ref.

PUF-PAS 20 sites in Brno, Czech Oct. to Nov. 2004, 28 days PCBs, OCPs, PAHs 322
Republic
SPMD 2 sites in Genoa, Italy, April 2007 to May 2008, nine PAHs 59
industrial area 1 month periods (28 to 36
days)
PUF-PAS 7 sites in Liberec, Czech Dec. 2005 to Dec. 2006, PAHs 323
Republic thirteen 28 day periods
PUF-PAS 23 sites in Zurich, Spring 2011 and spring 2013, PCBs 324
Switzerland two periods
XAD-PAS 3 sites in Strasbourg, France Every spring between 2013 PAHs, PCBs, OCPs, CUPs 325
and 2016, eleven 2 week
periods
PUF-PAS 6 sites in Ottawa, Ontario, Dec. 2002 to ar. 2003, 3 PFAS 162
Canada months
PUF-PAS 32 sites in Philadelphia/ Apr. to Jul. 2005, 96 to 98 PCBs 306
Camden, USA days
PUF-PAS 5 sites in Fairbanks, Alaska, Dec. 2008 to Mar. 2009, 4 PAHs 326
USA months
XAD-PAS 26 sites in Toronto, Ontario, Jul. to Oct. 2012, 65 to 92 VMS 28
Canada days
PUF-PAS 21 sites in Cleveland, Ohio Aug. to Sep. 2008, 20 to 22 PCBs 307
and 17 sites in Chicago, days (Cleveland) and 13 to 47
Illinois, USA days (Chicago)
PUF-PAS 13 sites in Chicago, Illinois, Jan. 2012 to Jan. 2014, 180 PAHs, BFRs, OPFR, OCPs 308
USA samples of 43  11 days.
PUF-PAS 8 sites in Toronto, Ontario, Aug. 2016 to Aug. 2017, six 2 PAHs, alk-PAHs, nitro-PAHs, 327
Canada month periods Oxy-PAHs,
dibenzothiophene (DBT),
alk-DBTs
PUF-PAS 6 sites in Concepcion, Chile Jan. to Mar. 2007, 2 months PCBs, PAHs, OCPs 328
PUF-PAS 6 sites in Temuco, Chile Apr. 2008 to Apr. 2009, three PAHs 329
4 month periods
PUF-PAS 5 sites in Santiago, 6 sites in Apr. 2008–Aug. 2009, four 4 PCBs, OCPs 330
Concepcion, 6 sites in month periods
Temuco, Chile
PUF-PAS 4 sites in Santiago de Cali, May to Sep. 2011, one 4 PCBs, OCPs, PAHs 331
Colombia month period
PUF-PAS 4 sites in Manizales and 1 June 2012 to Nov. 2014, total PCDD/Fs, dlPCBs 317,318
site in Bogota, Colombia of 27 periods (90 to 120 days)
PUF-PAS 11 sites in Mendoza, Dec. 2010 to Apr. 2011, 90 PCBs, HCB DDTs, PBDEs, 332
Argentina days
SPMD 6 sites in Bangkok, Thailand Mar. to Apr. 2000, 3 weeks PAHs 46
PUF-PAS 6 sites in and around May to Dec. 2005, four PAHs 315
Manila, Philippines (3 periods of 42 to 56 days
urban, 3 rural)
PUF-PAS 3 sites in Manipur, India Jan. to Dec. 2009, one month OCPs 333
for a year PAHs 334
PUF-PAS 5 sites in Kolkota, 1 site in Dec. to Mar. 2014, one 3 PBDEs, OCPs, PCBs 335
Sundarban, India month period
XAD-PAS 6 sites in Karachi and 4 sites Jan. to Oct. 2011, 10 months OCPs, PCBs 316
in Lahore, Pakistan
PUF-PAS 6 sites in Kathmandu, 4 sites Kathmandu and Pokhara: PAHs 336
in Pokhara, 3 sites in Aug. 2014 to Aug. 2015, PCBs, OCPs 337
Hetauda in Nepal Hetauda: Nov. 2015 to Aug.
2016, six 2 month periods
PUF-PAS 6 sites in Shanghai, China Aug. 2006 to Jun. 2007, three PAHs 310
3 month periods
PUF-PAS 6 sites in Tianjin, China Jul. 2006 to Jun. 2008, one 8 OCPs, PCBs 311
months & ve 3 month
periods
PUF-PAS 8 sites in Harbin, China Feb, 2007 to Jan. 2008, four 3 PAHs 314
month periods

1958 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

Table 14 (Contd. )

Sampler Location, no. of sites Time frame SVOCs Ref.

PUF-PAS 11 sites in Tianjin, China Jun. 2008 and Jan. 2009, two PCBs, PCDD/Fs 312
1 month periods
PUF-PAS 13 sites in Beijing, China Feb. 2011 to Mar. 2012, four PCDD/Fs, dlPCBs, PBDEs 309
seasonal periods
PUF-PAS 28 sites in Nanjing, China Jan. to Oct. 2015, four 2 to 3 PAHs 338
month periods
PUF-PAS 7 sites in Ningbo, China Nov. 2014 to Oct 2015, twelve PAHs 339
1 month periods
PUF-PAS 8 sites in Ezhou, Wuhan, Jun. to Aug. 2015, one period PAHs 340
Hubei, China
PUF-PAS 20 sites in Dalian, Liaoning, Jul. to Sep. 2017, one 40 day PAHs 341
China period
PUF-PAS 33 sites in Tianjin, China Jul. to Sep. 2016, Nov. 2016 OPEs 313
to Jan. 2017, two 90 day
periods
PUF-PAS 20 sites in Ulsan, South Jan. to Feb 2011, 49 days PAHs 319
Korea
PUF-PAS 20 sites in Ulsan, South Mar. to May 2013, 84 days PAHs, Cl- PAHs, Br-PAHs 320
Korea
PUF-PAS 36 sites in Seoul, South Aug. to Sep. 2006, Jan. to PAHs 342
Korea Feb. 2007, two 1 month
periods
PUF-PAS 7 sites in Tangshan, China Jul. to Oct. 2010, 3 months OCPs 343
PUF-PAS 22 sites in Istanbul, Turkey Sep. to Dec. 2014, four 1 PAHs, PCBs 344
month periods
PUF-PAS 40 sites in Aliaga, Turkey Jul. 2009 to May 2010, four 2 PAHs, PCBs 321
month periods
PUF-PAS 8 sites in Bursa, Turkey Feb. to Dec. 2014, ve 2 PCBs 345
month periods
XAD-PAS 15 sites in Minneapolis, 2013–2015, several 3 month PAHs 253
Minnesota, USA periods
PE-PAS 11 sites in Alexandria, Egypt Jul. 2010 to Jan. 2011, 21 PAHs alkyl-PAHs, 73
days thiophenes
OCPs 74

continental, we have listed those studies in Table 16 in Section of Toronto, likely because this is an area of such clear transition
G.1.3 above. in population density and industrial and agricultural
G.1.6 Global PAS network studies. Table 19 lists the PAS activity.453–458 Harrad et al. sampled a transect across the Bir-
network studies that were truly global in scale. Most of these mingham region.459,460 There is obviously some overlap between
were conducted as part of the Global Atmospheric Passive transect studies sampling along an urban-rural gradient and the
Sampling (GAPS) project, which is relying on PUF- urban network studies in Section G.1.1, if those networks include
PASs,127,208,444–447 XAD-PASs197,448,449 and SIP-PASs.139,147,151–153 urban, suburban and rural sites (Table 14).
Another global scale study with a focus on Africa, Latin America, At the largest scale, PAS transects explore air concentration
the Caribbean, and the Pacic Islands also relied on the PUF- variability with latitude. These studies have already been
PAS.450 In some cases, rst long-term time trends could be mentioned in the context of network studies in Section G.1. Most
derived from samples taken in different years.153,197,448 notable among those are the transects in Northern
Europe,40,42–45,150 along the length of Chile,195,400 and across North
G.2 PAS transects America.29,432,433 The northern part of the latter transect from
Southern Canada to the High Arctic is particularly useful because
Another common application of PASs involves a transect of
there are few, if any, emission sources that would confound the
sampling sites along which a change in air concentration may
recorded air concentration gradient. This transect thus consti-
be expected. Again, the scale of such transects varies widely. At
tutes a gradient of source proximity or remoteness449,461 and lends
the smallest scale, PASs have been used to quantify the impact of
itself to the derivation of empirical travel distances.29
roadways and other traffic arteries on the concentrations of
SVOCs.451,452 Some of the local scale studies in Section G.1.1
seeking to quantify the inuence of specic point sources (Table G.3 PAS vertical gradients
13) have relied on linear transects. Quite frequently, the air Another frequent theme of PAS studies for SVOCs is their use in
concentration variability along urban to rural transitions (Table the determinations of vertical concentration gradients. Again,
20) has been studied, with most of these studies sampling North these studies have a range of spatial scales.

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1959
Environmental Science: Processes & Impacts Critical Review

Table 15 Studies using PASs to measure the spatial variability of SVOCs in a region

Sampler Location, no. of sites Time frame SVOCs Ref.

Europe
PUF-PAS 6 sites in Northern Estonia Mar. to Aug. 2006, four 4 week PCBs, OCPs 353 and
periods 354
PUF-PAS 19 sites in Tuscany, Italy Apr. to Jul. 2008, 60 to 160 days PAHs, OCPs, PBDEs, PCBs 349
PUF-PAS 10 sites in Tuscany, Italy Apr. 2008 to Jul. 2009, four 3 to 5 PCBs, OCPs 351
month periods
PUF-PAS 4 sites in Puglia, Italy Jan. 2009 to Feb. 2010, four PCBs, PAHs 350
periods of 2 to 5 months
PUF-PAS 8 sites in Sicily, Italy Jul. to Dec. 2007, two 2 to 3 month PCBs, OCPs, PBDEs 365
periods
PUF-PAS 7 sites in Molise, Italy Apr. to Jul. 2011, one 3 month PAHs 366
period
PUF-PAS 6 sites around the Mar Menor 2010, four 1 month periods in PAHs, CUPs 237
lagoon, Cartagena, Spain each season
PUF-PAS 3 sites around Bothnian Bay, Jul. 2011 to Jan. 2013, ve 3 to 5 OCPs, CUPs, brominated anisoles 367
Sweden month periods
PUF-PAS 150 sites in Naples, Campania, Jul. 2015 to Apr. 2017, three-four 3 OCPs, PCBs, PAHs 368
Italy month periods
PUF-PAS 4 sites in Sabiñánigo, Aragón, Feb. 2016 to Nov. 2018, 29 HCHs, PeCB, HCB 369
Spain consecutive 1 month periods

Africa
PUF-PAS 3 sites in Northern Algeria Jul. 2008 to Nov. 2009, 28 to 56 PCBs, CUPs, PCDD/Fs 370
days
PUF-PAS 3 sites along coastal zone of Jan. to Dec. 2008, 4 weeks and 12 OCPs 355
Ghana weeks
PUF-PAS 4 sites in Gauteng Province, South May 2016 to Jan. 2017, two 100 day PBDEs, novel BFRs 371
Africa periods

Americas
PUF-PAS 4 sites in Southern Mexico 2002 to 2004, 31 to 116 days OCPs, PCBs 372
PUF-PAS 23 sites in the Yakima valley, Mar. to Apr. and Jun. to Aug. 2011, CUPs 244
Washington, USA 5 to 30 days
PUF-PAS 8 sites in Bahia Blanca, Buenos 2006–2007, 105 to 155 days, 191 to PCBs, OCPs 373
Aires province, Argentina 230 days
PUF-PAS 6 sites in Córdoba, Argentina Nov. to Dec. 2014, one period (44 PAHs, PCBs, OCPs, PBDEs 374
to 50 days)
XAD-PAS 10 sites in Quequén Grande basin, May 2013 to Sep. 2014, three 4 OCPs, CUPs, PCBs 357
Buenos Aires province, Argentina month periods
XAD-PAS 9 sites around Bahia Blanca, Jan. to Dec. 2015, four seasonal PAHs 375
Buenos Aires province, Argentina samples

Asia (other than China)

PUF-PAS 10 sites in Punjab, Pakistan Jan. to Mar. 2011, one 8 week OCPs 376
period
PUF-PAS 6 sites in the Indus Basin, Sep. to Oct. 2013, 56 days OCPs 356
Pakistan
PUF-PAS 6 sites in the Indus Basin, Oct. to Dec. 2012, 8 weeks PCBs 377
Pakistan
PUF-PAS 32 sites in Azad Jammu and Nov. 2016 to Jan. 2017, 8 weeks OCP, PCBs 364
Kashmir region, Pakistan
PUF-PAS Unknown number of sites in Unknown number of consecutive PAHs 378
Punjab, Pakistan 56 day periods
PUF-PAS 32 sites in Tamil Nadu, India Apr. 2009 to Jan. 2010, three PAHs 361
seasonal periods
PUF-PAS 9 sites within the Ganga River Apr. to Sept. 2013 (3 to 4 months) PAHs 358
basin, India and Feb. to May 2014 (1 to 2
months)
PUF-PAS 7 sites at lower reaches of Ganga Mar. to Apr. 2015, one 40 day OCPs 359
River, India period
PUF-PAS 19 sites along the south coast of Jan. to July 2006, two 3 month Coplanar PCBs, PCNs, PBDEs 352
South Korea periods

1960 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

Table 15 (Contd. )

Sampler Location, no. of sites Time frame SVOCs Ref.

PUF-PAS 10 sites in Gyeonggi Province, Mar. 2011 to Feb. 2013, 80 to 100 PCBs, PCDD/Fs 379
South Korea days
PUF-PAS 20 sites in Ulsan, South Korea Feb. 2011 to Feb. 2012, four 3 PAHs 380
month periods
PUF-PAS 41 sites in Kutahya, Turkey Jan. to Mar. and Jun. to Aug. 2014, PAHs, PCBs 363
two 2 month periods
PUF-PAS 4 sites in Kumluca, Antalya, May to Nov. 2013, 11 periods (1 to OCPs, PCBs 381
Turkey 3 months) at one site, two 3 month
samples at three sites
PUF-PAS 23 sites in Dilovasi district, Feb. 2015 to Feb. 2016, twelve 1 PAHs, PCBs 382
Kocaeli, Turkey month periods

China
PUF-PAS 40 sites in Hebei province 2007 to 2008, 46 to 123 days PAHs 346
PUF-PAS 7 sites in Hubei province Apr. 2012 to Mar. 2013, four OCPs 347
seasonal samples
PUF-PAS 14 sites in Gansu province Nov. 2010 to Mar. 2011, Jun. to PAHs 348
Sep. 2011
PUF-PAS 10 sites along the Lanzhou valley, 2013, two periods of 107 and 100 PAHs 383
Gansu province days
PUF-PAS 31 sites in Yangtze River Delta, Jul. 2011 to Jun. 2012, four PAHs 362
Jiangsu province seasonal periods
two seasonal periods SCCPs 360
PUF-PAS 11 sites within Guangdong Jan. to Apr. and Jul. to Sep. 2015, Levoglucosan, mannosan, 251
province two 2 month periods galactosan
PUF-PAS 18 sites along the Lanzhou valley, Winter 2014, summer 2015, two Nitrated PAHs 384
PAD-DD Gansu province 12 week periods

G.3.1 Small scale gradients above soil. A number of volatilisation and deposition, respectively. No attempt was
research groups have sought to quantify small scale (<2 m) made to account for the possibility that the SR is reduced so
vertical gradients of SVOCs above soil (Table 21). The rationale close the ground surface, even though a photograph of one of
is to deduce the direction and extent of diffusive air–soil the sampling sites shows a vegetation cover. Wang et al. used
exchange. This is a particularly challenging application of PASs, a slightly modied version of that sampler with double the
because the concentration differences on such a small scale surface area to record concentration gradients above a Tibetan
may be too small to be condently established, considering the pasture, but did use DCs to correct for differences in the SR with
limited precision of most PASs. This is particularly so, because height.171 Strangely, while SRs varied considerably (ranging
wind speeds decrease close to the ground, and therefore the SR from 1 to 3 m3 per day), they showed no apparent relationship
of a PAS will also decrease. It thus needs to be conrmed that with height. a-HCH and o,p0 -DDT showed lower levels close to
sequestered amounts that increase with PAS deployment height the ground during both summer and winter, indicative of
are not due to a gradient in the uptake kinetics, but truly reect deposition.
a depositional ux of the SVOC. In most cases, this will require Since the largest concentration difference can be expected to
the use of DCs to quantify the extent to which wind exposure of occur between the bulk atmosphere and the air right next to the
the PASs varies with height. soil surface, Donald and Anderson proposed to only sample at
This application of PASs has spawned a whole range of those two heights using PE strips placed in custom-made metal
custom-designed PAS, presumably because the most commonly boxes.70 DCs are used to correct for variable atmospheric
used PASs are too bulky and their sampling height too poorly turbulence in the boxes, which was much higher in air than in
constrained for nely resolved measurements. In particular, the soil air samplers. Above two contaminated soils, air concen-
PUF-based samplers used for ne-scale vertical gradients have trations at ground were found higher than at height, indicative
small housings that envelop the foam disk and are only open at of a potential for evaporation. At one site, the gradient was
the bottom. As a result, they have lower SRs than the regular PUF- reversed, indicating that PAHs and PCBs are likely to deposit.
PAS.169–171 Air–soil exchange uxes were estimated using molecular diffu-
Zhang et al. observed levels that decline with height at sion across a stagnant boundary layer presumed to be 1 mm
contaminated site and levels that increase with height at two thick. More recently, this approach has been used to measure
other sites, but those gradients were only apparent very close to potential emissions of chemicals from articial turf.71
the soil surface (<3 cm).169 They were interpreted as indicating

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1961
Environmental Science: Processes & Impacts Critical Review

Table 16 Studies relying on PAS network for SVOCs on a national scale

Sampler Location, no. of sites Time frame SVOCs Ref.

PUF-PAS 6 sites in Chile Dec. 2002 to Feb. 2003, 2 months OCPs, PCBs 400
XAD-PAS 20 sites in Chile Feb. 2006 to Jul. 2007, 1 year OCPs, CUPs, PAHs, PCBs 195
PUF-PAS 14 sites in Kuwait Feb. to Apr. 2004, 6 weeks PBDEs and PAHs 389
XAD-PAS 23 sites in Costa Rica Feb. 2004 to Feb. 2005, 1 year OCPs 390
PAHs 280
CUPs 391
XAD-PAS 7 sites in Costa Rica Oct. 2005 to Oct. 2006, 1 year OCPs, CUPs 401
PFAS 197
PUF-PAS 11 sites in Mexico 2005–2006, up to four consecutive 3 to 4 OCPs 402
month periods
PUF-PAS 8 sites in Canada (PEI, QC, ON, SK, BC) 2004 to 2005, 1 month Dacthal 397
PUF-PAS 10 sites in Central and Northern May to Aug. 2005, May to Herbicides 398
Canada (SK, AB, NV, NWT) Aug. 2007, 78 to 106 days
PUF-PAS 97 sites in China Jul. to Oct. 2005, 3 months PCBs 387
Dechlorane plus 399
PUF-PAS 40 sites in China Feb. to Mar., Apr. to May, Jul. to Aug., PAHs 403
Sep. to Nov. 2005, 8 weeks OCPs 404
PUF-PAS 90 sites in Northern China Jun. to Oct. 2011 PBDEs, TBE, DBDPE 388
PCNs 405
Bromomono-aromatics 406
PAHs, nitro-PAHs 407
PUF-PAS 18 sites in coastal India Jul. to Sep. 26, 2006, 6 weeks PCBs, PBDEs 408
PUF-PAS 68 sites in 7 Indian cities Dec. 2006 to Mar. 2007, 28 days OCPs 395
20 sites in 3 Indian cities, Winter 2006 (India) SCCPs, MCCPs 409
10 sites in Pakistan Winter 2011 (Pakistan)
PCNs 410
PBDEs 411
PUF-PAS 9 sites in Indian agricultural areas 2006–2007, four 3 month periods PCBs, OCPs 412
PUF-PAS 8 sites in Australia (WA, QLD, SA) Jan. to Feb. and Jul. to Aug. 2007, 40 to 50 days PAHs 268
PCBs 413
XAD-PAS 15 sites in Australia Jan. 2012 to Mar. 2013, 1 year (325 to 414 days) OCPs, PCBs 414
XAD-PAS 15 sites in Australia Jan. 2016 to Jun. 2017, 1 year (302 to 408 days) SCCPs, MCCPs 277
XAD-PAS 15 sites in Botswana May 2006 to May 2007, 1 year OCPs 210
PFAS 197
XAD-PAS 36 sites in 19 USA National parks Summer 2005-summer 2006, 1 year OCPs, CUPs, PAHs 396
PUF-PAS 13 locations in Azerbaijan Nov. to Dec. 2008, 4 weeks PCBs, OCPs 394
SIP-PAS PUF-PAS 10 cities in Korea Apr. to Jul. 2009, 90 days PFAS 239
PUF-PAS 55 sites in Japan Mar. to May, Aug. to Oct., Nov. 2008 to PCBs, PCNs 415
Jan. 2009, 8 weeks
PUF-PAS 13 sites in Ghana + 2 e-waste sites May to Jul. 2010, 56 days PCNs 392
OCPs 393
PCBs 416
XAD-PAS 5 sites in Luxembourg Jun. to Oct. 2008, 6 to 87 days Pesticides 385
PAHs 386
PUF-PAS 15 sites in Vietnam Jun. to Aug. 2012, Dec. 2012 to Feb. 2013, PCBs, OCPs 417
two 6 week periods
PUF-PAS 34 sites in 4 Nepalese cities Aug. 2014 to Oct. 2014, one 8 week period OCPs, PCBs 418
(Kathmandu, Pokhara, Birgunj, Biratnagar)
PUF-PAS 32 sites in 16 provinces in Turkey May 2014 to Apr. 2015, four 3 month periods OCPs, PCBs, PBDEs 419
PUF-PAS 62 sites in China 2016–2017, three periods of 7 to 8 weeks Unintentionally produced PCBs 420

Liu et al. used PE strips covered by a metal plate to sample where the air should to be reasonaly well-mixed, indicate a fairly
PAHs in air at numerous heights above the ground.76 SRs high uncertainty of the recorded concentrations.
derived from DCs were strongly dependent on height, ranging Wang et al. used the PUF-PAS to measure concentrations of
from 2 to 3 m3 per day at the lowest sampling heights to 20 m3 OCPs and PAHs above and below the canopy of a rice crop,
per day and higher for PASs deployed more than 50 cm above using DCs to allow for different SRs.463 Following expectations,
ground. Concentration gradients indicated deposition of lighter SRs were notably lower below the canopy than above it. Both
PAHs and volatilisation of heavier PAHs. Relatively large varia- lower and higher concentrations were observed below the
tions in the reported air concentrations at greater heights, canopy when compared to those measured above.

1962 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

Table 17 Studies relying on PAS network on a regional scale involving multiple countries

Sampler Location, no. of sites Time frame SVOCs Ref.

PUF-PAS 15 sites in the Laurentian Great Lakes Jul. 2002 and Jun. 2003, four 3 PCBs, PBDEs, OCPs 425
region (USA, Canada) month periods PCNs 426
Chiral pesticides 427
XAD-PAS 8 sites in Northeast Asia (South Korea, 2007 to 2008, 1 year PCBs, OCPs, dechlorane plus 423
Mongolia, China)
SPMD 10 sites from South of the UK to the Summer 1994 to summer 1996, 2 PCBs, a-HCH and g-HCH 40
north of Norway years
SPMD 12 sites from South of the UK to the Summer 1998 to summer 2000, 2 PCBs, HCB 42
north of Norway years
SPMD 11 sites from South of the UK to the Summer 2000 to summer 2002, 2 PCBs, PBDEs, OCPs 43
north of Norway years
SPMD 11 sites from South of the UK to the Summer 2002 to summer 2004, 2 PCBs, OCPs PBDEs, PAHs 44
north of Norway years
SPMD 11 sites from South of the UK to the Summer 2004 to summer 2008, PCBs, PBDEs 45
north of Norway two 2 year periods
SIP-PAS 10 sites from South of the UK to the Jul. 2008 to Jun. 2009, 1 year PCBs, PBDEs, HCB 150
north of Norway
SPMD 40 sites in 5 European countries Sep. to Dec. 1999 and Jun. to Aug. PAHs, nitro-PAHs 47
(Austria, Czech Republic, Poland, 2000, 21 days
Slovakia, Sweden)
PUF-PAS 23 sites in 8 Northern European Aug. to Oct. 2004, 56 to 82 days PCBs, PAHs, PBDEs, HCB, p,p0 - 428
countries (Ireland, UK, Denmark, DDT, p,p0 -DDE
Norway, Sweden, Russia, Finland,
Estonia)
PUF-PAS 11 sites around the Aegean Sea Jul. to Aug. 2012, one 1 month OCPs, PCBs, PBDEs, PAHs 103
(Greece, Turkey) period
PUF-PAS 5 sites Mexico City, Mexico, Mar. to Apr. 2006, 42 to 50 days OCPs, PCBs, PBDEs 429
Gothenburg, Sweden and Lancaster,
UK
PUF-PAS 15 sites in 4 West African Countries May to Jul. 2008, 90 days PCBs 421
(Ghana, Gambia, Sierra Leone, Ivory
Coast)
PUF-PAS 19 sites in 4 West African countries 2012, one 8 week period in rainy OCPs 422
(Togo, Benin, Nigeria, and Cameroon) season, two 8 week periods in
Cameroon
PUF-PAS 155 sites in 22 Central and Eastern 2006–2008; 6 periods of 28 day PAHs, OCPs, PCBs 430
Europe countries
PUF-PAS 45 sites in 3 East Asian countries Mar. to May, 2008, Aug. to Oct. PBDEs 424
(China, South Korea and Japan) 2008, 51 to 90 days
PUF-PAS 6 sites in 4 East Asian countries (The 2011–2012, two 2 month periods PCBs 431
Philippines, Japan, Malaysia, (46 to 98 days)
Vietnam)

The modied Bowen ratio method for ux estimation, which contaminated soil, because (i) a wind speed/SR gradient would
relies on vertical concentration gradients, can become prob- only counter and not create a decrease in the sequestered
lematic when the time resolution of the concentration amounts with PAS deployment height, i.e. a measured contami-
measurements is too coarse.464 In particular, if during the nation gradient is less likely to be a sampling artefact, and (ii)
sampling period, the air–soil exchange ux changes direction, SVOC evaporation is likely to occur continuously, i.e. it is less
the concentration gradient cannot be interpreted in terms of likely that the net ux direction changes during a deployment.
a ux.464 In the case of SVOCs, it is quite likely that air–soil None of the studies reporting SVOC concentration gradients
exchange uxes experience day and night differences. PAS above soil measured with PASs seems to be aware of the limita-
sampling periods are almost always longer than a day. If, tions of deriving uxes from long term average concentrations.
however, it can be assumed that the direction of net ux does G.3.2 Gradients on towers. At a somewhat larger scale, ve
not change during the sampling period, the performance of the studies have used towers to make vertical SVOC concentration
modied Bowen ratio method is not strongly affected by the measurements with PASs94,130,135,465,466 (Table 22). The issue of
length of sampler deployment.464 The condition to measure wind speed gradients possibly resulting in an SR gradient (as
meaningful vertical concentration differences with PASs is discussed in Section G.3.1) applies here as well. For example,
therefore most likely fullled when sampling above a highly Chaemfa et al. attributed lower sequestered amounts of PCBs at

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1963
Environmental Science: Processes & Impacts Critical Review

Table 18 Studies involving continental scale PAS networks for SVOCs

Sampler Location, no. of sites Time frame SVOCs Ref.

XAD-PAS 40 sites in North and Central May to Jul. 2000 to May to Jul. OCPs 29
America (Canada, USA, Mexico, 2001, 1 year a-HCH, g-HCH 432
Belize, Costa Rica) PCBs, PBDEs 433
PUF-PAS 71 sites in 22 European countries Jun. to Jul. 2002, 6 weeks PCBs, PBDEs, OCPs 434
PAHs, PCNs 435
POG 38 sites in 19 European countries Jun. 2002, 1 week PCNs, PCBs, PAHs 95
PUF-PAS 86 sites in 34 European countries Summer 2006, 3 months PCBs, OCPs, PAHs 436
PUF-PAS 77 sites in East Asia (China, South Sep. to Nov. 16, 2004, 56 days PCBs, OCPs, PBDEs 437
Korea, Singapore)
SIP-PAS 46 sites in East Asia (China, Mar. to Aug. 2009, 100 days PFAS 143
Taiwan, Japan, India)
PUF-PAS 106 sites in East Asia (Japan, Mar. to May 2008, 8 weeks PCBs, PCNs 438
China, Korea, Taiwan)
PUF-PAS 176 sites in Asia (China, India, Sep. 2012vAug. 2013, 4 seasonal PAHs, alkylated PAHs 439
Japan, Vietnam, South Korea) periods
PUF-PAS 26 sites in 15 African countries Jan. to Jun. 2008, 28 days to 3 PAHs, PCDD/Fs, PCBs, OCPs 440
months
PUF-PAS 20 sites in 12 African countries 2010–2018, 2 to 3 month periods CUPs 441
(166 samples)
PUF-PAS 7 sites in Latin America 2014, two to four 3 month periods OPFRs 442
PUF-PAS, SIP-PAS 9 sites in Latin America 2014/2015, up to eight 3 month HCBD, PCA, dicofol, OCPs, PCBs 31
periods, SIP for one period BFRs, OFRs in PUFs, cVMS, PFAS 443
in SIPs

the lowest sampling height to differences in wind speed with specic SRs to derive air concentrations largely eliminated
height.135 Moeckel et al. quantied the gradient in SR with the a gradient of increasing sequestered amounts of PCBs and
help of DCs and noted a four-fold increase of the SR of the PUF- OCPs with elevation in a Swedish forest.130 On a tower in Beijing,
PAS between ground level and 100 m (2 to 8 m3 per day).130 levels were still slightly higher closer to the ground, suggestive
Similarly, Li et al. noted an increase in SR from around 7 m3 per of urban ground sources.466 Both of these studies suggest that
day at 15 m to 11 to 12 m3 per day at 320 m.466 Using height- DC-corrected SRs are imperative when trying to measure vertical

Table 19 Studies involving global scale PAS networks for SVOCs

Sampler No. of sites Time frame SVOCs Ref.

PUF-PAS 7 sites 2002, 2 to 7 months OCPs 444

PUF-PAS 41 sites in 28 countries Dec. 2004 to Mar. 2005, 3 OCPs, PCBs, PBDEs 208
months (70 to 232 days) PCNs 445
PUF-PAS 41 sites in 36 countries Dec. 2004 to Dec. 2005, 4 OCPs, PCBs, PBDEs 127
consecutive 3 month periods
SIP-PAS and PUF-PAS 20 sites Mar. 2009 to Oct. 2009, 86 to 130 PFAS, PCBs 139
days VMS 147
OCPs, PBDEs 151
CUPs 152
XAD-PAS 36 to 46 sites 2005 to 2008, 4 consecutive 1 year OCPs, CUPs 448
periods
XAD-PAS 34 to 35 sites 2005 to 2006, two 1 year periods PAHs 449
PUF-PAS 31 sites in 11 countries 3 to 6 months PCBs, PCDD/Fs, DDT-compounds 450
XAD-PAS 17 to 46 sites 2006 to 2011, six consecutive 1 PFAS 197
year periods
PUF-PAS 40 sites 2005 to 2006, four consecutive BFRs 446
seasonal periods
PUF-PAS 48 sites 2014, several consecutive 3 PBDEs, HBCD, BFRs, chlorinated 447
month periods FRs, OPEs
SIP-PAS 21 sites Apr. to Jul. 2015, one 3 month PFAS, VMS 153
period, 2013,
four consecutive 3 month
periods

1964 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

Table 20 Studies deploying PASs along urban to rural transects

Sampler Location, no. of sites Time frame SVOCs Ref.

PUF-PAS SPMD 7 sites North of Toronto, Jul. to Oct. 2000, 120 days PCBs, OCPs 453
Canada
PUF-PAS 7 sites North of Toronto, Jun. 2000 to Jul. 2001, three 4 PAHs, PCBs, OCPs 454
Canada month periods PBDEs 456
PUF-PAS 7 sites North of Toronto, Apr. to May 2002, 1 month PCBs, PBDEs 455
Canada
PUF-PAS 8 sites North of Toronto, May to Oct. 2003, seven CUPs 457
Canada consecutive 1 month periods
PUF-PAS 10 sites across Birmingham, Aug. 2003 to Oct. 2004; PBDEs 459
UK eleven 1 month periods PCBs 460
PUF-PAS 19 sites North, East and West Oct. 2007 to Oct. 2008, four 3 PCBs, PBDEs, PAHs 458
of Toronto, Canada month periods

Table 21 Studies using PASs to measure air concentration gradients of SVOCs above soil

Sampler Location/time frame Sampling height SVOCs Ref.

SPMD 1 contaminated site in 20 and 150 cm HCHs, HCB, DDTs 462

landll, Bitterfeld, Germany
PUF disks in stainless steel 1 contaminated and 2 0.5, 2, 3, 4, 5, 8, 12, 15, 170, PAHs 169
can with open mesh-covered uncontaminated sites, 20, 25, 30, 40, 45 cm, 6 per
bottom (3 cm diameter Beijing, China, one 2 month height
opening) period
Regular PUF-PAS 1 rice paddy eld, Above (120 cm) and below PAHs, OCPs 463
Guangzhou, China, two 6 (40 cm) the rice crop canopy,
week periods 6 per height
PUF disks in ABS can with Sealed PVC chamber within 5–8, 15–18, 30–33 cm, no HCHs, DDTs 170
bottom that has holes greenhouse over soil spread replicates
on the ground, six 10 day
periods
PUF disk in stainless steel Pasture in Tibet, two 2 2, 3, 5, 8, 10, 15, 20, 50, 80, OCPs 171
housing with open mesh- month periods 120, 200 cm, 3 per height
covered bottom (4 cm pooled
diameter opening)
PE strip, 5 per metal box with 3 historically contaminated 4 replicates right above soil PAHs, PCBs 70
open bottom over soil, 5 per sites (Anniston, Alabama and 3 replicates at ca. 150 cm
box in upright metal Wyckoff, Washington, above it
chimney in air Mosie, Oregon) in USA, one
14 day period
PE strips placed beneath 1 urban green space, 5, 10, 15, 20, 25, 30, 35, 40, PAHs 76
a metal plate Shanghai, China, seven 2 to 45, 50, 60, 70, 80, 90, 100,
8 week periods 150, 200 cm

gradients with the PUF-PAS, whose SR becomes highly suscep- the XAD-PAS, but the PUF-PAS and SPMDs have also found use.
tible to wind at the wind speeds encountered at height.130,466 Because DCs cannot be used with the XAD-PAS, other ways to
G.3.3 Vertical gradients in mountains. At an even larger correct for variable SRs may be required. Sites where a PAS is
vertical scale, PASs have been used repeatedly to study air exposed to very high wind can oen be identied when all target
concentration gradients along mountain slopes (Table 23). chemicals have elevated sequestered amounts.194 The deploy-
Oen, those studies could also be classied as transects as ment time-normalised sequestered amount of a chemical that
dened in Section G.2 above. Studies have been conducted on can be assumed to have uniform air concentrations along the
mountains in Europe,467,468 Western Canada,452,469–471 South elevation gradient has been used to adjust SRs. The amounts of
American mountains,195,472–475 the Southern and Eastern slope of HCB sequestered by a XAD-PAS can oen serve this purpose, as
the Himalayas,65,194,476–481 as well as on the island of Hawaii.482 long as there are no major sources of HCB in the vicinity of the
The largest elevational range exceeds 5000 m on the Southern transect.194,195,482
slope of the Himalayas.476,479 A common theme among those studies is the investigation of
The issue of SR variability with elevation can also arise along mountain cold-trapping, especially when combined with
mountain transects, where the wind speed may vary between measurements of concentrations in soils (e.g. ref. 391, see also
sites at different elevation. The majority of these studies used Section G.8.2). Remarkably, air concentrations of endosulfan

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1965
Environmental Science: Processes & Impacts Critical Review

Table 22 Studies involving PASs measuring vertical gradients of SVOC concentrations on a tower

Sampler Location, sampling heights Time frame SVOCs Ref.

POG CN Tower, Toronto, Canada Oct. 2001, three 1 week periods PAHs, PCBs, OCPs 94
6 heights (30, 90, 150, 210, 270, 360
m)
PUF-PAS CN Tower, Toronto, Canada May to Sep. 2005, ve 1 month OCPs 465
5 heights (30, 90, 150, 210, 270 m) periods
PUF-PAS Meteorological tower Hazelrigg eld Apr. to Jun. 2006, one 54 day period PCBs, OCPs 135
station, Lancaster, UK
5 heights (3, 5, 10, 20, 30 m)
PUF-PAS, SPMD Norunda Common, Sweden May to Jul. 2007, one 40 day period PCBs, OCPs 130
8–9 heights (2.5, 9, 14, 19, 25, 44, 68,
73, 96 m)
PUF-PAS Meteorological tower, Bejing, China Dec. 2006 to Jan., May to Jul., Jul. to PCBs, OCPs 466
9 heights (15, 47, 80, 120, 160, 200, Aug. 2007, three 2 month periods
240, 280, 320 m)

are oen found to increase with elevation,65,195,470 consistent as offices and residences,489 or between the indoors and
with its high potential for mountain cold trapping.195,391 Some outdoors.146,161,490 The ultimate goal is oen the assessment of
studies compare forested sites with sites in clearings to explore personal inhalation exposure for large, epidemiological health
the inuence of vegetation.65,468,471,478 Others compare the levels studies.491,492 In this context, there is also interest in portable,
from sampling sites on different sides of a mountain.65,471 In personal PASs that measures concentrations of SVOCs in the
many cases, source proximity is found to govern the air breathing zone of an individual.166,167
concentration variability along such transects.471 For example, A wide variety of PASs have been used indoors, including
proximity to traffic arteries is the main variable controlling SPMDs,24,49,488,493 PE-based PASs,490,494 SIP-PASs,145,146,495 the
PAHs concentrations in Canadian mountain air452 and on original PUF-PAS24 or modication of it with different types of
Hawaii, the highest concentrations of many OCPs occur in the housing.156,163,164 Most recently, also a number of PASs based on
vicinity of major towns.482 PDMS have been introduced112,114 and used.496 While some
studies compared different PASs with each other,113 others
compared PAS-derived indoor air concentrations with concen-
G.4 Spatial variability in indoor air concentrations and trations in other media that may serve for exposure assessment,
inhalation exposure such as surface wipes, window lms or house dust.497
Air concentrations vary of course also in different indoor envi-
ronments, depending mostly on the presence and strength of
indoor sources, but also on characteristics of the indoor envi- G.5 Aspects related to using PASs for probing spatial
ronment, such as room ventilation rate or the availability of variability
sorbing compartments. Interest in air concentration variability G.5.1 Reporting amounts sorbed per time period vs. volu-
in indoor environments is driven by the desire to characterise metric concentrations. As shown in eqn (5), an SR is required to
human inhalation exposure to SVOCs, considering that most convert the amount of an SVOC quantied in the sorbent of
humans spend the majority of their time indoors. While for a PAS operating as a kinetic sampler into a volumetric air
most SVOCs, exposure routes other than inhalation are concentration. As discussed in Sections F.2 and F.3, SRs oen
considered dominant and health-relevant, inhalation exposure have considerable uncertainty. If the main objective of a study is
can be an important contributor to total exposure, e.g. if the recording of spatial patterns, as may be surmised for most
concentration levels are very high.485 Occupational environ- of the studies listed in Sections G.1 to G.4 above, the conversion
ments are particularly susceptible to having high indoor of sequestered amounts into volumetric air concentrations may
concentrations of SVOCs. not always be necessary. The key question then is whether the
PASs have been used in several studies seeking to measure spatial concentration differences that a study seeks to resolve
air concentrations of SVOCs indoors (Table 24). While access to are larger than the uncertainty in the SR. If that is the case, the
electricity is usually not difficult in indoor environments and uncertainty in the SR is less of a concern and it may be sufficient
the use of active air sampling techniques is oen feasible,486 to report the concentration patterns in units of mass seques-
PAS-based approaches tend to be less noisy and obtrusive and tered per unit of time (e.g. per day or per the nominal deploy-
allow for cost-effective sampling in a larger number of loca- ment period such as one month or one year). Especially, if
tions,487 e.g. in various locations within a room108 or within a generic SR is to be applied i.e. not one that is specic to
a building,488 or, more typically, in a number of different a particular deployment, it may in fact be more transparent to
buildings.145,161,489 A common theme is also the use of PASs to not report volumetric air concentrations: the time-normalised
compare levels in different types of indoor environments, such sequestered amount has less uncertainty than the calculated

1966 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

Table 23 Studies using PASs to record gradients in SVOC air concentrations along mountain slopes

PAS Location, no. of sites Time frame SVOCs Ref.

SPMD 3 sites (1600, 2240, 2600 m) in the Mar. 2002 to Jun. 2003, 85, 161 and 210 HCB, PCBs 467
Pyrenees, Catalonia, Spain days
PUF-PAS 12 or 13 sites at four elevations (700, Sep. to Oct. and Oct. to Nov. 2003, 52 and PCBs, HCB, p,p0 -DDE, 468
1010, 1420, 1790 m) on Mont Mars, Alps, 33 days p,p0 -DDT
Italy
XAD-PAS 8 sites (570–1951 m) on Mount Aug. 2003 to Aug. 2004, one 1 year period OCPs 469
Revelstoke, BC, Canada
6 sites (1109–2561 m) in Yoho National PAHs 452
Park, BC, Canada
8 sites (1402–2902 m) on Observation
Peak, Banff National Park, Alberta,
Canada
XAD-PAS 5 sites (800–2740 m) in Canadian Rocky Mar. to Aug. 2004, one 5 month period PFAS, HCHs, 470
and Purcell Mtns., British Columbia, endosulfan
Canada
PUF-PAS 4 sites (1820, 2600, 4650, 5200 m) on east Feb. 2005 to Jan. 2006, four 3 month OCPs, PCBs 472
side of the Andes, Bolivia periods
XAD-PAS 9 sites (1242–4485 m) on Balang Mtn., 2005 to 2008, ve 6 month periods OCPs, PCBs 194
Wolong Nature Reserve, Sichuan, China
PUF-PAS 4 sites (400–2200 m) in Serra dos Orgaos Jun. to Aug. 2007 and Dec. 2007 to Mar. PCBs, PBDEs 473
National Park, Brazil 2008, two 90 day periods
4 sites (600–1800 m) in Sao Joaquim OCPs 474
National Park, Brazil
XAD-PAS 6 sites (135–5100 m) on the Southern May to Nov. 2012 OCPs, PCBs 476
slope of the Himalayas in Nepal PAHs 477
PUF-PAS 3 sites (3800, 4200, 4400 m) on S-slope, 2 Jan. 2008 to Jan. 2012, sixteen 3 month OCPs, PCBs 478
sites (4200, 4400 m) on N-slope of Sygera periods
Mtn., Tibet, China
XAD-PAS 18 sites (1983–4553 m) on Shergyla Mtn., Jul. 2010 to May 2011, two periods of 4 OCPs, PCBs, PBDEs, 65
Tibet, China and 7 months HBCD
XAD-PAS 6 sites (48–4405 m) in Northern Chile Feb. 2006 to Jul. 2007, one 1 year period OCPs, CUPs, 195
7 sites (10–1874 m) in Central Chile PAHs, PCBs
5 sites (50–760 m) in Southern Chile
XAD-PAS 3 sites (500, 1310, 2052 m) in Swiss Alps, Jul. 2006 to Oct. 2006, one 4 month OCPs, PCBs 483
Switzerland period
XAD-PAS 9 sites (0 to 3400 m) on Hawaii Big Island May to Sep. 2011 OCPs, PAHs, PBDEs 193 and
482
XAD-PAS 18 or 20 sites (1983–4553 m) on Shergyla Jul. 2012 to Jul. 2015, one 1 year period SCCPs, MCCPs 480 and
Mtn., Tibet, China 481
LDPE-PAS 4 sites (700 to 2400 m) on Serra do Mar May to Jun. 2012, one 30 to 40 day period PAHs 484
and 4 sites on (990 to 1700 m) on Serra
Geral, Brazil
XAD-PAS 4 sites (60–600 m) on Mt. Anderson, BC, Aug. 2007 to Aug. 2008, 1 year period PAHs, PCBs 471
Canada
4 sites (350–950 m) on Grouse Mtn.
4 sites (90–700 m) on Four Brothers Mtn.
3 sites (400–800 m) on Blue Grouse Mtn.
1 site on Chilkoot Trail (435 m)
XAD-PAS 8 to 9 sites (100–5200 m) on the S-slope of May 2012 to Nov. 2014, ve half year OCPs, PCBs 479
the Himalayas in Nepal periods

concentration, because the latter inherits the uncertainty of the compare the exposure of different birds using time-normalised
SR. Several early PAS studies had reported spatial results in sequestered amounts.518
amount per sampler or amount per sampler per The issue of SR uncertainty becomes paramount when small
time,29,280,390,391,432,433,448,469 but only recently has there been a re- concentration differences are to be discerned. This issue is
emergence of support for not necessarily reporting volumetric further compounded if the concentration differences occur
air concentrations.514–517 This approach is clearly advisable along a spatial gradient where factors with an inuence on the
when using a PAS whose SR under a given set of circumstances SR (most notably wind speed, but possibly also temperature)
is not well established or even unknown. A good example is the may be expected to vary systematically. Examples are:
study attaching a PAS to a gull, where it was entirely possible to

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1967
Environmental Science: Processes & Impacts Critical Review

– Comparing indoor and outdoor deployments:490,492 indoor measured across the country, suggesting that these sites are
environments tend to experience much lower air turbulence representative of the more remote parts of Botswana and Costa
and therefore lower SRs, which could lead to the underestima- Rica.
tion of indoor concentrations when compared to outdoor One potential application of a PAS network is in fact to assist
deployment. in the selection of a representative sampling site or in the
– Elevational transects of any scale (see Section G.3) can be determination of the number of sites that is required to
expected to experience gradients of wind speed, with wind appropriately represent a particular area. For example, the
speeds increasing with height. This is particularly problematic purpose of a PAS network across Chicago was to establish how
when trying to observe air concentration gradients that increase representative an active air sampling site is of the city as
with elevation. a whole.308 A network of PASs across the Czech Republic was
– Comparing deployments under a forest canopy to deploy- used to address the issue of representativeness more compre-
ments in unsheltered locations.65 The SR for the PUF-PAS is hensively by not only comparing absolute concentration levels
predicted to be much lower in forested regions and much of different SVOCs but also time trends obtained from those
higher over the ocean.126 sites.516 Cluster analysis based on levels and trends revealed that
When concentration differences between sites with different the fourteen sites belonged to one of three types, whereby sites
wind exposure are small, it is necessary to derive deployment- belonging to a cluster share characteristics related to remote-
specic SRs, e.g. through the use of DCs, or choose a sampler ness, landscape, population, and pollution sources and there-
design with an SR that is not, or at least less, dependent on wind fore also their pollution prole. An analysis like this could assist
speed. For example, the application of DC-derived, elevation- in optimizing the elimination of sampling sites from a network,
specic SRs changed the concentration gradient with height by nding the number and location of sites that yield the most
along a tower.130 information with the least number of sites.516
It is important to realise that during the interpretation of G.5.3 Mapping and interpolation. Related to this issue of
spatial PAS data, one is not limited to comparing concentrations site representativeness is the question of whether it is justied
or time-normalised sequestered amounts, but it is possible to to interpolate the levels measured by PASs at different sampling
investigate compositional changes. If wind speed inuences SRs, sites, e.g. by geo-statistical methods, such as kriging. This is
it does so for all target chemicals and not just for one. Compo- likely only appropriate if it can be assured that there are no
sitional patterns thus are not, or at least less, affected by spatial major sources of an SVOC between any pair of sampling sites,
variability in wind. Cluster analysis is particularly useful when e.g. in a small-scale network with one or a few clearly dominant
interpreting compositional changes across a PAS network.477,519,520 emission sources (e.g. for some of the studies summarised in
Of course, there are oen objectives of PAS studies that go Table 13). A few studies have attempted to create urban
beyond deciphering of spatial patterns and these may require concentration maps through interpolation (Fig. 10A).342,360 The
the calculation of volumetric air concentrations. Examples are: results are not always very convincing,301,340 mostly because
(i) data obtained from different sampling methodologies are to relatively large concentration differences can occur over small
be compared,521 (ii) PAS data are to be combined with data from distances and the number of sampling sites is generally insuf-
adjoining environmental media such as water or soil in order to cient. When seeking to map urban air pollution, a minimum
derive information on air–surface exchange, which is discussed number of sampling sites of 0.29 multiplied with the urban
in detail in Section G.8 below, or (iii) PAS-derived data on surface to be monitored in km2 has been recommended, based
inhalation exposure are to be used in risk assessment, which is on work with VOCs and NO2.522 None of the urban PAS networks
discussed in Sections G.4 and G.5.5. for SVOCs (Table 14) comes close to fullling this criterion. A
G.5.2 Sampling site representativeness and scale. The more defensible way of displaying the spatial variability of PAS
number of sites in PAS networks varies widely from less than networks is to use a map that visualises the variability in the
a handful to more than 100. No matter how large the number of measured levels, but does not seek to interpolate values in areas
sites within a network may be, the issue of representativeness of where no sampling was done (see Fig. 10B for a good example).
sites is important. This is particularly relevant for large scale A more sophisticated strategy for urban pollution mapping is
networks, where a site is oen meant to represent a large area. land use regression analysis, where air pollution monitoring
In some of the continental or global networks listed in Sections data are combined with stochastic models that rely on predictor
G.1.5 and G.1.6, an entire country may oen be represented by variables that are typically obtained through geographic infor-
a single sampling site. Depending on the nature of that site, mation systems.523 While quite common for classical air
completely different results could be obtained. For example, pollutants (such as VOCs, PM2.5 and NO2), land use regression
within GAPS, the French site oen shows the highest concen- has, to the best of our knowledge, only been applied once to PAS
trations, merely by virtue of being located within Paris.127,447 data for SVOCs, namely to identify potential sources of PCBs,
Gawor et al.197 analyzed PASs from a global and two national PBDEs, PAHs, and polycyclic musk compounds in Toronto.524 A
networks for PFAS and used it to assess to what extent one site large fraction of the observed variability in PCB air concentra-
from the global GAPS network in each of Botswana and Costa tions could be related to the distribution of PCBs in use/storage/
Rica represents the national average as obtained from national building sealants, whereas that of the musk compounds was
networks in those two countries. In both cases, the levels of strongly related to population density.524
PFAS measured at the GAPS site were lower than the levels

1968 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

Table 24 Studies using PASs to study SVOC air concentrations indoors

PAS Location, no. of sites Time frame SVOCs Ref.

SPMDs 10 homes and outdoor sites in One 11 day period PAHs 493
Makwanpur region, Nepal
PUF-PAS w/o double 74 homes and 7 outdoor locations One 3 week period (outdoor PBDEs 161
bowl in Ottawa, Ontario, Canada 3 months)
59 homes for FOSE/FOSA MeFOSE, EtFOSE, EtFOSA, 162
MeFOSEA
52 homes for FTOHs 10:2 and 8:2 FTOH 136
SPMDs 15 homes in Hagfors, Sweden One 2 week period PAHs, PCBs 49
PUF-PAS 31 homes, 33 offices, 25 cars, 3 One 28 day period PCBs, PBDEs 489
public indoor spaces in
Birmingham, UK
PUF-PAS w/o DB 35 homes and 11 outdoor One 6 to 7 week period PAHs, OCPs, PCBs, PBDEs 429
locations in Mexico City, Mexico,
Gothenburg, Sweden and
Lancaster, UK
SPMDs 52 homes situated along the One 30 day period PAHs, PCBs, OCPs, some 488
border between Arizona and non-target
Mexico (four locations in each
home, pooled)
SIP-PAS 59 homes and 6 outdoor locations One 4 week period (outdoor Neutral and ionic PFAS 145
in Vancouver, Canada 3 months)
SIP-PAS 6 homes, 2 public indoor spaces One 20 day period (indoor) Neutral PFAS 239
and 6 outdoor sites in 3 major one 90 day period (outdoor)
cities
PUF-PAS 17 indoor and 18 outdoor sites in
10 cities in South Korea
SIP-PAS 20 homes, 12 offices, and 10 One 1 month period, 12 Neutral and ionic PFAS 146
outdoor locations in Birmingham, monthly periods at 8 sites
UK
PUF-PAS 20 homes, laboratories and offices One 27 to 38 day period PCBs, PBDEs 498
in Toronto, Canada Synthetic musk 288
Triolein coated vial 2 car interiors, 4 occupational One 37 day period PAHs 499
indoor, 12 outdoor sites in
Malaysia
PUF-PAS (single bowl in 63 homes in Bloomington, One 28 day period PBDEs, BFRs 500
North America) Indiana, United States, Toronto, 13 OPFRs 501
Canada, Brno, Czech Republic, 1 OCPs, PCBs 502
or 2 rooms per home, 23 homes in
Toronto, Canada, 20 homes in
Brno, Czech Republic
PUF-PAS Inside and outside of 6 schools in Multiple periods of 48 days PCBs, hydroxylated PCBs 492
East Chicago, Indiana and average (22 to 114 days)
Columbus Junction, Iowa, USA
PUF-PAS Indoors and outdoors at schools Quarterly (homes), bi- PCBs 491
and homes (n ¼ 293), Greater quarterly (schools), 45 days
Chicago Metropolitan Area, USA
PUF-PAS 20 homes, offices, 10 outdoor in Two 56 day period PBDEs 503
suburban and urban in Shanghai,
China
PUF-PAS 15 indoor samples (university, One 61 day period PAHs, phthalate 504
hospital, school) and 2 outdoor
samples (playing area at
university) in Ouargla city, Algeria
PUF-PAS 34 homes in Pokhara, Birgunj, One 2 month period PBDEs, HFRs 505
Biratnagar and Kathmandu, Nepal
PUF-PAS 20 homes (20 living rooms and 13 One 90 day period PBDEs 506
bedrooms) and 20 offices in
Hangzhou, China
PUF-PAS Three offices (low-use, medium- Three consecutive 28 day BFRs 507
PUF/GFF-PAS use and high-use) in Beijing, periods
China

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1969
Environmental Science: Processes & Impacts Critical Review

Table 24 (Contd. )

PAS Location, no. of sites Time frame SVOCs Ref.

SIP-PAS 57 homes in Kuopio, Finland One 21 day period PFAS 508

PUF-PAS 16 sites in offices and residential Two 60 day periods SCCPs, MCCPs 509
buildings in Beijing, China
PUF-PAS 16 residences in Iowa City, Iowa One 6 week period PCBs 165
PUF-PAS and PDMS 51 homes in Ottawa and Toronto, One 3 week period Phthalates, PBDEs, novel 112
Canada BFRs, OPFRs
LDPE-PAS 33 apartments, 33 car interiors, 60 One 2 month period OPFRs 494
offices, 30 outdoor locations in
Alexandria, Egypt
SIP-PAS 3 buildings in Uppsala, Sweden One 140 day period FTOHs, cVMS, BFRs, 495
OPFRs
LDPE-PAS 6 urban sites with matched indoor Seven 24 hour samples for 63 PAHs 490
and outdoor samples in Eugene, 1 week
Oregon
PUF-PAS modied 14 residential houses and 14 One 2 to 3 month period PAHs, PCBs, OCPs, CUPs 163
offices in Brisbane and Canberra,
Australia
PUF-PAS 25 homes in Harbin, China One 30 day period PAHs, phthalates, PBDEs, 510
OPEs
PUF-PAS 8 residential buildings and 3 Two 44 to 45 day periods PCBs 511
outdoor locations in Bursa, Turkey
PDMS 63–71 apartments in Toronto, Two 1 week periods PBDEs, PAHs, BFRs, 496
Canada phthalates
PUF-PAS (PUF-cyl) Two Swedish vessels (Icebreaker Two 6 to 7 day periods PAHs, alkylated PAHs 512
and cruise ship)
PUF-PAS 11 biological, 15 chemical and 14 One 21 day period Phthalates 513
physical laboratories, 3 offices

While not quite a land use regression, several PAS studies distances of OCPs calculated with models designed to assess
have sought to regress PAS-derived concentration data for PAHs a chemical's long-range transport potential.29 At other times,
with spatially resolved data on population density. PAHs models are used to describe the observed spatial variability. For
concentrations measured at different sites of the global GAPS example, a radial dilution model was applied to t the SVOC
network have been regressed against four different means to concentration variability across Toronto and to probe the rela-
quantitatively express the proximity of those sites to people.449 tive importance of sources within and outside of the city's
The remoteness index, which is derived from a global atmo- central business district.458
spheric transport model, explained as much as 60% of the G.5.5 PAS networks and human exposure and toxicity. One
variability in the concentration of uoranthene. Despite the of the motivations for interpolating PAS measurements of air
simplicity of this variable, the number of people living within 20 pollution levels is their potential use for human exposure
km of a sampling site explained even more of the variability in assessment and therefore use in epidemiological research.
the concentrations of the sum of PAHs measured around the However, not for all air pollutants can a stationary network of
lower Great Lakes.82 Concentrations of SCCPs measured across PASs allow for a reliable prediction of personal exposure
Australia also correlated well with a number of variables, levels.525 For many SVOCs, air concentrations generally are less
including population density, number of roads and site classi- pertinent for human exposure and health effects, as inhalation
cation (i.e. remote, rural and urban).277 is rarely a dominant pathway for exposure. The PAHs are an
G.5.4 PAS networks and spatial modelling. There have exception. Accordingly, numerous PAS studies have sought to
been several studies that used the spatial information from PAS estimate inhalation exposure and the health risk associated
networks in modelling studies. Sometimes this involves the with exposure to airborne PAHs.336,348,361–363,439,526 This is typically
comparison of the measured spatial distribution with one ob- done by summing the products of the concentration of carci-
tained from models. Fine-scale emission estimates and nogenic PAHs with their relative potency factors in order to
a dynamic atmospheric model have been used to simulate the obtain an equivalent concentration of benzo[a]pyrene, which in
distribution of acenaphthene in the atmosphere of Nanjing, turn can be converted into a chronic daily inhalation intake
China,338 and of DDTs, HCB, PCBs and PBDEs in Mendoza, rate. Sometimes, a value for the cancer risk due to inhalation is
Argentina.332 Empirical travel distances could be estimated derived as well.
from the air concentration variability across a latitudinal tran- Even though inhalation is generally not believed to represent
sect in North America to evaluate the characteristics travel a large fraction of the total intake of compounds with dioxin-

1970 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

like toxicity, some studies similarly calculate a toxic equivalence PASs have also been used for the assessment of exposure of
concentration (TEQ) using the product of the concentrations of birds to SVOCs. Stationary PUF-PAS deployed in Hamilton
the toxic PCDD/Fs, PCBs, and PCNs congeners and their toxic harbour were used to assess the exposure of double-crested
equivalent factors (TEFs), which in turn is used to calculate the cormorant colonies to PAHs,533 whereas a miniaturised
TEQ intake by inhalation.309,312,345,382,416,527 In other studies, the sampler strapped to the back of gulls was used to observe that
inhalation exposure and risk are estimated for individual PASs attached to birds foraging in or around landlls accu-
substances285,299 rather than for equivalent concentrations. mulated higher levels of PBDEs, but not of three other BFRs.518
Some studies performed probabilistic exposure-risk model- While the applications described in Sections G.1 to G.5 in
ling.363,382 Again, oen a cancer risk is estimated for dioxin-like some way take advantage of the ability of PASs to efficiently
substances. In other studies, the estimated inhalation exposure provide spatial information on SVOCs in the atmosphere, not
is compared with reference doses to estimate risk.304 all PAS studies are motivated by the need for characterisation of
In the above studies, the health risk assessment is based on spatial variability. Some of the other application areas of PASs
the SVOC air concentrations measured with the PAS. Another are discussed in Sections G.6 to G.8.
approach to probing the spatial variability in inhalation toxicity
is to perform in vitro bioassays using the extracts and dialysates G.6 Extreme remote region monitoring (polar regions, Tibet,
obtained from PASs (Table 25). In the nine studies we found in mountains)
the literature, either SPMDs, PUF-PASs or PE-PASs were
Because they do not require power, PASs have found use in the
deployed mostly in areas with elevated SVOC levels, namely
monitoring of SVOCs in very remote or otherwise difficult to
urban location271,322,413,528–530 or sites impacted by industrial
access locations. Examples are high mountain areas, and polar
activity.327,531,532 The bioassays explored genotox-
research stations and eld camps. One challenge for sampling
icity,322,327,413,528,529,531 dioxin-like toxicity modulated via the aryl
in such locales is oen that access is very difficult and expensive
hydrocarbon receptor (AhR),271,413,530,532 estrogenicity,530,532 and
and it may be only possible to visit sampling sites once a year.
cytotoxicity.327
Long term deployments require a PAS with long tlinear, i.e. a high
The PAS extracts were in most cases also subjected to
uptake capacity and the capability to work at extremely low
chemical analysis, most frequently for PAHs. In some cases,
temperatures. DCs oen do not work well at low temperatures,
more extensive chemical characterisation was undertaken, e.g.
because of insufficient loss rates,126 yet these remote sites oen
in addition to PAHs, extracts have been analysed for PCBs and
experience very high winds and are therefore in particular need
OCPs,322 alkylated PAHs and benzothiophenes,327 and alkylated
of deployment-specic SRs. Its high uptake capacity explains
PAHs and OPE.271 By subjecting the extracts to H2SO4 treatment
why the XAD-PAS is frequently used in such situations.520,534–540
prior to the bioassay, it is possible to investigate whether
It is probably not by coincidence that PE, PUF and SIP-PASs are
recalcitrant or degradable compounds in the extracts are
only used during the short and milder summer seasons.90,541,542
responsible for the detected biological activity.413,530 Eventually,
Another challenge for passive sampling in remote areas are
the biological activity of samples deployed in different locations
the oen very low SVOC concentrations, which, when combined
or during different seasons413 is compared. Studies typically also
with relatively low SRs, can result in sequestered amounts below
investigate whether biological activity and SVOC concentrations
detection limits. The FTS can achieve sampling volumes in the
in samples correlate.
1000s of m3 and is therefore used when regular site access is

Fig. 10 Two examples of how spatial variability in air concentrations as determined by PAS networks has been displayed, either using inter-
polation360 (left) or not363 (right). Reproduced from ref. 360 and 363 with permission from Elsevier, Copyright 2017 and 2020.

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1971
Environmental Science: Processes & Impacts Critical Review

Table 25 Studies that use assays to probe the biological activity of extracts obtained through passive air sampling

Sampler Location, no. of sites Bioassays Ref.

SPMDs 5 sites in Caserta, Italy, One 3 week period Microtox (inhibition of bioluminescence 528
of V. scheri)
Mutatox (genotoic reversion of dark
mutants of V. scheri)
PUF-PAS 20 sites across Brno, Czech Republic, one 28 day period Bacterial genotoxicity test using E. coli 322
in Oct./Nov. 2004 sulA::lacZ
SPMDs 4 sites in Vilnius, Lithuania, four 8 week periods in Chromosome aberrations and sister 529
different seasons in 2002–2003 chromatid exchanges (SCEs) in human
blood lymphocytes in vitro
SMART: somatic mutations and
recombination in Drosophila
melanogaster wing cells in vivo
SPMDs 6 sites in an occupational setting of a coke plant, incl. Comet assay (in vitro DNA-damaging 531
an office as a control, one 24 day period in winter activity in human cells)
Mutagenicity (Ames test TA98 strain, w/o
S9)
PUF-PAS 6 offices and inside and outside of a suburban home in Estrogenic (E-SCREEN-MCF7- BOS) 530
Brisbane, Queensland, Australia, one 40 to 50 day AhR (CAFLUX-H4G1.1c2)
period in Apr. to Aug. 2007
PUF-PAS 4 urban, 3 rural, and 1 background sites across Indirect acting genotoxicity (umuC assay) 413
Australia, two 40 to 50 days periods in Jan. to Feb. 2007 AhR
and Jul. to Aug. 2007
PUF-PAS 1 background and 1 urban/industrial site in each of AhR (anti-)estrogenicity, (anti-) 532
Lithuania, Slovakia, Romania, and Serbia, one 5 month androgenicity
period in Mar. to Aug. 2006
PUF-PAS 15 sites across the Athabasca Oil sands region, one 60 Salmonella mutation 327
day period in Apr. to May 2014 Mammalian lactate dehydrogenase
cytotoxicity
PE-PAS 9 sites in the Cleveland, Ohio area, one 60 day period in AhR 271
Jul. to Sep. 2013

possible (e.g. in research stations that are operating year- developing PASs for SVOCs therefore was the prospect to
round), but electricity supply is difficult. Even if a research determine long-term, interannual temporal trends in air
station has electric power, it is oen desirable to sample at a site concentrations much cheaper and with less effort. One of the
sufficiently removed from station buildings and human activi- advantages of PASs in this regard is that they can provide time-
ties (and therefore electricity) to avoid the possibility of averaged concentrations over long time periods. The challenge
contamination. The FTS has been used on the Tibetan plateau of deriving long term trends from active sampling campaigns is
and in Arctic, sub-Arctic and Antarctic research stations.176–181 that the latter usually sample only during a small fraction of
Table 26 summarises studies that have used PASs in time and therefore the time periods of active sampling need to
extremely remote environments. A vertical mountain transect is be representative, which requires a relatively large number of
a particular type of study in remote regions. These studies are samples.448
discussed in Section G.3.3 above and summarised in Table 23. On the other hand, atmospheric concentration changes over
Some of the studies in extremely remote regions have been time may be quite subtle and, considering the uncertainty of
going on for several years and allow for the observation of time SRs for many PASs, it may be questioned whether the precision
trends,181,539 which is further discussed in Section G.7 below. of PASs is sufficient to detect such changes. Shunthirasingham
et al. have argued that the uncertainty of the SR is of diminished
concern when the main interest is in the trend and not in the
G.7 Time trend monitoring
absolute concentration levels and if the PAS sampling sites are
Long term time trends in the air concentrations of SVOCs are an not changed.448 The reason is that the factors that result in SR
important method to assess the effectiveness of measures to variability (most notably wind speed) should be reasonably
reduce emissions of those compounds. Traditionally, this is constant from year to year at the same site. Therefore, the
being done using active air sampling campaigns. However, such uncertainty of the SR is not propagated to the uncertainty of the
campaigns require a long-term commitment of fairly signicant temporal trend. In fact, time trends can be derived from time-
resources in terms of personnel and nances. With the Stock- normalised sequestered amounts, i.e., no SRs are required at
holm Convention, there is a greatly increased need for such all.448,515,516
time trend monitoring in the context of the Convention's
effectiveness evaluation. An important motivation for

1972 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

Table 26 Studies using PASs in extremely remote environments

Sampler Location, no. of sites Time frame SVOCs Ref.

SPMD 3 sites (Weissuhjoch, May to Nov. 2005, Dec. 2005 OCPs 60

Switzerland, Zugspitze, to May 2006, May 2005 to
Germany, Sonnblick, May 2006, Three periods
Austria)
XAD-PAS 3 sites at each of Ny-Ålesund Aug. 2005 to Aug. 2006 (Ny- PCBs, OCPs 534
and King George Island, Ålesund), Dec. 2004 to Dec.
Antarctica 2005 (King George)
XAD-PAS 3 sites at each of King George Dec. 2004 to Dec. 2007 (Ny- PCBs, OCPs 535
Island, Ny-Ålesund, and Ålesund), Aug. 2005 to Jun.
Chuuk, South Pacic 2007 (King George), Feb.
2006 to Oct. 2009 (Chuuk),
three 1 year periods
XAD-PAS 1 site at Cape Vera, Devon 2006 to 2008, two 1 year HCB, p,p0 -DDE, a-HCH, 537
Island, Nunavut, Canada periods heptachlor epoxide, PCBs
XAD-PAS 16 sites on the Tibetan Jul. 2007 to Jun. 2008, 1 year PCBs, OCPs, PBDEs 536
Plateau, China period
13 sites PAHs 543
16 sites 2007 to 2012, ve 1 year PCBs, OCPs 520
periods
SIP-PAS 16 sites on the Tibetan May to Jul. 2011, May to Jul. VMS, PFAS 544
Plateau, China 2013, two 3 month periods
FTS 1 site at Nam Co Lake, Tibet, Oct. 2006 to Feb. 2008, PCBs, PAHs, OCP 180
China een 1 month periods PBDEs, TBPH, EHTeBB, 176
BTBPE
FTS 1 site at Alert, Nunavut, Oct. 2007 to Dec. 2008, OCPs, PBDEs, TBPH, 177
Canada eleven consecutive one- EHTBB, BTBPE
month periods
FTS 1 site at Little Fox Lake, Aug 2011 to Dec. 2014, 42 PBDEs, non-BDE FRs, OCPs 179
Yukon, Canada consecutive one-month
periods
FTS 2 sites at Toolok eld Jul. to Aug. 2013, three PBDEs 178
station, Alaska, USA consecutive 18 day periods
FTS 1 site at Casey station, Dec. 2009 to Nov. 2014, PCBs, OCPs 181
Antarctica thirty-three consecutive
samples (4 to 6 weeks, some
3 months)
PUF-PAS 5 sites on King George Dec. 2009 to Feb. 2010, 2 PCBs, PBDEs 541
Island, Antarctica months
XAD-PAS 4 sites on King George 2009–2010, 1 year PCBs 538
Island, Antarctica
PUF-PAS 2 sites in National Parks Sep. 2013 to Oct. 2015, eight OCPs, CUPs 475
Itatiaia (2470 m) and Serra 3 month periods
dos Orgaos (2200 m), Brazil
XAD-PAS 7 sites on King George Dec. 2010 to Jan. 2018, 7 PCBs, OCPs, PBDEs 539
Island, Antarctica consecutive 1 year periods
LDPE 1 site at Cape Bounty, Aug. 2015, 21 days, Jun. to PCBs, HCB, a-HCH, p,p0 - 90
Melville Island, Nunavut, Aug. 2016, 59 days DDE
Canada
SIP-PAS 6 sites, Terra Nova Bay, East Dec. 2010 to Jan. 2011, One 1 OCPs, PCBs 542
Antarctica month period
XAD-PAS 5 sites, Ny-Ålesund and Aug. 2014 to Aug. 2015, 1 OPEs 549
London Island, Svalbard year

Kalina et al. investigated to what extent PAS-derived long- uncertain because of low levels close to the detection limit. In
term trends agree with those derived from active air sampling particular, the study conrmed that time-normalised seques-
campaigns.515,517 The rst comparison involved 534 active air tered amounts can be used for trend determination from PASs
samples and 132 PUF-PASs taken over a period of more than 10 and suggested that it can be advantageous to use annually
years in Košetice in the Czech Republic.515 The trends obtained averaged PAS values. Kalina et al. expanded on this work by
from the two sampling methods generally agreed well, except comparing time trends from active and passive air sampling at
when break-through of lighter SVOCs in the active sampling ve additional sites in Northern Europe.517 While again good
occurred or when levels of high molecular weight PAHs were too agreement in the trends from both sampling method was noted,

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1973
Environmental Science: Processes & Impacts Critical Review

the comparison was limited somewhat by the shorter time declined signicantly, but those of p,p0 -DDE did not show
periods of side-by-side sampling. trends and HCB and the PAH chrysene even showed increasing
By now several multi-year trends obtained by PASs have been levels. Having time trends from multiple sites within a relatively
reported. Table 27 lists the studies that have derived temporal small area allowed for the investigation of the spatial variability
trends from at least four consecutive years of passive sampling. in those trends: three types of sampling sites could be identied
The longest such trend so far could be obtained from the which differ from each other in terms of the absolute levels and
transect of SPMD deployments at 11 sites in the UK and Norway, time trends. For example, four remote mountain sites were
which comprises 12 years of sampling between 1994 and characterised by low levels of all SVOC types and decreasing
2008.150 This study clearly established declining trends of both levels of PAHs.516
PCBs and PBDEs in Northwestern Europe. Interannual time PASs have particularly been useful for time trend determi-
trends could also be derived from the XAD-PAS deployed within nation in remote regions, where active sampling campaigns are
the global GAPS project for pesticides448 and neutral per- oen not feasible. Examples are Antarctica,181,539 Tibet520 or
uoroalkyl substances.197 Somewhat confusingly, there are two Mongolia.549 In Antarctic air, concentrations of PCBs, HCHs,
national networks in Spain deriving temporal trends of SVOCs DDTs and endosulfans declined over a seven-year period,
in the atmosphere. These networks, which have been sampling whereas those of HCB did not.539 At a site in Mongolia, levels of
over the same time frame, using the same PUF-PAS, the same HCB even increased, those of HCHs and DDTs declined,
seasonal resolution, the same target compounds and even at whereas there was no signicant trend for the PCBs.549 Over
a similar number of urban and rural locations, reported a ve-year period, concentrations of HCB and HCHs declined
preliminary results from ve years of sampling in 2016.545,546 on the Tibetan plateau, but no such trend was apparent for
More recently, one of the networks has reported on seven years DDT-related substances.520
of continuous monitoring for PCBs, PBDEs, PCDD/Fs and In some cases, while there may not be an extended period of
selected OCPs at 12 locations across the country.548 Among the continuous sampling, it can be possible to infer time trends if
ndings are divergent trends for different PBDE congeners sampling sites are revisited aer a hiatus of several years. For
belonging to different technical mixtures,547 as well as signi- example, levels of pesticides measured with PASs in Brazilian
cant declines in the levels of dioxin-like PCBs and PCDD/Fs.548 mountains in 2014/15 could be compared475 with those
Time trends of OCPs, PAHs, and PCBs have been derived measured with the same technique in 2008.474 Similarly, time
from PUF-PASs deployed for multiple years at 14 sampling sites trends of cVMS and PFAS could be derived443 by comparing
across the Czech Republic.516 Levels for most PCBs and HCHs

Table 27 Studies deriving interannual time trends in atmospheric SVOC contamination from PASs

Sampler Location, no. of sites Time frame Compounds Ref.

SPMDs (n ¼ 66) 11 sites along UK-Norway 1994–2008, 14 years, two- PCBs, PBDEs 45
transect year resolution, gap: 1996–
1998
XAD-PAS (n ¼ 60) 15 sites distributed globally 2005–2008, 4 years, annual OCPs, CUPs 448
resolution
XAD-PAS (n ¼ 78) 13 sites distributed globally 2006–2011, 6 years, annual PFAS 197
resolution
XAD-PAS (n ¼ 80) 16 background sites in Tibet, 2007–2012, 5 years, annual PCBs, OCPs 520
China resolution
PUF-PAS (n ¼ 17) 1 urban site in São Paulo, 2010–2015, 5 years, seasonal PCBs, PCDD/Fs, OCPs 550
Brazil resolution
PUF-PAS (n ¼ 211) 4 urban and 7 background 2008–2013, 5 years, seasonal DDTs, HCHs, HCB, PCBs, 545
sites across Spain resolution PBDEs, PCDD/Fs
PUF-PAS 5 urban and 7 background 2008–2013, 5 years, seasonal DDTs, HCHs, HCB, PCBs, 546
sites across Spain resolution PBDEs, PCDD/Fs
PUF-PAS (n ¼ 132) 1 site in the Czech Republic 2003–2013, 10 years, OCPs, PCBs, PAHs 515
monthly resolution (28 days)
PUF-PAS 14 sites in the Czech At most: 2004–2014 OCPs, PCBs, PAHs 516
Republic At least: 2007–2010, monthly
resolution (28 days)
PUF-PAS 5 sites in Northern Europe 2012–2016, min. of 5 years, PCBs, OCPs, PBDEs 517
seasonal resolution
FTS (n ¼ 33) 1 site in Antarctica Dec. 2009 to Nov. 2014 PCBs, OCPs 181
PUF-PAS (n ¼ 321) 5 urban and 7 background 2008–2015, 7 years, seasonal PBDEs 547
sites across Spain resolution dlPCBs, PCDD/Fs 548
XAD-PAS (n ¼ 44) 6 to 7 sites on the Fildes 2007–2017, 7 years, annual PCBs, PBDEs, HCB, HCHs 539
Peninsula, Antarctica resolution DDTs, endosulfans
PUF-PAS (n ¼ 35) 1 site at Lake Hovsgol, 2008–2015, 8 years, three to PCBs, OCPs 549
Mongolia ve 2 month periods per year

1974 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

levels in PASs from 21 GAPS sites in 2013 and 2015 with those water equilibrium of PCBs and OCPs in the Canadian High
reported at the same sites in 2009.139,147 Arctic.90 While studies involving paired air and water samples
have been dominated by PE-based PAS, not all studies used
them. PUF-PAS-derived air concentrations have been combined
G.8 PASs in studies of air–surface exchange with water concentrations obtained from a passive water
Quite regularly, surface compartments, such as soil and water, sampler based on PDMS sheets to derive air–water fugacity
are sampled at PAS deployment sites, oen with the intention to ratios for selected PAHs, PCBs, OCPs, and PBDEs at three sites
learn something about the equilibrium status of SVOCs in the Aegean Sea.103 Accounting for the gas-phase fraction of
between the atmosphere and these compartments and possibly the air concentration measured by PUF-PAS was possible,
estimate air–surface exchange uxes. because active air sampling conducted at the same time allowed
G.8.1 PASs in studies of air–water exchange. The very rst for the derivation of that fraction. Atmospheric concentrations
eld study of a PAS involved the placement of SPMDs in the measured by PUF-PAS have also been combined with grab water
water and the overlying atmosphere of a coastal lagoon.35 samples to estimate air–water fugacity ratios of several OCPs in
Another early study also deployed SMPDs in air and water in the the lower Ganga River.359
vicinity of a chemical factory to record the levels of PAHs36 and G.8.2 PASs in studies of air–soil exchange. It is quite
while both these studies estimated concentrations, no attempt common to analyze soil samples taken at or in the vicinity of
was made to relate the concentration in different media with passive air sampling sites,301,353,364,377,459,460,476,551 for example to
each other. On the other hand, Booij and van Drooge used four compare the relative composition of PCB mixtures in different
simultaneous deployments of SPMDs in air, water and within media.387 In particular, spatial trends of concentrations in air
the sea surface microlayer to probe the equilibrium status of and soil along elevation gradients have been used to study the
HCB and PCBs in the Western Wadden Sea of the Netherlands.37 occurrence of mountain cold trapping.390,391,452,471
The study indicated no enrichment of these SVOC at the water Measured air–soil concentration ratios can be compared
surface and a phase distribution close to equilibrium. with those that might be expected if air and soil were in equi-
PE was rst used to sample both air and water at the same librium.280 It is also possible to calculate fugacity gradients
location and derive an activity or fugacity gradient for PCBs between soil and air and infer the likely ux direction from that
and PAHs in Narragansett Bay.72,77 Differences in the activity gradient. Daly et al. were the rst to do this with air concen-
or fugacity of a chemical between different compartments trations derived from PASs and also explored the considerable
indicate deviations from thermodynamic equilibrium and uncertainty inherent in this approach.390 Not only is the esti-
therefore the potential for diffusive exchange between those mation of a fugacity from a soil concentration dependent on
compartments. Another early study using this approach was numerous assumptions and possibly quite uncertain partition
for PCDD/Fs in Newark Bay, New Jersey, USA, and revealed coefficients, the uncertainty of the SR of a PAS directly propa-
very strong fugacity gradients favouring volatilisation from gates to the uncertainty of the air fugacity value. Therefore, only
the bay to the atmosphere.78 The approach is also suitable for calculated fugacity gradients that deviate strongly from equi-
sampling from a moving ship by deploying the PE-PAS on the librium should be interpreted as indicating the potential for
ship's mast and sampling the water either by towing a PE volatilization or deposition.390 This approach has since been
strip behind the ship or pumping sea water through a pipe adopted repeatedly,340,341,344,346,347,356,363,375,376,394,543 not always
containing the PE strip.79 This approach was applied to study with full awareness of the signicant uncertainties it entails. It
the air–water exchange of PBDEs along a cruise track across is also important to stress that only the gas-phase concentration
the tropical Atlantic Ocean.79 contributes to a chemical's fugacity in air. Since there is now
PE has been deployed in paired air and water passive considerable evidence that the PUF-PAS samples particle-bound
samplers along the shores and on several on-lake locations of substances (Section F.6), an air concentration derived from
Lakes Ontario and Erie (number of sites varied from 13 to 22) a PUF-PAS should only be used to calculate a fugacity if the
during several 2 month long periods in 2011 and 2012 chemical is entirely in the gas-phase.
(number of deployments varied from 3 to 4 depending on Sometimes, generic mass transfer coefficients are applied to
location). Those samples have been analyzed for PAHs,82 the fugacity gradient to estimate uxes across the air–soil
OCPs,80 PCBs,81,85 PBDEs and other BFRs83 and musk interface.346,394 This is quite questionable, as (1) air concentra-
compounds.84 In a similar set-up, PE samplers were deployed at tion averaged over extended periods of time may not be
eleven coastal and three open-lake sites on Lake Superior appropriate for ux estimation, considering that the fugacity
during three deployments in 2011, with quantication of PAHs, gradient will undergo signicant changes as air concentration
PBDEs87 and PCBs and OCPs.88 Error propagation analysis and and temperature are variable at a time scale shorter than the
Monte Carlo simulations have been used to estimate the PAS deployment period,464 (2) the mass transfer coefficients are
uncertainty of air–water exchange uxes that have been calcu- highly site-specic, and (3) the uncertainty of the fugacity ratio
lated from the air and water concentration obtained from PE- is directly propagated to the ux. Such ux values should
based passive samplers.86 therefore not be interpreted as anything but rough order of
PE-PASs have more recently been used to conrm the net magnitude estimates.
deposition of OCPs and PAHs from the atmosphere to moun- Another approach to measuring soil–air exchange uxes is
tain lakes in Brazilian National Parks89,484 and to assess the air– not based on the analysis of soil samples, but on recording

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1975
Environmental Science: Processes & Impacts Critical Review

a gradient of air concentrations above the soil. These studies than were taken up cumulatively in four PUF-PASs deployed
were reviewed in Section G.3.1. sequentially for one week each at the same time and place.262
Together the studies by Holt et al.,514 Melymuk et al.262 and
Karásková et al.552 are a sobering illustration of the large
H. Recommendations and outlook uncertainties one might expect even from one of the most
regarding PASs for SVOCs thoroughly characterised PAS when deployed and analyzed by
a highly experienced group.
H.1 Comprehensive and quantitative evaluation of PASs for Melymuk et al. summarised many of the challenges beset-
SVOCs ting the passive air sampling for SVOCs and highlighted the
It is now more than 25 years since the rst PAS for SVOCs has appreciable uncertainties and biases to which they give rise.284 It
been introduced34 and since then well in excess of 300 studies is important that the community of PAS researchers and users is
have used PASs for SVOCs in a wide range of applications (see responding appropriately to the repercussions of the ndings of
Section G). Some of these applications include governmental these evaluation studies. For the community of PAS researchers,
monitoring programs, such as the programs for Global Atmo- it means that future effort should be dedicated to quantifying
spheric Passive Sampling (GAPS)208 and Joint Oilsands Moni- and reducing the uncertainties of the passive air sampling of
toring (JOSM),292 as well as large epidemiological investigations SVOCs based on a comprehensive understanding of how it
such as the Airborne Exposure to Semi-volatile Organic Pollut- works. New sampler designs and modications of existing
ants (AESOP) study.492 One would think that this represents designs should be targeted to address the identied short-
a mature eld, where the community of PAS developers and comings of the existing designs. This is further discussed in
users has a solid understanding of the strengths, limitations Section H.2. For the community of PAS users, it implies the
and uncertainties of this sampling technique. However, it is need to become aware of, and knowledgeable about, the limi-
only as a result of a number of comprehensive evaluation tations of the tools they apply and consequently to become
studies conducted during the past few years254,262,514,552 that more transparent and explicit in reporting the uncertainties of
increased awareness and a better quantitative understanding of the results they report. Section H.3 expands on this further.
these limitations and uncertainties have emerged.
Holt et al. used the long time series of concurrent air samples
taken in Košetice in the Czech Republic to assess the agreement H.2 Recommendations and outlook regarding the design of
between a PUF-PAS deployed for one month and four one-day PASs for SVOCs
HiVol samples taken during the same time.514 For most PCBs, This review revealed a surprisingly large number of designs for
OCPs and PAHs, concentrations obtained from the two sampling PASs for SVOCs. Tables 1, 2, 4 and 5 list almost 40 different
techniques were statistically signicantly different from each designs that have been proposed. Many of these designs are only
other, irrespective of whether the effective air volume Vair was the subject of one or two publications, i.e. these designs remain
derived using site-specic or generic SRs and what assumptions for the most part poorly characterised and nd only very limited, if
were made concerning the relative SR of particle-bound any, applications. On the other hand, there are a number of PAS
substances. Correlations between concentrations obtained from designs for which a large body of work exists, that created valuable
passive and active sampling were oen poor and linear regres- information on SRs and their variability, e.g. with respect to the
sions hardly ever showed a slope close to 1. While the average of inuence of wind or to the differences between different types of
12 monthly PUF-PASs and the average of the 48 active air samples SVOCs (see Section F), as well empirical quantitative data and
taken during the same year were generally within the same order estimation techniques on the uptake capacity of their sorbent
of magnitude, for less volatile PAHs discrepancies larger than an materials (see Section E). Some of these have also been rigorously
order of magnitude were common. Some of these discrepancies evaluated (Section H.1). As a result, it is possible to apply these
may be attributed to uncertainties introduced during analytical PASs with a reasonably clear understanding of their limitations
quantication and the different temporal coverage of the two and the expected uncertainty of the data they yield.
sampling approaches (14% overlap), but they are too large to be This is not to say that these PAS designs could not be
fully explained by these sources of uncertainty.514 improved upon. On the contrary, some of the most frequently
When trying to calibrate the PUF-PAS for a number of per- used PAS designs are quite compromised. The rst PAS for
uorinated substances, Karásková et al. oen failed to observe SVOCs, the SPMD, has turned out to have too complex and too
an increase in the sequestered amount with increasing variable an uptake kinetic and to pose too many analytical
deployment time indoors.552 Also, the uptake curves obtained challenges that it has ceased to be a competitive choice for most
for outdoor deployments showed poor replicate precision and applications, despite a fairly substantive body of work charac-
a very large range in SRe (0.7 to 30 m3 per day). Because many of terising it. Even the PUF-PAS, which has become the default
these peruorinated compounds are charged and have choice of many PAS studies, leans heavily on the side of high
extremely limited volatility, these results hint at the large vari- SRs in the balancing of the conicting objectives of high vs.
ability in the uptake of non-gaseous substances in PAS, espe- tightly controlled SRs and therefore these SRs are variable and
cially indoors, where air turbulence is very limited and/or highly highly uncertain (see Section C.1). While PUF is a popular
spatially variable. Melymuk et al. found much lower levels of choice as a sorbent because of its ease of handling and low cost,
several PAHs and BFRs in PUF-PASs deployed for one month its low uptake capacity necessitates complex SR estimations for

1976 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

more volatile SVOCs that introduce considerable uncertainty PE of the Radiello is itself too large and would interfere with the
(see Section F.3). diffusion of the SVOC to the sorbent. However, it may be feasible
Introducing a new sampler design only makes sense, if two to construct a radial diffusive barrier from material that has less
conditions are fullled. There has to be a: capacity to sorb SVOCs than PE.
(1) Clear motivation for proposing a new design, i.e. it has to Apart from their potential analysis by thermal desorption,
fundamentally improve on the existing, well-characterised PAS PDMS-based PASs show considerable promise for indoor
designs for at least one type of application. Sadly, the evidence applications, because of the ease of handling and because fast
suggests that some of the researchers proposing new designs diffusion within the polymer should assure that diffusion
are not even always aware of the full slate of PAS designs that across the air boundary layer is rate-limiting uptake. A hurdle to
have been described in the literature and therefore may be ill- their credibility currently is a general disregard of the need to
equipped to make a genuinely useful contribution. constrain the thickness of that air boundary layer to a small
(2) Commitment to undertake the experiments that are range, which even in indoor settings would be required to
required for comprehensive characterisation of a new design. At increase precision, and the failure to observe the theoretically
a minimum that needs to include (i) a calibration study with expected quantitative relationship between surface area and
replicated samplers being retrieved aer various lengths of time uptake rate, at least when the results of various studies are
and with simultaneous measurement of air concentration with compared (see Table 3). If that issue could be resolved, it should
active air sampler, as this is what is necessary to obtain infor- be quite easily possible to modify the surface area to volume
mation on tlinear and tequilibrium for SVOCs of different volatility ratio of the PDMS sorbent in order to tailor a PAS for a particular
and at different temperatures, (ii) an exploration of the vari- set of deployment conditions (i.e. a specic set of SVOCs
ability in SRs that can be expected in different types of sampled for a particular length of time, given a particular LOD
deployments. of the detection technique).
There is certainly a need for a PAS for SVOCs that can
generate air concentrations with a much smaller uncertainty
than the current slate of sampler designs. Such a PAS would H.3 Recommendations and outlook regarding applications
be useful for applications where small concentration differ- of PASs for SVOCs
ences need to be discerned (e.g. when measuring vertical As Section G illustrates, PASs for SVOCs now nd use for a wide
gradients, see Section G.3) or when highly accurate numbers variety of applications. A perusal of those studies reveals that
are required (e.g. when the potential for diffusive air–surface the considerable uncertainty of results obtained with most
exchange of SVOCs is to be deduced, see Section G.8). Such currently used PASs is oen profoundly underappreciated.
a sampler would almost certainly need to apply effective wind Partly, this may be attributable to an overestimation of the
sheltering and a diffusive barrier in order to achieve the performance capability of a PAS. More typically, it is a result of
tightly controlled SRs necessary for high precision (see the reluctance of many PAS users to quantify or even consider
Section B.4). Because that would inevitably lead to small SRs, that uncertainty. Going forward, there should be an expectation
the analysis of the amount of SVOC on the sorbent of such that certain elements are present in a PAS-based study, if it is to
a sampler will need to achieve very low LODs. There are now be published in the peer-reviewed literature or if the generated
several studies that have shown that an analytical approach data serve regulatory purposes.
that thermally desorbs the sequestered chemical from A good PAS study will replicate at least some of the sampler
a sorbent onto a GC-MS is capable of delivering such low deployments and have a large number of eld blanks. When
LODs. This is most easily achieved by a sorbent that can be targeting SVOCs that are known to be susceptible to blank
placed in the tubes used by commercial thermal desorption contamination (e.g. cVMS) and/or when sampling very low
instruments. Sorbents that have been thermally desorbed concentrations (see e.g. Section G.6), it may be advisable to have
include silicone coated stir bars, 52,55,108–110 Tenax TA-coated a eld blank for every deployed sampler.
GFFs,199,200 and commercially available Tenax TA lled Importantly, the uncertainty of the reported results needs to
cartridges.201 be explicitly stated. If results are reported in sequestered
Oen and maybe surprisingly, there has been no attempt to amounts (normalised to time), it may be sufficient to estimate
control the diffusion distance to these sorbents. In studies where that uncertainty from the standard deviation of replicated
this was attempted, e.g. by using different types of PE-based sampler deployments. If volumetric air concentrations are re-
diffusive barriers,55 this has not been successful, as evidenced ported, the uncertainty of the applied SRs or effective sampling
by high and unexplained SR variability. Interestingly, some of the volumes needs to be considered in addition to the sampling and
thermally stable sorbents that have been proposed, such as the analytical uncertainty. As Section F.3 illustrates the uncertainty
commercially available Tenax TA-lled cartridges and silicone of SRs is oen very large. If an effective sampling volume is
rods, would be compatible with the commercial Radiello diffusive estimated for a PAS operating in the curvi-linear uptake regime,
barriers, which for example nds use for passively sampling the uncertainty of the variables going into that estimation need
VOCs.273 Yet, there appears to have been no study that explored to be propagated to the nal reported air concentration. These
the feasibility of a PAS for SVOCs that combines a thermally stable variables include the KSG value, which is oen estimated from
sorbent in rod-shaped form with a Radiello diffusive barrier. It is a highly uncertain empirical regression equation and then
possible, and maybe even likely, that the sorptive capacity of the adjusted to the temperature of deployment using an equally

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1977
Environmental Science: Processes & Impacts Critical Review

uncertain energy of sorption,263 and it includes the intrinsic SR DUSG Internal energy of transfer from the sorbed phase to
determined from the loss of depuration compounds. The study the gas-phase
by Liu et al. illustrates how this can be done.86 Vair Effective sampling volume of a passive air sampler
An honest and comprehensive appraisal of uncertainty will VS Geometric bulk volume of a passive sampling
most likely reveal that in many instances a PAS-derived air sorbent
concentration for SVOCs may essentially be an order of Dz Thickness of the stagnant air boundary layer
magnitude estimate. This is especially true for particle-bound surrounding a sorbent
substances. It is important to stress that this does not invali-
date the usefulness of PASs as in many applications even highly
uncertain data are perfectly sufficient to meet the project I.2 Abbreviations for different samplers and sorbent
objectives. As was discussed in Sections G.5.1 and G.7, amounts materials
sequestered in a PAS have much less uncertainty than air
concentrations and are oen appropriate to answer the ques- ACF Activated carbon felt
tions related to spatial variability and temporal trends that PASs AL Articial leaf
are frequently called upon to address. However, honest uncer- EVA Ethyl vinyl acetate
tainty estimates will also reveal that some of the interpretations FTS Flow through sampler
of PAS-based studies will have to be questioned. Small differ- GFF Glass bre lter
ences in levels measured at forested and open sites may be HiVol High volume sampler
caused by SRs that depend on wind and not necessarily be an LoVol Low volume sampler
indication of the forest lter effect.65,468,478 The same applies to MECOP Membrane-enclosed copolymer
small differences in paired indoor/outdoor PAS deployments. PAS Passive air sampler
Even air–soil or air–water fugacity fractions of 0.1 or 0.9 may not PAS-DD Passive dry deposition sampler
be condently distinguished from equilibrium, if the air PDMS Polydimethylsiloxane
concentration is derived from a PAS. If efforts to develop a high PE, LDPE Polyethylene, low-density polyethylene
precision PAS for SVOCs, as advocated in Section H.2, should PES Passive emission sampler
succeed it will be time to revisit those PAS applications that rely POG Polymer-coated glass
on measured air concentrations with small uncertainty. PUF Polyurethane foam
SBSE Stir bar sorptive extraction
SIP Sorbent impregnated polyurethane
I. Acronyms, abbreviations, variables SPMD Semi-permeable membrane device
I.1 Variables SPME Solid phase micro extraction
VERAM Versatile, easy and rapid atmospheric monitor
XAD Styrene–divinylbenzene co-polymer
A Geometric surface area of the sorbent exposed to the
gas-phase
AS Specic surface area of an adsorbent
CG Concentration of chemical in the gas-phase I.3 Abbreviations for compounds and compound groups
CS Concentration of chemical in the sorbent of
a passive air sampler SVOCs Semi-volatile organic chemicals
D Molecular diffusivity in the gas-phase VOCs Volatile organic chemicals
DHSG Enthalpy of transfer from the sorbed phase to the
gas-phase BFRs Brominated ame retardants
kl Rate constant describing chemical loss from CUPs Current use pesticides
a passive air sampler HFR Halogenated ame retardants
ku Rate constant describing chemical uptake in MCCPs Medium chain chlorinated paraffins
a passive air sampler OCPs Organochlorine pesticides
KSG Equilibrium sorption coefficient between a sorbent OPE Organophosphate esters
and the gas-phase OPFRs Organophosphate ame retardants
mS Amount of chemical taken up in the sorbent of PAHs Polycyclic aromatic hydrocarbons
a passive air sampler PBDEs Polybrominated diphenyl ethers
R Ideal gas constant PCBs Polychlorinated biphenyls
SR Inherent sampling rate of a passive air sampler PCDD/Fs Polychlorinated dibenzo-p-dioxins and
SRe Empirical sampling rate of a passive air sampler dibenzofurans
Dt Deployment time of a passive air sampler PCNs Polychlorinated naphthalenes
tlinear Length of the linear uptake period of a passive air PFAS Peruorinated and polyuorinated alkyl
sampler substances
tequilibrium Time needed to approach equilibrium between SCCPs Short-chain chlorinated paraffins
sorbent and gas-phase VMS Volatile methyl siloxanes
T Temperature

1978 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

ADBI Celestolide 2 G. R. Carmichael, M. Ferm, N. Thongboonchoo, J.-H. Woo,

AHTN Tonalide L. Y. Chan, K. Murano, P. H. Viet, C. Mossberg, R. Bala,
ATE 2,4,6-Tribromophenylallyl ether J. Boonjawat, P. Upatum, M. Mohan, S. P. Adhikary,
ATII Traseolide A. B. Shrestha, J. J. Pienaar, E. B. Brunke, T. Chen, T. Jie,
BATE 2-Bromoallyl-2,4,6-tribromo-phenyl ether D. Guoan, L. C. Peng, S. Dhiharto, H. Harjanto,
BTBPE 1,2-Bis(2,4,6-tribromo-phenoxy)ethane A. M. Jose, W. Kimani, A. Kirouane, J.-P. Lacaux,
D3 Hexamethylcyclotrisiloxane S. Richard, O. Barturen, J. C. Cerda, A. Athayde,
D4 Octamethylcyclotetrasiloxane T. Tavares, J. S. Cotrina and E. Bilici, Measurements of
D5 Decamethylcyclopentasiloxane sulfur dioxide, ozone and ammonia concentrations in
DBDPE Decabromodiphenylethane Asia, Africa, and South America using passive samplers,
DDT Dichlorodiphenyltrichloroethane Atmos. Environ., 2003, 37, 1293–1308.
DDE Dichlorodiphenyldichloroethylene 3 S. Mukerjee, L. A. Smith, G. A. Norris, M. T. Morandi,
DPTE 2,3-Dibromopropyl-2,4,6-tribromophenyl ether M. Gonzales, C. A. Noble, L. M. Neas and A. H. Özkaynak,
EHTBB 2-Ethylhexyl-2,3,4,5-tetrabromobenzoate Field method comparison between passive air samplers
EtFOSA N-Ethyl peruorooctane sulfonamide and continuous monitors for VOCs and NO2 in El Paso,
EtFOSE N-Ethyl peruorooctane sulfonamidoethanol Texas, J. Air Waste Manage. Assoc., 2004, 54, 307–319.
8:2 FTOH Peruorooctyl ethanol
4 M. A. Puchalski, M. E. Sather, J. T. Walker,
10:2 FTOH Peruorodecyl ethanol
C. M. B. Lehmann, D. A. Gay, J. Mathew and
GEM Gaseous elemental mercury
W. P. Robarge, Passive ammonia monitoring in the
HBB Hexabromobenzene
United States: Comparing three different sampling
HBCD Hexabromocyclododecane
devices, J. Environ. Monit., 2011, 13, 3156–3167.
HCBD Hexchlorobutadiene
HCB Hexachlorobenzene 5 D. S. McLagan, M. E. E. Mazur, C. P. J. Mitchell and
HCH Hexachlorocyclohexane F. Wania, Passive air sampling of gaseous elemental
HHCB Galaxolide mercury: A critical review, Atmos. Chem. Phys., 2016, 16,
MCPA 2-Methyl-4-chlorophenoxyacetic acid 3061–3076.
MDM Octamethyltrisiloxane 6 G. Zhang and X. Liu, Passive atmospheric sampling of
MD2M Decamethyltetrasiloxane persistent organic pollutants (POPs), Prog. Chem., 2009,
MD3M Dodecamethylpentasiloxane 21, 297–306.
MeFOSA N-Methyl peruorooctane sulfonamide 7 L. Tuduri, M. Millet, O. Briand and M. Montury, Passive air
MeFOSEA N-Methylperuorooctane sulfonamidethylacry-late sampling of semi-volatile organic compounds, TrAC, Trends
MeFOSE N-Methyl peruorooctane sulfonamidoethanol Anal. Chem., 2012, 31, 38–49.
PBEB Pentabromoethylbenzene 8 M. Partyka, B. Zabiegała, J. Namieśnik and A. Przyjazny,
PBT 2,3,4,5,6-Pentabromotoluene Application of passive samplers in monitoring of organic
PCA Pentachloroanisole constituents of air, Crit. Rev. Anal. Chem., 2007, 37, 51–78.
PeCBz Pentachlorobenzene 9 A. Kot-Wasik, B. Zabiegała, M. Urbanowicz, E. Dominiak,
TBBP-A Tetrabromobisphenol-A A. Wasik and J. Namieśnik, Advances in passive sampling
TBE 1,2-Bis(2,4,6-tribromophenoxy)ethane in environmental studies, Anal. Chim. Acta, 2007, 602,
TBECH Tetrabromoethyl-cyclohexane 141–163.
TBCO 1,2,5,6-Tetrabromocyclooctane 10 S. Seethapathy, T. Górecki and X. Li, Passive sampling in
p-TBX 2,3,5,6-Tetrabromo-p-xylene
environmental analysis, J. Chromatogr. A, 2008, 1184, 234–
TBPH Bis-(2-ethylhexyl)tetrabromophthalate
253.
11 F. A. Esteve-Turrillas and A. Pastor, Passive air sampling,
Compr. Anal. Chem., 2016, 73, 203–232.
Conflicts of interest 12 F. A. Esteve-Turrillas, V. Yusá, A. Pastor and M. de la
There are no conicts of interest to declare. Guardia, New perspectives in the use of semipermeable
membrane devices as passive samplers, Talanta, 2008, 74,
443–457.
Acknowledgements 13 C. Qu, A. L. Doherty, X. Xing, W. Sun, S. Albanese, A. Lima,
We are grateful to the Natural Sciences and Engineering S. Qi and B. De Vivo, Polyurethane foam-based passive air
Research Council of Canada for funding most of our work on samplers in monitoring persistent organic pollutants:
passive air sampling of SVOCs over the past 20 years. Theory and application, in Environmental Geochemistry:
Site Characterization, Data Analysis and Case Histories, ed.
References B. De Vivo, H. Belkin and A. Lima, Elsevier, 2nd edn, 2018,
ch. 20, pp. 521–542.
1 R. M. Cox, The use of passive sampling to monitor forest 14 F. Salim and T. Górecki, Theory and modelling approaches
exposure to O3, NO2 and SO2: a review and some case to passive sampling, Environ. Sci.: Processes Impacts, 2019,
studies, Environ. Pollut., 2003, 126, 301–311. 21, 1618–1641.

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1979
Environmental Science: Processes & Impacts Critical Review

15 J. R. Aboal, E. Concha-Graña, F. De Nicola, S. Muniategui- 28 I. S. Krogseth, X. Zhang, Y. D. Lei, F. Wania and K. Breivik,
Lorenzo, P. López-Mahı́a, S. Giordano, F. Capozzi, A. Di Calibration and application of a passive air sampler (XAD-
Palma, R. Reski, H. Zechmeister, J. Martı́nez-Abaigar and PAS) for volatile methyl siloxanes, Environ. Sci. Technol.,
J. A. Fernández, Testing a novel biotechnological passive 2013, 47, 4463–4470.
sampler for monitoring atmospheric PAH pollution, J. 29 L. Shen, F. Wania, Y. D. Lei, C. Teixeira, D. C. G. Muir and
Hazard. Mater., 2020, 381, 120949. T. F. Bidleman, Atmospheric distribution and long-range
16 S. E. Gingrich, M. L. Diamond, G. A. Stern and transport behavior of organochlorine pesticides in North
B. E. McCarry, Atmospherically derived organic surface America, Environ. Sci. Technol., 2005, 39, 409–420.
lms along an urban-rural gradient, Environ. Sci. Technol., 30 J. E. Balmer, H. Hung, K. Vorkamp, R. J. Letcher and
2001, 35, 4031–4037. D. C. G. Muir, Hexachlorobutadiene (HCBD)
17 M. Unger and O. Gustafsson, PAHs in Stockholm window contamination in the Arctic environment: A review,
lms: Evaluation of the utility of window lm content as Emerging Contam., 2019, 5, 116–122.
indicator of PAHs in urban air, Atmos. Environ., 2008, 42, 31 C. Rauert, T. Harner, J. K. Schuster, A. Eng, G. Fillmann,
5550–5557. L. E. Castillo, O. Fentanes, M. Villa Ibarra,
18 E. Terzaghi, M. Scacchi, B. Cerabolini, K. C. Jones and A. Di K. S. B. Miglioranza, I. Moreno Rivadeneira, K. Pozo and
Guardo, Estimation of polycyclic aromatic hydrocarbon B. H. Aristizábal Zuluaga, Atmospheric concentrations of
variability in air using high volume, lm, and vegetation new persistent organic pollutants and emerging
as samplers, Environ. Sci. Technol., 2015, 49, 5520–5528. chemicals of concern in the Group of Latin America and
19 C.-Y. Huo, Y. Sun, L.-Y. Liu, E. Sverko, Y.-F. Li, W.-L. Li, Caribbean (GRULAC) region, Environ. Sci. Technol., 2018,
W.-L. Ma, Z.-F. Zhang and W.-W. Song, Assessment of 52, 7240–7249.
human indoor exposure to PAHs during the heating and 32 Y.-G. Ma, Y. D. Lei, H. Xiao, F. Wania and W.-H. Wang, Critical
non-heating season: Role of window lms as passive air review and recommended values for the physical-chemical
samplers, Sci. Total Environ., 2019, 659, 293–301. property data of 15 polycyclic aromatic hydrocarbons at 25

20 M. E. Bartkow, K. Booij, K. E. Kennedy, J. F. Müller and C, J. Chem. Eng. Data, 2010, 55, 819–825.
D. W. Hawker, Passive air sampling theory for 33 N. Li, F. Wania, Y. D. Lei and G. L. Daly, A comprehensive
semivolatile organic compounds, Chemosphere, 2005, 60, and critical compilation, evaluation and selection of
170–176. physical chemical property data for selected
21 J. M. Armitage, S. J. Hayward and F. Wania, Modeling the polychlorinated biphenyls, J. Phys. Chem. Ref. Data, 2003,
uptake of neutral organic chemicals on XAD passive air 32, 1535–1590.
samplers under variable temperatures, external wind 34 J. D. Petty, J. N. Huckins and J. L. Zajicek, Application of
speeds and ambient air concentrations (PAS-SIM), semipermeable membrane devices (SPMDs) as passive air
Environ. Sci. Technol., 2013, 47, 13546–13554. samplers, Chemosphere, 1993, 27, 1609–1624.
22 D. S. McLagan, C. P. J. Mitchell, H. Huang, Y. D. Lei, 35 H. F. Prest, L. A. Jacobson and J. N. Huckins, Passive
A. S. Cole, A. Steffen, H. Hung and F. Wania, A high sampling of water and coastal air via semipermeable
precision passive air sampler for gaseous mercury, membrane devices, Chemosphere, 1995, 30, 1351–1362.
Environ. Sci. Technol. Lett., 2016, 3, 24–29. 36 J. Caslavsky, Z. Zdrahal and M. Vytopilova, Application of
23 M. A. Snow, Development, characterization and eld testing SPMDs for PAH sampling in the DEZA chemical factory,
of a passive air sampler for monitoring occupational Polycyclic Aromat. Compd., 2000, 20, 123–141.
exposure to mercury vapors, MSc thesis, University of 37 K. Booij and B. L. van Drooge, Polychlorinated biphenyls
Toronto, Department of Chemistry, April 2020, p. 108. and hexachlorobenzene in atmosphere, sea-surface
24 M. Shoeib and T. Harner, Characterization and comparison microlayer, and water measured with semi-permeable
of three passive air samplers for persistent organic membrane devices (SPMDs), Chemosphere, 2001, 44, 91–98.
pollutants, Environ. Sci. Technol., 2002, 36(19), 4142–4151. 38 W. A. Ockenden, H. F. Prest, G. O. Thomas, A. Sweetman
25 L. Melymuk, M. Robson, P. A. Helm and M. L. Diamond, and K. C. Jones, Passive air sampling of PCBs: Field
Evaluation of passive air sampler calibrations: Selection calculation of atmospheric sampling rates by triolein-
of sampling rates and implications for the measurement containing semipermeable membrane devices, Environ.
of persistent organic pollutants in air, Atmos. Environ., Sci. Technol., 1998, 32, 1538–1543.
2011, 45, 1867–1875. 39 R. Lohmann, B. P. Corrigan, M. Howsam, K. C. Jones and
26 F. Wania, L. Shen, Y. D. Lei, C. Teixeira and D. C. G. Muir, W. A. Ockenden, Further developments in the use of
Development and calibration of a resin-based passive semipermeable membrane devices (SPMDs) as passive air
sampling system for persistent organic pollutants in the samplers for persistent organic pollutants: Field
atmosphere, Environ. Sci. Technol., 2003, 37, 1352–1359. application in a spatial survey of PCDD/Fs and PAHs,
27 T. Gouin, F. Wania, C. Ruepert and L. E. Castillo, Field Environ. Sci. Technol., 2001, 35, 2576–2582.
testing passive air samplers for current use pesticides in 40 W. A. Ockenden, A. Sweetman, H. F. Prest, E. Steinnes and
a tropical environment, Environ. Sci. Technol., 2008, 42, K. C. Jones, Toward an understanding of the global
6625–6630. atmospheric distribution of persistent organic pollutants:
The use of semipermeable membrane devices as time-

1980 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

integrated passive samplers, Environ. Sci. Technol., 1998, 32, 53 S.-D. Choi, S.-Y. Baek and Y.-S. Chang, Inuence of a large
2795–2803. steel complex on the spatial distribution of volatile
41 W. A. Ockenden, B. P. Corrigan, M. Howsam and polycyclic aromatic hydrocarbons (PAHs) determined by
K. C. Jones, Further developments in the use of passive air sampling using membrane-enclosed coplolmer
semipermeable membrane devices as passive air (MECOP), Atmos. Environ., 2007, 41, 6255–6264.
samplers: Application to PCBs, Environ. Sci. Technol., 54 F. A. Esteve-Turrillas, S. Ly-Verdú, A. Pastor and M. de la
2001, 35, 4536–4543. Guardia, Development of a versatile, easy and rapid
42 S. N. Meijer, W. A. Ockenden, E. Steinnes, B. P. Corrigan atmospheric monitor for benzene, toluene, ethylbenzene
and K. C. Jones, Spatial and temporal trends of POPs in and xylenes determination in air, J. Chromatogr. A, 2009,
Norwegian and UK background air: Implications for 1216, 8549–8556.
global cycling, Environ. Sci. Technol., 2003, 37, 454–461. 55 H. Paschke and P. Popp, New passive samplers for
43 F. Jaward, S. N. Meijer, E. Steinnes, G. Thomas and chlorinated semivolatile organic pollutants in ambient
K. C. Jones, Further studies on the latitudinal and air, Chemosphere, 2005, 58, 855–863.
temporal trends of persistent organic pollutants in 56 Y. Wang, A. J. Easteal and X. D. Chen, Ethylene and oxygen
Norwegian and U.K. background air, Environ. Sci. permeability through polyethylene packaging lms, Packag.
Technol., 2004, 38, 2523–2530. Technol. Sci., 1998, 11, 169–178.
44 R. Gioia, E. Steinnes, G. O. Thomas, S. N. Mejier and 57 K. Booij, H. E. Hofmans, C. V. Fischer and E. M. Van
K. C. Jones, Persistent organic pollutants in European Weerlee, Temperature-dependent uptake rates of
background air: derivation of temporal and latitudinal nonpolar organic compounds by semipermeable
trends, J. Environ. Monit., 2006, 8, 700–710. membrane devices and low-density polyethylene
45 J. K. Schuster, R. Gioia, K. Breivik, E. Steinnes, membranes, Environ. Sci. Technol., 2003, 37(2), 361–366.
M. Scheringer and K. C. Jones, Trends in European 58 W. Levy, B. Henkelmann, G. Pster, M. Kirchner, G. Jakobi,
background air reect reductions in primary emissions of A. Niklaus, J. Kotalik, S. Bernhö, N. Fischer and
PCBs and PBDEs, Environ. Sci. Technol., 2010, 44, 6760– K.-W. Schramm, Monitoring of PCDD/Fs in a mountain
6766. forest by means of active and passive sampling, Environ.
46 H. S. Söderström and P.-A. Bergqvist, Polycyclic aromatic Res., 2007, 105, 300–306.
hydrocarbons in a semiaquatic plant and semipermeable 59 M. T. Piccardo, A. Stella, M. Pala, D. Balducci and F. Valerio,
membrane devices exposed to air in Thailand, Environ. Field use of semipermeable membrane devices (SPMDs) for
Sci. Technol., 2003, 37, 47–52. passive air sampling of polycyclic aromatic hydrocarbons:
47 H. Söderström, J. Hajslova, V. Kocourek, B. Siegmund, Opportunities and limitations, Atmos. Environ., 2010, 44,
A. Kocan, M. W. Obiedzinski, M. Tysklind and 1947–1951.
P.-A. Bergqvist, PAHs and nitrated PAHs in air of ve 60 W. Levy, B. Henkelmann, G. Pster, S. Bernho,
European countries determined using SPMDs as passive M. Kirchner, G. Jakobi, R. Bassan, N. Kräuchi and
samplers, Atmos. Environ., 2005, 39, 1627–1640. K.-W. Schramm, Long-term air monitoring of
48 H. Söderström and P.-A. Bergqvist, Passive air sampling organochlorine pesticides using Semi Permeable
using semipermeable membrane devices at different Membrane Devices (SPMDs) in the Alps, Environ. Pollut.,
wind-speeds in situ calibrated by performance reference 2009, 157, 3272–3279.
compounds, Environ. Sci. Technol., 2004, 38, 4828–4834. 61 S. Fiedler, G. Pster and K.-W. Schramm, Partitioning of
49 B. Strandberg, P. Gustafson, H. Söderström, L. Barregard, uorotelomer alcohols (FTOH) to semipermeable
P.-A. Bergqvist and G. Sällsten, The use of semipermeable membrane devices (SPMD), Environ. Sci. Pollut. Res., 2010,
membrane devices as passive samplers to determine 17, 420–428.
persistent organic compounds in indoor air, J. Environ. 62 X. Zhu, C. Zhou, B. Henkelmann, Z. Wang, X. Ma, G. Pster,
Monit., 2006, 8, 257–262. K.-W. Schramm, J. Chen, Y. Ni, W. Wang and J. Mu,
50 W. L. Cranor, D. Á. Álvarez, J. N. Huckins and J. D. Petty, Monitoring of PAHs proles in the urban air of Dalian,
Uptake rate constants and partition coefficients for vapor China with active high-volume sampler and
phase organic chemicals using semipermeable membrane semipermeable membrane devices, Polycyclic Aromat.
devices (SPMDs), Atmos. Environ., 2009, 43, 3211–3219. Compd., 2013, 33, 265–288.
51 A. Cicenaite, J. N. Huckins, D. Á. Álvarez, W. L. Cranor, 63 Q. Xu, X. Zhu, B. Henkelmann, K.-W. Schramm, J. Chen,
R. W. Gale, V. Kauneliene and P.-A. Bergqvist, Feasibility Y. Ni, W. Wang, G. Pster, J. Mu, S. Qin and Y. Li,
of a simple laboratory approach for determining Simultaneous monitoring of PCB proles in the urban air
temperature inuence on SPMD-air partition coefficients of Dalian, China with active and passive samplings, J.
of selected compounds, Atmos. Environ., 2007, 41, 2844– Environ. Sci., 2013, 25, 133–143.
2850. 64 X. Zhu, T. Shi, B. Henkelmann, G. Pster, W. Wang, J. Mu,
52 L. Wennrich, P. Popp and C. Hafner, Novel integrative X. Li, L. Han and K.-W. Schramm, Monitoring of
passive samplers for the long-term monitoring of organochlorine pesticides in the urban air of Dalian,
semivolatile organic air pollutants, J. Environ. Monit., China with semipermeable membrane devices, Chin. Sci.
2002, 4, 371–376. Bull., 2014, 59, 2957–2965.

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1981
Environmental Science: Processes & Impacts Critical Review

65 N. Zhu, K.-W. Schramm, T. Wang, B. Henkelmann, 78 C. L. Friedman, M. G. Cantwell and R. Lohmann, Passive
X. Zheng, J. Fu, Y. Gao, Y. Wang and G. Jiang, sampling provides evidence for Newark Bay as a source of
Environmental fate and behavior of persistent organic polychlorinated dibenzo-p-dioxins and furans to the New
pollutants in Shergyla Mountain, southeast of the Tibetan York/New Jersey, USA, atmosphere, Environ. Toxicol.
Plateau of China, Environ. Pollut., 2014, 191, 166–174. Chem., 2012, 31, 253–261.
66 J. F. Müller, K. Manomanii, M. R. Mortimer and 79 R. Lohmann, J. Klánová, P. Kukučka, S. Yonis and
M. S. McLachlan, Partitioning of polycyclic aromatic K. Bollinger, Concentrations, uxes, and residence time
hydrocarbons in the polyethylene/water system, Fresenius. of PBDEs across the tropical Atlantic Ocean, Environ. Sci.
J. Anal. Chem., 2001, 371, 816–822. Technol., 2013, 47, 13967–13975.
67 M. E. Bartkow, D. W. Hawker, K. E. Kennedy and 80 M. Khairy, D. Muir, C. Teixeira and R. Lohmann, Spatial
J. F. Müller, Characterizing uptake kinetics of PAHs from trends, sources, and air–water exchange of organochlorine
the air using polyethylene-based passive air samplers of pesticides in the Great Lakes basin using low density
multiple surface area-to volume ratios, Environ. Sci. polyethylene passive samplers, Environ. Sci. Technol.,
Technol., 2004, 38, 2701–2706. 2014, 48, 9315–9324.
68 M. E. Bartkow, K. C. Jones, K. E. Kennedy, N. Holling, 81 M. Khairy, D. Muir, C. Teixeira and R. Lohmann, Spatial
D. W. Hawker and J. F. Müller, Evaluation of performance distribution, air-water fugacity ratios and source
reference compounds in polyethylene-based passive air apportionment of polychlorinated biphenyls in the lower
samplers, Environ. Pollut., 2006, 144, 365–370. Great Lakes basin, Environ. Sci. Technol., 2015, 49, 13787–
69 L. B. Paulik, C. E. Donald, B. W. Smith, L. G. Tidwell, 13797.
K. A. Hobbie, L. Kincl, E. N. Haynes and K. A. Anderson, 82 C. A. McDonough, M. A. Khairy, D. C. G. Muir and
Emissions of polycyclic aromatic hydrocarbons from R. Lohmann, Signicance of population centers as
natural gas extraction into air, Environ. Sci. Technol., 2016, sources of gaseous and dissolved PAHs in the lower Great
50, 7921–7929. Lakes, Environ. Sci. Technol., 2014, 48, 7789–7797.
70 C. E. Donald and K. A. Anderson, Assessing soil-air 83 C. A. McDonough, G. Puggioni, P. A. Helm, D. Muir and
partitioning of PAHs and PCBs with a new fugacity R. Lohmann, Spatial distribution and air–water exchange
passive sampler, Sci. Total Environ., 2017, 596–597, 293– of organic ame retardants in the lower Great Lakes,
302. Environ. Sci. Technol., 2016, 50, 9133–9141.
71 C. E. Donald, R. P. Scott, G. Wilson, P. D. Hoffman and 84 C. A. McDonough, P. A. Helm, D. Muir, G. Puggioni and
K. A. Anderson, Articial turf: chemical ux and R. Lohmann, Polycyclic musks in the air and water of the
development of silicone wristband partitioning lower Great Lakes: Spatial distribution and volatilization
coefficients, Air Qual. Atmos. Hlth., 2019, 12, 597–611. from surface waters, Environ. Sci. Technol., 2016, 50,
72 E. J. Morgan and R. Lohmann, Detecting air–water and 11575–11583.
surface-deep water gradients of PCBs using polyethylene 85 Y. Liu, S. Wang, C. A. McDonough, M. Khairy, D. Muir,
passive samplers, Environ. Sci. Technol., 2008, 42, 7248– P. A. Helm and R. Lohmann, Gaseous and freely-dissolved
7253. PCBs in the lower Great Lakes based on passive sampling:
73 M. A. Khairy and R. Lohmann, Field validation of Spatial trends and air-water exchange, Environ. Sci.
polyethylene passive air samplers for parent and alkylated Technol., 2016, 50, 4932–4939.
PAHs in Alexandria, Egypt, Environ. Sci. Technol., 2012, 46, 86 Y. Liu, S. Wang, C. A. McDonough, M. Khairy, D. Muir and
3990–3998. R. Lohmann, Estimation of uncertainty in air-water
74 M. A. Khairy and R. Lohmann, Feasibility of using low exchange ux and gross volatilization loss of PCBs: A case
density polyethylene sheets to detect atmospheric study based on passive sampling in the lower Great
organochlorine pesticides in Alexandria, Egypt, Environ. Lakes, Environ. Sci. Technol., 2016, 50, 10894–10902.
Pollut., 2013, 181, 151–158. 87 Z. Ruge, D. Muir, P. Helm and R. Lohmann,
75 M. A. Khairy and R. Lohmann, Field calibration of low- Concentrations, trends, and air-water exchange of PAHs
density polyethylene passive samplers for gaseous POPs, and PBDEs derived from passive samplers in Lake
Environ. Sci.: Processes Impacts, 2014, 16, 414–421. Superior in 2011, Environ. Sci. Technol., 2015, 49, 13777–
76 Y. Liu, S. Xie, L. Zheng, T. Li, Y. Sun, L. Ma, Z. Lin, 13786.
P. Grathwohl and R. Lohmann, Air–soil diffusive 88 Z. Ruge, D. Muir, P. Helm and R. Lohmann,
exchange of PAHs in an urban park of Shanghai based on Concentrations, trends, and air-water exchange of PCBs
polyethylene passive sampling: Vertical distribution, and organochlorine pesticides derived from passive
vegetation inuence and diffusive ux, Sci. Total Environ., samplers in Lake Superior in 2011, Environ. Sci. Technol.,
2019, 689, 734–742. 2018, 52, 14061–14069.
77 R. Lohmann, M. Dapsis, E. J. Morgan, V. Dekany and 89 R. O. Meire, M. Khairy, A. C. Targino, P. M. A. Galvão,
P. J. Luey, Determining air–water exchange, spatial and J. P. M. Torres, O. Malm and R. Lohmann, Use of passive
temporal trends of freely dissolved PAHs in an urban samplers to detect organochlorine pesticides in air and
estuary using passive polyethylene samplers, Environ. Sci. water at wetland mountain region sites (S-SE Brazil),
Technol., 2011, 45, 2655–2662. Chemosphere, 2016, 144, 2175–2182.

1982 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

90 Y. Ma, D. A. Adelman, E. Bauerfeind, A. Cabrerizo, 103 G. Lammel, O. Audy, A. Besis, C. Efstathiou,

C. A. McDonough, D. Muir, T. Soltwedel, C. Sun, K. Eleheriadis, J. Kohoutek, P. Kukučka, M. D. Mulder,
C. C. Wagner, E. M. Sunderland and R. Lohmann, P. Přibylová, R. Prokeš, T. P. Rusina, C. Samara,
Concentrations and water mass transport of legacy POPs A. Sofuoglu, S. C. Sofuoglu, Y. Taşdemir, V. Vassilatou,
in the Arctic Ocean, Geophys. Res. Lett., 2018, 45, 12972– D. Voutsa and B. Vrana, Air and seawater pollution and
12981. air–sea gas exchange of persistent toxic substances in the
91 J. B. Wilcockson and F. A. P. C. Gobas, Thin-lm solid- Aegean Sea: spatial trends of PAHs, PCBs, OCPs and
phase extraction to measure fugacities of organic PBDEs, Environ. Sci. Pollut. Res., 2015, 22, 11301–11313.
chemicals with low volatility in biological samples, 104 P. A. Martos and J. Pawliszyn, Time-weighted average
Environ. Sci. Technol., 2001, 35(7), 1425–1431. sampling with solid-phase micro-extraction device:
92 T. Harner, N. J. Farrar, M. Shoeib, K. C. Jones and implications for enhanced personal exposure monitoring
F. A. P. C. Gobas, Characterization of polymer-coated to airborne pollutants, Anal. Chem., 1999, 71, 1513–1520.
glass as a passive air sampler for persistent organic 105 A. Khaled and J. Pawliszyn, Time-weighted average
pollutants, Environ. Sci. Technol., 2003, 37, 2486–2493. sampling of volatile and semi-volatile airborne organic
93 N. J. Farrar, T. Harner, A. J. Sweetman and K. C. Jones, Field compounds by the solid-phase microextraction device, J.
calibration of rapidly equilibrating thin-lm passive air Chromatogr. A, 2000, 892, 455–467.
samplers and their potential application for low-volume 106 P. A. Martos and J. Pawliszyn, Calibration of solid phase
air sampling studies, Environ. Sci. Technol., 2005, 39, 261– microextraction for air analyses based on physical
267. chemical properties of the coating, Anal. Chem., 1997, 69,
94 N. J. Farrar, T. Harner, A. J. Sweetman and K. C. Jones, Field 206–215.
deployment of thin lm passive air samplers for persistent 107 J. Wang, L. Tuduri, M. Mercury, M. Millet, O. Briand and
organic pollutants: A study in the urban atmospheric M. Montury, Sampling atmospheric pesticides with
boundary layer, Environ. Sci. Technol., 2005, 39, 42–48. SPME: Laboratory developments and eld study, Environ.
95 N. J. Farrar, K. Prevedouros, T. Harner, A. J. Sweetman and Pollut., 2009, 157, 365–370.
K. C. Jones, Continental scale passive air sampling of 108 K. Desmet, N. De Coensel, T. Gorecki and P. Sandra,
persistent organic pollutants using rapidly equilibrating Proling the spatial concentration of allethrin and
thin lms (POGs), Environ. Pollut., 2006, 144, 423–433. piperonyl butoxide using passive sorptive sampling and
96 T. St. George, P. Vlahos, T. Harner, P. Helm and B. Wilford, thermal desorption capillary GC-MS, Chemosphere, 2008,
A rapidly equilibrating, thin lm, passive water sampler for 71, 2193–2198.
organic contaminants; characterization and eld testing, 109 X.-M. Li, Q.-H. Zhang, P. Wang, Y.-M. Li and G.-B. Jiang,
Environ. Pollut., 2011, 159, 481–486. Determination of polycyclic aromatic hydrocarbons in air
97 K. E. Kennedy, D. W. Hawker, J. F. Willer, M. E. Bartkow and by stir bar sorptive extraction-thermal desorption-gas
R. W. Truss, A eld comparison of ethylene vinyl acetate chromatography tandem mass spectrometry, Chin. J. Anal.
and low-density polyethylene thin lms for equilibrium Chem., 2011, 39, 1641–1646.
phase passive air sampling of polycyclic aromatic 110 J. Matsiko, H. Li, P. Wang, H. Sun, S. Zheng, D. Wang,
hydrocarbons, Atmos. Environ., 2007, 41, 5778–5787. W. Zhang, Y. Hao, Y. Li, Q. Zhang and G. Jiang, Stir bar
98 S. Genualdi and T. Harner, Rapidly equilibrating sorptive extraction and thermal desorption – gas
micrometer lm sampler for priority pollutants in air, chromatography/mass spectrometry for determining
Environ. Sci. Technol., 2012, 17, 7661–7668. phosphorus ame retardants in air samples, Anal.
99 T. P. Rusina, F. Smedes, J. Klánová, K. Booij and Methods, 2018, 10, 1918–1927.
I. Holoubek, Polymer selection for passive sampling: A 111 J. O. Okeme, A. Saini, C. Yang, J. Zhu, F. Smedes, J. Klánová
comparison of critical properties, Chemosphere, 2007, 68, and M. L. Diamond, Calibration of polydimethylsiloxane
1344–1351. and XAD-Pocket passive air samplers (PAS) for measuring
100 J. F. Narváez Valderrama, K. Baek, F. J. Molina and gas- and particle-phase SVOCs, Atmos. Environ., 2016, 143,
I. J. Allan, Implications of observed PBDE diffusion 202–208.
coefficients in low density polyethylene and silicone 112 J. O. Okeme, C. Yang, A. Abdollahi, S. Dhal, S. A. Harris,
rubber, Environ. Sci.: Processes Impacts, 2016, 18, 87–94. L. M. Jantunen, D. Tsirlin and M. L. Diamond, Passive air
101 T. P. Rusina, F. Smedes, M. Koblizkova and J. Klánová, sampling of ame retardants and plasticizers in
Calibration of silicone rubber passive samplers: Canadian homes using PDMS, XAD-coated PDMS and
Experimental and modeled relations between sampling PUF samplers, Environ. Pollut., 2018, 239, 109–117.
rate and compound properties, Environ. Sci. Technol., 113 J. O. Okeme, L. V. Nguyen, M. Lorenzo, S. Dhal, Y. Pico,
2010, 44, 362–367. V. H. Arrandale and M. L. Diamond, Polydimethylsiloxane
102 M. T. O. Jonker, S. A. van der Heijden, M. Kotte and (silicone rubber) brooch as a personal passive air sampler
F. Smedes, Quantifying the effects of temperature and for semi-volatile organic compounds, Chemosphere, 2018,
salinity on partitioning of hydrophobic organic chemicals 208, 1002–1007.
to silicone rubber passive samplers, Environ. Sci. Technol., 114 K. Vorkamp, L. Odsbjerg, M. Langeland and P. Mayer,
2015, 49, 6791–6799. Utilizing the partitioning properties of silicone for the

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1983
Environmental Science: Processes & Impacts Critical Review

passive sampling of polychlorinated biphenyls (PCBs) in from the rst year of the GAPS study, Environ. Sci.
indoor air, Chemosphere, 2016, 160, 280–286. Technol., 2009, 43, 796–803.
115 J. N. Huckins, M. W. Tubergen and G. K. Manuweera, 128 L. Tuduri, T. Harner and H. Hung, Polyurethane foam
Semipermeable membrane devices containing model (PUF) disks passive air samplers: Wind effect on sampling
lipid: A new approach to monitoring the bioavailability of rates, Environ. Pollut., 2006, 144, 377–383.
lipophilic contaminants and estimating their 129 J. Klánová, P. Eupr, J. Kohoutek and T. Harner, Assessing
bioconcentration potential, Chemosphere, 1990, 20, 533– the inuence of meteorological parameters on the
552. performance of polyurethane foam-based passive air
116 G. Eriksson, S. Jensen, H. Kylin and W. Strachan, The pine samplers, Environ. Sci. Technol., 2008, 42, 550–555.
needle as a monitor of atmospheric pollution, Nature, 1989, 130 C. Moeckel, T. Harner, L. Nizzetto, B. Strandberg,
341, 42–44. A. Lindroth and K. C. Jones, Use of depuration
117 P. Kömp and M. S. McLachlan, Interspecies variability of compounds in passive air samplers: Results from active
the plant/air partitioning of polychlorinated biphenyls, sampling-supported eld deployment, potential uses, and
Environ. Sci. Technol., 1997, 31, 2944–2948. recommendations, Environ. Sci. Technol., 2009, 43, 3227–
118 F. Böhme, K. Welsch-Pausch and M. S. McLachlan, Uptake 3232.
of airborne semivolatile organic compounds in agricultural 131 C. Chaemfa, E. Wild, B. Davison, J. L. Barber and
plants: eld measurements of interspecies variability, K. C. Jones, A study of aerosol entrapment and the
Environ. Sci. Technol., 1999, 33, 1805–1813. inuence of wind speed, chamber design and foam
119 H. Kylin, K. Söderkvist, A. Undeman and R. Franich, density of polyurethane foam passive air samplers used
Seasonal variation of the terpene content, an overlooked persistent organic pollutants, J. Environ. Monit., 2009, 11,
factor in the determination of environmental pollutants 1135–1139.
in pine needles, Bull. Environ. Contam. Toxicol., 2002, 68, 132 J. Thomas, T. M. Holsen and S. Dhaniyala, Computational
155–160. uid dynamic modeling of two passive samplers, Environ.
120 J. F. Müller, D. W. Hawker, D. W. Connell, P. Kömp and Pollut., 2006, 144, 384–392.
M. S. McLachlan, Passive sampling of atmospheric SOCs 133 A. A. May, P. Ashman, J. Huang, S. Dhaniyala and
using tristearin-coated breglass sheets, Atmos. Environ., T. M. Holsen, Evaluation of the polyurethane foam (PUF)
2000, 34, 3525–3534. disk passive air sampler: Computational modeling and
121 M. Stracquadanio, D. Bergamini, E. Massaroki and experimental measurements, Atmos. Environ., 2011, 45,
C. Trombini, Field evaluation of a passive sampler of 4354–4359.
polycyclic aromatic hydrocarbons (PAHs) in an urban 134 S. Hazrati and S. Harrad, Calibration of polyurethane foam
atmosphere (Bologna, Italy), J. Environ. Monit., 2005, 7, (PUF) disk passive air samplers for quantitative
910–915. measurement of polychlorinated biphenyls (PCBs) and
122 D. Gilbert, G. Witt, F. Smedes and P. Mayer, Polymers as polybrominated diphenyl ethers (PBDEs): Factors
reference partitioning phase: polymer calibration for an inuencing sampling rates, Chemosphere, 2007, 67, 448–
analytically operational approach to quantify multimedia 455.
phase partitioning, Anal. Chem., 2016, 88, 5818–5826. 135 C. Chaemfa, J. L. Barber, T. Gocht, T. Harner, I. Holoubek,
123 T. P. Rusina, F. Smedes and J. Klánová, Diffusion J. Klánová and K. C. Jones, Field calibration of polyurethane
coefficients of polychlorinated biphenyls and polycyclic foam (PUF) disk passive air samplers for PCBs and OC
aromatic hydrocarbons in polydimethylsiloxane and low- pesticides, Environ. Pollut., 2008, 156, 1290–1297.
density polyethylene polymers, J. Appl. Polym. Sci., 2010, 136 M. Shoeib, T. Harner, S. C. Lee, D. Lane and J. Zhu, Sorbent-
116, 1803–1810. impregnated polyurethane foam disk for passive air
124 R. Lohmann, Critical review of low-density polyethylene's sampling of volatile uorinated chemicals, Anal. Chem.,
partitioning and diffusion coefficients for trace organic 2008, 80, 675–682.
contaminants and implications for its use as a passive 137 L. Ahrens, T. Harner, M. Shoeib, M. Koblizkova and
sampler, Environ. Sci. Technol., 2012, 46, 606–618. E. J. Reiner, Characterization of two passive air samplers
125 T. F. Bidleman and C. E. Olney, High-volume collection of for per- and polyuroalkyl substances, Environ. Sci.
atmospheric polychlorinated biphenyls, Bull. Environ. Technol., 2013, 47, 14024–14033.
Contam. Toxicol., 1974, 11, 442–450. 138 R. Liu, Y. Lin, R. Liu, F. Hu, T. Ruan and G. Jiang,
126 N. J. Herkert, S. N. Spak, A. Smith, J. K. Schuster, T. Harner, Evaluation of two passive samplers for the analysis of
A. Martinez and K. C. Hornbuckle, Calibration and organophosphate esters in the ambient air, Talanta, 2016,
evaluation of PUF-PAS sampling rates across the Global 147, 69–75.
Atmospheric Passive Sampling (GAPS) network, Environ. 139 S. Genualdi, S. C. Lee, M. Shoeib, A. Gawor, L. Ahrens and
Sci.: Processes Impacts, 2018, 20, 210–219. T. Harner, Global pilot study of legacy and emerging
127 K. Pozo, T. Harner, S. C. Lee, F. Wania, D. C. G. Muir and persistent organic pollutants using sorbent-impregnated
K. C. Jones, Seasonally resolved concentrations of polyurethane foam disk passive air samplers, Environ. Sci.
persistent organic pollutants in the global atmosphere Technol., 2010, 44, 5534–5539.

1984 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

140 A. Dreyer, M. Shoeib, S. Fiedler, J. Barber, T. Harner, 153 C. Rauert, M. Shoeib, J. K. Schuster, A. Eng and T. Harner,
K.-W. Schramm, K. C. Jones and R. Ebinghaus, Field Atmospheric concentrations and trends of poly- and
intercomparison on the determination of volatile and peruoroalkyl substances (PFAS) and volatile methyl
semivolatile polyuorinated compounds in air, Environ. siloxanes (VMS) over 7 years of sampling in the Global
Chem., 2010, 7, 350–358. Atmospheric Passive Sampling (GAPS) network, Environ.
141 V. Langer, A. Dreyer and R. Ebinghaus, Polyuorinated Pollut., 2018, 238, 94–102.
compounds in residential and non-residential indoor air, 154 Y. Li, J. O. Okeme, H. Liu, M. Diamond and J. Zhu, Fixation
Environ. Sci. Technol., 2010, 44, 8075–8081. of XAD-4 power on lter paper using methyl cellulose for
142 L. Vierke, A. Lutz, M. Shoeib, E. J. Reiner, R. Guo, the passive air sampling of semi-volatile organic
W.-U. Palm, R. Ebinghaus and T. Harner, Air compounds in indoor air, Int. J. Environ. Anal. Chem.,
concentrations and particle–gas partitioning of 2016, 96, 1145–1155.
polyuoroalkyl compounds at a wastewater treatment 155 S. Tao, Y. Liu, W. Xu, C. Lang, S. Liu, H. Dou and W. Liu,
plant, Environ. Chem., 2011, 8, 363–371. Calibration of a passive sampler for both gaseous and
143 J. Li, S. Del Vento, J. Schuster, G. Zhang, P. Chakraborty, particulate phase polycyclic aromatic hydrocarbons,
Y. Kobara and K. C. Jones, Peruorinated compounds in Environ. Sci. Technol., 2007, 41, 568–573.
the Asian atmosphere, Environ. Sci. Technol., 2011, 45, 156 M. A.-E. Abdallah and S. Harrad, Modication and
7241–7248. calibration of a passive air sampler for monitoring vapor
144 L. Ahrens, M. Shoeib, T. Harner, S. C. Lee, R. Guo and and particulate phase brominated ame retardants in
E. J. Reiner, Wastewater treatment plant and landlls as indoor air: Application to car interiors, Environ. Sci.
sources of polyuoroalkyl compounds to the atmosphere, Technol., 2010, 44, 3059–3065.
Environ. Sci. Technol., 2011, 45, 8098–8105. 157 S. Tao, J. Cao, W. T. Wang, J. Y. Zhao, W. Wang, Z. H. Wang,
145 M. Shoeib, T. Harner, G. M. Webster and S. C. Lee, Indoor H. Y. Cao and B. Xing, A passive sampler with improved
sources of poly- and peruorinated compounds (PFCS) in performance for collecting gaseous and particulate phase
Vancouver, Canada: Implications for human exposure, polycyclic aromatic hydrocarbons in air, Environ. Sci.
Environ. Sci. Technol., 2011, 45, 7999–8005. Technol., 2009, 43, 4124–4129.
146 E. Goosey and S. Harrad, Peruoroalkyl substances in UK 158 A. Eng, T. Harner and K. Pozo, A prototype passive air
indoor and outdoor air: Spatial and seasonal variation, sampler for measuring dry deposition of polycyclic
and implications for human exposure, Environ. Int., 2012, aromatic hydrocarbons, Environ. Sci. Technol. Lett., 2014,
45, 86–90. 1, 77–81.
147 S. Genualdi, T. Harner, Y. Cheng, M. MacLeod, 159 J. Dotel, P. Gong, X. Wang, B. Pokhrel, C. Wang and
K. M. Hansen, R. van Egmond, M. Shoeib and S. C. Lee, J. Nawab, Determination of dry deposition velocity of
Global distribution of linear and cyclic volatile methyl polycyclic aromatic hydrocarbons under the sub-tropical
siloxanes in air, Environ. Sci. Technol., 2011, 45, 3349–3354. climate and its implication for regional cycling, Environ.
148 Y. Cheng, M. Shoeib, L. Ahrens, T. Harner and J. Ma, Pollut., 2020, 261, 114–143.
Wastewater treatment plants and landlls emit volatile 160 M. J. Climent, C. Coscollà, A. López, R. Barra and R. Urrutia,
methyl siloxanes (VMSs) to the atmosphere: Legacy and current-use pesticides (CUPs) in the
Investigations using a new passive air sampler, Environ. atmosphere of a rural area in central Chile, using passive
Pollut., 2011, 159, 2380–2386. air samplers, Sci. Total Environ., 2019, 662, 646–654.
149 L. Ahrens, T. Harner and M. Shoeib, Temporal variations of 161 B. H. Wilford, T. Harner, J. Zhu, M. Shoeib and K. C. Jones,
cyclic and linear volatile methyl- siloxanes in the Passive sampling survey of polybrominated diphenyl ether
atmosphere using passive samplers and high-volume air ame retardants in indoor and outdoor air in Ottawa,
samplers, Environ. Sci. Technol., 2014, 48, 9374–9381. Canada: Implications for sources and exposure, Environ.
150 J. K. Schuster, R. Gioia, T. Harner, S. C. Lee, K. Breivik and Sci. Technol., 2004, 38, 5312–5318.
K. C. Jones, Assessment of sorbent impregnated PUF disks 162 M. Shoeib, T. Harner, B. H. Wilford, K. C. Jones and J. Zhu,
(SIPs) for long-term sampling of legacy POPs, J. Environ. Peruorinated sulfonamides in indoor and outdoor air and
Monit., 2012, 14, 71–78. indoor dust: Occurrence, partitioning, and human
151 M. Koblizkova, S. Genualdi, S. C. Lee and T. Harner, exposure, Environ. Sci. Technol., 2005, 39, 6599–6606.
Application of sorbent impregnated polyurethane foam 163 X. Wang, A. P. W. Banks, C. He, D. S. Drage, C. L. Gallen,
(SIP) disk passive air samplers for investigating Y. Li, Q. Li, P. K. Thai and J. F. Müller, Polycyclic
organochlorine pesticides and polybrominated diphenyl aromatic hydrocarbons, polychlorinated biphenyls and
ethers at the global scale, Environ. Sci. Technol., 2012, 46, legacy and current pesticides in indoor environment in
391–396. Australia – occurrence, sources and exposure risks, Sci.
152 M. Koblizkova, S. C. Lee and T. Harner, Sorbent Total Environ., 2019, 693, 133588.
impregnated polyurethane foam disk passive air samplers 164 R. E. Dodson, V. Bessonneau, J. O. Udesky, M. Nishioka,
for investigating current–use pesticides at the global M. McCauley and R. A. Rudel, Passive indoor air
scale, Atmos. Pollut. Res., 2012, 3, 456–462. sampling for consumer product chemicals: a eld

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1985
Environmental Science: Processes & Impacts Critical Review

evaluation study, J. Exposure Sci. Environ. Epidemiol., 2019, brominated ame retardants in the Arctic atmosphere,
29, 95–108. Environ. Sci. Technol., 2012, 46, 7669–7676.
165 N. J. Herkert, J. C. Jahnke and K. C. Hornbuckle, Emissions 178 C. L. Davie-Martin, K. J. Hageman, Y.-P. Chin, B. J. Nistor
of tetrachlorobiphenyls (PCBs 47, 51, and 68) from polymer and H. Hung, Concentrations, gas–particle distributions,
resin on kitchen cabinets as a non-Aroclor source to and source indicator analysis of brominated ame
residential air, Environ. Sci. Technol., 2018, 52, 5154–5160. retardants in air at Toolik Lake, Arctic Alaska, Environ.
166 P. Bohlin, K. C. Jones, L. O. Levin, R. Lindahl and Sci.: Processes Impacts, 2016, 18, 1274–1284.
B. Strandberg, Field evaluation of a personal passive air 179 Y. Yu, H. Hung, N. Alexandrou, P. Roach and K. Nordin,
sampler for PAH exposure in workplaces, J. Environ. Multiyear measurements of ame retardants and
Monit., 2010, 12, 1437–1444. organochlorine pesticides in air in Canada's Western sub-
167 B. Strandberg, A. Julander, M. Sjöström, M. Lewné, H. Koca Arctic, Environ. Sci. Technol., 2015, 49, 8623–8630.
Akdeva and C. Bigert, Evaluation of polyurethane foam 180 H. Xiao, S. C. Kang, Q. G. Zhang, W. W. Han, M. Loewen,
passive air sampler (PUF) as a tool for occupational PAH F. Wong, H. Hung, Y. D. Lei and F. Wania, Transport of
measurements, Chemosphere, 2018, 190, 35–42. semivolatile organic compounds to the Tibetan Plateau:
168 M. Sorais, A. Rezaei, J. O. Okeme, M. L. Diamond, Monthly resolved air concentrations at Nam Co, J.
R. Izquierdo, J.-F. Giroux and J. Verreault, A miniature Geophys. Res., 2010, 115, D16310.
bird-borne passive air sampler for monitoring 181 S. Bengtson Nash, S. Wild, D. Hawker, R. Cropp, A. Bignert,
halogenated ame retardants, Sci. Total Environ., 2017, H. Hung, F. Wania, H. Xiao, P. Bohlin-Nizzetto and
599/600, 1903–1911. S. Broomhall, Persistent organic pollutants in the East
169 Y. Zhang, S. Deng, Y. Liu, G. Shen, X. Li, J. Cao, X. Wang, Antarctic atmosphere: Inter-annual observations from
B. Reid and S. Tao, A passive air sampler for 2010–2015 using high ow-through passive sampling,
characterizing the vertical concentration prole of Environ. Sci. Technol., 2017, 51, 13929–13937.
gaseous phase polycyclic aromatic hydrocarbons in near 182 T. S. Geoghegan, K. J. Hageman and A. J. Hewitt, Testing
soil surface air, Environ. Pollut., 2011, 159, 694–699. ow-through air samplers for use in near-eld vapour
170 B. Yang, B. Han, N. Xue, L. Zhou and F. Li, Air–soil dri studies by measuring pyrimethanil in air aer
exchange of organochlorine pesticides in a sealed spraying, Environ. Sci.: Processes Impacts, 2014, 16, 422–432.
chamber, J. Environ. Sci., 2015, 27, 241–250. 183 F. Esen, Development of a passive sampling device using
171 C. F. Wang, X. P. Wang, J. Ren, P. Gong and T. D. Yao, Using polyurethane foam (PUF) to measure polychlorinated
a passive air sampler to monitor air–soil exchange of biphenyls (PCBs) and organochlorine pesticides (OCPs)
organochlorine pesticides in the pasture of the central near landlls, Environ. Forensics, 2013, 14, 1–8.
Tibetan plateau, Sci. Total Environ., 2017, 580, 958–965. 184 F. Esen, Y. M. Evci and Y. Taşdemir, Evaluation and
172 S. Tao, Y. N. Liu, C. Lang, W. T. Wang, H. S. Yuan, application of a passive air sampler for polycylic aromatic
D. Y. Zhang, W. X. Qiu, J. M. Liu, Z. G. Liu, S. Z. Liu, hydrocarbons (PAHs), J. Environ. Sci. Health, Part A: Toxic/
R. Yi, M. Ji and X. X. Liu, A directional passive air Hazard. Subst. Environ. Eng., 2017, 52, 1022–1029.
sampler for monitoring polycyclic aromatic hydrocarbons 185 S. Oono, K. H. Harada, M. A. M. Mahmoud, K. Inoue and
(PAHs) in air mass, Environ. Pollut., 2008, 156, 435–441. A. Koizumi, Current levels of airborne polyuorinated
173 H. Xiao, H. Hung, T. Harner, Y. D. Lei, G. W. Johnston and telomers in Japan, Chemosphere, 2008, 73, 932–937.
F. Wania, A ow-through sampler for semi-volatile organic 186 J.-O. Levin, C.-A. Nilsson and K. Andersson, Sampling of
compounds in air, Environ. Sci. Technol., 2007, 41, 250–256. organic substances in workroom air using amberlite XAD-
174 H. Xiao, H. Hung, T. Harner, Y. D. Lei and F. Wania, Field 2 resin, Chemosphere, 1977, 6, 595–598.
testing a ow-through sampler for semi-volatile organic 187 J. E. Woodrow, J. N. Seiber, D. G. Crosby, K. W. Moilanen,
compounds in air, Environ. Sci. Technol., 2008, 42, 2970– C. J. Soderquist and C. Mourer, Airborne and surface
2975. residues of parathion and its conversion products in
175 H. Xiao, H. Hung, Y. D. Lei and F. Wania, Validation of a treated plum orchard environment, Arch. Environ.
a ow-through sampler for pesticides and Contam. Toxicol., 1977, 6, 175–191.
polybrominated diphenyl ethers in air, Atmos. Environ., 188 J. C. Chuang, S. W. Hannan and N. K. Wilson, Field
2009, 43, 2401–2409. comparison of polyurethane foam and XAD-2 resin for air
176 H. Xiao, L. Shen, Y. Su, E. Barresi, M. DeJong, H. Hung, sampling for polynuclear aromatic hydrocarbons, Environ.
Y. D. Lei, F. Wania, E. J. Reiner, E. Sverko and S.-C. Kang, Sci. Technol., 1987, 21, 798–804.
Atmospheric concentrations of halogenated ame 189 R. Dobson, A. Scheyer, A. Laure Rizet, P. Mirabel and
retardants at two remote locations: The Canadian High M. Millet, Comparison of the efficiencies of different
Arctic and the Tibetan Plateau, Environ. Pollut., 2012, 161, types of adsorbents at trapping currently used pesticides
154–161. in the gaseous phase using the technique of high-volume
177 H. Xiao, H. Hung, F. Wania, R. Lao, E. Sablic, E. Sverko, sampling, Anal. Bioanal. Chem., 2006, 386, 1781–1789.
Y. D. Lei, P. Fellin and E. Barresi, Field evaluation of 190 A. Ramı́rez Restrepo, S. J. Hayward, J. M. Armitage and
a ow-through sampler for measuring pesticides and F. Wania, Evaluating the PAS-SIM model using a passive

1986 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

air sampler calibration study for pesticides, Environ. Sci.: phases collected in polyurethane foam used in passive air
Processes Impacts, 2015, 17, 1228–1237. samplers, Water, Air, Soil Pollut., 2018, 229–282.
191 X. Zhang, T. N. Brown, A. Ansari, B. Yeun, K. Kitaoka, 204 N. J. Herkert, A. Martinez and K. C. Hornbuckle, A model
A. Kondo, Y. D. Lei and F. Wania, Effect of wind on the using local weather data to determine the effective
chemical uptake kinetics of a passive air sampler, sampling volume for PCB congeners collected on passive
Environ. Sci. Technol., 2013, 47, 7868–7875. air samplers, Environ. Sci. Technol., 2016, 50, 6690–6697.
192 X. Zhang, C. Wong, Y. D. Lei and F. Wania, Inuence of 205 X. Zhang, M. Tsurukawa, T. Nakano, Y. D. Lei and F. Wania,
sampler conguration on the uptake kinetics of a passive Sampling medium side resistance to uptake of semi-
air sampler, Environ. Sci. Technol., 2012, 46, 397–403. volatile organic compounds in passive air samplers,
193 X. Zhang, M. Hoang, Y. D. Lei and F. Wania, Exploring the Environ. Sci. Technol., 2011, 45, 10509–10515.
role of the sampler housing in limiting uptake of 206 J. O. Okeme, E. M. Webster, J. M. Parnis and
semivolatile organic compounds in passive air samplers, M. L. Diamond, Approaches for estimating PUF-air
Environ. Sci.: Processes Impacts, 2015, 17, 2006–2012. partitions coefficient for semi-volatile organic
194 W. J. Liu, D. Chen, X. D. Liu, X. Y. Zheng, W. Yang, compounds: A critical comparison, Chemosphere, 2017,
J. N. Westgate and F. Wania, Transport of semivolatile 168, 199–204.
organic compounds to the Tibetan Plateau: Spatial and 207 X. M. Zhang and F. Wania, Modeling the uptake of semi-
temporal variation in air concentrations in mountainous volatile organic compounds by passive air samplers:
Western Sichuan, China, Environ. Sci. Technol., 2010, 44, Importance of mass transfer processes within the porous
1559–1565. sampling media, Environ. Sci. Technol., 2012, 46, 9563–
195 C. Shunthirasingham, R. Barra, G. Mendoza, M. Montory, 9570.
C. E. Oyiliagu, Y. D. Lei and F. Wania, Spatial variability 208 K. Pozo, T. Harner, F. Wania, D. C. G. Muir, K. C. Jones and
of semivolatile organic compounds in the Chilean L. A. Barrie, Toward a global network for persistent organic
atmosphere, Atmos. Environ., 2011, 45, 303–309. pollutants in air: Results from the GAPS study, Environ. Sci.
196 P. Gong, X. Wang, X. Liu and F. Wania, Field calibration of Technol., 2006, 40, 4867–4873.
XAD-based passive air sampler on the Tibetan Plateau: 209 C. Persoon and K. C. Hornbuckle, Calculation of passive
wind inuence and conguration improvement, Environ. sampling rates from both native PCBs and depuration
Sci. Technol., 2017, 51, 5642–5649. compounds in indoor and outdoor environments,
197 A. Gawor, C. Shunthirasingham, S. J. Hayward, Y. D. Lei, Chemosphere, 2009, 74, 917–923.
T. Gouin, B. Mmereki, W. Masamba, C. Ruepert, 210 C. Shunthirasingham, B. T. Mmereki, W. Masamba,
L. E. Castillo, S. C. Lee, M. Shoeib, T. Harner and F. Wania, C. E. Oyiliagu, Y. D. Lei and F. Wania, Fate of pesticides
Neutral peruoroalkyl substances in the global atmosphere, in the arid subtropics, Botswana, Southern Africa,
Environ. Sci.: Processes Impacts, 2014, 16, 404–413. Environ. Sci. Technol., 2010, 44, 8082–8088.
198 J. Wagner and D. Leith, Passive aerosol sampler. Part I: 211 L. Sprunger, A. Proctor, W. E. Acree and M. H. Abraham,
Principle of operation, Aerosol Sci. Technol., 2001, 34, 186– Characterization of the sorption of gaseous and organic
192. solutes onto polydimethyl siloxane solid-phase
199 K. Arnoldsson, R. Magnusson, L. Hägglund, C. Lejon and microextraction surfaces using the Abraham model, J.
H. Wingfors, Initial evaluation of an axial passive sampler Chromatogr. A, 2007, 1175, 162–173.
for PAHs and OPAHs using substrates with and without 212 J. O. Okeme, J. M. Parnis, J. Poole, M. L. Diamond and
gas sampling capacity and varying diffusion distances, L. M. Jantunen, Poly-dimethylsiloxane–air partition ratios
Atmos. Pollut. Res., 2015, 6, 673–681. for semi-volatile organic compounds by GC-based
200 R. Magnusson, K. Arnoldsson, C. Lejon, L. Hägglund and measurement and COSMO-RS estimation: rapid
H. Wingfors, Field evaluation and calibration of a small measurements and accurate modelling, Chemosphere,
axial passive air sampler for gaseous and particle bound 2016, 156, 204–211.
polycyclic aromatic hydrocarbons (PAHs) and oxygenated 213 P. C. Tromp, H. Beeltje, J. O. Okeme, R. Vermeulen,
PAHs, Environ. Pollut., 2016, 216, 235–244. A. Pronk and M. L. Diamond, Calibration of
201 A. Sonnette, M. Millet, R. Ocampo, L. Alleman and polydimethylsiloxane and polyurethane foam passive air
P. Coddeville, Tenax-TA spiking approach of thermal samplers for measuring semi volatile organic compounds
desorption coupled to GC–MSMS for the quantication of using a novel exposure chamber design, Chemosphere,
PAHs in indoor air and dust, Polycyclic Aromat. Compd., 2019, 227, 435–443.
2017, 37, 170–177. 214 D. Zhao, J. C. Little and S. S. Cox, Characterizing
202 T. Primbs, S. Genualdi and S. M. Simonich, Solvent polyurethane foam as a sink for or source of volatile
selection for pressurized liquid extraction of polymeric organic compounds in indoor air, J. Environ. Eng., 2004,
sorbents used in air sampling, Environ. Toxicol. Chem., 130, 984–989.
2008, 27, 1267–1272. 215 I. Kamprad and K.-U. Goss, Systematic investigation of the
203 G. E. Arias-Loaiza, E. Beristain-Montiel, S. Gómez-Arroyo sorption properties of polyurethane foams for organic
and O. A. Amador-Muñoz, A thermo-cavitation method to vapors, Anal. Chem., 2007, 79, 4222–4227.
determine organochlorine pesticides in gas and particle

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1987
Environmental Science: Processes & Impacts Critical Review

216 T. F. Bidleman, O. Nygren and M. Tysklind, Field estimates 230 A. Stenzel, K.-U. Goss and S. Endo, Determination of
of polyurethane foam air partition coefficients for polyparameter linear free energy relationship (pp-LFER)
hexachlorobenzene, alpha-hexachlorocyclohexane and substance descriptors for established and alternative
bromoanisoles, Chemosphere, 2016, 159, 126–131. ame retardants, Environ. Sci. Technol., 2013, 47(3), 1399–
217 J. M. Parnis, D. Mackay and T. Harner, Modelling PUF 1406.
passive air samplers: Temperature dependence of 231 A. Stenzel, K.-U. Goss and S. Endo, Experimental
polyurethane foam–air partitioning of PAHs and their O-, determination of polyparameter linear free energy
N-, and S-derivatives computed by COSMO-RS, Atmos. relationship (pp-LFER) substance descriptors for
Pollut. Res., 2016, 7, 155–161. pesticides and other contaminants: New measurements
218 J. M. Parnis, A. Eng, D. Mackay and T. Harner, and recommendations, Environ. Sci. Technol., 2013,
Characterizing PUF disk passive air samplers for alkyl- 47(24), 14204–14214.
substituted PAHs: measured and modelled PUF-AIR 232 N. Ulrich, S. Endo, T. N. Brown, N. Watanabe, G. Bronner,
partition coefficients with COSMO-RS, Chemosphere, 2016, M. H. Abraham and K.-U. Goss, UFZ-LSER database v. 3.2.1,
145, 360–364. Helmholtz Centre for Environmental Research-UFZ,
219 A. P. Francisco, T. Harner and A. Eng, Measurement of Leipzig, Germany, 2017, accessed on 27.01.2020, http://
polyurethane foam–air partition coefficients for www.ufz.de/lserd.
semivolatile organic compounds as a function of 233 K.-U. Goss, Predicting equilibrium sorption of neutral
temperature: application to passive air sampler organic chemicals into various polymeric sorbents with
monitoring, Chemosphere, 2017, 174, 638–642. COSMO-RS, Anal. Chem., 2011, 83, 5304–5308.
220 A. Saini, J. Clarke and T. Harner, Direct measurements of 234 C. Loschen and A. Klamt, Prediction of solubilities and
polyurethane foam (PUF)–air partitioning coefficients for partition coefficients in polymers using COSMO-RS, Ind.
chemicals of emerging concern capable of equilibrating Eng. Chem. Res., 2014, 53, 11478–11487.
in PUF disk samplers, Chemosphere, 2019, 234, 925–930. 235 Y. Chen, C. Wang and F. Wania, Prediction of partitioning
221 J. F. Pankow, W. Luo, L. M. Isabelle, K. M. Hart and coefficients between XAD-2 and air for organic compounds
D. F. Hagen, Gas-solid retention volumes of organic using COSMOtherm, CGCS internship report, University of
compounds on styrene-divinylbenzene and Toronto, 2016, p. 27.
ethylvinylbenzene–divinylvenzene copolymer sorbent 236 M. Mari, M. Schuhmacher, J. Feliubadalo and
beads, J. Chromatogr. A, 1996, 732, 317–326. J. L. Domingo, Air concentrations of PCDD/Fs, PCBs and
222 L. Shen, Y. D. Lei and F. Wania, Sorption of chlorobenzene PCNs using active and passive air samplers, Chemosphere,
vapors on styrene-divinylbenzene polymer, J. Chem. Eng. 2008, 70, 1637–1643.
Data, 2002, 47, 944–949. 237 A. Carratalá, R. Moreno-González and V. M. León,
223 S. J. Hayward, Y. D. Lei and F. Wania, Sorption of a diverse Occurrence and seasonal distribution of polycyclic
set of organic chemical vapors onto XAD-2 resin: aromatic hydrocarbons and legacy and current-use
Measurement, prediction and implications for air pesticides in air from a Mediterranean coastal lagoon
sampling, Atmos. Environ., 2011, 45, 296–302. (Mar Menor, SE Spain), Chemosphere, 2017, 167, 382–395.
224 K.-U. Goss and R. P. Schwarzenbach, Linear free energy 238 J. He and R. Balasubramanian, A comparative evaluation of
relationships used to evaluate equilibrium partitioning of passive and active samplers for measurements of gaseous
organic compounds, Environ. Sci. Technol., 2001, 35, 1–9. semi-volatile organic compounds in the tropical
225 S. Endo and K.-U. Goss, Applications of polyparameter atmosphere, Atmos. Environ., 2010, 44, 884–891.
linear free energy relationships in environmental 239 S.-K. Kim, M. Shoeib, K.-K. Kim and J.-E. Park, Indoor and
chemistry, Environ. Sci. Technol., 2014, 48, 12477–12491. outdoor poly- and peruoroalkyl substances (PFASs) in
226 T. H. Uber, T. Hüffer, S. Planitz and T. C. Schmidt, Korea determined by passive air sampler, Environ. Pollut.,
Characterization of sorption properties of high-density 2012, 162, 144–150.
polyethylene using the poly-parameter linearfree-energy 240 A. Saini, J. O. Okeme, E. Goosey and M. L. Diamond,
relationships, Environ. Pollut., 2019, 248, 312–319. Calibration of two passive air samplers for monitoring
227 Y. Choi, Y.-M. Cho and R. G. Luthy, Polyethylene–water phthalates and brominated ame-retardants in indoor
partitioning coefficients for parent- and alkylated- air, Chemosphere, 2015, 137, 166–173.
polycyclic aromatic hydrocarbons and polychlorinated 241 D. T. Waite, N. R. Droneck, L. Tuduri, J. F. Sproul,
biphenyls, Environ. Sci. Technol., 2013, 47(13), 6943–6950. D. F. Chau and D. V. Quiring, Comparison of active versus
228 M. Schneider and K.-U. Goss, Systematic investigation of passive atmospheric samplers for some current-use
the sorption properties of Tenax TA, Chromosorb 106, pesticides, Bull. Environ. Contam. Toxicol., 2005, 74, 1011–
Porapak N, and Carbopak F, Anal. Chem., 2009, 81, 3017– 1018.
3021. 242 C. Chaemfa, J. L. Barber, C. Moeckel, T. Gocht, T. Harner,
229 H. C. Tülp, K.-U. Goss, R. P. Schwarzenbach and K. Fenner, I. Holoubek, J. Klánová and K. C. Jones, Field calibration
Experimental determination of LSER parameters for a set of of polyurethane foam disk passive air samplers for
76 diverse pesticides and pharmaceuticals, Environ. Sci. PBDEs, J. Environ. Monit., 2009, 11, 1859–1865.
Technol., 2008, 42(6), 2034–2040.

1988 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

243 C. Chaemfa, J. L. Barber, K.-S. Kim, T. Harner and 255 G. Liu, G. Zhang, J. Lia, X. Li, X. Peng and S. Qi, Spatial
K. C. Jones, Further studies on the uptake of persistent distribution and seasonal variations of polycyclic
organic pollutants (POPs) by polyurethane foam disk aromatic hydrocarbons (PAHs) using semi-permeable
passive air samplers, Atmos. Environ., 2009, 43, 3843–3849. membrane devices (SPMD) and pine needles in the Pearl
244 J. L. Armstrong, M. G. Yost and R. A. Fenske, Development River Delta, South China, Atmos. Environ., 2006, 40, 3134–
of a passive air sampler to measure airborne 3143.
organophosphorus pesticides and oxygen analogs in an 256 S. Harrad and M. A.-E. Abdallah, Calibration of two passive
agricultural community, Chemosphere, 2014, 111, 135–143. air sampler congurations for monitoring concentrations
245 P. Bohlin, O. Audy, L. Škrdlı́ková, P. Kukučka, P. Přibylová, of hexabromocyclododecanes in indoor air, J. Environ.
R. Prokeš, Š. Vojta and J. Klańová, Outdoor passive air Monit., 2008, 10, 527–531.
monitoring of semi volatile organic compounds (SVOCs): 257 P. Bohlin, K. C. Jones and B. Strandberg, Field evaluation of
a critical evaluation of performance and limitations of polyurethane foam passive air samplers to assess airborne
polyurethane foam (PUF) disks, Environ. Sci.: Processes PAHs in occupational environments, Environ. Sci. Technol.,
Impacts, 2014, 16, 433–444. 2010, 44, 749–754.
246 J. Heo and G. Lee, Field-measured uptake rates of PCDDs/ 258 S.-K. Kim and J.-E. Park, Comparison of two different
Fs and dl-PCBs using PUF-disk passive air samplers in passive air samplers (PUF-PAS versus SIP-PAS) to
Gyeonggi-do, South Korea, Sci. Total Environ., 2014, 491/ determine time-integrated average air concentration of
492, 42–50. volatile hydrophobic organic pollutants, Ocean Sci. J.,
247 T. Harner, K. Su, S. Genualdi, J. Karpowicz, L. Ahrens, 2014, 49, 137–150.
C. Mihele, J. Schuster, J.-P. Charland and J. Narayan, 259 P. Bohlin, O. Audy, L. Škrdlı́ková, P. Kukučka, Š. Vojta,
Calibration and application of PUF disk passive air P. Přibylová, R. Prokeš, P. Čupr and J. Klánová, Evaluation
samplers for tracking polycyclic aromatic compounds and guidelines for using polyurethane foam (PUF) passive
(PACs), Atmos. Environ., 2013, 75, 123–128. air samplers in double-dome chambers to assess semi-
248 Y. M. Evci, F. Esen and Y. Taşdemir, Monitoring of long- volatile organic compounds (SVOCs) in non-industrial
term outdoor concentrations of PAHs with passive air indoor environments, Environ. Sci.: Processes Impacts,
samplers and comparison with meteorological data, Arch. 2014, 16, 2617–2626.
Environ. Contam. Toxicol., 2016, 71, 246–256. 260 J. L. Armstrong, R. A. Fenske, M. G. Yost, M. Tchong-French
249 A. E. Sakin, F. Esen and Y. Taşdemir, Effects of sampling and J. Yu, Comparison of polyurethane foam and XAD-2
interval on the passive air sampling of atmospheric PCBs sampling matrices to measure airborne
levels, J. Environ. Sci. Health, Part A: Toxic/Hazard. Subst. organophosphorus pesticides and their oxygen analogs in
Environ. Eng., 2017, 52, 673–679. an agricultural community, Chemosphere, 2013, 92, 451–
250 A. Abdollahi, A. Eng, L. M. Jantunen, L. Ahrens, M. Shoeib, 457.
J. M. Parnis and T. Harner, Characterization of 261 N. Jariyasopit, Y. Liu, J. Liggio and T. Harner, Stability of
polyurethane foam (PUF) and sorbent impregnated PUF polycyclic aromatic compounds in polyurethane foam-
(SIP) disk passive air samplers for measuring type passive air samplers upon O3 exposure, Atmos.
organophosphate ame retardants, Chemosphere, 2017, Environ., 2015, 120, 200–204.
167, 212–219. 262 L. Melymuk, P. Bohlin-Nizzetto, R. Prokeš, P. Kukučka,
251 H. Jiang, G. Zhong, J. Wang, H. Jiang, C. Tian, J. Li, S. Zhao, P. Přibylová, Š. Vojta, J. Kohoutek, G. Lammel and
Z. Yu, L. Morawska and G. Zhang, Using polyurethane J. Klánová, Uncertainties in monitoring of SVOCs in air
foam-based passive air sampling technique to monitor caused by within-sampler degradation during active and
monosaccharides at a regional scale, Environ. Sci. passive air sampling, Atmos. Environ., 2017, 167, 553–565.
Technol., 2018, 52, 12546–12555. 263 T. Harner, Template for calculating PUF and SIP disk sample
252 Y. Wang, X. Zhu, Y. Gao, H. Bai, P. Wang, J. Chen, H. Yuan, air volumes, Version 2, April 6, 2020, https://
L. Wang, X. Li and W. Wang, Monitoring gas- and www.researchgate.net/prole/Tom_Harner/publications.
particulate-phase short-chain polychlorinated paraffins in 264 S. J. Hayward, T. Gouin and F. Wania, Levels and seasonal
the urban air of Dalian by a self-developed passive variability of pesticides in the rural atmosphere of Southern
sampler, J. Environ. Sci., 2019, 80, 287–295. Ontario, J. Agric. Food Chem., 2010, 58, 1077–1084.
253 K. M. Ellickson, C. M. McMahon, C. Herbrandson, 265 K. Prevedouros, E. Brorström-Lundén, C. J. Halsall,
M. J. Krause, C. M. Schmitt, C. J. Lippert and G. C. Pratt, K. C. Jones, R. G. M. Lee and A. J. Sweetman, Seasonal
Analysis of polycyclic aromatic hydrocarbons (PAHs) in and long-term trends in atmospheric PAH concentrations:
air using passive sampling calibrated with active evidence and implications, Environ. Pollut., 2004, 128, 17–
measurements, Environ. Pollut., 2017, 231, 487–496. 27.
254 M. E. Bartkow, J. N. Huckins and J. F. Müller, Field-based 266 F. Wania, J.-E. Haugen, Y. D. Lei and D. Mackay,
evaluation of semipermeable membrane devices (SPMDs) Temperature dependence of atmospheric concentrations
as passive air samplers of polyaromatic hydrocarbons of semivolatile organic compounds, Environ. Sci. Technol.,
(PAHs), Atmos. Environ., 2004, 38, 5983–5990. 1998, 32, 1013–1021.

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1989
Environmental Science: Processes & Impacts Critical Review

267 I. S. Krogseth, A. Kierkegaard, M. S. McLachlan, K. Breivik, 279 X. H. Zhu, G. H. Ding, W. Levy, G. Jakobi and
K. M. Hansen and M. Schlabach, Occurrence and K.-W. Schramm, Relationship of air sampling rates of
seasonality of cyclic volatile methyl siloxanes in Arctic air, semipermeable membrane devices with the properties of
Environ. Sci. Technol., 2013, 47, 502–509. organochlorine pesticides, J. Environ. Sci., 2011,
268 K. Kennedy, D. W Hawker, M. E Bartkow, S. Carter, 23(supplement), S40–S44.
Y. Ishikawa and J. F Müller, The potential effect of 280 G. L. Daly, Y. D. Lei, L. E. Castillo, D. C. G. Muir and
differential ambient and deployment chamber F. Wania, Polycyclic aromatic hydrocarbons in Costa
temperatures on PRC derived sampling rates with Rican air and soil: A tropical/temperate comparison,
polyurethane foam (PUF) passive air samplers, Environ. Atmos. Environ., 2007, 41, 7339–7350.
Pollut., 2010, 158, 142–147. 281 M. Z. Markovic, S. Prokop, R. M. Staebler, J. Liggio and
269 N. Domı́nguez-Morueco, M. Schuhmacher, J. Sierra, T. Harner, Evaluation of the particle inltration efficiency
M. Nadal and J. L. Domingo, Assessment of PAH loss in of three passive samplers and the PS-1 active air sampler,
passive air samplers by the effect of temperature, Atmos. Atmos. Environ., 2015, 112, 289–293.
Pollut. Res., 2016, 7, 142–146. 282 Q. Li, K. Yang, J. Li, X. Zeng, Z. Yu and G. Zhang, An
270 K. Booij and F. Smedes, An improved method for assessment of polyurethane foam passive samplers for
estimating in situ sampling rates of nonpolar passive atmospheric metals compared with active samplers,
samplers, Environ. Sci. Technol., 2010, 44, 6789–6794. Environ. Pollut., 2018, 236, 498–504.
271 C. A. McDonough, D. G. Franks, M. E. Hahn and 283 E. O. Gaga, T. Harner, E. Dabek-Zlotorzynska, V. Celo,
R. Lohmann, Aryl hydrocarbon receptor-mediated activity G. Evans, C.-H. Jeong, S. Halappanavar, N. Jariyasopit and
of gas-phase ambient air derived from passive sampling Y. Su, Polyurethane Foam (PUF) disk samplers for
and an in vitro bioassay, Environ. Toxicol. Chem., 2019, 38, measuring trace metals in ambient air, Environ. Sci.
748–759. Technol. Lett., 2019, 6, 545–550.
272 N. T. Petrich, S. N. Spak, G. R. Carmichael, D. Hu, 284 L. Melymuk, P. Bohlin, O. Sáńka, K. Pozo and J. Klańová,
A. Martinez and K. C. Hornbuckle, Simulating and Current challenges in air sampling of semivolatile organic
explaining passive air sampling rates for semivolatile contaminants: Sampling artifacts and their inuence on
compounds on polyurethane foam passive samplers, data comparability, Environ. Sci. Technol., 2014, 48,
Environ. Sci. Technol., 2013, 47, 8591–8598. 14077–14091.
273 A. Pennequin-Cardinal, H. Plaisance, N. Locoge, 285 L. Vilavert, M. Nadal, M. Schuhmacher and J. L. Domingo,
O. Ramalho, S. Kirchner and J.-C. Galloo, Performances Seasonal surveillance of airborne PCDD/Fs, PCBs and PCNs
of the Radiello® diffusive sampler for BTEX using passive samplers to assess human health risks, Sci.
measurements: inuence of environmental conditions Total Environ., 2014, 466/467, 733–740.
and determination of modelled sampling rates, Atmos. 286 L. Gao, Q. Zhang, L. Liu, C. Li and Y. Wang, Spatial and
Environ., 2005, 39, 2535–2544. seasonal distributions of polychlorinated dibenzo-p-
274 H. Skov, B. T. Sørensen, M. S. Landis, M. S. Johnson, dioxins and dibenzofurans and polychlorinated biphenyls
P. Sacco, M. E. Goodsite, C. Lohse and K. S. Christiansen, around a municipal solid waste incinerator, determined
Performance of a new diffusive sampler for Hg0 using polyurethane foam passive air samplers,
determination in the troposphere, Environ. Chem., 2007, Chemosphere, 2014, 114, 317–326.
4, 75–80. 287 M. Shoeib, J. Schuster, C. Rauert, K. Su, S.-A. Smyth and
275 H. Plaisance, The effect of the wind velocity on the uptake T. Harner, Emission of poly and peruoroalkyl
rates of various diffusive samplers, Int. J. Environ. Anal. substances, UV-lters and siloxanes to air from
Chem., 2011, 91, 1341–1352. wastewater treatment plants, Environ. Pollut., 2016, 218,
276 D. S. McLagan, C. P. J. Mitchell, H. Huang, B. Abdul 595–604.
Hussain, Y. D. Lei and F. Wania, The effects of 288 F. Wong, M. Robson, L. Melymuk, C. Shunthirasingham,
meteorological parameters and diffusive barrier reuse on N. Alexandrou, M. Shoeib, E. Luk, P. Helm,
the sampling rate of a passive air sampler for gaseous M. L. Diamond and H. Hung, Urban sources of synthetic
mercury, Atmos. Meas. Tech., 2017, 10, 3651–3660. musk compounds to the environment, Environ. Sci.:
277 L. M. van Mourik, X. Wang, C. Paxman, P. E. G. Leonards, Processes Impacts, 2019, 21, 74–88.
F. Wania, J. de Boer and J. F. Müller, Spatial variation of 289 J. Klánová, J. Kohoutek, L. Hamplova, P. Urbanova and
short- and medium-chain chlorinated paraffins in I. Holoubek, Passive air sampler as a tool for long-term
ambient air across Australia, Environ. Pollut., 2020, 261, air pollution monitoring: Part 1. Performance assessment
114141. for seasonal and spatial variations, Environ. Pollut., 2006,
278 X. H. Zhu, G. H. Ding, W. Levy, G. Jakobi, I. Offenthaler, 144, 393–405.
W. Moche, P. Weiss and K.-W. Schramm, QSPR study 290 N. Domı́nguez-Morueco, S. Augusto, L. Trabalón,
about sampling rates of semipermeable membrane E. Pocurull, F. Borrull, M. Schuhmacher, J. L. Domingo
devices for monitoring of organochlorine pesticides in and M. Nadal, Monitoring PAHs in the petrochemical
Alps air, Chin. Sci. Bull., 2011, 56, 1884–1889. area of Tarragona County, Spain: comparing passive air

1990 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

samplers with lichen transplants, Environ. Sci. Pollut. Res., in the south-west coast of India, Environ. Sci. Pollut. Res.,
2017, 24, 11890–11900. 2019, 26, 22772–22782.
291 S.-Y. Baek, S.-D. Choi, H. Park, J.-H. Kang and Y.-S. Chang, 302 N. A. O. Morin, P. L. Andersson, S. E. Hale and H. P. H. Arp,
Spatial and seasonal distribution of polychlorinated The presence and partitioning behavior of ame retardants
biphenyls (PCBs) in the vicinity of an iron and steel in waste, leachate, and air particles from Norwegian waste-
making plant, Environ. Sci. Technol., 2010, 44, 3035–3040. handling facilities, J. Environ. Sci., 2017, 62, 115–132.
292 J. K. Schuster, T. Harner, K. Su, C. Mihele and A. Eng, First 303 H. P. Arp, N. A. Morin, P. L. Andersson, S. E. Hale, F. Wania,
results from the oil sands passive air monitoring network K. Breivik and G. D. Breedveld, The presence, emission and
for polycyclic aromatic compounds, Environ. Sci. Technol., partitioning behavior of polychlorinated biphenyls in
2015, 49, 2991–2998. waste, leachate and aerosols from Norwegian waste-
293 J. K. Schuster, T. Harner, K. Su, A. Eng, A. Wnorowski and handling facilities, Sci. Total Environ., 2020, 136824.
J.-P. Charland, Temporal and spatial trends of polycyclic 304 H. Q. Anh, K. Tomioka, N. M. Tue, L. H. Tuyen, N. K. Chi,
aromatic compounds in air across the Athabasca Oil T. B. Minh, P. H. Viet and S. Takahashi, A preliminary
Sands Region reect inputs from open pit mining and investigation of 942 organic micro-pollutants in the
forest res, Environ. Sci. Technol. Lett., 2019, 6, 178–183. atmosphere in waste processing and urban areas,
294 N. Jariyasopit, Y. Zhang, J. W. Martin and T. Harner, northern Vietnam: Levels, potential sources, and risk
Comparison of polycyclic aromatic compounds in air assessment, Ecotoxicol. Environ. Saf., 2019, 167, 354–364.
measured by conventional passive air samplers and 305 H. Q. Anh, I. Watanabe, T. N. Tue, L. H. Tuyen, P. H. Viet,
passive dry deposition samplers and contributions from N. K. Chi, T. B. Minh and S. Takahashi, Polyurethane
petcoke and oil sands ore, Atmos. Chem. Phys., 2018, 18, foam-based passive air sampling for simultaneous
9161–9171. determination of POP- and PAH-related compounds: A
295 J. Hu, M. Zheng, W. Liu, Z. Nie, C. Li, G. Liu and K. Xiao, case study in informal waste processing and urban areas,
Characterization of polychlorinated dibenzo-p-dioxins and northern Vietnam, Chemosphere, 2020, 247, 125991.
dibenzofurans, dioxin-like polychlorinated biphenyls, and 306 S. Du, S. J. Wall, L. A. Cacia and L. A. Rodenburg, Passive air
polychlorinated naphthalenes in the environment sampling for polychlorinated biphenyls in the Philadelphia
surrounding secondary copper and aluminum metropolitan area, Environ. Sci. Technol., 2009, 43, 1287–
metallurgical facilities in China, Environ. Pollut., 2014, 1292.
193, 6–12. 307 C. Persoon, T. M. Peters, N. Kumar and K. C. Hornbuckle,
296 T. H. Nøst, A. K. Halse, S. Randall, A. R. Borgen, Spatial distribution of airborne polychlorinated biphenyls
M. Schlabach, A. Paul, A. Rahman and K. Breivik, High in Cleveland, Ohio and Chicago, Illinois, Environ. Sci.
concentrations of organic contaminants in air from ship Technol., 2010, 44, 2797–2802.
breaking activities in Chittagong, Bangladesh, Environ. 308 A. A. Peverly, Y. Ma, M. Venier, Z. Rodenburg, S. N. Spak,
Sci. Technol., 2015, 49, 11372–11380. K. C. Hornbuckle and R. A. Hites, Variations of ame
297 H. Sun, Y. Li, P. Wang, S. Zheng, J. Matsiko, D. Wang, retardant, polycyclic aromatic hydrocarbon, and pesticide
W. Zhang, Y. Hao, Q. Zhang and G. Jiang, Atmospheric concentrations in Chicago's atmosphere measured using
levels and distribution of Dechlorane Plus in an E-waste passive sampling, Environ. Sci. Technol., 2015, 49, 5371–
dismantling region of East China, Sci. China: Chem., 5379.
2017, 60, 305–310. 309 Y. Hao, Y. Li, T. Wang, Y. Hu, H. Sun, J. Matsiko, S. Zheng,
298 A. De la Torre, I. Navarro, P. Sanz, M. A. Arjol, J. Fernández P. Wang and Q. Zhang, Distribution, seasonal variation and
and M. A. Martı́nez, HCH air levels derived from Bailı́n inhalation risks of polychlorinated dibenzo-p-dioxins and
dumpsite dismantling (Sabiñánigo, Spain), Sci. Total dibenzofurans, polychlorinated biphenyls and
Environ., 2018, 626, 1367–1372. polybrominated diphenyl ethers in the atmosphere of
299 M. Petrovic, M. Sremacki, J. Radonic, I. Mihajlovic, Beijing, China, Environ. Geochem. Health, 2018, 40, 1907–
B. Obrovski and M. Vojinovic Miloradov, Health risk 1918.
assessment of PAHs, PCBs and OCPs in atmospheric air 310 X. Y. Wang, Q. B. Li, Y. M. Luo, Q. Ding, L. M. Xi, J. M. Ma,
of municipal solid waste landll in Novi Sad, Serbia, Sci. Y. Li, Y. P. Liu and C. L. Cheng, Characteristics and sources
Total Environ., 2018, 644, 1201–1206. of atmospheric polycyclic aromatic hydrocarbons (PAHs) in
300 P. Wang, N. Zhao, Y. Cui, W. Jiang, L. Wang, Z. Wang, Shanghai, China, Environ. Monit. Assess., 2010, 165, 295–
X. Chen, L. Jiang and L. Ding, Short-chain chlorinated 305.
paraffin (SCCP) pollution from a CP production plant in 311 X. Zheng, D. Chen, X. Liu, Q. Zhou, Y. Liu, W. Yang and
China: Dispersion, congener patterns and health risk G. Jiang, Spatial and seasonal variations of
assessment, Chemosphere, 2018, 211, 456–464. organochlorine compounds in air on an urban–rural
301 M. M. Gune, W.-L. Ma, S. Sampath, W. Li, Y.-F. Li, transect across Tianjin, China, Chemosphere, 2010, 78,
H. N. Udayashankar, K. Balakrishna and Z. Zhang, 92–98.
Occurrence of polycyclic aromatic hydrocarbons (PAHs) in 312 L. Ding, Y. M. Li, P. Wang, X. M. Li, Z. S. Zhao, T. Ruan and
air and soil surrounding a coal-red thermal power plant Q. A. Zhang, Spatial concentration, congener proles and
inhalation risk assessment of PCDD/Fs and PCBs in the

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1991
Environmental Science: Processes & Impacts Critical Review

atmosphere of Tianjin, China, Chin. Sci. Bull., 2013, 58, 325 M. Lévy, J. Al-Alam, C. Ridacker, S. Massemin and M. Millet,
971–978. Use of XAD®-2 passive air samplers for monitoring
313 Y. Liang, H. Wang, Q. Yang, S. Cao, C. Yan, L. Zhang and environmental trends of PAHs, PCBs and pesticides in
N. Tang, Spatial distribution and seasonal variations of three different sites in Strasbourg and its vicinity (east of
atmospheric organophosphate esters (OPEs) in Tianjin, France), Atmos. Environ., 2018, 195, 12–23.
China based on gridded eld observations, Environ. 326 T. Gouin, D. Wilkinson, S. Hummel, B. Meyer and A. Culley,
Pollut., 2020, 263, 114460. Polycyclic aromatic hydrocarbons in air and snow from
314 W.-L. Ma, H. Qi, Y.-F. Li, L.-Y. Liu, D.-Z. Sun, D.-G. Wang, Fairbanks, Alaska, Atmos. Pollut. Res., 2010, 1, 9–15.
Z. Zhang, C.-G. Tian and J.-M. Shen, Seasonal and spatial 327 N. Jariyasopit, T. Harner, D. Wu, A. Williams,
variations of air concentrations of polycyclic aromatic S. Halappanavar and K. Su, Mapping indicators of toxicity
hydrocarbons in Northeastern Chinese urban region, Bull. for polycyclic aromatic compounds in the atmosphere of
Environ. Contam. Toxicol., 2011, 86, 43–49. the Athabasca oil sands region, Environ. Sci. Technol.,
315 E. C. Santiago and M. G. Cayetano, Polycyclic aromatic 2016, 50(20), 11282–11291.
hydrocarbons in ambient air in the Philippines derived 328 K. Pozo, T. Harner, A. Rudolph, G. Oyola, V. H. Estellano,
from passive sampler with polyurethane foam disk, R. Ahumada-Rudolpha, M. Garrido, K. Pozo, R. Mabilia
Atmos. Environ., 2007, 41, 4138–4147. and S. Focardi, Survey of persistent organic pollutants
316 J. Nasir, X. Wang, B. Xu, C. Wang, D. R. Joswiak, S. Rehman, (POPs) and polycyclic aromatic hydrocarbons (PAHs) in
A. Lodhi, S. Shaq and R. Jilani, Selected organochlorine the atmosphere of rural, urban and industrial areas of
pesticides and polychlorinated biphenyls in urban Concepción, Chile, using passive air samplers, Atmos.
atmosphere of Pakistan: Concentration, spatial variation Pollut. Res., 2012, 3, 426–434.
and sources, Environ. Sci. Technol., 2014, 48, 2610–2618. 329 K. Pozo, V. H. Estellano, T. Harner, L. Diaz-Robles,
317 J. Cortés, C. M. González, L. Morales, M. Abalos, E. Abad F. Cereceda-Balic, P. Etcharren, K. Pozo, V. Vidal,
and B. H. Aristizábal, PCDD/PCDF and dl-PCB in the F. Guerrero and A. Vergara-Fernández, Assessing
ambient air of a tropical Andean city: Passive and active Polycyclic Aromatic Hydrocarbons (PAHs) using passive
sampling measurements near industrial and vehicular air sampling in the atmosphere of one of the most wood-
pollution sources, Sci. Total Environ., 2014, 491/492, 67–74. smoke-polluted cities in Chile: The case study of Temuco,
318 J. Cortés, M. Cobo, C. M. González, C. D. Gómez, M. Abalos Chemosphere, 2015, 134, 475–481.
and B. H. Aristizábal, Environmental variation of PCDD/Fs 330 K. Pozo, G. Oyola, V. H. Estellano, T. Harner, A. Rudolph,
and dl-PCBs in two tropical Andean Colombian cities using P. Přibylová, P. Kukučka, O. Audy, J. Klánová,
passive samplers, Sci. Total Environ., 2016, 568, 614–623. A. Metzdorff and S. Focardi, Persistent Organic Pollutants
319 S.-D. Choi, H.-O. Kwon, Y.-S. Lee, E.-J. Park and J.-Y. Oh, (POPs) in the atmosphere of three Chilean cities using
Improving the spatial resolution of atmospheric passive air samplers, Sci. Total Environ., 2017, 586, 107–114.
polycyclic aromatic hydrocarbons using passive air 331 Á. L. Álvarez, K. Pozo, M. I. Paéz, V. H. Estellano, Y. Llanos
samplers in a multi-industrial city, J. Hazard. Mater., and S. Focardi, Semivolatile Organic Compounds (SVOCs)
2012, 241/242, 252–258. in the atmosphere of Santiago de Cali, Valle del Cauca,
320 Q. T. Vuong, S. J. Kim, T. N. T. Nguyen, P. Q. Thang, Colombia along north-south transect using polyurethane
S.-J. Lee, T. Ohura and S.-D. Choi, Passive air sampling of foam disk as passive air samplers, Atmos. Pollut. Res.,
halogenated polycyclic aromatic hydrocarbons in the 2016, 7, 945–953.
largest industrial city in Korea: Spatial distributions and 332 M. F. Ruggeri, N. B. Lana, J. C. Altamirano and
source identication, J. Hazard. Mater., 2020, 382, 121238. S. E. Puliato, Spatial distribution, patterns and source
321 Y. M. Aydin, M. Kara, Y. Dumanoglu, M. Odabasi and contributions of POPs in the atmosphere of Great
T. Elbir, Source apportionment of polycyclic aromatic Mendoza using the WRF/CALMET/CALPUFF modelling
hydrocarbons (PAHs) and polychlorinated biphenyls system, Emerging Contam., 2020, 6, 103–113.
(PCBs) in ambient air of an industrial region in Turkey, 333 N. L. Devi, S. Qi, P. Chakraborty, G. Zhang and I. C. Yadav,
Atmos. Environ., 2014, 97, 271–285. Passive air sampling of organochlorine pesticides in
322 P. Čupr, J. Klánová, T. Bartos, Z. Flegrova, J. Kohoutek and a northeastern state of India, Manipur, J. Environ. Sci.,
I. Holoubek, Passive air sampler as a tool for long-term air 2011, 23, 808–815.
pollution monitoring: Part 2. Air genotoxic potency 334 N. L. Devi, S. Qi, P. Chakraborty, G. Zhang and I. C. Yadav,
screening assessment, Environ. Pollut., 2006, 144, 406–413. Atmospheric polycyclic aromatic hydrocarbons (PAH) in
323 O. Sanka, L. Melymuk, P. Čupr, A. Dvorska and J. Klánová, Manipur of the Northeast India: Monitoring on urban,
Dispersion modeling of selected PAHs in urban air: A new rural, and mountain sites, Polycyclic Aromat. Compd.,
approach combining dispersion model with GIS and 2014, 34, 12–34.
passive air sampling, Atmos. Environ., 2014, 96, 88–95. 335 K. Pozo, S. K. Sarkar, V. H. Estellano, S. Mitra, O. Audy,
324 P. S. Diefenbacher, A. C. Gerecke, C. Bogdal and P. Kukučka, P. Přibylová, J. Klánová and S. Corsolini,
K. Hungerbühler, Spatial distribution of atmospheric Passive air sampling of persistent organic pollutants
PCBs in Zürich, Switzerland: Do joint sealants still (POPs) and emerging compounds in Kolkata megacity
matter?, Environ. Sci. Technol., 2016, 50, 232–239. and rural mangrove wetland Sundarban in India: An

1992 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

approach to regional monitoring, Chemosphere, 2017, 168, Atmospheric concentrations and air–soil gas exchange of
1430–1438. polycyclic aromatic hydrocarbons (PAHs) in remote, rural
336 B. Pokhrel, P. Gong, X. Wang, C. Wang and S. Gao, village and urban areas of Beijing–Tianjin region, North
Polycyclic aromatic hydrocarbons in the urban China, Sci. Total Environ., 2011, 409, 2942–2950.
atmosphere of Nepal: Distribution, sources, seasonal 347 C. Qu, X. Xing, S. Albanese, A. Doherty, H. Huang, A. Lima,
trends, and cancer risk, Sci. Total Environ., 2018, 618, S. Qi and B. De Vivo, Spatial and seasonal variations of
1583–1590. atmospheric organochlorine pesticides along the plain-
337 B. Pokhrel, P. Gong, X. Wang, S. N. Khanal, J. Ren, C. Wang, mountain transect in central China: Regional source vs.
S. Gao and T. Yao, Atmospheric organochlorine pesticides long-range transport and air–soil exchange, Atmos.
and polychlorinated biphenyls in urban areas of Nepal: Environ., 2015, 122, 31–40.
Spatial variation, sources, temporal trends, and long- 348 X. Mao, Z. Yu, Z. Ding, T. Huang, J. Ma, G. Zhang, J. Li and
range transport potential, Atmos. Chem. Phys., 2018, 18, H. Gao, Sources and potential health risk of gas phase
1325–1336. PAHs in Hexi Corridor, Northwest China, Environ. Sci.
338 B. Li, S. Wu, S. Zhou, T. Wang and C. Wang, Spatiotemporal Pollut. Res., 2016, 23, 2603–2612.
distribution and dynamic modeling of atmospheric 349 V. H. Estellano, K. Pozo, T. Harner, S. Corsolini and
gaseous polycyclic aromatic hydrocarbons in a rapidly S. Focardi, Using PUF disk passive samplers to
urbanizing city: Nanjing, China, Environ. Geochem. Health, simultaneously measure air concentrations of persistent
2018, 40, 2603–2616. organic pollutants (POPs) across the Tuscany Region,
339 L. Tong, C.-H. Peng, Z.-W. Huang, J.-J. Zhang, X.-R. Dai, Italy, Atmos. Pollut. Res., 2012, 3, 88–94.
H. Xiao, N.-B. Xu and J. He, Identifying the pollution 350 V. H. Estellano, K. Pozo, C. Silibello, M. D. Mulder,
characteristics of atmospheric polycyclic aromatic C. Efstathiou, M. P. Tomasino, F. Funaro, I. Donadio and
hydrocarbons associated with functional districts in S. Focardi, Characterization of urban pollution in two
Ningbo, China, Bull. Environ. Contam. Toxicol., 2019, 103, cities of the Puglia region in Southern Italy using eld
34–40. measurements and air quality (AQ) model approach,
340 D. Wu, H. Liu, Z. Wang, J. Zhang, C. Zhan, S. Liu, T. Liu, Atmos. Pollut. Res., 2014, 5, 34–41.
J. Zheng, R. Yao and J. Cao, Atmospheric concentrations 351 V. H. Estellano, K. Pozo, P. Přibylová, J. Klánová, O. Audy
and air–soil exchange of polycyclic aromatic and S. Focardi, Assessment of seasonal variations in
hydrocarbons (PAHs) in typical urban-rural fringe of persistent organic pollutants across the region of Tuscany
Wuhan-Ezhou region, Central China, Bull. Environ. using passive air samplers, Environ. Pollut., 2017, 222,
Contam. Toxicol., 2020, 104, 96–106. 609–616.
341 Y. Wang, M. Bao, Y. Zhang, F. Tan, H. Zhao, Q. Zhang and 352 S.-Y. Baek, S.-D. Choi, S.-J. Lee and Y.-S. Chang, Assessment
Q. Li, Polycyclic aromatic hydrocarbons in the atmosphere of the spatial distribution of coplanar PCBs, PCNs, and
and soils of Dalian, China: Source, urban-rural gradient, PBDEs in a multi-industry Region of South Korea using
and air–soil exchange, Chemosphere, 2020, 244, 125518. passive air samplers, Environ. Sci. Technol., 2008, 42,
342 P. Q. Thang, S.-J. Kim, S.-J. Lee, C. H. Kim, H.-J. Lim, 7336–7340.
S.-B. Lee, J. Y. Kim, Q. T. Vuong and S.-D. Choi, 353 O. Roots, A. Roose, A. Kull, I. Holoubek, P. Čupr and
Monitoring of polycyclic aromatic hydrocarbons using J. Klánová, Distribution pattern of PCBs, HCB and PeCB
passive air samplers in Seoul, South Korea: Spatial using passive air and soil sampling in Estonia, Environ.
distribution, seasonal variation, and source identication, Sci. Pollut. Res., 2010, 17, 740–749.
Atmos. Environ., 2020, 229, 117460. 354 O. Roots, T. Lukki, P. Přibylová, J. Borůvková, P. Kukučka,
343 X. Jiang, L. Gao, G. Liu, C. Li and M. Zheng, A case study on O. Audy, J. Kalina, J. Klánová, I. Holoubek, A. Sweetman
the long-term inuence of historical intensive usage of and O. Schleicher, Measurements of persistent organic
organochlorine pesticides on current ambient air, Aerosol pollutants in Estonian ambient air (1990–2013), Proc. Est.
Air Qual. Res., 2014, 14, 1047–1054. Acad. Sci., 2015, 64, 184–199.
344 B. Cetin, F. Ozturk, M. Keles and S. Yurdakul, PAHs and 355 A. Adu-Kumi, R. Kares, J. Literak, J. Boruvkova,
PCBs in an Eastern Mediterranean megacity, Istanbul: P. O. Yeboah, D. Carboo, O. Akoto, G. Darko, S. Osae and
Their spatial and temporal distributions, air–soil J. Klánová, Levels and seasonal variations of
exchange and toxicological effects, Environ. Pollut., 2017, organochlorine pesticides in urban and rural background
220, 1322–1332. air of southern Ghana, Environ. Sci. Pollut. Res., 2012, 19,
345 A. Birgül, P. B. Kurt-Karakuş, H. Alegria, E. Gungormus, 1963–1970.
H. Celik, T. Cicek and E. C. Güven, Polyurethane foam 356 J. Sultana, J. H. Syed, A. Mahmood, U. Ali, M. Y. A. Rehman,
(PUF) disk passive samplers derived polychlorinated R. N. Malik, J. Li and G. Zhang, Investigation of
biphenyls (PCBs) concentrations in the ambient air of organochlorine pesticides from the Indus Basin, Pakistan:
Bursa-Turkey: Spatial and temporal variations and health Sources, air–soil exchange uxes and risk assessment, Sci.
risk assessment, Chemosphere, 2017, 168, 1345–1355. Total Environ., 2014, 497/498, 113–122.
346 W. Wang, S. Simonich, B. Giri, Y. Chang, Y. Zhang, Y. Jia, 357 M. F. Silva-Barni, M. Gonzalez, F. Wania, Y. D. Lei and
S. Tao, R. Wang, B. Wang, W. Li, J. Cao and X. Lu, K. S. B. Miglioranza, Spatial and temporal distribution of

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1993
Environmental Science: Processes & Impacts Critical Review

pesticides and PCBs in the atmosphere using XAD-resin southern Italy: Implications for sources and
based passive air sampler: The case study of Quequén environmental processes, Environ. Int., 2019, 124, 84–97.
Grande River watershed (Argentina), Atmos. Pollut. Res., 369 I. Navarro, A. de la Torre, P. Sanz, M. A. Arjol, J. Fernández
2018, 9, 238–245. and M. A. Martı́nez, Organochlorine pesticides air
358 B. M. Sharma, L. Melymuk, G. K. Bharat, P. Přibylová, monitoring near a historical lindane production site in
O. Sáňka, J. Klánová and L. Nizzetto, Spatial gradients of Spain, Sci. Total Environ., 2019, 670, 1001–1007.
polycyclic aromatic hydrocarbons (PAHs) in air, 370 Y. Moussaoui, L. Tuduri, Y. Kerchich, B. Y. Meklati and
atmospheric deposition, and surface water of the Ganges G. Eppe, Atmospheric concentrations of PCDD/Fs, dl-
River basin, Sci. Total Environ., 2018, 627, 1495–1504. PCBs and some pesticides in northern Algeria using
359 S. N. Khuman and P. Chakraborty, Air-water exchange of passive air sampling, Chemosphere, 2012, 88, 270–277.
pesticidal persistent organic pollutants in the lower 371 Z. J. Katima, O. I. Olukunle, O. L. Kalantzi, A. P. Daso and
stretch of the transboundary river Ganga, India, J. O. Okonkwo, The occurrence of brominated ame
Chemosphere, 2019, 233, 966–974. retardants in the atmosphere of Gauteng Province, South
360 S. Niu, R. Chen, Y. Zou, L. Dong, R. Hai and Y. Huang, Africa using polyurethane foam passive air samplers and
Spatial distribution and prole of atmospheric short- assessment of human exposure, Environ. Pollut., 2018,
chain chlorinated paraffins in the Yangtze River Delta, 242, 1894–1903.
Environ. Pollut., 2020, 259, 113958. 372 H. A. Alegria, F. Wong, L. M. Jantunen, T. F. Bidleman,
361 S. Sampath, G. Shanmugam, K. K. Selvaraj and M. S. Figueroa, G. G. Bouchot, V. C. Moreno,
B. R. Ramaswamy, Spatio-temporal distribution of S. M. Waliszewski and R. Infanzon, Organochlorine
polycyclic aromatic hydrocarbons (PAHs) in atmospheric pesticides and PCBs in air of southern Mexico (2002-
air of Tamil Nadu, India, and human health risk 2004), Atmos. Environ., 2008, 42, 8810–8818.
assessment, Environ. Forensics, 2015, 16, 76–87. 373 N. Tombesi, K. Pozo and T. Harner, Persistent Organic
362 S. Niu, L. Dong, L. Zhang, C. Zhu, R. Hai and Y. Huang, Pollutants (POPs) in the atmosphere of agricultural and
Temporal and spatial distribution, sources, and potential urban areas in the province of Buenos Aires in Argentina
health risks of ambient polycyclic aromatic hydrocarbons using PUF disk passive air samplers, Atmos. Pollut. Res.,
in the Yangtze River Delta (YRD) of eastern China, 2014, 5, 170–178.
Chemosphere, 2017, 172, 72–79. 374 C. N. Pegoraro and E. D. Wannaz, Occurrence of persistent
363 Y. Dumanoglu, E. O. Gaga, E. Gungormus, S. C. Sofuoglu organic pollutants in air at different sites in the province of
and M. Odabasi, Spatial and seasonal variations, sources, Córdoba, Argentina, Environ. Sci. Pollut. Res., 2019, 26,
air–soil exchange, and carcinogenic risk assessment for 18379–18391.
PAHs and PCBs in air and soil of Kutahya, Turkey, the 375 M. M. Orazi, A. H. Arias, A. L. Oliva, A. C. Ronda and
province of thermal power plants, Sci. Total Environ., J. E. Marcovecchio, Characterization of atmospheric and
2017, 580, 920–935. soil polycyclic aromatic hydrocarbons and evaluation of
364 U. Ali, A. J. Sweetman, K. C. Jones and R. N. Malik, Higher air–soil relationship in the Southwest of Buenos Aires
atmospheric levels and contribution of black carbon in soil- province (Argentina), Chemosphere, 2020, 240, 124847.
air partitioning of organochlorines in Lesser Himalaya, 376 J. H. Syed, R. N. Malik, D. Liu, Y. Xu, Y. Wang, J. Li,
Chemosphere, 2018, 191, 787–798. G. Zhang and K. C. Jones, Organochlorine pesticides in
365 K. Pozo, M. Palmeri, V. Palmeri, V. H. Estellano, air and soil and estimated air–soil exchange in Punjab,
M. D. Mulder, C. I. Efstathiou, G. Sará, T. Romeo, Pakistan, Sci. Total Environ., 2013, 444, 491–497.
G. Lammel and S. Focardi, Assessing persistent organic 377 U. Ali, J. H. Syed, A. Mahmood, J. Li, G. Zhang, K. C. Jones
pollutants (POPs) in the Sicily island atmosphere, and R. N. Malik, Inuential role of black carbon in the soil-
Mediterranean, using PUF disk passive air samplers, air partitioning of polychlorinated biphenyls (PCBs) in the
Environ. Sci. Pollut. Res., 2016, 23, 20796–20804. Indus River Basin, Pakistan, Chemosphere, 2015, 134, 172–
366 S. Loppi, K. Pozo, V. H. Estellano, S. Corsolini, G. Sardella 180.
and L. Paoli, Accumulation of polycyclic aromatic 378 A. Kamal, J. H. Syed, J. Li, G. Zhang, A. Mahmood and
hydrocarbons by lichen transplants: Comparison with R. N. Malik, Prole of atmospheric PAHs in Rawalpindi,
gas-phase passive air samplers, Chemosphere, 2015, 134, Lahore and Gujranwala districts of Punjab province
39–43. (Pakistan), Aerosol Air Qual. Res., 2016, 16, 1010–1021.
367 T. F. Bidleman, H. Laudon, O. Nygren, S. Svanberg and 379 Y. Min, J. Heo and M. Lee, Determination of toxic
M. Tysklind, Chlorinated pesticides and natural congeners of 17 PCDDs/PCDFs and 12 dl-PCBs using
brominated anisoles in air at three northern Baltic polyurethane foam passive air samplers in ten cities
stations, Environ. Pollut., 2017, 225, 381–389. around Seoul, Sci. Total Environ., 2014, 491/492, 17–27.
368 C. Qu, S. Albanese, A. Lima, D. Hope, P. Pond, A. Fortelli, 380 T. N. T. Nguyen, H. O. Kwon, G. Lammel, K. S. Jung, S. J. Lee
N. Romano, P. Cerino, A. Pizzolante and B. De Vivo, The and S. D. Choi, Spatially high-resolved monitoring and risk
occurrence of OCPs, PCBs, and PAHs in the soil, air, and assessment of polycyclic aromatic hydrocarbons in an
bulk deposition of the Naples metropolitan area, industrial city, J. Hazard. Mater., 2020, 393, 122409.

1994 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

381 E. Can-Güven, K. Gedik and P. B. Kurt-Karakuş, 393 J. N. Hogarh, N. Seike, Y. Kobara, G. K. Ofosu-Budu,
Polyurethane foam disk passive sampler derived air D. Carboo and S. Masunaga, Atmospheric burden of
concentrations of persistent organic pollutants in an organochlorine pesticides in Ghana, Chemosphere, 2014,
agricultural region with hot climate, Atmos. Pollut. Res., 102, 1–5.
2019, 10, 1913–1920. 394 G. Aliyeva, R. Kurkova, I. Hovorkova, J. Klánová and
382 B. Cetin, S. Yurdakul, E. Gungormus, F. Ozturk and C. Halsall, Organochlorine pesticides and polychlorinated
S. Sofuoglu, Source apportionment and carcinogenic risk biphenyls in air and soil across Azerbaijan, Environ. Sci.
assessment of passive air sampler-derived PAHs and PCBs Pollut. Res., 2012, 19, 1953–1962.
in a heavily industrialized region, Sci. Total Environ., 395 P. Chakraborty, G. Zhang, J. Li, Y. Xu, X. Liu, S. Tanabe and
2018, 633, 30–41. K. C. Jones, Selected organochlorine pesticides in the
383 L. Wang, Y. Zhao, X. Yi, Z. Wang, Y. Yi, T. Huang, H. Gao atmosphere of major Indian cities: Levels, regional versus
and J. Ma, Spatial distribution of atmospheric PAHs and local variations and sources, Environ. Sci. Technol., 2010,
their genotoxicity in petrochemical industrialized 44, 8038–8043.
Lanzhou valley, northwest China, Environ. Sci. Pollut. Res., 396 J. E. Schrlau, L. Geiser, K. J. Hageman, D. H. Landers and
2017, 24, 12820–12834. S. M. Simonich, Comparison of lichen, conifer needles,
384 P. Liu, Y. Ju, Y. Li, Z. Wang, X. Mao, H. Cao, C. Jia, T. Huang, passive air sampling devices, and snowpack as passive
H. Gao and J. Ma, Spatiotemporal variation of atmospheric sampling media to measure semi-volatile organic
nitrated polycyclic aromatic hydrocarbons in semi-arid and compounds in remote atmospheres, Environ. Sci. Technol.,
petrochemical industrialized Lanzhou City, Northwest 2011, 45, 10354–10361.
China, Environ. Sci. Pollut. Res., 2019, 26, 1857–1870. 397 Y. Yao, T. Harner, J. Ma, L. Tuduri and P. Blanchard,
385 C. Schummer, L. Tuduri, O. Briand, B. M. Appenzeller and Sources and occurrence of dacthal in the Canadian
M. Millet, Application of XAD-2 resin-based passive atmosphere, Environ. Sci. Technol., 2007, 41, 688–694.
samplers and SPME-GC-MS/MS analysis for the 398 P. G. Messing, A. Farenhorst, D. T. Waite and J. F. Sproull,
monitoring of spatial and temporal variations of Current-use herbicides in air as inuenced by their
atmospheric pesticides in Luxembourg, Environ. Pollut., estimated agricultural use at various distances from six
2012, 170, 88–94. sampling locations, Water, Air, Soil Pollut., 2014, 225,
386 C. Schummer, B. M. Appenzeller and M. Millet, Monitoring 2013–2022.
of polycyclic aromatic hydrocarbons (PAHs) in the 399 N. Ren, E. Sverko, Y.-F. Li, Z. Zhang, T. Harner, D. Wang,
atmosphere of southern Luxembourg using XAD-2 resin- X. Wan and B. E. McCarry, Levels and isomer proles of
based passive samplers, Environ. Sci. Pollut. Res., 2014, 21, dechlorane plus in Chinese air, Environ. Sci. Technol.,
2098–2107. 2008, 42, 6476–6480.
387 Z. Zhang, L. Liu, Y.-F. Li, D. Wang, H. Jia, T. Harner, 400 K. Pozo, T. Harner, M. Shoeib, R. Urrutia, R. Barra, O. Parra
E. Sverko, X. Wan, D. Xu, N. Ren, J. Ma and K. Pozo, and S. Focardi, Passive-sampler derived air concentrations
Analysis of polychlorinated biphenyls in concurrently of persistent organic pollutants on a North-South transect
sampled Chinese air and surface soil, Environ. Sci. in Chile, Environ. Sci. Technol., 2004, 38, 6529–6537.
Technol., 2008, 42, 6514–6518. 401 C. Shunthirasingham, T. Gouin, Y. D. Lei, C. Ruepert,
388 Y. Zhao, J. Ma, X. Qiu, Y. Lin, Q. Yang and T. Zhu, Gridded L. E. Castillo and F. Wania, Current-use pesticides
eld observations of polybrominated diphenyl ethers and transport to Costa Rica's high-altitude tropical cloud
decabromodiphenyl ethane in the atmosphere of North forest, Environ. Toxicol. Chem., 2011, 30, 2709–2717.
China, Environ. Sci. Technol., 2013, 47, 8123–8129. 402 F. Wong, H. A. Alegria, T. F. Bidleman, V. Alvarado,
389 B. Gevao, A. Al-Omair, A. J. Sweetman, L. Al-Ali, M. Al- F. Angeles, A. A. Galarza, B. Bandala, I. D. L. C. Hinojosa,
Bahloul, M. Helaleh and J. Zafar, Passive sampler-derived I. G. Estrada, G. G. Ryes, G. Gold-Bouchot,
air concentrations for polybrominated diphenyl ethers J. V. M. Zamora and E. R. Esinoza, Passive air sampling of
and polycyclic aromatic hydrocarbons in Kuwait, Environ. organochlorine pesticides in Mexico, Environ. Sci.
Toxicol. Chem., 2006, 25, 1496–1502. Technol., 2009, 43, 704–710.
390 G. L. Daly, Y. D. Lei, C. Teixeira, D. C. G. Muir, L. E. Castillo, 403 X. Liu, G. Zhang, J. Li, H.-R. Chang, S.-H. Qi, X.-D. Li and
L. M. M. Jantunen and F. Wania, Organochlorine pesticides K. C. Jones, Polycyclic aromatic hydrocarbons (PAHs) in
in the soils and atmosphere of Costa Rica, Environ. Sci. the air of Chinese cities, J. Environ. Monit., 2007, 9, 1092–
Technol., 2007, 41, 1124–1130. 1098.
391 G. L. Daly, Y. D. Lei, C. Teixeira, D. C. G. Muir, L. E. Castillo 404 X. Liu, G. Zhang, J. Li, L.-L. Yu, Y. Xu, X.-D. Li, Y. Kobara
and F. Wania, Accumulation of current-use pesticides in and K. C. Jones, Seasonal patterns and current sources of
neotropical montane forests, Environ. Sci. Technol., 2007, DDTs, chlordanes, hexachlorobenzene, and endosulfan in
41, 1118–1123. the atmosphere of 37 Chinese cities, Environ. Sci.
392 J. N. Hogarh, N. Seike, Y. Kobara and S. Masunaga, Technol., 2009, 43, 1316–1321.
Atmospheric polychlorinated naphthalenes in Ghana, 405 Y. Lin, Y. Zhao, X. Qiu, J. Ma, Q. Yang, M. Shao and T. Zhu,
Environ. Sci. Technol., 2012, 46, 2600–2606. Spatial distribution of polychlorinated naphthalenes in the

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1995
Environmental Science: Processes & Impacts Critical Review

atmosphere across North China based on gridded eld to atmospheric polychlorinated biphenyls (PCBs) in Ghana,
observations, Environ. Pollut., 2013, 180, 27–33. Environ. Sci. Pollut. Res., 2018, 25, 16316–16324.
406 Y. Lin, X. Qiu, Y. Zhao, J. Ma, Q. Yang and T. Zhu, 417 W. Wang, Y. Wang, R. Zhang, S. Wang, C. Wei, C. Chaemfa,
Polybromobenzene pollutants in the atmosphere of North J. Li, G. Zhang and K. Yu, Seasonal characteristics and
China: Levels, distribution, and sources, Environ. Sci. current sources of OCPs and PCBs and enantiomeric
Technol., 2013, 47, 12761–12767. signatures of chiral OCPs in the atmosphere of Vietnam,
407 Y. Lin, X. Qiu, Y. Ma, J. Ma, M. Zheng and M. Shao, Sci. Total Environ., 2016, 542, 777–786.
Concentrations and spatial distribution of polycyclic 418 I. C. Yadav, N. L. Devi, J. Li, G. Zhang and K. Breivik,
aromatic hydrocarbons (PAHs) and nitrated PAHs Possible emissions of POPs in plain and hilly areas of
(NPAHs) in the atmosphere of North China, and the Nepal: Implications for source apportionment and health
transformation from PAHs to NPAHs, Environ. Pollut., risk assessment, Environ. Pollut., 2017, 220, 1289–1300.
2015, 196, 164–170. 419 P. B. Kurt-Karakuş, T. Ugranli-Cicek, S. C. Sofuoglu,
408 G. Zhang, P. Chakraborty, J. Li, P. Sampathkumar, H. Celik, E. Gungormus, K. Gedik, A. Sofuoglu,
T. Balasubramanian, K. Kathiresan, S. Takahashi, H. E. Okten, A. Birgul, H. Alegria and K. C. Jones, The
A. Subramanian, S. Tanabe and K. C. Jones, Passive rst countrywide monitoring of selected POPs:
atmospheric sampling of organochlorine pesticides, Polychlorinated biphenyls (PCBs), polybrominated
polychlorinated biphenyls, and polybrominated diphenyl diphenyl ethers (PBDEs) and selected organochlorine
ethers in urban, rural, and wetland sites along the coastal pesticides (OCPs) in the atmosphere of Turkey, Atmos.
length of India, Environ. Sci. Technol., 2008, 42, 8218–8223. Environ., 2018, 177, 154–165.
409 C. Chaemfa, Y. Xu, J. Li, P. Chakraborty, J. H. Syed, 420 S. Zhao, K. C. Jones, J. Li, A. J. Sweetman, X. Liu, Y. Xu,
R. N. Malik, Y. Wang, C. Tian, G. Zhang and K. C. Jones, Y. Wang, T. Lin, S. Mao, K. Li, J. Tang and G. Zhang,
Screening of atmospheric short- and medium-chain Evidence for major contributions of unintentionally
chlorinated paraffins in India and Pakistan using produced PCBs in the air of China: Implications for the
polyurethane foam based passive air sampler, Environ. national source inventory, Environ. Sci. Technol., 2020, 54,
Sci. Technol., 2014, 48, 4799–4808. 2163–2171.
410 Y. Xu, J. Li, P. Chakraborty, J. H. Syed, R. N. Malik, Y. Wang, 421 R. Gioia, S. Eckhardt, K. Breivik, F. M. Jaward, A. Prieto,
C. Tian, C. Luo, G. Zhang and K. C. Jones, Atmospheric L. Nizzetto and K. C. Jones, Evidence for major emissions
polychlorinated naphthalenes (PCNs) in India and of PCBs in the west African region, Environ. Sci. Technol.,
Pakistan, Sci. Total Environ., 2014, 466/467, 1030–1036. 2011, 45, 1349–1355.
411 P. Chakraborty, G. Zhang, H. Cheng, P. Balasubramanian, 422 N. Isogai, J. N. Hogarh, N. Seike, Y. Kobara, F. Oyediran,
J. Li and K. C. Jones, Passive air sampling of M. J. Wirmvem, S. N. Ayonghe, J. Fobil and S. Masunaga,
polybrominated diphenyl ethers in New Delhi, Kolkata, Atmospheric monitoring of organochlorine pesticides
Mumbai and Chennai: Levels, homologous proling and across some West African countries, Environ. Sci. Pollut.
source apportionment, Environ. Pollut., 2017, 231, 1181– Res., 2018, 25, 31828–31835.
1187. 423 S.-Y. Baek, J. Jurng and Y.-S. Chang, Spatial distribution of
412 K. Pozo, T. Harner, S. C. Lee, R. K. Sinha, B. Sengupta, polychlorinated biphenyls, organochlorine pesticides, and
M. Loewen, V. Geethalakshmi, K. Kannan and V. Volp, dechlorane plus in Northeast Asia, Atmos. Environ., 2013,
Assessing seasonal and spatial trends of persistent 64, 40–46.
organic pollutants (POPs) in Indian agricultural regions 424 Q. Li, J. Li, C. Chaemfa, G. Zhang, Y. Kobara, J.-J. Nam and
using PUF disk passive air samplers, Environ. Pollut., K. C. Jones, The impact of polybrominated diphenyl ether
2011, 159, 646–653. prohibition: A case study on the atmospheric levels in
413 K. Kennedy, M. Macova, M. E. Bartkow, D. W. Hawker, China, Japan and South Korea, Atmos. Res., 2014, 143, 57–
B. Zhao, M. S. Dension and J. F. Müller, Effect based 63.
monitoring of seasonal ambient air exposures in Australia 425 T. Gouin, T. Harner, P. Blanchard and D. Mackay, Passive
sampled by PUF passive air samplers, Atmos. Pollut. Res., and active air samplers as complementary methods for
2010, 1, 50–58. investigating persistent organic pollutants in the Great
414 X. Wang, K. Kennedy, J. Powell, M. Keywood, R. Gillett, Lakes Basin, Environ. Sci. Technol., 2005, 39, 9115–9122.
P. Thai, P. Bridgen, S. Broomhall, C. Paxman, F. Wania 426 T. Harner, M. Shoeib, T. Gouin and P. Blanchard,
and J. Müller, Spatial distribution of selected persistent Polychlorinated naphthalenes in Great Lakes air:
organic pollutants (POPs) in Australia's atmosphere, Assessing spatial trends and combustion inputs using
Environ. Sci.: Processes Impacts, 2015, 17, 525–532. PUF disk passive air samplers, Environ. Sci. Technol.,
415 J. N. Hogarh, N. Seike, Y. Kobara and S. Masunaga, 2006, 40, 5333–5339.
Seasonal variation of atmospheric polychlorinated 427 T. Gouin, L. Jantunen, T. Harner, P. Blanchard and
biphenyls and polychlorinated naphthalenes in Japan, T. F. Bidleman, Spatial and temporal trends of chiral
Atmos. Environ., 2013, 80, 275–280. organochlorine signatures in Great Lakes air using
416 J. N. Hogarh, N. Seike, Y. Kobara, D. Carboo, J. N. Fobil and passive air samplers, Environ. Sci. Technol., 2007, 41,
S. Masunaga, Source characterization and risk of exposure 3877–3883.

1996 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

428 R. A. Gioia, A. J. Sweetman and K. C. Jones, Coupling 440 J. Klánová, P. Čupr, I. Holoubek, J. Borůvková, P. Přibylová,
passive air sampling with emission estimates and R. Kares, T. Tomsej and T. Ocelka, Monitoring of persistent
chemical fate modeling for persistent organic pollutants organic pollutants in Africa. Part 1: Passive air sampling
(POPs): A feasibility study for Northern Europe, Environ. across the continent in 2008, J. Environ. Monit., 2009, 11,
Sci. Technol., 2007, 41, 2165–2171. 1952–1963.
429 P. Bohlin, K. C. Jones, H. Tovalin and B. Strandberg, 441 S. Fuhrimann, J. Klánová, P. Přibylová, J. Kohoutek,
Observations on persistent organic pollutants in indoor M. A. Dalvie, M. Röösli and C. Degrendele, Qualitative
and outdoor air using passive polyurethane foam assessment of 27 current-use pesticides in air at 20
samples, Atmos. Environ., 2008, 42, 7234–7241. sampling sites across Africa, Chemosphere, 2020, 258,
430 P. Přibylová, R. Kares, J. Borůvková, P. Čupr, R. Prokeš, 127333.
J. Kohoutek, I. Holoubek and J. Klánová, Levels of 442 C. Rauert, T. Harner, J. K. Schuster, K. Quinto, G. Fillmann,
persistent organic pollutants and polycyclic aromatic L. E. Castillo, O. Fentanes, M. V. Ibarra,
hydrocarbons in ambient air of Central and Eastern K. S. B. Miglioranza, I. M. Rivadeneira, K. Pozo,
Europe, Atmos. Pollut. Res., 2012, 3, 494–505. A. P. Puerta and B. H. A. Zuluaga, Towards a regional
431 C. S. Kwan, H. Takada, R. Boonyatumanond, Y. Kato, passive air sampling network and strategy for new POPs
K. Mizukawa, M. Ito, L. Q. Dung, M. P. Zakaria and in the GRULAC region: Perspectives from the GAPS
E. C. Santiago, Historical occurrences of polybrominated Network and rst results for organophosphorus ame
diphenyl ethers and polychlorinated biphenyls in Manila retardants, Sci. Total Environ., 2016, 573, 1294–1302.
Bay, Philippines, and in the upper Gulf of Thailand, Sci. 443 C. Rauert, T. Harner, J. K. Schuster, A. Eng, G. Fillmann,
Total Environ., 2014, 470/471, 427–437. L. E. Castillo, O. Fentanes, M. Villa Ibarra,
432 L. Shen, F. Wania, Y. D. Lei, C. Teixeira, D. C. G. Muir and K. S. B. Miglioranza, I. Moreno Rivadeneira, K. Pozo and
T. F. Bidleman, Hexachlorocyclohexanes in the North B. H. Aristizábal Zuluaga, Air monitoring of new and
American atmosphere, Environ. Sci. Technol., 2004, 38, legacy POPs in the Group of Latin America and Caribbean
965–975. (GRULAC) region, Environ. Pollut., 2018, 243, 1252–1262.
433 L. Shen, F. Wania, Y. D. Lei, C. Teixeira, D. C. G. Muir and 444 T. Harner, K. Pozo, T. Gouin, A.-M. Macdonald, H. Hung,
H. Xiao, Polychlorinated biphenyls and polybrominated J. Cainey and A. Peters, Global pilot study for persistent
diphenyl ethers in the North American atmosphere, organic pollutants (POPs) using PUF disk passive air
Environ. Pollut., 2006, 144, 434–444. samplers, Environ. Pollut., 2006, 144, 445–452.
434 F. M. Jaward, N. J. Farrar, T. Harner, A. J. Sweetman and 445 S. C. Lee, T. Harner, K. Pozo, M. Shoeib, F. Wania,
K. C. Jones, Passive air sampling of PCBs, PBDEs, and D. C. G. Muir, L. A. Barrie and K. C. Jones,
organochlorine pesticides across Europe, Environ. Sci. Polychlorinated naphthalenes in the global atmospheric
Technol., 2004, 38, 34–41. passive sampling (GAPS) study, Environ. Sci. Technol.,
435 F. M. Jaward, N. J. Farrar, T. Harner, A. J. Sweetman and 2007, 41, 2680–2687.
K. C. Jones, Passive air sampling of polycyclic aromatic 446 S. C. Lee, E. Sverko, T. Harner, K. Pozo, E. Barresi,
hydrocarbons and polychlorinated naphthalenes across J. A. Schachtschneider, D. Zaruk, M. DeJong and
Europe, Environ. Toxicol. Chem., 2004, 23, 1355–1364. J. Narayan, Retrospective analysis of “new” ame
436 A. K. Halse, M. Schlabach, S. Eckhardt, A. J. Sweetman, retardants in the global atmosphere under the GAPS
K. C. Jones and K. Breivik, Spatial variability of POPs in Network, Environ. Pollut., 2016, 217, 62–69.
European background air, Atmos. Chem. Phys., 2011, 11, 447 C. Rauert, J. K. Schuster, A. Eng and T. Harner, Global
1549–1564. atmospheric concentrations of brominated and
437 F. M. Jaward, G. Zhang, J. J. Nam, A. J. Sweetman, chlorinated ame retardants and organophosphate esters,
J. P. Obbard, Y. Kobara and K. C. Jones, Passive air Environ. Sci. Technol., 2018, 52, 2777–2789.
Sampling of polychlorinated biphenyls, organochlorine 448 C. Shunthirasingham, C. E. Oyiliagu, X. Cao, T. Gouin,
compounds, and polybrominated diphenyl ethers across F. Wania, S. C. Lee, K. Pozo, T. Harner and D. C. G. Muir,
Asia, Environ. Sci. Technol., 2005, 39, 8638–8645. Spatial and temporal pattern of pesticides in the global
438 J. N. Hogarh, N. Seike, Y. Kobara, A. Habib, J.-J. Nama, atmosphere, J. Environ. Monit., 2010, 12, 1650–1657.
J.-S. Lee, Q. Li, X. Liu, J. Li, G. Zhang and S. Masunaga, 449 J. N. Westgate, C. Shunthirasingham, C. E. Oyiliagu, H. von
Passive air monitoring of PCBs and PCNs across East Waldow and F. Wania, Three methods for quantifying
Asia: A comprehensive congener evaluation for source proximity of air sampling sites to spatially resolved
characterization, Chemosphere, 2012, 86, 718–726. emissions of semivolatile organic contaminants, Atmos.
439 W.-J. Hong, H. Jia, W.-L. Ma, R. K. Sinha, H.-B. Moon, Environ., 2010, 44, 4380–4387.
H. Nakata, N. H. Minh, K. H. Chi, W.-L. Li, K. Kannan, 450 C. Bogdal, E. Abad, M. Abalos, B. van Bavel, J. Hagberg,
E. Sverko and Y.-F. Li, Distribution, fate, inhalation M. Scheringer and H. Fiedler, Worldwide distribution of
exposure and lung cancer risk of atmospheric polycyclic persistent organic pollutants in air, including results of
aromatic hydrocarbons in some Asian countries, Environ. air monitoring by passive air sampling in ve continents,
Sci. Technol., 2016, 50, 7163–7174. Trends Anal. Chem., 2013, 46, 151–161.

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1997
Environmental Science: Processes & Impacts Critical Review

451 P. Barthel, S. Thüns, C. Shunthirasingham, J. N. Westgate, Antonie”’ in Bitterfeld, Germany, Environ. Pollut., 2006,
F. Wania and M. Radke, Application of XAD-resin based 144, 414–422.
passive air samplers to assess local (roadside) and 463 Y. Wang, S. Wang, C. Luo, J. Li, L. Ming, G. Zhang and X. Li,
regional patterns of persistent organic pollutants, The effects of rice canopy on the air–soil exchange of
Environ. Pollut., 2012, 166, 218–225. polycyclic aromatic hydrocarbons and organochlorine
452 S.-D. Choi, C. Shunthirasingham, G. L. Daly, H. Xiao, pesticides using paired passive air samplers, Environ.
Y. D. Lei and F. Wania, Levels of polycyclic aromatic Pollut., 2015, 200, 35–41.
hydrocarbons in Canadian mountain air and soil are 464 D. J. Bolinius, A. Jahnke and M. MacLeod, Comparison of
controlled by proximity to roads, Environ. Pollut., 2009, eddy covariance and modied Bowen ratio methods for
157, 3199–3206. measuring gas uxes and implications for measuring
453 T. Harner, M. Shoeib, M. Diamond, G. Stern and uxes of persistent organic pollutants, Atmos. Chem.
B. Rosenberg, Using passive air samplers to assess urban- Phys., 2016, 16, 5315–5322.
rural trends for persistent organic pollutants. 1. 465 E. Moreau-Guigon, A. Motelay-Massei, T. Harner, K. Pozo,
polychlorinated biphenyls and organochlorine pesticides, M. L. Diamond, M. C. Chevreuil and H. Blanchoud,
Environ. Sci. Technol., 2004, 38, 4474–4483. Vertical and temporal distribution of persistent organic
454 A. Motelay-Massei, T. Harner, M. Shoeib, M. L. Diamond, Pollutants in Toronto. 1. Organochlorine pesticides,
G. Stern and B. Rosenberg, Using passive air samplers to Environ. Sci. Technol., 2007, 41, 2172–2177.
assess urban-rural trends for persistent organic pollutants 466 Y. Li, Q. Zhang, D. Ji, T. Wang, Y. Wang, P. Wang, L. Ding
and polycyclic aromatic hydrocarbons. 2. Seasonal trends and G. Jiang, Levels and vertical distributions of PCBs,
for PAHs, PCBs, and organochlorine pesticides, Environ. PBDEs, and OCPs in the atmospheric boundary layer:
Sci. Technol., 2005, 39, 5763–5773. Observation from the Beijing 325-m meteorological tower,
455 T. Gouin, T. Harner, G. L. Daly, F. Wania, D. Mackay and Environ. Sci. Technol., 2009, 43, 1030–1035.
K. C. Jones, Variability of concentrations of 467 B. L. van Drooge, J. O. Grimalt, K. Booij, L. Camarero and
polybrominated diphenyl ethers and polychlorinated J. Catalan, Passive sampling of atmospheric
biphenyls in air: Implications for monitoring, modeling organochlorine compounds by SPMDs in a remote high
and control, Atmos. Environ., 2005, 39, 151–166. mountain area, Atmos. Environ., 2005, 39, 5195–5204.
456 T. Harner, M. Shoeib, M. L. Diamond, M. Ikonomou and 468 F. M. Jaward, A. Di Guardo, L. Nizzetto, C. Cassani,
G. Stern, Passive sampler derived air concentrations of F. Raffaele, R. Ferretti and K. C. Jones, PCBs and selected
PBDEs along an urban–rural transect: Spatial and organochlorine compounds in Italian mountain air: the
temporal trends, Chemosphere, 2006, 64, 262–267. inuence of altitude and forest ecosystem type, Environ.
457 T. Gouin, M. Shoeib and T. Harner, Atmospheric Sci. Technol., 2005, 39, 3455–3463.
concentrations of current-use pesticides across south- 469 G. L. Daly, Y. D. Lei, C. Teixeira, D. C. G. Muir and F. Wania,
central Ontario using monthly-resolved passive air Pesticides in Western Canadian mountain air and soil,
samplers, Atmos. Environ., 2008, 42, 8096–8104. Environ. Sci. Technol., 2007, 41, 6020–6025.
458 L. Melymuk, M. Robson, P. A. Helm and M. L. Diamond, 470 M. Loewen, F. Wania, F. Wang and G. Tomy, Altitudinal
PCBs, PBDEs, and PAHs in Toronto air: Spatial and transect of atmospheric and aqueous uorinated organic
seasonal trends and implications for contaminant compounds in Western Canada, Environ. Sci. Technol.,
transport, Sci. Total Environ., 2012, 429, 272–280. 2008, 42, 2374–2379.
459 S. Harrad and S. Hunter, Concentrations of 471 B. Abdul Hussain, J. N. Westgate, S. J. Hayward,
polybrominated diphenyl ethers in air and soil on a rural- C. Shunthirasingham, T. N. Brown, H. Hung, Y. D. Lei
urban transect across a major UK conurbation, Environ. and F. Wania, Polycyclic aromatic hydrocarbons and
Sci. Technol., 2006, 40, 4548–4553. polychlorinated biphenyls in soils and atmosphere of
460 A. Jamshidi, S. Hunter, S. Hazrati and S. Harrad, Western Canadian mountains: The role of source
Concentrations and chiral signatures of polychlorinated proximity, precipitation, forest cover and mountain cold-
biphenyls in outdoor and indoor air and soil in a major trapping, Atmos. Environ. X, 2019, 100004.
U.K. conurbation, Environ. Sci. Technol., 2007, 41, 2153– 472 V. H. Estellano, K. Pozo, T. Harner, M. Franken and
2158. M. Zaballa, Altitudinal and seasonal variations of
461 H. von Waldow, M. MacLeod, M. Scheringer and persistent organic pollutants in the Bolivian Andes
K. Hungerbühler, Quantifying remoteness from emission mountains, Environ. Sci. Technol., 2008, 42, 2528–2534.
sources of persistent organic pollutants on a global scale, 473 R. O. Meire, S. C. Lee, A. C. Targino, J. P. M. Torres and
Environ. Sci. Technol., 2010, 44, 2791–2796. T. Harner, Air concentrations and transport of persistent
462 A. Paschke, B. Vrana, P. Popp and G. Schüürmann, organic pollutants (POPs) in mountains of southeast and
Comparative application of solid-phase microextraction southern Brazil, Atmos. Pollut. Res., 2012, 3, 417–425.
bre assemblies and semi-permeable membrane devices 474 R. O. Meire, S. C. Lee, Y. Yao, A. C. Targino, J. P. M. Torres
as passive air samplers for semi-volatile chlorinated and T. Harner, Seasonal and altitudinal variations of legacy
organic compounds. A case study on the landll “Grube and current-use pesticides in the Brazilian tropical and
subtropical mountains, Atmos. Environ., 2012, 59, 108–116.

1998 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

475 Y. D. S. Guida, R. O. Meire, T. P. M. Torres and O. Malm, Air measurement based on empirical and theoretical factors,
contamination by legacy and current-use pesticides in Environ. Sci. Technol., 2015, 49, 113–122.
Brazilian mountains: An overview of national regulations 487 P. Bohlin, K. C. Jones and B. Strandberg, Occupational and
by monitoring pollutant presence in pristine areas, indoor air exposure to persistent organic pollutants:
Environ. Pollut., 2018, 242, 19–30. a review of passive sampling techniques and needs, J.
476 P. Gong, X.-P. Wang, S.-H. Li, W.-S. Yu, J.-I. Li, D. B. Kattel, Environ. Monit., 2007, 9, 501–509.
W.-C. Wang, L. P. Devkota, T.-D. Yao and D. R. Joswiak, 488 R. W. Gale, W. L. Cranor, D. Á. Álvarez, J. A. Huckins,
Atmospheric transport and accumulation of J. D. Petty and G. L. Robertson, Semivolatile organic
organochlorine compounds on the southern slopes of the compounds in residential air along the Arizona-Mexico
Himalayas, Nepal, Environ. Pollut., 2014, 192, 44–51. border, Environ. Sci. Technol., 2009, 43, 3054–3060.
477 B. Pokhrel, P. Gong, X. Wang, S. Gao, C. Wang and T. Yao, 489 S. Harrad, S. Hazrati and C. Ibarra, Concentrations of
Sources and environmental processes of polycyclic polychlorinated biphenyls in indoor air and
aromatic hydrocarbons and mercury along a southern polybrominated diphenyl ethers in indoor air and dust in
slope of the central Himalayas, Nepal, Environ. Sci. Pollut. Birmingham, United Kingdom: Implications for human
Res., 2016, 23, 13843–13852. exposure, Environ. Sci. Technol., 2006, 40(15), 4633–4638.
478 J. Ren, X. Wang, Y. Xue, P. Gong, D. R. Joswiak, B. Xu and 490 K. P. Messier, L. G. Tidwell, C. C. Ghetu, D. Rohlman,
T. Yao, Persistent organic pollutants in mountain air of R. P. Scott, L. M. Bramer, H. M. Dixon, K. M. Waters and
the southeastern Tibetan Plateau: Seasonal variations and K. A. Anderson, Indoor versus outdoor air quality during
implications for regional cycling, Environ. Pollut., 2014, wildres, Environ. Sci. Technol. Lett., 2019, 6, 696–701.
194, 210–216. 491 M. D. Ampleman, A. Martinez, J. DeWall, D. F. K. Rawn,
479 P. Gong, X.-P. Wang, B. Pokhrel, H. Wang, X. Liu, X. Liu and K. C. Hornbuckle and P. S. Thorne, Inhalation and
F. Wania, Trans-Himalayan transport of organochlorine dietary exposure to PCBs in urban and rural cohorts via
compounds: three-years of observations and model-based congener-specic measurements, Environ. Sci. Technol.,
ux estimation, Environ. Sci. Technol., 2019, 53, 6773–6783. 2015, 49, 1156–1164.
480 J. Wu, W. Gao, Y. Liang, J. Fu, Y. Gao, Y. Wang and G. Jiang, 492 R. F. Marek, P. S. Thorne, N. J. Herkert, A. M. Awad and
Spatiotemporal distribution and alpine behavior of short K. C. Hornbuckle, Airborne PCBs and OH-PCBs inside
chain chlorinated paraffins in air at Shergyla Mountain and outside urban and rural U.S. schools, Environ. Sci.
and Lhasa on the Tibetan Plateau of China, Environ. Sci. Technol., 2017, 51, 7853–7860.
Technol., 2017, 51, 11136–11144. 493 A.-L. Rantalainen, T. Hyötyläinen, M. Saramo and
481 J. Wu, D. Cao, W. Gao, K. Lv, Y. Liang, J. Fu, Y. Gao, Y. Wang I. Niskanen, Passive sampling of PAHs in indoor air in
and G. Jiang, The atmospheric transport and pattern of Nepal, Toxicol. Environ. Chem., 1999, 68, 335–348.
medium chain chlorinated paraffins at Shergyla 494 M. Khairy and R. Lohmann, Organophosphate ame
Mountain on the Tibetan Plateau of China, Environ. retardants in the indoor and outdoor dust and gas-phase
Pollut., 2019, 245, 46–52. of Alexandria, Egypt, Chemosphere, 2019, 220, 275–285.
482 X. Zhang, J. Barnes, Y. D. Lei and F. Wania, Semivolatile 495 B. Sha, A.-K. Dahlberg, K. Wiberg and L. Ahrens,
organic contaminants in the Hawaiian atmosphere, Fluorotelomer alcohols (FTOHs), brominated ame
Environ. Sci. Technol., 2017, 51, 11634–11642. retardants (BFRs), organophosphorus ame retardants
483 C. Shunthirasingham, F. Wania, M. MacLeod, Y. D. Lei, (OPFRs) and cyclic volatile methylsiloxanes (cVMSs) in
C. L. Quinn, X. Zhang, M. Scheringer, F. Wegmann, indoor air from occupational and home environments,
K. Hungerbühler, S. Ivemeyer, F. Heil, P. Klocke, Environ. Pollut., 2018, 241, 319–330.
G. Pacepavicius and M. Alaee, Mountain cold-trapping 496 Y. Wan, M. L. Diamond and J. A. Siegel, Elevated
increases transfer of persistent organic pollutants from concentrations of semivolatile organic compounds in
atmosphere to cow's milk, Environ. Sci. Technol., 2013, 47, social housing multiunit residential building apartments,
9175–9181. Environ. Sci. Technol. Lett., 2020, 7, 191–197.
484 R. O. Meire, M. Khairy, D. Aldeman, P. M. A. Galvão, 497 L. Melymuk, P. Bohlin-Nizzetto, Š. Vojta, M. Krátká,
J. P. M. Torres, O. Malm and R. Lohmann, Passive P. Kukučka, O. Audy, P. Přibylová and J. Klánová,
sampler-derived concentrations of PAHs in air and water Distribution of legacy and emerging semivolatile organic
along Brazilian mountain transects, Atmos. Pollut. Res., compounds in ve indoor matrices in a residential
2019, 10, 635–641. environment, Chemosphere, 2016, 153, 179–186.
485 L. Li, J. A. Arnot and F. Wania, Revisiting the contributions 498 X. Zhang, M. L. Diamond, M. Robson and S. Harrad,
of far- and near-eld routes to human aggregate exposure Sources, emissions, and fate of polybrominated diphenyl
to polychlorinated biphenyls (PCBs), Environ. Sci. ethers and polychlorinated biphenyls indoors in Toronto,
Technol., 2018, 52, 6974–6984. Canada, Environ. Sci. Technol., 2011, 45, 3268–3274.
486 R. E. Dodson, D. E. Camann, R. Morello-Frosch, J. G. Brody 499 H. H. Abd Hamid, L. W. Wan, I. P. Ren Kai, M. T. Latif and
and R. A. Rudel, Semivolatile organic compounds in N. Kannan, Distribution of gas phase polycyclic aromatic
homes: Strategies for efficient and systematic exposure hydrocarbons (PAHs) in selected indoor and outdoor air

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 1999
Environmental Science: Processes & Impacts Critical Review

samples of Malaysia: A case study in Serdang, Selangor and biphenyls (PCBs) in indoor and outdoor air in Bursa,
Bachang, Malacca, EnvironmentAsia, 2017, 10, 80–85. Turkey: Levels and an assessment of human exposure via
500 M. Venier, O. Audy, Š. Vojta, J. Bečanová, K. Romanak, inhalation, Atmos. Pollut. Res., 2020, 11, 71–80.
L. Melymuk, M. Krátká, P. Kukučka, J. Okeme, A. Saini, 512 B. Strandberg, C. Österman, H. K. Akdeva, J. Moldanová
M. L. Diamond and J. Klánová, Brominated ame and S. Langer, The use of polyurethane foam (PUF)
retardants in the indoor environment - Comparative study passive air samplers in exposure studies to PAHs in
of indoor contamination from three countries, Environ. Swedish seafarers, Polycyclic Aromat. Compd., 2020, DOI:
Int., 2016, 94, 150–160. 10.1080/10406638.2020.1739084.
501 M. Vykoukalová, M. Venier, Š. Vojta, L. Melymuk, 513 Y.-X. Feng, N.-X. Feng, L.-J. Zeng, X. Chen, L. Xiang,
J. Bečanová, K. Romanak, R. Prokeš, J. O. Okeme, Y.-W. Li, Q.-Y. Cai and C.-H. Mo, Occurrence and human
A. Saini, M. L. Diamond and J. Klánová, health risks of phthalates in indoor air of laboratories,
Organophosphate esters ame retardants in the indoor Sci. Total Environ., 2020, 707, 135609.
environment, Environ. Int., 2017, 106, 97–104. 514 E. Holt, P. Bohlin-Nizzetto, J. Borůvková, T. Harner,
502 O. Audy, L. Melymuk, M. Venier, S. Vojta, J. Bečanová, J. Kalina, L. Melymuk and J. Klánová, Using long-term air
K. Romanak, M. Vykoukalová, R. Prokeš, P. Kukučka, monitoring of semi-volatile organic compounds to
M. L. Diamond and J. Klánová, PCBs and organochlorine evaluate the uncertainty in polyurethane-disk passive
pesticides in indoor environments - A comparison of sampler-derived air concentrations, Environ. Pollut., 2017,
indoor contamination in Canada and Czech Republic, 220, 1100–1111.
Chemosphere, 2018, 206, 622–631. 515 J. Kalina, M. Scheringer, J. Borůvková, P. Kukučka,
503 W. Han, T. Fan, B. Xu, J. Feng, G. Zhang, M. Wu, Y. Yu and P. Přibylová, P. Bohlin-Nizzetto and J. Klánová, Passive air
J. Fu, Passive sampling of polybrominated diphenyl ethers samplers as a tool for assessing long-term trends in
in indoor and outdoor air in Shanghai, China: seasonal atmospheric concentrations of semivolatile organic
variations, sources, and inhalation exposure, Environ. Sci. compounds, Environ. Sci. Technol., 2017, 51, 7047–7054.
Pollut. Res., 2016, 23, 5771–5781. 516 J. Kalina, M. Scheringer, J. Borůvková, P. Kukučka,
504 A. Boudehane, A. Lounas, Y. Moussaoui, C. Balducci and P. Přibylová, O. Sáňka, L. Melymuk, M. Váňa and
A. Cecinato, Levels of organic compounds in interiors J. Klánová, Characterizing spatial diversity of passive
(school, home, university and hospital) of Ouargla city, sampling sites for measuring levels and trends of
Algeria, Atmos. Environ., 2016, 144, 266–273. semivolatile organic chemicals, Environ. Sci. Technol.,
505 I. C. Yadav, N. L. Devi, J. Li and G. Zhang, Occurrence and 2018, 52(18), 10599–10608.
source apportionment of halogenated ame retardants in 517 J. Kalina, K. B. White, M. Scheringer, P. Přibylová,
the indoor air of Nepalese cities: Implication on human P. Kukučka, O. Audy and J. Klánová, Comparability of
health, Atmos. Environ., 2017, 161, 122–131. long-term temporal trends of POPs from co-located active
506 J. Sun, Q. Wang, S. Zhuang and A. Zhang, Occurrence of and passive air monitoring networks in Europe, Environ.
polybrominated diphenyl ethers in indoor air and dust in Sci.: Processes Impacts, 2019, 21, 1132–1142.
Hangzhou, China: Level, role of electric appliances, and 518 M. Sorais, M. J. Mazerolle, J.-F. Giroux and J. Verreault,
human exposure, Environ. Pollut., 2016, 218, 942–949. Landlls represent signicant atmospheric sources of
507 S. Newton, U. Sellström, S. Harrad, G. Yu and C. A. de Wit, exposure to halogenated ame retardants for urban-
Comparisons of indoor active and passive air sampling adapted gulls, Environ. Int., 2020, 135, 105387.
methods for emerging and legacy halogenated ame 519 X. Liu and F. Wania, Cluster analysis of passive air
retardants in Beijing, China offices, Emerging Contam., sampling data based on the relative composition of
2016, 2, 80–88. persistent organic pollutants, Environ. Sci.: Processes
508 K. Winkens, J. Koponen, J. Schuster, M. Shoeib, Impacts, 2014, 16, 453–463.
R. Vestergren, U. Berger, A. M. Karvonen, J. Pekkanen, 520 X. Wang, J. Ren, P. Gong, C. Wang, Y. Xue, T. Yao and
H. Kiviranta and I. T. Cousins, Peruoroalkyl acids and R. Lohmann, Spatial distribution of the persistent organic
their precursors in indoor air sampled in children's pollutants across the Tibetan Plateau and its linkage with
bedrooms, Environ. Pollut., 2017, 222, 423–432. the climate systems: A 5-year air monitoring study, Atmos.
509 W. Gao, D. Cao, Y. Wang, J. Wu, Y. Wang, Y. Wang and Chem. Phys., 2016, 16, 6901–6911.
G. Jiang, External exposure to short- and medium-chain 521 S. J. Hayward, T. Gouin and F. Wania, Comparison of four
chlorinated paraffins for the general population in active and passive sampling techniques for pesticides in
Beijing, China, Environ. Sci. Technol., 2018, 52, 32–39. air, Environ. Sci. Technol., 2010, 44, 3410–3416.
510 H.-L. Li, L.-Y. Liu, Z.-F. Zhang, W.-L. Ma, E. Sverko, 522 C. Cocheo, P. Sacco, P. Perez Ballesta, E. Donato, S. Garcia,
Z. Zhang, W.-W. Song, Y. Sun and Y.-F. Li, Semi-volatile M. Gerboles, D. Gombert, B. McManus, R. Fernandez
organic compounds in infant homes: Levels, inuence Patier, C. Roth, E. de Saeger and E. Wright, Evaluation of
factors, partitioning, and implications for human the best compromise between the urban air quality
exposure, Environ. Pollut., 2019, 251, 609–618. monitoring resolution by diffusive sampling and resource
511 M. F. Sari, F. Esen, D. A. C. Del Aguila and P. B. Kurt requirements, J. Environ. Monit., 2008, 10, 941–950.
Karakus, Passive sampler derived polychlorinated

2000 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020
Critical Review Environmental Science: Processes & Impacts

523 G. Hoek, R. Beelen, K. de Hoogh, D. Vienneau, J. Gulliver, transport local pollution, Environ. Sci. Technol., 2008, 42,
P. Fischer and D. Briggs, A review of land-use regression 7125–7131.
models to assess spatial variation of outdoor air 535 S.-Y. Baek, S.-D. Choi and Y.-S. Chang, Three-year
pollution, Atmos. Environ., 2008, 42, 7561–7578. atmospheric monitoring of organochlorine pesticides and
524 L. Melymuk, M. Robson, P. A. Helm and M. L. Diamond, polychlorinated biphenyls in polar regions and the South
Application of land use regression to identify sources and Pacic, Environ. Sci. Technol., 2011, 45, 4475–4482.
assess spatial variation in urban SVOC concentrations, 536 X.-P. Wang, P. Gong, T.-D. Yao and K. C. Jones, Passive air
Environ. Sci. Technol., 2013, 47, 1887–1895. sampling of organochlorine pesticides, polychlorinated
525 R. Liard, M. Zureik, Y. Le Moullec, D. Soussan, M. Glorian, biphenyls, and polybrominated diphenyl ethers across the
A. Grimfeld and F. Neukirch, Use of personal passive Tibetan Plateau, Environ. Sci. Technol., 2010, 44, 2988–2993.
samplers for measurement of NO2, NO, and O3 levels in 537 K. L. Foster, L. E. Kimpe, S. K. Brimble, H. Liu,
panel studies, Environ. Res., 1999, 81, 339–348. M. L. Mallory, J. P. Smol, R. W. Macdonald and
526 N. Jariyasopit, P. Tung, K. Su, S. Halappanavar, G. J. Evans, J. M. Blais, Effects of seabird vectors on the fate,
Y. Su, S. Khoomrung and T. Harner, Polycyclic aromatic partitioning, and signatures of contaminants in a High
compounds in urban air and associated inhalation cancer Arctic ecosystem, Environ. Sci. Technol., 2011, 45, 10053–
risks: A case study targeting distinct source sectors, 10060.
Environ. Pollut., 2019, 252, 1882–1891. 538 Y. Li, D. Geng, Y. B. Hu, P. Wang, Q. Zhang and G. Jiang,
527 A. P. Francisco, A. C. Nardocci, M. Y. Tominaga, C. R. da Levels and distribution of polychlorinated biphenyls in
Silva and J. V. de Assunção, Spatial and seasonal trends the atmosphere close to Chinese Great Wall Station,
of polychlorinated dioxins, furans and dioxin-like Antarctica: Results from XAD-resin passive air sampling,
polychlorinated biphenyls in air using passive and active Chin. Sci. Bull., 2012, 57, 1499–1503.
samplers and inhalation risk assessment, Atmos. Pollut. 539 Y. Hao, Y. Li, X. Han, T. Wang, R. Yang, P. Wang, K. Xiao,
Res., 2017, 8, 979–987. W. Li, H. Lu, J. Fu, Y. Wang, J. Shi, Q. Zhang and
528 M. Isidori, M. Ferrara, M. Lavorgna, A. Nardelli and G. Jiang, Air monitoring of polychlorinated biphenyls,
A. Parrella, In situ monitoring of urban air in Southern polybrominated diphenyl ethers and organochlorine
Italy with the Tradescantia micronucleus bioassay and pesticides in West Antarctica during 2011–2017:
semipermeable membrane devices (SPMDs), Concentrations, temporal trends and potential sources,
Chemosphere, 2003, 52, 121–126. Environ. Pollut., 2019, 249, 381–389.
529 G. Slapšytė, E. Lastauskiene and J. Mierauskiene, 540 X. Han, Y. Hao, Y. Li, R. Yang, P. Wang, G. Zhang, Q. Zhang
Genotoxicity of airborne hydrophobic pollutants sampled and G. Jiang, Occurrence and distribution of
by semipermeable membrane devices (SPMDs) in Vilnius organophosphate esters in the air and soils of Ny-Ålesund
city, Biologija, 2006, 1, 41–46. and London Island, Svalbard, Arctic, Environ. Pollut.,
530 K. Kennedy, M. Macova, F. Leusch, M. E. Bartkow, 2020, 263, 114495.
D. W. Hawker, B. Zhao, M. S. Denison and J. F. Müller, 541 Y. Li, D. Geng, F. Liu, T. Wang, P. Wang, Q. Zhang and
Assessing indoor air exposures using passive sampling G. Jiang, Study of PCBs and PBDEs in King George Island,
with bioanalytical methods for estrogenicity and aryl Antarctica, using PUF passive air sampling, Atmos.
hydrocarbon receptor activity, Anal. Bioanal. Chem., 2009, Environ., 2012, 51, 140–145.
394, 1413–1421. 542 K. Pozo, T. Martellini, S. Corsolini, T. Harner,
531 S. Bonetta, E. Carraro, S. Bonetta, C. Pignata, I. Pavan, V. H. Estellano, P. Kukučka, M. D. Mulder, G. Lammel
C. Romano and G. Gilli, Application of semipermeable and A. Cincinelli, Persistent organic pollutants (POPs) in
membrane device (SPMD) to assess air genotoxicity in an the atmosphere of coastal areas of the Ross Sea,
occupational environment, Chemosphere, 2009, 75, 1446– Antarctica: Indications for long-term downward trends,
1452. Chemosphere, 2017, 178, 458–465.
532 A. Érseková, K. Hilscherová, J. Klánová, J. P. Giesy and 543 C. Wang, X. Wang, P. Gong and T. Yao, Polycyclic aromatic
J. Novák, Effect-based assessment of passive air samples hydrocarbons in surface soil across the Tibetan Plateau:
from four countries in Eastern Europe, Environ. Monit. Spatial distribution, source and air–soil exchange,
Assess., 2014, 186, 3905–3916. Environ. Pollut., 2014, 184, 138–144.
533 S. J. Wallace, S. R. de Solla, P. J. Thomas, T. Harner, A. Eng 544 X. Wang, J. Schuster, K. C. Jones and P. Gong, Occurrence
and V. S. Langlois, Airborne polycyclic aromatic and spatial distribution of neutral peruoroalkyl
compounds contribute to the induction of the tumour- substances and cyclic volatile methylsiloxanes in the
suppressing P53 pathway in wild double-crested atmosphere of the Tibetan Plateau, Atmos. Chem. Phys.,
cormorants, Ecotoxicol. Environ. Saf., 2018, 150, 176–189. 2018, 18, 8745–8755.
534 S.-D. Choi, S.-Y. Baek, Y.-S. Chang, F. Wania, 545 A. De la Torre, P. Sanz, I. Navarro and M. Á. Martı́nez, Time
M. G. Ikonomou, Y.-Y. Yoon, B.-K. Park and S. Hong, trends of persistent organic pollutants in Spanish air,
Passive air sampling of polychlorinated biphenyls and Environ. Pollut., 2016, 217, 26–32.
organochlorine pesticides at the Korean Arctic and 546 J. Muñoz-Arnanz, J. L. Roscales, M. Ros, A. Vicente and
Antarctic research stations: Implications for long-range B. Jiménez, Towards the implementation of the

This journal is © The Royal Society of Chemistry 2020 Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 | 2001
Environmental Science: Processes & Impacts Critical Review

Stockholm Convention in Spain: Five-year monitoring concentration of polychlorinated biphenyls and

(2008–2013) of POPs in air based on passive sampling, organochlorine pesticides in air over the Northern
Environ. Pollut., 2016, 217, 107–113. Hovsgol Region in 2008–2015, Russ. Meteorol. Hydrol.,
547 J. L. Roscales, J. Muñoz-Arnanz, M. Ros, A. Vicente, 2019, 44, 78–85.
L. Barrios and B. Jiménez, Assessment of POPs in air 550 M. Y. Tominaga, C. R. Silva, J. P. Melo, N. A. Niwa,
from Spain using passive sampling from 2008 to 2015. D. Plascak, C. A. M. Souza and M. I. Z. Sato, PCDD, PCDF,
Part I: Spatial and temporal observations of PBDEs, Sci. DL-PCB and organochlorine pesticides monitoring in São
Total Environ., 2018, 634, 1657–1668. Paulo City using passive air sampler as part of the Global
548 J. Muñoz-Arnanz, J. L. Roscales, A. Vicente, M. Ros, Monitoring Plan, Sci. Total Environ., 2016, 571, 323–331.
L. Barrios, L. Morales, E. Abad and B. Jiménez, 551 F. Wong, A. Alegria and T. F. Bidleman, Organochlorine
Assessment of POPs in air from Spain using passive pesticides in soils of Mexico and the potential for soil-air
sampling from 2008 to 2015. Part II: Spatial and temporal exchange, Environ. Pollut., 2010, 158, 749–755.
observations of PCDD/Fs and dl-PCBs, Sci. Total Environ., 552 P. Karásková, G. Codling, L. Melymuk and J. Klánová, A
2018, 634, 1669–1679. critical assessment of passive air samplers for per- and
549 E. A. Mamontova, E. N. Tarasova, A. A. Mamontov, polyuoroalkyl substances, Atmos. Environ., 2018, 185,
A. V. Goreglyad and L. L. Tkachenko, Variations in the 186–195.

2002 | Environ. Sci.: Processes Impacts, 2020, 22, 1925–2002 This journal is © The Royal Society of Chemistry 2020